362 96 3MB
English Pages [227] Year 2021
Agrarian Extractivism in Latin America
Amid the growing calls for a turn towards sustainable agriculture, this book puts forth and discusses the concept of agrarian extractivism to help us identify and expose the predatory extractivist features of dominant agricultural development models. The concept goes beyond the more apparent features of monocultures and raw material exports to examine the inherent logic and underlying workings of a model based on the appropriation of an ever-growing range of commodified and non-commodified human and non-human nature in an extractivist fashion. Such a process erodes the autonomy of resourcedependent working people, dispossesses the rural poor, exhausts and expropriates nature, and concentrates value in a few hands as a result of the unquenchable drive for profit by big business. In many instances, such extractivist dynamics are subsidized and/or directly supported by the state, while also dependent on the unpaid, productive, and reproductive labour of women, children, and elders, exacerbating unequal class, gender, and generational relations. Rather than a one-size-fits-all definition of agrarian extractivism, this collection points to the diversity of extractivist features of corporate-led, external-input-dependent plantation agriculture across distinct socio-ecological formations in Latin America. This timely challenge to the destructive dominant models of agricultural development will interest scholars, activists, researchers, and students from across the fields of critical development studies, rural studies, environmental and sustainability studies, and Latin American studies, among others. Ben M. McKay is Assistant Professor of Development and Sustainability in the Department of Anthropology and Archaeology at the University of Calgary in Canada. His research focuses on the political economy and ecology of agrarian change in Latin America, agrarian extractivism, and food sovereignty alternatives. He is the author of The Political Economy of Agrarian Extractivism: Lessons from Bolivia (2020) and co-editor of The Edward Elgar Handbook of Critical Agrarian Studies (2021) and Rural Transformations and Agro-Food Systems (2018). Alberto Alonso-Fradejas is Postdoctoral Researcher at the Human Geography and Planning Department, Faculty of Geosciences, Utrecht University. Alberto is also an Associate Researcher at the Transnational Institute (TNI) in Amsterdam, a Fellow of the Guatemalan Institute of Agrarian and Rural Studies (IDEAR), and Reviews Section Co-Editor for the Journal of Peasant Studies (JPS). Arturo Ezquerro-Cañete holds a double PhD in International Development Studies from Saint Mary’s University and the Autonomous University of Zacatecas. His research focuses on the dynamics of agrarian transformations and new peasant movements in Paraguay. His work has been published in scholarly journals such as Journal of Agrarian Change, Latin American Perspectives, and Estudios Críticos del Desarrollo.
Routledge Critical Development Studies Series Editors Henry Veltmeyer is co-chair of the Critical Development Studies (CDS)
network, Research Professor at Universidad Autónoma de Zacatecas, Mexico, and Professor Emeritus at Saint Mary’s University, Canada Paul Bowles is Professor of Economics and International Studies at UNBC, Canada Elisa van Wayenberge is Lecturer in Economics at SOAS University of London, UK The global crisis, coming at the end of three decades of uneven capitalist development and neoliberal globalization that have devastated the economies and societies of people across the world, especially in the developing societies of the global south, cries out for a more critical, proactive approach to the study of international development. The challenge of creating and disseminating such an approach, to provide the study of international development with a critical edge, is the project of a global network of activist development scholars concerned and engaged in using their research and writings to help effect transformative social change that might lead to a better world. This series will provide a forum and outlet for the publication of books in the broad interdisciplinary field of critical development studies – to generate new knowledge that can be used to promote transformative change and alternative development. The editors of the series welcome the submission of original manuscripts that focus on issues of concern to the growing worldwide community of activist scholars in this field. To submit proposals, please contact the Development Studies Editor, Helena Hurd ([email protected]). 9 Deconstructing Human Development From the Washington Consensus to the 2030 Agenda Juan Telleria 10 Revolutions in Learning and Education from India Pathways towards the Pluriverse Christoph Neusiedl 11 Agrarian Extractivism in Latin America Edited by Ben M. McKay, Alberto Alonso-Fradejas, and Arturo Ezquerro-Cañete For more information about this series, please visit: www.routledge.com/ Routledge-Critical-Development-Studies/book-series/RCDS
Agrarian Extractivism in Latin America Edited by Ben M. McKay, Alberto Alonso-Fradejas, and Arturo Ezquerro-Cañete
First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 selection and editorial matter, Ben M. McKay, Alberto Alonso-Fradejas, and Arturo Ezquerro-Cañete; individual chapters, the contributors The right of Ben M. McKay, Alberto Alonso-Fradejas, and Arturo Ezquerro-Cañete to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: McKay, Ben M., editor. | Alonso Fradejas, Alberto, editor. | Ezquerro-Cañete, Arturo, editor. Title: Agrarian extractivism in Latin America / edited by Ben M. McKay, Alberto Alonso-Fradejas, and Arturo Ezquerro-Cañete. Description: Milton Park, Abingdon, Oxon ; New York, NY : Routledge, 2021. | Series: Routledge critical development series | Includes bibliographical references and index. Subjects: LCSH: Agriculture—Economic aspects—Latin America. | Agriculture—Social aspects—Latin America. | Agriculture and state— Latin America. Classification: LCC HD1790.5 .A47 2021 (print) | LCC HD1790.5 (ebook) | DDC 338.1098—dc23 LC record available at https://lccn.loc.gov/2020053873 LC ebook record available at https://lccn.loc.gov/2020053874 ISBN: 978-0-367-42254-7 (hbk) ISBN: 978-1-032-00607-9 (pbk) ISBN: 978-0-367-82295-8 (ebk) Typeset in Times New Roman by Apex CoVantage, LLC
Contents
List of illustrations and figures List of tables Notes on contributors Acknowledgements Introduction
vii viii ix xii 1
BEN M. McKAY, ALBERTO ALONSO-FRADEJAS, AND ARTURO EZQUERRO-CAÑETE
1
The Biotechnological Agrarian Model in Argentina: fighting against capital within science
21
CARLA POTH
2
Extractive dynamics of agrarian change in Bolivia
45
BEN M. McKAY AND GONZALO COLQUE
3
Agrarian extractivism in the Brazilian Cerrado
64
SÉRGIO SAUER AND KARLA R.A. OLIVEIRA
4
Social reproduction, dispossession, and the gendered workings of agrarian extractivism in Colombia
85
DIANA OJEDA
5
Agrarian extractivism and sustainable development: the politics of pineapple expansion in Costa Rica ANDRÉS LEÓN ARAYA
99
vi
Contents
6
Gender inclusion in the sugarcane production of agrofuels in coastal Ecuador: illusionary promises of rural development within a new agrarian extractivism
117
NATALIA LANDÍVAR GARCÍA
7
Life purging agrarian extractivism in Guatemala: towards a renewable but unlivable future?
139
ALBERTO ALONSO-FRADEJAS
8
Extractive agave and tequila production in Jalisco, Mexico
165
DARCY TETREAULT, CINDY McCULLIGH, AND CARLOS LUCIO
9
Forestry extractivism in Uruguay
186
MARKUS KRÖGER AND MARIA EHRNSTRÖM-FUENTES
Index
208
Illustrations and figures
1.1 1.2 6.1 6.2 7.1 7.2 8.1 8.2
Sociogram: educational, academic, and work trajectories Percentage of GM seeds released in each phase Sugarcane cultivated area in hectares (2002–2019) Women using contaminated water sources to wash clothes in Hacienda La Indiana Research sub-regions, departments, and municipalities in Guatemala Interactive analysis of diverse productive relations around multiple forces of production in resource extractivism Tequila production and exports and agave consumption, 1995–2019 Agave harvested and area planted in municipalities of the DO, 2003–2017
30 38 122 132 145 146 172 174
Tables
1.1 6.1 7.1 8.1
Shared conceptions that are crystallized in regulation Membership in the associations of the Hacienda La Indiana Diverse productive relations around multiple forces of production Owners of main tequila brands
32 125 146 173
Contributors
Alberto Alonso-Fradejas is Postdoctoral Researcher at the Human Geography and Planning Department, Faculty of Geosciences, Utrecht University. Alberto is an Associate Researcher at the Transnational Institute (TNI) in Amsterdam, a Fellow of the Guatemalan Institute of Agrarian and Rural Studies (IDEAR), and Reviews Section Co-Editor for the Journal of Peasant Studies (JPS). https://orcid.org/0000-0002-5906-6608 Andrés León Araya has a PhD in Anthropology from the City University of New York (CUNY) and is currently the Chair of the Center of Political Research and Studies (Centro de Investigación y Estudios Políticos, CIEP) at the University of Costa Rica. His research interests concern the relation between the expansion of monoculture plantations and state formation in Central America. https:// orcid.org/0000-0002-1594-5184 Gonzalo Colque is a Bolivian researcher and Executive Director of Fundación TIERRA based in La Paz, Bolivia. He has an MA in Agrarian, Food, and Environmental Studies from the International Institute of Social Studies (ISS) in The Hague, the Netherlands, and has been deeply involved with peasant organizations and Indigenous movements in Bolivia. Maria Ehrnström-Fuentes is Assistant Professor at Hanken School of Economics. Her research focuses on the political implications of corporate social responsibility in local struggles against forestry extractivism in South America, and on grassroots organizations enacting socio-ecological change through regenerative projects in different parts of the world. She has also published articles on sustainability and the governance of ethicality in alternative food networks. Issues related to political ontology and the pluriverse are core themes that cut through all of her research. Arturo Ezquerro-Cañete holds a double PhD in International Development Studies from Saint Mary’s University and the Autonomous University of Zacatecas. His research focuses on the dynamics of agrarian transformations and new peasant movements in Paraguay. His work has been published in scholarly journals such as Journal of Agrarian Change, Latin American Perspectives, and Estudios Críticos del Desarrollo. https://orcid.org/0000-0003-4604-0270
x Contributors Markus Kröger is Associate Professor of Global Development Studies at the University of Helsinki and a Fellow at the Academy of Finland. He has written extensively on natural resource politics and social movements, especially in relation to mining and forestry in Latin America and the Arctic. He is the author of Iron Will: Global Extractivism and Mining Resistance in Brazil and India. Currently he focuses on the world-ecology and political ontology of global extractivisms and deforestation. https://orcid.org/0000-0001-7324-4549 Natalia Landívar García is a right-to-food activist and academic committed to socially engaged research. In 2016, she started a PhD at the Natural Resources Institute of the University of Manitoba. She has drawn upon her 15 years of work as an activist advocating for the right to food and peasant’s right to inform and articulate her academic research projects. Her MA at the Facultad Latinoamericana de Ciencias Sociales (FLACSO) Ecuador and her current PhD thesis shed light on agrarian policies and the land struggles of grassroots organizations in coastal Ecuador. https://orcid.org/0000-0001-6889-7131 Carlos Lucio has a master’s degree in Sociology from the University of Coimbra, Portugal, and a PhD in Social Sciences from the Center for Research and Advanced Studies in Social Anthropology (CIESAS), Mexico. His research interests include social environmental conflicts, biocultural diversity and Indigenous movements. Recent books are (2016) Conflictos socioambientales, derechos humanos y movimiento indígena en el Istmo de Tehuantepec, UAZ; (2018) Social Environmental Conflicts in Mexico. Resistance to Dispossession and Alternatives from Below (co-editor with D. Tetreault and C. McCulligh), Palgrave Macmillan. https://orcid.org/0000-0002-0356-0877 Cindy McCulligh is Professor and Researcher at the Center for Research and Advanced Studies in Social Anthropology (CIESAS) in Guadalajara, Mexico. She has her PhD in Social Science from CIESAS, and her master’s in Environmental Studies from York University. Her research interests include social environmental conflicts, water management, environmental regulation, industrial pollution, and environmental justice. https://orcid.org/0000-0001-6447-7842 Ben M. McKay is Assistant Professor of Development and Sustainability in the Department of Anthropology and Archaeology at the University of Calgary in Canada. His research focuses on the political economy and ecology of agrarian change in Latin America, agrarian extractivism, and food sovereignty alternatives. He is the author of The Political Economy of Agrarian Extractivism: Lessons from Bolivia (2020) and co-editor of The Edward Elgar Handbook of Critical Agrarian Studies (2021) and Rural Transformations and Agro-Food Systems (2018). https://orcid.org/0000-0002-5737-5255 Diana Ojeda is Associate Professor at Cider (Interdisciplinary Center for Development Studies), Universidad de los Andes, Bogotá, Colombia. She holds a PhD degree in Geography from Clark University. Her work bridges feminist political ecology, feminist geopolitics, and social studies of science in the
Contributors
xi
study of land grabbing and dispossession in the Colombian Caribbean. https:// orcid.org/0000-0003-2009-8060 Karla R.A. Oliveira has a master’s degree in Environment and Rural Development, is a specialist in Agrarian Law, and graduated in Forest Engineering at University of Brasilia. She worked with public policies and programs related to food and water security for traditional populations in the Amazon and the Cerrado. She is engaged in academic projects like the Matopiba Observatory, a network for researches and studies of social, economic, and political implications of the expansion of the agricultural frontier in the Cerrado. https://orcid.org/0000-0001-6349-1861 Carla Poth has a PhD in Social Sciences and is Professor of Sociology in the Instituto de Ciencias at the Universidad Nacional de General Sarmiento (UNGS). Poth is a social and environmental activist of the Multisectorial contra el Agronegocio – La 41-. Her main research topics include the expansion of the agrarian model in Argentina, public policies, national and transnational institutions, and regulatory frameworks in genetically modified seeds, freetrade agreements, and the production of knowledge in agribusiness. Sérgio Sauer has a PhD in Sociology and is a Professor at the University of Brasilia (UnB), in the Graduate Program for Environment and Rural Development (Mader), and in the Center for Sustainable Development (CDS). He holds a research scholarship from the Brazilian CNPq and is a Fellow of the human rights NGO Terra de Direitos. His main research themes are land (land grabbing) and environment issues, public policies for the countryside and agrarian social movements, and agribusiness. https://orcid.org/0000-0002-2014-3215 Darcy Tetreault is Professor and Researcher in the Department of Development Studies (Unidad Académica en Estudios del Desarrollo) at the Autonomous University of Zacatecas. His approach to research involves ongoing collaboration with popular forms of organized resistance to environmentally destructive activities and with collective efforts from below to construct alternatives. His academic interests include extractivism, environmental conflicts, agrarian transformation, and peasant and Indigenous movements. https://orcid.org/0000-0002-1157-7839
Acknowledgements
The seeds for this book were planted some years ago – in 2014 at the FLACSOISA Conference on “Global and Regional Powers in a Changing World” hosted by the University of Buenos Aires in Argentina. It was there, at a lunch between sessions, that Henry Veltmeyer proposed that we submit a book proposal on “agrarian extractivism” to the Routledge Critical Development Studies Book Series. The concept itself was in its very initial stages at the time, being developed separately by us, the editors, in our respective doctoral research projects on agro-commodity crop complexes in Bolivia, Guatemala, and Paraguay. We would like to thank Henry Veltmeyer for encouraging us to follow through with this book project, as well as the other book series co-editors, Elisa van Waeyenberge and Paul Bowles for supporting this project. This book benefitted immensely from a workshop titled, “Agro-food systems and rural futures: a comparative analysis into the extractive character of industrial agriculture”, held at the University of Calgary in September 2019. We are thankful to all those who participated in that workshop and, of course, all the contributing authors who made this collection possible. We certainly learned a lot from and with them. We are also grateful to Chelsea Klinke and Gertrude Samar for their help with the workshop and their excellent work as copyeditors for this volume. We would also like to thank the Social Sciences and Humanities Research Council of Canada (SSHRC), as well as the University of Calgary’s Latin American Research Centre (LARC), the Faculty of Arts and the Department of Anthropology and Archaeology for supporting the workshop and book project. We are thankful to the anonymous peer reviewers for providing comments, critiques, and suggestions on each chapter in this collection, and on the idea of the edited volume as a whole. Many thanks to the Routledge Editorial team, including Helena Hurd and Rosie Anderson for their help and support in finalizing this book project for publication. Finally, Ben M. McKay would like to thank Carolina for her advice, insights, and encouragement, as well as Sophia and Thomas for the motivation and inspiration. Alberto Alonso-Fradejas is grateful to Adriana for their enlightening conversations on this project and her generous influx of motivation to pursue it. Arturo Ezquerro-Cañete thanks Likda for all her patience, understanding, and emotional support throughout this project.
Introduction Ben M. McKay, Alberto Alonso-Fradejas, and Arturo Ezquerro-Cañete
Fuelled by linear theories of modernization and neo-Malthusian discourses on the need to feed a growing population through a focus on productivity rather than access, distribution and its biophysical contradictions, the dominant model of agriculture is not only resulting in a failure to meet the United Nations’ Sustainable Development Goals (SDGs) of the 2030 Agenda, but more importantly it is failing humanity and leading to the destruction of life on Earth. Global hunger is projected to reach over one billion by 2021, while poverty largely remains a rural problem with roughly 80% of the world’s extreme poor living in rural areas, most of whom depend on farming for their livelihood (FAO 2018). The COVID-19 pandemic has further exposed this system’s vulnerabilities and inequities as highly external-input dependent, centralized supply chains and just-in-time delivery systems have come to a halt. Lockdowns and border restrictions have constrained channels of distribution, not only affecting food availability but also leaving seasonal farmworkers even more precarious or jobless. Supply chain disruptions and falling consumer demand threaten farmer livelihoods, while consumers simultaneously face food shortages and higher prices in markets (Clapp 2020; UN 2020). Furthermore, agriculture, forestry and other land-use change, underpinned by a model dependent on fossil fuels, accounted for an estimated 23% of total net anthropogenic greenhouse gas emissions from 2007 to 2016 (IPCC 2019), while the expansion of large-scale monocrop agriculture is one of the main drivers of deforestation and biodiversity loss (FAO 2019, 2020). Across the spectrum, corporate-led, external-input plantation agriculture (CEPA) is generating severe socio-economic and socio-ecological crises. Yet, the model persists and remains atop the agenda for national governments and influential international organizations, especially when it comes to investments for agricultural research projects (BFED and IPES-Food 2020). Proponents claim that the CEPA model will develop industries which generate quality employment opportunities, forward and backward linkages, and value-added processing in the places in which it operates (World Bank 2007). This perspective maintains that the model’s deficiencies can be overcome with technical fixes such as “climate-smart” agriculture (World Bank 2015), combined with non-binding governance mechanisms and corporate codes of conduct like the Principles for Responsible Agricultural Investment (PRAI) and the Roundtable on Sustainable Palm Oil (RSPO), among
2 McKay et al. other strategies under the guise of sustainable development (see Alonso-Fradejas; León Araya, this volume). However, these depoliticized approaches and technical fixes have consistently failed to address important issues of power, access and inequality, while voluntary multi-stakeholder governance mechanisms are often riddled with loopholes and enable business-as-usual practices to continue garnished with corporate social responsibility washing (Margulis et al. 2013; Dauvergne 2018; Taylor 2018; Alonso-Fradejas 2020). Governments around the world, from the progressive left to the conservative right, continue to prioritize and support the CEPA model even in cases where their main constituents (e.g. farmers, peasants, rural labourers, Indigenous peoples, and landless rural dwellers) are increasingly dispossessed, excluded, and marginalized by the underlying logics and workings of this model. In Bolivia, for example, the so-called industrial agriculture is promoted as a means to achieve food security, food sovereignty, and rural prosperity (see McKay and Colque, this volume). In Brazil, even under progressive left governments of Luiz Inácio Lula da Silva and Dilma Rousseff, whose supporters included the Landless Rural Worker’s Movement (Movimento dos Trabalhadores Rurais Sem Terra, MST) among other agrarian justice organizations, support for agribusiness and its expansion into new frontiers actually increased (see Sauer 2019; Sauer and Oliveira, this collection). There are, of course, strategic political alliances at play in these contexts, but the rationale remains one of productivity (or a “Productive Revolution”, in the case of Bolivia), technical efficiency, and the need to industrialize the countryside. But is this model of development actually industrializing the countryside? What are its socio-economic and ecological implications and for whom? This book puts forth and discusses the concept of agrarian extractivism to help us identify and expose the predatory extractivist1 features of dominant agricultural development models.2 The concept goes beyond the more apparent features of monocultures and raw material exports to the inherent logic and underlying workings of a model based on the appropriation of commodified and non-commodified forces of production in an extractivist fashion. Such a process erodes farmer autonomy, dispossesses the rural poor, expropriates nature, and appropriates surplus value. In other words, it is extractivist in its relationship with human and non-human nature, increasingly deepening the rift in the social metabolic order as a result of its drive for profit. In many instances, such extractivist dynamics are subsidized and/or directly supported by the state, while also dependent on the unpaid, productive, and reproductive labour of women, children, and elders, exacerbating unequal gender and generational relations. Rather than a one-sizefits-all definition of agrarian extractivism, this collection points to the diversity of extractivist features of corporate-led, external-input-dependent, plantation agriculture across distinct socio-ecological formations in Latin America. Each chapter contributes to the concept in its own unique way from perspectives grounded in critical agrarian political economy and (feminist) political ecology. In fact, this conversation between agrarian and environmental studies is a main conceptual and methodological contribution of this volume.
Introduction 3 In the remainder of this introductory chapter we discuss industrial capital’s penetration into, and transformation of, agriculture and some of its contradictions. We then turn to the extractivist features of the dominant agricultural model and put forth the concept of agrarian extractivism to better characterize this model, arguing that it offers analytical and political utility in the debates concerning dynamics of agrarian and environmental change, food systems, climate stewardship, and sustainable development more broadly. In the fourth section, we introduce this collection, followed by a discussion of what we see as the key aspects necessary for a comprehensive analysis of agrarian extractivism.
Industrial capital’s transformation of agriculture The CEPA model, often referred to as “industrial agriculture” or “agro-industry”, remains the dominant model for agricultural development in the global agro-food system and the primary means of combatting global rural poverty and hunger, as promoted by the most influential development agencies and international financial institutions (World Bank 2007). The International Assessment of Agricultural Knowledge, Science and Technology for Development (IAASTD) Report defines industrial agriculture as a “form of agriculture that is capital-intensive, substituting machinery and purchased inputs for human and animal labor” (IAASTD 2009, 563–4). It is highly mechanized and specialized, reliant on monocrop cultivation, often large in scale, highly dependent on industrialized external inputs and controlled by big business. It is based on the principles of comparative advantage, which encourages countries to specialize their agricultural production; on distant markets which relies on the international trade of food as a commodity; and on a complex web of global commodity chains which are largely controlled by corporate mega-mergers (see Sandwell 2019). There have been diverse debates in the literature concerning the social, economic, and ecological implications of the agro-industrial model vis-à-vis alternative models based on cooperative, family and/or peasant farming and agroecological methods. This includes both old and new debates pertaining to the role and viability of family and/or peasant farming and food sovereignty alternatives in generating a surplus and for feeding the world (McMichael 2009; Bernstein 2014; Rubio Vega 2018); the fate of the peasantry and small-scale family farmers based on socio-economic or demographic factors of differentiation (Boltvinik 2016; Kay 2016; van der Ploeg 2018); the relationship between farm size and productivity (Woodhouse 2010); the biophysical contradictions of the agro-industrial model (O’Connor 1998; Leff 2006; Weis 2010); and the rising global interest and investments in farmland (Deininger and Byerlee 2011; Borras et al. 2012; Soto Baquero and Gómez 2012). For proponents of the agro-industrial model, the integration of smallholders into global commodity chains via contract farming schemes is deemed the most effective way forward for rural poverty alleviation and rural development. According to the World Bank (2007, 241), for example, “producer organizations and contract farming are essential for these smallholders to take part in value chains and cater to supermarket demands” and “contract
4 McKay et al. farming . . . almost always results in higher income compared to that of similar farmers not on contract” (Minot and Ronchi 2014, 5). The World Bank’s “agriculture for development” approach rests on a neo-institutional economics framework focusing on transaction costs, economies of scale, monetary efficiency, coordination failures, and access to technology, among other factors (World Bank 2007; Minot and Ronchi 2014). However, critics argue that this model is centred on a market-based, residual logic, and technical approach, neglecting important relational aspects of power, ecological change, access, and control (Gras et al. 2009; Medeiros 2015). The industrial transformation of agriculture – first via mechanization which reduced the need for labour, then through the dissemination of hybrid and genetically modified (GM) seeds, and finally by means of the dependence on agrochemicals – has led to “a series of partial, discontinuous appropriations of the rural labour and biological production processes” (Goodman et al. 1987, 2). This is what Goodman, Sorj, and Wilkinson refer to as “appropriationism” which is “constituted by the action of industrial capitals to reduce the importance of nature in rural production, and specifically as a force beyond their direction and control” (Goodman et al. 1987, 3). Reducing the importance of nature’s complexity through biological simplification and standardization is necessary for the “industrial” model to achieve economies of scale and enhance its technical efficiency. However, this biological simplification and appropriationism “necessitates the chronic use of a range of biophysical overrides, or what amount to perpetual short-term ‘fixes’” (Weis 2010, 319). Brazil’s “Miracle of the Cerrado”, for example, required chemical fertilizers in order to “fix” its acidic soils so that improved seed varieties could be planted, albeit with devastating environmental impacts for the world’s most biodiverse savannah (Nehring 2016). Monocultures, worldwide, have been plagued with new weeds, pests, and plant disease which have been overridden by more toxic and increased quantities of chemical pesticides, resulting in a pesticide treadmill for farmers (Nicholls and Altieri 1997; Bakker et al. 2020). The adoption of GM soybeans in Bolivia, for example, has been followed by a drastic increase in agro-chemical usage, which is not only contaminating the soils and waterways but also increasing production costs for farmers, and leading to relations of debt and dependency with agribusiness (see McKay 2018). With farmers increasingly dependent on GM seeds, agro-chemicals, and machinery, industrial capital has captured agriculture by partially eliminating its material base and part of the natural production process incompatible with capital accumulation (Goodman et al. 1987, 156). Yet, all of these biophysical overrides come with hidden costs, impacting rural livelihoods, human health, and the environment. These overrides and externalities call into question this model which prides itself on “efficiency”. As Tony Weis (2010, 321) writes: in order to simplify, standardize and mechanize agriculture, and increase productivity per worker, plant and animal, a series of biophysical barriers must be overridden. Efficiency gains therefore hinge on many unaccounted, nonrenewable and actively destructive fixes, with fossilized biomass having an
Introduction 5 indispensable role in this process. While it involves a complex calculation, one common estimate is that industrial agriculture requires an average of 10 calories of fossil fuels to produce a single calorie of food. Industrial capital’s transformation of agriculture has not only resulted in biophysical contradictions and inefficiencies, it has also altered the social relations of production, property, and power in the countryside. Mechanization, the standardization of labour processes, and the increasing use of external inputs have reduced the need for labour, while contract farming schemes and incorporating farmers into commodity chains have introduced new and subtle ways to control land, labour, and other natural resources without owning the land outright. Further, the capital-intensive model excludes many rural dwellers from even engaging in agriculture, forcing them to sell their labour power for a wage, as everything from the seed, inputs, machinery, and market access have come to be controlled by market oligopolies with unprecedented power and influence over the global agrifood system (Sandwell 2019; Clapp and Purugganan 2020). For example, it has been reported that corporate agribusiness giants ADM, Bunge, Cargill, and Louis Dreyfus control an estimated 75–90% of the global grain trade (Murphy et al. 2012; Wesz Jr 2016). The seed, agro-chemical, and research and development (R&D) sectors are similarly characterized by extremely high levels of market concentration. in 2015 and 2016 three colossal mergers were announced: Bayer-Monsanto, Dow-Dupont, and Syngenta-ChemChina (with ChemChina a new player on the global stage). While BASF, the last of the Big Six, isn’t merging it has been picking up parts of the other five companies that they have been selling off during their mergers, like 5.9 billion euros worth of Bayer’s seed and herbicide business. In early 2017, it was announced that ChemChina would merge with another Chinese chemical firm, Sinochem, to form the largest chemical company in the world (Sandwell 2019). For those who do participate in “independent” farming, the increased reliance on commodified inputs “upstream” and corporate control “downstream” erodes farmer autonomy. As Kloppenburg (2004, 34) puts it, “[t]he means of production come to confront the farmer as commodities – they can be purchased but they cannot be autonomously reproduced”. This process binds farmers to off-farm capital, eliminating important functions from the farm, while simultaneously allowing for the extraction of surplus value in the off-farm industrial setting and both direct and indirect squeeze of the farmer (Bernstein 1979; Kloppenburg 2004). As Bernstein argues: The pressures which result in the ‘squeeze’ on simple reproduction include those arising from the exhaustion of both land and labour given the techniques of cultivation employed, from rural ‘development’ schemes which
6 McKay et al. encourage or impose more expensive means of production (improved seeds, tools, more extensive use of fertilisers, insecticides, pesticides, etc.) with no assurance that there will be increased returns to labour commensurate with the costs incurred, and from deteriorating terms of exchange for peasant produced commodities. (Bernstein 1979, 427). More than the material commodification and control of agro-inputs, just four companies (BASF, Bayer, ChemChina, and DowDuPoint) have a combined budget of an estimated 20 times larger than the Consortium of International Agricultural Research Centres (CGIAR) and 15 times that of the United States Department of Agriculture’s Agricultural Research Service budget for crop science research, giving them significant control over the agricultural R&D industry (ETC Group 2015). As Kloppenburg argues, agricultural research has been “an important means of eliminating the barriers to the penetration of agriculture by capital” by commodifying agro-inputs and displacing productive activities off the farm and into an industrial setting (Kloppenburg 2004, 10). But while agricultural innovations are certainly important, it becomes problematic when a small, selfinterested group dictates the research agenda and the interlinked technologies and products which are available in the market (see Miller and Conko 2001). With significant influence and control over the agricultural research agenda, this market oligopoly can invest in shaping agriculture’s technical form through continued innovations involving their technological packages complete with patented seeds, agro-chemical inputs, and advanced mechanization (see Poth, this volume). This control over R&D and knowledge production represents the power agribusiness has over patented ideas and technological innovation, and ultimately authority over the terms of modern agricultural production. As commodification deepens across all aspects of agricultural production – knowledge, seeds, tools, livestock, land, and labour – they become subject to the discipline of the market. And when that market is controlled by just a few companies with oligopolistic power they have the ability to craft, extract, and ultimately appropriate rents, energy, and materials from human and non-human nature alike, to the extent they want, and often with impunity (see Alonso-Fradejas, this volume). Deeper analyses thus point to the highly extractivist features of the CEPA model which, rather than industrialize the countryside, depredates it. Referring to this type of agricultural development as industrial agriculture is misleading both analytically and politically and requires a new conceptualization to not only reveal and highlight the extractivist features of many instances of corporate-led agriculture today but also to contribute to building fair, climate-sound, culturally sound, and truly sustainable models for agrarian development.
Extractivist dynamics of agriculture Extractivism has been referred to as a “mode of appropriation” which points to the different forms of organizing the appropriation of distinct natural resources
Introduction 7 (physical materials, energy, and ecological processes) for human purposes in specific social and environmental contexts (Gudynas 2015). Since we do not “produce” natural resources but rather appropriate or extract them from nature, Gudynas rejects the notion of “mode of production” when referring to extractivism (ibid.,188). From this perspective, extractivism is not analogous to an industry since the industrial, value-added processes usually occur in faraway places from the extraction. This builds on Bunker’s argument that the “internal dynamics of extractive economies differ significantly from those of productive economies in their effects on the natural environment, on the distribution of human populations, on the construction of economic infrastructure, and therefore on the subsequent development potential of the affected regions” (1984, 1019). Bunker (1984, 1019) goes on to say that “when natural resources are extracted from one regional ecosystem to be consumed or transformed in another, the socioeconomic and ecological linkages to the extracted commodity tend to a loss of value in the region of origin and to accretion of value in the region of consumption or transformation”. Extractivism not only leads to uneven economic and ecological exchange but can also have devastating social consequences. Incomes often rise and fall rapidly, populations are displaced, ecosystems are destroyed, and political elites become susceptible to forms of corruption. For Bunker, these processes represent “modes of extraction” which he introduced to characterize the systemic connections between changes in “the class structures; the organization of labour; systems of property and exchange; the activities of the state; the distribution of populations; the development of physical infrastructure; and the kinds of information, beliefs, and ideologies which shape social organization and behaviour” (1984, 1020). In other words, extractivism encompasses particular exploitative social relations combined with unequal ecological and economic exchange. It is therefore important to consider both the social relations of production (the commodification of human nature) and the predatory and exploitative relation of extractive capital to the wealth of natural resources that constitute the common heritage of Planet Earth (the commodification of non-human nature). In addition to the traditional extractive sectors (oil, gas, mining), agriculture, forestry, and fishing have been framed as potential forms of extractivism in the literature (Acosta 2013; Svampa 2013). Gudynas (2010b, 2), for example, has used the term agricultural extractivism to refer to agriculture oriented towards monoculture, the use of transgenics, machinery, and chemical herbicides, with “little or no processing and exportation of the produce as a commodity”. Gudynas suggests that this is not an “industry” and using the term industry implies value-added industrialization – not primary production for export (ibid.). For Gudynas, agricultural activity which is characterized by a high volume/intensity of extraction, semi-processed and destined for export is considered extractivism, with particular reference to soybean plantations in Latin America (Gudynas 2010a, 2010b, 2013). Giarracca and Teubal (2014, 48) suggest the term “also applies to a certain type of agriculture in which essential resources such as water and fertile land, and biodiversity, are degraded by extractivism”. Petras and Veltmeyer (2014, 64) use the term agro-extractivism in the context of the agrarian question of the 21st century,
8 McKay et al. arguing that what governments such as China and other international investors “primarily seek are lands to meet their security need for agrofood products and energy, while multinational corporations in the extractive sector of the global economy are primarily concerned to feed the lucrative biofuel market by producing oil palm, sugarcane (for ethanol) and soya” or what we might refer to as “flex crops”. Maristella Svampa includes agribusiness and biofuels production in her understanding of the new extractivism in Latin America, “due to the fact that they consolidate a model that tends to follow a monoculture, the destruction of biodiversity, a concentration of land ownership and a destructive re-configuration of vast territories” and driven by what she calls the Commodities Consensus (Svampa 2013, 118–19).3 Agrarian extractivism has therefore been introduced under the umbrella of extractivism to refer broadly to large-scale, intensive monocrop production for export. But what is the “extractivist” character of agrarian extractivism? Are all types of large-scale chemical-intensive monocrop plantations extractivist? Evidently, this type of agricultural production can take diverse forms in terms of land control and use, labour relations, surplus distribution, and the social relations of production and consumption more generally. Some large-scale plantations may require a large labour force, or be cooperatively owned by the workers, re-investing the surplus in the domestic economy, creating forward and backward linkages, generating dynamic intersectoral synergies, and producing value-added consumer goods for the domestic market. Whereas this type of large-scale industrial agriculture is also haunted by many socio-ecological issues, it is still distinct from that which is highly mechanized requiring minimal wage labour, mainly export oriented with little or no processing, corporate controlled in a monopolized market, and highly dependent on external chemical-based inputs. All capital-intensive agricultural production may not be inherently extractivist as such, which is why it is important to specify the extractivist character of the production process. In some cases, rather than being dependent on the exploitation of labour in the production process, agrarian extractivism requires increasingly less labour as it is based on various combinations of financialized, high-technology, resourceseeking extractivist capital and the appropriation of resource rents. Rather than unlimited supplies of labour (keeping wages low) being transferred to the industrial sector for a productive and social transformation, agrarian extractivism increases and stagnates what Marx defined as the relative surplus population (Marx 1867, see Alonso-Fradejas, this volume). In other words, rather than having a labour reserve which could keep wages depressed and whereby capital accumulation is largely dependent on labour exploitation, agrarian extractivism is characterized by a paradox “in which places (or their resources) are useful, but the people are not, so that dispossession is detached from any prospect of labour absorption” (Li 2009, 69). In other cases, both productive and reproductive labour are indeed needed and exploited for agro-extractivist expansion, as will be elaborated further (see Ojeda; Landívar García; Alonso-Fradejas, this volume). These extractivist features of some forms of corporate agriculture undermine the very material bases for which their reproduction – and value appropriation – depends
Introduction 9 and at an unprecedented scale (O’Connor 1998). This is not only leading to soil exhaustion, ecological destruction, rural out-migration, and surplus populations but is also resulting in catastrophic and irreversible impacts for countless species, including our own.
This collection This collection consists of ten chapters, including this introduction. The remaining nine chapters feature case studies from across Latin America, analysing the extractivist features of various agrarian systems and crop complexes. Each chapter provides a unique perspective on the concept of agrarian extractivism, contributing to its analytical and political utility. In Chapter 1, Carla Poth analyses the Biotechnological Agrarian Model (BAM) in Argentina as an expression of agrarian extractivism. She highlights the role of science and what she calls “production–expropriation–appropriation of knowledge” as the root of capital’s penetration in agriculture and where the extractivist process begins. By controlling R&D and displacing, expropriating and appropriating traditional scientific knowledge, science and technology are anything but politically neutral. Serving the interests of capital, Poth argues, the development of biotechnologies has transformed agricultural value chains and agrarian relations of production, opening up new spaces for extractivist capital to capture value. Poth traces the development and institutionalization of this model in Argentina which, through state–capital alliances (including in R&D) consolidated and legitimized a new phase of agrarian capital accumulation by dispossession. For Poth, knowledge production and politics is a key aspect of agrarian extractivism. It is therefore not only important for emancipatory struggles to take place on the land, but also – as academics, scientists, and researchers – in the lab. McKay and Colque analyse the extractivist features of Bolivia’s soy complex in Chapter 2, revealing the socio-economic and socio-ecological implications of this model. Building on the existing literature on extractivism, they put forth a framework to characterize agrarian extractivism in Bolivia based on four interlinked features: (i) significant volumes of materials extracted, primarily for exports, with little or no processing; (ii) value-chain concentration and sectoral disarticulation; (iii) high intensity of environmental degradation; and (iv) deterioration of labour opportunities and labour conditions in the area or sector. Despite the strong relations between the Morales administration and peasant/Indigenous social movements, this chapter shows the contradictions between discourse and practice in Evo Morales’ Bolivia. The authors point to the state’s dependence on extractivism, not only for extractive rents but as part of a political project through forging alliances with dominant classes of agrarian capital. As the so-called agroindustrial model is used to legitimize continued expansion by state and capitalist actors, McKay and Colque argue that the concept of agrarian extractivism is not only analytically more precise and useful but also politically as well, as it connects and draws parallels with the broader historical debates on extractivism and extractive enclaves.
10 McKay et al. In Chapter 3, Sérgio Sauer and Karla R.A. Oliveira confront hegemonic narratives in Brazil that legitimize processes of accumulation associated with the agribusiness economy. While dominant discourses frame the expansion of agricultural frontiers as necessary and unavoidable consequences of the national development process, Sauer and Oliveira argue that this advancing form of agrarian extractivism is deepening the appropriation of nature with the rapid decline and destruction of water, land, and forests, with little to no benefit for the local society or economy. The widening of this agricultural frontier is analysed as a “sacrificed zone” for the expansion of soy production and cattle ranching. This form of agrarian extractivism is advancing into traditional territories and conservation areas, such as the Matopiba region of the Brazilian Cerrado, generating and exacerbating socio-environmental conflicts in the region. The authors zoom in on the Parnaíba River Spring National Park to examine conflictual relationships between extractive capitalist development and popular resistance related to the appropriation of land and water in the Matopiba sub-region. Diana Ojeda brings a feminist political ecology perspective to the analysis of agrarian extractivism in Chapter 4. Ojeda examines the ways in which gender relations are central to the functioning of agrarian extractivism through a case study of oil palm plantations in the Colombian Caribbean sub-region of Montes de María, arguing that agrarian extractivism relies on, and deepens, gender disparities and forms of gender-based violence. The land-use change associated with agrarian extractivism, in many cases from traditional food crops to agro-commodity plantations, results in many farmers becoming net-food buyers. Ojeda reveals how such land-use change has resulted in a heavier burden for women in the sphere of social reproduction. As it is the men who are hired to work on oil palm plantations, the reproductive work of (mainly) women subsidizes the plantation and is thus fundamental for the maintenance of the agro-extractivist model. Further, the lack of recognition and payment of women’s work in social reproduction renders women and other feminized subjects increasingly dependent on wageearning men and thus exacerbates unequal gender relations within the household. While these unequal gender roles result from dynamics of agrarian extractivism, Ojeda argues that they are also constitutive of it. More than the exploitation of labour and the appropriation of nature, agrarian extractivism’s reliance on social reproduction enables the extractivist capture of value on many levels. Ojeda’s feminist perspective brings new insights and valuable contributions to the extractivist character of plantation agriculture and challenges us to think deeper into the gendered forms of extraction. In Chapter 5, León Araya explores the relationship between agrarian extractivism and sustainable development, through the case of pineapple production in Costa Rica. In so doing, he interrogates current conceptions of agrarian extractivism at the same time that he links them to broader schools of thought and societal issues. For León Araya, existing literature on (agrarian) extractivism resonates strongly with the dependency theory – in terms of centre–periphery relations, uneven terms of exchange, and the rentist patterns of periphery capitalism. He argues that this literature is constrained by the “territorial trap” of dependentista thought
Introduction 11 which tends to divide domestic and international politics and spaces, and present the national-state as the main scale of analysis. In his case study of pineapple expansion, he shows how agro-extractivist frontiers are expanded through various mechanisms, while operating together with discourses of sustainable development which provides a cover of legitimacy by moving discussions away from what is being destroyed or extracted, to what is being produced or sustained. Arguing for an expanded concept of agrarian extractivism, León Araya proposes going beyond the simple extraction of raw materials for export, and the perspective that extractivism is an activity solely controlled by transnational capital, to a broader constellation of human and non-human processes that must come together in particular ways for capital accumulation to take place. In Chapter 6, Natalia Landívar García explores the new extractivist dynamics of sugarcane production in coastal Ecuador, arguing that a state-sponsored agro-fuels project facilitates agro-extractivist expansion. Despite the integration of small-scale producer associations into the agro-fuels project, they remain subject to a corporate-controlled production model, resulting in relations of debt and dependency, unequal gender relations, and environmental destruction. For Landívar García, the inclusion of small-scale farmers coupled with discourses of energy sovereignty and sustainability represents nothing more than an illusion of fostering inclusive and sustainable rural development, while pursuing an extractivist and male-dominated form of accumulation. In Chapter 7, Alonso-Fradejas discusses the implications of the rise of the flex (oil) palm and (sugar)cane complexes in transitions to sustainability for jobs, labour regimes, and socio-ecological reproduction in Guatemala since the mid-2000s. He uses three key sets of criteria to qualify the character and assess the intensity of the “extractiveness” of a mode/form of commodity production through labour and socio-ecological reproductive lenses: (i) the examination of the “social metabolism” of resource extraction; (ii) the analysis of labour regimes and “social-productive regimes”; and (iii) the investigation of the range, ways, and extent to which the land’s ground rent, financial interest, royalties from intellectual property rights, payments for environmental services, and state subsidies are crafted, extracted, and appropriated by the owners of cane and palm companies. He argues that the operations of the flex (sugar)cane and (oil)palm complexes involve a predatory form of agrarian extractivism which is driving a process of “impairing destruction”. This works by means of a job-poor, culturally insensitive, toilsome and unpaid labour-based “productive” model, and the manufacturing of environmentally and socially toxic landscapes, to fuel a purge of the countryside that leaves nothing and no one unscathed. In Chapter 8, Darcy Tetreault, Cindy McCulligh, and Carlos Lucio challenge the notion that the concept of agrarian extractivism should be restricted to crops destined for export with little or no processing. Based on a study of agave and tequila production in Jalisco, Mexico, they argue that the domestic processing of biomass does not necessarily compensate for the negative social and environmental impacts of upstream agricultural activities – it can in fact add to them. They seek to demonstrate this to be the case for industrialized agave–tequila
12 McKay et al. production by adopting an approach that traces the flows of materials, pollutants, and money. They examine the dynamics of control over the agave–tequila value chain, the portion of tequila that is exported, technologies, company–farmer and company–worker relations, relevant public policy, and socio-environmental impacts. Through this approach, the authors demonstrate how the expansion and intensification of agave production since the mid-1990s has been characterized by increasing volumes of biomass extraction, environmental degradation, the marginalization of small-scale agave farmers, and deteriorating working conditions for agave-field workers. They point to traditional mezcal production as a socially and ecologically sustainable alternative model. In Chapter 9, Kröger and Ehrnström-Fuentes examine the case of large-scale tree monocultures for pulp production in Uruguay and assess whether and how various definitions of agrarian extractivism are suited to explain forestry dynamics. They frame forestry extractivism as a particular kind of agro-extractivism with its own unique features, mechanisms for expansion, and relations of extraction. The authors reveal how forestry projects capture large swaths of land and other natural resources with a license to pollute, using sophisticated legitimization campaigns to expand extractivist frontiers in the name of the new green and bioeconomy and sustainable development. They argue that political ecological, world-ecological, and political ontological analyses are important for understanding the particularities and types of extractivisms, calling for comprehensive and systematic in-depth studies on different forms of extractivisms across different sectors. Each of these chapters provides a valuable contribution to our understanding of the extractivist features of corporate-led, external-input, plantation agriculture. In the section that follows, we synthesize the key aspects from these contributions which we believe are crucial for a comprehensive analysis of agrarian extractivism.
Seven key aspects for analysing agrarian extractivism This collection, the first of its kind on agrarian extractivism, highlights the diversity of approaches and perspectives in analysing the extractivist features of dominant agricultural models today. Bridging critical agrarian and environmental studies, this collection charts new territory and offers analytical insights which go well beyond the initial discussions in the literature on extractivism based on raw material exports and transnational capital. Drawing from the insights in this volume, we put forth here seven key aspects for analysing agrarian extractivism: (i) sectoral and commodity particularities; (ii) flows of capital; (iii) labour dynamics; (iv) resource access and property dynamics; (v) flows of knowledge; (vi) flows of non-human nature’s energy and materials; and (vii) territorial restructuring and developmental effects. Each chapter in this collection engages with several, but not necessarily all, of these aspects. Engaging with these aspects in our analyses going forward will help us understand and reveal the extractivist features of agrarian and environmental change, and challenge the discursive legitimacy of
Introduction 13 dominant models of agricultural development. In what follows, we briefly discuss these seven aspects with reference to the contributions in this volume, offering a guiding framework for future research on the topic. First, it is important that we extend the empirical gaze of agrarian extractivism beyond its current anchoring in recent trends in Central and South America, and in particular, as pointed out by Tetreault, McCulligh, and Lucio (this volume), in the corporate-controlled production of GM flex crops, which has come to epitomize the concept in much of the critical literature to date (Giarracca and Teubal 2014; Gudynas 2015; Ezquerro-Cañete 2016; McKay 2017, 2020; AlonsoFradejas 2020). Viewing other forms of agricultural commodities and components of the agro-food system through the extractivist lens invites empirical reassessment of the scope of the concept and introduces new sets of questions about its political and analytical utility. In this volume, the concept of agrarian extractivism is applied to crops as diverse as agave (Tetreault, McCulligh, and Lucio) and pineapple (León Araya), to forestry (Kröger and Ehrnström-Fuentes), as well as to the prototypical agro-commodity flex crops such as oil palm (Ojeda; AlonsoFradejas), soybeans (McKay and Colque; Poth; Sauer and Oliveira); and sugar cane (Alonso-Fradejas; Landívar García). The particularities of these crops and trees present a range of diversity in terms of knowledge, capital and labour flows and dynamics; the social metabolism of resource extraction; levels of processing, industrialization, and broader development effects; and the relations and politics around resource access and control. Despite such diversity, this collection reveals the socio-economic and socio-ecological extractivist features which underpin these sectoral and commodity particularities. A second aspect concerns the flows of capital. While many of the crop and tree complexes analysed are controlled by market oligopolies, it is not always transnational capital, but also domestic classes of capital that extract and appropriate rents, energy, and materials from human and non-human nature. Understanding where and by whom flows of capital are being controlled and appropriated is crucial to our understanding of their broader development impacts. While it may be that most of the value is channelled to the “centres” of capital accumulation, it is too simplistic to think that these are unilateral or unidirectional processes, as pointed out by León Araya (this volume). Nearly all contributions in this volume explicitly highlight the key role of the state in fostering and facilitating agrarian extractivism, often forming alliances with both transnational and domestic capital for economic and political gain. Undertanding these flows of capital and control over value is central to our analysis of agrarian extractivism. Differentiated labour dynamics are apparent throughout the collection, and represent a third key aspect for our analyses. While in some cases labour has become surplus to the needs for capital accumulation (McKay and Colque; Kröger and Ehrnström-Fuentes; Alonso-Fradejas, this volume; see also Ezquerro-Cañete 2016), other crop complexes remain labour intensive and rely on various forms of labour exploitation. In Mexico, for instance, the “jornalerización of agave production” has fragmented the workforce along the lines of Taylorism, reducing labour costs as many of the tasks requiring the least skills but greatest physical
14 McKay et al. exertion are transferred to Indigenous migrants from southern Mexico (Tetreault, McCulligh and Lucio, this volume). Furthermore, as Alonso-Fradejas (this volume) demonstrates, the visible, “productive” labour regime stands on the shoulders of pockets of unfree and unpaid “productive” labour of women and children, and especially of the gratuitous affective and material “reproductive” labour in the household, often carried out by women, children, and the elderly. Similarly, for Ojeda (this volume), the extractivist capture of value relies on social reproduction and the unequal gender relations within the household, emphasizing the inseparability of production and social reproduction. Landívar García (this volume) also explores the gender-differentiated experiences in the agro-extractivist sector, highlighting the undervaluation of female labour, and revealing the disproportionate effect on women, whose work ultimately subsidizes the extractivist model as their time and efforts are appropriated. Resource access and property dynamics represent a fourth aspect important for our analyses. This is a key feature in all chapters in this volume and includes diverse forms of (legal and illegal) appropriation of land-based natural resources, including forests (see Kröger and Ehrnström-Fuentes, this volume) and water (see Sauer and Oliveira; León Araya; Tetreault, McCulligh and Lucio; Alonso-Fradejas, this volume); and through contract arrangements of debt and dependency which involve value capture without dispossession (McKay and Colque; Landívar García, this volume). In Brazil, Sauer and Oliveira (this volume) emphasize the combination of investments, financial speculation, public resources, collusion, and violent expropriation as the means through which expanding agro-extractivist frontiers are established. Agrarian extractivism requires a profound transformation of land use and the appropriation of nature, changing the relations of access and control over natural resources. A fifth aspect is the control and influence over the flows of knowledge. Agrarian extractivism involves not only a “metabolic rift” in material human–nature relations, but also a widening “knowledge rift” between traditional and familiar extensive farming practices and intensive farming methods and technologies protected through intellectual property rights (Alonso Fradejas; Poth; Tetreault, McCulligh, and Lucio, this volume). For Poth, extractivist relations in the agrarian sector are not only sown in the soil but begin in the spheres of science and knowledge production. Poth shows how the control over knowledge production opens up new spaces for extractivist capital to capture value. Kröger and Ehrnström-Fuentes (this volume) point to the hierarchical ontological relations between Indigenous knowledge and those of modernity, with “science” firmly situated in the latter, thereby erasing the lived realities and knowledge of those who do not conform to the modern forestry apparatus. In his analysis of the politics of environmentalism, León Araya (this volume) argues that the concept of sustainable development is propped up by scientific knowledge which depoliticizes and justifies forms of coercion and promotes technical–managerial fixes to the negative socio-environmental impacts of agrarian change. Identifying, tracing, and analysing the flows and politics of knowledge and science is fundamental to our understanding of the contemporary dynamics of agrarian extractivism.
Introduction 15 A sixth aspect concerns the flows of non-human nature’s energy and materials. This implies an analysis into whether, how, and the extent to which external nature is exhausted, or extracted, by whom and with what implications (for whom) (Alonso-Fradejas, this volume). All chapters in this volume point to the widespread ecological destruction inherent to agrarian extractivism, emphasizing different dynamics and connections within diverse perspectives. Alonso-Fradejas (this volume) and Tetreault, McCulligh, and Lucio (this volume) examine the “social metabolism” of resource extraction, pointing to the vast imbalances and contradictions in energy and material flows driven by agro-extractivist expansion. In their chapter on forestry extractivism, Kröger and Ehrnström-Fuentes (this volume) dismantle the green image of forestry corporations, revealing the links between forestry, water, and soil extraction, as it plunders nature’s bounty with significant contributions to climate change. Using a feminist political ecology perspective, Ojeda (this volume) argues that we must take seriously the role of social reproduction in our analyses of agrarian extractivism, which requires bringing nature and gender into our analysis, along with class, race, and other forms of oppression. By doing so, Ojeda shows how agrarian extractivism deteriorates the social and ecological conditions that sustain life. A similar argument is made by Alonso-Fradejas (this volume) regarding the ways in which predatory agrarian extractivism constrains or eliminates the cultural and material conditions for socio-ecological reproduction. Analysing materials and energy flows is essential to our analyses as agrarian extractivism involves the appropriation of nature at an accelerated rate, scale, and reach which is unprecedented and threatens life on Earth. A seventh aspect involves an analysis into the forms of territorial restructuring and the broader developmental effects of agrarian extractivism. Territorial restructuring “seeks control over the places and spaces where surplus is produced by shaping and controlling the institutions and social relations that govern production, extraction and accumulation” (Holt-Giménez 2008, 6). In most cases this restructuring is facilitated by the state or even part of a state project. Understanding the process of territorial restructuring requires an engagement with the six aforementioned aspects, while also delving into the role and nature of the state and state–capital–society relations, which inevitably goes beyond state discourse. Most of the contributors to this volume, for example, have challenged the discursive framing used by various governments to legitimize corporate-controlled, external-input, plantation agriculture. Official discourses have portrayed this model as a pillar of economic growth and recovery (Sauer and Oliveira), promoting climate stewardship and sustainable development (León Araya; AlonsoFradejas, Kröger and Ehrnström-Fuentes, this volume), social inclusion (Landívar García, this volume), and contributing towards the achievement of food sovereignty (McKay and Colque, this volume). In Argentina, Poth (this volume) shows how the state has been a key player in supporting and promoting biotechnology “to feed the world”. She suggests that the state’s role in imposing private property, legitimizing or naturalizing forms of dispossession, and expanding commodification is fundamental for extractive capitalism and indeed facilitates a process of
16 McKay et al. territorial restructuring. Ojeda (this volume) points to the ways in which oil palm plantations in Colombia were forged through the articulation of paramilitary, state, and capitalist dispossession, sustained and expanded by a systematic attack on social reproduction. Territorial restructuring is integral to agrarian extractivism, generating extractivist enclaves and sectoral disarticulation in some cases (McKay and Colque, this volume) or a process of “impairing destruction” which unleashes a social and ecological purge of the countryside which adversely affects friends and foes alike, regardless of species, social class, gender, ethnicity, or livelihood (Alonso-Fradejas, this volume). Understanding these broader development impacts and the role of the state in these processes is vital for an analysis of agrarian extractivism.
Conclusion This book examines the nature and character of resource extractivism, as well as the ways and extent to which capital has penetrated and transformed agriculture and rural livelihoods across the Latin American region. We point to an emerging literature which refers to these dynamics as agrarian extractivism – a concept which offers both analytical and political utility in the debates regarding agrarian, climate, and environmental change, as well as development and sustainability more generally. Agrarian extractivism brings the predatory extractivist features of corporate-led, external-input plantation agriculture to the fore. The concept directly challenges the notion that this model is productive and efficient, and it triggers industrialization in the countryside. By commodifying nearly all aspects of the value chain, and appropriating others for free, capital extracts from nature and labour as industrial inputs circulate through the air, soil, and water, contaminating the ecological material base and (hyper-)exploiting, exhausting, or outright displacing labour. This type of agricultural model parallels the dynamics of other resource extractivist sectors such as mining and hydrocarbon extraction and should be conceptualized as such. More than just removing or extracting natural resources from the ecosystem, extractivism involves a broad complex of social relations and flows of knowledge, ideas, energy, and materials behind the ever-growing expansion of commodity frontiers. It includes the operations of resource-seeking capital, the modality of accumulation, and the social relations of production (i.e. extraction). The diversity of perspectives in this volume offer a comprehensive analysis of agrarian extractivism in recent decades which we hope strengthens its analytical and political utility and provokes others to engage with this concept in the debates and studies concerning agrarian, climate, and environmental change.
Notes 1 We make a subtle, yet important distinction between “extractive” and “extractivist”. While all forms of agriculture/commodity production involve resource extraction, some do so to a larger extent than others – in terms of scale, pace, and reach. We argue that it
Introduction 17 is analytically useful to differentiate between (necessarily) extractive and (profit-driven, highly) extractivist forms of agricultural production. 2 This collection lacks a contribution on fisheries extractivism in Latin America. See Barbesgaard (2019) and Ertör and Ortega‐Cerdà (2019) for a similar perspective in the case of fisheries in Myanmar and Turkey, respectively. It also lacks a contribution on intensive livestock production (Weis 2013). 3 For Svampa, the “Commodities Consensus” refers to “the beginning of a new economic and political order sustained by the boom in international prices for raw materials and consumer goods, which are increasingly demanded by industrialised and emerging countries” (Svampa 2013, 117).
References Acosta, A. 2013. Extractivism and neoextractism: two sides of the same curse. In M. Lang and D. Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Quito and Amsterdam: Transnational Institute & Rosa Luxemburg Foundation, 61–86. Alonso-Fradejas, A. 2020. “Leaving no one unscathed” in sustainability transitions: the life purging agro-extractivism of corporate renewables. Journal of Rural Studies, 81, 127–38. Bakker, L., W. van der Werf, P. Tittonell, K.A.G. Wyckhuys, and F.J.J.A. Bianchi. 2020. Neonicotinoids in global agriculture: evidence for a new pesticide treadmill? Ecology and Society, 25(3), 26. Barbesgaard, M. 2019. Ocean and land control-grabbing: the political economy of landscape transformation in Northern Tanintharyi, Myanmar. Journal of Rural Studies Studies, 69, 195–203. Bernstein, H. 1979. African peasantries: a theoretical framework. The Journal of Peasant Studies, 6(4), 421–43. Bernstein, H. 2014. Food sovereignty via the ‘peasant way’: a sceptical view. The Journal of Peasant Studies, 6150(May), 1–33. BFED & IPES-Food. 2020. Money Flows: What Is Holding Back Investment in Agroecological Research for Africa? Brussels: Biovision Foundation for Ecological Development & International Panel of Experts on Sustainable Food Systems. Boltvinik, J. 2016. Poverty and persistence of the peasantry: background paper. In J. Boltvinik and S.A. Mann (eds.), Peasant Poverty and Persistence in the 21st Century: Theories, Debates, Realities and Policies. London: Zed Books, 45–91. Borras, S.M.J., C. Kay, S. Gomez, and J. Wilkinson. 2012. Land grabbing and global capitalist accumulation: key features in Latin America. Canadian Journal of Development Studies/Revue canadienne d ’ études du développement, 33(4), 402–16. 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(5), 1017. Clapp, J. 2020. Spoiled milk, rotten vegetables and a very broken food system. New York Times, 8 May. Clapp, J., and J. Purugganan. 2020. Contextualizing corporate control in the agrifood and extractive sectors. Globalizations, 1–11. Dauvergne, P. 2018. The global politics of the business of “sustainable” palm oil. Global Environmental Politics, 18(2), 34–52. Deininger, K., and D. Byerlee. 2011. Rising Global Interest in Farmland: Can it Yield Sustainable and Equitable Benefits? Washington, DC: World Bank.
18 McKay et al. Ertör, I., and M. Ortega-Cerdà. 2019. The expansion of intensive marine aquaculture in Turkey: the next-to-last commodity frontier? Journal of Agrarian Change, 19(2), 337–60. ETC Group. 2015. Breaking Bad: Big Ag Mega-Mergers in Play Dow + DuPont in the Pocket? Next: Demonsanto? Ottawa: ETC Group. Ezquerro-Cañete, A. 2016. Poisoned, dispossessed and excluded: a critique of the neoliberal soy regime in Paraguay. Journal of Agrarian Change, 16(4), 702–10. FAO. 2018. Ending Extreme Poverty in Rural Areas. The World Bank. Rome: Food and Agriculture Organization. FAO. 2019. The State of the World’s Biodiversity for Food and Agriculture. Rome: Food and Agriculture Organization. FAO. 2020. The State of the World’s Forests. Rome: Food and Agriculture Organization. Giarracca, N., and M. Teubal. 2014. Argentina: extractivist dynamics of soy production and open-pit mining. In H. Veltmeyer and J. Petras (eds.), The New Extractivism: A PostNeoliberal Development Model or Imperialism of the Twenty-First Century? London: Zed Books. Goodman, D., B. Sorj, and J. Wilkinson. 1987. From Farming to Biotechnology: A Theory of Agro-Industrial Development. Oxford, UK: Basil Blackwell. Gras, C.S. et al. 2009. La Argentina rural: de la agricultura familiar a los agronegocios. Buenos Aires: Biblos. Gudynas, E. 2010a. The new extractivism of the 21st century: ten urgent theses about extractivism in relation to current South American progressivism. Americas Program Report, 21, 1–14 January. Gudynas, E. 2010b. Agropecuaria y nuevo extractivismo bajo los gobiernos progresistas de América del Sur. Territorios, 5, 37–54. Gudynas, E. 2013. Extracciones, extractivismos y extrahecciones. Observatorio del Desarollo, 18, 1–18. Gudynas, E. 2015. Extractivismos: Ecología, economía y política de un modo de entender el desarollo y la Naturaleza. Cochabamba: CEDIB. Holt-Giménez, E. 2008. Territorial Restructuring and the Grounding of Agrarian Reform: Indigenous Communities, Gold Mining and the World Bank. Amsterdam: Transnational Institute. IAASTD. 2009. Agriculture at a Crossroads. Washington, DC: Island Press. IPCC. 2019. Climate Change and Land. Geneva: Intergovernmental Panel on Climate Change. Kay, C. 2016. La transformación neoliberal del mundo rural: procesos de concentración de la tierra y del capital y la intensificación de la precariedad del trabajo. Revista Latinoamericana de Estudios Rurales, 1(1). Kloppenburg, J.R. 2004. First the Seed: The Political Economy of Plant Biotechnology. Second edition. Madison, WI: The University of Wisconsin Press. Leff, E. 2006. La Ecología Política en América Latina: Un campo en construcción. In H. Alimonda (ed.), Los tormentos de la materia: Aportes para una ecología política latinoamericana. Buenos Aires: CLACSO, 21–39. Li, T.M. 2009. To make live or let die? Rural dispossession and the protection of surplus populations. Antipode, 41(S1), 66–93. Margulis, M.E., N. McKeon, and S.M.J. Borras. 2013. Land grabbing and global governance: critical perspectives. Globalizations, 10(1), 1–23. Marx, K. 1867. Capital: A Critique of Political Economy. Volume I. Moscow: Progress Publishers. McKay, B.M. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211.
Introduction 19 McKay, B.M. 2018. Control grabbing and value-chain agriculture: BRICS, MICs and Bolivia’s soy complex. Globalizations, 15(1), 74–91. McKay, B.M. 2020. The Political Economy of Agrarian Extractivism: Lessons from Bolivia. Blackpoint and Winnipeg: Fernwood Publishing. McMichael, P. 2009. Feeding the world: agriculture, development, and ecology. In L. Panitch and C. Leys (eds.), Coming to Terms with Nature: Socialist Register. London: Merlin Press, 170–94. Medeiros, L.S.D. 2015. Luta por reforma agrária no Brasil contemporâneo: entre continuidades e novas questões. In C. Grisa and S. Schneider (organizadores), Políticas públicas de desenvolvimento rural no Brasil. Porto Alegre: Editora da UFRGS, 339–60. Miller, H.I., and G. Conko. 2001. Precaution without principle. Nature Biotechnology, 19(4), 302–3. Minot, N., and L. Ronchi. 2014. Contract Farming. Washington, DC: Viewpoint. Murphy, S., D. Burch, and J. Clapp. 2012. Cereal Secrets: The World’s Largest Commodity Traders and Global Trends in Agriculture. Oxfam Research Report. Nehring, R. 2016. Yield of dreams: marching west and the politics of scientific knowledge in the Brazilian Agricultural Research Corporation (Embrapa). Geoforum, 77, 206–17. Nicholls, C.I., and M.A. Altieri. 1997. Conventional agricultural development models and the persistence of the pesticide treadmill in Latin America. International Journal of Sustainable Development and World Ecology, 4, 93–111. O’Connor, J. 1998. Natural Causes: Essays in Ecological Marxism. New York: The Guilford Press. Petras, J., and H. Veltmeyer. 2014. Extractive Imperialism in the Americas: Capitalism’s New Frontier. Leiden: Koninklijke Brill. Rubio Vega, B. (ed.). 2018. América Latina en la mirada: las transformaciones rurales en la transición capitalista. Mexico City: Universidad Nacional Autónoma de México. Sandwell, K. 2019. Growing Power: Mega-Mergers and the Fight for Our Food System. Amsterdam. Available at: www.tni.org/files/publication-downloads/web_comic_book_ small.pdf. Sauer, S. 2019. Rural Brazil during the Lula administrations: agreements with agribusiness and disputes in agrarian policies. Latin American Perspectives, 46(227), 103–21. Soto Baquero, F., and Gómez, S. 2012. Dinámicas del mercado de la tierra en América Latina y el Caribe: concentración y extranjerización. Rome: FAO. Svampa, M. 2013. Resource extractivism and alternatives: Latin American perspectives on development. In M. Lang and D. Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Amsterdam and Quito: Transnational Institute and Rosa Luxemburg Foundation. Taylor, M. 2018. Climate-smart agriculture: what is it good for? Journal of Peasant Studies, 45(1), 89–107. UN. 2020. Impact of COVID-19 on Food Security and Nutrition (FSN). New York: United Nations. van der Ploeg, J.D. 2018. Differentiation: old controversies, new insights. Journal of Peasant Studies, 45(3), 489–524. Weis, T. 2010. The accelerating biophysical contradictions of industrial capitalist agriculture. Journal of Agrarian Change, 10(3), 315–41. Weis, T. 2013. The Ecological Hoofprint: The Global Burden of Industrial Livestock. London and New York: Zed Books. Wesz Jr, V.J. 2016. Strategies and hybrid dynamics of soy transnational companies in the Southern cone. The Journal of Peasant Studies, 43(2), 286–312.
20 McKay et al. Woodhouse, P. 2010. Beyond industrial agriculture? Some questions about farm size, productivity and sustainability. Journal of Agrarian Change, 10(3), 437–53. World Bank. 2007. World Development Report 2008: Agriculture for Development. World Development Report 2008. Washington, DC: World Bank. World Bank. 2015. Future of Food: Shaping a Climate-Smart Global Food System. Washington, DC: World Bank.
1
The Biotechnological Agrarian Model in Argentina Fighting against capital within science Carla Poth
Introduction Dominant discourses often claim that biotechnology has triggered an “agricultural revolution”. Mass media, agribusiness, farmers, and politicians highlight the incredible changes generated by the introduction of genetically modified (GM) seeds, and how necessary they are to feed the world’s growing population and to combat hunger and malnutrition. It’s enough to read articles like “Sólo la biotecnología salvará al mundo” (Only biotechnology will save the world) published in Argentina’s largest newspaper, Clarin (29/01/2001), or to listen to the former Minister of Agriculture Luis Etchevere saying “Tenemos la firme posición de defender la biotecnología para alimentar al mundo” (We believe firmly in defending biotechnology in order to feed the world) (MAGyP 2019) to understand how this discourse is presented by the mainstream. But, what does this revolution mean? There is no doubt that biotechnologies have completely transformed agricultural value chains. They have changed the ways of farming, the use of the land, soil and water, and require external synthetic inputs for production. They have transformed the classical logics of property and the roles of agrarian actors. This transformation is characterized by a series of biophysical overrides and technical fixes which extend and deepen processes of commodification, enabling capital to extract value from farmers, society, the state, and nature. As capital continues to transform agrarian life, these forms of extraction form part of the underlying capitalist relations and forms of production. But the extractivist relations in the agrarian sector not only materialize upon exporting large quantities of raw materials but also in the spheres of science and knowledge production. Argentina is an ideal site to analyse how this process unfolds, being the first country to introduce GM seeds and biotechnologies in South America. GM soy production (Roundup Ready variety) increased from 50,000 hectares in 1996 (when it was first introduced) to 1.7 million hectares in only one year. Today, Argentina has legalized 61 GM seed varieties (of soy, cotton, potato, safflower, alfalfa, and maize) and produces more than 24 million hectares of transgenic crops. These crops are tolerant to climate stress, herbicides, and pesticides (like
22 Carla Poth glyphosate, glufosinate ammonium, or 2,4-dichlorophenoxyacetic acid), and are resistant to insects and viruses.1 This model has proven to be effective in creating profits for oligopolistic capital, but has failed miserably to combat hunger and sustain a healthy environment. In fact, globally, industrial agriculture only produces some 30% of food, while using 70% of the land and water and deforesting more than 13 million hectares per year, resulting in a significant contribution to climate change (ETC Group 2017). This model has reduced agricultural employment and dispossessed farmers, triggering rural out-migration, while substituting diverse crop varieties for agro-commodity flex crops and cattle ranching. Finally, it has provoked environmental contamination and the proliferation of illnesses associated with the use of toxic agro-chemicals. In Argentina alone, the use of herbicides has increased from 50 million kilograms (kg) per year, in 1996, to 525 million kg, in 2018. These impacts have not gone uncontested as conflicts and collective movements have multiplied. Populations in towns, including teachers and researchers, doctors and other professionals are denouncing this model, while peasant movements are trying to stop its expansion, and Indigenous people and consumers are questioning the intellectual property of seeds. Finally, scientists like Rafael Lajmanivich, Damian Verzeñazi, Damian Marino, and Medardo Ávila Vazquez are criticizing the role of science and policies that are legitimizing this system. In this chapter, I present the concept Biotechnological Agrarian Model (BAM) as a way in which agrarian extractivism expands in Argentina. The BAM involves a process of change and restructuring in which biotechnologies, including GM seeds, shape agrarian relations of production. The importance of these technologies in the dynamics of agrarian change requires a deeper understanding into the laboratories and universities in which they are produced, and of the associated constant and fluid process of expropriation and reappropriation of knowledge. The next (second) section characterizes the BAM as an expression of extractive capitalism. It discusses extractivism as the way in which capital accumulation manifests in particular territories, focusing on the role of science and problematizing why extractivism is not possible without what I refer to as the “production– expropriation–appropriation of knowledge as exchange value”. I also explain how GM seeds are born from production–expropriation–appropriation of knowledge as exchange value, and why the BAM expresses the violence of accumulation by dispossession that shapes extractive capitalism. In the third section, I show how the protagonists of this model in Argentina (who have strengthened the integration of profit-driven science and agriculture) developed the institutional and regulatory frameworks to consolidate and expand this new phase of capital accumulation in agriculture. Here, again, I pay special attention to the role played by science in the state’s institutional and regulatory frameworks. I divide this part of the analysis into four different periods. Each one reflects different patterns of the model’s hegemony and reveals how capital must continuously recreate new ways of generating profits, transforming and expanding in agrarian relations, and imposing new dynamics of exploitation based on violence and the construction of legitimacy. The fourth section concludes the chapter with a call to researchers and
Biotechnological Agrarian Model in Argentina 23 scholar-activists to continue the struggle against forms of agrarian extractivism in the spheres of research and academia.
The Biotechnological Agrarian Model: a phase of extractive capitalism To understand the BAM we need to analyse the ways in which capital manifests in the countryside. Capital is always looking for the material conditions for its creation. Its existence depends on its capacity to expand and the possibility to build the conditions to absorb a surplus. To do this, it needs to control – temporarily and spatially – production and labour. But the same conditions that facilitate capital’s expansion and accumulation – labour and nature – can also lead to its destruction and crisis. That is why when we see capital’s manifestations in the territory we always see its intrinsic crisis. Crisis is the unavoidable and necessary way for continued capital accumulation. Capitalism, as the predominant mode of production shaping social relations, tries to overcome its accumulation crises by postponing in time and expanding in space its forms of realization. The so-called “spatiotemporal fix” becomes a metaphor for capital’s solutions to crises. As Harvey (2004) proposes, capital requires temporary displacements (capital investment in long-term projects or social expenses, such as education and research) and spatial displacements, through the opening of new markets, commodification, productive capacities, and new possibilities of resources and labour in new places. These displacements reveal that capital’s reproduction is possible due to processes of “accumulation by dispossession” (Harvey 2004) which requires the continuous expropriation and appropriation of territories, resources, and labour. As Luxemburg (1951, 368) puts it: Capitalism needs non-capitalist social strata as a market for its surplus value, as a source of supply for its means of production and as a reservoir of labor power for its wage system (. . .) Since the primitive associations of the natives are the strongest protection for their social organizations and for their material bases of existence, capital must begin by planning for the systematic destruction and annihilation of all the non-capitalist social units which obstruct its development. In Latin America, where the relative costs of expropriating territories and natural resources is low, extractivism is functional to capital since it reproduces the dynamics of dispossession required for capital accumulation (Seoane 2012). Extractivism expands with mechanism of accumulation by dispossession that “gain(s) immediate possession of important sources of productive forces such as land, game in primeval forests, minerals, precious stones and ores, products of exotic flora such as rubber, etc.”; it “liberate(s) labour power and coerce(s) it into service”; and it “introduce(s) a commodity economy” and “separate(s) trade and agriculture” (Luxemburg 1951, 369). Extractive capital attempts to enter non-commercial relations with the environment and human beings and eliminates
24 Carla Poth forms of resistance by turning them into capitalist mercantile society. Accumulation by dispossession imposes a new way to link social relations where exchange value, represented by money, is the dominant social nexus. For this reason, it is argued here that violence is at the heart of extractivism. The consolidation of extractivism involves the destruction of the material, cultural and environmental structure of socio-territorial networks, appropriating the environment, labour, skills, and knowledge, with similar mechanisms of theft and predation described by Karl Marx in Volume 1 of Capital. But, is contemporary extractive capital the same as the forms of predation described by Marx in the origins of capitalism? Is it even the same as the mechanisms of Latin America’s colonization? We could say that extractivism is a condition for the daily reproduction of capital due to the way in which accumulation by dispossession manifests in territories rich in natural resources. But we also have to say that, nowadays, the past mechanisms of theft and predation have evolved and expanded in modern societies in two new ways that are fundamental to understand. First, present forms of extractivism often entail accumulation by dispossession on a scale infinitely higher than the original accumulation process due in part to scientific and technological innovations. Scale, in this context, does not refer to a quantitative number, but to the complete resignification of a “commodity” and the antagonistic relations that consolidate this commodification and reproduction of capital. Second, although capital has consolidated the production of knowledge and technologies as a constitutive element of the dynamics of exploitation, the subsumption of nature, labour, and knowledge under capital has surpassed the limits of the imagination (Gilly and Roux 2009; Roux 2008). For this reason, it is fundamental to understand present ways of knowledge production through a permanent process of production–expropriation–appropriation of knowledge. This process helps us understand how the production of knowledge becomes contentious, subject to the antagonistic class forces in capitalist society. Knowledge, as an objective condition of labour, has to be separated from the same labour that generates or produces it and becomes a territory in dispute (see Giraldo and Rosset 2018). As knowledge is produced and expropriated by dominant class forces it becomes commodified and converted from “the pigmy property of the many into the huge property of the few” (Marx 1887, 541). This process involves a constant resignification and co-optation of knowledge: its reduction to a “simple way of calculation and technical control” where “modernization has repressed the variety, variability, and the indeterminacy of the world to adjust it to the demands of production” (Rullani 2004, 100). To understand the extractive features of capital it is important to analyse not only the specific function of this knowledge and the technologies it produces, but to recognize the concrete paths that shape the production of knowledge, the dynamics used in knowledge expropriation and dispossession, and the characteristics of a reappropriation of this knowledge (Poth 2019). Within extractivism, science crystallizes the production–expropriation–appropriation of knowledge as exchange value with different mechanisms to subsume labour to capital, and to subsume itself into the creation of value, deepening dispossession and
Biotechnological Agrarian Model in Argentina 25 consolidating the idea of knowledge as a commodity, ultimately expanding the reproduction of capital. This production of knowledge coalesces in universities and laboratories dependent on private sector funding. In effect, information produced is private and researchers are limited by these corporations that impose confidentiality agreements (Lander 2006). As Levins and Lewontin (1985) express, this process promotes individualized, meritocratic, pragmatic, and reductionist practices within labs, and consolidates the strategy of academic secrecy to retain the information. It supports scientific competition, aligning research with the interests of companies that invest in labs. In this way, capital controls what, for whom, and how researchers investigate. Capital needs to hoard knowledge, dispossess other culture’s knowledge and biopirate genetic resources (Augsten 2005; Roux 2008). Another way to analyse dispossession within research spaces is the institutionalization and movement of public knowledge into private spheres. In this way, if years ago the production of public knowledge allowed disputes about dynamics, objectives, and receivers of research, nowadays disputes are eliminated with the consolidation of public science that has to produce knowledge in association with private corporations for markets, often through public–private partnerships. As never before, the Schumpeterian paradigm2 of innovation crystallizes within the scientific system, connecting economic growth with access to innovations. This paradigm can be seen in the creation of associations between companies and research centres financed by states, technological transference, and the creation of goods with high added value (Langer 2011). Finally, legal mechanisms to appropriate knowledge, such as intellectual property rights, have consolidated forms of accumulation at the global level. To create value, a commodity must be able to circulate and spread, but under legal frameworks which guarantee and facilitate the ability to realize profits. As Rullani says, “the exchange value of knowledge is completely linked to the practical capacity to limit its free diffusion” (2004, 102). Thus, another condition required to guarantee extractive capitalism is the consolidation of the state and its legal frameworks to legitimize and naturalize this dispossession, imposing private property in new spaces, deepening and expanding commodification. As such, the systemic violence of extractive capital becomes institutionalized into legal frameworks, criminalizing those who claim the preexistent right of communitarian property (Giaretto and Poth 2015). Extractive capital is the heart of capital accumulation in Latin America. It is not a mere technique of extraction nor a simple mechanism for transnational corporations to extract natural resources. Extractivism is a territorial expression of the accumulation of capital as a whole. It shows the reconfiguration of capital’s reproduction in time and space, represents new logics of production, and shapes new forms of resistance against this process. How does this extractivism relate with agrarian production? Agrarian extractivism has been the way in which capital has consolidated in Latin American territories. Capital, within the agrarian chain, has deepened the
26 Carla Poth search for a differential income, concentrating control over every link of the agrarian chain by corporations and expanding the scale of production, including into new territories and consolidating new commodities (Giarraca y Teubal 2013; Palmisano 2016). Agrarian extractivism has also configured the penetration of capital with new strategies to control resources (land, seeds, biodiversity, etc.) and contradictory state dynamics to consolidate the accumulation of capital and support the political legitimacy of this model (McKay 2018). Those changes impose not only economic and territorial concentration and displacements, but also subsequent violations to human rights (health, food, water, and others) and the creation of selective democracies (Aranda 2015a) where participation is excluded. This form of agrarian extractivism also represents the constitution of an agrarian chain where biotechnologies and their main innovation –GM seeds – crystallize the antagonistic relations and the logics of capital accumulation. This is what is referred to as the BAM (López Monja et al. 2009). The BAM represents a bridge between the closed spaces of the laboratory and the large areas of agriculture. With biotechnologies, capital may control the process of life from its beginning, defining the conditions of existence of each modified organism. In this way “controlling the process of reproduction of life is simultaneously an objective and a mechanism to build power” (Ceceña 2004, 24). Biotechnologies manipulate, appropriate, and control nature, and include commodification mechanisms within the seeds, creating varieties that are in a symbiotic relation with the time and rhythms of capital. Therefore, seeds are turned into commodities from the beginning. GM seeds are an expression of the expansion of capital into the reproduction of life, defining with market logics the conditions of existence of these new biological lives. In the example of GM seeds, the market logic of capital embedded into living beings becomes apparent. As capital logics penetrate the seed, they then become inserted into capitalist markets. GM seeds enable agribusinesses to build what Lander (2006) calls the “total market utopia” where capitalist markets expand to other regions and spaces, incorporating commodities that were not possible to be appropriated or commercialized before, such as nature and knowledge. The widespread use of GM seeds has enabled concentrated capitals to create a “biotechnological closed package”, composed of modified seeds and chemical inputs that organize production and investment, and shape productive relations and logics of innovation (Gras and Hernandez 2013; Postone 2002). Clearly, GM seeds have bridged the gap between labs and fields. Agrarian capital is not only generated within the process of the agrarian value chain, but within the scientific system, where corporations invest to produce new biotechnologies. As Kloppenburg says, “agricultural research can also be seen as an important means of eliminating the barriers to the penetration of agriculture by capital” (2004, 10). Labs and universities became an organic element for capital accumulation within agrarian production, generating and centralizing biotechnological knowledge for corporations (Goldstein 1989; Lander 2006). This accumulation of knowledge is preceded by the expropriation and appropriation of localized traditional agrarian
Biotechnological Agrarian Model in Argentina 27 knowledge and genetic resources that are commodified to produce more profits. This relation leads directly to the appropriation of genetic materials and knowledge by companies. They rely on the knowledge of Indigenous and peasant communities in order to conduct experiments in laboratories and conclude that it is an “invention” (López Monja et al. 2009, 94). The creation of legal mechanisms to appropriate knowledge and natural resources deepens those concentrating tendencies of accumulation. For instance, seed laws and intellectual property rights are being homogenized all over Latin America, increasing the time of patentability, including the appropriation of living organisms (like plants or seeds) and, finally, establishing a “double protection”3 (Perelmuter 2013). Biotechnologies have enhanced the scale of accumulation by dispossession in agrarian territories. They have reconfigured the economic structure as a whole, consolidating an agrarian chain where labs and universities are fundamental places for accumulation strategies, creating new markets and market oligopolies. Those technologies (and GM seeds) also crystallize the production–expropriation– appropriation of knowledge as exchange value. They are created with a scientific view that naturalizes the production of private knowledge in public universities (e.g. Monsanto financing a Master’s degree in the Faculty of Agronomy at the University of Buenos Aires), or the private appropriation of research in public labs or funds, such as the Rosario Agrobiotechnology Institute (INDEAR).4 In this context, researchers’ work is oriented by corporate interests; legitimating, for example, the private intellectual property of innovations that were consolidated with public resources. The production–expropriation–appropriation of knowledge as exchange value is underscored by researchers who not only work in biotechnologies but also have partnered with corporations or are part of state regulatory institutions (Shurman and Munro 2010), consolidating a closed idea where biotechnologies and GM seeds are the only way to produce. Under this framework, the state legalizes and legitimizes accumulation by dispossession of knowledge and natural resources (seeds, water, and land). However, extractive capital cannot only be seen in the income links of the agrarian chain. The BAM also shows explicit and subtle violence of accumulation by dispossession in other phases of production (Lapegna 2019). On one hand, the BAM explicitly accumulates capital by displacing peasants and other rural dwellers, destroying the environment through widespread deforestation and biodiversity loss, and even murdering social activists (like Cristian Ferreyra). On the other hand, it has more subtle mechanisms of dispossession, generating relations of debt and dependency, destroying regional economies, and denying environmental problems (Pengue 2017). Finally, the BAM is creating a massive epidemiological problem and has been linked with the spread of zoonotic diseases such as COVID-19 that ends with a “silent genocide” (Rossi 2020; Wallace 2020). The BAM represents a form of agrarian extractivism. But when I say this I do not mean that the BAM is the extensive use of natural resources with low valorization after extraction for exportation by transnational corporations, as Eduardo
28 Carla Poth Gudynas (2011) describes. The BAM expresses the extractivism in agrarian regions because it is a way of capital accumulation and, as demonstrated in this chapter, it expresses the origins of capital reproduction in Latin America. It is impossible to understand the accumulation of capital in this region without looking at this specific way of extracting value, and its associated crises and antagonisms. In Argentina, the BAM started to gain momentum during the 1990s, when the state developed a regulatory framework releasing GM seeds and expanding the use of agro-chemicals. Those regulatory frames expressed the integration between the agrarian structure and the production of knowledge, privileging the commercial interests of GM seeds with the pre-eminence of the substantial equivalence principle. In the next section, I trace the consolidation of this model in Argentina throughout four key stages.
The BAM in Argentina: four stages of consolidation During the 1990s, Argentina initiated a process that transformed its agrarian structure, betting on biotechnological production to penetrate global markets in a strategic position. The consolidation of the agro-export model implied an improvement of productivity and competitiveness, based on the production of cheap goods. Since then, every subsequent administration has created the conditions to expand and consolidate the BAM, but they did not do so in the same way. In this section, I will show how regulatory frameworks were adapted to different conflicts that took part in global and local geographical levels over four temporal stages. The first part (1991–1997) shows the initial steps of the regulatory frameworks and the creation of the National Commission of Agrarian Biotechnology (CONABIA). In the second stage (1998–2003) I highlight the multiple conflicts emerging within the BAM and different attempts to solve them. The third period (2004–2008) is characterized by a moment of institutional restructuring and diversification, where the state responded in an integral way to certain challenges posed by diverse actors. In every period, I focus on the role that scientific knowledge played in guaranteeing the commercial interests of transnational corporations and consolidating the political domination of the BAM. Finally, the fourth part (2008–2019) discusses the emergence of public debates around agrochemicals. From there, regulatory frameworks and institutions of GM seeds had to face new types of obstacles and conflicts. The first period (1991–1997): the consensus of CONABIA In 1991, CONABIA was created within the Secretary of Agriculture to design all the regulatory frameworks related to GM seeds (res.124/91). Although it is considered an expert and technocratic commission rather than a political one, the commission has more power and influence in political decisions about genetically modified organisms (GMOs) than any other state institution. It evaluates risk assessments, establishes technical and biosafety standards, and acts as a consultant for other government bodies. It centralizes all the information related to
Biotechnological Agrarian Model in Argentina 29 GMOs, and governs its own operating mechanisms. This commission is so important that there is no political authority that challenged its decision-making.5 We may identify two reasons to start creating these regulatory frameworks related to GMOs. First, since the late 1980s there were a lot of countries like the United States, Canada, and Mexico (and even some countries in Europe) that started to discuss different regulatory frameworks for GMOs. These debates showed that the commercialization of GMOs would be an important topic in the subsequent decades. Second, there were local pressures from seed corporations that tried to conduct experiments with different GM seeds and vaccines. Although these companies approached different state institutions like the National Food Safety and Quality Service (SENASA)6 and the National Seed Institute (INASE),7 none of them had the expertise to control the production or commercialization of those organisms. Seed companies, like Nidera and Syngenta, created an open dialogue about GMOs and their impacts, and partnered with agrarian nongovernmental organizations (NGOs) such as the Argentine Association of Regional Consortiums for Agricultural Experimentation (AACREA),8 the Argentine Association of Direct Sowing Producers (AAPRESID), and the Argentinian Biotechnological Forum,9 to push forth an agenda for GMOs. CONABIA is composed of experts with a scientific profile. They were selected from three main areas (res. 124/91): the scientific system (universities, research organisms, and labs), the public and private sector (especially representatives of seed chambers and scientific NGOs that promote biotechnologies). For CONABIA’s creators the experts’ knowledge would guarantee impartiality (because it avoided conflicts of interests), as well as a broad spectrum of knowledge that could create consensus about biotechnologies (Poth 2019).10 But what kind of knowledge and training did these experts need to have in order to be included? Why did CONABIA think that this particular knowledge could form a consensus? To understand the expertise of CONABIA’s members and to identify where they derived their expertise from, I present later the “educational, academic and work trajectories” of ten members of the commission. These members were crystallized in a sociogram (see Figure 1.1) depicting the different sectors from where they were recruited (research, public sector, private sector). These trajectories show the main places of education, academic experience, and work from where they derived their expertise that was valued for their admission in CONABIA. The objective of this sociogram is to identify which places were shared by these members, and build networks where “information and ideas move easily through the specialty, giving a sense of community” (Granovetter 1973, 12). If we observe the educational trajectories at the bottom of the sociogram we can see that most of the members shared graduate training in public universities (in different regions of the country), attending similar disciplines: Agronomic Engineering (with a technical productive training) and Biological Sciences. Postgraduate education appears to be homogeneous, with technical training in universities and labs in foreign countries. These trajectories consolidate applied research related to Genetic Engineering and specializations in the fields of transgenics and
30 Carla Poth
Figure 1.1 Sociogram: educational, academic and work trajectories Source: Author’s own elaboration
Biotechnological Agrarian Model in Argentina 31 agro-chemicals. The academic trajectories of those who come from the public sector and research are affiliated with public institutions with multiple and diverse funding sources (public or private ones). Only in the private sector are some differences observed, because they do not necessarily have academic experience. With their work trajectories (at the top of the sociogram), we see that most of the members worked as researchers and, at least once during their career, they have carried out research within a public–private agreement. Most of them have worked in academic management. Finally, Figure 1.1 depicts only one individual case of research/funding activity within the private sector, which was with the regulatory agency of a transnational seed company; and that those belonging to the public sector (public scientific system) show a greater adaptation to different types of work (advising the private sector, research), work areas (global or national), and sectors (private or public). When analysing these trajectories, we recognize that, although members of CONABIA were called from different sectors, most have worked in similar areas and share a similar approach. This helps explain why these actors could build shared perspectives and practices. Putting this sociogram together with a diachronic view of their trajectories, we can identify that the knowledge that was compiled by these experts, the ways in which they define and approach biotechnologies, and the role knowledge plays is similar. Members of CONABIA say that knowledge has to have a “market application”. This knowledge is related with financial inputs from companies that want to create, in association with universities or institutions such as CONICET (National Scientific and Technical Research Council),11 public–private partnerships where knowledge production is intimately linked to the profitability that knowledge can generate in the market. In this way, public resources are used to generate knowledge that then becomes commodified and controlled by the private sector. In the context of GMOs, the experts of CONABIA also share the idea that GM seeds do not necessarily have negative social, environmental, or sanitary consequences. They argue that genetic engineering can reduce any differences between GM seeds and conventional ones, selecting desirable attributes and eliminating any risk. All of these shared perspectives are crystallized within the regulatory frameworks governing GM seeds. The selection of experts with technical education, confidentiality (as a way to keep information and guarantee corporate profits), and the consolidation of substantial equivalence12 as the backbone of the GM seed regulation show what I call production–expropriation–appropriation of knowledge as exchange value (see Table 1.1). The scientific knowledge acquired by these experts represents the production–expropriation–appropriation of knowledge as exchange value, and configure a regulatory framework that enables the rapid release of GM seeds. Substantial equivalence, as Barret and Abergel (2000, 5) note, “has augmented the explicit promotional bias of the federal government and the private sector (. . .). In the process, thorough assessment of environmental hazards and meaningful public dialogue, have been sidelined by the imperative to market GE (genetically engineered) crops quickly and competitively”.
32 Carla Poth Table 1.1 Shared conceptions that are crystallized in regulation Shared conceptions about biotechnologies and GMOs
Regulation
Knowledge has to be connected with production. If companies do not see benefits, they will not invest in research. To put the final technologies in markets is the main objective of researching. This relation seems to be natural. Companies may guarantee profits for their investment in research. That is why knowledge cannot be spread. Intellectual Property Rights are the best way to promote a safe investment. GM seeds have no risk because science can control genetic stability and guarantee predictability.
The experts selected to compose the CONABIA are related with the scientific system and the agrarian structure (they are not related with basic research) Regulatory frameworks guarantee confidentiality of data that is sensible for public. Substantial equivalence supports regulatory frames. If no difference can be demonstrated between GM crops and non-GM crops, then no risk is assumed.
Source: Author’s own elaboration
Scientific knowledge can create consensus and repels public participation. Those who disagree with, or do not understand, the dominant scientific discourse are excluded or marginalized. Within CONABIA, experts have created an epistemological antagonism that distinguishes between scientists and other actors, the latter being excluded from decision-making because they do not have the scientific arguments (although most of them are scientists from the same disciplines). This exclusionary expert knowledge was so strong that organizations such as the Agrarian Federation (FAA) and the Rural Society (SRA) – representing different agrarian interests – were not considered to participate in this commission.13 Instead, other agrarian associations, such as AACREA or AAPRESID, were considered to advise on those topics.14 Thus, scientific language within state institutions became the way to homogenize a political and economic perspective that supports the development of biotechnologies to promote the BAM and, at the same time, it was the filter to avoid the participation of the public that could threaten these regulatory frames. Science, therefore, created a critical harmony within the political structure of the BAM. With this consensus, Argentina released the GM Roundup Ready soybean, the first transgenic seed from Monsanto, in 1996. The second period (1998–2002): creative adaptation Although the initial adoption of GM seeds appeared without conflict, from 1998, everything started to change. In Argentina, while GM soy expanded more than 10 million hectares in one year, territorial conflicts began to emerge. Three main international problems jeopardized the commercialization of GMOs and pushed
Biotechnological Agrarian Model in Argentina 33 CONABIA to review its regulatory frameworks. First, they had to pay attention to the negotiations of the Cartagena Protocol on Biosafety (CPB). Its negotiations were relevant because the main objective of this protocol was to form an integrated perspective regarding the commercialization of GMOs at the international level, taking into account many diverse perspectives from different sectors and countries (Josling and Babinard 1999). Such diverse voices could be seen in the popular rhetoric of the time and included topics such as gene flow, food diversity, environmental risks, sanitary consequences, economic and social risks, cultural diversity, and democratic participation (Lövei et al. 2007). Panels of the CPB were followed by popular forums organized all over the world that showed and denounced the impacts of GMOs and problematized this extractive agrarian model. These experiences were essential, because they were a continuous channel of communication that globalized local struggles against the BAM, generating, multiplying, and empowering local struggles (Shurman and Munro 2010). Second, while the CPB was negotiated, the European Union (EU) started implementing a de facto moratorium to prohibit GM soy imports. Until 1998, the EU was the main export market for Argentina’s commodities, but this moratorium suspended the imports of GM crop for five years. Representatives of the EU said that the novelty of biotech products made their risk uncertain, justifying the implementation of the precautionary principle (Motta 2008). Finally, in 1998, Brazil entered these debates when GM soy entered into the country illegally from Argentina. Brazil’s Ministry of Science, Technology, Innovations and Communications then began the process to release their own GM soybean via the National Biosecurity Technical Commission (CTNBio) in July 2000. This process was met with resistance, particularly among peasant movements and environmental NGOs. Nonetheless, in 2003, under the Lula da Silva government, the commercialization of GM soybeans was approved. Finally, in 2005, Brazil approved a Biosafety Law that gave full power to CTNBio and ratified GM seed production and commercialization within the country (Menasche 2002; Poth 2010). As resistance crossed national borders and frontiers, organizations in Argentina were also influenced to push against the BAM. Labelling campaigns, lawsuits against illegal transgenic crops, the growth of peasant movements against agroextractivist expansion, the increased visibility of health impacts of pesticides, and the struggles of organic producers to keep their production safe from pesticides were some of the ways in which resistance started to emerge within the BAM in Argentina (Poth 2019). Argentina’s state had to react to these conflicts in order to maintain the legitimacy and market access for its GM crops. At the international level, Argentina became part of the so-called Miami Group (with Brazil, the United States, and Canada) which, within the CBP, stopped labelling and risk evaluation of GMOs and pushed to incorporate these regulations (or lack thereof) into the World Trade Organization (WTO) normative (Newell 2009; Gebre Egziabher 2007). It also made demands to the EU in the Dispute Settlement Body in the WTO in the context of their moratorium to stop GM imports into the region (Motta 2008). At
34 Carla Poth the national level, the Argentinian state had to reform its regulatory frameworks. Although members of CONABIA claimed that GM seeds had no environmental or health risks, they had to attune to the global consensus to avoid losing export markets. New regulatory frameworks tried to avoid commercial risks by using scientific–technical mechanisms that would ensure there were minimal environmental and health risks associated with GMOs, thereby generating certainty to global importers from Argentina. In this way, Argentina started to demand the isolation of GM crops to control the gene flow (res. 226/97), regulated the labelling of GM seeds for exportation (res. 34/98), and created post-harvest monitoring (res. 289/97). The state also created the requirements and criteria to authorize GMOs for food use (res. 511/98), including the National Service of Agri-Food Health and Quality (SENASA). Finally, it approved the “Phytosanitary Products Registration Manual” (res. 440/98 and 350/99) and included the Argentina Chamber of Agricultural, Livestock and Fertilizer Sanitation (CASAFE),15 recognizing the relation between GM seeds and pesticides. Although at a first sight we may see that these regulations were related to environmental or health concerns, when we make a deeper analysis we can see that the “environmental assessment” only focused on the relation between the GM crops with other crops for exportation, but not with the wider ecosystem. This was what was called the “agronomic risk” and it was based solely on the need to protect export-oriented crops. Labelling transgenic seeds was also exclusively for exports, not for local markets (Poth 2019). At every block in the road, the CONABIA was able to creatively adapt in order to continue facilitating the expansion of the BAM. But these adaptations have been plagued with unresolved tensions and contradictions. Its dominant discourses based on scientific innovation were no longer effective because the bases that supported this discourse were increasingly delegitimized and excluded. In this context, it was necessary to create a new regulatory framework that could bring more certainty to the BAM. The third period (2002–2009): learning from experience During the Argentine crisis of 2001,16 discussions concerning GMOs faded into the background and the BAM expanded to completely transform the agrarian sector. Despite increases in cultivation area and productivity, this period unveiled how profits were not distributed to the majority of farmers, but rather concentrated in the hands of corporations. The crisis also led to reinforcing Argentina’s insertion into global markets as a commodity exporter as investors took advantage of the crisis, and the upgrowth of the international price of commodities allowed a cycle of new profits for this sector (Bonnet 2015). The crisis in Argentina also resulted in the necessity to generate export revenues, but global markets had changed significantly. The insertion of new global producers of GM crops (like Brazil17 and Paraguay), and the emergence of new consumers (like China), demanded changes in the organizational structure of the BAM (Slipak 2012).
Biotechnological Agrarian Model in Argentina 35 In addition to this, institutional conflicts that emerged in previous phases were not resolved and structural reforms became necessary. From the diagnosis of the state, the reformulation of the institutions and legality of the BAM had two main strategies. First, the state recognized the diversity of struggles that arose during the implementation of the BAM. However, in a different way as in previous phases, it understood that it was necessary to give various actors in distinct class positions a political response, diversifying the institutional scaffolding and thinking about the BAM problem from a comprehensive perspective. Second, and relatedly, the state had to become a mediator between different sectors (from agrarian bourgeoisie18 to peasants, Indigenous, and environmental social organizations). Finally, the only way to appropriate more value from the exportation of commodities was to promote the production of knowledge as a state policy, confronting new competitors, like Brazil. In 2009, the creation of the Ministry of Agriculture allowed the state to intervene in every aspect of the agrarian chain, looking at the BAM in an integral way. This Ministry also had the ability to deal with conflicts in two ways. On the one hand, it started to incorporate multiple themes and debates that emerged, recognizing the existence of the diversity of agrarian and rural subjects with whom it was necessary to negotiate and debate politically. On the other hand, it involved a process in which state institutions redefined discursively the class struggle according to their logic and criteria. This containment allowed the composition of the institutional scaffolding, but the internalization of these conflicts became uncertain. New institutions started to answer to the BAM demands, proposing policies for the agrarian sector as a whole. The state started to have an active impulse to look for global markets for agro-exports and created new laws and family farming programmes that tried to co-opt and/or appease some environmental, peasant and Indigenous movements and political organizations. Finally, the state dealt with conflicts related to agro-chemicals at the federal level of governance. This further fragmented the articulation of these movements that, only in a few cases, managed to promote provincial or national resistances (Giaretto and Poth 2015). In this context, although CONABIA seemed to disappear it continued working but shaped its logics into this new phase. Its comprehensive strategy was to reinforce the “political logics” against the economic and technical decisions. That is why risk assessment or other technical responses were not enough to face all of these new challenges. CONABIA continued working but became protected by countless institutions that filtered the political conflicts associated with the BAM (Poth 2019). At the same time, the state started to have an active role in the investment process and in defining the scientific–technological directions. CONABIA guided this process. The creation of the Ministry of Science, Technology and Productive Innovation, in December 2007, with Lino Barañao as head of the portfolio, showed the hierarchy of the scientific–technological system in the political project. The proposal was to turn science and technology into a pillar of the new political project, to establish priority lines of research based on biotechnologies, and to train and incorporate scientific capacity to develop this project.19
36 Carla Poth But what kind of science and technology was financed? What was the logic of knowledge production that developed in this framework? The “Modern Biotechnology Development and Production Law”, approved in 2007, defined that the state would be in charge of guaranteeing investment in biotech products, and established tax exemptions for companies that invest in biotechnological research, while creating a fund to stimulate the sector. It also consolidated the cooperation between the public and private sectors.20 From here, the production–expropriation–appropriation of knowledge as exchange value, crystallized in CONABIA, turned into a coherent state policy and expanded using the increase of public financing and the systematic generation of research networks between public and private sectors. Research for practical purposes (related to markets), public–private technological development agreements (INDEAR as a successful experience), the strengthening of the appropriation of knowledge in pursuit of profits (through the reform of legal frameworks to strengthen intellectual property rights), and the recognition of biotechnologies as something to be developed within the scientific–technological system (expansion of university careers, growth exponential of public financing for the development of lines of research considered strategic) is the result of this state policy. From this development, the scientific system standardizes practices, rules, and research dynamics consistent with this logic of knowledge production consolidated in CONABIA and its experts. The fourth phase (2008–2019): nothing is closed Although the state developed a strong policy supporting the commodification of knowledge we cannot say that the scientific system aligned completely to this process. Science has to be thought as a way of production–expropriation–appropriation of knowledge, in a dialectical movement of change. The expansion of this policy of production–expropriation–appropriation of knowledge as exchange value tried to homogenize structures and practices within research spaces. Instead, however, it generated a fertile ground for the emergence of a new moment in the antagonism within the BAM, and what we call the “Carrasco effect” was essential to develop this new edge of the dispute. Andrés Carrasco was the President of Argentina’s CONICET and head of the Laboratory for Embryology at the University of Buenos Aires. He started to research the health impacts of agro-chemicals in embryos after seeing a map made by the environmental activist group, The Mothers of Ituzaingo, in Córdoba. This map, called “The map of death”, shows the ways in which people who lived in close proximity to GM soybean production were dying, and were getting sick due to agro-chemical exposure. In 2009, Carrasco’s results were published in mass media and widely distributed, triggering profound impacts and revolutionizing the scientific system. Representatives from key technical organizations of the BAM (AAPRESID, AACREA) and business chambers such as CASAFE reacted against Carrasco, and he was denounced by high-level state actors such as the Minister of Science, Lino Barañao, and agribusiness elites. The Ministry created a special commission to research agro-chemicals with the aim to evaluate the
Biotechnological Agrarian Model in Argentina 37 scientific information and claims by Carrasco related to glyphosate. Furthermore, CONICET declared publicly that Carrasco’s work was not financed or endorsed by their organization (Motta 2018). The final report of the commission concluded that “under responsible use of glyphosate, this product has a low risk for human health and the environment” (CNIA 2009, 3). This report was the official position of CONICET and the Ministry. However, Carrasco’s stature and findings influenced many scientists and doctors who also started to denounce the consequences of the BAM, questioning the type of science and technology being promoted at the state level (Skill and Grinberg 2013). Since then, a lot of related discourse started to emerge within universities and academic arenas. In 2010, the first National Meeting of Doctors from Fumigated Towns consolidated the relations between scientists and territorial organizations and concluded with the establishment of a Doctor’s network. They initiated Chairs of Food Sovereignty and new research about environmental and health impacts of agro-chemicals emerged. This new research tried to facilitate collaborative knowledge sharing between communities and the academy to identify new illnesses related to agro-chemicals (sanitary camps from University of Rosario, Monte Maiz research from the University of Cordoba), to identify water contamination associated with glyphosate (Ronco et al. 2016), to see the persistence of agro-chemicals in food and other agrarian products (Telam 2015; Aranda 2015b), and to see the presence of agro-chemicals in blood and urine. This relation between science and territorial organizations redefined the resistance to the BAM and brought a wider view of the problem and a critical visibility to the production of knowledge, creating as a result new ways to build it. In this way, Carrasco’s research provided not only new evidence to legitimate territorial struggles against the BAM, but opened a discussion within the scientific system about epistemological perspectives and the role of science in the agrarian chain (Poth and Manildo Forthcoming). This new debate within the scientific system transferred into other arenas and controversies, providing citizens with scientific arguments to support judicial and regulatory struggles. In 2012, there was a legal case against agricultural producers that fumigated in the Ituzaingó Annex neighbourhood, in Córdoba. This trial ended with a guilty sentence and was important because of its capacity to put the fumigation problem on the agenda and to consolidate new resistance networks. Researchers participated in the trial and accompanied similar activities on the streets, which were useful to show the complexity of the BAM and to denounce GM seeds and the role of CONABIA. Public agencies and private companies promoting the use of GM seeds were also denounced, including CONABIA for its lack of transparency and the absence of public participation and risk assessment (required by General Environmental Law, N° 25.675). Complementary to these legal mechanisms, the media exposed the conflict of interests and the secrecy that protected this public agency. What was CONABIA’s response to these debates? In 2010, CONABIA made the regulatory release of GMOs cheaper and easier, introduced more private
38 Carla Poth 1,64 9,84
11.48
1991-1997 (1) 1998-2003 (6)
77.04
2004-2009 (7) 2010- 2020 (47)
Figure 1.2 Percentage of GM seeds released in each phase Source: Data from MAGyP (2020)
institutions, and eliminated the second phase of evaluation for gene-stacked events.21 And these changes were effective. Of the 61 events released in Argentina, 47 were released in this last phase, including a new seed made in national labs, with the cooperation between the private and the public sector. As we can see in Figure 1.2, all of the regulatory changes tended to promote the releasing of GM seeds year by year. But CONABIA had to see all these demands. In 2013, it had to make public its rules of procedure and its composition (although the commission published the institutions, those members were not known) (res. N° 10/13). After that, it had to include a consultative mechanism to receive commentaries in the second phase of the risk assessment. Even with all of these changes, CONABIA has been recently criticized in a 2019 report by the National General Audit (organism of state control). This report highlights the absence of public participation, because consultative mechanism were not known or made accessible. The report also notes that CONABIA’s internal functioning is unknown to the public, what is in conflict with the “Access to Public Information Law” (N° 27.275). It reviews that federal governments are not represented22 and that regulation ignores international conventions putting in danger biodiversity and food security. In spite of these conflicts and critics, CONABIA continues to be one of the main promoters of the BAM. Finally, the coexistence since 2008 of the CONABIA (promoting the BAM) and the Secretariat of Family Farming (confronting the expansion of this model proposing the agro-ecological farming), represents a new complex duality of the agrarian chain. This contradictory policy is based in the idea that it is possible to support corporate-controlled monoculture dependent on biotechnological packages in parallel with an agrarian sector based on small-to-medium-scale family farms using agro-ecological production methods for regional economies.
Biotechnological Agrarian Model in Argentina 39 This political strategy was effective to contain some political organizations and to create a new institutional legitimacy. But it also tries to create a harmonious balance between two models that are mutually exclusive: on the one hand, the capitalist logic of BAM needs to expand to survive and, on the other, the agro-ecological farming endangers that expansion (questioning the use and ownership of land, the reproduction of seeds, and the use of agrochemicals). This contradiction opens the door to a new future story that is yet to be decided.
Nothing has concluded This chapter began with a discussion about biotechnologies and GM seeds as a revolution. In some way, radical changes have forced us to think about agrarian regions in a completely different way. Nothing will ever be the same, with patterns of profits spread all over natural resources or communitarian practices. Like never before, capital has penetrated agrarian relations, looking for new ways of accumulation and consolidating extractivism in all its forms. Production of knowledge has been its best ally. Today it is impossible to think of any aspect of the agrarian chain without connecting the farm (its logics, times, and organization) and what happens in labs or universities (what they are researching and for whom). Agrarian capital accumulation is a game that is played before seeds are sown. Biotechnologies have changed not only the agrarian structure but also the ways of making science itself. On land and in labs, capital tries to subsume every cell, spreading through everything, looking to create the conditions for accumulation. This indissoluble connection has to be part of our framework if we want to understand the underlying logic of dominant agricultural models and their tendency to be extractive in nature and character. In Argentina, state–capital alliances supporting the BAM have been exposed and the scientific “neutrality and objectivity” used to support this model has been discredited. The Argentinian state has consolidated the capitalist project with the BAM, using science as its rationale. Thus, political domination and economic exploitation of capital need science to consolidate the agrarian extractivist project. That is why those who are fighting against the BAM, and even against extractivism, are creating new ways to produce knowledge and are expropriating and reappropriating, in a critical way, some strategies of the hegemonic science. In those revolutionary moments, even within science, we need to remember the Marxist premise that says: the only reason for capital to expand is its intrinsic crisis of accumulation, generated by class antagonism. This crisis is the one that assures us that nothing has ended. The process of commodification in farms and labs has not concluded. This chapter shows how this antagonism unfolds, examining how complex forms of economics, politics, and science integrate in the total structure of accumulation of agrarian capitalism. It is clear why the production of knowledge is a main aspect of agrarian extractivism, but it is also an important territory of dispute. Conceptualizing this as a process of
40 Carla Poth production–expropriation–appropriation of knowledge allows us to understand this dispute and the ways in which science has been instrumental to agro-extractivist expansion. Struggles against extractivism have opened a “Pandora’s box”, putting science in dispute, exposing the politics and production of knowledge within science and evidencing that the “neutrality” that science uses is stained with capitalist domination. Some may argue that technology is not the problem, but rather how it is used. But, as I have showed in this chapter, this view can only be supported if we fail to see the capital antagonism spread within the production of knowledge. Disputes related to the BAM show that scientific knowledge and technologies are the problem because they are an important part of capital accumulation and, ultimately, extraction. Technologies and knowledge are not neutral. As researchers, we have the commitment to reveal the fetishized character that erases antagonisms within science and technologies. To reappropriate them, we have to recover our creative work and destroy our subsumption within the production of knowledge, thus destroying the patterns of domination that capital spreads.
Notes 1 This information originates from the Comisión Nacional de Investigación sobre Agroquímicos (CNIA). 2 This paradigm supports the relation between innovation and accumulation, from an endogenous view of technological change. Innovations are the key to economic growth and social benefits. From this view, the promotion of technological capacities stimulates competition, increasing productivity. In this way, research institutions are integrated with innovation systems by means of public policies that incorporate science with production, increase budgets designated to applied science and technology transfer, and increase the self-financing of scientific and technological institutions (Langer 2011). 3 This means that new regulation allows us to protect seeds with patent laws and with the Plant Variety Protection System. 4 Created in 2010, it is shown as one of the paradigmatic experiences of cooperation between the public sector (that trains and uses human resources) and Biosidus company that invests physical capital. INDEAR projects “would be oriented as a priority to respond to specific needs of the organized agro-industrial value chain”. INDEAR institutional video can be found at www.indear.com/ 5 This changed in 2017, when the Minister of Agriculture stopped the releasing of the GM wheat that had been accepted by the CONABIA. 6 SENASA is a state institution that regulates and certifies programmes and practices related to animal and vegetal health, including safety, hygiene, and quality of food. This institution also releases and classifies the agro-toxics used in agrarian production (and for other uses). 7 INASE is a state institution that overseas the seed market in Argentina, including seed certification and quality control, as well as the implementation of UPOV 78 in the country. 8 AACREA is a social association, composed of agronomists that promote the technological development of agrarian production. AAPRESID is another social association that promotes production with direct sowing. Both organizations were fundamental actors to promote the expansion of agrarian technological changes, consolidating a managerial way of production, expanding biotechnological innovations.
Biotechnological Agrarian Model in Argentina 41 9 This Forum is composed of seed corporations and institutions related to biotechnologies. From its beginning, it promotes relations between state, science, and the private sector, and spreads what it considers the benefits of biotechnologies for society. 10 The first institutions that composed it were: INTA (National Institute of Agrarian Technologies), University of Buenos Aires, the Argentinian Biotechnological Forum, CONICET (National Research Council of Science and Technologies), the National Direction of Agrarian Production and Commercialization, and SENASA. 11 CONICET is a public agency responsible for directing and coordinating the majority of scientific and technical research carried out in universities and institutes in Argentina. 12 Substantial equivalence requires that a new crop, particularly GMOs, must demonstrate that it is as safe as its traditional counterpart (OECD 2000). 13 Regulators said that those organizations were not interested in the technological policy for the agrarian chain. 14 Argentine Association of Regional Consortiums for Agricultural Experimentation is an institution that promotes the technological and managerial development of agribusiness. It is composed of small groups of farmers and technicians known as CREA groups. The Argentina Association of Direct Seeding Producers (Aapresid) is a nongovernmental non-profit Organization that promotes the agricultural paradigm to increase productivity, based on innovation, science, and knowledge management network. Most of the regulators within CONABIA participated in both organizations. 15 CASAFE is a private organization that groups national and international companies that produce fertilizers and chemicals for agrarian production. 16 The crisis of 2001 was a political, economic, social, and institutional crisis that was preceded by a popular movement demanding “Que se vayan todos” (“Everyone must leave”) that forced the resignation of the President Fernando de la Rua. This crisis is characterized by a deep recession and a representative conflict. The conclusion for this crisis was the currency devaluation in 2002, and an institutional review to rebuild the legitimacy of political institutions. The agrarian sector had a substantial role in the configuration of the crisis and its solutions. 17 Brazil not only increased its cultivated area but also developed its own GM seed with the state-owned research corporation EMBRAPA. 18 Organizations like FAA (Argentinian Agrarian Federation), SRA (Argentinian Rural Society), CRA (Argentinian Rural Confederation), and CONINAGRO (Agricultural Intercooperative Confederation) turned their production into biotechnologies, but they were not considered at the beginning of the BAM regulations. 19 In fact Lino Barañao is a molecular biologist, specialized in genetic engineering. 20 This role was ratified by CONICET, and by different programmes of biotech investment within the Ministry of Science and Technology and within the Ministry of Agriculture. 21 The stacked events are seeds that have more than one genetic modification in its DNA. CONABIA decided that if every event had been released separately, it was not necessary to evaluate risks when they are combined into one seed. 22 The National Constitution indicates that Federal governments manage natural resources.
References Aranda, D. 2015a. Tierra Arrasada. Buenos Aires: Sudamericana. Aranda, D. 2015b. Ocho de cada diez, con tóxicos. Página 12, 3 September. Available at: www.pagina12.com.ar/diario/sociedad/3-280798-2015-09-03.html. Augsten, F. 2005. El debate actual sobre el uso de las biotecnologías en la Unión Europea. Algunas implicaciones para los países del sur. In J. Villarreal et al. (eds.), Un mundo patentado. La privatización de la vida y del conocimiento. El Salvador: Heinrich Böll.
42 Carla Poth Barret, K., and E. Abergel. 2000. Breeding familiarity: environmental risk assessment for genetically engineered crops in Canada. Science and Public Policy, 27, 2–12. Bonnet, A. 2015. La insurrección como restauración. El Kirchnerismo. Buenos Aires: Prometeo. Ceceña, A.E. 2004. Estrategias de reproducción de una hegemonía sin límites. In A.E. Ceceña (ed.), Hegemonías y emancipaciones en el siglo XXI. Buenos Aires: CLACSO, 20–32. Clarín Opinión. 2001. Sólo la biotecnología salvará al mundo, 29 de enero. Available at: www.clarin.com/opinion/solo-biotecnologia-salvara-mundo_ (Accessed 15 November 2020). CNIA (Comisión Nacional de Investigación sobre Agroquímicos). 2009. Evaluación de la información científica vinculada al glifosato en su incidencia sobre la salud humana y el ambiente. CONICET. Buenos Aires. ETC Group. 2017. Quién nos alimentará. ¿La cadena industrial de producción de alimentos o las redes campesinas? 3rd edition. Available at: www.biodiversidadla.org/ Documentos/Quie_n_nos_alimentara_La_cadena_industrial_de_produccio_n_de_ alimentos_o_las_redes_campesinas_3_edicion_2017. Gebre Egziabher, T. 2007. The Cartagena protocol on biosafety: history, content and implementation from a developing country perspective. In T. Traavik and L. Li Ching (eds.), Biosafety First: Holistic Approaches to Risk and Uncertainty in Genetic Engineering and Genetically Modified Organisms. Tromso: Tapir Academic Press. Giaretto, M., and Poth, C. 2015. Lo territorial en disputa: aportes para la articulación entre luchas urbanas y luchas socioambientales. Despierta, 2. Brasil. Giarraca, N., y Teubal, M. 2013. Introducción: Las actividades extractivas en Argentina. En N. Giarraca y M. Teubal (coord.), Actividades extractivas en expansión. ¿Reprimarización de la economía Argentina? Buenos Aires: Antropofagia. Gilly, A., and Roux, G. 2009. El despojo de los cuatro elementos: capitales, tecnologías y mundos de la vida. In E. Basualdo and E. Arceo (eds.), Los condicionantes de la crisis en América Latina, Buenos Aires: CLASCO. Giraldo, O., and Rosset, P. 2018. Agroecology as a territory in dispute: between institutionality and social movements. Journal of Peasant Studies, 45(3), 545–64. Goldstein, D. 1989. Biotecnología, universidad y política. México: Siglo XXI. Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology, 78(6), 1360–80. Gras, C., and V. Hernandez. 2013. Los pilares del modelo agrobusiness y sus estilos empresariales. In El agro como negocio. Producción, sociedad y territorios en la globalización. Buenos Aires: Biblos, 17–48. Gudynas, E. 2011. El nuevo extractivismo progresista en América del Sur: Tesis sobre un viejo problema bajo nuevas expresiones. In Colonialismos del siglo XXI: Negocios extractivos y defensa del territorio en América Latina. Barcelona: Icaria Editorial. Harvey, D. 2004. El nuevo imperialismo: acumulación por desposesión. In Socialists Register. London: CLACSO. Josling, T., and Babinard, J. 1999. The Political Economy of GMO’s: Emerging Disputes Over Food Safety, the Environment and Biotechnology. Stanford: Institute for International Studies. Kloppenburg, J. 2004. First the Seed: The Political Economy of Plant Biotechnology, 1492–2000. Second edition. Wisconsin: University of Wisconsin Press. Lander, E. 2006. La ciencia neoliberal. In A.E. Ceceña (coord.), Los desafíos de las emancipaciones en un contexto militarizado. Buenos Aires: CLACSO, 45–94.
Biotechnological Agrarian Model in Argentina 43 Langer, A. 2011. Ciencia, universidad y economía. Voces en el Fenix, 8, 36–45. Lapegna, P. 2019. La Argentina transgénica. Buenos Aires: Siglo XXI Editores. Levins, R., and R. Lewontin. 1985. The Dialectical Biologist. London: Cambridge. López Monja, C., T. Perelmuter, and C. Poth. 2009. El avance de la soja transgénico: ¿Progreso científico o mercantilización de la vida?. Un análisis crítico sobre la biotecnología en Argentina. Buenos Aires: Centro Cultural de la Cooperación Editions. Lövei, G., T. Bohn, and A. Hilbeck. 2007. Biodiversity, ecosystem services and genetically modified organisms. In T. Traavik and L. Li Ching (eds.), Biosafety First: Holistic Approaches to Risk and Uncertainty in Genetic Engineering and Genetically Modified Organisms. Tromso: Tapir Academic Press. Luxemburg, R. 1951. The Accumulation of Capital. London: Routledge and Keagan Paul Ltd. MAGyP. 2019. Etchevehere: “Tenemos la firme posición de defender la biotecnología para alimentar al mundo”. Ministerio de Agricultura, Ganadería y Pesa. Available at: www.argentina.gob.ar/noticias/etchevehere-tenemos-la-firme-posicion-de-defender-labiotecnologia-para-alimentar-al-mundo (Accessed 28 November 2020). MAGyP. 2020. OMG Comerciales. Ministerio de Agricultura, Ganadería y Pesa. Available at: www.argentina.gob.ar/agricultura/alimentos-y-bioeconomia/ogm-comerciales (Accessed 28 November 2020). Marx, K. 1887. Capital. A Critique of Political Economy. Volume I. Moscow: Progress Publishers. McKay, B. 2018. Extractivismo agrario: dinámicas de poder, acumulación y exclusión en Bolivia. La Paz: Tierra. Menasche, R. 2002. Legalidade, legitimidade e lavouras transgenicas clandestinas. In H. Alimonda (comp.), Ecología Política: Naturaleza, Sociedad y Utopía. Buenos Aires: CLACSO. Motta, R. 2008. O risco nas fronteiras entre política, economía y ciencia: a controversia acerca da política sanitaria para alimentos genéticamente modificados. Master’s thesis, Social Science Institute, Universidade de Brasilia, Brasil. Newell, P. 2009. Bio-hegemony: the political economy of agricultural biotechnology in Argentina. Journal of Latin American Studies, 41, 27–57. OECD. 2000. Agricultural Policies in OECD Countries: Monitoring and Evaluation 2000: Glossary of Agricultural Policy Terms. OECD. Available at: https://stats.oecd.org/ glossary/detail.asp?ID=2604 (Accessed 20 October 2020). Palmisano, T. 2016. El agronegocio sojero en Argentina: Modelo extractivo en los mundos rurales. Revista Economía, 68(107), 13–33. Pengue, W. 2017. Cultivos transgénicos: La verdadera historia. Voces en el Fenix, 60(8), 62–72. Perelmuter, T. 2013. El rol de la propiedad intelectual en los actuales procesos de cercamientos. El caso de las semillas en la Argentina. In N. Giarraca and M. Teubal (eds.), Actividades extractivas en expansión. ¿Reprimarización de la economía? Buenos Aires: Antropofagia, 97–118. Postone, M. 2002. Tiempo, trabajo y dominación social. Una reinterpretación de la teoría crítica de Marx. Barcelona: Ediciones Jurídicas y Sociales. Poth, C. 2010. El modelo biotecnológico en América Latina: Un análisis sobre las posturas de los gobiernos de Lula y Kirchner en torno a los organismos genéticamente modificados y su relación con los movimientos sociales. In A.L. Bravo et al. (eds.), Los señores de la soja. La agricultura transgénica en America Latina. Buenos Aires: CICCUS, 261–308.
44 Carla Poth Poth, C. 2019. Biotecnología, ciencia y poder. Un análisis crítico sobre la regulación en torno a las semillas GM. Córdoba: Administración Pública y Sociedad. Poth, C., and L. Manildo. forthcoming. Impactos socioambientales y sanitarios del modelo de agronegocios: una lectura en clave de desigualdades y resistencias. In S. Feldman (comp.), Desigualdades en Argentina: actores, territorios y conflictos. Los Polvorines: EDIUNGS. Ronco, A.E., D. Marino, M. Abelando. et al. 2016. Water quality of the main tributaries of the Parana Basin. Glyphosate and AMPA in surface water and bottom sediments. Environmental Monitoring and Assessment, 188(8), 458–71. Rossi, E. (ed.). 2020. Antología toxicológica del glifosato. 5th edition. Buenos Aires: Naturaleza de Derechos. Available at: www.naturalezadederechos.org/antologia5.pdf. Roux, R. 2008. Marx y la cuestión del despojo. Claves teóricas para iluminar un cambio de época. Herramienta, 38, Año XII, Buenos Aires. Rullani, E. 2004. El capitalismo cognitivo, ¿un deja vú? In O. Blondeau et al. (eds.), Capitalismo cognitivo, propiedad intelectual y creación colectiva. Madrid: Traficantes de Sueños. Seoane, J. 2012. Neoliberalismo y ofensiva extractivista: Actualización de la acumulación por despojo, desafíos de Nuestra América. Theomai, 26. Second Semester. Shurman, R., and W. Munro. 2010. Fighting for the Future of Food: Activist Versus Agribusiness in the Struggle Over Biotechnology. Minneapolis: University of Minnesota. Skill, K., and E. Grinberg. 2013. Controversias sociotécnicas en torno a las fumigaciones con glifosato en Argentina. Una mirada desde la construcción social del riesgo. In G. Merlinsky (comp.), Cartografías del conflicto ambiental en Argentina. Buenos Aires: Ciccus, 91–117. Slipak, A. 2012. La emergencia de China como potencia: desafíos para el desarrollo que enfrenta la Argentina, IV Congreso Anual, 15, 16 y 17 de agosto de 2012, Facultad de Ciencias Económicas de la Universidad de Buenos Aires, Buenos Aires. Telam. 2015. Encuentran glifosato en algodón, gasas, hisopos, toallitas y tampones Iin Telam, 20 October. Available at: www.exactas.unlp.edu.ar/articulo/2015/10/21/ encuentran_glifosato_en_algodon__gasas__hisopos__toallitas_y_tampones. Wallace, R. 2020. Dead Epidemiologists: On the Origins of COVID-19. New York: Monthly Review Press.
Official documents Argentina.gob.ar “Tenemos la firme posición de defender la biotecnología para alimentar al mundo”. 9 de Septiembre de 2019 www.argentina.gob.ar/noticias/etchevehere-tenemosla-firme-posicion-de-defender-la-biotecnologia-para-alimentar-al-mundo Minimum Budgets for Environmental Protection of Native Forest, Law 26.331 Modern Biotechnology development and Production, Law 26.270 Bicentennial Plan for Science, Technology and Innovation (2006–2010) Strategical Program of Agrarian Biotechnology (2004–2015) Comisión Nacional de Investigación sobre Agroquímicos (CNIA)-2009. Administrative resolutions of Secretary of Agriculture: N° 124, 656, 669, 837, 328, 226, 289, 34, 511, 440, 396, 131, 350, 1265, 219, 412, 500, 39, 57, 644, 362, 249, 710, 510, 10.
2
Extractive dynamics of agrarian change in Bolivia Ben M. McKay and Gonzalo Colque
“Soya permite lograr la soberanía alimentaria del país” (Soy allows for the achievement of food sovereignty of the country). This was the message that came out of a Forum held in La Paz in 2014 hosted by Bolivia’s agribusiness representatives (ANAPO 2014). The “grano de oro” (golden grain), as they call it, not only helps the country achieve food sovereignty, but generates more than 100,000 direct and indirect jobs, it is claimed. Thanks to genetically modified (GM) seeds, they said, productivity has not only increased, but production costs have decreased, requiring less agro-chemicals, and all producers, whether small or large scale, have adopted and benefitted from these new technologies. The key message from the event is perhaps best captured by José Luis Landivar Bowles, President of the Instituto Boliviano de Comercio Exterior (Bolivian Institute of Foreign Trade, IBCE): The noble soybean crop plays a transcendental role in the country’s oilseed chain (cluster) as it is its production, the main link that initiates a series of other activities or links to generate value; under a system of sustainable production, that is, economically viable, environmentally sustainable and socially responsible, contributing to Bolivian food sovereignty with the production of protein of plant origin that will become food for humans, animal protein and a renewable energy source, apart from generating huge foreign exchange from exports with added value, taxes and quality jobs.1 (IBCE 2014) For agribusiness operating in Bolivia and its landholding elite, soybeans have become the most important cash crop and the only GM crop legally permitted in the country since 2005. Soybean plantations occupy over one-third of the entire cultivated area in the country, and the oilseed crop is not only the top agricultural export but one the country’s top three total exports (in terms of value), forming one of the pillars of the three-pronged extractivist development model along with hydrocarbons and minerals (INE 2020). Soybeans have transformed the country’s landscape, replacing traditional and conventional crops and forested lands, and have significantly changed the forms and relations of production, property, and power. This chapter analyses the extractive dynamics of Bolivia’s agricultural
46 Ben M. McKay and Gonzalo Colque model from the lens of agrarian political economy and ecology and challenges the claims put forth by both agribusiness representatives and the state. Rather than a form of industrial agricultural development which implies value-added processing, sectoral linkages, and employment generation, we argue that Bolivia’s soy complex is better characterized as a form of agrarian extractivism. Agrarian extractivism brings the extractive character of the so-called industrial agriculture to the forefront of our analysis, interrogating the socio-economic and socioecological implications of the agricultural model. In Bolivia, industrial agriculture, or agro-industry, is used as a type of discursive framing to legitimize the dominant model of agricultural development today. Fuelled by neo-Malthusian logics and theories of modernization, the dominant model of agricultural development hinges on achieving maximum yields and efficiencies through technical fixes and frontier expansion. Globally, the agro-food sector has become increasingly corporatized and concentrated, largely controlled by a few multinational firms that collaborate among themselves to control the upstream and downstream components of the agroindustrial complex (see Clapp 2018). The influence and power of agribusiness over all aspects of the value chain has equipped the agro-industrial model with legitimacy and authority in the countryside, as research and development has become largely controlled by just a few agribusiness conglomerates. Agrarian extractivism, as a concept, challenges this legitimacy and authority by revealing the extractive form and character of this model and its socio-economic and socio-ecological implications. The remainder of this chapter is structured as follows: in the next section, we analyse the extractive character of Bolivia’s soy complex, putting forth four key features of agrarian extractivism which we argue has both analytical and political utility as a concept. In the third section, we delve into some of the most recent dynamics of agrarian change including the widespread forest fires that scorched Santa Cruz in 2019, the new biofuel initiatives in 2018, and the recent legislation concerning genetically modified organisms (GMOs) passed in 2020 by the unelected interim government. We then conclude with some final thoughts on the analytical and political utility of agrarian extractivism as a concept with reference to the Bolivian case.
Agrarian extractivism in Bolivia Bolivia’s insertion into the corporatized agro-commodity food system is a relatively new development. It was only in 2005 that GM soybean seeds were legalized,2 introducing new technological packages and inputs, and incorporating many smallholders into new relations and forms of production. The integration of smallholders into the soy complex, who represent the rural majority, has led to an increased dependence on external inputs – that is, chemical-based pesticides and fertilizers, GM seeds, heavy machinery – as well as an increasingly monopolized control over the storage (silos), processing and access to external markets necessary for the export-oriented crop.
Extractive dynamics of agrarian change 47 This transition coincided with the election of Evo Morales and the Movimiento al Socialismo (Movement Towards Socialism, MAS) who came to power with overwhelming support from the country’s historically marginalized populations, including some of the most influential social movements representing Indigenous, peasants, and other rural dwellers (see McKay 2018a). Despite earlier attempts at agrarian reform in 1953 and 1996, Bolivia’s landholding structure remained highly unequal, based on minifundio and subsistence agriculture in the western highlands and large-scale, capital-intensive monocultures in the eastern lowlands, which, “do not interact or complement each other” (Kay and Urioste 2007, 69). Indeed, Bolivia’s agrarian reform was left unfinished and the rise of Evo Morales and the MAS to power came with much anticipation and hope for transformative change. Upon taking control of the state apparatus, the MAS government promised an “agrarian revolution”, which would transform the unequal landholding structure and incorporate concepts such as food sovereignty, rights of mother nature and vivir bien into the New Economic, Social, Communitarian, and Productive Model. Such promises, however, went unfulfilled, as the MAS entered into a strategic alliance with the agribusiness sector, shifting its discourse to one of frontier expansion and contradictorily associating the soy complex with food security and sovereignty. Rather than democratizing and taking control over its food system, Bolivia became more dependent than ever on food imports during the first term of the MAS, with food imports amounting to US$1.1 billion (Urioste 2011, 31). While important progress was made in terms of formalizing land titles for Indigenous territories, there was never a serious attempt at actually reforming the unequal landholding structure (see Urioste 2012; Colque et al. 2016). Rather, a state–capital alliance was formed which enabled the MAS to gain support and ease oppositional tensions among economically powerful groups, maintaining its political power and control, while enabling agribusiness and landholding elites to increase their control over land and the soy complex in the eastern lowlands.3 Such an alliance resulted in many internal frictions within the MAS and among many of its constituents. As Bolivia’s former vice minister of land, Alejandro Almaraz (2014, 54), put it after he parted ways and became a critic of the MAS, “The Indigenous and peasant project . . . has been defeated, and the dominant interests and power of the business sector, the oligarchs and the latifundistas have been reinstalled”. These dynamics of agrarian change have enabled agribusiness and landholding elites to increase their control over Bolivia’s land and agricultural complexes resulting in a model of agricultural development that is extractive, rather than industrial, in its form and character. Agrarian extractivism as conceptualized here builds on much of the literature on extractivism, particularly as a mode of accumulation (Acosta 2013) and appropriation (Gudynas 2015), as well as the three dimensions put forth by Gudynas (2013) regarding scale (volume of material extracted), ecological impacts (intensity of extraction), and resource destination (semi-processed for export). As a mode of accumulation, agrarian extractivism involves particular social relations of production and reproduction in the current phase of capitalist agriculture in
48 Ben M. McKay and Gonzalo Colque which the ground rents and surplus value are extracted and labour opportunities and conditions deteriorate or disappear via new forms of value-chain control and exclusion. Taking these factors together with Gudynas’ dimensions of extractivism and inspired by the work of Alonso-Fradejas (2015; this collection), we characterize agrarian extractivism in Bolivia with the following four key interlinked features: (i) significant volumes of materials extracted, primarily for exports, with little or no processing; (ii) value-chain concentration and sectoral disarticulation; (iii) high intensity of environmental degradation; and (iv) deterioration of labour opportunities and labour conditions in the area or sector. These four features are not meant to represent a one-size-fits-all definition for agrarian extractivism; rather, they bring to the fore the extractive character of the so-called industrial agriculture across socio-economic and socio-ecological spheres with a particular emphasis on scale (scale of extracted materials and/ or scale of capital involved), control (over value chains and control grabbing) sectoral linkages (or lack thereof), and the extent to which the sector represents an extractive enclave. The degree and form of extractivism in agrarian sectors will undoubtedly vary in different contexts and crop complexes, as this collection demonstrates. The soy complex in Bolivia must be understood in the context of the country’s broader national development strategy based on a neo-extractivist development model. The increased role of the state in the Bolivian economy since 2006 has coincided with a somewhat laissez-faire strategy in the agricultural sector. Unlike other extractive sectors (e.g. hydrocarbons) where the state appropriates part of the resource rents via taxes and royalty payments, land and agro-export taxes are nearly negligible. The legal mechanism for transferring part of the land rent to the state is the land tax, applicable only to medium- and large-scale land owners according to land size. However, the land tax is not enforced in practice. The municipal governments in charge of collection do not have the administrative capacity or mechanisms to enforce and collect the tax. For agricultural production and exports, there is a special tax regime known as the RAU (Régimen Agrario Unificado) which is also only applicable to medium- and large-scale landowners and combines the value-added tax (Impuestos al Valor Agregado, IVA), the transaction tax (Impuesto a las Transacciones, IT), the tax on business profits (Impuesto sobre las Utilidades de las Empresas, IUE), and another tax regime complementary to the value-added tax (Régimen Complementario al Impuesto al Valor Agregado, RC-IVA). According to municipal state actors in communities in Santa Cruz, however, this tax also lacks enforceability and, even when enforced, contributes very little to state revenues. In the country’s most fertile lands of Santa Cruz, for example, the RAU amounts to US$3.83 (25.42 Bs) per hectare per year – a small fraction compared to the resource rents extracted and the profits appropriated by landowners and agribusiness. The expansion of the soy complex has also halted the agrarian reform, or agrarian revolution, that was an important part of Morales’ political agenda when he came to power in 2006. The abandonment of Morales’ initial plan to carry out redistributive agrarian reform via expropriation, combined with the lack of state
Extractive dynamics of agrarian change 49 tax on land and agricultural production and new decrees established to legalize the use of new GMOs (see the following sections), are indicative of the power and influence of the agribusiness lobby in Santa Cruz. While the politics of this state–capital alliance go beyond the scope of this chapter see the work of Webber (2017), McKay (2018b), and Tilzey (2019) for some insightful critical analyses. In the ensuing sections, Bolivia’s soy complex is analysed in terms of the four interlinked features of agrarian extractivism. Significant volumes of raw material exports The first dimension of agrarian extractivism as defined here concerns the volume of raw materials extracted and destined for export with little or no processing. Volumes are considered large or significant relative to other agricultural-based exports, while processing is only significant if it generates value-added sectoral linkages and employment opportunities. In Bolivia, the area under soybean cultivation has dramatically increased over the past 20 years, with the volume of soybeans increasing from 995,500 metric tons (t) in 2000 to 1,921,330 t in 2019 while land area went from 490,500 hectares (ha) to 1,028,000 ha during the same period (ANAPO 2019). In 2019, 80% of all soybeans and derivatives were destined for export (ANAPO 2019). It is the value-added component of the production process that can trigger sectoral articulation, as complementary sectors engage in industrial processing and manufacturing, creating employment in the upstream and downstream sectors. When soybeans are semi-processed into oilcake and meal for export, there is no sectoral articulation and little employment generation. The oilcake must be further processed to be converted into animal feed or consumer products. The significant volume of soybeans produced, semi-processed, and destined for export represent the first of four features of agrarian extractivism in Bolivia. Value-chain concentration and sectoral disarticulation The second dimension of agrarian extractivism in Bolivia pertains to the concentration of value-chain control and the lack of sectoral articulation. The majority of value-chain components are not produced in Bolivia, meaning their associated surplus value is appropriated elsewhere. The majority of Bolivia’s GM seeds, for example, come from Argentina, and just four companies control 86% of the seed market, while agro-chemicals are imported from China (48%), Argentina (22%), Brazil (7%), and Paraguay (7%) (INIAF 2015) (SENASAG 2014). Further, four of the top six companies that control 85% of the soybean market for storage, processing and exports are owned by transnational agribusinesses, which include US-based multinationals ADM and Cargill (AEMP 2013; McKay 2018c). Heavy machinery – tractors, fumigators, and harvesters – that has replaced manual labour and changed the entire production process with substantial productivity increases is also imported, principally from Brazil (61%), the United States (10%), and Argentina (9%) (INE 2016).
50 Ben M. McKay and Gonzalo Colque Surplus value is produced during the production of these use values, which are imported to Bolivia in order to increase labour productivity and ultimately extract more value from soybean production. The natural fertility of Bolivian soil is a source of use value in the production process, which, through the application of labour power and agro-industrial inputs, produces surplus value represented by the soybean, an agro-commodity exchanged on international markets. Agro-industrial inputs and mechanization have substantially decreased the socially necessary labour time to produce soybeans, reducing the need for labourers and extracting more of nature’s use values through intensification, thus increasing the relative surplus value appropriated by agro-industrialists and capitalist producers. The natural fertility of the soil varies across geographical areas, enabling those who control more favourable soils to appropriate more surplus profit. The soil’s natural fertility increases the productivity of labour and enables the appropriation of surplus profits either by the capitalist producer or by those controlling the land in the form of ground rent (differential rent I) (Marx 1981). Furthermore, different capitals (seeds, agro-chemicals, machinery) can also produce more value (at least temporarily, due to decreasing soil fertility) on equal amounts of land using the same amount of labour power. This is the appeal of technological innovation in agriculture, such as high-yielding seed varieties, agro-chemical inputs, and advanced mechanization. This is another form of surplus profit extracted from nature (as a use value), which may be appropriated as surplus value by capitalist producers or as ground rent (differential rent II) by landowners (Fine and SaadFilho 2004). In other words, ground rent is the appropriation of surplus profits by landed property. In the context of Bolivia’s soy complex, the main source of value is appropriated through ground rent, though some labour is needed to extract the resource and bring it to the market in commodity form. The most central part of the production of soybeans, which is the basis for the entire complex, is land. Control over the land means control over where the soybean (and surplus value) is produced. The soil and the worker, as Marx (1976, 638) put it, are “the original sources of all wealth”. Land serves as the means of production, by providing nutrients for the soybeans to grow, and embodies part of the production process within the soil itself (Harvey 2006). Of course, other necessary inputs, including labour power, seeds, and machinery, are required, but land remains central. Yet formal land ownership has become less and less important in appropriating the surplus value from the production process due to processes of productive exclusion and value-chain control (see McKay and Colque 2016; McKay 2018c). Bolivia’s soy complex can thus be characterized by the importation of finished products (GM seeds, agro-chemicals, machinery), the circulation of this agrocapital through Bolivian soil controlled by a small minority of agro-capitalists, concentrated control of the production process by a few companies, and the export of the semi-processed soybean in its commodity form. The extraction of the surplus value from the production process and appropriation by a few domestic and multinational companies (value-chain concentration), along with the lack of
Extractive dynamics of agrarian change 51 forward and backward linkages in the domestic economy (sectoral disarticulation), represent a significant dimension of agrarian extractivism in Bolivia. High intensity of environmental degradation The third dimension of agrarian extractivism in Bolivia concerns the intensity of environmental degradation in the region. The intensity of environmental degradation refers to unsustainable farming practices that have lasting effects, most directly on the communities in close proximity to the production process but also beyond. Highly mechanized and GM soybean production is based on a myriad of unsustainable farming and land-use practices associated with declining soil fertility and erosion, contamination of water sources, high rates of deforestation, and loss of biodiversity, which contribute to climatic changes such as increased flooding and drought (Hecht 2005; Pengue 2005; Müller, Pacheco, et al. 2014). This is linked to the heavy use of synthetic fertilizers and agro-chemicals required to treat large-scale monocultures, the mechanization of production and the massive expansion of the agricultural frontier to serve export markets principally for animal feed and biodiesel (Catacora-Vargas et al. 2012). Since the legalization of GM seeds, the quantity of agro-chemicals used in production has far outpaced the cultivation area, more than tripling from 12.6 million kilograms in 2010 to 38.3 million kilograms in 2014, while area under cultivation increased by just 28% (SENASAG 2014). The ecological extraction taking place as a result of an expanding agro-industrial soy complex is apparent and is exhausting the soil and threatening the health and safety of communities. Soybean farmers attest to this, explaining that year after year new types of weeds and pests threaten their crops and require new types of herbicides and pesticides. One prominent community leader in Cuatro Cañadas, Paulino Sánchez, arrived from Potosí in 1983, receiving 50 ha from the government. He said that one of the more challenging issues for farmers today is the depletion of the soil’s fertility. “The soil is losing its nutrients”, he said. “There is compaction from machinery and people use a lot of chemicals, so yields are decreasing”. Since the legalization of GM seeds, yields have fluctuated between 1.3 and 2.7 metric tons per hectare, which is difficult for farmers’ economic security since for each hectare, one ton of harvest roughly covers the cost of production, he said. Those who can afford the best seeds and technologies, however, do not suffer the cost of this environmental degradation to the same extent, at least in the immediate term. Sánchez also explained that many people are worried about the risks of investing in production due to high costs and frequent periods of drought and floods. Some people have lost everything due to natural disasters and therefore do not want to risk their entire savings on a tractor or harvester with so much uncertainty. These natural disasters, particularly the effects of El Niño and La Niña, are increasingly affecting not only farmers and their harvests but entire communities. Since 1990, there have been a total of 25 floods, which have resulted in the death of 674 people and affected close to 3 million (EM-DAT 2016). While the expansion of Bolivia’s agricultural frontier and the resulting deforestation is certainly
52 Ben M. McKay and Gonzalo Colque not the only cause of the increased frequency and severity of floods and drought, forest loss greatly influences regional and global climates, as forests not only play an important role as a carbon sink but also return water to the atmosphere via the extraction of soil water by tree roots, referred to as “a transpiration service” (Malhi et al. 2008). A United Nations study reveals that in the past 30 years, Bolivia has lost over 6 million ha of forest and has one of the highest rates of deforestation per capita in the world (320 square metres per person per year) – 20 times higher than the global average (16 square metres per person per year) (UN-REDD 2010). Seventy-five per cent of this deforestation activity is located in Santa Cruz, with an average deforestation rate of 200,000 ha per year from 2000 to 2010 (Cuéllar et al. 2012). A study by Müller et al. (2013) found that from 1992 to 2004, 72.6% of the 1.88 million ha of forests cleared in Bolivia’s lowlands were removed due to medium- and large-scale mechanized agriculture (53.7%) and small-scale agriculture (18.9%), with cattle ranching causing the remaining 27.4%. This period coincides with the initial expansion of the agricultural frontier when soybean cultivated areas increased from 164,920 ha in 1992 to 602,000 ha in 2004 (ANAPO 2015). In the late 1990s, for example, three Mennonite communities abandoned over 100,000 ha of soybean land due to soil erosion, compaction, and exhaustion and moved north to clear new territory, selling their land to cattle ranchers (Fearnside 2001). However, for the period from 2005 to 2010, Müller et al. (2014) found that the principal drivers of deforestation in the lowlands had reversed, with cattle ranching representing 59.7% and mechanized agriculture (24.6%) and small-scale agriculture (15.9%) representing a combined 41.3%. As land prices increase and market conditions make growing soybeans more profitable than cattle ranching, cattle ranchers are pushed into new areas, triggering more deforestation and opening new areas for future soybean expansion (Fearnside 2001; Hecht 2005; Weis 2013). Deforestation has not only led to a loss of biodiversity, it also affects the communities that depend on forests for their livelihoods, particularly the Indigenous territories of Lomerío, Guarayos, and Isoso, which are located in and around the expanding agricultural frontier. A study by Vadillo et al. (2013) concludes that large-scale monoculture expansion is one of the major threats to the Indigenous territory of Lomerío and the Chiquitano people, who collectively have the rights to close to 260,000 ha, 60% of which is forest. Guarayos is another Indigenous territory threatened by agro-extractive expansion. Located to the north of the expansion zone, many farmers in Cuatro Cañadas and San Julián are seeking land in and around Guarayos, where the frontier is being extended as a result of illegal deforestation, occupation, and land deals between Indigenous leaders and soybean farmers. Despite such threats to Indigenous peoples and the biodiversity that many depend on for their livelihood, the government’s agenda as announced by Vice President García Linera and agribusiness representatives of the Agricultural Chamber of Eastern Bolivia in 2012 was to increase the agricultural frontier by 1 million ha per year until 2020 to
Extractive dynamics of agrarian change 53 “guarantee food sovereignty” (Vicepresidente 2012; Heredia Garcia 2014). This exemplifies the state’s attempt to justify and legitimize forms of capital accumulation through popular discourses, serving the interests of the landed elites and agribusiness while capital-poor farmers, rural dwellers, and Indigenous peoples remain excluded and further threatened by the environmental consequences of the expansion. The dynamics of deforestation run much deeper than the loss of forests and biodiversity. Forests provide important land cover, prevent erosion, absorb rainfall, and regulate weather and climate patterns. There is ample evidence that deforestation amplifies flood risk and exacerbates the severity of El Niño – Southern Oscillation (ENSO) climate cycles (Bradshaw et al. 2007; Malhi et al. 2008). Rapid rates of deforestation have coincided with increased floods in Bolivia, with the most catastrophic floods in recent history in 2007–2008 and 2014. In 2007, floods displaced over 100,000 families, killing 50 people and affecting 366,000 ha of cultivated land, while in 2008, floods resulted in the deaths of another 67 people, displacing 97 families as the river, the Rio Grande, which borders the principal soy-producing communities of El Puente, San Julián, Cuatro Cañadas, and Pailon, rose 3–4.5 metres (BID 2014). In 2014, 85 municipalities were affected, displacing some 24,036 families, destroying 713 homes, killing 44 people and affecting 352 ha of cultivated land (BID 2014). In May 2015, over 100 communities in the municipality of San Julián alone lost much of their harvest due to floods. Abraham Guzman of Nucleo 20 in San Julián lost his entire parcel (45 ha), while approximately 700 ha in his community were flooded. For small farmers, this results in almost an entire year’s income lost, while their initial investment in agro-inputs leads to indebtedness – sometimes to agro-industry such as ADM, Gravetal, Monica or FINO, or to other farmers – which could potentially lead to having to sell their land. The ENSO phenomenon and floods continue to increase in intensity and frequency. In the 1960s, the department of Santa Cruz experienced normal conditions without the ENSO climate phenomenon 74% of the time, while in the first decade of the 21st century, the region experienced the ENSO phenomenon 76% of the time with increased severity and frequency of floods and drought (ACF-IN 2009). The increasing occurrences and severity of floods and drought do not affect everyone equally. In the soy expansion zone and especially in the municipalities of Cuatro Cañadas and San Julián, the vast majority of small-scale farmers occupy plots of land in the flood zone around the Rio Grande, while large-scale farmers occupy the more fertile and slightly higher land to the east of Highway 9. Though deforestation due to the expansion of the agricultural frontier has been identified as a major contributor to the increasing frequency and intensity of flooding and erosion in Santa Cruz’s Land Use Plan (Plan de Uso del Suelo), new laws have been established that excuse illegal deforestation (Law 739), increase deforestation limits (Law 741), and extend the regulatory inspections to make sure land is meeting a socio-economic function (Función Económica y Social) from every two to every five years (Colque et al. 2016). Rather than implementing strict regulations against deforestation and promoting reforestation and sustainability
54 Ben M. McKay and Gonzalo Colque practices, the government is investing close to US$17 million in construction projects to protect communities and productive lands in order to facilitate the continued expansion of the agricultural frontier to serve the interests of the soy complex (ANAPO 2015). These residual solutions do not address the systemic ecological crises unfolding due to large-scale monoculture production and frontier expansion. Farmers are left with more uncertainties than ever – in terms of volatile yields due to decreasing soil fertility and increased agro-chemical requirements, drought and floods due to the ENSO system, dust storms, loss of biodiversity and an inability to diversify production, and the contamination of water sources partly due to widespread fumigation. A study by Mekonnen and Hoekstra (2011) of the Twente Water Centre in the Netherlands found that every ton of harvested soybeans requires a ton of water. Considering the total amount of forest cover lost, the decline in soil fertility, loss of biodiversity, and water contamination, the result will be a stark ecological deficit and potential for ecological crisis if this production model persists. The capitalist penetration of agriculture and the appropriation of seeds, fertilizers, pest control, and even labour by industry has led to the “discontinuous but persistent undermining of discrete elements of the agricultural production process, their transformation into industrial activities, and their re-incorporation into agriculture as inputs” (Goodman et al. 1987, 2). This process, what Goodman et al. (1987) call appropriationism, has provoked this irreparable rift in the socioecological metabolism. The substitution of natural for industrial inputs allows for the accelerated extraction of nature as a “free gift”, overriding previous ecological constraints, while the technological packages of synthetic fertilizers, GM seeds, agro-chemicals, and mechanization similarly override traditional farmer knowledge, practice, and labour requirements, rendering farmers increasingly dependent on or even obsolete to agribusiness. This separation and disregard for the socio-ecological metabolism leads to the neglect of natural regeneration and the symbiosis of agro-ecological processes as well as rapid environmental degradation through externalizing costs and technological fixes. Such biophysical overrides are unsustainable and its tendency to generate ecological crises and move into new greenfield sites exposes the accelerating contradictions of “industrial” capitalist agriculture (Weis 2010). With nearly 80% of Bolivia’s soybeans and derivatives destined for export, soybeans are not only extracted and appropriated for value realization elsewhere, but the mode of extraction diminishes the productive capacity of the natural resources in the long term, leading to ecological impoverishment and unequal ecological exchange between trading countries (Bunker 1984). As Bunker (1984, 1053) asserts in his important work on extractive export economies in the Amazon basin and their tendency for unequal net flows of matter and energy exchanges with productive or articulated industrial economies, “[w]e must consider the effects of the exploitation of labor and the exploitation of the entire ecosystems as separate but complementary phenomena, both of which affect the development of particular regions”. The extraction, or expropriation, of nature is a defining feature of agrarian extractivism in Bolivia. As Bunker wrote over 30 years ago, “The
Extractive dynamics of agrarian change 55 ecological and demographic consequences of these disruptions are likely to last far longer than the demand for the commodity or the particular mode of extraction which provides it” (1984, 1056). For Bolivia, the socio-economic and ecological impoverishment of its principal mode of extraction, mining, should serve as a stark reminder. The tragic underdevelopment and impoverishment of Bolivia’s once largest and richest mining city of Potosí or the more recent disappearance of Bolivia’s second-largest lake, Lago Poopó, for which hundreds of families depend for their livelihood, exemplify the harsh realities of extractivism. Deterioration of labour opportunities and conditions The fourth dimension of agrarian extractivism in Bolivia concerns the lack of labour opportunities and deteriorating labour conditions pertaining to the soy complex. From a labour perspective, there is nothing inherently undesirable about mechanized soybean production. To be sure, most people would much rather benefit from the increased labour productivity and less physically demanding conditions of labour associated with mechanized agriculture. In other words, it is not mechanization as a form of agricultural production that is in and of itself a problem but the social relations of production associated with the form within the broader socio-economic context. When the form of production substantially decreases the need for labour in a sectorally and socially disarticulated economy, it can result in surplus populations (Li 2009). And this is precisely what is happening in Bolivia’s soy complex. The claim from ANAPO and IBCE that soybeans generate 100,000 jobs is extremely inaccurate and misleading. Aside from the exaggeration of direct employment, the other significant employment claims in storage (1,431) and transportation (26,824) involve temporary and precarious jobs. In my research, I found that many of the storage and processing facilities employ 2 to 6 full-time staff, 7 to 14 part-time technicians, and another 7 to 14 part-time general labourers. Further, the transport truck drivers (internal) are hired during harvest, which spans only a few months of the year and does not offer any type of job security or benefits. Finally, the job numbers provided by ANAPO are for the soybean summer harvest – the busiest and most lucrative time of the year for agricultural production. For winter harvests, the estimates for total employment drop to just over 40,000 and include the same part-time and precarious jobs. Measurements of the employment generation of an industry should consist of stable, annual employment, not precarious one- to two-month seasonal labour opportunities. This dimension of agrarian extractivism is not limited to the decreased need for wage labourers in the production process, as it also relates to deteriorating conditions for workers – in terms of health, safety, and precariousness. Manual sugar cane cutting in Brazil, for example, still provides a livelihood for some 500,000 people, but conditions are extremely demanding both mentally and physically, often akin to slave-like conditions (Alves 2006; McGrath 2013). Data from Brazil’s Land Pastoral Commission (Comissão Pastoral da Terra) revealed that 10,010 workers were liberated from slave-like labour conditions in the sugar cane sector
56 Ben M. McKay and Gonzalo Colque from 2003 to 2010 (Brasil 2011). Furthermore, in reference to oil palm and sugar cane plantations in Guatemala, Alonso-Fradejas (2015, 492) asserts that “while labor and labor arrangements are flexibly organized to maximize surplus extraction, the working conditions are damaging workers’ physical and mental health in severe and even deadly ways”. Both of these dimensions of labour are considered part of agrarian extractivism. In Bolivia’s highly mechanized soy complex, it is labour’s lack of utility for capital accumulation that is generating exclusion and surplus populations. Together, these features point to a model of agricultural development that is unsustainable and detrimental for the rural majority with severe socio-economic and socio-ecological implications. It is not one of industrializing the countryside, but rather of value extraction, labour exclusion, and nature expropriation. But it is not only the soy complex that has these extractive features. Recent agrarian dynamics in Bolivia are exacerbating and reinforcing the extractive model of agriculture with important implications for rural livelihoods and futures.
Recent agro-extractive dynamics: forest fires, biodiesel and GMOs In the line of fire The boom in tropicalized soy ended up seducing the government of Evo Morales who, leaving aside his ecological rhetoric, forged a political and economic alliance with the agribusiness sector. This public–private coalition has turned the forests of Chiquitanía and the Amazon into emerging territories for the (agro) extractive economy. The gradual deregulation of forest protection laws as well as the fight against private appropriation of state-owned land has unleashed a rampant race for land grabbing, now part of the drivers that explain the intensification of wildfires. Although agrarian extractivism has been the subject of public attention for a few years, the 2019 forest fires alerted to the magnitude of the problem, not only in Bolivia but also in Brazil. The forests and regions connected to the agricultural borders of both countries suffered fires that caught the attention of the whole world. Bolivia’s forest fires have an intimate connection to those that spread across the Brazilian Amazon. This connection is not based on uncontrollable wildfires spreading across borders, which is what the Bolivian government claimed in a first defensive reaction; but rather, that both countries have a similar agenda: large-scale expansion of arable land and grasslands for extensive livestock and monocrop agriculture. Rather than a natural phenomenon or accidental event, this is a technique for frontier expansion facilitated by the state and carried out by landed elites. In Bolivia’s fertile lowlands of Santa Cruz, fires spread across 3.6 million ha, representing nearly 10% of the department’s total land area and a five-fold increase in forest fires compared to the previous year (TIERRA 2019). Over half (52.7%) of the area burned was public land, while the rest (47.3%) was private
Extractive dynamics of agrarian change 57 property (individual and collective), including agribusiness landholdings, Indigenous territories (Tierras Comunitarias de Origen, TCO), communitarian properties, and medium-to-small-scale landholdings (TIERRA 2019). While forest fires are nothing new in countries that practice slash and burn, known as chaqueo in Bolivia, these fires indeed represent an anomaly and are related to the intensification of human activity expanding into new frontiers to incorporate new lands for agricultural and livestock activities as well as to re-sprout the pastures of the cattle ranches. Added to this was the lack of precautionary measures for fire management, which led to fires that became uncontrollable. A new era of biodiesel development? At the end of 2018, agribusiness representatives and the government launched the beginning of the so-called “biodiesel era”. The public–private investment committed amounts to some US$2 billion, estimating that it would take at least 250,000 ha of new land for this project to start and be profitable. Fuelled by discourses of “green energy” and sustainability, the biofuel programme seeks to increase soybean and sugar cane production for biodiesel and anhydrous ethanol, respectively. Agribusiness lobbyist pressed the government to legalize new varieties of GM crops, while also pushing for flexibility of the norms and regulations on deforestation and burning, something that was granted in the period 2017–2019 (Law 741 and DS 3973 considered as “legal doors” for forest clearing and burning). Furthermore, the business portfolio establishes that the state will be the main buyer of the biofuel for use as an additive and partial substitute for imported diesel. Deregulations on deforestation combined with state discourses to expand the agricultural frontier for biofuels provided favourable conditions for the use of wildfires to open up new areas without consequences. The state’s acceptance of biofuels generated great expectations among the landowners of the Chiquitanía interested in enabling more farmland. This is in addition to the new trade deal with China to export beef which certainly provided incentives to open up new lands for livestock grazing. After the Morales government was replaced amid a political scandal with claims of electoral fraud, the transitional government of Jeanine Añez resumed the biofuel agenda in early 2020. The Añez government accelerated state support to incentivize private investment and ensure the state purchase of ethanol and future biodiesel production through state contracts that guarantee purchases at subsidized prices. However, this intention was aborted and uncertainty looms large. The fall in the international price of oil below US$30 a barrel and the uncertainty of the energy market in 2020 have jeopardized the viability of the biofuels project. Government authorities announced that they would ratify the state ethanol purchase agreements, and although they announced meetings with the sugar cane sector, they have not made public the terms on which the documents were signed, so there is no certainty about the amount of state purchase committed and the price to pay.
58 Ben M. McKay and Gonzalo Colque The push for GMOs On 7 May 2020, Supreme Decree 4232 was approved with the intention of accelerating the adoption of five GM crops: soybeans, corn, wheat, sugar cane, and cotton. This is in response to the demands from the agribusiness lobby for the use of biotechnology, which has been an ongoing negotiation with the state. In 2019, Evo Morales passed Supreme Decree 3874 in order to expedite the legalization of two new transgenic soybean seeds: HB4 and intact. This push to legalize GM crops has been combined with the liberalization of agro-exports, which occurred in two recent moments: the approval of Supreme Decree No. 3920 in 2019, which extends to 60% soybean free export quotas and in 2020, Jeanine Añez decreed the total liberalization of exports. One of the arguments used to expedite the adoption of the GM crops was that it would help combat the looming food crisis that the COVID-19 pandemic and quarantine would cause. The agribusiness lobby returned to the initial discourse of being the guarantors of food security and sovereignty and that in order to fulfil this social responsibility, they needed new state facilities to expand cultivation areas. The Reactivation Plan for the Agricultural Sector for COVID-19, presented by the National Agricultural Confederation (CONFEAGRO) to the government, includes a credit line of 500 million dollars for five years, a two years grace period and a 4% annual interest rate (CONFEAGRO 2020). This request has not gone uncontested by various sectors of civil society, including environmental activists, small producers, defenders of sustainable agriculture, peasants from the highlands, Indigenous communities of the Amazon, and lowlands among others. Arguments put forth by the agribusiness sector have been discredited and it has been made clear that this sector, in alliance with the state, is attempting to take advantage of the pandemic to advance its agenda to expand the agro-extractive frontier (see Mamani 2020). Quickly reviewing the list of requested crops, it can be deduced that the main interest is not to produce food to combat the food crisis but to expand the production of soybeans that are almost entirely exported to the international market, in addition to legalizing the planting of yellow transgenic corn which is mostly used to feed livestock. The new push to expand GM crops represents ongoing conflict in society that also casts doubt on whether the soy complex is economically viable or is supported by extra-economic mechanisms to accumulate profits. One important issue concerns the yields of transgenic soybeans that are not significantly higher than conventional cultivation. Soybean yields have been volatile, with average yields in 2019 amounting to just 1.91 tons per hectare which represents the lowest record since the legal planting of this GM crop in Bolivia began. After nearly 15 years of using GM seeds, yield increases have been trivial at just 16% above pre-2005 levels (Colque 2020). When considered with production costs, farmers are actually worse off than before. From 2002 to 2014, costs of production for soybean farmers increased from US$263 to US$475 per hectare, and their dependence on a single crop has rendered them net-food buyers and subject to volatile and highly processed foodstuffs (McKay 2020).
Extractive dynamics of agrarian change 59 This type of extractive agriculture is further accentuated by the widespread practice of the agribusiness and landholding elite to capture state support in their favour. Bolivia’s agribusiness sector is not only extractive in character and form, it is highly dependent on extra-economic privileges, especially at times when agro-commodity prices are low. The privileges of the sector are demonstrated by the aforementioned events. The forest fires, in addition to the disastrous ecological implications, reflect the relaxation of environmental regulations achieved through state–capital alliances and the exercise of the power and influence of agribusiness over the decisions of the national government. Biofuel expansion projects, in addition to reinforcing the agro-extractivist model at the expense of the rural (peasant) majority, involve state purchases at prices above the market. In light of these facts, agrarian extractivism does not resemble the narrative of the private sector – that it is a productive sector that generates multiple economic and social benefits in a sustainable way. Finally, the recent attempts to spread GM crops in Bolivia not only further threaten small-scale, peasant farming but also the rich diversity of traditional crops grown throughout the country. The widespread use of GM crops will undoubtedly render the country more dependent on food imports as GM monocultures destined for export replace traditional crops for domestic consumption.
Conclusion This chapter has revealed the socio-economic, political, and socio-ecological implications of the soy complex and Bolivia’s agricultural model more broadly. It is clear that Bolivia’s so-called grano de oro (golden grain) does very little to contribute to the country’s food security, let alone food sovereignty. In fact, the vast land-use changes from traditional to “industrial” crops has rendered the country more dependent on food imports to meet its domestic demand, while the corporate control over its most important crop has resulted in a loss of autonomy among the majority of rural dwellers, some of whom have become excluded and others caught in relations of debt and dependency. Rather than industrializing the countryside, this model based on capital-intensive large-scale GM monocultures for export is extractive in character and form and is better characterized as agrarian extractivism. As a concept, we contend that agrarian extractivism is both politically and analytically useful for understanding new dynamics of agrarian change brought on by this type of development model. It directly challenges the notion of agro-industrialization by exposing the lack of industrial linkages, employment generation, sustainability, and benefits for the domestic economy and society. Industrial inputs are imported and controlled by a market oligopoly, circulate through Bolivian soil, and extracted in commodity form for export. Bolivia’s soy complex resembles an extractive enclave, socially and sectorally disarticulated from the rest of the economy with production destined for export markets, while eroding the very material and productive base from which it depends. While dominant discourses from the state and classes of capital attempt to legitimize and justify this model, agrarian extractivism as a concept puts the
60 Ben M. McKay and Gonzalo Colque extractive character to the forefront of the discussion. In the current context of an extractivist development model which encroaches into Indigenous territories, excludes smallholders and leads to widespread deforestation, uniting against all forms of extractivism provides an all-encompassing injustice frame. While natural resource extraction in Bolivia is certainly not new, the way in which the state has continued to expand extractive frontiers reinforces existing forms of socio-economic and ecological exploitation. Whether mineral, hydrocarbon, or agricultural extraction, rural dwellers and classes of labour are faced with deteriorating livelihood options. Converging against such forms of extractivism may facilitate political alliances across various social groups affected by extractivism. This chapter hopes to contribute both analytically and politically to the ongoing debates in Bolivia by analysing the extractive dynamics of agrarian change in Bolivia and challenging discourses of agricultural industrialization and rural development.
Notes 1 El noble cultivo de la soya juega un rol trascendental en la cadena (clúster) de oleaginosas del país al ser su producción, el eslabón principal que da inicio a una serie de otras actividades o eslabones de generación de valor; bajo un sistema de producción sustentable, esto es, conómicamente viable, medioambientalmente sostenible y socialmente responsable, aportando a la soberanía alimentaria boliviana con la producción de proteína de origen vegetal que se transformará en alimento para los humanos, proteína animal y fuente de energía renovable, al margen de generar ingentes divisas por las exportaciones con valor agregado, impuestos y empleos de calidad. 2 Though legalized in 2005, GM soybeans were entering illegally years prior from neighbouring Argentina and Brazil. 3 For an analysis into the politics behind such a strategy and the dynamics of state–society–capital relations in Bolivia see McKay (McKay 2018b, 2020).
References ACF-IN. 2009. Santa Cruz: Plan departamental de contingencias para desastres naturales, proyecto Dipecho-Rio Grande – Mision Bolivia-Paraguay. Santa Cruz: ACF-IN. Acosta, A. 2013. Extractivism and neo extractivism: two sides of the same curse. In M. Lang and D. Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Amsterdam: Transnational Institute, 61–86. AEMP. 2013. Estudio Mercado Del Grano de Soya. La Paz: Autoridad de Fiscalizacion y Control Social de Empresas. Almaraz, A. 2014. Las luchas indígenas y campesinas en los dos ciclos de la Reforma Agraria. In Fundación TIERRA (ed.), Recientes Transformaciones Agrarias en Bolivia: Memoria. La Paz: Fundación TIERRA, 49–57. Alonso-Fradejas, A. 2015. Anything but a story foretold: multiple politics of resistance to the agrarian extractivist project in Guatemala. The Journal of Peasant Studies, 42(3–4), 489–515. Alves, F. 2006. Por que morrem os cortadores de cana? Saude e Sociedade, 15(3), 90–98. ANAPO. 2014. Soya permite lograr la soberanía aliementaria del país. ANAPO Noticias, 105.
Extractive dynamics of agrarian change 61 ANAPO. 2015. Memoria Anual 2015. Santa Cruz: Asociacion de Productores de Oleaginosas y Trigo. ANAPO. 2019. Memoria Anual 2019. Santa Cruz: Asociación de Productores de Oleadinosas y Trigo. BID. 2014. Perfil de risego de desastres para Bolivia ante inundaciones y deslizamientos en cuencas seleccionadas. Washington, DC: Banco Interamericano de Desarrollo. Bradshaw, C.J.A., N.S. Sodhi, K.S.H. Peh, and B.W. Brook. 2007. Global evidence that deforestation amplifies flood risk and severity in the developing world. Global Change Biology, 13(11), 2379–95. Brasil, R. 2011. Brazil’ s ethanol in the world: socio-environmental impacts of export sugarcane companies. Evaluation, 20. 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(5), 1017. Catacora-Vargas, G., P. Galeano, S.Z. Agapito-Tenfen, D. Aranda, T. Palau, and R. Onofre Nodari. 2012. Soybean Production in the Southern Cone of the Americas : Update on Land and Pesticide Use. Norway: GenOk – Centre for Biosafety. Clapp, J. 2018. Mega-mergers on the menu: corporate concentration and the politics of sustainability in the global food system. Global Environmental Politics, 18(2), 12–33. Colque, G. 2020. Magras cosechas de soya transgénica. La Paz: Fundación TIERRA [online]. Colque, G., E. Tinta, and E. Sanjinés. 2016. Segunda Reforma Agraria: Una historia que incomoda. La Paz: TIERRA. CONFEAGRO. 2020. Propuesta de Reactivacion del Sector Agropecuario Nacional por el COVID 19. Santa Cruz: Confederacion Agropecuaria Nacional. Cuéllar, S., A. Rodríguez, J. Arroyo, S. Espinoza, and D. Larrea. 2012. Mapa de deforestación de las tierras bajas y los yungas de Bolivia 2000–2005–2010: Proyección Sistema de Coordenadas Geográficas, Datum WGS84. Bolivia: Santa Cruz de la Sierra. EM-DAT. 2016. Emergency Events Database [online]. Available at: https://www.cred.be/ projects/EM-DAT. Fearnside, P.M. 2001. Soybean cultivation as a threat to the environment in Brazil. Environmental Conservation, 28(1), 23–38. Fine, B., and A. Saad-Filho. 2004. Marx’s Capital. Fourth edition. Lond and Sterling, VA: Pluto Press. Goodman, D., B. Sorj, and J. Wilkinson. 1987. From Farming to Biotechnology: A Theory of Agro-Industrial Development. Oxford, UK: Basil Blackwell. Gudynas, E. 2013. Extracciones, extractivismos y extrahecciones. Observatorio del Desarollo, 18, 1–18. Gudynas, E. 2015. Extractivismos: Ecología, economía y política de un modo de entender el desarollo y la Naturaleza. Cochabamba: CEDIB. Harvey, D. 2006. The Limits Capital. London: Verso. Hecht, S.B. 2005. Soybeans, development and conservation on the Amazon frontier. Development and Change, 36(2), 375–404. Heredia Garcia, H. 2014. Gobierno so abre a construer la agenda del millon de ha. El Deber [online]. Available at: www.eldeber.com.bo/economia/gobierno-abre-construir-agendadel.html (Accessed 10 January 2015). IBCE. 2014. Soya: Su Importancia Como Cadena de Valor Agroproductiva En Bolivia. Volume 227. Santa Cruz: Instituto Boliviano de Comercio Exterior. INE. 2016. Comercio Exterior. La Paz: INE.
62 Ben M. McKay and Gonzalo Colque INE. 2020. Instituto Nacional de Estadística [online]. Available at: https://www.ine. gob.bo/. INIAF. 2015. INIAF Semillas Santa Cruz, Solicitude de Importacion Recepcionadas. Santa Cruz: INIAF. Kay, C., and M. Urioste. 2007. Bolivia’s Unfinished agrarian reform: rural poverty and development policies. In A.H. Akram-Lodhi, S.M. Borras, and C. Kay (eds.), Land, Poverty and Livelihoods in the Era of Globalization: Perspectives from Developing and Transition Countries. London and New York: Routledge, 41–79. Li, T.M. 2009. To make live or let die? Rural dispossession and the protection of surplus populations. Antipode, 41(S1), 66–93. Malhi, Y., J. Timmons Roberts, R.A. Betts, T.J. Killeen, W. Li, and C. Nobre. 2008. Climate change, deforestation, and the fate of the Amazon. Science, 319(5860), 169–72. Mamani, M.I. 2020. Aprovecharse del pánico: la táctica de los agroindustriales para exigir más transgénicos. La Paz: Fundación TIERRA [online]. Marx, K. 1976. Capital: A Critique of Political Economy Volume One. Middlesex, England: Penguin Books. Marx, K. 1981. Capital: A Critique of Political Economy. Volume 3. London: Penguin Books. McGrath, S. 2013. Fuelling global production networks with slave labour?: migrant sugar cane workers in the Brazilian ethanol GPN. Geoforum, 44, 32–43. McKay, B.M. 2018a. The politics of convergence in Bolivia: social movements and the state. Third World Quarterly, 39(7), 1247–69. McKay, B.M. 2018b. The politics of agrarian change in Bolivia’s soy complex. Journal of Agrarian Change, 18(2), 406–24. McKay, B.M. 2018c. Control grabbing and value-chain agriculture: BRICS, MICs and Bolivia’s soy complex. Globalizations, 15(1), 74–91. McKay, B.M. 2020. The Political Economy of Agrarian Extractivism: Lessons from Bolivia. Blackpoint and Winnipeg: Fernwood Publishing. McKay, B.M., and G. Colque. 2016. Bolivia’s soy complex: the development of ‘productive exclusion’. Journal of Peasant Studies, 43(2), 583–610. Mekonnen, M.M., and A.Y. Hoekstra. 2011. The green, blue and grey water footprint of crops and derived crop products. Hydrology and Earth System Sciences, 15(5), 1577–600. Müller, R., T. Pistorius, S. Rohde, G. Gerold, and P. Pacheco. 2013. Policy options to reduce deforestation based on a systematic analysis of drivers and agents in lowland Bolivia. Land Use Policy, 30(1), 895–907. Müller, R., D.M. Larrea-Alcázar, S. Cuéllar, and S. Espinoza. 2014. Causas directas de la deforestación reciente (2000–2010) y modelado de dos escenarios futuros en las tierras bajas de Bolivia lowlands and modeling of future scenarios. Ecología en Bolivia, 49(1), 20–34. Müller, R., P. Pacheco, and J.C. Montero. 2014. El contexto de la deforestación y degradación de los bosques en Bolivia. Causas, actores e instituciones. Pengue, W.A. 2005. Transgenic crops in Argentina: the ecological and social debt. Bulletin of Science, Technology & Society, 25(4), 314–22. SENASAG. 2014. Database: Servicio Nacional de Sanidad Agropecuaria e Inocuidad Alimentaria [online]. Available at: https://www.developmentaid.org/#!/donors/ view/143406/senasag. TIERRA. 2019. Fuego en Santa Cruz. La Paz: Fundación TIERRA.
Extractive dynamics of agrarian change 63 Tilzey, M. 2019. Authoritarian populism and neo-extractivism in Bolivia and Ecuador: the unresolved agrarian question and the prospects for food sovereignty as counter-hegem ony. Journal of Peasant Studies, 46(3), 626–52. UN-REDD. 2010. National Programme Document. Bolivia: UN-REDD. Urioste, M. 2011. Concentracion y extranjerizacion de la tierra en Bolivia. La Paz: Fundación TIERRA. Urioste, M. 2012. Concentration and ‘foreignisation’ of land in Bolivia. Canadian Journal of Development Studies/Revue canadienne d’études du développement, 33(4), 439–57. Vadillo, A., J.M. Salgado, and S.N. Muiba. 2013. Gobernanza de los recursos naturales en Lomerío. La Paz: Fundacion TIERRA. Vicepresidente. 2012. Gobierno y CAO acuerdan ampliar la frontera agrícola para guarantizar soberanía alimentaria [online]. Available at: www.vicepresidencia.gob.bo/Gob ierno-y-CAO-acuerdan-ampliar-la (Accessed 16 August 2018). Webber, J.R. 2017. Evo Morales, transformismo, and the consolidation of agrarian capitalism in Bolivia. Journal of Agrarian Change, 17(2), 330–47. Weis, T. 2010. The accelerating biophysical contradictions of industrial capitalist agriculture. Journal of Agrarian Change, 10(3), 315–41. Weis, T. 2013. The Ecological Hoofprint: The Global Burden of Industrial Livestock. London and New York: Zed Books.
3
Agrarian extractivism in the Brazilian Cerrado Sérgio Sauer and Karla R.A. Oliveira
Introduction During the days of 20–23 November 2019, national news networks announced the deployment of “Operation Far West”, carried out by the Federal Police in a number of cities across the State of Bahia and in the Federal District. The police operation removed the presiding judge of the Court of Justice of Bahia and took five judges into custody. An investigation undertaken by the Federal Public Ministry alleged crimes of corruption, money laundering, and criminal organization. These were all related to “grilagem” – land grabbing through a scheme of defrauding documents – revealed in western Bahia presumed to involve judges, lawyers, and large-scale landowners (Camargo 2019). The scheme was built on the sale of judicial rulings that would legalize the grabbing of public and communal land in the agricultural frontier region known as the Matopiba. An acronym derived from the states of Maranhão, Tocantins, Piauí, and Bahia, this region has been seen in recent years as the “newest” agricultural frontier characterized by global trends at the intersection of farmland and finance (Mançano Fernandes 2019). According to the investigations, court rulings had legalized an estimated 360,000 hectares (ha) but the rulings-for-sale scheme involved more than 800,000 disputed ha (Min. Og Fernandes 2019). It became yet another story of illegal appropriation of land (land grabbing for speculative purposes) along a frontier that had been transformed for large-scale soy production, indicative of how agrarian extractivism has proceeded in Brazil. Despite land grabbing being a fundamental historical component, agrarian extractivism cannot be reduced to illegal appropriations of public and/or communal land and its speculative perspective. As part of the so-called agrarian question, it is rather a structural process of “accumulation by dispossession” (Harvey 2003) that includes diverse forms of (legal and illegal) appropriation of natural resources beyond land, such as forests, water, and minerals (underground resources) with social consequences. Brazilian agrarian extractivism is formed by a combination of investments in large-scale, capital-intensive monocrop production for export in very large estates and speculative processes (financialization) (Delgado 2013). It relies on support (public resources) and collusion (loosening of laws) from the
Agrarian extractivism in the Cerrado 65 state (Martins 2012), which has driven the expansion of agricultural frontiers, transforming land use, and deepening the appropriation of nature. Contemporary agrarian extractivism in Brazil emerged from the implementation of the Green Revolution in the middle of the 20th century. It has been based on land concentration, speculation, and “grilagem” (land grabbing), and it is not reduced to its productive dimension. Agrarian extractivism is closely related to financial speculation, especially after the 2008 economic crisis. Historically, it has been rooted in power pacts between the agrarian elites (large landowners), agribusiness, and the state (Delgado 2013). Even based on neoliberal narratives (Mançano Fernandes 2019), it enforces conservative political stands, and has resulted in poverty and inequality (Martins 2012). In recent decades, the Cerrado biome has become a driving economic force, gaining international attention as the newest and largest agricultural frontier (The Economist 2010). The region accounts for nearly a quarter of the country’s national territory and more than a half of Brazilian soy production (IBGE 2017). Despite its relatively acidic soil and underdeveloped infrastructure, the Cerrado in general and the Matopiba in particular, have become a major destination for largescale agribusiness enterprises (Favareto 2019), deepening processes of agrarian extractivism by opening new areas dedicated to crops bound for export (Flexor and Leite 2017). Furthermore, as places of “the degradation of the other” (Martins 2012) and where forms of accumulation by dispossession are rapidly unfolding, the Matopiba frontier is one of the places where agrarian extractivism is most prominent in Brazil. As it is common with land targeted for capitalist expansion, the Matopiba is marked by high levels of national and foreign investments combined with poverty, inequality, and low levels of environmental protection (Favareto 2019). Against the advancement of the agricultural frontier, rural communities are resisting and defending their territories and commons, especially the right of access to, and control over, quality water. It is a conflictual relation between extractive capitalist development and popular resistance (Veltmeyer and Zayago Lau 2019), in which the main actors are Indigenous people and traditional communities in the Matopiba region (Mançano Fernandes 2019). The trend towards financialization of land has compromised long-standing national commitments to agrarian reform and land demarcation, which stalled since Rousseff’s government (2012–2016) and the brief administration of Temer (2016–2018) and was completely abandoned by the Bolsonaro administration in 2019 (Sauer et al. 2019a). The political and economic crises that have taken hold of the country over the past decade have been leveraged as evidence for the need to loosen restrictions on foreign investment. As such, these fragile years have given momentum to processes of financialization and dispossession (Harvey 2003) of the Brazilian countryside (Flexor and Leite 2017), specifically along the Matopiba frontier (Favareto 2019). Official discourse has portrayed agribusiness as the single-most stable sector in an erratic economy that has otherwise thwarted international investment (Sauer 2018; Leite et al. 2019). The notion of “agribusiness” has had a different meaning
66 Sérgio Sauer and Karla R.A. Oliveira in Brazil than in the United States, for instance, where the concept was forged in the 1960s (Gryszpan 2012). The concept started being used in the middle of 1990s in Brazil, in an attempt to overcome the history of agricultural modernization associated with land concentration, land speculation, agrarian conflicts, and low productivity. Agribusiness was then linked to the discourse of modernization, technical efficiency – production based on private investments in technology, associated to the reductionist concept of “industrialization of agriculture” – and a profitable sector (Gryszpan 2012). According to Delgado (2013, 62), Brazilian agribusiness is a “virtual pact of political economy”, with crucial state and public support, which perpetuates the power of large landowners. Throughout the Temer administration, and in the years preceding it, praise for export-oriented agribusiness remained a pillar of the economic recovery rhetoric characterizing agrarian extractivism in Brazil. Processes of land concentration, deforestation, and agribusiness expansion were present throughout the administrations of Lula (2003–2010) and Rousseff (2011–2016); these trends drove rural populations from the countryside and incited land conflict. These trends were predicted to – and did, in fact – accelerate during the Temer presidency (see Sauer 2018). Under the Bolsonaro administration (2019–2022), the outlook remains grim – more tension and conflict in the Brazilian countryside is expected, as the administration has openly offered its political support to large landowners and has acted to legalize land grabs and foreign investments in land (Sauer et al. 2019a). This chapter will provide an analysis of how agrarian extractivism has been promoted along agricultural frontiers and its implications for traditional territories and conservation areas, specifically in the Matopiba region of the Brazilian Cerrado. The widening of the agricultural frontier will be analysed as part of the Cerrado as a “sacrificed zone” for the expansion of soy production and cattle ranching. Finally, the case of the Parnaíba River Spring National Park will be discussed as a representative case for explaining socio-environmental conflicts caused by the expansion of the agricultural frontier, and social resistance related to the appropriation of land and water in the Matopiba.
Agrarian extractivism: the appropriation of land and nature in the Brazilian agricultural frontier In Brazil, a consolidated policy of economic expansion targeting agriculture and livestock production appeared first in the 1940s and again with renewed momentum in the wake of the dictatorial military regime of 1964. According to official discourse, this policy had two key aims: (i) to occupy empty spaces throughout the national territory (especially those regions with cheap land) and (ii) to minimize land tenure conflicts in the south, southeast, and northeast regions of the country (Delgado 2013). The so-called March to the West encouraged the occupation of the central region of Brazil and guided different migration flows, focusing on the occupation of public and communal lands and exploration of areas of the Cerrado biome and the Amazon (Martins 1996).
Agrarian extractivism in the Cerrado 67 The implementation of green revolution technologies in the mid-south of the country beginning in the 1960s – part of the military dictatorship’s plan to industrialize or modernize agriculture – did not result in significant shifts in the expansion logic of the agricultural frontiers contemplated in the March to the West (Sauer and Leite 2012). Changes to the factors of production – the adoption of intensive mechanization, chemical fertilizers, and selected seeds – modernized the large properties (latifundia) (Martins 1996) but did not change land concentration or decrease frontier expansion itself. On the contrary, the modernization of large properties with subsidized credit and state support was linked to incentives (especially tax exemptions for industry and financial companies) intended for private investments and colonization projects (Hecht 2005), which served as a basis for the deepening of agrarian extractivism and land concentration (IBGE 2017). According to Martins (1996), in the 1970s, Latin American societies – including Brazil, despite its differences with Spanish America – were shaped by the logic of agricultural frontiers. Economic growth and social relations were marked by geographic mobilization and demographic displacement, as countries were encouraged to use “unoccupied” or “insufficiently occupied” lands (Hecht 2005), demonstrating complete disregard for ongoing land use by local populations, Indigenous people, peasants, traditional communities, and others. Besides being geographically located at the margins, agricultural frontiers are crucial to keep prices low with available land for appropriation, or private investments in “insufficiently occupied” land (meaning cheap or “free” land). These were promoted with public incentives subsidizing investments in modern technologies (machinery, chemicals, storage infrastructure) that boost agricultural growth and reduce the “yield gaps” present in non-capitalized estates and peasant farming (World Bank 2010). Such processes reveal the historic basis of agrarian extractivism in Brazil. According to many authors (Svampa 2019; Gudynas 2018; Acosta 2013), extractivism and its associated concepts – such as neo-extractivism and post-extractivism, including agrarian extractivism – are a product of a historical development trajectory present throughout Latin America since its colonization. The strategy behind this extractivism has implied an economic dependence on the exploitation of natural resources, associated with a violent expansion of the frontiers and the export of its natural resources, in addition to dispossession and genocide linked to the “invention of Europe and the expansion of capital” (Svampa 2019, 16). The economic cycles imposed by this capitalist logic brought great contrasts to a region that found itself divided between the development of its productive forces based on the profitability of an export system built upon its natural resources, on one hand, and extreme poverty, inequality, and social resistances (Veltmeyer and Zayago Lau 2019) on the other. This is how large-scale appropriation began, characterized by the export of raw materials and a great level of dependence on the global economy (Svampa 2019). As a result, extractivism does not merely define mining but is instead present in any kind of exploitation and appropriation of nature, including agriculture (Acosta 2013).
68 Sérgio Sauer and Karla R.A. Oliveira However, since the beginning of the 21st century, with a global increase of commodity prices – the so-called “commodities boom” – the region has witnessed the rise of new possibilities in which extractive capital is consolidating expansion and the appropriation of natural resources, especially with regards to land. This new phase of extractivism is referred to as neo-extractivism (Svampa 2019; Acosta 2013), especially in light of profit distribution through social policies. The model has been aided by planning and management, the intensive use of technologies, and the engagement of “unused” lands and other lands already open to agricultural and livestock production (Hecht 2005). These elements are further supported by governmental incentives aimed to expand frontiers and increase Brazilian competitiveness in the international agriculture commodities market (Sauer 2018). In this new phase of extractivism, some “ruptures and continuities” are emerging (Svampa 2019). These continuities are due to the colonial memory of extractivism in Latin American countries and an imaginary surrounding of an endless abundance of natural resources in the region, driving the expansion of the frontiers (Martins 2012). Ruptures are the consequence of a deepening land degradation, resulting in new forms of resistances and political and social disputes involving land, territory, and natural resources (Veltmeyer and Zayago Lau 2019). The hegemonic narratives supporting extractivism – and agrarian extractivism – are based in a series of arguments and political acts (Almeida 2011). They are searching to build and sustain public support (state subsidies and legal backing) that legitimizes processes of accumulation associated with extractive activities (Gudynas 2019). This extends to and includes support for agribusiness (mainly for large monocrops and cattle ranching), and for appropriation of land and natural resources, including illegal appropriations,1 based on the narrative that increasing production for export is the only solution (Svampa 2019). As argued by Almeida (2011, 102), such justifying apparatuses of political discourses, media support, academic researches, and judicial mechanisms could be understood as agro-strategies that act to promote the expansion of agricultural frontiers as “the only solution to food system crisis and all the problems related to food supply chains”. They constitute an important part of a broader political agenda – encompassing multilateral agencies and media – and are based on a narrative that is intimately linked to food-supply needs. These include discourses that support a loosening of legal mechanisms and state-supplied financial support to agribusiness. For Almeida (2011) such agro-strategies are formed by complex coalitions with shared interests that can unify conservative and populist governments, financial corporations, and media interests. Such alignment can lead to broad and diffuse readings of these issues, highlighting the effects of land concentration, exalting the possibilities of the commodities’ market, and minimizing the effects of climate change. Agro-strategies, as a combination of narratives and political actions, also encompass a set of initiatives aimed at removing judicial and formal obstacles to production while facilitating large land expanses to industrial interests (Almeida 2011). The political pact that encompasses this strategy became clear and concrete
Agrarian extractivism in the Cerrado 69 in the proposed changes to the Forest Code in 2012, especially the proposal to reduce the Legal Reserve (protected forest area) in the Amazon, making more land available for agriculture and pasture (Sauer and França 2012). Marked by ultra-liberal economic narratives and an exacerbated political and moral conservatism, the Bolsonaro administration (2019–2022) has further deepened this political pact while offering his explicit support to projects characterized by agrarian extractivism. To this end, he issued Provisional Measure 910, which regularized illegal occupations of public land and opened the possibility that similar occupations of up to 2,500 ha could be titled anywhere in Brazil (Sauer et al. 2019a). He has publicly encouraged the expansion of the agricultural frontier and an increase in soy and cattle production on Indigenous and traditional lands (commons) as well as in protected areas. His minister of agriculture – formerly president of the Parliamentary Agricultural Front more commonly known as the Rural Caucus – has unconditionally supported congressional bills aimed to deregulate foreign investment in land (Castro et al. 2017). Analysing recent processes of accumulation, Delgado (2013) defined political and economic dynamics of agrarian extractivism using the concept of “agribusiness economy”. This is based on a power pact with characteristics similar to those of an agro-strategy (Almeida 2011) and extractivism (Acosta 2013). According to Delgado (2013, 63), the agribusiness economy extrapolates upon a pure economic strategy to “build ideologically a hegemony from the top, bringing together large landholdings, agro-industrial chains closely linked to the foreign sector, and the bureaucracies of the State”. It is enabling “the accumulation of capital under the scope of these sectors melded by public money” (Delgado 2013, 62), based on a discourse of “environmental governance” as “the condition of possibility for neoextractivism” (Baletti 2014, 7) or agrarian extractivism (McKay 2017; Veltmeyer and Zayago Lau 2019). Agrarian extractivism in Brazil, combining productive and unproductive investments, is rooted “in the capture and overexploitation of natural comparative advantages” or in ground rent (Delgado 2013, 64), appropriating land and nature. Its extractive logic is based on production but also speculation, resulting in “accumulation by dispossession” (Harvey 2003). Agrarian extractivism includes investments by national and foreign agricultural companies for exports, the denationalization of the agro-industrial sector and land grabbing (Borras et al. 2012), and appropriating public and communal land, resulting in land conflicts and social inequality (Sauer and Castro 2020). After 1998, the agrarian extractivism represented by the agribusiness economy became a central feature of Brazil’s economic logic. Especially in terms of the trade balance (export), it represented a strategic pact between “the large agroindustrial capital, the system of public credit to agriculture and agro-industry, the property of land and the State” (Carvalho 2013, 34). The expansion of extractive capital was accelerated by an increase in the price of commodities on global markets and available “cheap” land, resulting in the “predominance of invested capital in land and natural resources, such as minerals and metals, hydrocarbons and agro-fuels”. This process of accumulation was associated with a “progressive
70 Sérgio Sauer and Karla R.A. Oliveira political cycle” (Veltmeyer and Zayago Lau 2019, 49), creating social-assistance policies that would give a novel character to agrarian extractivism (Sauer 2019). According to Delgado (2013), such “agribusiness economy”, based on a political power pact, employs a set of ideological devices, such as (i) an active agribusiness bench (the Rural Caucus) at the National Congress to support legal arrangements; (ii) an active agribusiness class association to represent interests and promote accumulation; (iii) a state bureaucracy aimed at expanding public credit and governmental support; (iv) passivity of public regulatory institutions, including a loosening of laws and the regularization of land grabbing; (v) a strong co-opting of academic circles; and (vi) collusion and support from the national media.2 Agrarian extractivism has led to a set of socio-environmental conflicts, particularly conflicts over land, territory, water, and other natural resources (Oliveira and Sauer 2020). As the agricultural frontier continues to expand, it encroaches onto the land and territory of riverside communities and other traditional populations (for instance, Maroon communities or quilombolas), Indigenous peoples and familybased rural producers who settled in the Amazon and Cerrado many decades ago (Martins 1996). The next section analyses processes of agrarian extractivism in one region of the Cerrado biome – Matopiba.
Agrarian extractivism, investments, and inequality in the Matopiba Large-scale agribusiness has expanded at break-neck speed across the Cerrado, transforming Brazil into the second-largest soy producer in the world. In a context of growing foreign investment and simultaneous economic depression, the socalled “miracle of the Cerrado” (The Economist 2010) was lauded and exported as a successful case of agricultural modernization driven by national, foreign, and transnational investments in land, technology, and infrastructure (Delgado 2013), albeit heavily supported by public resources (Sauer 2019, 2018). The transformation of the Cerrado into one of the most agriculturally productive landscapes in South America dates back to the 1960s, when a technical package accompanying the Green Revolution was implemented in Brazil (Martins 2012). While the financialization of the Brazilian countryside is a rather recent phenomenon, it is an outgrowth of long-standing institutional narratives calling for the industrialization, modernization, and internationalization of the agricultural sector (Delgado 2013). The rapid conversion of the Cerrado biome into an agribusiness stronghold fits as one of the most recent chapters in a large body of scholarly work on land concentration, land grabbing, agrarian extractivism, and the dispossession of the Brazilian countryside. Around 50% of the biome has been deforested by the expansion of agricultural frontiers promoted by the state over the last three or four decades (Favareto 2019). The transformation of the Cerrado, however, has been fraught with controversy (The Economist 2010), as the expansion of agribusiness enterprises has
Agrarian extractivism in the Cerrado 71 accelerated environmental degradation and the displacement of local and traditional communities (Castro et al. 2017). The lenient environmental protections in the Cerrado have had clear impacts on the landscape: after the Atlantic Forest, the Cerrado is the biggest and single-most altered biome in Brazil. Accounting for a mere 3.65% of the Cerrado – some 73,000 square kilometres – the Matopiba accounts for 62% of its deforestation. Between 2013 and 2015, the region lost 36,900 square kilometres of forest (Favareto 2019). After 1998 and particularly in the first decade of the 2000s, large-scale mechanized farmers increased their presence in the Cerrado region and with an expansion of the agricultural frontier came an increase in deforestation and land conflict. In recent years, this transformation has taken place across both rain forest (Amazon) and Cerrado biomes – especially in the Matopiba (Sauer 2018; Castro and Sauer 2020) – with severe environmental impacts (Baletti 2014). In 2015, the production of grains (mainly soy) in the Cerrado surpassed that of the southern region for the first time in the country’s history (Favareto 2019). While the Matopiba represents roughly one-third of the overall Cerrado – 730,000 of its total 2 million square kilometres – it has become an auspicious cornerstone in the rhetoric of agribusiness expansion into north-central Brazil (Pereira et al. 2017). Soy production across those four states has increased from an estimated 769,000 tons (t) in 1993 to more than 14.5 million t in 2017–2018. Soy production in the Matopiba has doubled in the last seven years and increased nearly 20-fold in the past 15 years, representing 11% of national production in 2017. Its cultivated area spans 4.3 million ha, or 12.3% of the total area cultivated with soy in 2017, but its cultivation has been concentrated in only a few municipalities (Favareto 2019). After 2010, the Matopiba became a natural target for new investments, given the lack of environmental protections and enforcement in the Cerrado, as well as the discourses of “unproductive”, “available” land in the region (resulting in several cases of land grabbing of public and communal land, as a result). In addition, the Matopiba has a strategic location in proximity to the northern port of São Luis (State of Maranhão) and other major departure ports for soy. This process of investments was further accelerated by state-led pushes to develop the region as a global agribusiness powerhouse in the context of an international commodities boom (Flexor and Leite 2017). The ongoing expansion of soy, corn, and cotton across the Matopiba gained further momentum in 2015 when the Brazilian state introduced the official Matopiba development plan for the region, based primarily on governmental incentives that supported agricultural expansion (Sauer et al. 2019a). At its unveiling, the government announced that up to 10.9 million were expected to be cultivated and come into production in the region by 2023. Beyond a new agricultural frontier, the Matopiba was becoming a territory of capitalist expansion and foreignization of the land (Pereira et al. 2017). After assuming power following the impeachment of Dilma Rousseff in 2016, Michel Temer veered sharply from the agrarian rhetoric of the Workers’ Party (PT). Land policies were deeply affected, including the extinction of the
72 Sérgio Sauer and Karla R.A. Oliveira Ministry of Agrarian Development (MDA), which was responsible for the land reform programmes. However, it kept and increased public support for agribusiness, especially credit and technical support from Brazilian Agricultural Research Corporation (Embrapa), aiding those who were investing in the Matopiba region (Sauer et al. 2019a). Among the priorities announced by the administration was the loosening of the 2010 restrictions that limited the foreign ownership of land and public policies (Leite et al. 2019). Despite having clear intentions to remove barriers to foreign ownership of land, internal resistance – notably from Blairo Maggi, billionaire soy farmer, former governor of Mato Grosso, owner of the Amaggi Group and at the time the Minister of Agriculture – meant that these efforts gained little traction (Sauer and Mészarós 2017). However, working more closely with the Rural Caucus, the political influence wielded by the agribusiness sector became more visible, making clearer Temer’s support for agrarian projects that are extractive in form and character. In 2017, with the support of Congress, Temer succeeded in passing legislation that facilitated the privatization of public plots of land illegally tenured in frontier regions, including land issued under agrarian reform (Leite et al. 2019). The ease with which this legislation could be applied was greatly bolstered by the modification of Provisional Measure 910 in 2019 by Bolsonaro, which allowed for the titling of public land holdings in all of Brazil. This legislation has also benefitted those illegally appropriating large tracks of lands (up to 2,500 ha) in the Amazonian, Cerrado, and Matopiba regions (Sauer et al. 2019a), with low or no payments at all. Aside from the large investments in land – including by foreigners – and its role in the global soybean market, the Matopiba region is characterized by high levels of poverty and social inequality. According to Favareto (2019), there are four different Matopibas and in the wake of agricultural expansion, there is more poverty and injustice than wealth and well-being in the region. In a total of 337 municipalities within Matopiba, only 45 could be classified as “rich” or as having a Gross Domestic Product (GDP) above the national average. The fact that production is concentrated in few municipalities impedes the distribution of benefits throughout the rest of the territory. The monocrops are also based on mechanized production and saving labour, generating few jobs. It is a contradictory synthesis of injunctions between the entrepreneurial and patrimonial state, the entrepreneurship of private individuals and companies (Favareto 2019), as well as old latifundiary practices (Martins 1996), demonstrating its extractive character in agrarian development in Brazil. The agribusiness opportunities present in the region, both realized and speculative, have had detrimental impacts on the most vulnerable sectors of the Brazilian countryside: traditional communities and quilombolas (Maroons), small-scale farmers, and Indigenous peoples (Oliveira 2018). These groups have suffered direct violence, in addition to threats, forced evictions, the contamination of water sources, and the expropriation of common lands. In many cases, loss of land is the result of the falsification of documents and land grabbing (Favareto 2019), as it is the case of land grabbing in the State of Bahia already mentioned (Camargo 2019; Mançano Fernandes 2019).
Agrarian extractivism in the Cerrado 73 The historical narrative that has depicted the Cerrado and Matopiba as “empty” or void of human activity has been used to legitimize the exclusion of these communities from national development projects and the advancement of policies intended to increase the legibility of the land (Martins 2012; Hecht 2005). State policies prioritizing the conservation of the Amazon biome while lowering environmental protection in the Cerrado have made this region, and particularly the Matopiba, a territory for large-scale agribusiness development in recent decades. Contrary to Amazonian states, where private farmers are obligated to maintain 80% of their properties as natural forest, in the states that comprise the Matopiba, a mere 20% of the Cerrado biome must be conserved on private property (Oliveira and Hecht 2016; Sauer and França 2012). In this context, political projects – especially the relaxation of norms and laws, a foundational element of the foreignization, financialization, and extractive character of land use and labour conditions – have resulted in the denationalization of national territory, reducing the capacity of regulation and weakening sovereignty (Sassen 2014). Thus, in contrast to the colonial period – or even more recent processes of expropriation – there have been new and more flexible governmental measures intended to legalize expropriation and land grabbing (Sauer et al. 2019a). Historically, land regimes – including laws, various forms of land appropriation and use, land rights, etc. – were based on a national logic that reflected the importance of a nation and its legal framework (Sauer and Castro 2020). This process described here, by contrast, represents a novelty in processes of capitalism as it currently operates (Castro et al. 2017). In the case of the Brazilian Cerrado, and the Matopiba more specifically, there are clear indications of state-driven capitalist processes of internationalization or denationalization of land and natural resources (Sassen 2014), making trends of agrarian extractivism more evident. Nevertheless, the appropriation of natural resources occurring in the Matopiba has been met with processes of resistance originating in traditional communities living in the region for generations. The revolt of the Correntina municipality – a place of high land concentration, agribusiness, and inequality3 – is one such example, demonstrating the resistance of traditional communities against agrarian extractivism in the Cerrado (Mançano Fernandes 2019; Pereira et al. 2017). On 2 November 2017, a group of approximately 1,000 people carried out direct action on an agribusiness enterprise, Fazenda Igarashi, which holds an estimated 2,500 ha in Correntina, a municipality in the State of Bahia, located in the Matopiba region. This farm used to pump water for irrigation from the Rio Arrojado, which was depleting water reserves of several rural communities that no longer had sufficient water for human and animal consumption. Tired of waiting for a solution from the responsible governmental bodies, the population plundered and destroyed the farm’s irrigation structure in order to underline how harmful the model of production had been to the environment and to the people (Mançano Fernandes 2019; Pereira et al. 2017). This act was – and continues to be – harshly criminalized by different political representations, but it succeeded in showing that the farm’s water consumption
74 Sérgio Sauer and Karla R.A. Oliveira was sufficient for supplying the entire population of Correntina. It has also received support from grassroots organizations, such as the Pastoral Commission on Land (CPT, a Catholic church organization), parliamentarians supporting popular causes and agrarian social movements. As part of the public support mechanism, a street demonstration was convened in the city, gathering about 20,000 people only a week later (Pereira et al. 2017). This revolt and demonstration denounced the social and environmental predatory logic of the productive model behind Brazilian agrarian extractivism.
Agrarian extractivism and social-environmental conflicts over land, water, and territory in the Matopiba In addition to nurturing the expansion of its agricultural frontier, since the 1960s, Brazil has also strengthened its environmental laws. Between 1974 and 1984 – known as the “destruction decade” due to the expansion of soy crops and cattle ranches – the country succeeded in establishing a number of important environmental protected areas, especially in the Amazon (Ramos 2014). While these areas were being increasingly monitored, the Cerrado was offered as a “sacrifice zone” (Oliveira and Hecht 2016) for the expansion of the agricultural frontier. This was largely due to its vast expanse and favourable conditions (ground slope) for modern technologies used on monocultures. Most of the protected areas in the Cerrado were established only in the 1990s (Eloy et al. 2016) as this region had long been a territory of investment, land acquisition, and land grabbing, an arena on which the characteristics of agrarian extractivism could be expressed. Agrarian extractivism produces negative social and ecological implications as it requires an increased consumption of natural resources (mainly land and water) that causes the degradation of ecosystems and is a significant emitter of greenhouse gases, contributing to climate change. This agro-food model, including the development of chemical agriculture and carbon-emitting fuels, releases soil carbon into the atmosphere, leading to further negative environmental consequences (McMichael 2018). This is a consequence of the Brazilian development paradigm and agrarian extractivism, broadly understood, which is based mainly on ground rent (Carvalho 2013) and land grabbing (Sauer and Borras 2016; Borras et al. 2012). Land grabbing is not a new phenomenon in Brazil. It is part of an existing (historical) pattern of land concentration, which persists even in the 21st century. Foreign investments are new mechanisms and means to maintaining this land concentration. Furthermore, the rising demand for land intensifies territorial conflicts in the same way that they reinforce the historical imbalance between social demand and supply (Sauer and Leite 2012). Land grabbing has also mainly taken place in the agricultural frontier, where water is an available resource. Since water is an – if not the most – important resource for agriculture,4 there are many conflicts related to access and distribution. The concept of “water grabbing” refers to all situations in which there is a powerful actor reallocating or taking control of the water resources for its
Agrarian extractivism in the Cerrado 75 own benefit, despite the right of local users’ and the ecosystem’s use of water. In general, those situations include capturing the decision-making power around water, which means the decision on how and for what purposes water can be used (Franco et al. 2013; Veldwish et al. 2018). As with other natural resources, water grabbing cannot be separated from the land-grabbing phenomenon. Veldwish et al. (2018, 62) state that “drawing insight from the discussion on land grabbing, we understand water grabbing as capturing control not just of water itself, but also of the power to decide how it will be used – by whom, when, for how long and for what purposes – in order to control the benefits”.5 The natural conservation policy in place in the Brazilian Cerrado does not limit the soy and cattle expansion areas. According to Eloy et al. (2016), policies for protected areas as well as the Brazilian Forest Code promote a “selection policy”, while national parks and forest reserves are generally established in areas where there is no possibility for an expansion of the agricultural frontier. On the other hand, policies to prevent forest fires are implemented in an exhaustive manner, directly affecting the productive practice of traditional agriculture. The Parnaíba River Spring National Park is an example of the policies detailed by Eloy et al. (2016) and of the socio-economic and ecological impacts of agrarian extractivism in the Matopiba region (Oliveira 2018). The location of the park is of particular ecological relevance, given that it is home to the headwaters of the Parnaíba River – one of the region’s most important rivers – and due to its vicinity to the Jalapão, an important recharge area for a number of water basins. The region is also the home of quilombola communities that have lived for nearly 200 years on these territories in symbiosis with the environment (Oliveira and Sauer 2020). Since the creation of the Parnaíba River Springs National Park, traditional communities (the commons)6 recount how their means of production and ways of living have been systematically repressed. Controlled burning, the farming of palm swaps and the pasturing of animals within protected areas were banned; the collection of wood for home use (cooking) was criminalized. Such measures have resulted in controversy and tension between traditional communities and park management with communities reporting hostility shown towards traditional ways of life and a clear preference (with no legal basis) shown for the expansion of soy (Oliveira 2018). According to some quilombola leaders, these communities have learned other ways of defending their livelihoods as they seek alliances rather than conflict with Park managers. They have changed their forms of resistance, combining the demands of their rights with actions for environmental conservation. In the past, their struggle was based on protecting themselves against the violence of the “large farmers”. Now, it relies on an understanding of their land rights, the (environmental) norms, and regulations of the Park. Thus, they have shifted their struggle and resistance from basing the arguments in the territory to using “the content of the paper” and environmental laws (Oliveira 2018, 114).
76 Sérgio Sauer and Karla R.A. Oliveira Furthermore, the self-identification or self-recognition as “traditional communities” has played an important role in their resistance. After a long process of conflicts involving the right to live on the land inside the Park and to use the biodiversity in their traditional ways of dwelling and producing, the communities started the procedures to be formally recognized as “quilombola communities” by the State. This status has traditional community guarantees, at least from the legal perspective, which protects their territory and territorial rights by the Brazilian Constitution (Oliveira 2018). The traditional management of the Cerrado vegetation is closely related to the details of its landscape. According to Eloy et al. (2019), the fire-based production systems have symbiotic relationships with the Cerrado environment. The wet sites near the paths followed by the springs (called veredas) are managed by a floodplain cutoff that provides temporary overflow protection. These areas are loci of fertile soils for food production and they are the only humid sites in the landscape. The high-altitude plains, known as chapadas (plateau areas), were traditionally used as cattle pastures (common use) and occasionally, there was a need to reduce the presence of biomass using fire (Eloy et al. 2019). So, while State-protected areas have the capacity for environmental conservation, the traditional management of landscape in the Cerrado biome offers a more holistic approach to ecological preservation and sustainability (Eloy et al. 2016). However, during the same time that this national park was created, in 2002, an intense expansion of soybean cultivation was underway in the Chapada das Mangabeiras, located just south of the park. When these plateaus were appropriated by agribusiness interests for the cultivation of soybeans, traditional communities found themselves corralled and excluded (Oliveira 2018). In 2010, the area had high rates of deforestation and authorities began to inspect a number of large farms, which eventually resulted in the reduction of the protected area in 2015 (Oliveira 2018). In the literature, these processes are treated as “Protected Areas Downgrading, Downsizing and Degazettement” (or PADDD, for short) for the conservation of biodiversity (Mascia and Pailler 2011). Downgrading refers to loosening legal restrictions on or de-regulating human activities within a protected area. Downsizing is understood as shrinking the size or extension of a protected area, resulting from a legal change in its size or dimension. Degazettement is the process of a total loss of legal protection and the elimination of a protected area. These processes are ongoing in Brazil, affecting conservation areas like the National Park closely related to the expansion of the agricultural frontier or the construction of infrastructure (WWF-Brasil 2017). Such occurrences became more common at the end of the 2000s with common justifications including the need for energy generation and building infrastructure. Beginning in 2016, the hegemonic narratives explicitly defended the expansion of agricultural frontiers for the production of export commodities (Sauer 2018). Currently, around 10% of territories under some kind of protection in Brazil are threatened by PADDD, which will result in biodiversity loss in these protected
Agrarian extractivism in the Cerrado 77 areas. Often, the reductions to these areas have taken place without consultations with civil society (WWF-Brasil 2017). The Parnaíba River Springs National Park is one clear example of “downsizing”, as its borders were altered by Law No. 13.090 in 2015. Signed off by then President Dilma Rousseff, the changes reduced and moved the borders of the park. The move came at the hands of the Rural Caucus in the National Congress, who acted in defence of large-scale landowners in the region. The total area was reduced to 76,000 ha, legalizing land grabbing (illegal occupation) and creating space for the expansion of soy in the southern part of the park while compensating for some of the lost area with an extension into traditional territories (Oliveira 2018). This process of “downsizing” did not include, at any moment, consultations with the traditional communities affected. Civil society in the area around the park knew of the changes only after the law was enacted and, according to testimony from local leaders, only economically powerful and agribusiness groups had access to the downsizing proposal and, as a result, communities were not involved in the decision-making process (Oliveira 2018). Beyond the loss of territory and communal land, the groundwater within the national park was contaminated in this process. Those living in the quilombola communities of Curupá, Brejinhos, Macacos, and Povoado do Prata – all near the park – recount that the majority of the springs that form the headwaters of the Parnaíba River either dried up or were polluted (Oliveira 2018). According to leaders from Povoado do Prata, beginning in the 2000s, the water level of the rivers bordering the communities has continued to drop. This has had consequences for traditional herding (raising free-roaming livestock on the plateaus), given that they had to take the animals increasingly long distances in search of water. Of the eight municipalities impacted by the park, there are clear economic and social inequalities that have emerged from the development processes, just as the expansion of monocultures is reflected in indexes on inequality and social productivity (Favareto 2019). Formosa do Rio Preto,7 for example, in the state of Bahia, was the sixth largest exporter of soy in Brazil in 2018 with a production of 214,000 t of grain (IBGE 2018). According to the Ministry of Social Development (MDS 2016), however, nearly 50% of the population lives in extreme poverty, offering yet another reflection of the concentrating effect of agrarian extractivism. Sauer and França (2012) assert that changes to the Brazilian Forest Code made in 2012 by the National Congress stemmed from a belief that nature represents an obstacle to development. Opposing this, the conservationist logic stated that areas for productive practices should be maintained as part of the environmental conservation measures. Attacks on conservationist measures were complemented by the use of environmental laws against traditional populations, targeting the rights and practices associated with the commons. The banning of traditional productive practices – such as fire management or common use of plateau lands – act as examples (Eloy et al. 2016). Political efforts to loosen judicial positions regarding environmental licensing also affected the Brazilian Cerrado. The application of environmental laws
78 Sérgio Sauer and Karla R.A. Oliveira disregarded numerous international conventions on climate ratified by the Brazilian government (Ramos 2014).8 Beyond the regularization of land grabbing,9 there was complete negligence of social-environmental conflicts in the process of licensing large-scale projects and activities that were effectively or potentially polluting the environment (Sauer and França 2012). Similar bills aiming to alter the extension or category of protected areas have also been placed on the agenda of the Bolsonaro administration. Since mid-2016, the country has been witnessing an unprecedented attack on protected areas. The offensive comes in large part from large-scale rural producers who have irregularly occupied these areas, together with mining companies and grabbers of public and communal lands. Potential outcomes, as noted, include the reduction of the size of these areas in addition to the level of protection currently enjoyed. In some cases, the areas will cease to be protected altogether. Currently, Congress is considering a number of bills that envision modifications to or the re-categorization of these areas (Oliveira and Sauer 2020), allowing private appropriation of nature. Land and water grabbing are the two processes that give visibility to agrarian extractivism in the agricultural frontier region of the Cerrado. Such processes are sustained by the agribusiness economy with the aim of freeing up the territory of traditional communities and Indigenous peoples, in addition to protected areas, for the large-scale production of soybeans bound for export. Land-use changes have caused environmental degradation – in particular, reducing and contaminating water supplies – and conflicts over land and territory (Oliveira and Sauer 2020). Communities are being driven from traditionally occupied territories – communal lands – and traditional practices to maintain biodiversity are being criminalized (the use of fire, for example), while the springs that feed the main rivers in the region are polluted. However, these activities have not gone uncontested, as traditional populations are mobilizing to defend their territorial rights and livelihoods (Mançano Fernandes 2019). The result is not only resistance but direct confrontations and challenges to agrarian extractivism in Brazil.
Conclusion Extractivism is not a recent phenomenon in Brazil, much as it is a well-known process across Latin America. Beginning with the colonization of the continent and the prospecting of the Brazil’s natural resources, economic activities of extractivist nature appropriated natural resources at a large scale with the goal of fostering an export market. This is a fundamental element in the ongoing occupation and appropriation of Latin America, a characteristic that is constantly renewed and revived. It was further amplified in the wake of the “commodities boom” of the 2000s, but a strong presence of the State at the time and the distribution of profits through socio-economic programmes provided a novel perspective to this logic of production. This long-standing trend of natural resource appropriation and the extraction of the socio-economic value are better characterized as agrarian extractivism than
Agrarian extractivism in the Cerrado 79 industrial agriculture as the former concept embodies an explicit exploitation of the land for agricultural commodities bound for export, with little to no benefit for local societies or economies. In Brazil, since the mid-20th century, this process has had support from and deep involvement in the agribusiness sector, which has built its power combining modern investments of capital to conservative agrarian forces, particularly political power of latifundia and “grilagem”, or land grabbing, using the state’s compliancy and complicity to appropriate public and communal land and natural resources. The “March to the West” instigated by the Brazilian government in the 1940s – and further propelled during the military dictatorship in the 1960s – treated the Cerrado as an ideal terrain to expand livestock operations and grain cultivation. The region was imagined as a “population void” with vast areas fit for mechanized agriculture, further encouraging the Cerrado to be considered as the ideal environment for agricultural frontier expansion, particularly for soy cultivation. The consequences have since been made clear: the country has become the second largest soy exporter in the world while the Cerrado is one of the most threatened biomes in Brazil with an estimated 50% of its native vegetation already deforested. The agribusiness economy is further sustained by judicial, institutional, and media apparatuses (as part of the agro-strategies) that have reliably delivered narratives that frame the expansion of the agricultural frontier as necessary and the appropriation of land as unavoidable in this development process. Such narratives support actions taken against traditional communities – by failing to recognize their rights – and protected areas – through downsizing and the loosening of legal protections. They have also resulted in actions taken by the Rural Caucus in the National Congress, which represents the most recent – and principal – expression of agrarian extractivism in Brazil. The result has included an opening of land – even that held illegally or in processes including land grabbing – for the production and exportation of commodities. The Cerrado in general and the Matopiba sub-region in particular have been deeply marked by social and economic inequality. This occurs, first, because there are few municipalities that benefit economically from the production of monocultures as control over land, resource rents, and surplus value is highly concentrated, meaning that the majority of cities across the Matopiba have not undergone any economic growth as a result. Second, the Matopiba is characterized by contradictions, which include a stimulated economy and the expansion of the agricultural frontier in some municipalities while extreme poverty continues to affect the majority of the population. This form of agrarian extractivism in Brazil criminalizes traditional practices of biodiversity management (generally, employing environmental laws) resulting in a loss of territory and biodiversity. Among the most serious environmental problems reported by traditional communities – even from inside the protected areas – are ongoing water shortages and the contamination of water resources. The effects of an advancing form of agrarian extractivism on territories dedicated to environmental protection are generating and exacerbating socioenvironmental conflicts in the region. Despite traditional communities protecting
80 Sérgio Sauer and Karla R.A. Oliveira the environment in their traditional territories for generations, they are directly impacted by the rapid decline and destruction of natural resources – water, land, and forests. In some parts of the Matopiba, the depletion of water resources has been questioned and resulted in building resistance to agribusiness and the expansion of the agricultural frontier. Nonetheless, traditional communities, that for centuries have occupied these territories, are denouncing the model of agrarian extractivism and attempting to reclaim their territories, as well as their access to and use of the communal waters and lands of the Cerrado.
Notes 1 The Provisional Measures n. 759, issued in December 2016 by Temer, and n. 910, published in December 2019 by Bolsonaro, are part of the historical logic of land appropriation based on the support of the state. These laws were issued to legalize (issuing land titles) illegal tenures of public areas of up to 2,500 ha, with very low payments, since the highest values will be 50% of the market value of the land (Leite et al. 2019). 2 The most explicit media support has from a publicity campaign on the channel Rede Globo, titled “Agro is tech. Agro is pop. Agro is everything”, which airs during high traffic hours and with considerable financing from a bank that does not, in fact, finance agribusiness (for more details, see Bruno 2012). 3 According to the IBGE Agricultural Census, in 2016, the municipality had one of the highest Gross Domestic Product (GDP) and the highest GDP per capita among the municipalities in the Matopiba region. The rural population represented almost 60% of the total population, but Correntina is among the highest in land concentration, since the large estates (size above 1,000 ha) occupied more than 75% of the total land area. The Gini index is very high with 0.927, in 2006 (very close to the level of maximum land concentration), with poverty reaching 45% of the rural population and 32% of the general population. The municipal Human Development Index was 0.603, below the national average, being an example of the contrast between economic growth and social inequality (Sauer et al. 2019b). 4 According to the National Water Agency (ANA 2018), 72% of water consumption in Brazil is used for irrigation, 11% for cattle grazing, 9% for cities, 7% for industrial production, and 1% for rural settlements, revealing the importance of water, especially for the expansion of intensive soy production in the Cerrado. 5 According to Veldwish et al. (2018), the fluid nature of the property has enormous impacts on access, allocation, reallocation, distribution, and quality of water. Furthermore, it contributes to controversial grabbing processes, which includes unequal power relationships, obscure distinctions between formal and informal rights, unclear administrative boundaries, and so on. 6 Quilombola communities are African slaves’ descendants, having their rights to the territories they traditionally occupy recognized by the Brazilian Constitution. They are part of “traditional people and communities”, social groups that have historical and particular relationships to land. They are practicing traditional agriculture and native pastures, combining techniques of fire use and seasonal management of nature. These systems are sustainable, conserving the biodiversity and water resources for centuries (Eloy et al. 2019, 2016). 7 Formosa do Rio Preto is one of the cities that was targeted in the “Operation Far West”, already mentioned, during which the Federal Policy detained judges, large estate owners, and land dealers involved in a land grabbing scheme that resulted in the appropriation of thousands of hectares of public land and the expulsion of traditional communities from their territories (Camargo 2019).
Agrarian extractivism in the Cerrado 81 8 The implementation of the Environmental Rural Registry (CAR), which was instituted in 2012 and overseen until 2019 by the Ministry of Environment, is a good example. Beyond the Bolsonaro administration shifting management of the programme putting it in the Ministry of Agriculture, despite it being a self-declaration system for rural reserves and properties, CAR has been used by large landowners to appropriate communal and public land in their declarations of environmental reserves (Sauer 2018; Sauer and França 2012). 9 The modification of Provisional Measure 910 in December 2019 allowed for the regularization of irregular land holdings across the country, including irregular soy plantings in areas near the park. The loosening of rules, beyond allowing for titling of grabbed land, has resulted in an increase in deforestation as cutting down the vegetation is considered a “productive action” and criteria that proves occupation of the land (Sauer et al. 2019a; Leite et al. 2019).
References Acosta, A. 2013. Extractivism and neoextractivism: two sides of the same course. In M. Lang and D. Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Amsterdam: TransNational Institute, TNI. Available at: www.tni.org/en/publication/ beyond-development (Accessed 12 December 2019). Almeida, A.W.B. 2011. A reconfiguração das agroestratégias: novo capítulo da guerra ecológica. In S. Sauer and W. Almeida (eds.), Terras e territórios na Amazônia: demandas, desafios e perspectivas. Brasília: Editora da UnB, 27–44. ANA – Agência Nacional de Águas. 2018. Conjuntura dos recursos hídricos no Brasil 2018. Brasília: ANA. Available at: http://conjuntura.ana.gov.br/ (Accessed 12 December 2019). Baletti, B. 2014. Saving the Amazon? Sustainable soy and the new extractivism. Environment and Planning A, 46, 5–25. Borras, S., J. Franco, S. Gómez, C. Kay, and M. Spoor. 2012. Land grabbing in Latin America and the Caribbean. Journal of Peasant Studies, 39(3–4), 845–72. Bruno, R. 2012. Movimento sou agro: marketing, habitus e estratégias de poder do agronegócio. In Anais do 36° Encontro Anual da ANPOCS. Águas de Lindóia, SP, 21–25 October, (unpublished). Camargo, I. 2019. PF prende juiz em ação contra venda de sentenças para grilagem de terras na Bahia. Site G1-BA and GloboNews. Salvador, 23 November. Available at: https://g1.globo.com/ba/bahia/noticia/2019/11/23/pf-prende-juiz-em-desdobramentode-operacao-que-afastou-presidente-e-mais-5-desembargadores-do-tj-da-bahia.ghtml (Accessed 25 November 2019). Carvalho, H.M. 2013. A expansão do capitalismo no campo e a desnacionalização do agrário no Brasil. Revista Reforma Agrária, 31–43, July. [Brasília, Associação Brasileira de Reforma Agrária (ABRA), Special edition]. Castro, L.F.P., E. Hershaw, and S. Sauer. 2017. Estrangeirização e internacionalização de terras no Brasil: oportunidades para quem? Estudos internacionais, 5(2), 74–102. Delgado, G. 2013. Economia do agronegócio (anos 2000) como pacto do poder com os donos da terra. Revista Reforma Agrária, 61–8, July. [Brasília, Associação Brasileira de Reforma Agrária (ABRA), Special edition]. The Economist. 28 August 2010. Brazilian Agriculture: The Miracle of the Cerrado. 28 August. Available at http://www.economist.com/node/16886442 (Accessed 30 July 2017).
82 Sérgio Sauer and Karla R.A. Oliveira Eloy, L., C. Aubertin, F. Toni, S.L.B. Lúcio, and M. Bosgiraud. 2016. On the margins of soy farms: traditional populations and selective environmental policies in the Brazilian Cerrado. The Journal of Peasant Studies, 43(2), 494–516, doi:10.1080/03066150.2015 .1013099. Eloy, L., I.B. Schmidt, S.L. Borges, M.C. Ferreira, and T.A. dos Santos. 2019. Seasonal fire management by traditional cattle ranchers prevents the spread of wildfire in the Brazilian Cerrado. Ambio, 48, 890–9, doi: https://doi.org/10.1007/s13280-018-1118-8. Favareto, A. 2019. Entre chapadas e baixões do Matopiba: Dinâmicas territoriais e impactos socioeconômicos na fronteira da expansão agropecuária no Cerrado. São Paulo: Ilustre Edit & GreenPeace. Fernandes, Min. Og. 2019. Pedido de busca e apreensão criminal n° 10-DF. Inquérito 1.258/DF. Salvador: Superior Tribunal de Justiça e Ministério Público Federal, 30 October. Available at: www.migalhas.com.br/arquivos/2019/11/decisao-og-fernandes-opera cao-faroeste.pdf (Accessed 25 November 2019). Flexor, G., and S.P. Leite. 2017. Land market and land grabbing in Brazil during the commodity boom of the 2000s. Contexto Internacional, 39(2), 393–420, May–August. Franco, J., L. Mehta, and G.J. Veldwish. 2013. The global politics of water grabbing. Third World Quarterly, 34(9), 1651–75, doi: 10.1080/01436597.2013.843852. Gryszpan, M. 2012. Origens e conexões norte-americanas do agribusiness no Brasil. Revista Pós Ciências Sociais UFMA, 9(17). Available at: www.periodicoseletronicos. ufma.br/index.php/rpcsoc/article/view/987. Gudynas, E. 2018. Development and nature: modes of appropriation and Latin American extractivisms. In J. Cupples, M.P. Schalscha, and M. Prietro (orgs.), The Routledge Handbook of Latin American Development. London: Routledge. Gudynas, E. 2019. Hasta la última gota. RevIISE – Revista De Ciencias Sociales Y Humanas, 13(13), 15–31. Available at: www.ojs.unsj.edu.ar/index.php/reviise/article/ view/323 (Accessed 12 December 2019). Harvey, D. 2003. O novo imperialismo. 3ª edition. São Paulo: Loyola Press. Hecht, S.B. 2005. Soybeans, development and conservation on the Amazon frontier. Development and Change, 36(2), 375–95. IBGE – Instituto Brasileiro de Geografia e Estatística. 2017. Censo Agropecuário 2017. Rio de Janeiro: IBGE. Avaiable at www.ibge.gov.br. IBGE – Instituto Brasileiro de Geografia e Estatística. 2018. Produção agrícola – lavoura temporária. Bahia: Formosa do Rio Preto. Available at: cidades.ibge.gov.br. Leite, A.Z., N. Tubino, and S. Sauer. 2019. Políticas públicas para terra e território: um olhar prospectivo sobre os próximos quatro anos no campo brasileiro. In Brasil: incertezas e submissão? São Paulo: Fundação Perseu Abramo, 171–89. Mançano Fernandes, B. 2019. Land grabbing for agro-extractivism in the second neoliberal phase in Brazil. Revista NERA, 22(50), 208–38, September–December. Martins, J.S. 1996. O tempo da fronteira: retorno à controvérsia sobre o tempo histórico da frente de expansão e da frente pioneira. Tempo Social: Revista de Sociologia da USP, 8(1), 25–70, May, São Paulo. Martins, J.S. 2012. Fronteira: a degradação do outro nos confins do humano. São Paulo: Contexto. Mascia, M.B., and S. Pailler. 2011. Protected area downgrading, downsizing and degazettement (PADDD) and its conservation implications. Conservation Letters, 4(1), 9–20, February. Available at: http://onlinelibrary.wiley.com/doi/10.1111/j.1755263X.2010.00147.x/full (Accessed 27 November 2017). McKay, B.M. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211.
Agrarian extractivism in the Cerrado 83 McMichael, P. 2018. Towards an ecology of development. In G.H. Fagan and R. Munck (orgs.), Handbook on Development and Social Change. Cheltenham, UK: Edward Elgar Publishing. MDS – Ministério do Desenvolvimento Social e Combate à Fome. 2016. Tabulador de dados do Cadastro Único. Brasilia: MDS. Available at: http://aplicacoes.mds.gov.br/ sagi/cecad/ sobre_tabcad.php (Accessed 24 September 2016). Oliveira, K.R.A. 2018. Fronteira agrícola e natureza: visões e conflitos no Parque Nacional das Nascentes do Rio Parnaíba. Dissertação de mestrado, Mestrado em Meio Ambiente e Desenvolvimento Rural (PPG-Mader), Universidade de Brasília, Brasília. Oliveira, G., and S. Hecht. 2016. Sacred groves, sacrifice zones and soy production: globalization, intensification and neo-nature in South America. The Journal of Peasant Studies, 43(2), 251–85, doi:10.1080/03066150.2016.1146705. Oliveira, K.R.A., and S. Sauer. 2020. A expansão da fronteira agrícola no Matopiba e seus impactos sobre unidades de conservação e comunidades quilombolas. In S. Guéneau, J.A.S. Diniz, and C.J.S. Passos (orgs.), Alternativas para o bioma Cerrado: Agroextrativismo e uso sustentável da sociobiodiversidade. Brasilia: Mil Folhas, IEB, 165–200. Pereira, L.I., L.D. Buscioli, C.F. Origuéla, J. Sobreiro Fo, and B.M. Fernandes. 2017. Disputas territoriais em Correntina-BA: territorialização do agronegócio, resistência popular e o debate paradigmático. Boletim DataLuta, Nera – Núcleo de Estudos, Pesquisas e Projetos de Reforma Agrária, 118. Available at: www.researchgate.net/ publication/321149145_DISPUTAS_TERRITORIAIS_EM_CORRENTINA_-_BA_ TERRITORIALIZACAO_DO_AGRONEGOCIO_RESISTENCIA_POPULAR_E_O_ DEBATE_PARADIGMATICO (Accessed 12 December 2019). Ramos, A. 2014. Políticas públicas para áreas protegidas no Brasil. In A Diversidade cabe na Unidade? Áreas protegidas no Brasil. Brasília: IEB, 153–65. Sassen, S. 2014. Expulsions: Brutality and Complexity in the Global Economy. Cambridge: Belknap Press, Harvard University Press. Sauer, S. 2018. Soy expansion into the agricultural frontiers of the Brazilian Amazon: the agribusiness economy and its social and environmental conflicts. Land Use Policy, 79, 326–38. Sauer, S. 2019. Rural Brazil during the Lula administrations: agreements with agribusiness and disputes in agrarian policies. Latin American Perspectives, 46(4), 103–21, 1 July. Available at: http://journals.sagepub.com/doi/full/10.1177/0094582X16685176 (Accessed 12 December 2019). Sauer, S., and S. Borras Jr. 2016. ‘Land grabbing’ e ‘green grabbing’: Uma leitura da ‘corrida na produção acadêmica’ sobre a apropriação global de terras. Campo-Território: Revista de Geografia Agrária, 11(23), July, 6–42. Sauer, S., and L.F.P. Castro. 2020. Land and territory: struggles, for land and territorial rights in Brazil. In O. De Schutter and R. Balakrishnan (eds.), Property Rights from Below: Commodification of Land and the Counter-Movement. New York: Routledge, 113–30. Sauer, S., and F.C. França. 2012. Código Florestal, função socioambiental da terra e soberania alimentar. Caderno CRH, 25(65), 285–307, May/August, Salvador. Sauer, S., and S.P. Leite. 2012. Agrarian structure, foreign investment in land, and land prices in Brazil. Journal of Peasants Studies, 39(3–4), 873–98, July–October. Sauer, S., and G. Mészáros. 2017. The political economy of land struggle in Brazil under workers’ party governments. Journal of Agrarian Change, 17(2), 397–414. Sauer, S., N. Tubino, A.Z. Leite, and G. Carrero. 2019a. MP 910: a grilagem de terras é ilimitada e incentiva o desmatamento. VioMundo, 19 December. Available at: www.
84 Sérgio Sauer and Karla R.A. Oliveira viomundo.com.br/voce-escreve/quatro-especialistas-em-questao-agraria-mp-910objetiva-a-grilagem-ilimitada-de-terras-publicas-e-desmatamento-de-grandes-areas. html (Accessed 19 December 2019). Sauer, S., A.Z. Leite, K. Oliveira, and A. Shankland. 2019b. The implications of closing civic space for sustainable development in Brazil. IDS and Act Alliance, mimeo. doi:10.13140/RG.2.2.31480.90885. Svampa, M. 2019. Las fronteras del neoextractivismo em América Latina: conflictos socioambientales, giro ecoterritorial y nuevas denpendencias. Bielefeld: Bielefeld University Press. https://doi.org/10.14361/9783839445266. Veldwish, G.J., J. Franco, and L. Mehta. 2018. Water grabbing: practices of contestation and appropriation of water resources in the context of expanding global capital. In R. Boelens, T. Perreault, and J. Vos (orgs.), Water Justice. Cambridge and New York: Cambridge University Press. Veltmeyer, H., and E. Zayago Lau. 2019. America Latina en el vórtice del capital extractivo y de la resistência. In S. Sauer (ed.), Desenvolvimento e transformações agrárias: BRICS, competição e cooperação no Sul Global. São Paulo: Outras Expressões, 45–70. World Bank. 2010. Rising Global Interest in Farmland: Can It Yield Sustainable and Equitable Benefits? Washington, DC: World Bank. Available at: siteresources.worldbank. org/INTARD/Resources/ESW_Sept7_final_final.pdf (Accessed 15 March 2012). WWF-Brasil. 2017. Unidades de Conservação sob risco: ofensiva contra áreas protegidas abrange uma área quase do tamanho de Portugal. Brasília: WWF-Brasil.
4
Social reproduction, dispossession, and the gendered workings of agrarian extractivism in Colombia Diana Ojeda
Introduction Studies of extractivism often point to its effects in relation to shifts in land tenure, environmental deterioration, labour issues, or health impacts. While these effects are not equally distributed along the lines of differentiation and power, few studies have focused on their gendered character (e.g. Erpel Jara 2018; Leguizamón 2019; Ulloa 2016). Moreover, despite gender’s centrality to the configuration of the social – and thus environmental – order, fewer works have analysed the role of gendered power relations in setting up the conditions that allow for the implementation, maintenance, and expansion of extractivist projects (e.g. León Araya 2017; Petzl forthcoming; Silva Santisteban 2018).1 In this chapter, I address the need for a feminist political ecology perspective on agrarian extractivism. This approach seems even more important due to the growing political traction that the concept has gained during the last decade in Latin America, as its wide use in social movement and activist circles attests. Despite its relevance for multiscalar strategies of territorial care and defence in the region, extractivism needs further theorization, and more attempts to contribute to building a “post-extractivist thought” seem more urgent than ever (Silva Santisteban 2018). The concept of extractivism originated within a long Latin American tradition of critiques to development and colonialism. In the context of the “commodity consensus” of the early 2000s (Svampa 2013), it became useful for understanding the economic and political landscape where different national economies turned to the production of raw materials such as copper, gold, silver, oil, corn, soy, and wheat. In general terms, extractivism depends on the control of “huge volumes of natural resources, which are not at all or only very partially processed and are mainly for export according to the demand of central countries” (Acosta 2015, 12). One of its key features is the extractive capture of value, that is, “value appropriation that occurs and proceeds without securing the material conditions that allow for the continuation of such value appropriation” (Ye et al. 2020, 5). Under this definition, extractivism constitutes “a particular way of structuring the processes of production and reproduction”, where the material reproduction of the resources involved is put at risk (Ye et al. 2020, 1). While this is true for capitalism in general, extractivism is often characterized by the concentration of
86 Diana Ojeda resource control accompanied by the externalization of social and environmental costs (Veltmeyer and Bowles 2014, 5). Another key feature in extractivism’s characterization has been the implied use of different forms of direct and indirect violence (LASA Forum 2019; Timm Hidalgo 2018; Zibechi 2019). Moreover, as Maristella Svampa’s (2013) commodity consensus points to, extractivism has become commonsensical; and, following Rocío Silva Santisteban (2018), it implies the use of state-led biopolitical strategies. Agrarian extractivism has captured growing scholarly attention in the context of heightened land-grabbing dynamics associated with the late 2000s crisis and the increased global demand for food, feed, and fuel. While most studies of extractivism have focused on mining and hyrdrocarbons, research on agrarian extractivism has proven their similarities in terms of the long-lasting effects on the renewability of resources. Authors such as Alberto Alonso-Fradejas (2018), Arturo EzquerroCañete (2016), and Ben M. McKay (2017), editors of this volume, have shown how agrarian extractivism involves the deterioration of nature and labour as part of the process of subtracting value through capitalist agricultural production. Seeking to contribute to this literature, I emphasize the centrality of both the material and social conditions of reproduction in the conceptualization and critique of extractivism. Drawing from Marxist feminist analyses, this chapter insists on the inseparability of production and social reproduction (see Hartmann 1979). This point of departure relates with Nancy Fraser’s (2014) take on Marx’s “hidden abode of production” (1976, 279). In her analysis, this move to an examination of the conditions of exploitation reveals a “dirty secret”: capital expands “via the noncompensation of a portion of workers’ labor time” (Fraser 2014, 147). However, calling for a more comprehensive conception of capitalism that gives prominence to its dependency on extra-economic conditions, Fraser emphasizes on Marx’s second move at the end of Capital vol. 1 through which “an even dirtier secret” is unveiled: “behind the sublimated coercion of wage labor lies overt violence and outright theft” (Fraser 2014, 147). According to her, this move from exploitation to expropriation requires an analysis of dispossession that gives centrality to such extra-economic conditions. These include “the natural processes that sustain life and provide the material inputs for social provisioning”, and the “solidary relations” and “affective dispositions” that contribute to shaping “the appropriately socialized and skilled human beings who constitute ‘labor’” (Fraser 2014, 157), both of which are profoundly shaped by gender roles, expectations and disparities. Additionally, this chapter poses an implicit dialogue with authors as diverse as Maria Mies (1986), Jason Moore (2015), and James O’Connor (1998), among others, who have contributed to a better understanding of the connections between Marxist economics and ecology. Drawing from ethnographic work carried out intermittently between 2013 and 2019 in Montes de María, a sub-region of the Colombian Caribbean, I examine the centrality of dispossession in the production of landscapes of oil palm cultivation. My understanding of dispossession as a violent process of sociospatial configuration under which what is being taken are the possibilities to sustain life, and gives centrality to social reproduction.
The gendered workings of agrarian extractivism 87 Social reproduction can be defined as the social and ecological structures, relations, and institutions that sustain life at the individual, communitarian, local, and planetary level. As Giovanna Di Chiro suggests, social reproduction is “the intersecting complex of political-economic, sociocultural, and material-environmental processes required to maintain everyday life and to sustain human cultures and communities on a daily basis and Intergenerationally” (2008, 281). As such, it ranges from the mere reproduction of bodies to the maintenance of the social order. Central to it are the fundamental, but largely undervalued and ignored practices that make up for “the fleshy, messy, and indeterminate stuff of everyday life” (Katz 2001, 711). The gendered socio-ecological relations that social reproduction illuminates thus lie at the centre of my argument. I contend that taking seriously the role of social reproduction requires bringing in both nature and gender – in its intersections with class, race, and other axes of oppression – to the analysis of agrarian extractivism. This requires paying closer attention to how processes and dynamics of dispossession create the background conditions of possibility for capitalist accumulation. In line with Silvia Federici’s work (2004), these processes and dynamics implicate the control of women’s labour and bodies, as well as the enclosure and deterioration of commons such as waterholes, riverbanks, and wildlands. I thus pay close attention to the way in which gender relations are key to the implementation and maintenance of agrarian extractivist projects, at the same time that result from them. The second section, following this introduction, focuses on the ways in which oil palm plantations in Montes de María were forged at the end of the 1990s through the articulation of paramilitary, state-led, and capitalist dispossession, and how they have been expanded and sustained by a systematic attack to social reproduction. I detail the role of gender-based violence in this process, particularly of sexual violence, in the spatial disciplining of female and other feminized bodies, as direct violence morphed into a subtler territorial regime. Section 3 pays attention to the ways in which feminized subjects’ work – including women, children, and the elderly – subsidizes capitalist accumulation within the oil palm plantation. Drawing on the particular figure of peperas (female oil palm fruit pickers) and analysing the transition from rice to oil palm production, I highlight the extraeconomic means through which the extractive capture of value is attained. Both sections also highlight agrarian extractivism’s disproportionate effects on women as it deepens gender disparities and potentiates gender-based violence. The concluding section reflects on the analytical possibilities that a feminist political ecology perspective can open up in relation to the workings of agrarian extractivism.
Gender-based violence and the implementation of oil palm plantations in Montes de María Montes de María is a region located in the north of Colombia, about an hour away from the city of Cartagena. Its contemporary history and geography have been marked by conflict between landowning elites, devoted to cattle ranching and
88 Diana Ojeda monocrops such as tobacco and rice, and peasant organizations, most of them legacy of 1968 state attempts of agrarian reform.2 More recently, Montes de María is best known for the unprecedented violence carried out by the paramilitary group Autodefensas Unidas de Colombia (AUC) at the end of the 1990s and the beginning of the 2000s. Between 1996 and 2005, the region reported 56 massacres, 6,000 selective murders, 215,000 cases of forced displacements, and 82,000 hectares of stolen and abandoned lands (Agencia Presidencial para la Acción Social y la Cooperación Internacional 2010; Centro Nacional de Memoria Histórica 2010). The lands stolen by paramilitary groups, acting with state-sanction, were quickly incorporated into the market, giving way to the implementation of oil palm plantations through a dual strategy of violent and legalized dispossession. Even if massacres, torture, sexual violence, and forced displacement, among other forms of violence, allowed the rapid expansion of oil palm (Bargent 2011; ILSA 2014; Verdad Abierta 2018), these mechanisms operated in tandem with different state-led strategies (Grajales 2011; ILSA 2012; Molano 2011; Verdad Abierta 2012), including direct state subsidies given to economic elites for further acquiring land under the scandalous case of Agro Ingreso Seguro; development projects such as Alianzas Productivas (productive alliances) between peasants and palm oil entrepreneurs that translated into insurmountable debt for the former; and securitization programmes like Zonas de Consolidación which were declared anti-Constitutional only a year after their implementation. The sociospatial – and thus environmental – order installed translated by the second half of the 2000s into ordinary, everyday forms of ongoing dispossession very much present today (Ojeda et al. 2015). As supposedly demobilized paramilitary groups were incorporated to the plantations’ security schemes, death threats, curtailed mobility, and restricted access to resources became the day-to-day experience for Montes de María’s peasant, black, and Indigenous communities (Defensoría del Pueblo 2015). In this context, sustained attacks to social reproduction and different forms of gender-based violence have been fundamental for opening up new spaces for agrarian extractivism as well as for maintaining them. Sexual violence, even if it is just one form of gender-based violence, makes this clear. A wide repertoire of sexual violence including rape, sexual slavery, forced cohabitation, forced nudity, sexual harassment, and forced maternity was perpetrated by the AUC men, but also by police and army officers acting in alliance with them, mostly against women in the region, but also against children and men (Centro Nacional de Memoria Histórica 2011, 2017). While literature on war and peace widely recognizes the systematic use of sexual violence as a mechanism used by armed actors to assert power and control, the analytical and judicial connection between sexual violence and land grabbing is rarely made. Studies of sexual violence in Colombia have gone beyond the prevalent narrative of women’s bodies as “war booty”, to point to the different ways in which sexual violence played a key role in the dispossession of local populations and the control of whole communities (Corporación Humanas 2013; Centro Nacional de Memoria Histórica 2011). The continuation of such violence after the cessation of direct attacks has also been studied for the country, particularly in relation to
The gendered workings of agrarian extractivism 89 the ways in which the armed conflict has become tightly articulated with genderbased violence occurring within spaces supposedly external to war, such as the community and the home (Centro Nacional de Memoria Histórica 2017, 2018). Taking this into account, sexual violence played an important role in the dispossession of the lands that were quickly turned into oil palm plantations, among others such as teak, in Montes de María. Nevertheless, beyond the monocrops’ implementation, the continuation of this and other forms of violence has been key to the maintenance and expansion of extractive ventures many years after the direct paramilitary control morphed into a less tight territorial regime. Extended through less evident, yet ongoing forms of violence, sexual violence continues to be an effective mechanism of disciplining women and restricting their use of land and water, among other key resources (Berman Arévalo and Ojeda 2020). This in relation as well to how violence, particularly against women and other feminized subjects such as children and sexuality/gender dissidents, becomes routinized in extractive sites, both in the form of direct violence and of the exacerbation of structural inequalities (Meertens 2001; Silva Santisteban 2018; Ulloa 2016). As stated, the use of sexual violence, and other forms of war against women’s bodies, served its purpose of spreading terror and forcedly displaying entire towns, and thus of dispossessing peasants from their land “opening up” the frontier of expansion of the oil palm plantations in the region.3 But as many of the sexual attacks occurred when women were working on their parcelas (plots), this was another way for women in particular losing access to their land. The lost plots for subsistence agriculture point towards the embodied form of violence that carried a very powerful message in terms of what Rita Segato calls a “pedagogy of cruelty” (2014, 23): sexual harassment, sexual assaults in the plots, curfews and the strict control over women’s mobilities and sexuality have worked as a very effective mechanism of disciplining women and circumscribing them to private spaces. In many of my conversations with women in the region they expressed a profound resentment about not having the possibility to work the land and a feeling of reclusion to the domestic space. Losing access to land, and with it what they saw as an important component of their campesina identity, affected them deeply. In their stories, they cherished their gardens as a way of reaffirming themselves as peasants and the possibility the plants give them to reconnect with moral economies which were dramatically disrupted by the war and involve material and symbolic networks necessary for tending for the sick, sharing food, and dealing with death and trauma (Orozco 2016). Multiple stories about pregnancy, birth, and abortion also spoke of this embodied dimension of dispossession. For them, the land also represented the possibility to control their sexual and reproductive life using their knowledge of plants. This possibility was lost along with the war and the expansion of the plantations, and has been only partially recovered through their gardens. The reclusion to the home did not mean that their productive labour was reduced, nor that their daily itineraries were simply cut. On the contrary, now that they could not cultivate and with the intensification of the monetization of the economy in the region, they had to engage in paid work as food vendors,
90 Diana Ojeda domestic workers, and other forms of care work in other towns and peri-urban centres. Their daily walks for water became longer too as the enclosure of jagüeyes (waterholes) meant more difficult access to the liquid. This was true as well for children and the elderly, whose working hours also expanded. For children, the longer walks to fetch water represented waking up even earlier, before entering school at 6 a.m. For older women in particular, it meant a greater load of childcare as mothers and other caretakers had to be away from town for longer periods of time. For all, getting food has implied more rebusque (rummaging), engaging in diverse activities for making ends meet, as well as a greater dependency on male partners and relatives which have been pushed to work cutting the oil palm in precarious jobs which are not available year around. The continuity of the paramilitary territorial regime has translated too on the continued use of different forms of gender-based violence to adjudicate certain spaces and roles to women within agrarian extractivism, as the case of peperas analysed in the next section shows. However, this is not limited to productive spaces. As many of the women identify a connection between their role in peasants’ and women’s organizations and the different forms of gender-based violence they have survived, they point as well to a reduction of political spaces. The same can be said to spaces fundamental for women’s sociability and children’s play such as waterholes and wildlands (el monte), which have been turned increasingly into neatly organized rows of oil palms. This restriction of women’s spaces of production, reproduction, political action, and sociability is a constitutive element of agrarian extractivism in the region, and not just a consequence of the expansion of oil palm plantations. At the same time, gender-based violence has played a central role in the articulation of paramilitary, state, and capitalist dispossession. Sexual violence, among other forms of gender-based violence that range from structural inequality of access to resources and paid jobs to feminicide, have been effective for disciplining and dispossessing women and other feminized subjects. In the next section, I explore their place within the inner workings of agrarian extractivism in the region.
Gender, social reproduction, and “the backstory of expropriation” With rice it was very different in the sense that what we [the women] picked up from the ground. . . [what was left after the harvest] we could eat. . . . The palm is different, see? We can’t eat from the palm (de la palma no podemos comer). (Maryuris, Maríalabaja, July 2014) In Maryuris’s account, the dirtier secret of capitalist accumulation is laid bare. The front story of exploitation of rice workers and palm workers is of course a key part of it, but only a small one. Instead, it is “the backstory of expropriation”, as Fraser puts it (2014, 146), which I argue helps us to better understand agrarian extractivism’s specificities. Maryuris defines herself as campesina
The gendered workings of agrarian extractivism 91 (peasant). She is in her 50s and usually refers to the times before oil palm with nostalgia. It is not that she idealizes the men’s working conditions within the rice plantations nor that she has forgotten the suffocating sensation of the pesticides that the fumigation plane left behind. Her place within the plantation, along with the other women (and children and the elderly), has not changed much as well: providing the reproductive work necessary for the productive work of male partners and relatives to occur, and coming afterwards to pick up from the ground the leftovers from the harvest. What is radically different is explicit in her brutal assertion: “we could eat”. This has to do with the very concrete fact that rice could be milled, cooked, and eaten; while the palm fruit, even if it can be manually converted into oil, does not represent a significant part of a meal. However, perhaps more important, it relates to the fact that, in an increasingly monetized economy, the palm fruit needs to become money that can buy rice and other staples to eat. And that possibility is mediated by gender positionings, roles, and relations. Rice monocrops were implemented in the region as part of the late 1960s state attempts of agrarian reform. Hand in hand with technological packages, state investments in irrigation infrastructure and the mechanization brought upon by the Green Revolution, rice cultivation and cattle ranching gave way in the following three decades to the control of big portions of land and water by a small landowning elite. Within this economy, landless (mostly male) peasants were daily labourers, leaseholder peons, or share-croppers. As it still happens today, the care work performed by women both for the hacienda and for the home was taken for granted and rarely paid. Also, as still happens today, women occupied the margins of the plantation system. Yet, women’s memories of rice harvesting speak of a meagre, yet dramatically different, room for manoeuver within rice production that they lost with the arrival of the palm. Maryuris’s account resonates with Eloísa Berman Arévalo’s (2017) findings in the region where implicit agreements between landowners and landless peasants allowed for women and children to harvest the leftovers of rice production. As Berman Arévalo (2017, 180–2) notes, these arrangements allowed landowners to turn waste into social legitimacy, at the same time those were used by campesinas as a way of dealing with the logics of mechanized rice production. As Eloísa, a woman from Montes de María, described: After they cut their rice, they shouted ‘¡campo libre!’ (fields are open!). That rice was for the people. If you found a row with good rice, you put it in that washbowl and just let yourself go, let yourself go till the row was done . . . It was beautiful seeing all those people there. In the afternoon, you saw everybody leaving with their sacs on their heads, everybody, everybody, women and kids and a few men all along the long road. (Eloísa cited in Berman Arévalo 2017, 181) Eloisa’s accounts, like Maryuris’s and many other women’s stories of rice harvesting in Montes de María, are full of nostalgia for better times before the oil palm.
92 Diana Ojeda While some practices of rice milling and harvesting still occur within domestic spaces, they are rare in an increasingly tighter space for social reproduction amidst the endless rows of palm. As noted, at least two elements remain the same. On the one hand, women (and children and the elderly among other feminized subjects) occupy the margins at already marginalized spaces and ecologies deemed as having natural “vocation” for large-scale agribusiness. Within both rice and oil palm plantations, as in other male-dominated productive activities, it is men who have access to the meagre salaries and women are left to pick up the leftovers. As Petzl (forthcoming) notes for the case of Montes de María, women and other feminized subjects are excluded from the socionatural arrangements brought up by the oil palm expansion and end up living “in the shadow of the palm”. On the other hand, deeming women, their bodies, and their work as expendable or disposable is fundamental for capital accumulation to occur. As feminist work has previously noted (e.g. Chattopadhyay 2018; Federici 2004; Wright 2006), it is through this gendered – and classed, racialized, etc. – positioning that waste can be turned into something of value: social value in the case of rice; and, as I show later, commercial value in the case of palm oil. Within the oil palm economy, it is men who are hired for the work of cutting, loading, and carrying the palm. Women are left to be peperas. They go in groups walking behind the men and picking up the fruit that falls from the bunches men are collecting and transporting. They have no protection, usually wearing flip-flops, unlike men who get high rubber boots as part of their personnel provision. Peperas are not contracted by palm owners; they are directly paid by the male workers (their partners, fathers, or brothers) according to the weight of the fruits picked. Male workers get paid as well by weight. In our conversations, women usually complained about the job conditions: the unbearable heat, the risk of snake bites, and the fact that a full workday leaves them no more than two dollars for them, when they do not do it as “a favour” in support of their loved ones. They perform this work for long periods of time, often turning to food vending and domestic work among other activities in exchange for cash. Peperas are not only subjected to the exploitation of men, who are themselves exploited; but in some cases they are forced to work picking the fruit. More than often, partners make women work for the plantation as a way of controlling where they are and with whom. “We were there in the same place. On the same farm . . . We entered together the job, and left together the job . . . he was at peace with that” (Juana cited in Petzl forthcoming). As Petzl notes, it is through their male counterparts that they have access to the plantation space. The control of their bodies, mobility, and schedules is reasserted through patriarchal relations that, although not new in the region, were actualized and reinforced after the paramilitary incursion and the oil palm expansion as women lost the space of the plot. Additionally, many of the moral economies that supported their livelihoods and general well-being have eroded: “In the days before, one got more, at least food to eat . . . one could go to the neighbour – there, where the palm is now – and he
The gendered workings of agrarian extractivism 93 gave you plantains, gave you milk, cheese . . . That was before” (Luisa cited in Petzl forthcoming). In relation to the women’s reproductive work, they saw that the arrival of the oil palm plantations extended their workday both directly and indirectly. Directly, as the absence of men in the spaces of the home and the plot, to work at the plantation for a full workday, represented for the women the need to take care of fixing the home, collecting firewood, and fishing, among other activities. “One here is both man and woman because one has to go find firewood, to fish, to go get the yucca. And they just do their job [at the plantation]” (Juana cited in Petzl forthcoming). Indirectly, as with the deteriorated ecologies and the progressive enclosure felt in the region, women have a heavier social reproduction load. Clearly, feminized work subsidizes the plantation through the appropriation of the time and effort of women, children, and the elderly who go fetch water and firewood, care for the sick, cook, clean, and deal with the physical and mental scars of paramilitary and extractivist violence. But this has occurred in a context where the expansion of the plantation has dramatically put at risk the mere subsistence of the local communities. As such, social reproduction is fundamental for the analysis of agrarian extractivism. Land and water grabbing by the plantation have severely restricted access to food (Quiroga and Vallejo 2019). Common spaces for hunting bushmeat and for chickens and pigs to roam free have been curtailed (Petzl forthcoming). The glyphosate cocktail the palm requires pollutes water and land, deteriorating local communities’ health conditions: men often suffer strong headaches, dizziness, and severe dehydration when working in the plantation; women and children suffer respiratory, digestive, and skin problems usually in relation to the quality of the water they consume or use for washing the clothes or playing. From time to time, they face massive fish kills, so they know well that the water and their bodies are being poisoned. Moreover, the disproportionate effects of agrarian extractivism over women and children are also evident in terms of the different ways in which deteriorating living conditions, increased dependency on money, and thus on male partners and relatives, and the reclusion to domestic spaces are probably connected to increased levels of physical and psychological violence against them. The heightened processes and dynamics of dispossession brought up by agrarian extractivism can be understood in terms of Bina Fernandez’s (2018) use of the concept of depletion. Based on her work in the Gujarat region of India, Fernandez asserts that “dispossession is not only the structural violence of the appropriation of land and resources as delineated in Marxist scholarship, but also, importantly, the depletion of capacities due to gendered, differential exposure to the currents of structural violence, which are visible precisely in the realm of social reproduction” (Fernandez 2018, 17). Fernandez’s approach makes evident the need to examine not just “violent, grand structural moments of enclosure”, but the “incremental, under-the-radar process of erosion of the capacity of social reproduction that is ongoing in the everyday lives of marginalized populations” (Fernandez 2018, 18). Similarly, León Araya (2017), in his research on the Bajo Aguán’s region in Honduras, shows how the arrival of oil palm plantations “transformed the
94 Diana Ojeda relations between men and women and created a distinction between productive/ paid work and reproductive/unpaid work” as well as productive and reproductive spaces (León Araya 2017, 178). As he argues, it is through this distinction that the dispossession and subordination of peasants was made possible. As this section has shown, such gendered experiences of dispossession (see also Elmhirst et al. 2017; Levien 2017; Li 2017; Julia and White 2012) are fundamental for a full comprehension of agrarian extractivism as it is through dispossession that the extraction of value from nature is attained. In the concluding section, I turn to what a feminist political ecology perspective can bring to the analysis of agrarian extractivism.
Conclusions This chapter has pointed to the disproportionate effects of agrarian extractivism over women and feminized subjects. The mobilization of fear, monetization of livelihoods, widespread precarity, and the enclosure of spaces like water wells, riverbanks, and wildlands have dramatically affected the capacity of local communities to sustain life, and in particular has translated for women into spatial restrictions, extended productive and reproductive work, as well as increased dependency towards men, and thus more vulnerability to genderbased violence. However, as I have shown, these gendered roles, relations, and spaces are not just the result of agrarian extractivism; they are its conditions of possibility. Drawing from feminist studies, I have centred my attention on the ways in which social reproduction is absolutely necessary for capital accumulation. From a feminist political ecology perspective, I have pointed to the ways in which agrarian extractivism deteriorates the mere social and ecological conditions that sustain life. If this is true for capitalism as a whole, what makes agrarian extractivism different from other agrarian capitalist projects? One of the conditions for – as well as the results of – the expansion of the oil palm plantations is the blatant deterioration of local living conditions. With agrarian extractivism the possibility to sustain life is severely put at risk. While large monocrops of rice existed in the region along with extensive cattle-ranching, the hacienda system guaranteed basic living conditions. This shift from production to reproduction, the extractive capture of value, is what I argue results central for agrarian extractivism’s functioning. And, as I have shown, gender (working along with other systems of domination) is not subsidiary, but constitutive, of it. In that sense, I argue that a feminist perspective on agrarian extractivism allows for a better understanding of its logics and forms of operation. It opens up the possibility to examine its gendered functioning. Furthermore, such an approach can potentially make visible the everyday, ordinary geographies through which the extractivist socio-environmental regime is questioned and subverted on a day-today basis. If it is through the dispossession of women’s bodies, spaces, social ties, and reproductive labour that palm oil plantations are implemented and sustained, it is through them that they can be contained and subverted.
The gendered workings of agrarian extractivism 95
Acknowledgements I thank Andrés León Araya for his comments and suggestions to preliminary versions of this chapter. They significantly contributed to my argument. I also want to thank an anonymous reviewer and all participants of the workshop on Agrarian Extractivism carried out at University of Calgary in September 2019, particularly the editors of this volume, for their careful reading and feedback.
Notes 1 I emphasize gender’s constitutive role in the configuration of the social order (Scott 1986; see also Arango and Viveros 2012). I argue that this social order is at the same time an environmental one, particularly in relation to how gendered subjects and gendered natures are mutually constituted (Harris 2006; Nightingale 2006; Rocheleau et al. 1995). 2 The Asociación Nacional de Usuarios Campesinos – ANUC (National Association of Peasant Users) is the most important organization that emerged in the second half of the 20th century, bringing together plantation workers’ unions, cooperatives, and smaller organizations. Despite originating from a state initiative within attempts of agrarian reform, the ANUC ended up being an important space of peasant struggle in Montes de María (Fals Borda 2002; Rivera Cusicanqui 1982; Zamosc 1987). 3 I have carried out ethnographic work mainly with women, in part because of my own embodied positioning in the field, but also because of my particular political and academic interests. Despite this focus, sex/gender dissidents, youth, and children, among other feminized subjects, have also participated in my research and have informed my thinking. Further research is needed to fully account for their realities, lived experiences, and knowledge production.
References Acosta, A. 2015. Después del saqueo: Caminos hacia el posextractivismo. Perspectivas, Análisis y Comentarios Políticos América Latina, 1. Agencia Presidencial para la Acción Social y la Cooperación Internacional. 2010. Unidades agrícolas familiares, tenencia y abandono forzado de tierras en Colombia. Bogotá: Acción Social. Alonso-Fradejas, A. 2018. El proyecto de capitalismo agroextractivo: una mirada a la cuestión agraria contemporánea desde Guatemala. In América Latina en la mirada: Las Transformaciones Rurales en la transición Capitalista. México: UNAM. Arango, L.G., and M. Viveros. 2012. El género: una categoría útil para las ciencias sociales. Maguaré, 26(2), 267–71. Bargent, J. 2011. Montes de María: ¿quién heredó la riqueza manchada de sangre de las AUC? Kavilando, 3(2), 85–9. Berman Arévalo, E. 2017. Making Space in the “Territorial Cracks”: Afro-Campesino Politics of Land and Territory in the Colombian Caribbean. Doctoral dissertation, Department of Geography, The University of North Carolina, Chapel Hill. Berman Arévalo, E., and D. Ojeda. 2020. Ordinary geographies: care, violence and agrarian extractivism in “post-conflict” Colombia. Antipode, 52(6), 1583–602. Centro Nacional de Memoria Histórica. 2010. Memorias de despojo y resistencia campesina en la costa Caribe (1960–2010). Bogotá: CNMH. Centro Nacional de Memoria Histórica. 2011. Mujeres y guerra: Víctimas y resistentes en el Caribe colombiano. Bogotá: CNMH.
96 Diana Ojeda Centro Nacional de Memoria Histórica. 2017. La guerra inscrita en el cuerpo: Informe nacional de violencia sexual en el conflicto armado. Bogotá: CNMH. Centro Nacional de Memoria Histórica. 2018. Un carnaval de resistencia: Memorias del reinado trans del río Tuluní. Bogotá: CNMH. Chattopadhyay, S. 2018. Violence on bodies: space, social reproduction and intersectionality. Gender, Place & Culture, 25(9), 1295–304. Corporación Humanas. 2013. La violencia sexual. Una estrategia paramilitar en Colombia: Argumentos para imputarle responsabilidad penal a Salvatore Mancuso, Hernán Giraldo y Rodrigo Tovar. Bogotá: Corporación Humanas. Defensoría del Pueblo. 2015. Sistema de Alertas Tempranas, March 30 Available at: https:// www.defensoria.gov.co/public/pdf/Documento-SAT-15.pdf. Di Chiro, G. 2008. Living environmentalisms: coalition politics, social reproduction, and environmental justice. Environmental Politics, 17(2), 276–98. Elmhirst, R., M. Siscawati, B. Basnett, and D. Ekowati. 2017. Gender and generation in engagements with oil palm in East Kalimantan, Indonesia: insights from feminist political ecology. The Journal of Peasant Studies, 44(6), 1135–57. Erpel Jara, Á. (ed.). 2018. Mujeres en defensa de territorios: Reflexiones feministas frente al extractivismo. Santiago: Fundación Heinrich Böll. Ezquerro-Cañete, A. 2016. Poisoned, dispossessed and excluded: a critique of the neoliberal soy regime in Paraguay. Journal of Agrarian Change, 16(4), 702–10. Fals Borda, O. 2002 [1986]. Retorno a la tierra: Historia doble de la Costa. Volume 4. Bogotá: El Áncora Editores, Universidad Nacional de Colombia, Banco de la República. Federici, S. 2004. Women, land-struggles and globalization: an international perspective. Journal of Asian and African Studies, 39(1–2), 47–62. Fernandez, B. 2018. Dispossession and the depletion of social reproduction. Antipode, 50(1), 142–63. Fraser, N. 2014. Behind Marx’s hidden abode. New Left Review, 86, 55–72. Grajales, J. 2011. The rifle and the title: paramilitary violence, land grab and land control in Colombia. Journal of Peasant Studies, 38(4), 771–92. Harris, L. 2006. Irrigation, gender, and social geographies of the changing waterscapes of Southeastern Anatolia. Environment and Planning D: Society and Space, 24(2), 187–213. Hartmann, H. 1979. The unhappy marriage of Marxism and feminism: towards a more progressive union. Capital & Class, 3(2), 1–33. Instituto Latinoamericano por una Sociedad y un Derecho Alternativos – ILSA. 2012. Primer Informe Montes de María: Entre la consolidación del territorio y el acaparamiento de tierras, Aproximación a la Situación de Derechos Humanos y del DIH en la region (2006–2012). Bogotá: ILSA. Instituto Latinoamericano por una Sociedad y un Derecho Alternativos – ILSA. 2014. Montes de María: Un escenario de riesgo para la exigibilidad de los derechos de la población víctima del conflicto armado. Bogotá: ILSA. Julia, W., and B. White. 2012. Gendered experiences of dispossession: oil palm expansion in a Dayak Hibun community in West Kalimantan. The Journal of Peasant Studies, 39(3–4), 995–1016. Katz, C. 2001. Vagabond capitalism: and the necessity of social reproduction. Antipode, 33(4), 708–27. LASA Forum. 2019. Dossier: Violencia contra líderes y lideresas defensoras del territorio y el ambiente en América Latina. LASA Forum, 50(4).
The gendered workings of agrarian extractivism 97 Leguizamón, A. 2019. The gendered dimensions of resource extractivism in Argentina’s soy boom. Latin American Perspectives, 46(2), 199–216. León Araya, A. 2017. Domesticando el despojo: palma africana, acaparamiento de tierras y género en el Bajo Aguán, Honduras. Revista Colombiana de Antropología, 53(1), 151–85. Levien, M. 2017. Gender and land dispossession: a comparative analysis. The Journal of Peasant Studies, 44(6), 1111–34. Li, T.M. 2017. Intergenerational displacement in Indonesia’s oil palm plantation zone. The Journal of Peasant Studies, 44(6), 1158–76. Marx, K. 1976. Capital. Volume 1. London: Penguin Harmondsworth. McKay, B. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211. Meertens, D. 2001. Facing destruction, rebuilding life: gender and the internally displaced in Colombia. Latin American Perspectives, 28(1), 132–48. Mies, M. 1986. Patriarchy and Accumulation on a World Scale: Women in the International Division of Labor. London and New York: Zed Books. Molano, A. 2011. Montes de María: un modelo de desarrollo que concentra tierras y mano de obra. El Espectador, 21 de mayo. Moore, J. 2015. Capitalism in the Web of Life: Ecology and the Accumulation of Capital. London: Verso. Nightingale, A. 2006. The nature of gender: work, gender, and environment. Environment and Planning D: Society and Space, 24(2), 165–85. O’Connor, J. 1998. Natural Causes: Essays in Ecological Marxism. New York: Guilford. Ojeda, D., J. Petzl, C. Quiroga, A.C. Rodríguez, and J.G. Rojas. 2015. Paisajes del despojo cotidiano: acaparamiento de tierra y agua en Montes de María, Colombia. Revista de Estudios Sociales, 54, 107–19. Orozco, A.S. 2016. Mulheres reabitando os sonhos: Um recorrido pelo cuidado da vida. Estudio de caso no município de María la Baja, Colombia. MA thesis, Rural Development, Faculdade de Ciências Econômicas da Universidade Federal do Rio Grande do Sul, UFRGS, Porto Alegre. Petzl, J. forthcoming. La vida a la sombra de la palma. Subjetividades y paisajes del despojo cotidiano en María La Baja, Montes de María. In D. Ojeda, C. Quiroga, and D. Vallejo (eds.), Montes de María: una región en disputa. Bogotá: Editorial Javeriana. Quiroga, C., and D. Vallejo. 2019. Territorios de agua: infraestructura agrícola, reforma agraria y palma de aceite en el municipio de Marialabaja (Bolívar). Revista Colombiana de Antropología, 55, 59–89. Rivera Cusicanqui, S. 1982. Política e ideología en el movimiento campesino: el caso de la ANUC (As. Nacional de Usuarios Campesinos). Bogotá: CINEP. Rocheleau, D., B. Thomas-Slayter, and E. Wangari. 1995. Feminist Political Ecology: Global Issues and Local Experience. London and New York: Routledge. Scott, J. 1986. Gender: a useful category of historical analysis. The American Historical Review, 91(5), 1053–75. Segato, R. 2014. Las nuevas formas de la guerra y el cuerpo de las mujeres. Puebla: Pez en el árbol. Silva Santisteban, R. 2018. Mujeres y conflictos ecoterritoriales: Impactos, estrategias, resistencias. Lima: AECI. Svampa, M. 2013. Consenso de los commodities y lenguajes de valoración en América Latina. Nueva Sociedad, 244(4), 30–46. Timm Hidalgo, A. 2018. Feminicidio extractivista: Reflexiones sobre la violenciahacia las mujeres defensoras del agua y los territorios. In Ángela Erpel Jara (ed.), Mujeres en
98 Diana Ojeda defensa de territorios: Reflexiones feministas frente al extractivismo. Santiago: Fundación Heinrich Böll. Ulloa, A. 2016. Feminismos territoriales en América latina: defensas de la vida frente a los extractivismos. Nómadas, 45, 123–39. Veltmeyer, H., and P. Bowles. 2014. Dynamics of extractivist resistance: linking Latin America and Northern British Columbia. Paper presented at a Conference of the International Studies Association (ISA), Universidad de Buenos Aires, Buenos Aires, 21–25 July. Verdad Abierta. 2012. La tierra y el conflicto en los Montes de María. Available at: https:// verdadabierta.com/la-tierra-y-el-conflicto-en-los-montes-de-maria/. Verdad Abierta. 2018. Los acuatenientes: La historia del agua, la tierra y la agroindustria de la palma de aceite en María La Baja, Bolívar. Available at: https://verdadabierta. com/especiales-v/2018/acuatenientes/#contenido. Wright, M. 2006. Disposable Women and Other Myths of Global Capitalism. New York: Routledge. Ye, J., J. van der Ploeg, S. Schneider, and T. Shanin. 2020. The incursions of extractivism: moving from dispersed places to global capitalism. The Journal of Peasant Studies, 47(1), 155–83. Zamosc, L. 1987. La cuestión agraria y el movimiento campesino en Colombia: luchas de la Asociación Nacional de Usuarios Campesinos (ANUC), 1967–1981. Bogotá: Instituto de Investigaciones de las Naciones Unidas para el Desarrollo Social. Zibechi, R. 2019. El estado de excepción como paradigma político del extractivismo. Movimiento Mesoamericano contra el Modelo Extractivo Minero. Available at: https://movimientom4.org/2019/04/el-estado-de-excepcion-como-paradigma-politicodel-extractivismo/.
5
Agrarian extractivism and sustainable development The politics of pineapple expansion in Costa Rica Andrés León Araya
Introduction In early 2017, a study conducted by a group of public institutions using satellite imagery and coordinated by the UNDP claimed that between 2000 and 2015, nearly 6,000 hectares of forest had been lost to the expansion of pineapple production in Costa Rica. According to this study, the amount of land dedicated to this crop had quadrupled during that period of time, reaching a staggering 58,000 hectares in 2015; a number glaringly higher from the 37,000 hectares that the agricultural census had reported just three years before (Araya 2017). This data was used by environmental groups such as the Costa Rican Ecological Federation (FECON) to argue that the expansion of pineapple was gravely affecting the country’s ecosystems, as well as poisoning the water sources of many communities in the rural areas (Araya 2017). It also forced the National Chamber of Pineapple Producers and Exporters (CANAPEP) to issue a response to what they saw as an attack. In a newspaper article that appeared less than a month after, Abel Chaves, president of CANAPEP, responded claiming that the report was inaccurate and that the so-called expansion was the result of the “migration” by producers from other crops such as coffee, citrus trees, macadamia nuts, rice, beans, sugar cane, among others, to pineapples. He furthered his claim by citing that Costa Rica’s forest cover had increased from 1.2 to 1.4 million hectares in that same period of time. Chaves concluded that, “the word expansion has been used politically in the case of pineapples”, and that “an incorrect message is being presented. The image of a sector that has been successful is being tarnished” (Rojas 2017). Around this same time, the then Minister of Agriculture, Luis Felipe Arauz (2017) wrote an op-ed in which he argued that the discussion regarding the pineapple industry had been polarized between the CANAPEP that exalted its economic importance and the environmental groups that were against its expansion due to the “supposed” environmental impacts. According to him, it was necessary to find an equilibrium between both sectors. In the rest of the article, he argues first in favour of the economic importance of the activity (basically creating jobs and exports) and against the claims of the environmentalists. For example, regarding the aforementioned study, the Minister argued that it “does not prove that the
100 Andrés León Araya source of the deforestation is the pineapple activity, since it could have been the result of another activity previous to pineapples during those 15 years”. He finished his piece with a reflection on FECON’s proposal of enforcing a moratorium on the expansion of the crop until the health and environmental impacts of its cultivation were assessed: “I do not think that this solves the problems noted. Should we prohibit the growth of an activity that creates thousands of jobs and resources for the country? Or do we control and resolve its problems?” What is clearly missing from both Chaves’ and Arauz’s accounts is what happened to the people and communities whose livelihood depended on those previous activities, as well as the economic value that was lost in the transformation of the land use, from activities such as cattle ranching, maize and milpa or oranges, to pineapple. They also fail to address the numerous claims of water poisoning and appalling working conditions that have been levelled against the industry and that have little to do with deforestation (Acuña 2006; Aguilar and Arroyo 2014; Arguedas 2015; Carazo et al. 2016; Diepens et al. 2014; EcheverríaSáenz et al. 2012; Maglianesi-Sandoz 2013; Silvetti 2015). It would appear that these are the cases in which, according to Chaves, the word expansion is used politically. The underlying idea in the official discourse of both the government and the sector is that economic growth through agricultural exports is compatible with the conservation of nature through the creation of conservation areas. This idea is very much in sync with the framework of sustainable development, according to which sustained economic growth can be combined with the conservation of nature, if the proper technological and managerial fixes are applied. As such, what should be a very political discussion – what to produce, how, and for whom – is displaced by a techno-political field that “is structured around the perceived inevitability of capitalism and a market economy as the basic organizational structure of the social and economic order, for which there is no alternative” (Swyngedouw 2010, 215). This chapter explores the relation between agrarian extractivism and sustainable development, through the case of pineapple production in Costa Rica. My main argument is twofold. First, that by moving the discussion away from issues such as health, nature, and labour, and towards technological fixes, sustainable development renders invisible the stories of the people and landscapes upon which the pineapple activity has expanded. Second, the discourse of sustainable development attempts to give coherence to the fragmenting and destructive dynamics upon which the national strand of agrarian extractivism is predicated. The chapter is organized into four sections. First, I characterize the discussion on agrarian extractivism, in search of some key elements to analyse the Costa Rican case from a critical perspective that further delves into the transnational and technological aspects of its operations. Next, drawing from Gramscian understandings of common sense, I present the version of sustainable development that became hegemonic in Costa Rica from the 1990s onwards. In a third moment, I bring together the previous discussions on sustainable development and agrarian extractivism to shed light on the process of expansion of the pineapple activity in the country, and
Agro-extractivism, sustainable development 101 point to the specific mechanisms through which the implementation and expansion of the plantations have been enabled, creating a pineapple republic. Fourth, I show how, by reducing issues such as the large-scale dispossession of rural populations to environmental problems, the sustainable development discourse both opens and closes the possibilities of opposing the pineapple expansion. I finish with a discussion regarding pending issues in the study of agrarian extractivism. 1 Agrarian extractivism: from a literal to an expanded definition The notion of extractivism is based on the idea that there is a difference within capitalism between the extraction of value from labour (surplus value), as in the case of industrial activities, and the direct extraction of value from nature. Of course, industrial capitalism is based on the transformation of raw materials into commodities. However, while in industry the transformation of raw materials entails “adding” value, in terms of labour, technology, and infrastructure, among others; in extractivist activities, rather than addition, what we find is subtraction, as the sources of value – labour and nature – are destroyed (McKay 2017). In Latin America, authors such as Gudynas (2009) and Svampa (Svampa and Viale 2014) have defined extractivism in terms of a mass of natural resources that are extracted and exported with little added value and minimal linkages to the “national economies”. Initially we used to think about mining, with the boom in South America of monocrops such as soy in the 1990s, the notion was extended to include forms of agribusiness in what is known as agrarian extractivism. As such, agrarian extractivism resonates strongly with the dependency theory. Particularly in terms of centre–periphery relations, the uneven terms of global exchange, and the rentist patterns of periphery capitalism. All of these topics have been used fruitfully to discuss the global dynamics of capital accumulation and the political economy of the region (Coronil 1997; Moore 2015). At the same time, agrarian extractivism repeats one of the main biases of dependentista thought: The “territorial trap”. That is, the dominant tendency within social and economic sciences of separating domestic and international politics and spaces, and presenting the national-state as the dominant scale of analysis (Agnew 1994). As a result, the transnational and spatially fragmented aspects of capitalism are underplayed. This has been a point reiterated by globalization theorists. However, it would be a mistake to think that the spatial fragmentation and dispersal patterns of production, distribution, and consumption processes are something new, or related specifically to the turn of the century. Rather these are enduring aspects of capital accumulation, particularly of agricultural commodities. For example, the expansion of the sugar plantation economy in the Caribbean during the 18th century was related to the need to lower the price of industrial labour in Europe. Further, many of the disciplinary techniques that were later used in the European factory were first developed on Caribbean plantations (Mintz 1985; Moore 2010). We find another good example in the case of bananas. According to John Soluri (2009), the global trade of bananas was developed in Central America through the initial
102 Andrés León Araya combination of medium- and small-scale Honduran farmers and US capital located in ship trading and railroad construction, resulting in the giant fruit companies that we know nowadays. Also, in the 1930s and 1940s, when pathogens such as Panama disease began wreaking havoc on the banana plantations, it was in labs located in the Global South where practical solutions were found. Finally, according to Steve Striffler (1997), the vertical disintegration of transnational firms that began in the 1980s was the result of a set of land occupations staged by landless peasants and organized workers in United Fruit’s Ecuador banana plantations. Thus, although production is mainly concentrated in peripheries, and most of the value is funnelled towards the centres, it is simplistic to think that these are unilateral or unidirectional processes. Further, it underplays the levels of creativity – usually for furthering capital accumulation – that are unleashed by these types of enterprises. As such, our understanding of agrarian extractivism must go beyond the simple extraction of raw materials for export. Sandro Mezzadra, Brett Neilson, and Verónica Gago (Gago and Mezzadra 2017) have proposed that we differentiate between a literal approach to extractivism that describes historical and contemporary forms of forced removal of raw materials from the earth’s depths and biosphere, and an expanded concept of extractivism that includes the notion of “operations of capital, which not only describes specific and analytically isolable processes through which capital ‘hits the ground’ in diverse material contexts but also enables investigation of how such operations concatenate and spread within larger formations of capitalism” (Mezzadra and Neilson 2017, 3). Here, our focus moves away from discretely located dynamics of extraction of raw materials to a broader constellation of human and non-human elements and processes, located in different spaces and that must come together in particular, albeit not necessarily in expected ways, for capital accumulation to take place (Tsing 2005). In this chapter, I am particularly interested in exploring how a sense of coherence is given to one such constellation. Further, how both this coherence and the spatial and political fragmentation of agrarian extractivism are actually crucial for its operations. As shown in the next section, the Costa Rican case allows for a closer analysis of how landscapes of supposedly sustainable agrarian extractivism are assembled. 2 The geographies of “Desarrollo sostenible a la tica” According to Alonso Ramírez’s (2017) study on market-based conservation in Costa Rica, it is impossible to separate the country’s process of structural adjustment during the 1980s and 1990s, from the sustainable development discourse: “While initial policy measures were centered on developing non-traditional agricultural exports (NTAE), these policies’ intensification of historical problems of unproductive deforestation led the adjustment to change its approach towards a new form of capital articulation through commodified and financialized forms of ‘conservation commodities’” (2017, 65). With policies such as the establishment of payment for environmental services (PES) and the fostering of agroforestry
Agro-extractivism, sustainable development 103 systems, the idea was “to promote forest cover as a means of safeguarding the goal of macroeconomic stabilization achieved by the adjustment” (ibid., 65). This was a multiscale process, in which the protection of nature was to be advanced through the creation of new types of commodities, such as PES, but also by promoting ecotourism into the rural areas, as a way of remedying the effects of extensive cattle ranching and monocrop expansion (Nygren 1998). In this way, it was argued, nature would stop being seen as idle and unproductive, and become a valued resource that could improve the well-being of rural communities, at the same time as it attracted foreign investment (Herrera-Rodríguez 2013). The idea that the degradation of nature resulted from the lack of valuation of resources such as the forest or land would become instilled into most of the conservation institutions, public policies, and laws that were created during this period. However, as Anja Nygren (1998, 204) argues, even within the supposed consensus regarding sustainable development, there are struggles between different “state institutions, international aid agencies, and nongovernment organizations, pursuing different goals and responding to different ideologies” She located four such ideologies in the country. (i) Environmentalism for nature, in which the central idea was the “protection of the country’s natural forests and conservation of its wildlife and diversity” (1998, 205). (ii) Environmentalism for profit, where “sustainable development means economic revitalization, where private enterprises and multinational companies are encouraged to invest in ecotourism, forest extractivism and biobusiness, in the name of making the country’s biodiversity an economically profitable commodity and loading the country’s environmental beauty with international market value” (ibid., 207). (iii) Alternative environmentalism, in which the Western divide between nature and culture is the reason behind the destruction of nature, and it is in Indigenous knowledge where the necessary answers and healing can be found. Finally, (iv) environmentalism for the people, where sustainability has to be sought in small-scale activities such as social forestry and small-scale agriculture. Nygren points out that environmentalism for profit is the dominant perspective in the country, with the support of both capital and most government institutions, and that whenever environmentalism gets in the way of profit, it can be moved to the background. At the same time, these understandings of sustainable development point towards very distinct political projects. In fact, according to Mauricio Herrera-Rodríguez (2013), the only thing that they have in common is that they all strive to promote “sustainability”, whatever that might mean. The enduring question is then, how can contradictory political projects cohabit the same development strategy with no apparent conflict? According to Antonio Gramsci (1971, 348), “common sense” is a set of immediate, unconnected facts that are taken for granted, and non-critical assertions on reality and the “way in which things are”, where “in a whole range of judgements common sense identifies the exact cause, simple and to hand, and does not let itself be distracted by fancy quibbles and pseudo-profound, pseudo-scientific metaphysical mumbo-jumbo”. Common sense is conservative in the sense that it gives the idea that things are “naturally” as they are, and thus, becomes crucial to the question of hegemony.
104 Andrés León Araya For Gramsci, domination in capitalist societies can never be seen as a simple matter of coercion, for consensus must also be created. Hegemony, in a sense, is the particular ways in which coercion and consent are articulated and sometimes equilibrated in different space–times to shape the relations between dominant and subaltern groups in an effort – never entirely successful or complete – to secure and reify a particular status quo (Buci-Glucksmann 1980). This is why common sense is so important, as it plots a sort of roadmap by which everyday life is read and understood. However, and this is crucial, common sense is far from an even or homogenous formation. Rather, To some degree, we all live in a commonsense world, just not the same one . . . We all continually channel the stream of events that wash over us into familiar narratives, making sense of what would otherwise appear as random. The knowledge we draw on to do this is derived both from the particular circles in which we move, and our own life experiences as these are mediated by the narratives available to us. Over time this knowledge comes to constitute a solid, emotionally persuasive core against which we test both what happens to us, and how others explain the world to us . . . At any historical moment, even within the same place, there will be multiple narratives, some closely connected and overlapping, some conflicting and contradictory, but all of which are, to some rational being, self-evident truths. (Crehan 2016, 47) In Costa Rica, a sustainable development common sense began to take form in the 1990s. This common sense was not built from scratch, and has as its foundations the historically apparent truth that the country’s progress and identity are tied to agricultural export crops – particularly coffee – (Quesada 2008), as well as the assertion made by many historians that a general sense of the importance of the conservation of nature was already present in the Costa Rican liberal reforms of the 19th century (Evans 2010; Viales Hurtado 2001). It would be during the government of José María Figueres (1994–1998) that these two apparent historical “truths” were fused together with the widespread idea of Costa Rican exceptionality,1 in what the president called “desarrollo sostenible a la tica”: sustainable development, Costa Rican style. Basically, desarrollo sostenible a la tica was the combined implementation of the general tenets of the environmentalism for profit model, with the deepening of neoliberal reforms, fashioned as a continuation of the country’s exceptional history. During his government, Figueres signed the country’s third Structural Adjustment Project with the World Bank, which included a set of very unpopular major cuts in the public sector, as well as the liberalization of the banking sector. He also prioritized foreign policy, signing a Free Trade Agreement with Mexico, and had an active role in the creation of the Central American Integration System in 1995. Here, sustainable development became the discourse in charge of stitching together foreign and domestic policy, as well as the idea of continuity and change – from developmentalist to neoliberal – within the development model.
Agro-extractivism, sustainable development 105 Echoing the Brundtland Report of 1987, desarrollo sostenible a la tica was officially defined as the search for “more general welfare in the present, while we ensure the great equilibriums that make our development possible in the long run”. According to the government, these equilibriums were the following: (i) political-institutional, looking to extend democratic decisionmaking; (ii) of the social structure, trying to breach the internal inequality gaps; (iii) economic, to attain long periods of sustained economic growth; and (iv) environmental, looking to harmonize social and economic life with the natural world (Monge 2015, 7). It was during the 1990s that the government began to actively promote and brand the country abroad as a “green” tourist destination and that much of the current Costa Rican conservation framework, including institutions, laws, and policies, was created. This sustainable development common sense has not been limited to the arena of the state and public policy. Through the inclusion of environmental topics in the curricula of both elementary and high school courses and the creation of media campaigns oriented towards educating the population regarding the importance of taking care of resources such as water and electricity, and, while it has widely circulated among most of the country’s social groups, it has been particularly effective among the urban middle classes. The geographical expression of sustainable development has been profoundly uneven and patchy at best. Urban areas have come to be dominated by the service economy, with activities such as call centres, light industrial assembling (hospital beds, for example), and commerce. While rural areas have been transformed either into export-oriented plantations (mainly bananas, palm oil, pineapples, and other tropical fruits), national parks, or tourist enclaves catering either to the adventurous spirits, or those who prefer the amenities of the first world in the midst of luscious nature. Investment in infrastructure follows this pattern, as the better roads spread out from the capital towards the plantations, ports, national parks, and tourist enclaves, while the means of communication within and between rural communities are neglected, to say the least. The result is a set of fragmented landscapes in which the experience, and thus the understanding of the country is profoundly differentiated along class and location lines, with a clear fracture between the urban and the rural (Herrera-Rodríguez 2013). It is within this context that we should understand the expansion of pineapple production in the country. 3. Creating a pineapple republic While growing pineapples in Costa Rica has a very long history, its specialization towards exports can be dated to the 1970s (Aravena Bergen 2005). During this decade, the transnational Fresh del Monte began exploring the possibility of moving their stagnant production of pineapples from Hawaii to places in Latin America with lower land and labour costs, as well as more favourable climate conditions (Bartholomew et al. 2012). The fact that a subsidiary of del Monte was already present and the country’s “political stability” made Costa Rica an attractive destination. By the late 1970s, the first plantations were created in Buenos
106 Andrés León Araya Aires of Puntarenas, in the south of the country, by the Pineapple Development Company (PINDECO) and expanded from there to different parts of the country in the following decades (Guevara et al. 2017). To understand why and how pineapples expanded so quickly, we need to look both at the process of structural adjustment that Costa Rica experienced during the 1980s and 1990s, and at the transformations of the global pineapple value chain. I will begin with the latter. Before the 1970s, due to the short shelf life of the fruit, the global pineapple value chain was dominated by canned fruit and juices. The pineapple companies were searching for a fruit variety that could be more “scalable” (Tsing 2012), in the sense of achieving a more homogeneous format in terms of maturation time, taste, quality, size, durability, and shelf life. In 1961, Dole, del Monte, and the Maui Pineapple Company formed a consortium to fund the creation of the Pineapple Research Institute (PRI), based in Hawaii. The PRI was later dissolved in 1975, but not before developing a set of new hybrid varieties. A particular hybrid, the “73–114”, caught the attention of the scientists due to its sweet flavour and yellow colouring to the point that it was later named the MD2, after Millie Dillard, the wife of the CEO of the Maui Pineapple Company (Amar et al. 2015). However, this variety was not immediately oriented towards commercial production. It would be until the late 1970s and 1980s, when several of the varieties created by the PRI were taken to Costa Rica to test their qualities, that it became evident to the Costa Rican agronomists in charge of the plantation’s research of the hybrid’s great potential. Not only did it have a sweeter flavour, and contained larger amounts of Vitamin C, it had a larger and more homogenous size than other varieties. To top it all, it had a longer shelf life, thus, potentially opening up markets further away, such as Europe. By the mid-1990s, del Monte had shifted all of its Costa Rican production towards the MD2 variety. At the same time, it secured a very controversial patent over the rights to the hybrid, which would translate into a long legal dispute with Dole. Finally, in 2003, del Monte was forced to drop their pretensions of having exclusive control over the MD2 hybrid, opening the door for other companies such as Dole and Fyffes to create and commercialize their own varieties of golden pineapples (Greig 2004; Solera and Porras 2017). The impact of the MD2 variety cannot be downplayed. It transformed the world market for fresh fruit, but also, the price at which it could be sold, which was more than three times higher than for other varieties (Amar et al. 2015). For example, in 1997, the MD2 variety represented only 20% of del Monte’s pineapple sales, but 64% of their profits (Solera and Porras 2017, 45). Locally, the impact of MD2 cultivation by del Monte had a concentrating effect, as less than 10% of the producers came to control more than 90% of the surface dedicated to the crop (CANAPEP, n.d.). Regarding the neoliberal agrarian transformation of Costa Rica, what came to be known as the process of structural adjustment was basically the deepening of the articulations of the national economy with the global market through economic liberalization, as well as an uneven process of privatization and institutional reform. In terms of the rural areas, the idea was to promote tourism, as well as a set of “new” export-oriented crops, such as pineapples, oranges, and
Agro-extractivism, sustainable development 107 palm oil, in an attempt to diversify the economic structure and take advantage of the national “competitive advantages” (basically, cheap labour and land) (Edelman 1999; León Araya 2015). A set of incentives were given to promote these crops. For example, Export Contracts (Law 6955), which exempted exporters from paying import tariffs for raw materials and equipment, as well as sales taxes; Tax Installment Payments (Certificados de Abono Tributario), which basically acted as exports subsidies; and the classification of processing and packaging plants as free zones (Law 7210) (Obando 2017). At the same time, support for the production of staple crops and other products oriented towards the domestic market diminished. For example, credit for the staple crops sector was cut by nearly 90% between 1980 and the 2000s, and technical support shifted away from these products, and towards the new export-oriented crops (Cerdas Sandí 2015; Edelman 1999). This transformation took place in a period in which the country’s agricultural land was contracting anyways. Between 1984 and 2014, the country’s total farmland decreased by 20% (around 600,000 hectares) (GRUTA 2017), in part due to conservation, but also as the broader process of urbanization unfolded. As such, the rural areas became a place of dispute between three forms of production. First, a set of small and medium producers of staple crops, other foodstuffs, and cattle, whose production was oriented mainly towards subsistence and the domestic market. Second, a group of larger producers, including cooperatives and domestically own companies dedicated to the production of the “traditional” monocultures (coffee, sugar cane, and rice). Finally, a group of transnational companies specialized mainly in the production of tropical fruits (bananas, pineapple, and oranges) for export, but that also controls most of the national food industry through their control of the supermarket sector (Alvarado and Charmel 2002). The resulting balance, as seen in the last decade or so, has been a significant increase in the market power of the third group, a complicated balance for the second one, with some producers managing to thrive, while others sink, and the general crisis of the first set of producers (GRUTA 2017). As a result, Costa Rica went from being a country that produced most of its foodstuffs before the 1980s, to one that imports about 34% of its rice, 69% of its maize, 73% of its beans, and 77% of cereals that are consumed domestically (Chacón 2014). It is also clear that it is in this power dispute, and not in the expansion of the almost inexistent agrarian frontier (deforestation), that we will find the explanation behind the expansion of the pineapple industry (Fagan et al. 2013). However, what is the dispute between different political and production logics is not presented, nor understood as such in the country. Here the common sense of sustainable development plays a significant role. For the particular case of pineapple production, the history of the activity in the last few decades takes the form of a process that hides in plain sight. Under the idea and discourse that pineapple production is highly profitable – although it is dependent on state subsidies – and strives towards sustainability, it went from a marginal crop in the 1970s to the country’s third most important export in 2018, with a total value of over one billion dollars.
108 Andrés León Araya As mentioned before, the export-oriented production of the crop began in Buenos Aires in the late 1970s, with the installation of PINDECO. From there, seeing the high prices that the fruit was receiving in the global markets, it expanded towards the northern region of the country (the Cantons2 of Los Chiles, Upala, and Guatuso), in the border region with Nicaragua. By the 2000s, it expanded into the country’s northern Atlantic region (especially the Cantons of Guácimo, Siquirres, and Pococí) (Valverde et al. 2016). What all these places have in common is that they were traditional agricultural regions that combined the presence of monocultures with significant production for the domestic market. However, this expansion would not have been possible if not for the technological changes that took place with the introduction of the MD2 variety. Before, the types of climatic conditions in which pineapples could be developed at a large scale were limited. However, with the new variety and its technological package, overly dependent on agro-chemical use (around 30 kg/ha as against the national average of roughly 18 kg/ha), the agrarian frontier was expanded through a technological fix. We can identify four mechanisms through which the expansion has taken place: 1
2
3
The beachhead: Sale of devalued lands. Besides climatic conditions, the reason that PINDECO began its operations in Buenos Aires was the large availability of cheap land. This area had been formerly a set of mainly underutilized cattle ranches with absentee owners, who were willing to sell lands that they deemed basically useless, on the cheap. Transition from above: From one monocrop to another. As it became evident that the activity was profitable, it began to expand into areas where other monocrops, such as rice, oranges, or bananas were grown, as well as into areas dominated by extensive cattle ranching. In these cases, the transition towards pineapples followed two paths: (i) through sales, as many struggling medium and large producers of traditional crops (for example, rice and cattle ranchers) were seduced to either rent or sell their land by the attractive prices being offered; (ii) activity conversion, by those producers who were able to access either state incentives or foreign financial support, and converted their stagnant or less profitable ventures into pineapple plantations. In this group we can find various long-standing members of the country’s landowning elite, who have held important posts in the national government.3 Subordinated incorporation. With the implementation of the structural adjustment programmes in the 1980s and 1990s, many of the peasant settlements that had been created by the state in the previous decades began to struggle (Edelman 1999). Not only was state support dwindling, but the prices of crops such as maize and beans were deteriorating, due to the lowering tariffs. As such, many of these settlements, particularly in the northern regions, were approached by pineapple companies who offered them financial and technical support, in exchange for exclusivity contracts. This did not only lead to relations of dependency in relation to the larger companies, but also, in many cases, to indebtedness and eventually to having to sell the land (Acuña 2006; Rojas 2006; Solera and Porras 2017; Valverde et al. 2016).
Agro-extractivism, sustainable development 109 4
Toxic landscapes. Pineapple production creates the conditions of possibility for its own expansion. The fumigant-dependent form of production of the plantations produces a “topography of difference”, where “inside are scenes of modular discipline; outside life – human and nonhuman – continues in its unruly riot, but now swept by plantation encouraged plagues” (Tsing et al. 2019, S187). For example, the waste produced when harvesting the fruit attracts a large type of fly (Stomoxys calcitrans) that feeds on cattle and stops it from gaining weight, ruining many nearby small and medium producers and forcing them to sell their lands. Also, the massive amount of pesticides used in the crop’s production drifts into nearby communities, contaminating households and fields, making any other agricultural activity almost unviable. Further, the use of heavy machinery by the plantations (harrows, backhoes, spray booms, etc.) destroys the ballast and dirt roads that they share with the neighbouring communities, making them unusable during the rainy season. As a result, it becomes much more challenging for other farmers to take their harvest to the market. This process has had a profound effect on the landscapes of the regions discussed. The social worlds of the neighbouring communities have been rattled and transformed significantly. For example, peasant households who used to work independently found themselves resorting to wage labour as a way of survival (Valverde et al. 2016). Particularly in the Caribbean, people who had occupied lands in an attempt to escape the labour conditions of the banana enclave found themselves landless again and forced to see their daughters and sons repeat their history in the pineapple plantations (Aguilar and Arroyo 2014). As we saw, it has also affected the country’s capacity to produce its own food, tilting ever more the scale towards the transnational companies that control both agricultural exports and imports. This, in turn, also transformed the dynamics between rural and urban spaces. As less and less of the national diet came to depend on domestic production, and the country’s reality came to be understood as predominantly urban and service oriented, the rural spaces came to be seen as either spaces of nature (pristine national parks and empty beaches) or export-oriented plantations. However, what could not be hidden is the discontent generated by the crop’s expansion.
4. The politics of environmentalism For the Costa Rican case – and to paraphrase Antonio Gramsci (1971) – I would argue that sustainable development operates in the interregnum created by the crisis of authority that resulted from the process of structural adjustment. It appeals to the need to construct consensuses and equilibriums, at the same time as scientific knowledge is used to justify and depoliticize the increasing levels of coercion needed to sustain domination. For the case of monocultures, such as pineapples, the formula has been to approach the social and environmental effects of their expansion as technical and managerial in nature.
110 Andrés León Araya As such, when people organize and protest against different effects of the monocrop plantations that are not immediately or obviously “environmental”, like the poisoning of their water sources or the pesticide drift towards the neighbouring communities and schools, they are immediately codified as “environmentalists”, that is as people who are concerned with the environment and nothing else. We find a good example in the case of the National Front of Sectors Affected by Pineapple Production (FRENASAPP), which was created in 2008 in the North Caribbean region in reaction to the poisoning by pesticides of the water sources of various peasant communities. While they mainly used the language of the right to health to frame their grievances, they were quickly presented as an “environmental organization”, which was clearly an oversimplification. This is not to say that environmental concerns are not present in their complaints. Rather, to enclose these grievances as merely or mainly environmental hides other elements that are also present, such as the dynamics of dispossession that have made the pineapple industry possible in the first place. Of course, this is why the sustainable development common sense is so effective. Grouping differentiated grievances against water poisoning, labour conditions, lack of economic linkages with local economies, among others, in a single category, not only depoliticizes them but presents solutions framed in the mould of “environmentalism for profit” as universal. At the same time, the language of environmentalism defines the contours between consensus and coercion. As I just mentioned, grievances presented in the language of environmentalists are able to enter public discussion. However, those grievances that do not conform and are unequivocally posed as political issues are repressed both physically and symbolically, for example, by presenting them as anti-progress or “communists”, or, as in the case of the public institutions, by dismissing most of their protests against monocrop plantations, on the grounds that no scientific evidence was presented. For example, in 2009, when the aforementioned communities organized under FRENASAPP protested that their water sources were contaminated, using as evidence a study done by the Autonomous National University of Costa Rica (UNA), the then minister of health responded: “we are delivering drinking water to them, because they are scared of drinking the one from the place, and there has been such a psychosis, that they even correlate spots in the skin that result from fungus, with showering in water that they say has Bromacil [a commercial brand of pesticide]” (Boeglin 2015). This is a perverse logic in which the affected communities and actors must demonstrate that they are victims, while the state, in practice, becomes the steward of these activities. Nevertheless, the language of environmentalism can also operate as a bridge between rural experiences and urban middle-class sensibilities. In fact, the largest protests against the expansion of pineapple production have taken place in the capital, San José, and with significant participation of environmentally conscious middle class. Although what urbanites, peasants, or Indigenous groups understand by this can be quite different, environmentalism both forges and limits the type of urban–rural coalitions that can be created. This is, in part, a problem of representation. Rural areas in Costa Rica tend to be presented to the general public in digital and mass media through images of nature (exuberant forests and empty
Agro-extractivism, sustainable development 111 beaches) and monoculture (coffee, sugar cane, oil palms, and pineapples), but rarely, with the people in charge of producing those landscapes (Mitchell 1996). Further, when presented in the media, it is as unruly and violent groups that block roads and are against progress. The result is an alienating experience in which monocultures stand for money, and sustainability stands for protected wildlands, seemingly without contradictions, with the experiences of rural communities being left out of the picture. As such, the country’s development strategy is organized around the promotion of agro-exports and tourism in the rural areas, combined with a service economy in the cities, which is highly dependent on the attraction of foreign investment. Foodstuff and consumption goods, in general, are supposed to be imported on the cheap, for which revenue must be created, through exports and foreign investment, to ensure the macroeconomic health of the system and job creation. Politically, the legitimacy of the system is based on the support of the urban working and middle classes, who are supposed to understand their well-being in relation to the promotion of exports and the attraction of foreign investment, while at the same time they do not have to carry the brunt of the environmental and economic impacts. As a result, activities such as monocrops are highly attractive and presented as opposite to the traditional visions of pollution related to factories and chimneys. This spatial fragmentation of the country and the image of coherence, which the sustainable development framework provides, organize how hegemony operates in and around monocrop plantations. It maps the relations between different activities and the positions that different actors have in relation to development: either as promoters (plantation owners) or backward minded opponents (community organizations). Accordingly, it also defines and legitimizes the support and dialogue with the former, and the repression and criminalization of the latter.
Conclusions As I write these lines, two events dominate my newsfeed. On the one hand, the United Nations selected Costa Rica as the 2019 Champions of the Earth Award for Policy Leadership due to the country’s programme to decarbonize its economy by 2050. On the other hand, on 6 November, the National Environmental Technical Secretariat (SETENA) approved a project presented by PINDECO to plant around 500 hectares of pineapple in Palmar Sur, in the country’s South Pacific region, while ignoring the objections presented by various environmental activists, archaeologists, and biologists regarding the irreparable environmental and cultural damage that the proposed pineapple plantation would have. What is significant about this project is that it is located close to various archeological sites that were declared World Heritage Sites by UNESCO in 2014 due to the enigmatic stone spheres that can be found all over the region. Further, the proposed plantation is close to the Terraba-Sierpe Wetlands, declared a Ramsar site in 1995. As argued in this chapter, what allows these two apparently contradictory events to coexist within the Costa Rican development strategy is the sustainable
112 Andrés León Araya development framework. While the plan to decarbonize the economy by 2050 is based on abstract cosmopolitan imagery, where technology will play an important part, PINDECO’s proposed plantation is located in a region hardly known by most urban-based Costa Ricans and visiting tourists. Further, since many pineapple companies are currently striving to have their production certified as “carbon neutral”, it could be claimed that plantations, not archaeological or Ramsar sites, are the path towards decarbonizing the economy. This is how sustainable development and agrarian extractivism operate together in Costa Rica through the creation of a common sense that allows contradictory and even opposing practices to take place at the same time, under a veneer of coherence. By presenting the problem of development as the combination between sustained economic growth and the conservation of nature, sustainable development displaces the discussion away from what is being destroyed, and towards what is being produced. Previous landscapes and ways of being in the world are either invisibilized or presented as backward minded and inefficient, thus justifying the dynamics of dispossession upon which the expansion of monocrop plantations is predicated. New plantation landscapes are celebrated, as they supposedly produce economic growth and jobs, and strive towards sustainability through the use of technology, ignoring or downplaying the destructive forces unleashed by the topographies of difference I have presented. At the same time, the Costa Rican case is a good example of the transnational forces at play behind the operation of agrarian extractivism. Nature is clearly being extracted through the destruction of the Costa Rican soils for the profit of transnational companies and the dietary delight of faraway consumers. However, as I have shown, much more is also happening. For example, the MD2 variety was created in Hawaii, but it only became a viable commodity in Costa Rica, after being developed by Costa Rican scientists. Further, the development and expansion of the activity in the country would not have been possible without the country’s public universities, their scientific research, and the preparation of highly skilled and specialized professionals. In fact, with the global expansion of pineapple production, many Costa Rican agronomists have been hired and have migrated to countries such as Ghana and Indonesia to help create and run new plantations. Furthermore, Costa Rican scientists are a recurrent feature in the sector’s global events, such as the International Pineapple Symposium. Also, the traditional image of extractivism as an activity solely controlled by transnational capital needs to be complicated. Although PINDECO is by far the largest pineapple producer in the country, Costa Rican capital is broadly involved in the activity. Moreover, most of the political operators that created the legal and institutional conditions of possibility of the expansion are also involved as owners of plantations. In this same line, the transition towards extractivism cannot be seen in unilateral and lineal terms. In Costa Rica, the expansion of pineapple production has followed four different trajectories: land sales, transition from above, subordinated incorporation, and the production of toxic landscapes. Predominantly, extractivist dynamics are seen as only destructive. However, it is
Agro-extractivism, sustainable development 113 clear that their operations are also productive, in the sense of creating new morethan-human orderings, with unexpected effects. This is an aspect that I believe, deserves more attention than what it has been given. Finally, we must also move beyond the territorial trap’s tendency that is present in many agrarian extractivist works, in which the national space is homogenous. For the Costa Rica case, the legitimacy and political support of the government and elite’s agrarian extractivist project come from the urban middle classes, who are highly dependent on the service economy and food imports, and have a highly alienated perspective of the country’s rural areas as either empty beaches, pristine parks, modern plantations or inefficient peasant, and Indigenous farms. As such, the transformation of the latter into any of the other ones is not only acceptable, but also desirable, as it is congruent with the sustainable development common sense, which highlights abstract economic growth and hectares of forest cover, and obviates the violence unleashed upon concrete communities.
Notes 1 Much of Costa Rica’s national identity is built around the idea of its exceptionality in relation to the rest of the Central American region in terms of the lack of armed conflicts and higher levels of social development. However, as different authors have also argued, this discourse has become a powerful tool to co-opt and defuse counter-hegemonic movements that may threaten the status quo (Acuña 2002; Jiménez 2005). 2 Costa Rica is administratively divided into Provinces, then Cantons, and lastly districts. 3 For example, Alfredo Volio, a member of the executive board of the company Upala Agrícola, was Minister of Agriculture and Husbandry, as well as a member of the board of directors of the Banco Nacional, the country’s largest public bank. In this group, we also find Carlos González Pérez, a former member of the e Central Bank’s board of directors, and owner of one of the country’s largest rice companies.
Bibliography Acuña, G. 2006. Producción de piña en caribe y pacífico sur de Costa Rica. Revista Ambientico, 158, 2–3. Acuña, V.H. 2002. La invención de la diferencia costarricense, 1810–1870. Revista de Historia, 45, 191–230. Agnew, J. 1994. The territorial trap: the geographical assumptions of international relations theory. Review of International Political Economy, 1(1), 53–80. Aguilar, D., and R. Arroyo. 2014. El amargo sabor de la piña: Movimiento socioambiental en el Caribe costarricense (Trabajo Social). San José: Universidad de Costa Rica. Alvarado, I., and K. Charmel. 2002. The rapid rise of supermarkets in Costa Rica: impact on horticultural markets. Development Policy Review, 20(4), 473–85. Amar, A.T., P.S. Tong, and C. Ng. 2015. The MD2 “super sweet” pineapple (Ananas comosus). UTAR Agriculture and Science Journal, 1(4), 14–17. Arauz, L.F. 2017. El equilibrio de piña y ambiente, 12 July. Available at: http://gobierno.cr/ el-equilibrio-de-pina-y-ambiente/ (Accessed 27 February 2018). Aravena Bergen, J. 2005. La expansión piñera en Costa Rica: La realidad de los perdedores de la agroindustria exportadora de la piña. San José: COECO-CEIBA.
114 Andrés León Araya Araya, J. 2017. Expansión piñera se comió 5.568 hectáreas de bosque. Semanario Universidad, 8 March. Available at: https://semanariouniversidad.com/pais/expansion-pinerase-comio-5-568-hectareas-bosque/. Arguedas, D. 2015. Pineapple industry leaves Costa Rican communities high and dry. IPS Press Service, 25 May. Available at: www.ipsnews.net/2015/05/pineapple-industryleaves-costa-rican-communities-high-and-dry/. Bartholomew, D.P., R.A. Hawkins, and J.A. Lopez. 2012. Hawaii pineapple: the rise and fall of an industry. HortScience, 47(10), 1390–8. Boeglin, N. 2015. La piña de costa rica ante la comisión interamericana de derechos humanos. International Network of Human Rights, 14 April. Available at: http://ridh.org/ news/la-pina-de-costa-rica-ante-la-comision-interamericana-de-derechos-humanos/. Buci-Glucksmann, C. 1980. Gramsci and the State. London: Lawrence & Wishart. Carazo, E., J. Aravena, V. Dubois, J. Mora, F. Parrado, and A. Mora. 2016. Condiciones de producción, impactos humanos y ambientales en el sector piña en costa rica. San José: Oxfam. Cerdas Sandí, D. 2015. El proceso de financierización de la banca pública costarricense: El caso del banco nacional, 1995–20051. Revista de Ciencias Sociales, 4(150). Chacón, K. 2014. El desafío de garantizar la seguridad alimentaria y nutricional. San José: Estado de la Nación. Coronil, F. 1997. The Magical State: Nature, Money, and Modernity in Venezuela. Chicago: University of Chicago Press. Crehan, K. 2016. Gramsci’s Common Sense: Inequality and Its Narratives. Durham: Duke University Press. Diepens, N.J., S. Pfennig, P.J. Van den Brink, J.S. Gunnarsson, C. Ruepert, and L.E. Castillo. 2014. Effect of pesticides used in banana and pineapple plantations on aquatic ecosystems in Costa Rica. Journal of Environmental Biology, 35(1), 73. Echeverría-Sáenz, S., F. Mena, M. Pinnock, C. Ruepert, K. Solano, E. De la Cruz, . . . C. Barata. 2012. Environmental hazards of pesticides from pineapple crop production in the Río Jiménez watershed (Caribbean Coast, Costa Rica). Science of the Total Environment, 440, 106–14. Edelman, M. 1999. Peasants Against Globalization: Rural Social Movements in Costa Rica. Stanford: Stanford University Press. Evans, S. 2010. The Green Republic: A Conservation History of Costa Rica. Austin: University of Texas Press. Fagan, M.E., R.S. DeFries, S.E. Sesnie, J.P. Arroyo, W. Walker, C. Soto, . . . A. Sanchun. 2013. Land cover dynamics following a deforestation ban in northern Costa Rica. Environmental Research Letters, 8(3), 034017. Gago, V., and S. Mezzadra. 2017. A critique of the extractive operations of capital: toward an expanded concept of extractivism. Rethinking Marxism, 29(4), 574–91. Gramsci, A. 1971. Selections from the Prison Notebooks of Antonio Gramsci. Q. Hoare and G. Nowell-Smith (eds.). London: Lawrence & Wishart. Greig, I. 2004. Pineapple wars redux. Chronica Horticulturae, 44(2), 5. GRUTA. 2017. Costa Rica: Expansión del capitalismo en el campo y sus estrategias territoriales. San José: Grupo de Estudios Agrarios, UCR. Gudynas, E. 2009. Diez tesis urgentes sobre el nuevo extractivismo. Extractivismo, Política y Sociedad, 187–225. Guevara, A., R. Arce, and P. Guevara. 2017. Economic, Social and Environmental Impact of Pineapple Farming in Costa Rica. San José: INCAE.
Agro-extractivism, sustainable development 115 Herrera-Rodríguez, M. 2013. Sustainable development in Costa Rica: a geographic critique. Journal of Latin American Geography, 12(2), 193–219. León Araya, A. 2015. Desarrollo geográfico desigual en Costa Rica: El ajuste estructural visto desde la Región Huetar Norte. San José: Editorial de la Universidad de Costa Rica. Maglianesi-Sandoz, M.A. 2013. Desarrollo de las piñeras en Costa Rica y sus impactos sobre ecosistemas naturales y agro-urbanos. Biocenosis, 27. McKay, B.M. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211. Mezzadra, S., and B. Neilson. 2017. On the multiple frontiers of extraction: excavating contemporary capitalism. Cultural Studies, 31(2–3), 185–204. Mintz, S.W. 1985. Sweetness and Power. New York: Viking. Mitchell, D. 1996. The Lie of the Land: Migrant Workers and the California Landscape. Minneapolis: University of Minnesota Press. Monge, C. 2015. Desarrollo sostenible a la tica: Geopolítica y ambiente en la Administración Figueres Olsen (1994–1998). Revista Rupturas, 5(1), 1–21. Moore, J.W. 2010. The end of the road? Agricultural revolutions in the capitalist worldecology, 1450–2010. Journal of Agrarian Change, 10(3), 389–413. Moore, J.W. 2015. Capitalism in the Web of Life: Ecology and the Accumulation of Capital. London: Verso Books. Nygren, A. 1998. Environment as discourse: searching for sustainable development in Costa Rica. Environmental Values, 201–22. Obando, A. 2017. El Estado detrás de la piña: El conflicto socioambiental del monocultivo de piña los cantones de Upala, Guatuso y Los Chiles (2000–2015) (Licenciatura en ciencias políticas). San José: Universidad de Costa Rica. Quesada, R. 2008. Ideas económicas en Costa Rica (1850–2005). San José: EUNED. Ramírez, A. 2017. A Political Ecology of Neoliberal Multiculturalism: Social Inclusion and Market-Based Conservation in Indigenous Costa Rica (To obtain the degree of Doctor). Rotterdam: Erasmus University. Rojas, J.A. 2006. Ayer peones, hoy productores y exportadores de piña. Ambientico, 158, 4–5. Rojas, P. 2017. Piñeros refutan estudio: “La expansión no ha existido”. CRHoy.Com, 4 May. Available at: www.crhoy.com/ambiente/pineros-refutan-estudio-la-expansionno-ha-existido/. Silvetti, F. 2015. La expansión de monocultivos de exportación en Argentina y Costa Rica: conflictos socioambientales y lucha campesina por la justicia ambiental. Mundo Agrario, 16(32). Solera, M.C., and R.D. Porras. 2017. Posibilidades locales de desarrollo en presencia de enclaves: Caso de la Asociación de Productores de Piña de la comunidad de Utrapez, ubicada en la Zona Sur de CR. Perspectivas Rurales Nueva Época, 29, 43–72. Soluri, J. 2009. Banana Cultures: Agriculture, Consumption, and Environmental Change in Honduras and the United States. Austin: University of Texas Press. Striffler, S. 1997. In the Shadows of State and Capital: The United Fruit Company and the Politics of Agricultural Restructuring in Ecuador, 1900–1995. Durham: Duke University Press. Svampa, M., and E. Viale. 2014. Maldesarrollo: La Argentina del extractivismo y el despojo. Volume 3088. Buenos Aires: Katz Editores. Swyngedouw, E. 2010. Apocalypse forever? Theory, Culture & Society, 27(2–3), 213–32. Tsing, A.L. 2005. Friction: An Ethnography of Global Connection. Princeton: Princeton University Press.
116 Andrés León Araya Tsing, A.L. 2012. On nonscalability: the living world is not amenable to precision-nested scales. Common Knowledge, 18(3), 505–24. Tsing, A.L., A. Mathews, and N. Bubandt. 2019. Patchy anthropocene: landscape structure, multispecies history, and the retooling of anthropology. Current Anthropology, 60(20), 186–97. Valverde, K., A. Jiménez, and M. Porras. 2016. La expansión por omisión: Territorios piñeros en los cantones Los Chiles, Upala y Guatuso, Costa Rica (2004–2015). San José: Estado de la Nación. Viales Hurtado, R.J. 2001. Las bases de la política agraria liberal en Costa Rica. 1870– 1930. 1 Una invitación para el estudio comparativo de las políticas agrarias en América Latina. Diálogos Revista Electrónica de Historia, 2(4).
6
Gender inclusion in the sugarcane production of agrofuels in coastal Ecuador Illusionary promises of rural development within a new agrarian extractivism Natalia Landívar García1
Introduction The Ecuadorian state promotes the expansion of sugarcane in Guayas province while using a discourse of energy sovereignty, environmental sustainability, and rural development. In 2010, the Ecuadorian government established the Ecopaís programme that aimed to lower the costs of imported high-octane gasoline through a 5% sugarcane-based ethanol blend mandate, while reducing carbon emissions and strengthening the sugarcane agro-industry with new investments. Agro-fuels form part of the productive and energy transformation process pursued by the PAIS Alliance, a self-described political “movement”, or party in power since 2006–2007. The successive governments of the PAIS Alliance have promoted the transformation of the productive and energy matrix in efforts to overcome the problems generated by Ecuador’s historic primary-export and extractive-led economic model. As argued by its proponents, this policy differs from traditional capital accumulation since it is based on principles of “solidarity, responsibility and social inclusion” (Marx Carrasco 2014). The goal is to create diversified and greater value-added production and develop new knowledge and capacities with regards to technology, eco-efficiency, and biodiversity (SENPLADES 2012). Most scholars, however, argue that the much desired productive diversification and industrialization have been minimal and that Ecuador continues to depend heavily on the exportation of primary products (Larrea and Greene 2018; Larrea and Larrea 2017; Dávalos and Albuja 2014). In fact, according to the Central Bank of Ecuador (2020) the export of oil, bananas, and shrimp continues to be the country’s main exports, with oil as the principal source of export revenue. Indeed, the development approach implemented by PAIS Alliance has been referred to as “new or neo-extractivism” (Acosta 2013; Gudynas 2009; Veltmeyer 2013; Veltmeyer and Petras 2014). It departs from extractivism, the historic mode of accumulation pursued by industrial centres to exploit natural resource-endowed countries, causing national economic distortions, destruction of nature, and
118 Natalia Landívar García dispossession of peoples’ livelihoods (Acosta 2013). The “new” part has been adopted by progressive left governments, in contrast to neoliberal governments, as a new form of extractivism that still rely on natural resource extraction; yet it involves the participation of a sovereign state in the redistribution of extractive rents for social policies and infrastructure to achieve value-added industrialization and economic growth (Gudynas 2009; Arsel et al. 2016). National sovereignty over extractive rents constitutes one of the most popular discourses of the PAIS Alliance (Dávalos and Albuja 2014). Certainly, the fiscal resources that were made available from high oil prices between 2010 and 2012, as well as bilateral loans along with the improved fiscal policy and administrative capacity of the state, allowed the Ecuadorian state to invest in public infrastructure and social programmes (Larrea and Greene 2018). However, according to Larrea and Greene (2018), the concentration of land and capital remained extremely high despite PAIS Alliance’s promise of productive matrix transformation and an “agrarian revolution”. North and Grinspun (2016) argue that the new extractivism involves developmentalist policies that reinforce historic patterns of asset concentration while ignoring the transformative potential of agrarian reforms for longterm poverty reduction. This is true for Ecuador, where extractivist projects are violently being imposed with increasing evidence of its negative consequences for the environment and rural livelihoods (Artacker 2018; Bravo 2020; Davidov 2013; Arsel 2012; Wilson and Bayón 2017) while the “social debt”2 to the agrarian sector continues to be a reality for many. A new area of research has emerged to analyse how the new extractivism appears in the agrarian sector. With specific reference to Argentine soy production, Gudynas (2010) notes that external market demand has triggered an ongoing and massive geographical expansion and increasing production of capital- and input-intensive transgenic soy destined exclusively for export, with limited or no processing, few possibilities for capturing the soy export surplus, and scant opportunities for employment generation. In an attempt to distinguish such dynamics from previous processes of agrarian transformation, an emergent literature has started to deal with the constitutive characteristics of the new agrarian extractivism (Alonso-Fradejas 2018; Ezquerro-Cañete 2016; McKay 2017). In his study of the soy expansion promoted by a progressive government in Bolivia, McKay (2017) presents a characterization of a new agrarian extractivism that involves massive volumes of raw materials that are exported with little or no processing and that, compared to other crops, concentrate more cultivated land and earn the greatest amount of export revenues. Moreover, only a few domestic and transnational companies concentrate control over the value chain, thus appropriating the surplus value generated in all activities that make up the soy chain. According to McKay (2017), all this results in the degradation of the environment due to unsustainable farming practices, and the deterioration of work opportunities and conditions. New agrarian extractivism unfolds within rapid and profound agrarian transformations at a global level that has increased the demand for flex or multiple purpose crops and the profit of traditional and emergent agro-industrial capital
Gender inclusion in sugarcane production 119 (Borras Jr. et al. 2016; Gillon 2016). For Borras et al. (2016), “flex crops” refer to the capacity to vary a crop’s multiple uses – food, agro-fuel, feed for animals, and other industrial applications, given the existing physical, technological, and financial conditions of production. The land-use and crop-use changes associated with flex monocultures are considered to be the primary driver of land and natural resource grabbing that have led to negative social and environmental effects (Borras Jr et al. 2012, 2016). Furthermore, the incorporation – and potential benefits (or harms) – of small-scale farmers as suppliers of raw materials in the flex crop value chain, remains a challenge for scholarly research (Borras Jr. et al. 2016). In this regard, using an agrarian political economic perspective, White and Dasgupta (2010) propose to look at the conditions under which crops are produced and processed, and the actors who control the added value in the production and processing that convert the raw material into agro-fuels and other products. These are relevant aspects for a characterization of the new dynamics of agrarian extractivism in the Ecuadorian context. Under the pretence of benefiting small-scale sugarcane associations from the agro-fuel policy, the Ecuadorian state implemented the Agroenergía programme from 2013 to 2018. My research focuses on the associations in the Hacienda La Indiana in the coastal Guayas province that acquired land under the statesponsored land policy Plan Tierras and became part of the agrarian programme Agroenergía to complement the higher demand of sugarcane for ethanol production. Following McKay (2017), this chapter analyses the new extractive dynamics of the current sugarcane production in Guayas province and the inclusion of small-scale producer associations in developing the agro-fuel programme Ecopaís. I contend that an extractive agro-fuel project is currently being implemented by the Ecuadorian state with a social inclusive component used to legitimize the oligopolistic power of sugarcane mills and their distilleries. Furthermore, I argue that, through public resources, the sugarcane mills and their distilleries control not only the production of sugarcane but also the processing whereby this crop becomes ethanol and realizes added value. This research adds to the recent literature on the extractive character of flex crop monocultures by offering new insights into the social–economic–ecological implications of including small-scale farmer associations, especially, the disproportionate impacts on women involved in an extractive and male-dominated dynamic. The chapter is based on qualitative research conducted between September 2018 and August 2019. In addition to reviewing relevant existing data sources, 31 semi-structured individual interviews were conducted with actors who were directly or indirectly involved in the state agro-fuel programme. These actors included representatives of national peasant organizations, public sector officials, representatives of the private sector, small-scale alcohol producers, and agrarian experts. Among these actors, I interviewed seven women and nine men from four associations that acquired land in the Hacienda La Indiana, who speak about their particular gender-differentiated experiences in sugarcane production. The remainder of this chapter is organized as follows. First, I provide a brief overview of the historical context of sugarcane production in Guayas. Second,
120 Natalia Landívar García I analyse the fundamental changes under which ethanol production is unfolding to facilitate the extraction of surplus value and extractive rent by the sugarcane mills and their distilleries. Third, I turn to an analysis of the integration of smallscale producer associations in the production of sugarcane through the Agroenergía programme. In this part, I analyse the increasing economic dependency on agro-industrial inputs, the gendered division of labour, as well as the implications of a highly capital- intensive and external-input-intensive agricultural model on nature, human health, and social reproduction. The last section provides a concluding discussion of my results.
From sugarcane to the ethanol: a brief history of extraction The sugarcane industry has a long history of influence in the capitalist development of Ecuador, shaping access to and control over land and production, as well as social conditions in the rural sector (Guerrero 1978). Although the origins of sugarcane production in the country can be traced to the 17th century in the highlands, industrialization of sugar and alcohol production began only at the end of the 19th century in the coastal provinces (Fisher 1983). As documented by Guerrero (1978), the industry was initially directly linked to the cocoa boom in the coastal provinces of Guayas and Los Ríos, for it was the hegemonic class of cocoa agro-exporters and its related banking bourgeoisie with political influence that started investing in sugarcane production. The cocoa boom period (1840– 1910) had led to the concentration of the most productive coastal lands, acquired through economic and violent mechanisms such as the acquisition of confiscated properties in mortgage default, forced purchase of land below market prices, and land grabbing from peasants and Indigenous communities that did not have formal land titles (Maiguashca 2012). During the first decades of the 20th century, the Banco Comercial y Agrícola, the former owner of San Carlos sugarcane mill, started to expand its investments in modern machinery for sugarcane production, benefiting from cheap labour and the then high international price of sugar (Guerrero 1978, 2017) In his analysis of the sugarcane mills during 1900–1954, Guerrero (1978) documents that in the 1950s, only two sugarcane mills, San Carlos and Valdéz, successfully consolidated their market positions by appropriating capital from the exploitation of the labour force and the acquisition of large tracts of land from small-scale sugarcane mills that were not able to compete in the market. A process of proletarianization ensued as peasant tenants were evicted from the haciendas of cocoa producers that fell into bankruptcy and were bought by the large sugarcane mills. For the most part, the displaced coastal peasants and Indigenous people from the highlands, who were hired as permanent or temporary sugarcane workers, were men, since there is no record of the presence of female workers in the sugarcane plantations (Flórez Holguín 2011). Guerrero (1978) documents monopolistic market practices, such as planned restrictions of sugar production to generate shortages, speculation, and price increases, along with media campaigns, all of which facilitated control of the market by the San Carlos and Valdéz
Gender inclusion in sugarcane production 121 sugarcane mills. Under these conditions, sugarcane became the most important export product during the 1960s and mid-1970s, after the banana crisis and before the oil boom started in the 1970s. This momentum prompted the former military government (1963–1969) to nationalize two major sugarcane mills, TAINA and AZTRA (Azucarera Tropicana Americana) (Fisher 1983). The 1960s was also a period characterized by violence against sugarcane workers (Harari et al. 2012) and the increased influence of agro-industrial enterprises in agrarian and land policies and institutions. In that regard, North (1985) argued that the political and economic influence of the dominant capitalist class of the coast – that includes sugarcane mill owners – was a determining factor in undermining the potentially democratizing impacts of the agrarian reforms that were pursued by military governments in the 1960s and early 1970s. At the same time, the Ecuadorian state made resources available for the modernization of these powerful agro-industrial groups that benefited from public policies at the expense of peasant livelihoods (North 1985). For Flórez Holguín (2011), these policies were crucial for stabilizing the position of the sugarcane mills in the national economy. During the 1990s, neoliberal agrarian policies and a deregulated land market exacerbated the existing high level of land and market concentration in the hands of rural elites and shaped the agrarian structure for years to come. Indeed, these were the conditions under which the three sugarcane mills San Carlos, Valdéz, and Coazúcar (formerly AZTRA) consolidated an oligopolistic market for sugarcane in the last decades of the 20th century (Nicolalde Herrera 2014).
Setting new rules for ethanol extraction Sugarcane production, along with palm oil and banana, occupy the largest extension of land in the country (INEC 2020). As shown in Figure 6.1, although the cultivated area of sugarcane has fluctuated during the last 18 years, it had increased significantly compared to the first decade of the new century, when agro-fuels were declared of national interest as a result of the forceful lobbying efforts of the sugarcane mills. Sugarcane production increased from 5 million metric tonnes (MT) in 2002 to more than 9 million MT in 2019 while the cultivated area increased from 84,420 to 126,000 hectares during the same period. Figure 6.1 also shows that it is in Guayas province where the cultivated area doubled. Today, the same official source calculates that the coastal province of Guayas produces 80% of Ecuador’s sugarcane and allocates 90% of its cultivated land area to this flex crop. The National Agricultural Census (2000) documented that the high levels of land concentration in Guayas province are directly linked to the production of sugarcane. The census also reveals that the districts where the three most prominent sugarcane mills – San Carlos, Valdez, and Coazúcar3 – are located are among those with the highest levels of land concentration in the country. The apparent stability of agro-industry, as well as government incentives for agro-fuel production, according to a sugarcane expert, expanded the agricultural land frontier and brought new small- and medium-scale sugarcane producers into the sector (interview with Francisco Alemán, Guayaquil, 1 July 2019). Sugarcane
122 Natalia Landívar García 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0
Guayas Province
Total Hectares
Figure 6.1 Sugarcane cultivated area in hectares (2002–2019) Source: Data compiled from National Statistic and Census Institute (INEC)
expansion fostered by the Ecopaís programme, however, has not altered the concentrated ownership of land, which remains mostly in the hands of the largest sugarcane mills, San Carlos, Valdéz, and Coazúcar. Even though there is no recent official data available on the concentration of land in the hands of the three largest sugarcane mills, it is estimated that in 2008 they owned around 57,000 hectares, which represents 55% of the total hectares used for the production of sugarcane (Figueroa De La Vega 2008). Small- and medium-scale sugarcane producers from surrounding areas supply the remainder of the sugarcane production to the mills. However, it is not only the appropriation of land and control of production that stand out in the production of sugarcane. Since the majority of sugarcane production by the main mills is harvested mechanically (CINCAE 2019), employment in the sector and the region has been reduced significantly, affecting, in particular, the “zafreros” or cane cutters. Much of the field labour is now performed mainly by foreign workers (primarily from Venezuela and Peru), who are paid less than Ecuadorians and are outsourced, that is, they do not have formal labour relations with the sugarcane mills (interview with Ángel Rivero, Guayaquil, 26 December 2018). Despite labour regulations introduced by the PAIS Alliance government, the precariousness of the labour force persists in new forms of subcontracting that allow sugarcane mills to lower labour costs without making major investments (Harari et al. 2012). In 2017, 14,450 workers were involved in the production of sugarcane, half of them casual and informal workers (MAG 2017). The same source reveals that only 12% of the workers were women, confirming that work related to the production of sugarcane is male dominated.
Gender inclusion in sugarcane production 123 The processing facilities of ethanol are concentrated in the three main distilleries of Soderal, Codana, and Producargo that are owned by the three largest mills San Carlos, Valdéz, and Coazúcar. These distilleries were established at the end of the 1980s and beginning of the 1990s for the recycling of the molasses obtained from the residue of sugarcane processing. Initially, molasses was used exclusively for ethyl alcohol production destined for the alcohol, pharmacological, and cosmetology markets. With new technological advances, sugar mills and their distilleries have expanded their product portfolio so that by the early 2000s they began distilling anhydrous ethanol to supply the chemical industry. With the beginning of the Ecopaís programme in 2010, the main distilleries started using anhydrous alcohol for the generation of ethanol. The ethanol is sold to EP-Petroecuador, the public enterprise in charge of the production of the sugarcane agro-fuel and its distribution and commercialization in the national petroleum market. The support of the PAIS Alliance government under the leadership of Rafael Correa (2007–2017) was crucial for the development of the ethanol agro-industry. This becomes evident when we look into the 2015 visit of former Vice President Jorge Glass4 to the Soderal distillery, owned by the San Carlos sugarcane mill. As recognized by the San Carlos sugar mill board president, the incentives created by the PAIS Alliance government encouraged the Soderal distillery to expand its investments (Ingenio San Carlos 2015). There were essentially two kinds of incentives. First, through Executive Decree 675 of May 2015, a new ethanol price was set to match the US Gulf coast price index published by the agency Argus.5 Although the price depends on the daily ethanol average pricing determined by Argus, the government guarantees the distilleries a “floor” price of US$0.90 cents per litre. Earlier, ethanol prices had depended on decreasing international sugar prices. Second, both distilleries Codana and Soderal signed investment agreements of US$7 million and US$12 million, respectively, with the Ministry for Industry and Productivity (MIPRO),6 the institution that coordinates the implementation of the Ecopaís programme. As a result of this public–private partnership arrangement, in 2016, Codana and Soderal doubled their ethanol processing capacity from its previous level, from 34.65 million litres to 74.45 million litres per year. Furthermore, in 2016, the combined total private and public investment of US$350 million was an important contribution to the implementation of the Ecopaís programme (Ordóñez Saénz 2016). These incentives clearly demonstrate the ways in which the distilleries appropriate public resources and surplus value from the production of ethanol. According to the Alcohol Producers Association of Ecuador (APALE),7 established in 2005 by the main sugarcane mills, the ethanol floor price corresponds to the distilleries production costs for producing a litre of ethanol (interview with Carlos León, Guayaquil, 21 January 2019). Representatives of APALE argue that the ethanol industry represents savings for the state by reducing currency outflows for imports of high-octane gasoline. However, so far, no official information about the real production costs of the distilleries nor the state benefits in producing agrofuels have been publicly disclosed. Furthermore, insignificant employment generation through the expansion of processing facilities has been acknowledged by
124 Natalia Landívar García MIPRO (Mendieta 2019). Thus, as a way to compensate for this situation, MIPRO demands that the distilleries buy artisanal alcohol produced by small-scale associations from the highland provinces at a price of US$1.04 per litre, and at the level of 4% of the total amount they provide to EP-Petroecuador.8 According to one representative of the distilleries, the associations provide a low-quality alcohol that represents an extra cost of US$0.04 cents per litre in the ethanol processing. However, as he recognizes, it is the economy of scale that makes ethanol production a profitable business for the distilleries (interview with Emilio Oneto, Milagro, 28 June 2019). During the third mandate of the PAIS Alliance government under Lenin Moreno, potential investments totalling US$460 million to the sugarcane industry were projected. However, neither new investments nor previous agreements with distilleries and EP-Petroecuador were renewed immediately in 2018 due to opposition from the Ministry of Hydrocarbons to the Ecopaís policy. It seems that the opposition to Ecopaís is the response of the entire petrochemical sector, including state institutions, that are interested in ensuring their main source of income, namely petroleum-derived fuel (GAIN 2018). The above may be the case, but the opposition to Ecopaís needs to be considered also in light of the government’s austerity policies that questioned the feasibility of maintaining subsidized gasoline prices, including the Ecopaís price.9 More importantly, the Ministry of Hydrocarbons’ attempt to reduce the ethanol floor price of US$0.90 cents paid to the distilleries, due to the economic losses this represents for the state, has become the most contentious issue between this institution and the distilleries. According to one media report, the government has complained that the price paid to the distilleries is significantly higher than the cost of importing ethanol and the poor production capacity of the distilleries has reduced the actual blend of ethanol in Ecopaís (España 2018). Certainly, the Ministry of Hydrocarbons’ posture hints at an inefficiency in the production of ethanol. Control over the production and processing of ethanol is exercised by the three distilleries owned by historic oligopolistic sugarcane mills. The extractive nature of ethanol can be identified, in part, in the increased production volume and cultivated area of sugarcane, especially in Guayas province, triggered by the establishment of the Ecopaís programme. More importantly, sugarcane mills and their distilleries continue to appropriate surplus value derived not only from their land but also the land of small- and medium-scale producers who supply them with sugarcane, thereby benefiting from the continuation of the historic levels of land concentration and labour precariousness in Guayas. The officially set ethanol price and public–private partnership are mechanisms through which the distilleries have extracted state resources to expand their processing facilities for ethanol. This, in turn, was made possible through lobbying and political alliances that allow the distilleries to make ethanol into a profitable business despite their inefficient productive capacity. Artisanal alcohol purchases from small-scale producer associations were supposed to be the way through which the distilleries would reinvest their revenues back into the economy. However, the low quota of artisanal alcohol assigned to the associations, the lack of adequate technology to
Gender inclusion in sugarcane production 125 produce high-quality alcohol, and recent regulations for alcohol imports, are all reasons to doubt the sustainability of this social inclusion component. In the following sections, I will continue to characterize the extractive nature of ethanol as I analyse the integration of the small-scale sugarcane producer associations of Hacienda La Indiana in ethanol production through the Agroenergía programme.
Examining the integration of small-scale producers into the ethanol production In 2011, through the Plan Tierras, four associations – 27 de Octubre, Buen Vivir, La Indiana, and Arroceros de Cone – acquired the Hacienda La Indiana in the parish of Puerto Inca, in the province of Guayas. The 1,400-hectare sugarcane Hacienda was confiscated from the Isaías Group, a group of corrupt bankers who were held responsible for the country’s financial meltdown in the late 1990s. A single associative title was granted to the four associations composed of 250 individuals in total, 30% of whom were women.10 Table 6.1 shows that all but one of the four associations have a female membership of nearly or over 30%. This can be attributed to the constitutional principles of gender equality that influenced the development of policies within the Ministry of Agriculture during the Rafael Correa administration, leading to the incorporation of at least 30% female membership in the associations that benefited from Plan Tierras (Deere 2017). The pooling of female and male members’ economic revenues through joint cultivation of sugarcane was aimed at guaranteeing payments on the mortgage taken out by the associations through the Plan Tierras (Ericsson 2013). Two years after the Hacienda La Indiana was acquired by the associations, the national programme Agroenergía was launched in the Hacienda. Agroenergía, complementary to Ecopaís, was a programme pursued by the Ministry of Agriculture from 2013 to 2018. Its overall objective was to increase the land under sugarcane production annually by 10,000 hectares, to reach 83,000 hectares, and increase sugarcane production to 800 million litres of ethanol by 2020 (Diario El Telégrafo 2013). Agroenergía included an agreement of joint implementation, in which the Ministry of Agriculture committed itself to provide agricultural inputs and technical support while the associations in Hacienda La Indiana Table 6.1 Membership in the associations of the Hacienda La Indiana
Buen Vivir 27 de Octubre Coop. La Indiana Arroceros de Cone Total
Total members
Women
Men
% Women
38 51 115 46 250
12 6 45 13 76
26 45 70 33 174
32 12 39 28 30
Source: Interviews with the associations in the Hacienda La Indiana
126 Natalia Landívar García were responsible for the production of cane seedlings that would be transplanted elsewhere. The goal of making the Hacienda La Indianainto a “semillero de caña”– the seedbed for the provision of sugarcane for the production of agro-fuels – raised expectations that 250 men and women would not only have access to land, but that they would also receive support and a guaranteed market from the state. Indeed, Agronergía was developed on the premise that the state would invest in the creation of a public sugarcane mill that would be able to absorb the increased production of sugarcane (interview with SC, Guayaquil, 2019). Thus, the inclusion of small-scale producers in sugarcane production under an associative title, along with the possibility of starting a state-led process capable of creating employment and value-added industrialization was an innovative initiative. However, the shortcomings of the programme were notorious by 2015 when the cane seedlings could not be transplanted elsewhere due to lack of available fields (interview with SC, Guayaquil, 2019). Also, in the absence of a public sugarcane mill, the Ministry of Agriculture asked the neighbouring sugarcane mill, Coazúcar, to establish contract farming arrangements with the associations.11 Up to the present date, the associations continue to sell cane to Coazúcar under an annual harvest contract. Participation within this corporate-controlled production model has resulted in increased dependency on agro-industrial inputs, a gendered division of labour, and a double burden on women. These arrangements are destroying the environment and the reproductive conditions of the women and men in the Hacienda La Indiana, and are precisely the kinds of relations that are constituent elements of new agro-extractive dynamics in Ecuador. Increasing economic dependency on agro-industrial inputs Seeds make up an important part of the costs for sugarcane producers, especially during the first planting season. In 2013, the Agroenergía programme provided funding to replace the sugarcane stalks in the Hacienda La Indiana that were already 8–10 years old (interview with Franklin Martínez, La Indiana, 21 May 2019). Today, seed costs are among the main concerns of the association members since they cannot afford to replant the fields with certified seeds.12 Sugarcane producers who supply the three main sugarcane mills depend on the certified seed varieties produced by these same mills. Different varieties of domestically certified seeds have been developed during the last decade by a research centre sponsored by the sugarcane mills (CINCAE 2019). The introduction of domestic seed varieties was intended to replace the Ragnar seed varieties imported from other countries, while guaranteeing the availability of seeds to the sugarcane mills for their own production and commercial activities (Diario El Telégrafo 2011).13 While wealthier sugarcane producers buy certified seeds directly from the mills, the majority of producers must enter into relations of debt and dependency with the sugarcane mills who have oligopolistic control over the sugarcane complex (interview with Ángel López, Guayaquil, 24 June 2019).
Gender inclusion in sugarcane production 127 Association members also complain of the high costs of pesticides and fertilizers as prices increase steadily and are concentrated in the hands of just a few companies (MAGAP 2016; Nicolalde Herrera 2014). According to the Ministry of Agriculture (2016), fertilizer import quantities have doubled since the early 2000s. Similarly, Naranjo (2017) reports a progressive increase in pesticide imports, especially since 2013 with the implementation of the productive matrix transformation that included agro-fuels. Pesticides come mainly from China and Colombia and are sold by only nine companies that concentrate 65% of the national market of agricultural inputs (Naranjo 2017). One female association member captures the difficulties that small-scale sugarcane producers face in a highly capital- and input-intensive production system by stating, “this land is not for everyone, you can’t work this land if you are poor” (interview with EEE, La Indiana, 1 April 2019). Another concern shared by all association members is the Ministry of Agriculture’s recent decision to reduce the sugarcane price (Ministerial Agreement Nr. 131 of 29 July 2019). This decision responds to the multiple efforts of the sugarcane industry and their distilleries to lower the costs of the raw material, roughly half of which is supplied by independent sugarcane producers. Before this ministerial decision, a minimum price support policy existed for sugarcane and other agricultural products – a policy that was implemented during the previous Correa-headed government of PAIS Alliance. The minimum price aimed to stabilize the sugarcane sector while galvanizing political support from the National Union of Sugarcane Producers of Ecuador. However, this price support hid important weaknesses in the agrarian price structure during Correa’s mandate (interview with Rafael Guerrero, Guayaquil, 21 November 2019). Furthermore, due to an apparent crisis14 in the sugar market, the three sugarcane mills declared insolvency right after the 2018 harvest, and as a consequence, they did not pay their sugarcane providers. This meant that the small- and medium-scale producers could not recover their investments, could not repay their bank loans, and did not have the financial resources to begin the 2019 harvest season. Despite the formal access to land granted to the producer associations, the sugarcane mills ultimately exercise control over land vis-à-vis their influence over the terms of production and control over agro-industrial inputs. The oligopolistic conditions under which this capitalist agro-industrial model operates allow them to determine the input prices, directly influencing the production cost structure of sugarcane providers. Through the farming contracts facilitated by the Agroenergía programme, the sugarcane mill, Coazúcar, can squeeze the four associations into a clearly dependent and very vulnerable position, as the mills continue to exercise political influence in public institutions. All of this is putting at risk the livelihoods of small producers as they struggle to pay their loans and maintain their access to land. However, it is not only the production value that is appropriated by the sugarcane mills. As mentioned earlier, sugarcane mills and their distilleries are the main suppliers in the national ethanol market; thus, they also control the processing through which the raw material realizes added value. As Ángel Ortega, member of the 27 de Octubre Association clearly puts it, to generate income, “they [the
128 Natalia Landívar García sugarcane mills] have the value-added, they have the energy, paper; they don’t only have the sugar” (interview with Ángel Ortega, La Indiana, 21 May 2019). Exacerbating the gendered division of labour and the double burden on women While some association members had years of experience in sugarcane production since they had worked for the San Carlos or Coazúcar sugarcane mills, others gained skills and knowledge as sugarcane producers only with the acquisition of the Hacienda. The work in the sugarcane fields includes maintenance activities that involve soil preparation, removal of the weed called “cauca”, cleaning of water canals, application of fertilizers, irrigation, and fumigation with herbicides. Prior to manual cane cutting and harvesting comes the desoje, whereby sugarcane producers with machete in hand remove the dry leaves to permit the burning of the sugarcane fields. Only a reduced number of association members, mainly those who live in the Hacienda, undertake the desoje maintenance work and are paid by the association to which they belong. For example, in the association 27 de Octubre, only 6 out of 51 members work on the 120 hectares assigned for sugarcane production by this association. The sugarcane mills hire workers outside the association to cut and transport the cane while members of the associations are commonly in charge of supervising this work. While interviewing association members about women’s participation in production, some do recognize that the management of production has a strictly gendered labour division whereby women are assigned only limited participation. For example, in the association 27 de Octubre, where there are only 7 women among 51 members, no women are involved in sugarcane production, except when “mingas15 [are] organized to cut the weed” (interview Luciano Briones, La Indiana, 20 May 2019). Franklin Martínez, a member of the same Association, states his perspective on women’s work on sugarcane fields as follows: “how can you expect a woman to carry an entire bag of fertilizer, I mean, it’s an inherently masculine job” (interview Franklin Martínez, La Indiana, 21 May 2019). FV, an elderly female member of this association, who is one of the few members based in the Hacienda, has often been assigned the task of providing meals to male members in charge of the maintenance work. When she was asked whether she was working in the sugarcane fields, she responded: “I do not work, first because I am old, and second, because I am a woman” (interview with FV, La Indiana, 20 May 2019). Franklin Martínez’s and FV’s accounts uncover the undervaluation of female labour (Deere 1995) whose contributions as cooks and food providers are not recognized as part of the production activities undertaken by associations. There are different perceptions in the other associations about the engagement of women in the hard physical labour demanded by sugarcane production. On the one hand, the participation of women is seen as exceptional. The 2013 harvest is illustrative. At that time, when the Agroenergía programme was already in progress, all members of the associations, including women, were needed to cut and harvest the sugarcane (interview with Georgina Roso, La Indiana, 23 April 2019).
Gender inclusion in sugarcane production 129 This was an unforeseen response to avoid possible cane production losses due to the lack of effective planning by the Ministry of Agriculture’s Agroenergía programme. On the other hand, while referring to women working in the field, others affirm that “women work harder than men” and “are not afraid of the machete” (interview with Ramona Macías, La Indiana, 3 April 2019). Women who were in charge of the maintenance work, in the words of one male member, are “unas muchachonas” (interview witn Kleber Bustamante, La Indiana, 2 April 2019) referring to the outstanding strength of young women who engage in this physically demanding task. However, when examining this in greater detail, it becomes clear that the work to which women are assigned in sugarcane production depends on their age, health, education level, and social status. Indeed, the women who perform hard physical work in sugarcane production are those with more urgent economic needs and fewer labour opportunities elsewhere. Nevertheless, women have taken on more leadership roles in the associations and greater responsibilities in their daily organizational functioning. In this regard, it is worth mentioning the Buen Vivir association because it has integrated women into all five management board positions. Association members agree that the presence of women in these leadership roles has made the administration of funds more transparent and effective. Before assuming the leadership, women were critical of the former association administration that was led only by men, but they were also afraid of being excluded for openly expressing their opinions about the management of funds (interview with Patricia Ronquillo, La Indiana, 24 April 2019). Today, according to Patricia Ronquillo from Association Buen Vivir, some former male leaders of her association still continue to distrust women’s performance in leadership roles. While in other associations the administration has remained in the hands of male members, Buen Vivir also has a female administrator, a position that may be described as executive director whose principal role is to legally represent and run the association. Other spaces of power remain dominated by men in all associations, such as participation in the irrigation and water distribution council (juntas de agua) where women are completely excluded. Whether or not female members perform physical maintenance or administrative work, or are in charge of the provision of food for the workers, they are the most affected by the increased pressures to maintain a livelihood from the production of sugarcane. Indeed, this new agrarian extractive approach hides the double burden that involves the integration of women into assigned gendered spaces of sugarcane production. Following Federici (2018), this means that female members undertake work shifts in the associations where they need to demonstrate their capacities, either in the field or the office, to be as effective as male members while they simultaneously continue to perform multiple unpaid reproductive activities in their households and also in the associations. Productive and reproductive women’s work, according to Deere, “serve to enhance the household’s level of reproduction, mitigating the effects of a low return to family labour implicit in low prices for peasant production or low wages” (Deere 1995, 60). This can also be seen within the associations where unrecognized women’s reproductive work in the cane production process helps to maintain the functioning
130 Natalia Landívar García of the associations. These findings support the contribution by Ojeda (this volume) concerning the disproportionate effects of agrarian extractivism on women, whose work ultimately subsidizes the extractivist model as their time and efforts are appropriated. As Ojeda demonstrates, agrarian extractivism deteriorates social and ecological conditions, thus hindering the possibilities to sustain life. This is a point I will further develop in the last section. Deterioration of environmental quality, and productive and reproductive conditions Association members raised a number of other concerns related to meagre economic benefits due to sugarcane’s low productivity and sugarcane production– related health issues. A former worker at the Valdez sugarcane mill says: “one hectare is currently producing between 40 and 50 tonnes when it should be producing between 80 and 90 tonnes” (interview with JSB, La Indiana, 23 April 2019). By stating “the earth is tired, we do not have good production, and on the top of that, they [the sugarcane mills] do not pay us for the little we produce” (interview with EEE, 1 April 2019), a female association member makes clear the link between decreasing returns and lost fertility and soil erosion due to the intensive use of chemical inputs. Indeed, according to the National Statistic and Census Institute (INEC 2015), 45% of the pesticides most commonly used in sugarcane production can be considered highly toxic.16 Pesticides are manually applied in the sugarcane fields to eliminate plagues and weed, but also to accelerate the maturation of the plants. When referring to the measures taken to avoid harm to people’s health due to the direct application of chemicals, association members mention the use of basic garments such as masks and aprons. However, they recognize that there is a lack of adequate equipment due to the negligence of association administrators, the person responsible for providing health protection. According to a member of Association 27 de Octubre, Luciano Briones, “we do not have any protection . . . when I worked for the sugarcane mill, I received half-litre of milk [an apparently common form of protection],17 but here I do not receive anything” (interview with Luciano Briones, La Indiana, 20 May 2019). In the interviews I conducted, other members even refer to the death of people who labour under hard working conditions and are exposed to chemicals without adequate protection. Wilman Sarango, current president of Association La Indiana, recognizes that they are contaminating the soil due to the vast quantities of agro-chemicals used in sugarcane production. However, he thinks chemicals are important to becoming more efficient and to increasing profits, thus achieving a higher standard of living. By stating that “you need to become accustomed to live in the countryside”, an elderly woman, FV from the Association La Indiana believes that, despite the level of contamination to the rural environment, she is better off in the Hacienda La Indiana than in Milagro, the city where the sugarcane mill Valdez is located and where she used to live with her husband. Her comments clearly signal the environmental effects of the industry as a whole. She remembers the days
Gender inclusion in sugarcane production 131 “when sugarcane fields were burned, ashes would fall in your backyards . . . if your windows were open, you could get them in your dishes” (interview with FV, La Indiana, 20 May 2019). Indeed, a common denominator among populations adjacent to the sugarcane mills has been complaints against the contamination caused by sugarcane production. In particular, the population of Milagro city has been denouncing chronic respiratory illnesses, such as bronchitis, caused by the emission of dust and particles in sugarcane production (Harari et al. 2012). These affirmations reflect the position of some association members who have meekly accepted that integration into sugarcane production means abandoning their right to a healthy environment. However, most members are sceptical, not only about the economic benefits achieved with their inclusion as sugarcane producers – as shown earlier – but on the long-term effects on the soil, water sources and the air,18 and the consequences for people’s health, including cancer and ulcers due to the burning sugarcane fields and the application of pesticides. The indiscriminate use of pesticides and the subsequent water contamination of the estuaries are the main environmental concerns for most of the women I interviewed. According to them, environmental conditions are among the key reasons that prevent members from settling in the Hacienda and why they choose to keep living in nearby localities. For a female member of the La Indiana association, “the water transported by the canals, . . . is not suitable for human consumption because they fumigate, wash the spray pumps, the boots, the containers without any precaution” (interview with Ramona Macías, La Indiana, 3 April 2019). She goes on to say that “women use this contaminated water for bathing, cleaning, to do the laundry, they even drink this water . . . No one has told us that this water is contaminated, that something can happen to us, that we can get ill” (interview with Ramona Macías, La Indiana, 3 April 2019). Another female member wishes that “there were organic products so that not many chemicals would be applied into the soil . . . it is sad to see” (interview with EEE, La Indiana, 1 April 2019). As feminist political ecology scholars state “(w)omen carry a disproportionate share of responsibilities for resource procurement and environmental maintenance . . . and yet they have very limited formal rights (and limited political and economic means) to determine the future of resource availability and environmental quality” (Rocheleau et al. 1996, 13). Gender-differentiated interests and concerns of association members regarding the evident environmental destruction result from the gendered division of labour and the associated social norms within society at large. For female members responsible for the essential daily living conditions of their families and associations, the quality of the environment determines the possibility of social reproduction.19 Nonetheless, for the most part, environmental destruction and resulting health issues are ignored or silenced. They are not discussed within the associations and remain hidden concerns, talked about, if at all, only among individuals and perhaps within families. This resonates with Camacho’s (2017) argument that contamination through the use of toxic chemicals in modern agricultural production becomes “naturalized” in the narratives and practices of those who use them while it remains hidden to public opinion and
132 Natalia Landívar García
Figure 6.2 Women using contaminated water sources to wash clothes in Hacienda La Indiana Note: Women in the photo were not participants in the study.
Gender inclusion in sugarcane production 133 public institutions. For Camacho (2017), environmental contamination becomes yet another form of dispossession of resources needed for production and social reproduction that occurs in a progressive, subtle, and daily manner.
Concluding remarks This chapter has shown that current ethanol production for the Ecuadorian market is characterized by the replication of historic patterns of accumulation and the incorporation of dynamics of economic, social, and environmental extraction that we associate today with agrarian extractivism. The Ecopaís agro-fuel programme has significantly expanded the cultivated area for sugarcane. However, as argued, the mills continue to own most of the land, thus appropriating the surplus value of the sugarcane produced by them and the small- and medium-scale suppliers. The PAIS Alliance governments have actively promoted this extractive project through the ethanol-based agro-fuel Ecopaís using a narrative of energy sovereignty, environmental sustainability, and rural development. Most importantly, I have emphasized that public resources were used in the form of an overpriced ethanol price and public–private partnerships arrangement that expanded the processing capacities of the private distilleries owned by the sugarcane mills. Despite internal disputes between the distilleries and certain state institutions during the Lenin Moreno administration (2017 to the present date), sugarcane mills have succeeded, through strong lobbying and their political connections, in maintaining the support of public institutions as important allies. While political influence of those controlling the sugarcane complex is nothing new in the capitalist agrarian history of Ecuador, the recent changes in ethanol production represent a new characteristic of this extractivist project. Today, the three largest sugarcane mills and their distilleries control the processing facilities in which sugarcane becomes ethanol and realize the added value in the name of energy sovereignty. This chapter also has shown that the new agrarian extractivism in the case of ethanol is characterized by the inclusion of small-scale farmers under the illusion of fostering overall rural development. PAIS Alliance has used this social inclusion discourse to compensate for the weak employment generation in the ethanol processing process. This social component is used to legitimize the appropriation of surplus value and state resources by the sugarcane mills and their distilleries as they consolidate their oligopolistic power through this new extractive dynamic. Moreover, the examination of the Agroenergía programme in Hacienda la Indiana clearly demonstrates that, despite the increased participation of women, there are severe limitations and serious problems in engaging in this agro-extractive model. A principal concern is the meagre economic revenue that members of the association receive due to their increasing dependency on agricultural inputs and the reduction of sugarcane prices that benefit the mills that take advantage of their oligopolistic power and political influence. I have also demonstrated that the risk and fear of losing livelihoods from the production of sugarcane, of paying their debts, and being unable to maintain their access to land is putting more pressure on female members. A pronounced gendered division of labour and social
134 Natalia Landívar García class differences among women who take on the dual burden that implies their integration into sugarcane production, and the destruction of both environmental quality and their productive and reproductive conditions, form part and parcel of this extractive and male-dominated mode of accumulation. Nevertheless, female members have started to question their participation in sugarcane production and are seeking ways of opting out, in search of a sustainable model that will allow them to live on and from the land.
Notes 1 I want to thank the book editors who invited me to participate in this research project, as well as the reviewers for their patience during the drafting of this article. Special thanks to Dr. Diana Ojeda, Dr. Liisa North, and my advisors, Dr. Annette Desmarais and Dr. Iain Davidson-Hunt. I also want to thank the International Development Research Center (IDRC) for generously financing my research project through a Doctoral Research Award in 2018. I also would like to acknowledge the financial support received from the Social Sciences and Humanities Research Council of Canada and the Canada Research Chair Program. 2 Here, I refer to the accumulated effects of cutbacks in social spending in the agrarian sector. 3 Other three mills, IANCEM, Monterrey and Isabel María, do not have a significant sugarcane production (Figueroa De La Vega 2008). 4 Former Vice President (2013–2017) Jorge Glass was one of the foremost proponents of the productive matrix transformation promoted by the PAIS Alliance. In 2017, he was suspended from his official duties and sentenced to six years imprisonment for receiving bribes from the Brazilian company Odebrecht. 5 Argus is an agency that publishes global energy and commodity news and price reporting. It offers an ethanol assessment, which is published daily and is the basis to calculate the ethanol price paid to the distilleries. 6 Before the organizational restructuring of PAIS Alliance government under Lenin Moreno, this ministry was the Coordination Ministry of Production, Employment and Competitiveness. 7 As of 20 February 2020, APALE claims in its website to represent all the actors involved in the agro-fuels sugar commodity chain complex – i.e. sugarcane producers, artisan alcohol producers, sugarcane mills, and sugarcane workers (www.apale.org). 8 In 2019, MIPRO allocated a quota to 14 associations generating direct employment for approximately 2,600 people (Mendieta 2019). 9 Although gasoline prices (Extra and Ecopaís) were adjusted in December 2018, it was in October 2019 when, under the conditionalities set by the International Monetary Fund agreement, the PAIS Alliance government fully eliminated the price subsidies for gasoline, including Ecopaís. Widespread social protest and unrest were determining factors forcing the government to rescind this measure. 10 Members of the associations came from surrounding rural cities (Milagro) and parishes, as well as from distant sites, such as semi-urban villages of Guayaquil and the rice-producing zone of Salitre, where most of them are still living and have other sources of income in and outside agriculture. 11 They held a single farming contract with Coazúcar until December 2019 when contracts were separately assigned to each association. 12 Sugarcane lasts about three to five years before requiring re-seeing. 13 There is another research centre financed by the National Union of Sugarcane Producers of Ecuador, but its seed production capacities are not significant. 14 According to the Global Agricultural Information Network (2019), fiscal policy decisions, the global surplus of sugar, and its resulting falling international price, have
Gender inclusion in sugarcane production 135
15 16 17 18
19
affected the internal sugar market. However, more research is needed to understand the real effects of this crisis on the sugarcane mills given the state protectionist measures that cover their competitive disadvantages in the sugar market, especially in the Andean Community of Nations. A minga is a collective work in which all community members participate. It is a word used in both the highlands and the coast of Ecuador. According to the informants, the herbicides used in the sugarcane production are Ametrina, Terbutrina, Amina, Diuron, Pendimetalinay, and Atrazina 90. For reasons that are not clear, milk is used to as an energy drink and to protect the stomach of sugarcane workers. Some members stressed that chemicals can be transported through the wind, thus, there are effects on health caused not only by the manual spraying of herbicides on the sugarcane field but also from the aerial fumigations of neighbouring banana plantations. For a broader definition of social reproduction, see Giovanna Di Chiro (2008), and also Ojeda (this volume).
References Acosta, A. 2013. Extractivism and neoextractivism: two sides of the same curse. In Miriam Lang and Dunia Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Amsterdam: Transnational Institute, 61–86. Alonso-Fradejas, A. 2018. El Proyecto de Capitalismo Agroextractivo: Una Mirada a La Cuestión Agraria Contemporánea Desde Guatemala. In Blanca Rubio (ed.), América Latina En La Mirada: Las Transformaciones Rurales En La Transición Capitalista. México, DF: Universidad Nacional Autónoma de México. Arsel, M. 2012. Between ‘Marx and markets’? The state, the ‘left turn’ and nature in Ecuador. Tijdschrift Voor Economische En Sociale Geografie, 103(2), 150–63. https://doi. org/10.1111/j.1467-9663.2012.00709.x. Arsel, M., B. Hogenboom, and L. Pellegrini. 2016. The extractive imperative in Latin America. Extractive Industries and Society, 3(4), 880–7. https://doi.org/10.1016/j. exis.2016.10.014. Artacker, T. 2018. El Aparato Desarrollo Políticas Agrarias Progresistas. Una Crítica Desde El Postdesarrollo a Las Políticas Públicas de Los Gobiernos de Correa En Ecuador y Mujica En Uruguay. Ecuador Debate, 105, 69–90. Banco Central del Ecuador. 2020. Información Estadística Mensual No. 2022, Agosto. Quito. Available at: https://contenido.bce.fin.ec/home1/estadisticas/bolmensual/ IEMensual.jsp. Borras Jr., S.M., C. Kay, S. Gómez, and J. Wilkinson. 2012. Land grabbing and global capitalist accumulation: key features in Latin America. Canadian Journal of Development Studies/Revue Canadienne d’études Du Développement, 33(4), 402–16. https://doi.org/ 10.1080/02255189.2012.745394. Borras Jr., S.M., J.C. Franco, S.R. Isakson, L. Levidow, and P. Vervest. 2016. The rise of flex crops and commodities: implications for research. The Journal of Peasant Studies, 43(1), 93–115. https://doi.org/10.1080/03066150.2015.1036417. Bravo, E. 2020. Cuando El Mar Entra a La Tierra. Producción Camaronera En La Cuenca Baja Del Río Guayas, Tierras Campesinas y Comunas Ancestrales. Quito: Acción Ecológica. Camacho, J. 2017. Acumulación Tóxica y Despojo Agroalimentario En La Mojana, Caribe Colombiano. Revista Colombiana de Agropología, 53(1), 123–50.
136 Natalia Landívar García Carrasco, C.M. 2014. La Acumulación Originaria Del Socialismo Del Siglo XXI (Parte II). El Telégrafo, 17 February. Available at: www.eltelegrafo.com.ec/noticias/carlos-marx/1/ la-acumulacion-originaria-del-socialismo-del-siglo-xxi-parte-ii. CINCAE (Centro de investigación de la caña de azúcar del Ecuador). 2019. Informe Anual 2018. El Triunfo: CINCAE. Dávalos, P., and V. Albuja. 2014. Ecuador: extractivist dynamics, politics and discourse. In Henry Veltmeyer and James Petras (eds.), The New Extractivism: A Post-Neoliberal Development Model or Imperialism of the Twenty-First Century? London: Zed Books, 144–71. Davidov, V. 2013. Mining versus oil extraction: divergent and differentiated environmental subjectivities in ‘post-neoliberal’ Ecuador. Journal of Latin American and Caribbean Anthropology, 18(3), 485–504. https://doi.org/10.1111/jlca.12043. Deere, C.D. 1995. What difference does gender make? Rethinking peasant studies. Feminist Economics, 1(1), 53–72. https://doi.org/10.1080/714042214. Deere, C.D. 2017. Women’s land rights, rural social movements, and the state in the 21stcentury Latin American Agrarian reforms. Journal of Agrarian Change, 17, 258–78. https://doi.org/10.1111/joac.12208. Di Chiro, G. 2008. Living environmentalisms: coalition politics, social reproduction, and environmental justice. Environmental Politics, 17(2), 276–98. https://doi. org/10.1080/09644010801936230. Diario El Telégrafo. 2011. Nuevas Semillas Caña Azucar Mejoran Produccion, 7 December. Available at: www.eltelegrafo.com.ec/noticias/economia/4/nuevas-semillas-de-canade-azucar-mejoran-produccion. Diario El Telégrafo. 2013. Caña de Azúcar Propicia Cambio de Matriz Energética Del País, 6 October. Available at: www.eltelegrafo.com.ec/noticias/economia/4/ cana-de-azucar-propicia-cambio-de-matriz-energetica-del-pais. Ericsson, K. 2013. Tierra+Campesino=¿Soberanía Alimentaria? El Uso de La Tierra de Hacienda La Indiana Después de La Transferencia de Tierras a Asociaciones Campesinas. Quito: Facultad Latinoamericana de Ciencias Sociales (FLACSO) Ecuador. España, S. 2018. La Ecopaís Tiene Menos Biocombustible Del Anunciado. Expreso, 14 November. Ezquerro-Cañete, A. 2016. Poisoned, dispossessed and excluded: a critique of the neoliberal soy regime in Paraguay. Journal of Agrarian Change, 16(4), 702–10. https://doi. org/10.1111/joac.12164. Federici, S. 2018. Marx and feminism. TripleC, 16(2), 468–75. Figueroa De La Vega, F. A. 2008. “Tablero de Comando” Para La Promoción de Los Biocombustibles En Ecuador. Santiago de Chile: United Nations. Fisher, S. 1983. Estado, Clases e Industria. La Emergencia Del Capitalismo Ecuatoriano y Los Interéses Azucareros. Quito: Editorial El Conejo, Facultad Latinoamericana de Ciencias Sociales (FLACSO) Ecuador and Instituto Latinoamericano de Investigaciones Sociales (ILDIS). Holguín, F., and G. Lorena. 2011. Los Trabajadores de La Zafra: Identidad Obrera En La Industria Azucarera Ecuatoriana. El Caso de AZTRA (1964–1977). Quito: Facultad Latinoamericana de Ciencias Sociales (FLACSO) Ecuador. GAIN (Global Agricultural Information Network). 2018. Ethanol Industry Emerging in Ecuador. Quito: USDA Foreign Agricultural Service. Available at: https://gain.fas. usda.gov/RecentGAINPublications/EthanolIndustryEmerginginEcuador_Quito_Ecua dor_1-29-2018.pdf.
Gender inclusion in sugarcane production 137 GAIN (Global Agricultural Information Network). 2019. Ecuadorian Sugar Consumption Continues to Decline. Quito. Available at: https://apps.fas.usda.gov/newgain api/api/report/downloadreportbyfilename?filename=Sugar Annual_Quito_Ecuador_ 3-19-2018.pdf. Gillon, S. 2016. Flexible for whom? Flex crops, crises, fixes and the politics of exchanging use values in US corn production. The Journal of Peasant Studies, 43(1), 117–39. https://doi.org/10.1080/03066150.2014.996555. Gudynas, E. 2009. Diez Tesis Urgentes Sobre El Nuevo Extractivismo. Contextos y Demandas Bajo El Progresismo Sudamericano Actual. In Extractivismo, Política y Sociedad. Quito: Centro Andino de Acción Popular (CAAP) and Centro Latinoamericano de Ecología Social (CLAES), 187–225. Gudynas, E. 2010. Agropecuaria y Nuevo Extractivismo Bajo Los Gobiernos Progresistas de América Del Sur. Territorios, 5, 37–54. Guerrero, R. 1978. Los Ingenios En El Desarrollo Del Capitalismo En El Ecuador 1900– 1954. In II Encuentro de Historia y Realidad Económica y Social Del Ecuador. Cuenca: IDIS and Universidad de Cuenca, 527–94. Guerrero, R. 2017. Del Banco Comercial y Agrícola Al Ingenio San Carlos (1925–1950). Ecuador Debate, 102, 123–36. Harari, R., H. Harari, N. Harari, F. Harari, and R. Freire. 2012. Machete, Sudor y Enfermedad: Condiciones Laborales de Los Trabajadores y Trabajadoras de La Zafra En Ecuador. Milagro: Federación de Trabajadores Agroindustriales, Campesinos e Indígenas Libres del Ecuador (FENACLE), Fondo de Cooperación al Desarrollo (FOS) and Corporación para el Desarrollo de la Producción y el Medio Ambiente Laboral (IFA). INEC (Instituto Nacional de Estadísticas y Censos). 2000. Censo Nacional Agropecuario 2000. Quito. Available at: www.ecuadorencifras.gob.ec/censo-nacional-agropecuario/. INEC (Instituto Nacional de Estadísticas y Censos). 2015. Uso y Manejo de Agroquímicos En La Agricultura 2014. Quito: INEC. INEC (Instituto Nacional de Estadísticas y Censos). 2020. Encuesta de Superficie y Producción Agropecuaria Continua (ESPAC) 2019. Quito. Available at: www.ecuadorenci fras.gob.ec/estadisticas-agropecuarias-2/. Ingenio S.C. 2015. Visita Del Vicepresidente de La República. Available at: www.sancarlos.com/category/noticias. Larrea, C., and G. Natalia. 2018. Concentration of assets and poverty reduction in postneoliberal Ecuador. In Liisa L. North and Timothy D. Clark (eds.), Dominant Elites in Latin America: From Neo-Liberalism to the “Pink Tide”. Cham: Palgrave Macmillan, 93–118. Larrea, C., and A.I. Larrea. 2017. Hemos Sembrado El Petróleo En El Ecuador? In Carlos Larrea (ed.), Está Agotado El Periodo Petrolero En Ecuador? Alternativas Hacia Una Sociedad Más Sustentable y Equitativa: Un Estudio Multicrítico. Quito: Ediciones La Tierra and Universidad Andina Simón Bolívar, 33–56. MAG (Ministerio de Agricultura y Ganadería). 2017. Boletín Situacional Caña de Azúcar. Quito: Sistema de Información Pública Agropecuaria – SIPA. MAGAP (Ministerio de Agricultura Ganadería Acuacultura y Pesca). 2016. La Política Agropecuaria Ecuatoriana: Hacia El Desarrollo Territorial Rural Sostenible 2015– 2025, I Parte. Quito: MAGAP. Maiguashca, J. 2012. La Incorporación Del Cacao Ecuatoriano Al Mercado Mundial Entre 1840 y 1925, Según Los Informes Consultares. Procesos, 35, 67–97.
138 Natalia Landívar García McKay, B.M. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211. https://doi.org/10.1016/j.worlddev.2017.04.007. Mendieta, F. 2019. Informe Sector Alcoholero (ECOPAIS). Quito: Ministerio de Producción, Comercio Exterior, Inversión y Pesca. Naranjo, A. 2017. La Otra Guerra. Situación de Los Plaguicidad En Ecuadsor. Quito: Acción Ecológica. Nicolalde, H., and L. David. 2014. Análisis Económico de La Cadena Productiva de La Caña de Azúcar Bajo Un Enfoque Estructuralista y Matriz de Análisis de Política, Período 2006–2012. Quito: Pontificia Universidad Católica del Ecuador. North, L.L. 1985. Implementación de La Política Económica y La Estructura Del Poder Político En El Ecuador. In Louis Lefeber (eds.), La Economía Política Del Ecuador. Campo, Región, Nación. Quito: Corporación Editora Nacional, 425–57. North, L.L., and R. Grinspun. 2016. Neo-extractivism and the new Latin American developmentalism: the missing piece of rural transformation. Third World Quarterly, 37(8), 1483–504. https://doi.org/10.1080/01436597.2016.1159508. Ordóñez Saénz, J. 2016. Proyecto Ecopaís. Quito: Ministerio Coordinador de Producción, Empleo y Competitividad. Rocheleau, D., B. Thomas-Slayter, and E. Wangari. 1996. Gender and environment. a feminist political ecology perspective. In Feminist Political Ecology: Global Issues and Local Experiences. London: Routledge, 3–23. SENPLADES. 2012. Transformación de La Matriz Productiva. Revolución Productiva a Través Del Conocimiento y El Talento Humano. Quito: Ediecuatorial. Veltmeyer, H. 2013. The political economy of natural resource extraction: a new model or extractive imperialism? Canadian Journal of Development Studies, 34(1), 79–95. https://doi.org/10.1080/02255189.2013.764850. Veltmeyer, H., and J. Petras. 2014. The New Extractivism: A Post-Neoliberal Development Model or Imperialism of the Twenty-First Century? London: Zed Books. White, B., and A. Dasgupta. 2010. Agrofuels capitalism: a view from political economy. The Journal of Peasant Studies, 37(4), 593–607. https://doi.org/10.1080/03066150.201 0.512449. Wilson, J., and M. Bayón. 2017. La Selva de Los Elefantes Blancos: Megraproyectos y Extractivismos En La Amazonía Ecuatoriana. Quito: Ediciones Abya Yala e Instituto de Estudios Ecologistas del Tercer Mundo.
7
Life purging agrarian extractivism in Guatemala Towards a renewable but unlivable future? Alberto Alonso-Fradejas
Introduction Climate, energy, environmental, food, and financial/economic crises took centre stage in 2008 and thrived for years to follow.1 Furthermore, the crises sparked a renewed global interest in natural resources for business, poverty reduction, and climate change adaptation and mitigation purposes (Borras et al. 2018). This is especially the case with the core role the resource-intensive bio, blue, and green economies are called to play within the emerging “4th Industrial Revolution” “characterized by a fusion of technologies that is blurring the lines between the physical, digital, and biological spheres [at an unprecedented] velocity, scope, and systems impact” (Schwab 2016). In this conjuncture, global demand for renewable energy and materials soars, and natural resource extractivism is framed as a vehicle of transition to socially and environmentally sound futures, championing the needs of humanity and Planet Earth. Crops and trees quickly become one such transformational vehicle – particularly their newer, ever-growing, and flexibly interchangeable uses as carbon sinks and sources of bioenergy and biomaterials complementing their traditional uses as food, feed, fibre, fuel, and timber. As a result, corporate “flex crops and commodities complexes”2 consolidate and upgrade within former strongholds and venture into new territories to take down the convergent social and ecological crises. But in so doing, they contribute to a large extent to the latest global resource rush. The “land grabbing” and “new extractivism” research agendas in response to the global resource rush have brought issues of contemporary agrarian, environmental, and climate change into the spotlight. However, both streams of literature have run parallel to one another. While they have offered important insights, their findings have often been disconnected and, therefore, partial in addressing a common problem. Bridging critical and intersectional political economy, ecology, and sociology perspectives I explore here the implications of the rise of the flex crops and commodities complexes in sustainability transitions for jobs, labour regimes, and socio-ecological reproduction more broadly. This is particularly relevant considering the centrality of renewables in the 4th Industrial Revolution’s bio, blue, and green economies, the 2015 Paris Climate Agreement, and especially the 2030 Agenda for Sustainable Development’s pledge to “leave no one behind”.
140 Alberto Alonso-Fradejas Grounded in a comparative and longitudinal research in Guatemala during 2005–2014, I suggest that the rise of the flex crops and commodities complexes fuels a distinct form of predatory resource extractivism with far-reaching implications for human and non-human nature and the climate. In brief, flex (sugar)cane and (oil) palm complexes thrived from the mid-2000s onwards under a favourable global conjuncture and the auspices of a mighty Guatemalan oligarchy with thick ties to transnational financiers. Paradoxically framed as a silver bullet for both sustainable development and climate change mitigation, the ways in which labour, land, financial capital, knowledge, and non-human nature are mobilized into flex cane and palm commodity production result in a predatory, life purging model of “agrarian extractivism”.3 This is because the burgeoning flex cane and palm complexes are driving a process of “impairing destruction” by means of a job-poor, culturally insensitive, toilsome, and unpaid labour-based “productive” model, and the manufacturing of environmentally and socially toxic landscapes. Impairing destruction unleashes a social and ecological purge of the countryside which adversely affects friends and foes alike and regardless of species, social class, gender, ethnicity, or livelihood. However, in the densely populated, structurally unequal, and largely job-scarce Guatemalan context of the early 21st century, which might resonate elsewhere, the purge strikes especially hard the thousands of working families (and particularly women) who are deemed redundant for the new renewables-led corporate socio-ecological order. In what follows, I discuss these claims. First, I briefly review the contributions from the “land grabbing” and “new extractivism” research agendas. Then, I describe my analytical approach, methodology, and methods. After this, I trace core aspects of contemporary agro-extractivism in the Guatemalan post-colonial history. To this follows the analysis of how the predatory agro-extractivism of cane and palm companies shapes jobs, labour regimes, and socio-ecological reproduction. Then I discuss why, how, and to what extent the flex cane and palm complexes are driving a process of impairing destruction which results in a socioecological purge that, while it does not go unchecked, leaves nothing and no one unscathed. I conclude with a reflection on the broader implications of the new renewables-led corporate order for socio-ecological reproduction, climate stewardship, and sustainable development.
Investigating the latest global resource rush The current resource rush under convergent global crises has spurred a good deal of political and scholarly debate. To this end, two research agendas have gained relevance from the mid-2000s onwards. One concerns the investigation of a new global land rush. The other one involves analysing the massive wave of resource extractivism that is spilling over the world. On one hand, there is the “research rush on the global land rush” (Edelman et al. 2013, 1528). Following GRAIN’s (2008) report alerting the public to the “global land-grab for food and financial security”, the phenomenon was initially picked up by actors who have an agenda to frame and influence land deals as
Life purging agrarian extractivism 141 potential drivers of development and growth.4 Epistemologically, these assessments relied on neoclassical and/or neo-institutional economics perspectives, which assume that outcomes of land deals depend on the state of governance and market competition in the places where they unfold. Hence, land deals were ontologically divided into lawful, transparent, and desirable “large-scale land acquisitions”, and regretful “land grabs” that do not abide by the rule of law and market fair play (von Braun 2008). In addition to these mainstream assessments, there were also those undertaken from critical perspectives. But during what Edelman et al. (2013, 1520) call “the making sense period” between 2007 and 2012, much critical and mainstream research on the global land rush suffered from a series of analytical flaws. These include the need to situate inquiries within longer term historical trajectories of agrarian, environmental, and climate change; to broaden our understanding of land grabs beyond land transactions for food and fuel, and beyond “dispossession” and “land”; and to pay more and better attention to research methodologies and the quality of available data (Edelman et al. 2013; Franco et al. 2015; Zoomers and Otsuki 2017; Brent et al. 2018). On the other hand, the notion of “extractivism” regains momentum in scholarly and political debates following the commodity boom at the turn of the 21st century. Of particular relevance in this regard are the contributions of the Latin American post-development school. One of its key authors, Eduardo Gudynas, defines conventional extractivism as “activities which remove great quantities of natural resources that are not then processed (or are done so in a limited fashion) and that leave a country as exports” (Gudynas 2010, 1). This is a definition endorsed by Alberto Acosta (Acosta 2013, 63) and Maristella Svampa – also key post-development authors – although the latter qualifies Gudyna’s definition by arguing that extractivism involves “the expansion of frontiers to territories formerly considered ‘unproductive’” (Svampa 2013, 118). These leading intellectuals argue that a “new extractivism” rose in the early 21st century in Latin American countries under left-leaning “pink tide” governments.5 The new extractivism follows its conventional peer’s “style of development based on the appropriation of Nature [only that] the state plays a more active role, and gives extractivism a greater legitimacy because it redistributes some of the surplus to the population” (Gudynas 2010, 1). For Svampa, differences between conventional and new extractivisms also occur along the fault lines of the structural shift “from the Washington Consensus with its focus on finance to the Commodities Consensus based on the large-scale export of primary products” (2013, 118. C.f. McKay et al. 2017). What they do all agree on, however, is that the new extractivism is not only about mineral and hydrocarbon extraction but also involves resource extractivism in agriculture, fishing, and forestry. Contributions by the post-development school authors and others are rich and manifold. But whereas the bulk of attention is devoted to the analysis of the new extractivism’s performance as a political project (Arsel et al. 2016), the ecological terms of exchange among countries (Rice 2007), and the social metabolism of extraction (Martínez-Alier et al. 2010), the social relations of production underpinning contemporary resource extractivism have received lesser and/or narrower
142 Alberto Alonso-Fradejas attention. Often, materialist analyses of the new extractivism have focused on structural drivers (Acosta 2013; Svampa 2013), and on the economic terms of exchange between countries (Veltmeyer and Petras 2014). Thus, labour, land, financial, and knowledge relations are generally unexplored or underexplored in the literature on new extractivism. This makes it easy to confuse means with ends and take extractivism for granted since colonial plunder rather than as a historically and geographically situated phenomenon. Despite the repeated calls by key authors against the absolutization and dehistorization of extractivism as an analytical category (Gudynas 2013; Veltmeyer and Petras 2014), much of the literature on new extractivism conceals more than it reveals about the various trajectories, geographical unevenness, and ecologically and socially differentiated outcomes of resource extractivism today.
Methodological notes on the labour and socio-ecological reproduction questions in resource extractivism My general analytical approach blends critical and intersectional political economy, ecology, and sociology perspectives to better fit the increasingly diverse, uneven, interconnected, and fluid socio-ecological formations of today’s world. From this approach, I aim to bridge the varying topics and temporal units of analysis particular to the literature on land grabbing and the new extractivism, as well as agrarian, environmental, development, and sustainability studies more generally. As anticipated, my focus is on the labour and socio-ecological reproduction implications of sustainability transitions. This emphasis stems from my broader aim to examine whether and how three major types of relationships behind the political economy, ecology, and sociology of sustainable development interplay in the current social and ecological crises to shape human and non-human agency and well-being, and vice versa. These various types of relationships have been framed as three “general contradictions” under different forms of capitalism over time and across geographies. One refers to the contradiction between capital and wage labour regarding the appropriation of surplus value in commodity production (Marx 1887). The second is between capital and Nature (O’Connor 1988).6 And the third contradiction is between capital and social reproduction (Fraser 2016).7 To be clear, the history of natural resource extraction for social reproduction purposes is as old as humanity. Different modes and forms of commodity production rely on natural resource extraction to a greater or lesser extent. The question that remains is how to qualify and account for the intensity of resource “extractiveness” of a mode or form of production, which ranges from indispensable to predatory extractivism (Gudynas 2013). For Gudynas, there are three necessary and simultaneous conditions for predatory resource extractivism, including “large scale of extraction, limited processing, and export destination” (2013, 5). To me, these may or may not be all part of the equation. What really matters is that these or any other relevant conditions do not take place in a “social and ecological
Life purging agrarian extractivism 143 vacuum”. Rather, they drive and express concrete social relations in a specific socio-ecological formation, within a particular time span, and in the context of a broader world-historical conjuncture.8 Thus, I use three key sets of criteria to qualify the character and assess the intensity of the “extractiveness” of a mode/form of commodity production through labour and socio-ecological reproductive lenses. One set includes criteria related to whether, how, and the extent to which external nature is exhausted, or the breadth of the “metabolic rift” (Marx 1894, 567) between resource extractivism and non-human nature. Analytically, my focus is on the ecological conditions for the reproduction and well-being of human and non-human life, including the natural conditions of commodity production. Hence, I am concerned with the examination of the “social metabolism” of resource extraction, meaning “the manner in which human societies organize their growing exchanges of energy and materials with the environment” (Martínez-Alier et al. 2010, 1). The sociometabolic perspective includes labour dynamics. But considering the centrality of the labour and social-reproductive questions in sustainable development, and for explanatory purposes, I analyse labour issues separately. Thus, another set focuses on jobs, “labour regimes”,9 and “social-reproductive regimes”,10 and includes three specific criteria. The first relates to the implications of resource extractivism for employment numbers. The interest here is on jobs in both resource extractivism and the broader economy. The second criterion investigates whose labour is mobilized and organized in which ways, including wages and labour conditions, in the actual extractivist process. The third criterion concerns whether, how, and the extent to which resource extractivism relies on the appropriation of neither hired nor paid affective and/or material labour, including but not limited to family labour. My third set of extractiveness criteria delves into the question of control over flows of capital in commodity production. It involves looking at resource extractivism through the lenses of Marx’s theory of value and surplus value. Hence, this set includes criteria regarding the range, ways, and extent to which different commodity value portions – other than those amounting to wages and the constant capital replacement fund – as well as state revenues, are appropriated by non-direct producers. For the case at hand, these are criteria for the analysis of whether, how, and to what extent land’s ground rent, financial interest, royalties from intellectual property rights, payments for environmental services and state subsidies are crafted, extracted, and ultimately appropriated by the owners of cane and palm companies. At the same time, the previous three sets of extractiveness criteria shape and are also strongly shaped by culture, place, and the ideological distribution outcomes of the clash between competing worldviews, knowledge, and “ways of knowing” (Martínez-Torres and Rosset 2014) which coalesce into hierarchies of well-being perspectives in specific places that are relatively stable over time. Thus, following Hart’s notion of “critical ethnographies of globalization”,11 my analysis is geographically and historically grounded in Guatemala from 2005 onwards. There are two reasons for this.
144 Alberto Alonso-Fradejas First, because it makes sense to analyse the sustainable development effects of renewables in Guatemala during the current world-historic conjuncture. Together with heightened mineral and hydrocarbon extraction, a renewables pandemic washes over Guatemala since 2005. It is especially from this year onwards that (sugar)cane and (oil) palm plantations and processing plants spread like wildfire, and the small Central American country becomes a leading world producer and exporter of multiple cane and palm commodities (Hurtado Paz y Paz 2008; Alonso-Fradejas et al. 2008; Alonso-Fradejas 2013; Mingorría et al. 2014). In this context, agro-ecological, social, and policy structures are reconfigured to meet the needs of what is broadly categorized as a “new economic model”. This is led by a select group of Guatemalan white oligarchic families with thick ties to foreign capital, in a country which is overwhelmingly populated by Indigenous and mestizo (i.e. “ladino”) peoples. And yet, the more fundamental dynamics at work behind the “new economic model” from 2005 onwards remained largely unexplained beyond these immediate factors. The second reason to carry out this research in Guatemala is because of its feasibility. I did comparative field research in the country for a decade, including before and during the convergent global crises from 2008 onwards. Hence, this manuscript relies on empirical material gathered in Guatemala and abroad, but especially on that collected through mixed methods research involving around 700 research subjects in the Guatemalan northern lowlands region (see map in Figure 7.1). My research methods included: (i) 111 individual and 49 group semi-structured interviews; (ii) participant observation and observed participation in a multitude of everyday and exceptional events; (iii) two waves of gender-differentiated household panel survey in 2010 and 2014 (n = 586 × 2); (iv) geographic information system analysis of land-use changes; (v) secondary source analysis; (vi) soil analysis; (vii) water analysis; and (vii) two participatory documentary films. As Figure 7.1 depicts, the northern lowlands include almost half of the Guatemala’s territory and they are mainly populated by the Maya-Q’eqchi’ peoples. The 1.6 million hectare Mayan Biosphere Reserve in the northern lowlands marks the agrarian frontier for thousands of (Q’eqchi’) shifting cultivators, a few owners of traditional haciendas and ranches, and a fewer number of (sugar)cane and (oil) palm companies from 2005 onwards. My methodological strategy is two-pronged. First, I trace core aspects of today’s agro-extractivism through Guatemala’s post-colonial history, beginning with the Liberal Revolution of 1871 and the ensuing dominance of the coffee economy. Then, I carry out an interactive analysis of productive relations in the flex cane and palm complexes in two steps. First, I examine labour, land, financial, knowledge, and ecological relations of production, distribution, property, and reproduction. For instance, I investigate “land” relations of production, distribution, property, and reproduction, as well as labour, land, financial, knowledge, and ecological “distribution” relations. Table 7.1 lists the 20 different productive relations under scrutiny.
Life purging agrarian extractivism 145
Figure 7.1 Research sub-regions, departments, and municipalities in Guatemala Source: Elaboration by Margot Stoete, Utrecht University
Second, I follow the rationale depicted in Figure 7.2 to examine the interplay among all of the previous productive relations in the flex cane and palm complexes and thereby their combined effects on jobs, labour, and socio-ecological reproduction. Thus, I understand productive relations in a broad fashion which involves the ways in which (i) productive forces are organized in the transformation of nature into commodities; (ii) different portions of the commodity value are distributed; (iii) ownership over productive forces and entitlements to commodity value portions are politically sanctioned; and (iv) the commodity value portions appropriated by different actors are used – after taxes and subsidies – for conspicuous consumption, life reproduction, and/or accumulation purposes. In other words, I approach productive relations as socio-ecological relations of commodity production (narrowly defined), distribution, property, and reproduction.12
146 Alberto Alonso-Fradejas Table 7.1 Diverse productive relations around multiple forces of production Multiple (→) and diverse (↓) productive relations
Labour (W)
Land (L)
Financial (F)
Knowledge (K)
Ecological (E)
Production (p) Distribution (D) Property (P) Reproduction (R)
W-p W-D W-P W-R
L-p L-D L-P L-R
F-p F-D F-P F-R
K-p K-D K-P K-R
E-p E-D E-P E-R
Source: Author elaboration
Figure 7.2 Interactive analysis of diverse productive relations around multiple forces of production in resource extractivism Source: Author elaboration
A brief historization of the labour and socioecological reproduction questions in the Guatemalan flex cane and palm complexes I am concerned here with an analysis of change and continuity in the productive relations behind contemporary agro-extractivism. To this end, I look back into two relevant periods in Guatemalan history. The first ranges from the liberal revolution of 1871–1943, before the triumph of the 1944 social-democratic revolution. In 1871, the liberals seized power and held it firmly for more than 70 years. Making a key state-building principle of the Lockean maxim of individual private property as a natural and absolute right, this period was marked by the
Life purging agrarian extractivism 147 massive enclosure of Indigenous people’s communal land. This process of primitive accumulation was largely driven by the labour demands of the burgeoning coffee economy. Coffee production for export thrived in a world-historic context in which British hegemony was contested by the post-Civil War United States and a unified Germany undergoing a rapid industrialization process. Capitalist relations of production generalized in banana plantations owned by the US United Fruit Company from 1906, and in coffee production from the 1930s onwards. But coffee haciendas relied on forced labour drafted by the state all throughout the 1871–1943 period (McCreery 1994). Resource extractivism in 1944–1954 revolved around the social-democratic project of “capitalism from below” that took place in Guatemala. In 1954, the US Central Intelligence Agency (CIA) orchestrated a coup, ousting the democratically elected government of Árbenz and rolling back the progressive reforms of the “Revolutionary Spring” decade. To this followed the crudest period of the Cold War in Guatemala until new democratic elections were held in 1985, epitomized in the internal armed conflict that broke out in 1962 between the military state and various Marxist guerrillas which joined forces under the umbrella of the Guatemalan National Revolutionary Unity (URNG) from 1982 onwards. The active involvement of the state with working people which for opposite reasons characterized the Revolutionary Spring and the Cold War in Guatemala finds little resonance today. Therefore, the second period of relevance for my analysis is the initial phase of neoliberal globalization. Arguably, this ranges in Guatemala from the beginning of the structural adjustment of the economy and polity in 1986, to the approval of the free-trade agreement with the United States (DR-CAFTA) in 2005. In 1985, a transition to civilian rule took place with a constitutional assembly and the celebration of general elections. The winning Christian-Democratic party set into action the liberalization, deregulation, and privatization reforms of the Washington Consensus policy package. Between 1987 and 1996, the government and the URNG engaged in peace negotiations. The peace and a return of liberal democracy allowed for the neoliberal globalization project to root itself in Guatemala. Among other things, this meant that the dumping of cheap grain imports by the United States flipped a switch for accumulation in the agri-food system, turning its emphasis from an export-oriented to an import-oriented business model. A series of powerful agro-industries, especially eggs/poultry and pork producers, emerged and/or consolidated in Guatemala following the massive inflows of cheap corn, rice, and wheat arriving from the United States (Alonso-Fradejas and Gauster 2006). The neoliberal restructuring of the countryside affected everyone, including traditional landlords and commercial farmers, and unleashed the “impairing destruction” process behind the first neoliberal purge of the Guatemalan countryside. But as usual, working people suffered the most. A rural exodus occurred as a result, in which hundreds of thousands of farmers outcompeted by cheap food imports and jobless farmworkers fled to Guatemala City and the United States.
148 Alberto Alonso-Fradejas Continuity Current productive relations in the flex cane and palm complexes resemble those from 1871 to 1943 in five ways. First, cane and palm companies continue to rely on debt-peonage to some extent. Additionally, most plantation workers remain temporarily employed and must farm to subsidize their wage income. Second, landed property remains a means of wealth and prestige for the national oligarchy, especially for the elders who continue to treat it as their natural and absolute right (Alonso-Fradejas 2021). Third, foreign financiers also play a key role in the provision of funds to cane and palm companies amid a limited and more costly domestic credit supply. Fourth, as with coffee, cane and palm commodity production relies on specialized knowledge and technologies protected through intellectual property rights. And fifth, cane and palm companies use environmental resources and services, and dispose of waste and contaminants at zero cost. Productive relations during 1986–2005 also inform today’s in six ways. First is the generalization of capitalist labour relations, and specifically of a flexible and piecemeal labour regime inaugurated by sugar cane companies in the 1990s. Second, land relations continue to hinge on the formalization of individual land property rights and a reliance on the market as the prime land access and redistribution mechanism. Third, as part of large oligarchic family business groups that also include banks, most cane and palm companies enjoy preferential access to funding from domestic financiers as well, just as the poultry and pork agro-industries did. Fourth, access to knowledge and technologies enclosed through intellectual property rights keeps offering a competitive edge and acting as entry barriers into the flex cane and palm complexes. Fifth, as it was the case with industrial chicken and pig feedlots and earlier sugar cane companies, pollutants and waste from cane and palm plantations and processing plants are dumped or released into the ecosystem for free and with impunity. Sixth, cane and palm companies expand the transnationalization trend that poultry and pork companies, together with a few sugar cane mills, inaugurated in the 1990s. Change The particularity of the current productive relations in the Guatemalan flex crop complexes mirrors a broader context in which two extra-economic factors stand out. On one side, the main coping strategy of the masses of purged cultivators and farmworkers – namely fleeing – is increasingly constrained. The agrarian frontier has been legally closed as of 1990, and trespassers into the Mayan Biosphere Reserve and other nature conservation enclosures are criminalized as “ecoterrorists” or narco-collaborators. Guatemala City is increasingly at pains to host the masses of desperate newcomers in its sprouting network of violence-ridden and job-poor slums. And the national security imperative following the attacks of 11 September 2001 in the United States, together with the ensuing economic
Life purging agrarian extractivism 149 recession that began in 2008 there are all expressed through the tightening of border and immigration policing. This results in the yearly expulsion of thousands of unauthorized Guatemalan migrants living in the United States (Ratha et al. 2016). On the other side, the gradual rise within the national oligarchy of a younger intellectual vanguard restructures the ways in which cane and palm companies do business.13 Broadly speaking, this new generation of oligarchs is responsible for the upgrade of their families’ domestic agro-industries into transnational agribusinesses. Unlike older cane and palm tycoons forged at the heat of Cold War’s military dictatorships, this younger generation comes of age during Guatemala’s democratic transition under neoliberal globalization. As a result, they tend to follow a political rationale that prioritizes concessions to working people in the short term as the path towards enhanced business resilience and profitability and class hegemony in the long run (Alonso-Fradejas 2021). Thus, the distinctiveness of the current productive relations in the flex crop complexes is rooted in four dynamics. First, wage labour in cane and palm plantations is mainly organized through capitalist relations, its unfree labour remnants apart. As it was the case with the 19th century coffee economy, cane and palm plantation workers are usually recruited via labour contractors. In some cases, although not so often as in the past, contractors advance cash payments to be later deducted from the worker’s wage and the companies keep the workers’ identity cards until they consider the debt is settled. This means that debtpeonage persists today. But following labour mobilization at the grassroots and pressure from international watchdogs, cane and palm companies upgraded their salaries and working conditions from 2012 onwards to meet the minimum Guatemalan legal standards. As a result of this labour regime fix, value in today’s flex cane and palm complexes is “officially” generated by mostly free, wage labour. Second, whereas the 19th century coffee planters lacked more labour than they did land (Palma et al. 2004), today’s cane and palm companies enjoy a superabundance of cheap labour and suffer from a limited and contentious supply of land. Third, the flex cane and palm complexes implement a knowledge fix that enhances plantations’ yields while simultaneously increasing their resilience to climate and environmental disruptions. Conversely, Maya-Q’eqchi’ shifting cultivators are trapped in a knowledge rift between traditional and familiar extensive farming practices, and more recent, more costly, and stranger intensive farming methods and technologies. Fourth, most of the value generated in cane and palm commodity production is now controlled by cane and palm companies, and increasingly securitized to be used as a funding tool hedged against inflation and price variability. For instance, the current transnationalization strategy of the Guatemalan flex cane and palm complexes includes the offshoring of the special purpose vehicle firms used in land securitization operations in tax havens like Panama.
150 Alberto Alonso-Fradejas
The predatory agro-extractivism of the flex cane and palm complexes I discuss here the main outcomes of my interactive analysis of productive relations in the flex cane and palm complexes. To this end, I follow my three sets of criteria to qualify and account for the intensity of extractiveness of a mode/form of production. Control over value and capital flows Cane and palm companies craft, extract, and appropriate an increasingly diverse mix of commodity value portions and state revenues. Exceptions notwithstanding, they control flows of ground rent from farmland, interest from investments (e.g. in real estate or financial assets), royalties from intellectual property rights (e.g. over plant varieties and biomass fractioning and transformation processes and technology), and payments for ecosystem services (e.g. biogas generation from palm oil mill effluent (POME)). They additionally enjoy tax exemptions, preferential funding through national and international public funds, as well as state subsidies for energy and transport infrastructure mega-projects and public social grants for workers and their families. As a result, cane and palm companies can either limit or eliminate external rent claims like interests from financiers, ground rent from landlords, or taxes from the state, and reap super-profits in return. Furthermore, I flagged earlier that cane and palm companies increasingly financialize these rents to fund their operations, for instance, by securitizing land’s ground rent and payments for environmental services. At the same time, they almost completely eliminate any potential spill-over effect of their wealth concentration in the local economy (Alonso-Fradejas et al. 2008). Jobs and labour dynamics The expansion of cane and palm companies beginning in 2005 made employment numbers on cane and palm plantations grow. But these shrink after the companies achieve world-record productivity gains from 2012 onwards.14 Higher labour productivity means that fewer workers are needed. For instance, the number of cane cutters fell from 65,000 in 1990 to 35,000 in 2012, even though the area under cane cultivation doubled during the same period (ASAZGUA 2012). Higher labour productivity is the outcome of the aforementioned labour regime fix that swaps better salaries and to a lesser extent improved working conditions for harder, longer, and more casual working days. Since mechanization of cane and palm cultivation tends to be costly, unfeasible, or both, companies depend on the extension and intensification of the working day to hike labour productivity. Hence, labour productivity increases rely on flex and piecemeal work, and this is a key historical trait from 1986 to 2005 as I have argued earlier. However, this plantation labour regime becomes the rule rather than the exception from 2012 onwards (Hurtado Paz y Paz and Sánchez 2011; Alonso-Fradejas 2013).
Life purging agrarian extractivism 151 Despite or rather because of the 2012 plantation labour regime fix, risky and strenuous work on cane and palm plantations can have a serious and sometimes fatal effect on the health of workers – especially when wages are tied to working more and faster.15 Palm harvesting entails chopping down palm fruit bunches that weigh up to 40 kg and letting them free fall from 15 to 25 m. Reports of workers hit by falling palm fruit bunches are common. Fruit bunches are then loaded into water-buffalo carts and transported to the roads where trucks heading for the mill await. A truck can carry about 3,000 fruit bunches that need to be uploaded manually from the water-buffalo carts. Bruises and sprains are routine injuries for those charged with this task. In other cases, workers are asked to apply between 15 and 20 sacks of fertilizer, each one weighing around 50 kilos. Even the most experienced workers report eye and respiratory disorders and skin rashes following fertilizer application. In addition to physical exhaustion from lifting heavy weights under demanding tropical conditions marked by heat and humidity, there are the risks of cuts from thorny fronds and snakebites while walking around in the underbrush on the palm but also cane plantations. In fact, harsh work characterizes cane harvesting too. “A worker cutting 6 tonnes of cane a day in a 200- by- 6 metre area walks approximately for 4.4 km and is required to make around 66,666 machete hits and body flexions” (Alves 2006, 94–5). Dehydration-related disorders are also reported by palm plantation workers, though these seem to be less severe than those affecting cane cutters. Initially documented in Nicaragua, a fatal dehydration-related chronic kidney disease (“Mesoamerican Nephropathy”) is killing cane cutters by the hundreds in Central America (Elinder and Wernerson 2019). Additionally, the labour regime fix increased the corporate appropriation for free of productive and reproductive labour of the plantation workers’ families. On one hand, cane and palm companies appropriate the unpaid “productive” labour of children and women assisting wage-earning adult men in plantation work. This is nothing new to the sexual and generational divisions of labour in the Guatemalan northern lowlands (Alonso-Fradejas et al. 2008; Grandia 2012). But there are meaningful differences with respect to how this used to work under a time– rate wage system. For instance, unpaid family labour would traditionally support wage-earning men to finish their daily job assignments faster so they could dedicate the remaining part of the workday to the family farm. Children’s labour would usually perform as a reinforcement once school was out, and women’s labour would be devoted mainly to “reproductive” tasks, often including tending an orchard at home. But in current cane and palm plantation work under piecemeal wage systems, for many men, unpaid family labour becomes essential to achieve the equivalent of a legal minimum wage. Were it not for the support of their partners and children, plantation wage-workers would have either had to allocate extra time for their duties or hire an assistant. This is why many children quit school during the periods their fathers work for cane and palm companies. On the other hand, companies can keep piece rates low thanks to the unpaid productive and reproductive labour of women, children, and elders in the male
152 Alberto Alonso-Fradejas plantation wage-workers’ households. In families whose male adult members migrate for plantation wage-work or stay but are employed in jobs demanding long working days, adult women have taken over family farming tasks and responsibilities. For many women, this means having to extend already long and overloaded working days. In the Polochic area, “households working in oil palm plantations, and particularly women, have no time for community activities, personal care, or resting, even when they desire so, since they prefer saturating their time than abandoning or significantly reducing maize cultivation” (Mingorría et al. 2014, 863). Children wake up at 4 a.m. to fetch water and collect firewood so women can prepare coffee and tortillas before men head to the plantation by 5 a.m. And especially when head-of-household women need to take over the family farm work, household elders keep working until their last breath in a diversity of tasks such as tending the home orchard, water and firewood collection, cooking, cleaning, and taking care of children, the sick, and the injured. The social metabolism of flex cane and palm commodity production The transformation of nature into cane and palm commodities requires large amounts of energy and materials from within and outside the agro-ecosystem (Alonso-Fradejas et al. 2010). The land-use changes associated with a total clearing of land to pave the way for cane and palm plantations involve a major appropriation of environmental resources and services. Cane and palm farming and transformation similarly require extremely large quantities of soil and water nutrients. Hence, in addition to stockpiling those nutrients that exist in a plantation’s agro-ecosystem, chemical fertilizers are applied to the soil,16 while streams are diverted and underground water pumped into cane and palm fields for irrigation purposes.17 Additionally, the brunt of waste and pollutants generated throughout the process of cane and palm commodity production is transferred at zero cost, and usually with impunity, to the ecosystem and its human and non-human population in the operating areas and beyond. Most often, the land-use change to cane and palm plantations (see next section) limits the capacity of the (agro)ecosystems to perform as carbon sinks (e.g. through deforestation and peatland drainage), and thereby leads to higher carbon dioxide emissions. Cane and palm farming and processing are also waste- and contaminant-heavy processes. The soil itself serves as a dumping site and among other forms of waste and pollutants it absorbs those resulting from agro-chemical input use are striking, according to soil analysis outcomes.18 Cane leaves and stems and palm fronds are increasingly left in the soil after harvest or pruning for fertilization purposes. But the trade-off is clear: if excessive materials are used as feedstock for off-grid energy production in the cane and palm transformation plants, then there is the need to increase the use of external fertilizers in the plantations. Likewise, freshwater bodies act as carriers and depositories of waste and pollutants when cane and palm are farmed and processed. Outcomes of water analysis carried out in freshwater bodies flowing
Life purging agrarian extractivism 153 through plantations and wastewater from palm oil mills clearly show low levels of water oxygen and oxidation–reduction potential, and high levels of water’s pH acidity and temperature.19
Purging the countryside in the name of sustainability and development I move now to twine together the outcomes of the previous analyses of each set of extractiveness criteria into an integrated argument about why, how, and the extent to which cane and palm companies are driving a life purging form of predatory agro-extractivism. The rise of the flex cane and palm complexes erodes (self-)employment and undermines ecological and social reproduction regardless of whether the latter is subsistence- or business-oriented. Rather than through the macroeconomic conditions posed by the “Dutch Disease” and “Resource Curse” theses,20 these adverse effects on jobs, labour, and socio-ecological reproduction are best explained by the consideration of cane and palm companies’ predatory agro-extractivism as a process of “impairing destruction”.21 This works by means of a job-poor, culturally insensitive, toilsome, and unpaid labour-based “productive” model and the manufacturing of environmentally and socially toxic landscapes to “leave no one unscathed”. A constrained, and constraining, “productive” economy As I flagged earlier, higher labour productivity means fewer jobs in the flex crop complexes. On top of this, cane and palm are far less labour intensive than the crops commonly farmed by working people. In the Guatemalan northern lowlands cane and palm require just 36 and 52 workdays per hectare per year, respectively, whereas maize, for instance, demands 112 and chilli 184 workdays per hectare per year (Alonso-Fradejas et al. 2008; Dürr 2015). More generally, cane and palm companies constrain the conditions of possibility for other livelihoods in agriculture, forestry, fishing, or tourism. Cane and palm companies grab control over large tracts of land and restructure the local labour regime. They also hoard financial capital, royalties from intellectual property rights, payments from environmental services, and state revenues as part of powerful oligarchic family business groups that join forces under the almighty “Coordinating Committee of Financial, Industrial, Commercial and Agricultural Chambers of Guatemala” (CACIF) (Alonso-Fradejas 2015). The ensuing territorial restructuring adds wood to the fire of entrenched challenges for socioecological reproduction in the companies’ operating areas, including a skewed distribution of land resources, jobs, and wealth; a gender and generational division of labour that overburns young and female bodies; and limited access to education, health, and other social services. Additionally, the predatory agroextractivism of the flex crop complexes downplays the culturally and ecologically sound well-being-levelling role of various community institutions. These include
154 Alberto Alonso-Fradejas the yearly allocation of family land plots according to household size and composition features while keeping a communal forest reserve and fallows area, as well as a diversity of mutual support arrangements such as labour exchanges, free land leases, and the sponsoring of community works or festivities by the relatively wealthier (Alonso-Fradejas 2015). Therefore, the “productive” economy of the flex cane and palm complexes not only leaves behind the majority of the local population, but also weakens the communal fabric and institutions that enable the reproduction of “lowlander” (MayaQ’eqchi’ and mestizo) communities and identities, and a degree of value sharing which can make the difference between life and death for many underprivileged families. Toxic landscapes The flex cane and palm complexes produce environmentally and socially toxic landscapes (and waterscapes).22 On one side, the land-use changes associated with expanding plantations and the hyper-intensive forms of cane and palm farming and processing shape the weather (e.g. rain and rainfall), and constrain the abilities of the plantation agro-ecosystem to renew its stocks of energy and materials (e.g. soil nutrients). Hence, cane and palm companies search for ways to increase their resilience to extreme weather events (e.g. droughts and floods) and environmental disruption, while simultaneously increasing yields and reducing costs so they can stay in business and remain successful. To this end, they implement the knowledge fix I noted earlier and adopt soil conservation, biological pest control, and other so-called “sustainable intensification” and “climate-smart” production practices. Despite, and often because of, these acclimatization and greening efforts, cane and palm companies trigger a series of “environmental cost-shifting relations”.23 I identify four environmental cost-shifting relations associated with four separate mechanisms of dumping waste and contaminants into the soil. First, aerial spraying of agro-chemicals over cane plantations adversely affects people, crops, livestock, and forests. Second, the use of glyphosate-based herbicide in plantations negatively impacts the health of workers, villagers, and non-human species, as a broad-spectrum pesticide. Third, sustainable intensification farming practices, such as biological pest control and the use of superfluous cane and palm biomass and mill residues for soil fertilization, still involve the transfer of environmental burdens (e.g. proliferation of snakes and flies, respectively, which adversely affect people and cattle). Fourth, the costly and lengthy process of recovering the soil’s fertility after decades of intensive farming, which for palm plantations includes the uprooting of deep-growing and intertwining palms’ roots, often results in exhausted plantations simply left idle. Additionally, water flowing through plantations and from processing plants filters into underground aquifers and/or is released into rivers and streams, unleashing two more environmental cost-shifting relations. The first one concerns the adverse implications on aquatic life and biodiversity. These have been labelled by the UN in Guatemala as an “ecocide” (OHCHR 2015),24 and also damage the
Life purging agrarian extractivism 155 human use of water for farming, fishing, and animal rearing. The second is the negative implications of polluted water on human health (e.g. when drinking or bathing). Herrera and Silva report a higher incidence of gastrointestinal pathologies, hair loss, skin rashes, and eye disorders for villagers after land-use changed to palm plantations in the northern lowlands (2014, 13).25 On the other side, cane and palm companies contribute to the making of socially toxic landscapes by means of three severe social cost-shifting relations. First, the adverse effects of cane and palm farming and processing on human health and safety go beyond plantation workers. They are also felt by nearby, and sometimes also faraway, residents. Some main health issues include the illnesses associated with polluted aquifers, the plague of flies triggered by palm fronds and fruit bunches left to rot in the plantations after the kernel is extracted, and in the case of cane plantations, the aerial spraying of agro-chemicals and smoke from burning cane fields during harvest.26 Second, the flex cane and palm complexes compromise local and national food security. Land-use change analysis outcomes show that the land transformed to cane production between 2005 and 2010 in the Polochic area had been used for staple food crops and pastures in 2000. New palm plantations in the whole of the northern lowlands between 2005 and 2010 substituted forest, staple food crops, and the shifting cultivators’ fallows. Food production is also compromised by the pest of flies, plantation agro-chemicals and residues spilling over food farms and cattle ranches, and the exhausted plantation soils in which farming is not easy or feasible anymore. Increasingly scarce employment constrains working people’s ability to buy increasingly more costly food. And the public-private small scale palm contract farming program (PROPALMA) is pumped with national food security funds and framed as a “pro-poor policy to stop land grabbing”.27 Third, the Head Nurse of the government clinic in Panzós (Polochic area) explained in 2009 that “since the cane company settled in, brothels mushroomed and morbidity rates of sexually transmitted diseases skyrocketed [. . .] we are even witnessing the first HIV cases in the zone affecting migrant plantation-workers and their partners”. Similarly, women sex-workers from Fray zone argued that “before the palm companies came, the few existing ‘bares’ [brothels] and ‘chicas’ [women sex-workers] were enough. With palm’s arrival business is booming! Not only more bares opened, but also many buses packed with additional chicas arrive every fifteenth of the month when the palm lads get paid”.28 Indeed, nightlife burgeons on pay days, not least with drunk fights and gun shootings. Alcohol abuse during pay-day celebrations results in, or worsens, (domestic) violence against women and children. In the words of a Maya-Q’eqchi’ woman from the Polochic area: “most of our men working for the cane and the palm companies get drunk every 15th, and we receive them at home waiting for ‘the worst’ to happen”.29 Leaving no one unscathed . . . nor sitting idly by Flex cane and palm complexes’ predatory agro-extractivism blindly purges human and non-human life, including of friends and foes alike. For instance, the increasingly harsh conditions for socio-ecological reproduction that I described
156 Alberto Alonso-Fradejas earlier also damage the interests and well-being of non-Indigenous landed upper classes in the banana, cattle, or coffee business. But in the generally job-scarce, extremely unequal, and densely populated Guatemalan context of the early 21st century, the life purging agro-extractivism of the flex cane and palm complexes strikes especially hard the thousands of mostly Indigenous working families (and particularly women) who are deemed unfit and thereby redundant or “surplus” for the requirements of the new renewables-led corporate order. This is nothing new nor specific to Guatemala. For instance, I flagged earlier how, as elsewhere around the world during the onset of neoliberal globalization (Davis 2006), the structural adjustment of the Guatemalan economy in 1986– 2005 pushed many working people into the latent section of the relative surplus population, or that which “struggles to reproduce itself through farming and is always ready to provide the cheapest labour within a potentially expanded labourforce” (Marx 1887, 444). But from the mid-2000s onwards, the life purging agroextractivism of cane and palm companies downgrades many working people from the latent to the stagnant section of the relative surplus population, or that “with extremely irregular employment [and] characterized by maximum of workingtime and minimum of wages” (ibid.).30 Hence, the rise of the flex cane and palm complexes both swells and stagnates the reserve army of labour while simultaneously pushing the relative surplus population to the limits of subsistence through impairing destruction. Not surprisingly, cane and palm companies are facing strong opposition. This is organized around accommodative and more challenging political standpoints, both of which include a heterogenous constituency in terms of class, ethnicity, gender, and generational attributes. I discuss the politics behind the current rise of the flex cane and palm complexes in detail elsewhere (Alonso-Fradejas 2015, 2021), including the politics across non-Indigenous dominant classes (e.g. owners of cane and palm companies vs. owners of banana, coffee, or cattle haciendas). It suffices to say here that the most adamant challengers of the corporate renewables’ agro-extractivist purge frame their struggle for an alternative, transformative life project as “defence of territory”. Aiming to “move from practices of cultural resistance to the full exercise of collective rights in the living territory”,31 defence of territory brings together two interrelated visions. One is for the self-determined government of land access and control relations, in which land and other natural resources are simultaneously understood as means of production, Nature, and political territory. The other is a locally-grounded vision of food sovereignty to democratize, rejuvenate, and decrease the social metabolism of the (re)productive activities in the living territory (Alonso-Fradejas 2015).32 However, organizing in defence of territory is neither easy nor safe. Cane and palm companies have realized that their ability to successfully stay in business is increasingly contingent on their abilities to achieve license to operate at the local level and to reproduce the natural and personal conditions of production of their businesses. Hence, they have developed an “authoritarian corporate populist” political agenda (Alonso-Fradejas 2021). Authoritarian corporate populism is especially keen on manufacturing working people’s consent to the new renewables-led
Life purging agrarian extractivism 157 corporate order of the oligarchy. This involves political concessions to the underprivileged through public grants and multi-stakeholder governance, as in populist political regimes as usual. But additionally, and distinctively, authoritarian corporate populism involves meaningful corporate concessions on the realm of private relations of production. These work through the fixes in labour and knowledge relations that I discussed earlier, and act to soften the blow of cane and palm farming and processing on people and the environment. Violence, nonetheless, remains foundational to authoritarian corporate populism even if it is now cloaked in the rule of law (Alonso-Fradejas 2021).33
Conclusion: towards a renewable but unlivable future? I have explored the implications of the heightened reliance on renewables, and particularly on crop biomass, in big business-led transitions to sustainability for jobs, labour regimes, and socio-ecological reproduction. In Guatemala from 2005 onwards, the rise of corporate flex cane and palm complexes has benefitted mainly the few who managed to capitalize on land sales or leases (Grünberg et al. 2012), and/or to get a relatively stable and remunerative job (e.g. as non-manual plantation workers or in processing plants). But overall, the expansion of cane and palm companies has had adverse consequences for the most part of human and nonhuman nature. Flex cane and palm farming and processing involve a predatory form of agrarian extractivism which is driving a process of impairing destruction. This involves a job-poor, culturally insensitive, toilsome, and unpaid labourbased “productive” economy which not only leaves behind most local residents, but also exhausts especially young and female bodies, and erodes the communal fabric and institutions that allow the reproduction of more culturally and ecologically sound communities and identities and a degree of local value sharing which is helpful to everyone and vital for the underprivileged. Additionally, it is also the underprivileged who get the short end of the stick in the environmental and social cost-shifting relations behind the manufacturing of toxic landscapes by cane and palm companies. As a result, the rise of the flex cane and palm complexes both increases and stagnates the reserve army of labour while simultaneously pushing the relative surplus population to the limits of subsistence. Many refuse to accept this outcast condition and struggle to either better accommodate themselves to a greener but overly unfair and yet unsustainable future, or to challenge it through defence of territory contention for social and environmental justice. These findings call into question business-as-usual climate stewardship and sustainable development initiatives that constrain the (re)production of fairer and more climate-proof, culturally sound, and youth-friendly life projects.
Acknowledgements I am most grateful to Jun Borras, Max Spoor, Murat Arsel, Marc Edelman, Christina Schiavoni, Jan Douwe van der Ploeg, Ben McKay, Salena Tramel, Zoe Brent,
158 Alberto Alonso-Fradejas Andrés León Araya, Jochen Dürr, Elyse Mills, Jose Luis Caal, Henry Veltmeyer, Sérgio Sauer, and Annelies Zoomers for their invaluable support and comments on earlier versions of this work. I am also very grateful to the participants at the workshop on Agrarian Extractivism in Calgary University in September 2019, and to my colleagues at Utrecht University, ISS, TNI, and IDEAR for their encouragement and inspiring comments. Thank you to Gertrude Samar and Chelsea Klinke for their great copy-editing work. Last but not the least, I am especially indebted to the Maya-Q’eqchi’ and other research participants from Guatemala and elsewhere. Any remaining errors and omissions are my own.
Notes 1 Overlapping with the global COVID-19 pandemic from late 2019/early 2020. 2 These involve crops and trees, with “multiple uses (food, feed, fuel, fibre, industrial material, etc.) that can be flexibly interchanged” (Borras et al. 2016, 94). See also Goodman et al. (1987). 3 Or “agro-extractivism”. For a genealogy of the concept and current uses see AlonsoFradejas et al. (2008), McKay (2017), Ye et al. (2019), and the fellow chapters in this volume. 4 Including, for instance, the World Bank (Deininger and Byerlee 2011) and The International Food Policy Research Institute (IFPRI) (von Braun 2008). 5 Argentina, Bolivia, Brazil, Ecuador, Uruguay, and Venezuela. 6 “Based on the process of capitalist-created scarcities of external and human nature”, this contradiction pushes towards an economic crisis which “assumes the form of a ‘liquidity crisis’ or under-production of capital” (O’Connor 1988, 18–19; see also Moore 2016). Indeed, O’Connor examined the contradictions between capital and its natural and social conditions of production. This is why I use the term “socio-ecological” reproduction. However, I rely on the richer frames and analyses of Fraser and other critical feminist intellectuals for the examination of the contradictions between capital and its social conditions of production. 7 “On the one hand, social reproduction is a condition of possibility for sustained capital accumulation; on the other, capitalism’s orientation to unlimited accumulation tends to destabilize the very processes of social reproduction on which it relies” (Fraser 2016, 100). 8 For instance, McKay argues that the current flex soy complex in Bolivia involves a type of “agrarian extractivism [which is] characterized by four interlinked dimensions: (1) large volumes of materials extracted destined for export with little or no processing; (2) value chain concentration and sectoral disarticulation (3) high intensity of environmental degradation; and (4) deterioration of labour opportunities and/or labour conditions” (McKay 2017, 199). 9 Meaning the “specific methods of mobilizing labour and organizing it in production, and their particular social, economic and political conditions” (Bernstein 1988, 31–2). 10 Fraser (2016) understands these as the institutionalization of the social relations of affective and material labour which set the social-reproductive conditions for different forms of capitalist production over time and across geographies. My interest is on the conditions for the social reproduction of labour power and of “communities and identities” (Bakker and Gill 2003, 26). I still differentiate here between “productive” and “reproductive” labour for explanatory purposes. Nonetheless, I agree with Mezzadri’s argument regarding “that only interpretations of social reproduction activities and realms as value-producing can advance our understandings of labour relations of contemporary capitalism” (Mezzadri 2019, 33 emphasis in original). I am very grateful to Zoe Brent for our inspiring and ongoing conversation on the social reproduction question.
Life purging agrarian extractivism 159 11 These draw on “detailed historical ethnographies to illuminate multileveled processes in ways that are enabling of critical practices and the formation of alternatives” (Hart 2002, 819). 12 Usefully summarized by Bernstein in four fundamental critical political economy questions: “i) who does what? ii) who gets what? iii) who owns what? and iv) what do they do with it?” (Bernstein 2010, 22–3). 13 Indeed, class power might not change face but it ages. And so a new generation of oligarchs, men and women between 25 and 45 years old, have gradually taken over key managerial positions in the family businesses since the turn of the century. They are usually alumni of the elitist and libertarian Francisco Marroquin University in Guatemala, or of European and U.S. Ivy League universities from which they often also hold postgraduate degrees. For instance, the CEO of the leading Guatemalan cane company holds a BS in Mechanical Engineering and an MS in Industrial Engineering from Cornell University in the United States where he studied during 1988–1993 (Jaramillo 2016). 14 Achieving 7 palm oil tons/hectare in Guatemala vs. the 4 tons/hectare world’s average (GREPALMA 2016). And according to the Head of Human Resources of a large Guatemalan cane company, “the most skilled of our cane cutters can harvest 15 tons of cane a day! These are well above world average figures of 2 or 3 tons” (interview in January 2014). 15 In 2015 the UN in Guatemala condemned “the practice of conditioning salaries on reaching productivity goals imposed unilaterally by the [cane and palm] companies. As a result, overtime is not remunerated and workers’ physical integrity and health have been affected” (UNHRC 2015, 16). 16 Cane and palm cultivation in the karstic soils of the northern lowlands depends heavily on chemical fertilizers. Cane production demands “large amounts of fertilizer and other agro-chemicals” (interview with owner, head agronomic engineer, and security chief of Polochic’s Chabil Utzaj cane company, February 2008). The neighbouring palm company applies as much as 429 kg/ha/year of NPK fertilizer and 34.3 kg of boron (interview with Head Engineer of NaturAceites Polochic Palm Oil mill, March 2008). 17 In 2003, or before the tremendous flex crop expansion from 2005 onwards, cane was the first and palm the third largest irrigation water user in Guatemala. Together they used 56% of the total irrigation water. By 2010 cane remained the top user and palm use expanded more than two-fold – reaching second place. In 2010 the two flex crops hoarded 61% of the country’s irrigation water (Gálvez and Andrews 2015, 159). 18 Together with mineral fertilizers, flex cane and palm companies apply fungicides, pesticides, and herbicides. And at least in Fray, Polochic, Ixcán, and Sayaxché zones, palm companies use glyphosate-based herbicide (Monsanto’s “Roundup”). Large volumes of these agro-chemicals which are not synthetized by canes and palms are released into the atmosphere as volatized nitrogen, filtered underground, or spilled into the Caribbean Sea or the Pacific Ocean via freshwater currents. 19 Following an agreement brokered by the author and Jose Luis Caal between the Institute of Agrarian and Rural Studies (IDEAR) and the Institute of Hydro-biological Research (IIH) of Guatemala’s National University (USAC), samples were collected in June 2013 in areas in which freshwater bodies run through palm plantations, close to palm oil mills, and elsewhere. The parameters evaluated included: (i) total dissolved solids (TDS) in milligrams/litre; (ii) salinity in grams/kilogram; (iii) electrical conductivity in microsiemens/centimetre; (iv) dissolved oxygen in milligrams/litre; (v) oxidation–reduction potential in millivolts; (vi) water pH; and (vii) water temperature in degree Celsius. 20 Which discuss the effects of national currency appreciation following increased exports of commodities on the competitiveness of other domestic exports (Auty 1993). 21 This concept revisits the claim that capitalism is an endless and organic process of “creative destruction” popularized by Schumpeter. For him, capitalism involves an
160 Alberto Alonso-Fradejas
22 23
24
25
26 27 28 29 30
31 32
33
“evolutionary process [. . .] that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one” (Schumpeter 1943, 1952 83, 82). Hence, impairing destruction applies to situations in which a new socio-ecological order destroys the old one at the same time that it constrains or eliminates its very own socio-ecological conditions of possibility. These are cases in which the efforts that highly predatory capital makes to save itself from itself result in an exclusive, unfair, and toxic socio-ecological order that threatens its own reproduction, let alone the reproduction of alternative life projects. I am grateful to Andres León Araya and Diana Ojeda for suggesting this formulation. Martínez-Alier and O’Connor introduced the notion of “ecological distribution” to refer to the “social, spatial and temporal asymmetries or inequalities in the use by humans of environmental resources and services (whether traded or not), for example, in the depletion of natural resources [. . .] and in the burdens of pollution” (1999, 381). Hence, I understand the social relations of contaminants and waste transfer as environmental cost-shifting (distribution) relations. In June 2015, millions of fishes and other aquatic and amphibious animals floated dead through 150 km of La Pasion River flowing through northern Guatemala and Mexico. They suffocated due to “malathion”, a chemical component used in palm oil mill effluent oxidation lagoons, which spilled over into plantation drainages and the river (CMI 2015). These environmental cost-shifting relations triggered by cane and palm companies call into question the outcomes of the environmental impact assessments (EIAs) prepared by allegedly independent third parties, let alone of those directly elaborated by the companies like the self-EIA of Polochic’s Chabil Utzaj cane company approved by the government (interview with Chabil Utzaj’s representatives, February 2008). Interview with chief physician of the government clinic in Telemán (Polochic area), June 2009. Interview with PROPALMA Director, September 2009. Interview in October 2013. Similar accounts were offered by women sex-workers in Panzós (April 2008), Sayaxché (June 2011), and Chisec (October 2013). Group interview in March 2008. Similar issues were reported by women in Ixcán (June 2011), Chisec (December 2009 and June 2011), and Sayaxché (June 2011). Poverty rates in Guatemala grew from 51% in 2006 to 59.3% in 2014 (World Bank 2019). “Chronic child malnutrition is at about 50%, the highest rate in Latin America, and fourth-highest rate in the world. The proportion of the population living with hunger has increased by 80% over the past 20 years, from about 17% in 1991, to 30.5% in 2012” (Taft-Morales 2014, 12). On demographic, employment, wages, costs of the food/living baskets, inequality, and poverty trends see the National Statistics Bureau of Guatemala and the World Bank’s National Living Standards Measurement Study surveys. Statement by an influential Maya-Q’eqchi’ ideologue, lawyer, and Member of Congress during a workshop in Guatemala City, June 2013. The Maya-Q’eqchi’ refer to land in Spanish as “trabajadero” (workplace) and in Q’eqchi’ as “li ch’och” (the Earth). Traditionally, they understand land as a living and life-giving entity that must be cared for and respected by them, Her children; the “R’al Ch’och” (Children of the Earth). Life, human and otherwise, is guarded by the “Tzuultaq’as”, the Mountain-Valley spirits which inhabit the territories where the Maya-Q’eqchi’ reside. Rule of law violence unfolds through direct repression as well as through the criminalization of protest and protesters. In 2014 alone, 814 attacks had been “directed at human rights defenders who work in the main problems affecting the country’s human rights, such as those dedicated to defending the rights of Indigenous peoples, territory, land, and environment” (IACHR 2015, 18).
Life purging agrarian extractivism 161
Bibliography Acosta, A. 2013. Extractivism and neoextractism: two sides of the same curse. In M. Lang and D. Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Quito and Amsterdam: Transnational Institute & Rosa Luxemburg Foundation, 61–86. Alonso-Fradejas, A. 2013. “Sons and Daughters of the Earth”: Indigenous Communities and Land Grabs in Guatemala. Land & Sovereignty in the Americas series. Oakland and Amsterdam: FoodFirst & TNI. Available at: https://foodfirst.org/publication/sons-anddaughters-of-the-earth-indigenous-communities-and-land-grabs-in-guatemala/. Alonso-Fradejas, A. 2015. Anything but a story foretold: multiple politics of resistance to the agrarian extractivist project in Guatemala. Journal of Peasant Studies, 42(3–4), 489–515. Alonso-Fradejas, A. 2021. The Rise of Authoritarian Corporate Populism. Latin American Perspectives (forthcoming). Alonso-Fradejas, A., and S. Gauster. 2006. Perspectivas para la agricultura familiar campesina de Guatemala en un contexto DR-CAFTA. ASC & IDRC/CRDI. Guatemala: IDEAR. Alonso-Fradejas, A., F. Alonzo, and J. Dürr. 2008. Caña de azúcar y palma africana: Combustibles para un nuevo ciclo de acumulación y dominio en Guatemala. Guatemala: IDEAR and Magna Terra Editores. Alonso-Fradejas, A., G. Gamboa, and S. Mingorría. 2010. Los agrocombustibles y la profundización del capitalismo agrario flexible en territorios campesinos de Guatemala. In VIII Congreso Latinoamericano de Sociología Rural (ALASRU). Brasil: Porto de Galinhas. Alves, F. 2006. Por que morrem os cortadores de cana. Saúde e Sociedade, 15(3), 90–8. Arsel, M., B. Hogenboom, and L. Pellegrini. 2016. The extractive imperative in Latin America. The Extractive Industries and Society, 3(4), 880–7. ASAZGUA. 2012. Ingenios activos y empleo. Asociacion de Azucareros de Guatemala. Available at: www.azucar.com.gt/ingenios.html. Auty, R.M. 1993. Sustaining Development in Mineral Economies: The Resource Curse Thesis. London: Routledge. Bakker, I., and S. Gill (eds.). 2003. Power, Production, and Social Reproduction: Human in Security in the Global Political Economy. New York: Palgrave Macmillan. Bernstein, H. 1988. Labour regimes and social change under colonialism. In B. Crow, M. Thorpe, and D. Wield (eds.), Survival and Change in the Third World. Cambridge: Cambridge Polity Press, 30–49. Bernstein, H. 2010. Class Dynamics of Agrarian Change. Halifax: Fernwood. Borras Jr, S.M., J.C Franco, S.R. Isakson, L. Levidow, and P. Vervest. 2016. The rise of flex crops and commodities: implications for research. Journal of Peasant Studies, 43(1), 93–115. Borras Jr., S.M., T. Moreda, A. Alonso-Fradejas, and Z.W. Brent. 2018. Converging social justice issues and movements: implications for political actions and research. Third World Quarterly, 39(7), 1227–46. Brent, Z.W., A. Alonso-Fradejas, G. Colque, and S. Sauer. 2018. The “tenure guidelines” as a tool for democratising land and resource control in Latin America. Third World Quarterly, 39(7), 1367–85. CMI. 2015. La Pasión: Desastre ecológico y social. Centro de Medios Independientes. Available at: https://cmiguate.org/la-pasion-desastre-ecologico-y-social/.
162 Alberto Alonso-Fradejas Davis, M. 2006. Planet of Slums. London: Verso. Deininger, K., and D. Byerlee. 2011. Rising Global Interest in Farmland: Can It Yield Sustainable and Equitable Benefits? Agriculture and Rural Development. Washington, DC: World Bank. Available at: https://siteresources.worldbank.org/DEC/Resources/RisingGlobal-Interest-in-Farmland.pdf. Dürr, J. 2015. Diez mitos y realidades sobre las cadenas agroalimentarias en Guatemala. Revista de Estudios Sociales, 79(I), 75–93. Edelman, M., C. Oya, and S.M. Borras Jr. 2013. Global Land Grabs: historical processes, theoretical and methodological implications and current trajectories. Third World Quarterly, 34(9), 1517–31. Elinder, C.G., and A.O. Wernerson. 2019. Mesoamerican Nephropathy. Erasmus University Rotterdam Medical Center. Available at: www.uptodate.com/contents/mesoameri can-nephropathy (Accessed 28 September 2019). Franco, J.C., T. Kramer, A. Alonso-Fradejas, H. Twomey, and P. Vervest. 2015. The Challenge of Democratic and Inclusive Land Policymaking in Myanmar. Amsterdam: TNI. Available at: https://www.tni.org/files/download/the_challenge_of_democratic_and_ inclusive_land_policymaking_in_myanmar.pdf. Fraser, N. 2016. Contradictions of capital and care. New Left Review, 99–117, July–August. Gálvez, J., and K.L. Andrews. 2015. Perfil del agro y la ruralidad de Guatemala 2014: situación actual y tendencias. Guatemala: URL & IICA. Goodman, D., B. Sorj, and J. Wilkinson. 1987. From Farming to Biotechnology: A Theory of Agro-Industrial Development. Oxford: Basil Blackwell. GRAIN. 2008. Seized: The 2008 Land Grab for Food and Financial Security. Barcelona: GRAIN. Available at: www.grain.org/article/entries/93-seized-the-2008-landgrab-forfood-and-financial-security. Grandia, L. 2012. Enclosed: Conservation, Cattle, and Commerce Among the Q’eqchi’ Maya Lowlanders. Seattle: University of Washington Press. GREPALMA. 2016. La palma en Guatemala. Gremial de Palmicultores de Guatemala. Available at: www.grepalma.org/palmicultura-en-guatemala?o=4 (Accessed 21 November 2018). Grünberg, G. et al. 2012. Tierra e igualdad: Desafíos para la administración de tierras en Petén, Guatemala. Guatemala and Washington, DC: World Bank. Gudynas, E. 2010. The New Extractivism of the 21st Century: Ten Urgent Theses about Extractivism in Relation to Current South American Progressivism. Americas Program Report. Washington, DC: Center for International Policy. Gudynas, E. 2013. Extracciones, extractivismos y extrahecciones: un marco conceptual sobre la apropiación de recursos naturales. Observatorio del desarrollo, 18, 1–18. Hart, G. 2002. Geography and development: development/s beyond neoliberalism? Power, culture, political economy. Progress in Human Geography, 26(6), 812–22. Herrera, A.H., and R. Silva. 2014. Huellas de plaguicidas en el Refugio de Vida Silvestre Punta de Manabique y sus efectos Socioeconómicos. Guatemala: Universidad de San Carlos (USAC). Hurtado Paz y Paz, L. 2008. Dinámicas agrarias y reproducción campesina en la globalización: el caso de Alta Verapaz, 1970–2007. Guatemala: F&G Editores. Hurtado Paz y Paz, L., and G. Sánchez. 2011. ¿Qué tipo de empleo ofrecen las empresas palmeras en el municipio de Sayaxché, Petén? Guatemala: Action Aid. IACHR. 2015. Situation of Human Rights in Guatemala: Diversity, Inequality and Exclusion. No. OEA/Ser.L/V/II. Doc. 43/15. Washington, DC: Inter-American Human Rights Commission, 1–221.
Life purging agrarian extractivism 163 Jaramillo, V. 2016. Luis Fernando Leal, el CEO de la expansión de Ingenio Magdalena. Estrategia & Negocios. Available at: www.estrategiaynegocios.net/lasclavesdeldia/994790-330/ luis-fernando-leal-el-ceo-de-la-expansión-de-ingenio-magdalena. Martínez-Alier, J., and M. O’Connor. 1999. Distributional issues: an overview. In J.C.J.M. van den Bergh (ed.), Handbook of Environmental and Resource Economics. Cheltenham: Elgar, 381–92. Martínez-Alier, J., G. Kallis, S. Veuthey, M. Walter, and L. Temper. 2010. Social metabolism, ecological distribution conflicts, and valuation languages. Special Section: Ecological Distribution Conflicts, 70(2), 153–8. Martínez-Torres, M.E., and P.M. Rosset. 2014. Diálogo de Saberes in La Vía Campesina: food sovereignty and agroecology. Journal of Peasant Studies, 41(6), 979–97. Marx, K. 1887 [1867]. Capital: A Critique of Political Economy. Volume I: The Process of Production of Capital. Moscow: Progress Publishers. Marx, K. 1894. Capital: A Critique of Political Economy. Volume III: The Process of Capitalist Production as a Whole. F. Engels (ed.). New York: International Publishers. McCreery, D. 1994. Rural Guatemala, 1760–1940. Stanford, CA: Stanford University Press. McKay, B.M. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211. McKay, B.M., A. Alonso-Fradejas, Z.W. Brent, S. Sauer, and Y. Xu. 2017. China and Latin America: towards a new consensus of resource control? Third World Thematics: A TWQ Journal, 1(5), 592–611. Mezzadri, A. 2019. On the value of social reproduction: informal labour, the majority world and the need for inclusive theories and politics. Radical Philosophy, 2(4), 33–41, Spring. Mingorría, S., G. Gamboa, B. Martín-López, and E. Corbera. 2014. The oil palm boom: socio-economic implications for Q’eqchi’ households in the Polochic valley, Guatemala. Environment, Development and Sustainability, 16(4), 841–71. Moore, J.W. 2016. Introduction: Anthropocene or capitalocene? Nature, history, and the crisis of capitalism. In J.W. Moore (ed.), Anthropocene Or Capitalocene? Nature, History, and the Crisis of Capitalism. Oakland, CA: PM Press, 1–13. O’Connor, J. 1988. Capitalism, nature, socialism: a theoretical introduction. Capitalism Nature Socialism, 1(1), 11–38. OHCHR. 2015. Press Conference on the Ecocide of La Pasión River: Findings by the OHCHR. Guatemala: Office of the United Nations High Commissioner for Human Rights (press release 21 July). Palma, G., A. Taracena, and E. Baumeister. 2004. Cambios en la tenencia de la tierra: tendencias históricas. Guatemala: UNDP-ASDI. Ratha, D., S. De, S. Plaza, K. Schuettler, W. Shaw, H. Wyss, and S. Yi. 2016. Migration and Remittances: Recent Developments and Outlook. Migration and Development Brief. No, 26. Washington, DC: World Bank, 1–48. Rice, J. 2007. Ecological unequal exchange: consumption, equity, and unsustainable structural relationships within the global economy. International Journal of Comparative Sociology, 48(1), 43–72. Schumpeter, J.A. 1952. Socialism, Capitalism and Democracy. Fourth edition. London: George Allen & Unwin. Schwab, K. 2016. The Fourth Industrial Revolution: What It Means and How to Respond, Agenda. World Economic Forum. Available at: www.weforum.org/agenda/2016/01/ the-fourth-industrial-revolution-what-it-means-and-how-to-respond/. Svampa, M. 2013. Consenso de los Commodities y lenguajes de valoración en América Latina. Nueva sociedad, 244, 30–46.
164 Alberto Alonso-Fradejas Taft-Morales, M. 2014. Guatemala: Political, Security, and Socio-Economic Conditions and U.S. Relations. Washington, DC. Available at: https://fas.org/sgp/crs/row/R42580. pdf. UNHRC. 2015. Annual Report of the United Nations High Commissioner for Human Rights, Addendum: Report of the United Nations High Commissioner for Human Rights on the Activities of His Office in Guatemala. Geneva: UN Human Rights Council. Available at: www.ohchr.org/EN/HRBodies/HRC/. . ./Session28/. . ./A_HRC_28_3_Add_1_ ENG.doc. Veltmeyer, H., and J. Petras. 2014. The New Extractivism in Latin America. London: Zed. von Braun, J. 2008. Food and Financial Crises: Implications for Agriculture and the Poor, Food Policy Report. Washington, DC: International Food Policy Research Institute (IFPRI). Available at: www.ifpri.org/publication/food-and-financial-crises. World Bank. 2019. Guatemala: Sustainable Development Statistics. Available at: https:// data.worldbank.org/country/guatemala?view=chart (Accessed 11 February 2020). Ye, J., J.D. van der Ploeg, S. Schneider, and T. Shanin. 2019. The incursions of extractivism: moving from dispersed places to global capitalism. The Journal of Peasant Studies, 1–29. Zoomers, A., and K. Otsuki. 2017. Addressing the impacts of large-scale land investments: re-engaging with livelihood research. Geoforum, 83, 164–71.
8
Extractive agave and tequila production in Jalisco, Mexico Darcy Tetreault, Cindy McCulligh, and Carlos Lucio
In many parts of the state of Jalisco, the rural scenery is dominated by fields of “blue” agave. These monoculture plantations provide the raw material (Agave tequilana Weber var. azul) for producing Mexico’s emblematic spirit: tequila. In response to growing demand for this alcoholic beverage, blue agave plantations have expanded rapidly since the turn of the millennium. They have also become more “extractive” in nature, we argue, because of the way intensified production has affected the environment, smallholder farmers, and field workers. Further, we argue in this chapter that, because industrialized production of tequila is so polluting, because the value added in manufacturing the spirit is concentrated in the hands of a few large mostly foreign firms that dominate the market, and because the largest volumes of tequila are exported to the United States, the processing of agave into tequila in Mexico actually accentuates the extractive nature of the entire value chain. Accordingly, this chapter seeks to challenge the notion that only agricultural production primarily destined for export, without any valueadded downstream through industrialized processing, is “extractive” in nature. Our approach draws from ongoing field research carried out in the state of Jalisco and elsewhere in Mexico, an extensive literature review, and analysis of relevant data made available by governmental agencies and business organizations. We examine the degree to which transnational companies control the agave–tequila value chain, the portion of tequila that is exported, the technologies employed in production and processing, company–farmer and company–worker relations, relevant public policy, water use and contamination, and other environmental and social impacts. We begin with a discussion of the concept of agro-extractivism, drawing links to the related concept of social metabolism and pointing to how this leads to an investigation that examines the flows of materials, contaminants, and money. From there, the next two sections trace the history and material flows of agave–tequila production in Jalisco, from the Colonial Era through to the neoliberal period. The extractive nature of extensive monoculture agave cultivation and industrialized tequila production is examined in Section 4, illustrating the major environmental and social impacts. In Section 5, we outline the alternative model represented by small-scale agave farming and traditional mezcal production. Finally, the conclusions sum up our argument for expanding the concept of agro-extractivism
166 Tetreault et al. to include some crops channelled to domestic industrial processing and, at the same time, for focusing it more precisely on the negative social and environmental impacts of accelerated biomass extraction by large corporations for the sake of profit.
Distilling the concept of (agro)extractivism Gudynas (2010, 13) defines extractivism as “activities that remove large quantities of natural resources that are not processed (or processed only to a limited degree), particularly for export”. His analysis is focused on the South American experience during the first decade of the 21st century, when governments in the region from across the political spectrum responded to a primary commodities boom by taking measures to increase the rate of extracting materials from the earth, mostly for export. This is what Svampa (2018) refers to as “the commodities consensus”. While Gudynas’ (2010) focus is on the mining and petroleum sectors, he observes a form of “agricultural extractivism” in the expansion of transgenic soy, where production is controlled by transnational firms, highly mechanized, intensive in its use of chemical herbicides, involves little or no domestic processing, and is mostly destined for export. Indeed, corporate-controlled genetically modified soy has come to epitomize the concept of agro-extractivism (see e.g. Giarracca and Teubal 2014; Petras and Veltmeyer 2014, 62–100; Gudynas 2015, 45–7; Ezquerro-Cañete 2016; McKay 2017). With a focus on soy production in Bolivia, McKay (2017, 203) defines agro-extractivism in the following terms: “(1) large volumes of materials extracted destined for export with little or no processing; (2) value-chain concentration and sectoral disarticulation (3) high intensity of environmental degradation; and (4) the deterioration of labor opportunities and labor conditions in the area/sector”. There is no doubt that these are the most salient characteristics and consequences of massive soy production in Bolivia and elsewhere in the so-called “Soy Republic” of South America, including Brazil, Argentina, and Paraguay. The question is: do these terms adequately describe all forms of agro-extractivism? Our argument in this chapter is that even though agave and tequila from Mexico is not “destined for export with little or no processing”, it does nevertheless entail the extraction of large volumes of materials, value-chain concentration, high-intensity environmental degradation, deteriorating labour conditions, and – something not mentioned in the definition provided earlier, but examined by, for example, Ezquerro-Cañete (2016) – deteriorating conditions for smallholder farmers; and as such it should be considered a form of extractivism. Blue agave production in Mexico is not the only manifestation of agroextractivism that brings into question the criterion of “little or no [domestic] processing”. Two other crops that are frequently mentioned in the literature as being produced in an agro-extractive manner are palm oil and sugar cane (Petras and Veltmeyer 2014, 63; Gudynas 2015, 133; Svampa 2018, 22; Alonso-Fradejas 2019). Like soy, these two crops are considered to be “flex crops”, since they can be channelled to different value chains (food, animal feed, or biofuels), depending
Extractive agave and tequila production 167 on international market signals. However, the agro-extractivism literature to date generally turns a blind eye to the domestic industrial processes that convert these crops into biofuels and, in the case of sugar cane, into sugar, rum, and processed foods. This is odd considering, for example, that Brazil is the second largest producer of biofuels in the world after the United States. Furthermore, as several authors have suggested, extractivism is not a new phenomenon; it can be traced back over 500 years to the Conquest, when Europeans began extracting materials from the Americas for their exchange value in Europe (Bunker 1984; Acosta 2013; Girvan 2014; Gudynas 2015; Svampa 2018). As Girvan (2014) points out, sugar was at the centre of the Atlantic mercantile economy that flourished during the Colonial era. Whether this constituted “extractive imperialism” or not “is a matter of definition”, he affirms (2014, 51), “but arguably it satisfied several of the criteria”. One of these, however, was not little to no domestic processing of raw materials. Sugar mills were established throughout Latin America and the Caribbean during the Colonial period to process sugar cane into sugar, as were distilleries to produce rum, and as these processes became increasingly industrialized over time, their polluting consequences intensified. With this in mind, this chapter uses the case of agave–tequila production as a source of empirical evidence to argue for (re)conceptualizing agro-extractivism with a focus on the negative social and environmental consequences stemming from large-scale biomass extraction and processing, taking into consideration the technologies employed, the social relations of extraction and evolving regulatory frameworks. This resonates with Giarracca and Teubal’s (2014, 48) proposal to apply the concept to “a certain type of agriculture in which essential resources such as water and fertile land, and biodiversity, are degraded”. It implies investigating the extractive nature of agricultural activities by tracing the volumes and flows of biomass, water, and contaminants, as well as the flow of money, taking into consideration the consequences for workers and smallholder farmers. From this perspective, the connection between the rapid expansion in Latin America of “activities that remove large quantities of natural resources” (Gudynas 2010, 13), and the acceleration of the global economy’s social metabolism, comes into view. The concept of social metabolism, which was first employed in the social sciences by Marx and Engels, refers to a relation of material exchange between human society and the biophysical environment, mediated by labour (Marx 1977, 283–92). It is a concept that has been used as a point of departure for ecological economists who, especially since the 1990s, have developed methods for quantifying the flows of materials through national economies (FischerKowalski 1998; Fischer-Kowalski and Hüttler 1999). In this way, Krausmann et al. (2009) have demonstrated that despite the technological advances to make processes of material flow more ecologically efficient in the mould of “sustainable development”, the global economy’s social metabolism (measured in terms of the weight of material throughput) has continued to accelerate in the neoliberal era. This tendency received further impetus in the context of the 2003–2012 primary commodities boom, largely driven by rapid economic growth in China, which
168 Tetreault et al. increasingly has become a destination for agro-export commodities from South America. Put simply, the global economy’s increasing social metabolism – ultimately driven by capitalism’s intrinsic orientation to endless capital accumulation – has rapidly expanded the extractive frontier in Latin America under the structural conditions of neoliberal globalization. In this context, Dittrich and Bringuezu (2010) observe a net flow of materials from countries in the global South to those in the North, with the notable exceptions of Australia and Canada, two industrialized countries that are net exporters of materials. Likewise, Martínez Alier and Walter (2016) estimate that the rate of extracting materials from Latin America increased four-fold between 1970 and 2008, reaching over eight billion metric tons in the last year of their analysis. During this period, they observe that the region (particularly South America) experienced persistent and increasing physical trade deficits, measured as the difference between the weight of exports and that of imports. They and other political ecologists characterize this trend as “ecologically unequal exchange”, noting that it does not take into account environmental and social “externalities” on the local level. In Marxist terms, the net flow of materials from the global South to the North – and of financial assets valorized through the appropriation of resource rents – amounts to a form of “ecological imperialism” or what Veltmeyer and Petras (2014) call the “imperialism of the twenty-first century”. To be sure, ecological imperialism and agro-extractivism have manifested in distinct ways in Mexico, largely because of the country’s proximity to and commercial relations with the United States, but also because of the course of class struggle in Mexico. To begin with, as a result of the historical struggle for land, which culminated in the Mexican Revolution of 1910–1920 and in the agrarian reform that ensued, over half of the country’s territory is in the hands of ejidos, which are collective landholdings belonging to groups of smallholder farmers, most of whom hold usufruct rights over individual parcels of land. Because of this unique agrarian structure, which has remained relatively stable in spite of changes made to the Mexican Constitution and to the Agrarian Law in 1992 to allow for the privatization of ejidal land, Mexico has not been subject to “land grabbing” in the same way as in South America, where large-scale land acquisitions by foreign governments and transnational corporations are linked to agro-extractivism (Petras and Veltmeyer 2014, 62–6). As Borras and his collaborators (2012, 407) note, “(re)concentration of capital and land [in Mexico] occurs not through land grabbing, but through foreign and domestic corporate control of the agribusiness value chain”. As illustrated later, this is the case for blue agave cultivation and tequila production, which is controlled by large tequila companies who rent ejidal and privately owned land. From a broader perspective, Mexico stands out in the Latin American context because its economy has not been “reprimarized” to the same extent as those of many countries in South America. Between 2000 and 2017, primary commodities represented only 21.4% of Mexico’s total exports, compared to around 70% for Argentina, Colombia, and Uruguay; over 80% for Chile and Peru, and over 90%
Extractive agave and tequila production 169 for Bolivia, Ecuador, and Venezuela in the same period.1 This is because Mexican exports continue to be dominated by manufactured goods produced in the maquiladora sector, especially automotive parts, electronics, and electrical appliances.2 In 2017, 80% of all Mexican exports were destined for the United States, including 78% of all animal, vegetable, and food products. Instead of producing mostly flex crops for export to Asian markets as in South America, Mexican agro-food exports are made up mostly of “luxury” goods for US markets, particularly berries, fruits and vegetables, and alcoholic beverages, of which the most important are tequila and beer (Otero et al. 2013). In 2017, tequila was the fifth most important agro-food export from Mexico in terms of value, representing 4.1% of the value of exports, with tequila exports valued at US$1.34 billion (SIAP 2018). Our investigation is focused on the state of Jalisco, which is not only the state with the highest value of agricultural production in Mexico but also the origin and centre of blue agave production and where all but a handful of tequila distilleries are located. A history of agave and tequila production Pre-Hispanic Indigenous groups in Mesoamerica used the agave plant for a variety of purposes, including to make clothing, as a food source, and to produce fermented alcoholic beverages. There is debate over whether or not distillation technology was developed before the Conquest. In any case, the Philippine technology introduced by the Spanish spread rapidly throughout the region during the 17th century. By the end of the 18th century, elite families of Spanish descent (including the Cuervo family) were operating four distillery-taverns in the vicinity of the town of Tequila, producing what was known at the time as mezcal from Tequila (Luna Zamora 1999, 45). From then on, the cultivation of agave and its distillation into tequila – a regionally specific type of mezcal – have been the main economic activities in the area around Tequila and in the neighbouring municipalities of Amatitán and El Arenal. During the first half of the 19th century, including the tumultuous years of struggle for independence from Spain (1810–1821), the planting of agave proliferated throughout this region, as did the number of distilleries (Luna Zamora 2015, 79). Agave was grown on large haciendas, where dispossessed Indigenous people and landless mestizo peasants worked, and also on small- and mediumsized “ranches” (Luna Zamora 1999, 2015, 58, 75). Artisanal technology was employed to process agave into tequila; for example, cooking agave hearts (called “piñas”) underground, crushing them with mallet blows in hollowed-out tree trunks, fermenting the juice in leather containers or in underground pits lined with stones, and distilling alcohol with rudimentary stills made of wood and iron (Luna Zamora 2015, 81, 87). In response to growing demand in Guadalajara and beyond, larger tequila manufacturers began introducing innovations to industrialize this process in the 1860s, including wood-saving brick-built ovens, steam engines to move large circular heart-crushing stones, and modern stills imported from France.
170 Tetreault et al. The region’s agrarian structure remained relatively stable until the last two decades of the 19th century (Lizama Silva 2019, 189), when, during the Porfirio Díaz dictatorship (1876–1911), a small number of tequila manufacturers were able to acquire huge tracts of land in order to vertically integrate the entire agave–tequila production process. This was the period during which the largest and most powerful tequila companies were established or consolidated, including Herradura, Sauza, and José Cuervo. In addition, growing demand for tequila encouraged investors to plant agave and establish distilleries elsewhere in the state of Jalisco, especially in the Highlands of Jalisco, located to the northeast of Guadalajara. With the development of railway and maritime port infrastructure, the large tequila companies in Jalisco began to export their products to the United States in growing volumes, reaching over 13 thousand litres by the year 1900 (Luna Zamora 1999, 81). Production declined thereafter. The Mexican Revolution from 1910 to 1920 and the land reform carried out in the aftermath combined to create an extended period of economic instability for hacienda owners and tequila manufacturers. The number of agave plants dropped from 70 million to 4 million between 1900 and 1940, while tequila production fell 8.71 million litres to just 1.65 million litres (Luna Zamora 2015, 122). From the 1930s onwards, the breaking up of haciendas and their redistribution to landless peasants had the effect of disarticulating the vertically integrated production of agave and tequila. The beneficiaries of the agrarian reform, who collectively obtained usufruct rights over land in the form of ejidos, were inclined to plant maize, beans, and other subsistence crops. Thus, when the demand for tequila rose – as it did in the United States during the Second World War – tequila manufacturers were obliged to enter into various forms of contracts with ejidal farmers in order to entice them into planting agave (Luna Zamora 1999, 158–9). This relationship of dependence on smallholder farmers to supply agave was reinforced during the period of state-led import-substituting industrialization (ISI) from 1940 to 1982. Given the long maturation period of blue agave, the absence of regional long-term planning and coordination between large tequila producers led to cycles of agave over-supply and shortage (Luna Zamora 1999; Zizumbo Villarreal et al. 2009; Bowen and Gaytán 2012; Gaytán and Bowen 2015). The most effective measure taken by tequila manufacturers to deal with agave shortages was to adulterate their product by adding sugars from other plants, especially sugar cane. In an effort to confront this tendency, in 1949, the first federal standard regulating tequila defined it as a mezcal obtained in its entirety from different varieties of agave tequilana Weber. However, in the context of increasing US demand for tequila in the 1950s and 1960s, stemming largely from the growing popularity of margaritas and other mixed drinks, large tequila companies pressured for changes to this standard, giving rise to the 1964 norm allowing tequila to be degraded with sugars from other sources, up to a maximum of 30%. This norm was further relaxed in 1970 to permit the use of up to 49% of foreign sugars. In this way, tequila manufacturers were able to increase the total volume of their production by a factor of 13 between 1950 and 1980 (reaching 60 million litres in the last
Extractive agave and tequila production 171 year), even though agave cultivation only increased by a factor of 2.5 during the same period (Luna Zamora 2015, 152–3). Paradoxically, the Denomination of Origin (DO) for tequila was established in 1974, just four years after reducing the required agave-derived content and about the same time as foreign capital began to penetrate the industry. The largest tequila firms in Jalisco petitioned for it through the Tequila Industry’s Regional Chamber (now the National Chamber of the Tequila Industry), successfully obtaining the first product outside of Europe to be protected by a DO (Gaytán and Bowen 2015). The DO sought to restrict the use of the label “tequila” to agave distillates produced from at least 51% Agave tequilana Weber var. azul grown anywhere in the state of Jalisco and in specific municipalities of the neighbouring states of Guanajuato, Michoacán, and Nayarit. In 1977, the DO was expanded to include several municipalities in the state of Tamaulipas, in spite of the region’s lack of history in agave or tequila production and only due to the lobbying of wealthy agro-industrialists who had planted agave there in anticipation of increasing US demand (Gaytán and Bowen 2015, 277). As Bowen and Valenzuela Zapata (2009, 118) observe, the large size of the area included in the DO for tequila, “allows the tequila firms to adopt a ‘frontier’ production strategy, externalizing ecological and social costs and moving on to another area after resources have been exhausted”. In the mid-1980s, the tequila industry employed about 1,500 factory workers, 760 administrative employees, 82 technical professions, and approximately 15,000 agricultural workers (Luna Zamora 1999, 184). At that time, smallholder farmers still often intercropped agave with maize, beans, and other crops during the first four years of its growth; and many farmers only planted agave on hillsides and marginal lands. Agro-chemicals were used, but much less so than today, since plant diseases were not as problematic as they would become in the following decades when monoculture agave plantations spread to completely dominate certain regions of Jalisco and beyond.
Foreign takeover and the intensification of agave and tequila production Structural adjustments were applied to the Mexican economy during the 1980s, in the context of the debt crisis, to initiate the transition from state-led industrialization to a market-led strategy based on the principles of free trade, privatization, deregulation, and labour flexibility. In this context, domestic tequila sales lagged, while exports to the United States rose spectacularly, exceeding the volumes sold in Mexico from the mid-1980s onwards (Luna Zamora 2015, 196–7). The DO for tequila was strengthened in 1993 with the creation of the Tequila Regulatory Council (CRT), charged with protecting the DO in Mexico and abroad, ensuring compliance with tequila norms and providing information for producers (CRT 2016). Bowen and Gaytán (2012, 84) highlight how the CRT – which brings together tequila companies, agave producer associations, and government agencies – is in fact governed by supply-chain actors, thus representing a clear case where “private companies themselves are the major regulatory agent”. This has
172 Tetreault et al.
1600
350
1400
300
1200
250
1000
200
800
150
600
100
400
50
200
0
Tequila produced
Tequila exported
Agave consumption
Figure 8.1 Tequila production and exports and agave consumption, 1995–2019 Source: Based on statistics of the Tequila Regulatory Council
0
Thousands of tons
400
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Milions of liters at 40% abv
led to the flexibilization of standards governing tequila in the interests of transnational investors, while simultaneously protecting the industry from “tequila-like” products from other regions (Gaytán 2017). Since NAFTA came into effect, the production and exportation of agave distillates have soared. Total production of tequila more than tripled between 1995 and 2019, increasing from 104.3 to 351.7 million litres. In 2019, just over 70% of tequila produced in Mexico was exported (see Figure 8.1). Of this volume, 83% went to the United States. A new trend has been the “premiumization” of the tequila market, reflected in the increase in exports of 100% agave tequilas, from 1.9% in 1995 to 52.7% in 2019. This has driven an increase in the consumption of agave, which has almost quintupled from 278,700 tons in 1995 to 1.34 million tons in 2019. Processes of concentration and centralization in the ownership of big tequila companies have taken place since the turn of the millennium. In 2005, four firms (Cuervo, Sauza, Herradura, and Cazadores) controlled about two-thirds of the total tequila market (Bowen and Valenzuela Zapata 2009). These firms have entered into alliances and mergers with the largest transnational liquor companies in the world. Bacardi acquired 100% ownership of Cazadores in 2002; Beam Global (now Beam Suntory) bought Sauza in 2005; and Brown Forman acquired all of Herradura in 2006 (Bowen and Gaytán 2012). The tequila industry has been described as oligopolistic, given the high degree of concentration (Orozco 2019; Macías and Valenzuela 2009). While in 2019 the CRT reported that 142 distilleries marketed 1,441 brands of tequila, market information indicates that just ten companies controlled over three quarters of the market (measured in terms of volume), and only two of these are domestically owned: Becle and Casa Centinela (see Table 8.1) (Becle 2018). Mexico’s
Extractive agave and tequila production 173 Table 8.1 Owners of main tequila brands Main brands
Company
Owners
Headquarters
Global market share (as % of volume – 2018)*
José Cuervo, 1800, Maestro Tequilero/ Dobel, Centenario Sauza, Tres Generaciones, 100 Años, Hornitos El Tesoro Patrón
Becle, SAB de CV
Beckmann Family
Mexico
29.8%
Sauza Tequila
Beam Suntory
United States
12.4%
Bacardi Limited
Bermuda
8.9%
United States
7.1%
United Kingdom
5.8%
La Alteña Patrón Spirits International AG Bacardi Limited
Tequila Cazadores, Corzo, 4 Vientos Herradura, El Jimador, Brown-Forman Pepe Lópz, Don Eduardo, Tesoro del Dragón Don Julio Diageo PLC Casamigos Casamigos Spirits Company DeLeón The Deleon Spirits Company San Matías, Del Rey, Casa San Matías Pueblo Viejo Corazón Tequila Sazerac Siete Leguas Siete Leguas Distillery Pernod Ricard Olmeca Altos Tequila Avión Centinela, Cabrito Casa Centinela El Mayor, Exotico, Destiladora Juárez González Lux
Diageo Plc
Diageo/Sean United States Combs Sazerac
Luxco
United States
4.7%
France
3.0%
Mexico United States
2.5% 1.9%
Source: Authors’ elaboration * Based on information reported by Becle.3
National Institute of Statistics and Geography (INEGI) reports that tequila and mezcal manufacturing formally employed 5,728 people nationwide in 2019, equal to just 0.01% of the country’s economically active population. The largest player, Cuervo, which claims to have a 30% market share, is in Mexican hands, owned by the Beckmann family (listed in 2019 as number 452 of global billionaires) (Becle 2018). After an agreement with UK-based Diageo covering distribution outside Mexico expired in 2013, Cuervo took over distribution
174 Tetreault et al. and later traded its stake in the Don Julio brand to Diageo for Irish whiskey Bushmills. More recently, in 2017 Diageo acquired Casamigos, a brand started by actor George Clooney, for US$1 billion, and in 2018 Bacardi purchased the popular Patrón label for a reported US$5.1 billion (Kirsch 2018). In 2004, Tequila was recognized by Mexico’s Secretary of Tourism as a “Pueblo Mágico” (Magic Town) to help promote tourism. Two years later, UNESCO awarded World Heritage status to the “agave landscape” of the Amatitán-Tequila valley, strangely celebrating the aesthetics of what is essentially a green monoculture desert. This has led to a lucrative tourist industry. According to Gaytán and Bowen (2015), two companies, Tequila Herradura (Brown-Forman) and Cuervo, have been able to capture the bulk of tourist flow (see also Hernández López 2009). The growing demand for tequila has fuelled an increase in agave harvested, peaking at nearly 2 million tons in 2014, while at the same time the area planted in the DO has tended to decrease since its high point of nearly 150,000 ha in 2008 (see Figure 8.2). Map 1 presents the geography of agave extraction for the 2003–2017 period, where areas in southern Jalisco and neighbouring states play a key role in addition to the traditional cultivation areas near Tequila (Valles region) and the Highlands, particularly the municipalities of Arandas (Altos Sur) and Atotonilco (Ciénega region), home to the factories of, for example, Cazadores, Patrón, and Don Julio. The cyclical crises in agave production have led to increasing vertical integration of the largest tequila companies. At the same time, intermediaries known as “coyotes” have become key players for acquiring the agave not grown under contract with a distillery. The presence of coyotes is stimulated by the large distillers, as they are able to drive down the price paid to small independent producers,
Millions of tons
140.00 120.00
1.5 100.00 80.00 1 60.00 40.00
0.5
20.00 0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Tons harvested
Area planted
Figure 8.2 Agave harvested and area planted in municipalities of the DO, 2003–2017 Source: Authors’ elaboration based on data from SIAP (2019)
Thousands of hectares
160.00 2
Extractive agave and tequila production 175 who are unable to sell their agave directly to tequila companies and often unable to comply with the administrative requirements of the CRT and other government agencies (Orozco 2019; Hernández López 2014). The role of intermediaries is also bolstered by the fact that contracts with tequila companies are often treated as options to buy without risk for the companies, who will purchase the agave if needed or leave it when the harvest is not required (Orozco 2019). In general, as observed by Macías and Valenzuela (2009, 468), tequila companies are able to “act as a cartel and surpass the negotiating power of the agave producer organizations”.
The social and environmental impacts of industrialized agave and tequila production Until the 2000s, almost all of the agave used to make tequila was grown in and around Tequila and in the Highlands of Jalisco. However, after the agave shortage between 1999 and 2003 and in the context of rapidly increasing external demand within the framework of NAFTA, the largest tequila companies began taking control of agave production and expanding it into other parts of Jalisco and into neighbouring states recognized by the DO (Bowen and Gerritsen 2007; Bowen and Valenzuela Zapata 2009; Gaytán and Bowen 2015). Geographical expansion was accompanied by increasing corporate control over the production process via “reverse leasing arrangements”, which oblige smallholder farmers to relinquish control not only of their lands but of all aspects of agave cultivation. As Bowen and her collaborators observe (Bowen and Gerritsen 2007; Bowen and Valenzuela Zapata 2009; Gaytán and Bowen 2015), through such leasing arrangements, big tequila companies are in a position to externalize environmental costs during the six- to eight-year period the plants now take to mature. The negative social and environmental consequences of corporate-controlled intensive agave production can be summed up in the following terms: 1
2
Displacement of traditional crops partially oriented to subsistence, especially maize, beans, and squash. Martínez Rivera et al. (2007) found that, when agave was introduced between 1999 and 2002 into the municipality of Tonaya, Jalisco, the area planted with corn fell by half. While in the recent past, smallholder farmers tended to plant agave mostly on hillsides and poorer quality soils, since the late 1990s it has been increasingly introduced on flat and fertile land where tractors can be used (Hernández López 2014, 40–1). Also, the practice of intercropping agave with other crops has been abandoned (Hernández López 2014, 45–6). Marginalization of small-scale agave farmers, whose role is relegated to renting land and whose knowledge of traditional (and more ecologically sustainable) agave farming methods, is being delegitimatized and lost (Bowen and Gerritsen 2007; Martínez Rivera et al. 2007; Bowen and Valenzuela Zapata 2009; Hernández López 2014).
176 Tetreault et al. 3
4
5
Loss of (genetic) biodiversity. First, the widespread planting of blue agave has displaced the planting of other species of agave, including the “green agave” (Agave angustifolia Haw) that was traditionally planted in southern Jalisco (Hernández López 2014; Rodríguez Contreras et al. 2017). Second, changes in land use for agave plantations are one of the major causes of deforestation in Jalisco (Graf and Santana 2019). Third, in and around agave plantations, scientists have detected a decrease in the diversity and abundance of birds and beetles (Martínez Rivera et al. 2007). And, fourth, genetic diversity has diminished within the “blue agave” species because of the widespread use of an asexual form of plant reproduction through the propagation of hijuelos that grow around the mother plant, as well as clonal micropropagation techniques in vitro. Intensive use of agro-chemicals. After a massive pest infestation killed 25% of the agave population in Jalisco between 1993 and 1999, large tequila companies began applying increasing quantities of agro-chemicals (Bowen and Valenzuela Zapata 2009; Hernández López 2014). The intensive monoculture cultivation systems used by these companies make the agave more susceptible to pests and disease, which in turn increases the need to apply pesticides (Bowen and Valenzuela Zapata 2009; Rodríguez Contreras et al. 2017). In the municipality of Autlán, agave growing companies apply as many as 17 different agro-chemicals, including extremely toxic and carcinogenic pesticides, two of which are prohibited in the United States (Martínez Rivera et al. 2007). Soil erosion. The intensive use of herbicides strips the soil of most of its vegetative cover, making it more susceptible to erosion. In addition, where agave is planted on hillsides, corporate-controlled agave cultivation systems tend to establish furrows that are parallel to the slope of the hills, in order to facilitate harvesting by providing a path to roll down agave hearts, even though this also facilitates soil erosion (Bowen and Gerritsen 2007; Bowen and Valenzuela Zapata 2009; Rodríguez Contreras et al. 2017; Herrera Pérez et al. 2018).
As regards to social impacts, the field research of Bowen and Gerritsen (2007, 476) in southern Jalisco indicates that “reverse leasing arrangements exacerbate power imbalances between the local farmers and the tequila companies, by allowing the tequila companies to become less reliant on small agave producers and more self-sufficient in their supply of agave”. These arrangements are said to offer “limited positive impacts on smallholder incomes or agricultural productivity”, and “potentially allow firms to externalize environmental costs” (ibid., 476., see also Rodríguez Contreras et al. 2017). Nevertheless, smallholder farmers are enticed into entering into such arrangements for at least two reasons. First, in the context of NAFTA, the commercial production of maize and other traditional crops has become economically unfeasible, as low-cost imports from highly subsidized large-scale farmers in the United States brought producer prices down, at the same time as production costs
Extractive agave and tequila production 177 increased because of the dismantling of governmental agencies and programmes that had provided small-scale Mexican farmers with support. Second, the money paid by the tequila companies provides a secure source of income, freeing small landowners from carrying out agricultural work, which is something particularly attractive to older farmers at a stage of semi-retirement (Bowen and Gerritsen 2007; Rodríguez Contreras et al. 2017). Tequila companies pay landowners somewhere between 1,500 and 2,500 pesos per hectare annually (Luna Zamora 2015, 157). Thus, in the municipality of Autlán, smallholder farmers who rent their land to blue agave producers earn 10% more than they would by planting maize under the best of circumstances (Martínez Rivera et al. 2007). Moreover, under the contractual agreement, small landowners have been able to collect government subsidies from the Program for Direct Support for Rural Areas (PROCAMPO), which was introduced to cushion the effects of trade liberalization and to help farmers to transition to export crops with a comparative advantage. Hernández López (2014) found that smallholder farmers who rent their land to a tequila company or its intermediaries sometimes end up accepting jobs as day labourers (jornaleros) for the same company. This is part of a process that he dubs the jornalerización of agave production, which is characterized by the atomization of the tasks involved in intensive agave production, whereby different members of work crews perform specialized tasks under supervised direction, along the lines of Taylorism. In this scheme, labour costs are kept as low as possible by hiring Indigenous migrants from southern Mexico, generally to perform the jobs that require the least training and the greatest physical exertion: clearing land, planting, carrying plants, loading trucks with piñas (which can weigh as much as 40–60 kilograms), and applying agro-chemicals (Hernández López and Porraz Gómez 2011; Hernández López 2014). The Tequila Regulatory Council estimates that the sector employs 24,000 labourers in agave production, with variable wages, working conditions, and terms of contract. Since the late 1990s, big tequila companies have created their own agricultural departments and have hired agronomists to take charge of and expand the production of agave on rented land. In some cases, these departments hire labourers directly; in others, hiring is outsourced to intermediaries, so as to avoid having to pay for social security and more generally having to adhere to labour laws. In a survey carried out in the municipality of Tequila, Padilla Miranda (2017) found that 42% of agave-producing labourers did not have social security and only 22% had a permanent contract. The same survey found that the wages earned by most workers in agave fields range from 850 to 1,600 pesos per week; crew bosses and operators of machinery make between 1,350 and 2,100 pesos per week; and jimadores – the skilled labourers who perform the task of harvesting piñas – can earn up to 3,000 pesos per week. These estimates for wages match well with the ones reported by Hernández López (2014) in a state-wide investigation. As for Indigenous migrant workers, their average wage was estimated by Hernández López and Porraz Gómez (2011, 174) to be 1,000 pesos a week for 8- to 12-hour workdays, “without any kind of
178 Tetreault et al. social security”. Considering that the cost of a basic food basket in rural Mexico is calculated to be 1,100 pesos per month per person (equal to about US$52 dollars), the income earned by most labourers is barely enough to meet the basic food needs of a family of four. Finally, in our analysis of the extractive nature of agave production in Mexico, it is important to analyse the environmental impacts of industrialized tequila manufacturing. In order to do this, we must briefly describe this process, which begins with cleaving and roasting of the piñas (agave hearts). Traditionally, the pieces of the agave heart are roasted in earthen ovens heated with firewood, which imbues them with flavour; however, industrial processes use steel autoclave ovens or diffusers (hydrolysis of raw agave), because they are faster and more efficient (Bowen and Gaytán 2012; Gaytán and Bowen 2015). Mechanical grinders for milling have also replaced the stone mills (tahonas) used to extract juices from roasted piñas. Fermentation, traditionally done in wooden tanks, has transitioned to stainless steel vats, often with the addition of chemical accelerants, despite adversely affecting the quality of the beverage. Steel stills have also supplanted copper stills for distillation, although this too reduces the quality of the product (Lucio 2018). Tequila production generates residual materials and effluents that contaminate soil and water when not properly managed, which is commonly the case in Jalisco (Martínez Rivera et al. 2007; Rodríguez Contreras et al. 2017). After the piñas are milled, an acidic fibrous material is left behind called bagasse. One litre of tequila produces about three kilos of bagasse. Poorly managed disposal of this residual material “can lead to the contamination of aquifers through lixiviation, create phytosanitary risks, and negatively affect the fertility of agricultural land” (Rodríguez Contreras et al. 2017, 88). In fact, in 2015 the disposal of large quantities of bagasse in a site without proper protection contaminated the town of San Luis Tenango’s water well, forcing residents to obtain water from tanker trucks for a period of several months (ibid.). Moreover, one litre of tequila produces between 7 and 11 litres of liquid effluents, known as vinasses, which are even more harmful to the environment (Martínez Rivera et al. 2007; Rodríguez Contreras et al. 2017). Vinasse effluents have temperatures of up to 90oC, are highly acidic (pH less than 4), have a high organic load (chemical oxygen demand in the range of 20,000–35,000 mg/L), and contain heavy metals, making them harmful to aquatic life forms. Distilleries in Jalisco have permits to discharge 8.2 million cubic metres per year, equivalent to 260 litres per second. Researchers from the Mexican Institute of Water Technology (IMTA) estimated in 2013 that only 15% of tequila vinasses received treatment (Estrada et al. 2013). The vast majority of tequila distilleries are located in the Santiago River basin, considered the most polluted river in Mexico (McCulligh 2020). Studies in this basin have detected highly polluting tequila effluents, violating the Mexican wastewater discharge standard (CEA-IMTA 2011; CEA-AyMA 2007). In February 2020, the Jalisco state government included the effluents of Cuervo, Patrón, and Siete Leguas on a list of 29 companies in violation of discharge standards in this basin.
Extractive agave and tequila production 179 The vast majority of discharge permits for tequila producers stipulate the effluent must be infiltrated into soil or used in agricultural irrigation, a practice that quickly degrades the soil and is often averted through clandestine discharge into water bodies. The head of the ecology department in the Highlands municipality of Arandas, for example, estimates that 50% of tequila vinasses are illegally discharged into water bodies.4 Many large distilleries in the Highlands region outsource the problem by taking their vinasses to a “treatment plant” in the municipality of Ayotlán, which is authorized to dispose of them via agricultural use. However, this plant was provisionally shut down in 2018 by the Jalisco Environmental Protection Agency (PROEPA) when it was found to be discharging the untreated vinasses into a nearby river.5 Finally, the National Chamber of the Tequila Industry (CNIT) exerted its influence in 2018 to oppose modifications to the national wastewater standard in Mexico (NOM-001-SEMARNAT-1996), claiming that updating the current lax norm would put the “viability of the industry” at risk and could lead to the closure of companies unable to comply or for which treatment costs would exceed production costs.6
The peasant alternative: agro-ecological agave and traditional mezcal production Tequila is just one type of agave spirit, the one that has been most industrialized. There are many others, including, raicilla, bacanora, tuxca, zihuaquio, tepe, barranca, quitupan, and comiteco. In Mexico, families of smallholder farmers and master distillers (many of whom are Indigenous) produce a variety of mezcals in 26 states of the republic, using 53 species of agave, 48 of which are of wild origin. In Oaxaca alone, more than 42,000 families derive their livelihoods from mezcal production (Gobierno de Oaxaca 2015, 55). In the model of agro-ecological forestry management that has emerged around small-scale agave production in Mexico, the plants are often reproduced using seeds to promote genetic diversity. Numerous species of agave plants are then intercropped with food crops in diversified agro-ecosystems, which helps to increase the presence of beneficial flora and fauna (including pollinators), reduce the incidence of pests, conserve ecosystem services, and protect water resources, while also sequestering CO2 and favouring edaphic biodiversity (Zizumbo Villarreal et al. 2009; Torres 2018; Lucio 2018; Nobel 2011). In the processing of agave into mezcal, traditional producers use artisanal technology to manufacture relatively small amounts of distilled spirits, between 500 and 1000 litres per month. The technologies employed typically include earthen ovens heated with firewood, manual milling or the use of stone-wheel mills for grinding cooked agave hearts, fermentation in wooden tanks or stone-lined wells, and distillation in copper stills, occasionally with clay or wooden still-heads. The producers of traditional mezcal generate relatively small quantities of liquid and material wastes. As such, they can be managed in an ecologically sustainable way. Vinasse effluents are neutralized by mixing the ashes from agave-cooking ovens to reduce their acidity, and then spread on agricultural fields as a biological
180 Tetreault et al. fertilizer. Bagasse is used to make bricks to construct buildings or briquettes to reduce the use of firewood. Many small-scale producers let some bagasse pile up beside their processing units, where they produce edible mushrooms during the rainy season. To be sure, the commercial success of traditional mezcals in recent years entails significant risks in both environmental and social terms. The boom in sales has placed greater pressure on wild agave plants, which take between 15 and 20 years to grow and are becoming more intensively exploited to meet the demands of consumers in search of exotic flavours. Also, traditional producers are under constant pressure to intensify production processes and accelerate extraction rates, compromising the carrying capacity of local ecosystems. Fair-trade networks are just beginning to be constructed, building on the pioneering efforts of individual producers and intermediaries, and on at least one experience of collective organizing in the Sansekan Tinemi mezcal brand. In an effort to bring together small-scale agave farmers and mezcal producers who seek to protect the biocultural heritage of traditional mezcal production, the Environmental Studies Group developed a strategy for agave forest management in the state of Guerrero and convened the first National Meeting of Agave Forestry Managers, which took place in February 2015. The Non-Timber Forest Products Research Network created in 2016 has held three more national meetings, which serve as a space to exchange knowledge, information on management techniques, experiences of collective organization, and concerns and proposals about the DO for mezcal and its corresponding regulatory framework. This DO, established in 1994, also tends to favour the interests of large distillers instead of protecting the knowledge and techniques of small-scale producers. While part of the alternative pursued by the research network is to pressure for changes to the existing regulatory framework, traditional mezcal producers and fair-trade merchants also seek to circumvent DO regulations by pursuing alternative labelling options under the category of “agave spirits”.
Conclusions In this chapter, we have sought to challenge the notion that, to be considered extractive, biomass produced in an ecologically and socially destructive manner must be export oriented with little or no domestic processing. We have argued that this stipulation is arbitrarily restrictive and turns a blind eye not only to the extractive nature of corporate-controlled agricultural production of crops destined mostly for domestic industrial processing, but also to the social and environmental impacts of the industrialization process itself. As an alternative approach to conceptualizing agro-extractivism, we have proposed retaining and building on the core notion that it has to do with extracting from the earth “large quantities of natural resources”. From this view, the closely related concept of social metabolism links the expansion of the (agro)extractive frontier to the increasing volumes of materials flowing through the global economy, thereby exposing the limitations of technological advances aimed at greater
Extractive agave and tequila production 181 ecological efficiency within the framework of a dominant mode of production that is geared to boundless capital accumulation. Epistemologically, our approach involves evaluating the extractive nature of agro-industrial activities in a holistic manner that revolves around tracing the flows of materials, pollutants, and money. It seeks to take into consideration the social relations of extraction and production, the technology employed, and relevant regulatory schemes, and it focuses on the social and environmental consequences. Accordingly, our analysis has highlighted how the production of agave and tequila in Mexico has become progressively more extractive over time, with a leap towards predatory forms of agro-extractivism under the structural conditions of neoliberal globalization. Since NAFTA came into effect in the mid-1990s, the extraordinary increase in the rate of extracting agave biomass has been made possible by technological and organizational changes whose negative social and environmental impacts include the marginalization of smallholder agave farmers, deteriorating labour conditions for field workers, loss of (genetic) biodiversity, intensive agro-chemical use, deforestation, and soil erosion. Through reverse leasing arrangements, big tequila companies have vertically integrated, taking greater control of agave production. Simultaneously, foreign capitals acquired many of these companies, facilitating the outflow of tequila and profits from Mexico. Furthermore, industrialized tequila production exacerbates the extractive nature of the entire agro-industrial process by generating highly toxic effluents that are dumped untreated into waterways, as well as poorly managed material wastes that are also contaminating. Traditional mezcals constitute an alternative for conscientious consumers and a defence strategy for smallholder farmers in the face of aggressive corporate strategies to control market shares of agave distillates. Smallholder farmers throughout Mexico, but especially in Indigenous regions, apply agro-ecological forest management to produce multiple varieties of genetically diverse agave in an ecologically and socially sustainable manner. Small-scale mezcal producers manage the relatively small volumes of effluents and bagasse generated to avoid contaminating water and soil. Traditional mezcals conserve a vast diversity of aromas and flavours and, above all, an enormous cultural and gastronomic heritage in their preparation. With the rise in their popularity, networks are currently under construction to bring together traditional mezcal producers, action researchers, and promoters of fair trade who seek to conserve the sustainability and cultural knowledge of traditional mezcal production.
Notes 1 Based on data available on the website of the Economic Commission for Latin America and the Caribbean (www. www.cepal.org/en/datos-y-estadisticas). 2 As argued by Tello (1996), the exportation of manufactured goods from Mexico’s maquiladora sector constitutes a form of exporting cheap labour, which is embodied in maquila products. This, combined with the massive flow of migrant workers to the United States, makes cheap labour Mexico’s most important export (Tello 1996; Cypher 2010). Cypher (2010, 637) goes so far as to suggest that “Central America and Mexico are actually
182 Tetreault et al.
3 4 5 6
caught in what might be termed ‘sub-primarization’”. However, as demonstrated in this chapter while cheap embodied labour is indeed exported from Mexico through the agave–tequila value chain, so are large volumes of primary materials, which are also cheap because tequila companies are effectively able to externalize environmental costs. www.cuervo.com.mx/documents/presentations/2019/Corporate%20Prez%20JPM%20 Conference%202019.pdf, consulted June 2020. Interview, September 2019. https://semadet.jalisco.gob.mx/prensa/noticia/742, consulted February 2020. Comment of CNIT to the National Regulatory Improvement Commission on the proposed change to the NOM-001-SEMARNAT-1996, http://187.191.71.192/expediente/ 21218/recibido/59416/B000181068, consulted February 2020.
References Acosta, A. 2013. Extractivism and neoextractism: two sides of the same curse. In M. Lang and D. Mokrani (eds.), Beyond Development: Alternative Visions from Latin America. Quito and Amsterdam: Fundación Rosa Luxemburg and Transnational Institute, 87–104. Alonso-Fradejas, A. 2018. El proyecto de capitalismo agroextractivo: una mirada a la cuestión agraria contemporánea desde Guatemala. In B. Rubio (coord.), América Latina en la mirada: Las transformaciones rurales en la transición capitalista. Mexico City: UNAM, Instituto de Investigaciones Sociales. Becle. 2018. Corporate Presentation. Available at: www.ircuervo.com/documents/presen tations/2018/Corporate%20Presentation%20Nov%2018%20-%20Santanders%20 Consumer%20Day.pdf. Borras, S., C. Kay, S. Gómez, and J. Wilkinson. 2012. Land grabbing and global capitalist accumulation: key features in Latin America. Canadian Journal of Development Studies, 33(4), 402–16. https://doi.org/10.1080/02255189.2012.745394. Bowen, S., and M.S. Gaytán. 2012. The paradox of protection: national identity, global commodity chains, and the tequila industry. Social Problems, 59(1), 70–93. https://doi. org/10.1525/sp.2012.59.1.70. Bowen, S., and P. Gerritsen. 2007. Reverse leasing and power dynamics among blue agave farmers in western Mexico. Agriculture and Human Values, 24(4), 473–88. doi:10.1007/ s10460-007-9088-7. Bowen, S., and A. Valenzuela Zapata. 2009. Geographical indications, terroir, and socioeconomic and ecological sustainability: the case of tequila. Journal of Rural Studies, 25(1), 108–19. https://doi.org/10.1016/j.jrurstud.2008.07.003. 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), 1017–64. Comisión Estatal de Agua y Saneamiento, Jalisco and AyMA Ingeniería y Consultoría. 2007. Estudio de actualización de clasificación del río Verde y la parte Alta del río Santiago, Jalisco. Guadalajara: CEAS. Comisión Estatal del Agua and Instituto Mexicano de Tecnología del Agua. 2011. Actualización del estudio de calidad del agua del Río Santiago (desde su nacimiento en el Lago de Chapala, hasta la Presa Santa Rosa). Mexico City: IMTA. Consejo Regulador del Tequila. 2016. Agave y tequila: Un binomio de éxito. Guadalajara: Editorial Temacilli. Cypher, J. 2010. South America’s commodities boom: developmental opportunity or path dependent reversion? Canadian Journal of development Studies, 30(3–4), 635–62.
Extractive agave and tequila production 183 Dittrich, M., and S. Bringuezu. 2010. The physical dimension of international trade part 1: direct global flows. Ecological Economics, 69(9), 1838–47. https://doi.org/10.1016/j. ecolecon.2010.04.023. Estrada, E., M.A. Garzón, G. Mantilla, and M.S. Salazar. 2013. Desarrollo de una tecnología de tratamiento para la industria de bebidas alcohólicas (tequila, informe 2013. IMTA. Available at: http://repositorio.imta.mx/handle/20.500.12013/1550. Ezquerro-Cañete, A. 2016. Poisoned, dispossessed and excluded: a critique of the neoliberal soy regime in Paraguay. Journal of Agrarian Change, 16(4), 702–10. https://doi. org/10.1111/joac.12164. Fischer-Kowalski, M. 1998. Society’s metabolism: the intellectual history of material flow analysis, part I, 1860–1970. Journal of Industrial Ecology, 2(1), 61–78. doi:10.1162/ jiec.1998.2.1.61. Fischer-Kowalski, M., and W. Hüttler. 1999. Society’s metabolism. the intellectual history of material flow analysis, part II, 1970–1998. Journal of Industrial Ecology, 2(4), 107–36. https://doi.org/10.1162/jiec.1998.2.4.107. Gaytán, M.S. 2017. The transformation of tequila: from hangover to highbrow. Journal of Consumer Culture, 17(1), 62–84. https://doi.org/10.1177/1469540514556169. Gaytán, M.S., and S. Bowen. 2015. Naturalizing neoliberalism and the de-Mexicanization of the tequila industry. Environment and Planning A, 47(2), 267–83. https://doi. org/10.1068/a130281p. Giarracca, N., and M. Teubal. 2014. Argentina: extractivist dynamics of soy production and open-pit mining. In H. Veltmeyer and J. Petras (eds.), The New Extractivism: A PostNeoliberal Development Model or Imperialism of the Twenty-First Century? London: Zed Books, 47–79. Girvan, N. 2014. Extractive imperialism in historical perspective. In J. Petras and H. Veltmeyer (eds.), Extractive Imperialism in the Americas: Capitalism’s New Frontier. Leiden: Brill, 49–61. Gobierno de Oaxaca. 2015. Quinto Informe de Gobierno. Available at: www.gabinocue. org/wp-content/uploads/2016/01/AVANCES-YLOGROS-5-2015.pdf. Graf, S., and E. Santana. 2019. Naturaleza y medio ambiente. In A. Acosta (ed.), Jalisco a futuro 2018–2030, Construyendo el porvenir, Volumen 1, Diagnósticos. Guadalajara: Universidad de Guadalajara. Gudynas, E. 2010. El nuevo extractivismo del siglo XXI. Diez tesis sobre el extractivismo bajo el progresismo sudamericano actual. Revista Memoria, 242–3(12–17), 24–30. Gudynas, E. 2015. Extractivismos. Ecología, economía y política de un modo de entender el desarrollo y la Naturaleza. Cochabamba: CEDIB. Hernández López, J.J. 2009. Tequila: Centro mágico, pueblo tradicional. ¿patrimonialización o privatización? Andamios, 6(12), 41–67. Hernández López, J.J. 2014. La jornalerización en el paisaje agavero. Actividades simples, organización compleja. Mexico City: CIESAS. Hernández López, R.A., and I.F. Porraz Gómez. 2011. ¿De paisano a paisano? Explotación laboral y exclusión social de jornaleros chiapanecos en Jalisco. Nómadas, 34, 167–81. Herrera Pérez, L., E. Valtierra Pacheco, I. Ocampo Fletes, M.A. Tornero Campante, J.A. Hernández Plascencia, and R. Rodríguez Macías. 2018. Agricultural contract plans for the production of Agave Tequilana Weber in the region of Tequila, Jalisco. Agricultura, Sociedad y Desarrollo, 15(4), 619–37. doi: 10.22231/asyd.v15i4.903. Kirsch, N. 2018. Bacardi to acquire Patrón Tequila In $5.1 billion deal: Billionaire owner will cash out. Forbes, 22 January. Available at: www.forbes.com/sites/
184 Tetreault et al. noahkirsch/2018/01/22/bacardi-to-acquire-patron-tequila-for-5-1-billion-billionaireowner-will-cash-out/#7dc402513590. Krausmann, F., S. Gringrich, N. Eisenmenger, K.H. Erb, H. Haberl, and M. FischerKowalski. 2009. Growth in global material use, GDP and population during the 20th century. Ecological Economics, 68(10), 2696–705. doi:10.1016/j.ecolecon.2009.05.007. Lizama Silva, G. 2019. Reforma agraria en Tequila (Jalisco, México), 1915–1980. Revista de Historia, 26(2), 183–217. doi: 10.4067/S0717-88322019000200183. Lucio, C. 2018. Traditional Mezcal Production in Zapotitlán de Vadillo. In D. Tetreault, C. McCulligh, and C. Lucio (eds.), Social Environmental Conflicts in Mexico: Resistance to Dispossession and Alternatives from Below. New York: Palgrave Macmillan, 255–76. https://doi.org/10.1007/978-3-319-73945-8_8 Luna Zamora, R. 1999. La historia del tequila, de sus regiones y sus hombres. Second edition. Mexico: CONACULTA. Luna Zamora, R. 2015. La construcción cultural y económico del tequila. Mexico: Universidad de Guadalajara, Núcleo Universitario Los Belenes. Macías, A., and A.G. Valenzuela. 2009. El tequila en tiempos de la mundialización. Comercio Exterior, 59(6), 459–72. Martínez Alier, J., and M. Walter. 2016. Social metabolism and conflicts over extractivism. In F. Castro, B. Hogenboom, and M. Baud (eds.), Environmental Governance in Latin America. New York: Palgrave Macmillan, 58–85. https://doi.org/10.1007/978-1137-50572-9_3. Martínez Rivera, L.M., P. Gerritsen, J.J. Rosales Adame, A. Moreno, S. Contreras, A. Solís, L.E. Rivera, O.J. Cárdenas, L.I. Iñiguez, Cuevas, C. Palomera, E. García, A. Aguirre, and J.L. Olguín. 2007. Implicaciones socioambientales de la expansión del cultivo de agave azul (1995–2002) en el municipio de Tonaya, Jalisco, México. In P. Colunga-García Marín, A. Larqué, L.E. Eguiarte, and D. Zizumbo Villarreal (eds.), En lo ancestral hay un futuro: del tequila, los mezcales y otros agaves. Mexico City: CICY, CONACYT, CONABIO, SEMARNAT, INE, 265–86. Marx, K. 1977. Capital: A Critique of Political Economy. Volume I. New York: Vintage Books. McCulligh, C. 2020. Alcantarilla del progreso: corporaciones, corrupción institucionalizada y la lucha por el río Santiago. Buenos Aires: Ediciones CICCUS; Alcalá de Henares: Instituto Universitario de Investigación en Estudios Latinoamericanos (IELAT), y Red WATERLAT-GOBACIT. http://doi.org/10.5072/zenodo.572379. McKay, B. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211. https://doi.org/10.1016/j.worlddev.2017.04.007. Nobel, P.S. 2011. Sabiduría del desierto, agaves y cactos: CO2, agua, cambio climático. Texcoco: Editorial del Colegio de Postgraduados. Orozco, J.L. 2019. Panorámica actual de la industria tequilera. In Manual Técnico del Tequilero. Mexico City: Consejo Regulador del Tequila, 7–39. Otero, G., G. Pechlaner, and E.C. Gürcan. 2013. The political economy of ‘food security’ and trade: uneven and combined dependency. Rural Sociology, 78(3), 263–89. doi:10.1111/ruso.12011. Padilla Miranda, L. 2017. Calidad de vida de los jornaleros del agave en el municipio de Tequila, Jalisco. Master’s thesis in Social Territorial Studies, University of Guadalajara, Los Valles University Centre, Jalisco, Mexico. Petras, J., and H. Veltmeyer (eds.). 2014. Extractive Imperialism in the Americas: Capitalism’s New Frontier. Leiden: Brill.
Extractive agave and tequila production 185 Rodríguez Contreras, F.E., L.M. Martínez Rivera, and C. Palomera García. 2017. Contextualización socioambiental del agave en Tonaya, Jalisco, México. Región y Sociedad, 29(70), 71–102. http://dx.doi.org/10.22198/rys.2017.70.a406. Servicio de Información Agroalimentaria y Pesquera. 2018. Atlas Agroalimentario 2012– 2018. Mexico: SAGARPA. Servicio de Información Agroalimentaria y Pesquera. 2019. Cierre de la producción agrícola. Available at: https://nube.siap.gob.mx/cierreagricola/ (Accessed 11 January 2020). Svampa, M. 2018. Las fronteras del neoextractivismo en Ámerica Latina: Conflictos socioambientales, giro ecoterritorial y nuevas dependencias. Guadalajara: Universidad de Guadalajara. Tello, C. 1996. La economía mexicana: hacia el tercer milenio. Nexos, 223, 47–55. Torres, I. 2018. El manejo forestal y el monocultivo intensivo de magueyes mezcaleros: beneficios y perjuicios de prácticas contrastantes. In C. Illsley, I. Torres, J. Hernández, P. Morales, R. Varela, I. Ibáñez, and H. Nava (eds.), Manual de manejo campesino de magueyes mezcaleros forestales. Mexico City: Grupo de Estudios Ambientales. Veltmeyer, H., and J. Petras (eds.). 2014. The New Extractivism: A Post-Neoliberal Development Model or Imperialism of the Twenty-First Century? London: Zed Books. Zizumbo Villarreal, D., M.P. Colunga García, O. Vargas Ponce, J.J. Rosales Adame, and R.C. Nieto-Olivares. 2009. Tecnología agrícola tradicional en la producción de vino mezcal (mezcal y tequila) en el sur de Jalisco, México. Revista de Geografía Agrícola, 42, 65–82.
9
Forestry extractivism in Uruguay Markus Kröger and Maria Ehrnström-Fuentes
Introduction In this chapter, we assess how prominent definitions of (agro)extractivism are suited to explain forestry extractivism, and the shared and particular qualities of forestry extractivism as it manifests itself in large-scale tree monocultures for pulp production in Uruguay. The definitions of (agro)extractivism by McKay (2017), Alonso-Fradejas (2018), Gudynas (2015), Svampa (2012), and others are assessed. We identify forestry extractivism, including industrial pulpwood plantations and cellulose pulp production facilities, as one distinct form of agroextractivism, with some notable differences in comparison to the production of soybean, corn, and other crops via monoculture plantations. One key difference is the prevalence of much higher capital investments in the pulping part of forestry extractivism, wherein pulp investments costing several billion euros are the key expansion mechanisms of pulpwood plantations (Kröger 2013a, 2013b). These pulpwood plantations of eucalyptus, pine, and acacia are the most prominent feature of forestry extractivism, in terms of hectares occupied, and the longterm planning needed to establish and root them (Kröger 2014). Other forms of forestry extractivism are tree plantations serving primarily as sources of carbon substitution, for example, as charcoal material to supplant coke in the iron oresteel making complexes, or as material for producing pellets, wood chips, or other energy and electricity production raw material (Kröger 2016). However, it should be noted that in the same way as all natural resource extraction, agriculture or mining is not extractivist in character; the label “extractivist” should also not be used for all forestry techniques and forest- and tree-based economic sectors, and land-use activities and processes. This chapter will help to clarify what constitutes forestry extractivism in the context, polity, and sector of Uruguayan pulpwood tree plantation expansion. We will also make some general comments on extractivism and particularly agro-extractivism, enlightening how forestry extractivism helps to nuance and pinpoint key features that conceptualizing extractivism more broadly should take into consideration. We will not provide herein systematic comparison of different agro-extractivist sectors, but do offer some notes and methodological comments on how such studies should be carried out. This discussion is based primarily on
Forestry extractivism in Uruguay 187 our own field research and case studies on forestry investment expansion sites around the globe, especially in Uruguay and Chile (Ehrnström-Fuentes 2015, 2016, 2019, 2020; Ehrnström-Fuentes and Kröger 2017, 2018), Brazil (Kröger 2013a, 2013b, 2013c; Marini et al. 2017), Southeast Asia, Africa and elsewhere (Overbeek et al. 2012; Kröger 2014, 2016), and Finland (Kröger and Raitio 2017), as well as our studies in other extractivist sectors, such as mining and agriculture, in different polities of the planet. We focus on the case of Uruguay, as the largest landowners of Uruguay are multinational paper and pulp companies with headquarters in and with other deep ties to Finland. Currently, about 1 million hectares of the country’s 17.1 million hectares are covered with pine and eucalyptus plantations that are mainly in the hands of Finnish companies (Uruguay XXI 2016). The corporations UPM and Stora Enso are the main owners of these plantations destined to pulpwood production in Uruguay. Stora Enso holds a joint venture with the Chilean CMPC corporation, called Montes del Plata at the estuary of the Plata river, while UPM has one mill in Fray Bentos, alongside the Uruguay river. UPM is currently building another pulp mill, which, if finished, will be the world’s biggest, in the middle of the country in the province of Tacuarembó. Both of the UPM projects have been highly contentious, the first one, called Botnia (and first owned primarily by another Finnish company, Metsä Group, owned mostly by a Finnish forest owners’ cooperative), creating a major international conflict between Argentina and Uruguay (Kröger 2007; Pakkasvirta 2008), and the second creating major internal opposition and protests, for several reasons (Bacchetta et al. 2019). The particularity of Uruguay is that the pulp mills are located in tax-free and free-trade zones: they are not part of the tax regime of Uruguay in the same manner as, for example, soybean plantations in Uruguay or similar pulp investments are in Brazil. Therefore, what Uruguay exports are eucalyptus trunks, to be processed and shipped away by the corporations as pulp, a commodity used in paper, cardboard, and tissue production around the world, especially in China.
Forestry extractivism: a particular kind of agro-extractivism Gudynas One of the most well-known and applied short definitions of extractivism is offered by Eduardo Gudynas. This definition has been created within the discussions around what is industry and what is merely raw material extraction without notable socio-economic benefits, which is resisted. This definition is applied especially in Latin America, due to the five centuries long history of extractivism in different forms, where raw materials were exported from the continent for the benefit of colonial, imperial, or neoliberal hegemonic or dominant global powers. Gudynas (2018, 26) defines that “we are faced with an extractivism when three simultaneous characteristics occur: an extraction of natural resources in large volumes or high intensity, where half or more are exported to global markets, and are as raw materials or commodities”.
188 Markus Kröger and Maria Ehrnström-Fuentes These characterizations clearly apply to Uruguayan eucalyptus plantations and pulp production. The extracted volume is large and the intensity high, and practically all the production is exported, first as tree trunks from Uruguay to the freetrade zones where the pulp mills are, and from there as pulp to the transnational space of global markets. In the last two pages of the book Extractivismos (2015), Gudynas further emphasizes that particular valuations and ethics underlie extractivisms. He argues that it is essential to analyse the alternative valuations that people resisting extractivism have, so as to understand what extractivism is. We will come to this point of Gudynas in defining extractivism later in this chapter. The earlier definition of Gudynas (2018) is simple and clear, and easily applicable and understandable. It serves to not over-extend the concept, or use it too vaguely. However, this conceptualization is problematic if a production practice would otherwise be destructive, large-scale, and socio-ecologically problematic, but not destined to export markets. In this sense, the definition provided earlier may be too nationalistically focused, given that there are many internal colonial frontiers inside countries, and not just between them. Leaving these cases out from the analysis of extractivism would create an unnecessary gap on how extractivism is understood and examined in both practice and theory. For example, much of the forestry practices around the globe would, according to Gudynas’ definition, not be considered as extractivist, as in many places, especially in the global North, pulp is further processed into paper and other pulp-based higher value products, and maybe even consumed, within that same country. Yet, we acknowledge that the previous definition holds analytical merits due to its clarity and the possibility it gives to pinpoint how the new wave of export-oriented mega-pulp mills targeting both Global North and South suggests a deepening of extractivist stances globally and how Finland and Uruguay differ in terms of added value to the national economy. The global pulping boom suggests that forestry extractive operations are becoming larger in scale, leading to more severe environmental degradation, and contributing relatively more to the national foreign trade balance (and thus being more relevant for national macroeconomic policies). We will next move to more specific definitions of agrarian extractivism from the general definition of Gudynas. McKay McKay et al. (2021, intro, this book) call for a more rigorous definition of agrarian extractivism. Two exemptions are definitions given by McKay (2017) and Alonso-Fradejas (2018). We will next present these definitions and then analyse how they explain forestry extractivism. McKay (2017) names agrarian extractivism as defined by four interlinked features: 1 2
large volumes of materials extracted destined for export with little or no processing; value-chain concentration and sectoral disarticulation;
Forestry extractivism in Uruguay 189 3 4
high intensity of environmental degradation; and deterioration of labour opportunities and labour conditions in the area/sector.
Pulp production in Uruguay fulfils all these four categories. Uruguay exports eucalyptus trunks to the free-trade zones, which then produce pulp (and pollution and carbon emissions). The value chain is highly concentrated in the hands of two to three foreign corporations only, which are disarticulated from the rest of the economy (Ehrnström-Fuentes and Kröger 2018). Pulp investments in the global South are typically enclave investments, which do not support the creation of local industries or development, but buy the required capital goods for the building and maintenance of the mills from the core localities within the system of global pulping, which Kröger (2014) calls forestry imperialism. In this system, Finland, Sweden, and Austria are the key countries which have the corporations producing the pulping machinery: the Finnish Valmet corporation has produced about 80% of the pulping capacity of the world, while Austria’s Andritz, with production lines in Finland and Sweden, is another key pulping machinery producer. The Finnish-Swedish Poyry Consulting is by far the leading project planning engineering corporation. Also, the chemical producers are located in these countries. United States and Canada also have some technology producers, but most of the value chain and key functions have been very focused on a select group of countries and corporations that have headquarters in tied relationships to national production networks in these countries (for pulping, especially in Finland, Sweden, and Austria). This is thus a niche form of creating value, of particular relevance in the case of how forestry extractivism is organized through different multinational corporations with their head offices and main shareholders group up the value chain located in the global North. Eucalyptus plantations cause major environmental degradation, due to the application of agro-toxics, and acidification, salinization, and erosion of soils, as well as by decreases in stream flow and availability of clean water (Jackson et al. 2005; Kröger 2014). Water scarcity is another direct environmental consequence of eucalyptus plantations, leading to the need to supply rural communities caught in the middle of the plantations with externally sourced water from tank trucks in Uruguay, for example (Ehrnström-Fuentes 2019). Eucalyptus plantations covering hundreds of thousands of hectares occupy disproportionately large sections of municipal territories in a limited territory, making it hard or impossible for other forms of land uses to co-exist. This situation forces a semi-voluntary exodus of farmers, as well as other rural dwellers, who no longer can find jobs at local farms (Ehrnström-Fuentes 2019). Tree plantation jobs in Uruguay are seasonal, lowpaid, requiring extensive travelling, and based on precarious working conditions (Ehrnström-Fuentes and Kröger 2018): silvicultural workers are the most impoverished and have poorest health and economic indicators of all worker categories in Uruguay (Cardeillac Gulla et al. 2015). Corporations and politicians claim new pulp investments would be required to generate jobs, but figures show that the number of jobs for wood extraction for pulping has decreased between 2007 and 2016 by 992, totalling 1,669 in 2016. At the same time, however, 1,100 new jobs
190
Markus Kröger and Maria Ehrnström-Fuentes
were created for pulp processing, and 1,050 for other pulp-related forestry activities (Bacchetta et al. 2019, 36). We will make further use of the four features stated previously in providing a more detailed analysis of the existing pulp investments and projects in Uruguay after this section on applying definitions of agrarian extractivism to forestry in Uruguay. Alonso-Fradejas Other key definitions of extractivism help to further elucidate forestry extractivism in Uruguay. We will next move to assess the definition given by AlonsoFradejas (2018), wherein agro-extractivism is seen as having three key features: 1 2 3
the extraction and appropriation of the surplus value, rents, and state revenues, including by means of financialization; the appropriation of productive and reproductive labour; and the contamination and exhaustion of external nature’s energy and materials as well as damaging workers’ health and vitality.
Forestry extractivism in Uruguay also fulfils all these three categories. Especially the point about appropriating state revenues is important here, as the Uruguayan state is in an extremely disadvantageous contractual relation with the pulp-producing corporations (Bacchetta et al. 2019). The plantations and pulp-producing operations generate very low or no tax revenues, and are indirectly highly subsidized (e.g. through state-funded infrastructures for the logistical services of the operations). The appropriation of labour refers to significantly reconfiguring the kind of state-funded education and professional jobs that are offered in the extractivist context (Balch 2018). In Uruguay, the state has focused on periodically rolling out a new pulp investment after the first one, so that the construction workers who have built one mill are not unemployed once the project is finished but can find a future job in the next project. This is a kind of path dependency, technological lock-in, and an agro-industrial enclave without economically important interindustry synergy benefits (Kröger 2007; Balch 2018), where the pulp companies ensure that the state supports their further expansion, by ensuring thousands of job opportunities during the construction phase. However, after the construction phase is over the state needs to find new jobs for them. In fact, Balch (2018) has pointed out how the corporations adjust job creation figures to their own agenda and through carefully designed corporate social responsibility (CSR) activities, employ tactics to detach themselves from fulfilling the promise of employing large numbers of people, while still paying attention to portraying an image of themselves as responsible. Also, new professional education courses have been offered, and educational sectors and states in Finland and Uruguay have signed several contracts to repurpose the Uruguayan educational sector to make more workers available for the forestry sector (Fermi 2019). This has, of course, many opportunity costs and
Forestry extractivism in Uruguay 191 appropriates the free education offered by the state for the purposes of extractivist operations, and their smooth running. The appropriation of labour production has myriad and deep political impacts. Bigger trade unions have thus supported continued pulp investments, and put pressure, especially on the linked progressive parties that have composed the Uruguayan government during the years when pulp investment decisions were made. The 2020 elections brought in a conservative government for the first time in 15 years, but this has not yet changed the state–pulp business relations, as the government’s support for the expansion of the forestry sector has continued. Point (3) is a bit divided: in the most clearly extractivist parts of forestry operations, the work is dangerous to health and the environment due to the application of pesticides and other features. It is difficult and problematic to find or present “hard facts” numbers on these issues, due to the political means of underreporting and overreporting of key numerical data by their institutional producers, and the hiding of data. For example, the impacts on water pollution by the Plata River pulp mills are not revealed to the public, but confidential, although observed by a joint commission of Argentina and Uruguay, established after their major international conflict around the Botnia pulp mill (Pakkasvirta 2008). Svampa Maristella Svampa’s (2012) definition of extractivism includes additional considerations. Besides defining extractivism as large-scale and oriented towards export markets and as an activity where the environmental impacts are felt by local populations, Svampa makes explicit links to “intensive occupation of territories, land grabbing, concentration of landownership” and the large transnational corporations involved in the politics of extractivisms. Svampa’s definition of extractivism as a form of territorial occupation indicates and emphasizes that land is not “empty” prior to the arrival of extractive operations but that extensive areas of land are occupied for the use of extraction. This focus on “territory” opens up space for an understanding of the ontological politics involved in struggles over land affected or threatened by extractivism (Ehrnström-Fuentes 2019). In the Latin American social movement literature, the term “territory” does not just refer to the physical boundaries, jurisdiction, or ownership structures but also includes the symbolic meanings, practices, and relations to non-human agents of that particular place (Ehrnström-Fuentes 2020; Kröger and Lalander 2016). Thus, territorial occupations are not just about who owns or controls the land through ownership or law enforcement, but extractive occupations of territory also change the very ontology of the place by altering the symbolic meanings, practices, and human–other-than-human relations (Ehrnström-Fuentes 2019). Svampa’s definition emphasizes how states prioritize large transnational corporations over local populations and the weakening effects this has on democracy (see also McKay 2017). Yet, the transnational corporations are not the only agents involved in the politics over forestry extractivism, neither are they passive receivers of the benefits granted by the states. Transnational forestry corporations
192
Markus Kröger and Maria Ehrnström-Fuentes
together with a myriad of different supportive non-state actors (e.g. forestry development consultants, suppliers, financial agents, corporate-benevolent media, and NGOs) are involved in shaping both the institutional environment in which they operate (Pakkasvirta 2008),1 and the very ontology, or reality, that enables their own existence (Böhm and Brei 2008; Ehrnström-Fuentes 2019). Thus, through carefully designed social development projects and marketing campaigns labelled as CSR, forestry corporations actively engage in local politics seeking to influence the public opinion and their legitimacy among key stakeholders (Balch 2018; Ehrnström-Fuentes 2019). We discuss the implications of this ontological approach for examining forestry extractivism in Uruguay at the end of this chapter.
Forestry extractivism: particularities The expansion of tree plantations and pulp mills, especially in Uruguay, reveal several factors which are important in highlighting the extractive character of these investments, but which are not yet captured fully or in a nuanced enough way in the previous definitions of (agro)extractivism. The discussion of these in the following helps broaden the understanding of how forestry can also take extractivist forms, and the differences that extractivisms can take. On a broad level, forestry extractivism converges with other forms of (agro)extractivisms, but it is good to provide analysis on the differences between these extractivisms as well, while understanding that the broader concept of extractivism captures well the much broader similarities between these, under which there are nuanced differences. However, there is still a need for systematic in-depth studies on different forms of extractivisms, across different sectors. Such studies should partake from first studying the various sectors within the same polity and context (e.g. soybeans and pulp in Uruguay), and then proceed with comparisons to settings which share the expansion of these same sectors (e.g. to soybeans and pulp in Brazil). One should avoid comparing a different sector in a different context with another polity and sector, for the many methodological issues this creates, particularly the impossibility of controlling the key changes in explanatory factors: thus making comparisons between, for example, Central American oil palm extractivism and Uruguayan pulpwood expansion are methodologically flawed and not recommended (Kröger 2020a). When approaching the phenomenon of forestry extractivism, which should also be done by bounding the case to a specific context as in here, first it is important to note that forestry extractivism here (and in most other contexts) includes two separate but interconnected operations: tree plantations and cellulose pulp production. These sectors have distinct impacts on the local population and the environment, and they create different kinds of political economic dynamics and contentious politics. Yet, in studies examining the socio-economic–environmental effects of the whole sector, the analysis must include both operations and their interconnected dynamics. The following list opens up key mechanisms of how these two interconnected sectors of forestry
Forestry extractivism in Uruguay 193 extractivism is expanded, several of which show how they differ from other forms of extractivism. Specific laws and trade deals in a bound international order The first key difference to note by the example of Uruguayan pulp investments is the role of bilateral state–corporate cooperation in promoting, supporting, and protecting the birth of the sector long before the actual decisions are taken on the investments. In Uruguay, the state has paved the way for the birth of the forestry sector by: (i) creating laws that enabled the spread of large industrial tree plantations; (ii) signing bilateral trade agreements with the home country of the investing corporations (Finland); and (iii) establishing free-trade zones in which pulp mills can be set up and operate without fiduciary responsibilities (EhrnströmFuentes and Kröger 2018). These state-level supportive mechanisms are executed prior to investments, protecting the corporations from allegations of unlawful conduct as well as from potential financial losses due to political changes in the operating environment of the host country (Ehrnström-Fuentes and Kröger 2018). Conflicts are common to investments in forestry (Ehrnström-Fuentes 2015; Gritten and Mola-Yudego 2010; Joutsenvirta and Vaara 2008; Leys and Vanclay 2010; Myllylä and Takala 2011; Kröger 2013c), and for the corporations, there is a need to safeguard themselves from the social and political risk associated with these conflicts. Uruguayan pulp investing highlights a much more bilateral and longer term international binding of production roles between Finland and Uruguay than, for example, soybean or palm oil expansion in Latin America, whose ownership and international and national power structures are much more spread and less bound and tight than in this exemplary and globally important case of forestry extractivism. Long-term setting-up through stages While agro-extractivist ventures can often be established in a matter of years or even in a year, it takes more than a decade, even several decades, to establish the required material basis for modern pulp mills, even when using fast-growth species such as eucalyptus. Forestry extractivism is thus introduced in several stages in comparison with other extractive modes: 1
Strategic phase. The making of land available and suitable for large-scale monoculture tree plantations undergoes a carefully designed strategic plan during which suitable areas for tree plantations are identified and mapped, regional land-use plans are changed, legal arrangements are implemented, and economic incentives (subsidies) for tree plantations are set up. Consultancy firms act as the architects of these strategies, creating a coherent narrative about the benefits of forestry, which are needed to attract investor interest and to garner support among local lawmakers that shape the institutional landscape in favour of forestry investments (Carrere and Lohmann 1996).
194 Markus Kröger and Maria Ehrnström-Fuentes
2
3
This normally takes several decades, while, for example, soybean plantations have been expanded in South America through a much faster strategic planning phase. Tree planting phase. During this stage, rural communities undergo an ontological transformation in how community members engage and relate to the land. Farming communities that used to sustain themselves through food production practices are transformed into tree plantation communities sustained through industrial low-waged labour moving in and out of the community (Cardeillac Gulla et al. 2015). The changes in landownership from settled smallholder farms to distant multinational corporations impact how and by whom land is managed, how the ecology of the land is practised, and territories are constituted. Thus, as communities, ecologies and their associated human and other-than-human actors are increasingly linked to forestry operations; this also influences the political preferences for the further expansion of forestry and the arrival of (contested) pulp mills. Land ownership turns markedly concentrated, much more than in soybean or other extractivist sectors, where regional and national elites and even middle-sized farmers have remained much more prominent than in the corporate-controlled forestry extractivism (Kröger 2020b). Pulp mill planning and construction phase. During this phase, the desirability of the whole forestry sector is likely to be publicly debated. In contrast to other agro-extractivist sectors, forestry destined to pulp production, due to the long planning and building phase, may offer more possibilities for public deliberation over the desirability and impacts of the sector on the local community and the environment. However, although broad similarities are shared between pulp investment dynamics across different contexts, this deliberative space is not offered automatically across the board, but depends on polity differences, and especially on whether contentious agency is being built, i.e. whether there is the will and capacity to question pulp investments by the civil society and progressive state actors (Kröger 2013a). As planning for pulp-based economies takes a long time, specific politics are ushered in. For example, in contrast to the environmental impact assessments (EIAs) executed in the highly contentious mining sectors, the forestry practices and work relations are already settled in the country and in many local communities at this point (see point 2). This creates more political interest skewed towards adding more value to the existing operations, going from logging to pulp production. Thus, in contrast to the mining sector (see Kröger 2020a), the forestry sector is well established with linkages to the local economy long before the desirability of the sector is publicly debated during the evaluations of the EIA. In comparison to agro-extractivism in the form of agricultural crops, such as the soybeans, corn, and cotton planted on commoditizing resource frontiers in Brazil’s Mato Grosso state, for example (Kröger 2020b), there seems to be however much more contestation of mega-investment plans and in particular of the construction of pulp mills than for resisting the expansion of the crop-based agro-extractivisms that are not so dependent on
Forestry extractivism in Uruguay 195
4
high-cost and tightly controlled capital goods such as pulping lines and specific forestry techniques. Post mill-construction phase. Conflicts are most likely to emerge only after the mills have been constructed, and the dire reality of unrealized promises is revealed to the public (Kröger 2010). This opens up a new era of contentious politics or extensive usage of CSR policies aimed at curbing mobilizations, or both. The execution of future projects becomes increasingly difficult within the same region, as visible in Uruguay.
Forestry enclosures The amount of land occupied by plantations is locally and regionally vast, as in other agro-extractivist models, and, in the case of Uruguay as well as most pulp-producing regions in the Global South, in large part centralized to the forestry corporations, who through direct ownership have more control of the whole chain of custody. High levels of land concentration to single corporations controlling hundreds of thousands of hectares in a 100–200 kilometre radius is a commonality of pulping in the Global South, distinguishing forestry extractivism as a particularly corporate-controlled form of extractivism in the array of (agro) extractivisms. In Uruguay, the occupation of vast areas of land is strikingly clear: according to UPM, its owned and rented tree plantations cover 385,000 hectares, while Montes del Plata is estimated to control between 200,000 and 300,000 hectares, while the total extension of eucalyptus and pine plantations is around 1 million hectares (Bacchetta et al. 2019, 201). Another 4.1 million hectares of the country’s 17.1 million hectares have been declared as a forestry priority (Uruguay XXI 2016). What this means is that land areas previously inhabited by other non-forestry-related practices (e.g. cattle grazing and agricultural practices destined to food production) are deemed less important. Through governmental incentives and land-use policies, the symbolic meanings of land-related practices are increasingly tied to forestry operations as a national priority. In areas destined to forestry operations, alternative non-forestry-related land practices are not able to resist these kinds of territorial occupations (Ehrnström-Fuentes 2019; O’Neill 2015). Criticism of land ownership concentration and foreignization has been raised both by academics (Piñeiro 2012) and mainstream media (Toivonen 2016). In recent years, forestry corporations have started to offer cooperation programmes to local landowners. They explain that by planting eucalyptus trees on their land, the landowners will be able to “diversify” their traditional production (usually cattle and agriculture) through sustainable eucalyptus production using UPM’s high-quality seedlings (Elhordoy 2019). Through these programmes, local landowners plant and manage eucalyptus plantations destined to pulp mills on their land (UPM Forestal Oriental 2018). These types of diversification of farm productivity into tree plantations are also enabled by financial assistance provided by the Inter-American Development Bank in close cooperation with the Uruguayan Agricultural Ministry.
196 Markus Kröger and Maria Ehrnström-Fuentes The forestry enclosures do not only include tree plantations, but all associated infrastructure that encloses rural territories in forestry-related practices. For example, investments in roads and railways that serve the forestry sector are enabled to use public–private partnerships between the state and international development funds (Elhordoy 2019). In Uruguay, pulp mills, also financed by international development funds (Elhordoy 2019), are constructed on tax-free zones that represent another type of enclosure in which only practices related to the production of pulp destined to export markets are allowed to exist. Sophisticated legitimization campaigns The forestry sector presents itself as a central player in the new “green” bioeconomy that uses biomass, or renewable materials, “to create sustainable fossilfree alternatives in a variety of end uses” (UPM 2020). Thus, in order to appear sustainable, the forestry sector has heavily invested in various forms of CSR and played an active part in developing international certified standards (e.g. FSC, PEFC) and networks that promote sustainable forest management on global scales (e.g. World Business Council for Sustainable Development, The Forests Dialogue initiative, and the WWF’s New Generation Plantations platform) (Stora Enso 2020). Certifications also play vital roles in the creation of a sectorial culture and philosophy tied to forestry in different contexts and at multiple levels (Carrere and Lohmann 1996; Pakkasvirta 2008). The narratives used to frame forestry as socially acceptable through media further strengthen the symbolic meanings of sustainability and ethicality as defined within the sector itself (Ehrnström-Fuentes and Kröger 2017; Ehrnström-Fuentes 2015), crowding out alternative ways of framing sustainability based on other land-based practices that still exist in places endangered by forestry expansion (Ehrnström-Fuentes 2019). In comparison to, for example, oil palm or soybean agro-extractivisms, the terms of debate in forestry are not so much about the need to avoid deforestation by extractivist expansion, as the forestry sector is much more likely to be associating itself (and being associated) as an agent of “reforestation”, or expander of “planted forests”, or even simply “forests”, although in practice the sector does cause direct and indirect deforestation, and destruction of native ecosystems such as grasslands (Kröger 2014). In fact, pulp and paper companies have, in the past, repeatedly been exposed to reputationally damaging consumers and publishing house boycotts for their involvement in clear-cutting old growth and native forests (Halme 2002; Joutsenvirta 2006; Stine 2011). Thus, as a consequence, the sector as a whole, not just in Uruguay, has over a long period of time included social responsibility in their strategic management decisions and marketing campaigns to mould “public opinion in favor of the pulp and paper sector, hiding the negative environmental, economic and social impacts of this type of development” (Böhm and Brei 2008, 358). How forestry corporations use CSR projects to legitimize the presence of pulp mills in local communities in Uruguay has been documented in detail by
Forestry extractivism in Uruguay 197 Balch (2018). Also, Ehrnström-Fuentes and Kröger’s (2017) study of oppositional voices in Uruguay and Chile show how the corporate activities designed to create a “social license to operate” (SLO) are used to silence grievances and legitimize plantations and pulp mills to key influential stakeholders elsewhere (investors and end customers). This sophisticated corporate counter-tactics have led to major revamping of the local political setting, remoulding the political landscape in its favour, destabilizing counter-mobilizations that expose the adverse social and environmental impact of forestry operations (Ehrnström-Fuentes and Kröger 2017, 2018). Similar kinds of legitimization campaigns do also exist in other extractivist sectors, such as mining and agriculture (e.g. round-table discussions on sustainable palm oil and soybean that would supposedly not be produced in deforested areas). In the case of forestry extractivism, these campaigns have started to focus more and more on framings that seek to legitimize and coin these corporations as forerunners in the bioeconomy, as key bio- and green-economy parcels of the struggle against fossil-fuel-based economies, which would further help in mitigating the climate change (Ehrnström-Fuentes 2019). This is a distinct struggle and dynamics for legitimization campaigns than the campaigns by other forms of extractivism, such as mining or animal feed production, whose negative climate impacts (e.g. deforestation) are more self-evident, and emphasized internationally. Despite the investments in CSR and other stakeholder relations, conflicts with environmental movements and local populations, whether visible or more covert, are still commonplace to the instalment and running operations of pulp mills, and the expansion and sustaining tree plantations in Uruguay and elsewhere. These conflicts seem to be more visible internationally and more present than conflicts around soybean plantation expansion in Uruguay or elsewhere. For example, a Google search on 17 July 2020 on “Uruguay pulp conflict” versus “Uruguay soybean conflict” results in 1.5 m hits versus a bit less than 0.9 m hits. “Pulp conflict” in general produces 17.8 m hits, while “soybean conflict” produces 7.55 m hits. “Palm oil conflict” produces however 29 m hits, and “mining conflict” 137 m hits, which shows the much higher visibility of these two sectors than pulping, globally. We will next move to a more detailed analysis of the political ecology and political ontology of forestry extractivism, which are key issues, or maybe the key issues, to consider when analysing and defining extractivism.
The political ecology of forestry extractivism The definitions of agro-extractivism, as well as the definition of extractivism by the various authors discussed earlier, offer tools for pinpointing, especially the political, economic, and social aspects of extractivisms. These analyses of the politics involved in extractivism also offer possibilities for deepening the analysis of political ecology (studying the effects/politics of extractivism on the ecology, and vice versa, within the same ontology) (see e.g. Alonso-Fradejas 2018) or a political ontology (studying the effects/politics of extractivism on and with
198
Markus Kröger and Maria Ehrnström-Fuentes
alternative ontologies or alternative ways of being) approach (see e.g. various works by Gudynas and Svampa). A political ecology approach for examining extractivisms emphasizes the socio-ecologically destructive characteristic of extractive operations and to use these as the basis of inquiry. Most existing political-economic analyses of extractivism give some importance to ecological matters, from the viewpoint of how the destructive aspect of extractivism is its key defining feature, affecting the political and accumulative dynamics in crucial ways. Ye et al. (2020) list ten key features of extractivism, one of these being the creation of barrenness, places that are depleted, destroyed, or polluted by extractive operations. Dunlap and Jakobsen (2019) emphasize the destructive and violent characteristic of extraction, and so does Jason W. Moore (2015) in his work on the Capitalocene and the web of life. For Moore, capitalism is a frontier, a destructive project in its core: it needs to extract, blunder, and appropriate new value by expanding resource-extracting frontiers, which cause destruction within the web of life. Alonso-Fradejas (2018) weaves together a more detailed political ecological analysis of soil, hydrology impacts of sugar cane, and palm oil production with a Marxist conceptual analysis of this metabolic rift. Research on extractivism should engage with the relevant ecological impact studies of the type of extractive operation in question, uniting this to political economic analyses. We will next assess monoculture tree plantations more in detail in this vein. The key issue to consider in the political ecology of forestry extractivism is that monocultures of eucalyptus used in pulp production acidify and salinize soils and decrease stream flows (Jackson et al. 2005). They also deplete and decrease the availability of water, especially drinkable water, due to the use of pesticides in the plantations, and diminish groundwater tables due to their fast-growth cycle of eucalyptus clonal stands, and the breeding of eucalyptus variants that require much water to grow faster (Kröger 2014; Overbeek et al. 2012). This leaves behind landscapes and soils that are eroded (Jobbágy and Jackson 2003). The existing ecosystem services of the Uruguayan grasslands are depleted, and the resiliency of the systems is lost (Céspedes-Payret et al. 2009). This also explains how cheap pulp can be produced in these mega-mills. This is premised on a cheapening of nature, a process that is based on appropriation of nature’s unpaid work, in terms of Moore. This is how commodity frontiers have destroyed and depleted lived environments since the 15th century, as Moore (2015) describes, moving to new areas. In the case of South American pulp investment expansion, these have spread from the coasts and accessible river sites inland (Marini et al. 2017), as visible also in Uruguay. Another key type of extractivist appropriation and destruction of existing accumulated capital, in nature or ecosystems, is the loss of organic soil carbon when natural pampa grasslands previously used for pasture are converted into tree plantations. Studies in Uruguay on 20-year-old eucalyptus stands established on grasslands suggest that even though eucalyptus growth does capture carbon from the atmosphere into the tree trunks, beneath the surface the existing soil
Forestry extractivism in Uruguay 199 carbon rapidly dissolves: about 80% of the positive impact of tree planting in terms of carbon capture above the ground is cancelled due to the negative impact on carbon balances underground (Carrasco-Letelier et al. 2004). Similar impacts in terms of carbon emissions and balances of soils are also present in other forms of agro-extractivism, and elsewhere such as in sugar cane and oil palm plantations in Central America (Alonso-Fradejas 2018, 346). The focus of most analyses lies on what happens above the ground, to justify the investments, and to capture the rents from booming carbon markets and credits, for their supposedly beneficial effects on carbon sequestration and, thus, climate change mitigation (Overbeek et al. 2012; Kröger 2016). These carbon calculations require much more detailed and critical scrutiny. Moreover, pulp mills themselves produce about double the amount of carbon emissions as they produce pulp. Thus, a 1-million-ton-per-year-producing pulp mill produces over 2 million tons of carbon emissions (https://suomenluonto.fi/ uutiset/vahtikoira-miljardin-kuution-hiilivelka/). Furthermore, most of the produced pulp quickly returns to the atmosphere, when converted into cardboard and tissue products, which are mostly not recycled (Carrere and Lohmann 1996; Dauvergne and Lister 2011). Even when recycled, the processing of new materials contributes to more emissions (Overbeek et al. 2012). The green image of forestry corporations thus needs to be seriously challenged. The overall carbon impact is clearly negative, and these pulp investments are not a solution to curbing climate change, quite the contrary. Mega-pulp mills are unnecessary and unsustainable investments. In terms of the extractivist theoretical apparatus, pulping does just that what its name suggests: pulping landscapes, existing ecosystems, and carbon balances, appropriating existing organic matter in the soil, converted into pulp, for profit and gain which concentrates to particular corporate elites (Kröger 2013b). What is left behind in many places are eroded and polluted soils, uninhabitable, and water scarce rural landscapes (Carrere and Lohmann 1996), which are locked in into ever-further forestry extractivism, as many soils, especially in low-precipitation countries such as Uruguay, do not serve anymore for other purposes, or would require extensive investments in irrigation. This political ecological analysis leads us to a key observation. The specificity of forestry extractivism, at least in Uruguay, seems to be a deep capture and control of large swaths of land for several decades for the forestry corporations’ purposes. This is assured by rights to pollute, which protect the investments through the possibility of charging the Uruguayan government in the international arbitration court for investment disputes, if the people decide to democratically increase regulations, for example. The continuity is ensured by a technological lock-in, where the forestry methods used, described earlier, make it possible to introduce ever-more extractive eucalyptus stands, increase the amount of fertilizers and pesticides, and thus deplete soils for other land-use practices. The salinization, acidification, emissions from loss of soil carbon, and decrease of water levels are particularly worrying aspects here.2 The destruction inherent to extractivisms suggests that extractivisms are forms of expanding unproductive capital: they are not really intensive forms of
200 Markus Kröger and Maria Ehrnström-Fuentes production, but actually, forces which destroy and deplete capital that resides in soils, in the broader conceptualization of capital which also includes soils (as Marx already suggested) (Dowbor 2018). Eucalyptus plantations in Uruguay have led to the loss of fertility of these soils (Céspedes-Payret et al. 2012). Tree plantations or pulp investments are also deeply extractive ventures, a sort of plunder of nature’s bounty. We suggest that this political ecological analysis should serve as one key guideline in defining what kinds of natural resource exploitations should be called extractivist and which not. The political ecology of forestry extractivism also suggests another important point for analyses of extractivisms. In many cases, extractivist operations of different types of work in conjunction. For example, forestry extractivism is a tool or modality for water extractivism, due to the negative impacts of polluting and consuming, appropriating water through plantation expansion (see Farley et al. 2005, 2008). Water extractivisms can take many forms, but the requirement of ample rainfall and soils which do not need to be irrigated suggests that forestry extractivism (as many other agro-extractivisms) is almost always working as a form of water extractivism. However, eucalyptus for pulp production cannot be grown in just any kind of territories, since it requires about 800–1000 mm of precipitation per year (in contrast to 400 mm for soybean cultivation). Forestry consulting companies have mapped the areas around the world where eucalyptus monocultures would be possible, and there is a limited number of these places. Furthermore, in comparison to broad-leaf forest in similar precipitation, eucalyptus plantations are more likely to reduce water availability due to their high evapotranspiration, and be more threatened by drought (Liu et al. 2017). These political ecological characteristics of forestry extractivism further underline the importance of studying not only extractivism as a general process, but to study extractivisms, in their varied forms. Lastly, we will delve deeper into analysing the political ontological dynamics, which can open up forestry extractivism.
The political ontology of resistance to forestry extractivism To understand resistance to forestry and the politics of place that local groups engage in as they defend their life forms against forestry extractivism, it is not enough to examine these conflicts as merely environmental, or as expressions of “ecological distribution conflicts” (Scheidel et al. 2018). How “environment” and “ecology” are understood and practised by local actors depends largely on the practices that inform their way of being and relating to place (Blaser 2013b; de la Cadena 2015), which in turn informs how they respond to the arrival of extractivism in their community (Ehrnström-Fuentes 2016, 2020). The field of political ontology allows for a nuanced examination of the impacts, conflicts, and dynamics involved when local groups confront and resist the arrival of extractivism on their land (Ehrnström-Fuentes 2019, 2020). Drawing on Indigenous struggles against states’ efforts to colonize and develop Indigenous territories, Political Ontology (with capital letters) drawing
Forestry extractivism in Uruguay 201 particularly from Amerindian knowledge postulates that the dominant position that the modern myth/ontology holds vis-à-vis other ontologies singularizes the multiplicity of place-based worlds into one presumed universally applicable reality (Blaser 2010, 2013a; de la Cadena 2015; de la Cadena and Blaser 2018). In essence, in encounters with modernity (e.g. participatory democracy, national politics), Indigenous ways of being, knowing, and relating (to) the world are not considered relevant, legitimate, or even acknowledged to exist (Blaser 2010; de la Cadena 2015). Ontological openings which question the modern forestry apparatus, with its narrow and western-based, technical definitions of forests and trees, are currently starting to challenge global forest governance (González and Kröger 2020). Thus, when ethical assessments of what exist, what is good, and what is desirable are done based on modern ontological assumptions, this effectively excludes, or makes invisible, alternative voices from the political debates about how life should be lived in the community (Ehrnström-Fuentes 2015). This exclusion of alternative voices is particularly clear in the debate over the desirability of extractive projects, where the “necessity” to create economic growth and development to solve issues of poverty are commonly used to legitimize new projects, while ignoring the harms caused to the ways of reproducing life in the community (Gudynas 2015). The hierarchical ontological relations that exist between Indigenous communities and developmentalist states in other parts of Latin America are also present in Uruguay despite its lack of large Indigenous populations. The urban Uruguayan growth-driven techno-scientific entanglements with the forestry sector invisibilize many grievances felt by rural cattle farmers in regions assigned as a forestry priority. Due to their rural status as less productive and less capable of contributing to economic growth and development than the capital-intensive forestry assemblage, smallholder cattle farms and other oppositional voices have not been able to make their voices heard in the national media and political debates or mobilize a large unified movement against forestry investments (Ehrnström-Fuentes 2019). The conflicts that have emerged do, however, make visible non-human forces involved in politics of extractivism. As noted previously, conflicts over water are common features in the political ecology of forestry extractivism. These water conflicts are also manifestations of ontological conflicts between competing ways of performing the world through different human–water entanglements (Sepúlveda 2016). In Uruguay, the gradual decrease, and later absence of underground water is what has forced many cattle grazing farmers to sell and abandon their farms (Ehrnström-Fuentes 2019). Yet, this decrease of underground water availability has not officially been acknowledged as an issue related to forestry. Instead, drawing on scientific discourses, politicians, investing corporations, and even (some) scientists claim that changes in water precipitation, and thus availability thereof, is related to climate change rather than the arrival of monoculture tree plantations in Uruguay (Ehrnström-Fuentes 2019). This debate, revolving around the (ab)use of “science”, is firmly situated within how facts are established within modernity, and has ended up erasing the lived realities of farmers who attest that the plantations have made farming very difficult because of their experienced water
202 Markus Kröger and Maria Ehrnström-Fuentes scarcity. Thus, to understand the organization of resistance (or lack thereof) it is important to pay attention to the mobilizing force of non-human actors as extractivism threatens the existence of the multiple worlds that these actors collectively sustain (Ehrnström-Fuentes 2020, 2019). To understand how corporate agency affects these politics, it is important to note how forestry politics include elaborated CSR programmes that seek to erase conflicts and strengthen the forestry ontology among a wide array of civil society actors (Böhm and Brei 2008). This is partly due to the fact that in contrast to other agro-extractive fields in Uruguay (e.g. soy and rice), multinational corporations (and their associated business partners, as discussed earlier) are the main actors driving the expansion of the forestry sector, connecting large areas of land to extractive practices. Thus, the continuous expansion of the capitalist appropriation of nature is not an anonymous force of the “landed elite” but driven by clearly identifiable corporations and their appointed managers who, at any time, can be exposed as acting irresponsibly, as seen in the environmental campaigns previously discussed. By investing in community development projects (Balch 2018) and defining sustainability according to criteria set by the industry itself (Ehrnström-Fuentes and Kröger 2017), forestry corporations in Uruguay have used CSR to engage directly in the ontological politics of place (EhrnströmFuentes 2019). Thus, by shaping the conversations, meanings, and relations to “the things at stake” (Blaser 2013b) in places affected by their operations, corporations use CSR as a mechanism to connect humans and non-humans in ways that further strengthen the forestry assemblage (for a detailed account of how forestry corporations use CSR to disable local conflicts, see Balch 2018). These brief notes on the ontological conflicts here should be complemented by detailed political ontologies of extractivisms in different places.
Conclusions This chapter has identified forestry extractivism as a particular type of extractivism, which in some regards can be considered a specific instance or sub-category of agrarian extractivism. Uruguay was used as an example to identify how forestry extractivism functions in the case of pulp investments in that country based on extensive eucalyptus monocultures and paper pulp mills. Key features of forestry extractivism include: 1 2 3 4 5
specific trade deals, as pulp investments are costly; long-term setting-up through stages: master plans, enclosures, establishing pulp mills, and managing rising conflicts after the building; mills and plantations; ecological and carbon impacts; and massive legitimization campaigns.
We argued that political ecological, world-ecological, and political ontological analyses are important for defining what activities should be called extractivist,
Forestry extractivism in Uruguay 203 and what types of extractivisms are involved in each activity. We showed how the existing conceptualizations of (agro)extractictivism help guide research around forestry extractivism. The existing definitions were found to be helpful, and we recommend adopting them as a checklist of what aspects need to be considered in analysing forestry and other forms of extractivism. Pulping involves several forms of extractivisms, for example, based on a destructive relation with soils, water and carbon, forestry extractivism being thus definable as at least soil, water, and carbon extractivism. The carbon stored in trees is converted into paper products, and carbon is returned to the atmosphere in the production and consumption process, and depleted from soils due to the intensive production methods. If carbon storages would be increased, then that forestry practice would not be carbon extractivist. An example of this kind of process is, for example, the growth of hardwood trees in long-growth cycles in natural forests in Germany, described by Wohlleben (2016). To this list of extractivisms which are interrelated and form the possibilities of pulpwood plantation expansion, one can also add soil extractivism, as the soils are eroded. We also discussed ontological conflicts related to pulpwood expansion, which the case of Uruguayan expansion can shed light on. Political ontology is central for understanding (agro)extractivisms, especially in contexts where conflicts and grievances remain for many years, mostly in the shadows, as within Uruguay. From a political ontology perspective, forestry corporations’ ambitious legitimization campaigns are a direct response to the local conflicts around how pulpwood production and pulping affect, or threaten to affect, local ways of being in place. Thinking through the concept of (agro)extractivism puts emphasis on what is extracted ecologically. This analysis should be accompanied by a global political economic and resource geopolitics analysis of particular global extractivisms, such as forestry. This should also be tied to particular contexts, polities, and lived environments, which significantly influence especially the politics through which global extractivisms of different types are birthed and resisted. The literature on Uruguayan pulp investment was used as an example of this here. Further studies on agro-extractivism should focus on making systematic and carefully designed comparisons where both the polities and sectors compared are controlled for, comparing, for example, soybeans and eucalyptus in Brazil and Uruguay, which have both, but not any kind of resource-exploiting sector anywhere with any other kind of natural resource extraction. It is also good to separate and bound what “normal” resource extraction is, and what extractivist extraction is: several existing definitions, briefly reviewed and applied here, provide precise tools for this, which we recommend to use.
Notes 1 In the case of Uruguay’s forestry sector, the forestry consultant firms (e.g. Poyry Group) are the creators of strategies and ideologies, pulp, and paper companies (e.g. UPM, Stora Enso) as project implementers, chemistry companies (e.g. BASF, Bayer, Kemira), and machine producing companies (e.g. Valmet, Ponsse) as important suppliers of material goods, and financial investors (export guarantee agencies, international banks, and credit
204 Markus Kröger and Maria Ehrnström-Fuentes agencies) as provider of the necessary capital for these investments (Carrere and Lohmann 1996; Kröger 2007, 2010; Pakkasvirta 2008). In these networks, governments (e.g. Finland, Uruguay) enable the creation of links among the business actors and related associations (e.g. export promoting agencies, business associations, industry representatives, and worker unions) (for a more detail overview of the governmental role in “birthing forestry extractivism” see Ehrnström-Fuentes and Kröger 2018). 2 Other forms of extractivism have also very negative and specific ecological impacts, whose comparative analysis to eucalyptus monocultures is however beyond the scope of this article.
References Alonso-Fradejas, A. 2018. The Rise of Agro-Extractive Capitalism: Insights from Guatemala in the Early 21st Century. The Hague: International Institute of Social Studies. Bacchetta, Víctor L., M. Gustavo, and Y. William (eds.). 2019. La entrega: el Proyecto Uruguay-UPM. Available at: www.banktrack.org/download/la_entrega_el_proyecto_ uruguay_upm/la_entrega_el_proyecto_uruguayupm_10_2019.pdf. Balch, O. 2018. Pulp Frictions: An Ethnographic Critique of Corporate Social Responsibility and its Role in Legitimising Uruguay’s Largest Ever Foreign Investment. Dissertation thesis, University of Cambridge, Cambridge. Blaser, M. 2010. Storytelling Globalization from the Chaco and Beyond. Durham, NC: Duke University Press. Blaser, M. 2013a. Ontological conflicts and the stories of people in spite of Europe. Current Anthropology, 54, 547–67. Blaser, M. 2013b. Notes towards a political ontology of ‘environmental’ conflicts. In L. Green (ed.), Contested Ecologies: Dialogues in the South on Nature and Knowledge. Cape Town: HSRC Press, 13–27. Böhm, S., and V. Brei. 2008. Marketing the hegemony of development: of pulp fictions and green deserts, Marketing Theory, 8, 339–66. Cardeillac Gulla, J., A. Gallo, and A.J. Pérez. 2015. Permanencias en el agro uruguayo: un estudio de caso sobre el trabajo asalariado rural. Revista de Ciencias Sociales, 28(36), 77–98. Carrasco-Letelier, L., C. Eguren, C. Castiñeira, O. Parra, and D. Danario. 2004. Preliminary study of prairies forested with Eucalyptus sp. at the northwestern Uruguayan soils. Environmental Pollution, 127, 49–55. Carrere, R., and L. Lohmann. 1996. Pulping the South: Industrial Tree Plantations and the World Paper Economy. London: Zed Books. Céspedes-Payret, C., G. Piñeiro, M. Achkar, O. Gutiérrez, and D. Panario. 2009. The irruption of new agro-industrial technologies in Uruguay and their environmental impacts on soil, water supply and biodiversity: a review. International Journal of Environment and Health, 3(2), 175–97. Céspedes-Payret, C., G. Piñeiro, O. Gutiérrez, and D. Panario. 2012. Land use change in a temperate grassland soil: afforestation effects on chemical properties and their ecological and mineralogical implications. Science of the Total Environment, 438, 549–57. Dauvergne, P., and J. Lister. 2011. Timber. Cambridge: Polity Press. De la Cadena, M. 2015. Earth Beings: Ecologies of Practice Across Andean Worlds. Durham, NC and London: Duke University Press. De la Cadena, M., and B. Blaser. 2018. A World of Many Worlds. Durham, NC: Duke University Press.
Forestry extractivism in Uruguay 205 Dowbor, L. 2018. The Age of Unproductive Capital: New Architectures of Power. Newcastle upon Tyne: Cambridge Scholars Publishing. Dunlap, A., and J. Jakobsen. 2019. The Violent Technologies of Extraction: Political Ecology, Critical Agrarian Studies and the Capitalist Worldeater. Cham: Springer Nature. Ehrnström-Fuentes, M. 2015. Production of absence through media representation: a case study on legitimacy and deliberation of a pulp mill dispute in southern Chile. Geoforum, 59, 51–62. Ehrnström-Fuentes, M. 2016. Delinking legitimacies: a pluriversal perspective on political CSR. Journal of Management Studies, 53, 433–62. Ehrnström-Fuentes, M. 2019. Confronting extractivism – the role of local struggles in the (un)making of place. Critical Perspectives on International Business, ahead-of-print. https://doi.org/10.1108/cpoib-01-2018-0016. Ehrnström-Fuentes, M. 2020. Organising in defence of life: the emergence and dynamics of a territorial movement in Southern Chile. Organization [online] October 29. Ehrnström-Fuentes, M., and M. Kröger. 2017. In the shadows of social licence to operate: untold investment grievances in Latin America. Journal of Cleaner Production, 141, 346–58. Ehrnström-Fuentes, M., and M. Kröger. 2018. Birthing extractivism: the role of the state in forestry politics and development in Uruguay. Journal of Rural Studies, 57, 197–208. https://doi.org/10.1016/j.jrurstud.2017.12.022. Elhordoy, A. 2019. El aporte del BID y CAF al sector forestal en Uruguay. Revista Forestal, 7 May. Available at: www.revistaforestal.uy/industria/aporte-del-bid-caf-alsector-forestal-uruguay.html (Accessed 3 July 2020). Farley, K.A., E.G. Jobbágy, and R.B. Jackson. 2005. Effects of afforestation on water yield: a global synthesis with implications for policy. Global Change Biology, 11, 1565–76. Farley, K.A., G. Piñeiro, S.M. Palmer, E.G. Jobbágy, and R.B. Jackson. 2008. Stream acidification and base cation losses with grassland afforestation. Water Resources Research, 44(7). Fermi, M.J. 2019. Manos a la obra. Revista Forestal, 31 January. Available at: www. revistaforestal.uy/industria/manos-la-obra.html. González, N.C., and M. Kröger. 2020. The potential of Amazon indigenous agroforestry practices and ontologies for rethinking global forest governance. Forest Policy and Economics, 118. https://doi.org/10.1016/j.forpol.2020.102257. Gritten, D., and B. Mola-Yudego. 2010. Blanket strategy: a response of environmental groups to the globalising forest industry. International Journal of the Commons, 4(2), 729–57. doi:10.18352/ijc.216. Gudynas, E. 2015. Extractivismos: ecología, economía y política de un modo de entender el desarrollo y la naturaleza. Primera edición. Cochabamba, Bolivia: CEDIB, Centro de Documentación e Información Bolivia. Gudynas, E. 2018. Extractivisms : tendencies and consequences. In Reframing Latin American Development. Routledge Critical Development Studies. New York and Oxon: Routledge, 61–76. https://doi.org/10.4324/9781315170084-4. Halme, M. 2002. Corporate environmental paradigms in shift: learning during the course of action at UPM – Kymmene. Journal of Management Studies, 39, 1087–109. Jackson, R. et al. 2005. Trading water for carbon with biological carbon sequestration. Science, 310, 1944–7. doi:10.1126/science.1119282. Jobbágy, E.G., and R.B. Jackson. 2003. Patterns and mechanisms of soil acidification in the conversion of grasslands to forests. Biogeochemistry, 64, 205–29.
206 Markus Kröger and Maria Ehrnström-Fuentes Joutsenvirta, M. 2006. Ympäristökeskustelun yhteiset arvot. Doctoral dissertation, Acta Universitatis oeconomicae Helsingiensis, Helsinki, A-273. Joutsenvirta, M., and E. Vaara. 2008. Discursive (de)legitimation of a contested Finnish greenfield investment project in Latin America. Scandinavian Journal of Management, 25(1), 85–96. Kröger, M. 2007. Uruguayn ja Argentiinan sellutehdaskonflikti latinalaisamerikkalaisena maa-ja metsäkiistana. Kosmopolis, 37(2), 23–42. Kröger, M. 2010. The politics of pulp investment and the Brazilian landless movement (MST). Acta Politica, 39. Kröger, M. 2013a. Contentious Agency and Natural Resource Politics. London: Routledge. Kröger, M. 2013b. Globalization as the ‘pulping’ of landscapes: forestry capitalism’s north – south territorial accumulation. Globalizations, 10(6), 837–53. Kröger, M. 2013c. Grievances, agency and the absence of conflict: the new Suzano pulp investment in the Eastern Amazon. Forest Policy and Economics, 33, 28–35. Kröger, M. 2014. The political economy of global tree plantation expansion: a review. Journal of Peasant Studies, 41(2), 235–61. Kröger, M. 2016. The political economy of ‘flex trees’: a preliminary analysis. The Journal of Peasant Studies, 43(4), 886–909. Kröger, M. 2020a. Iron Will: Global Extractivism and Mining Resistance in Brazil and India. Ann Arbor: University of Michigan Press. Kröger, M. 2020b. Field research notes on Amazon deforestation during the Bolsonaro era. Globalizations, 17(6), 1080–3. Kröger, M., and R. Lalander. 2016. Ethno-territorial rights and the resource extraction boom in Latin America: do constitutions matter? Third World Quarterly, 37(4), 682–702. Kröger, M., and K. Raitio. 2017. Finnish forest policy in the era of bioeconomy: a pathway to sustainability? Forest Policy and Economics, 77, 6–15. Leys, A.J., and J.K. Vanclay. 2010. Land-use change conflict arising from plantation forestry expansion: views across Australian fencelines. International Forestry Review, 12(3), 256–69, 14. Liu, W., J. Wu, H. Fan, H. Duan, Q. Li, Y. Yuan, and H. Zhang. 2017. Estimations of evapotranspiration in an age sequence of Eucalyptus plantations in subtropical China. PLoS One, 12(4), e0174208. Marini Perpetua, G., A.M. Kröger, and A. Thomaz Junior. 2017. Estratégias de territorialização das corporações agroextrativistas na América Latina: o caso da indústria de celulose no Brasil. Revista NERA, 20(40), 61–87. McKay, B.M. 2017. Agrarian extractivism in Bolivia. World Development, 97, 199–211. Moore, J.W. 2015. Capitalism in the Web of Life: Ecology and the Accumulation of Capital. London and New York: Verso. Myllylä, S., and T. Takala. 2011. Leaking legitimacies: the Finnish forest sector’s entanglement in the land conflicts of Atlantic coastal Brazil. Social Responsibility Journal, 7(1), 42–60. O’Neill, E. 2015. When planting trees hurts the environment. The Nation, 9 December. Available at: www.thenation.com/article/archive/when-planting-trees-hurts-the-envi ronment (Accessed 23 June 2020). Overbeek, W., M. Kröger, and J. Gerber. 2012. An overview of industrial tree plantations in the global South: conflicts, trends and resistance struggles. Ejolt Report Number, 3. Pakkasvirta, J. 2008. Pulp & Ficton: Tarinoita globalisaatiosta ja sellutehtaasta. Helsinki: Gaudemus Helsinki University Press.
Forestry extractivism in Uruguay 207 Piñeiro, D. 2012. Land grabbing: concentration and ‘foreignisation’ of land in Uruguay, Canadian Journal of Development Studies/Revue Canadienne D’études du Développement, 33, 471–89. Sepúlveda, C. 2016. Swans, Ecological Struggles and Ontological Fractures: A Posthumanist Account of the Río Cruces Disaster in Valdivia, Chile. Disseration thesis, University of British Columbia, Vancouver. Scheidel, A., L. Temper, F. Demaria, and J. Martínez-Alier. 2018. Ecological distribution conflicts as forces for sustainability: an overview and conceptual framework. Sustainability Science, 13, 585–98. Stine, R. 2011. Social media and environmental campaigning: brand lessons from Barbie. Reuters Events Sustainable Business [Online] August 4. Available at: https:// www.reutersevents.com/sustainability/supply-chains/social-media-and-environmentalcampaigning-brand-lessons-barbie (Accessed 15 January 2021). Stora Enso. 2020. Forests, Plantation and Land Use. Available at: www.storaenso.com/en/ sustainability/environmental/forests-plantations-and-land-use (Accessed 23 June 2020). Svampa, M. 2012. Cuatro claves para leer América Latina. Nueva Sociedad, 268, marzoabril de 2017, ISSN: 0251-3552. Toivonen, J. 2016. Analyysi: Yli miljoonan hehtaarin maaomistus ulkomailla – onko Suomi nykyajan siirtomaaisäntä? YLE News [online]. Available at: https://yle.fi/uuti set/3-9319223 (Accessed 3 July 2020). UPM. 2020. Innovating for a Future Beyond Fossils. Available at: www.upm.com/aboutus/this-is-biofore/ (Accessed 23 June 2020). UPM Forestal Oriental. 2018. Informe Publico. Available at: www.upm.uy/siteassets/docu ments/informe-publico-2018.pdf (Accessed 23 June 2020). Uruguay XXI. 2016. Informe del sector forestal en Uruguay. Available at: www.cama ramercantil. com.uy/uploads/cms_news_docs/Sector-Forestal-2016.pdf (Accessed 23 June 2020). Wohlleben, P. 2016. The Hidden Life of Trees: What They Feel, How They Communicate – Discoveries from a Secret World. Vancouver: Greystone Books. Ye, J., J. van der Ploeg, S. Schneider, and T. Shanin. 2020. The incursions of extractivism: moving from dispersed places to global capitalism. The Journal of Peasant Studies, 47, 155–83.
Index
Note: Page numbers in italics indicate a figure and page numbers in bold indicate a table on the corresponding page. Page numbers followed by “n” indicate a note.
4th Industrial Revolution 139 2030 Agenda for Sustainable Development 139 Abergel, E. 31 accumulation: capital 4, 8 –9, 11, 13, 22 –6, 28, 39 –40, 53, 87, 94, 102; crisis of 39; by dispossession 23 –4, 27, 64 –5, 69; of knowledge 26; in Latin America 25; legitimize processes of 10; maledominated form of 11; mechanism of 23; modality of 16; processes of, 69; scale of 27; tendencies of 27 Acosta, A. 141 agave harvest 174 agrarian chain 26 agrarian change: agrarian extractivism in Bolivia 46 –56; agro-extractive dynamics 56 –9; extractive dynamics of 45 –60; overview 45 –6 agrarian extractivism 25 –6, 158n8; advancing form of 10; agrarian chain 26; agribusiness economy 69; analysis of 10, 12 –16, 87, 93; appropriation of land and nature 66 –70; aspects for 12 –16, 39; Biotechnological Agrarian Model (BAM) 9, 22, 27; in Bolivia 46 –56; in Brazil 64 –80; Brazilian agricultural frontier 66 –70; characteristics of 74; characterized 8, 9, 46; in Colombia 85 –95; concentrating effect of 77; concept of 2 –3, 9, 11, 13; defined 8, 188 –9; developmental effects of 15; disproportionate effects on women 87; ecological impacts of 75; economic dynamics of 69; empirical gaze of 13; features of 46; gendered
workings of 85 –95; gender roles 10; in Guatemala 139 –58; illusionary promises of rural development within 117 –34; inequality 70 –4; introduction 8; investments 70 –4; land and water grabbing 78; land-use change 10; life purging model of 139 –58; from literal to expanded definition 101 –2; in Matopiba 70 –8; one-size-fits-all definition for 48; overview 64 –6, 99 –101; political utility of 46; predatory form of 7, 11, 15, 16, 74, 140, 142, 150; social-environmental conflicts 70, 74 –8; sustainable development and 99 –113, 157; territorial restructuring 16; transformation of land 14; see also sustainable development agrarian revolution 47 –8, 118 agribusiness economy 69 –70, 79 agribusiness lobby, Santa Cruz 49, 58 agricultural commodities 13, 79, 101 agricultural development models 2 agricultural employment 22 agricultural innovations 6 agricultural modernization 66, 70 agricultural research 6 agricultural revolution 21 agriculture: capital accumulation in 22; capitalist 47; chemical 74; climatesmart 1; corporate 6, 8; economic expansion 66; external-input plantation 1; extractive 56, 59; extractivism 141; extractivist dynamics of 6 –9; extractivist sectors 187, 197; forms of 16n1; fossil fuels 1; industrial 2 –3, 5 –6, 8, 22, 46, 48, 54, 79; industrial capital’s transformation of 3 –6; industrialization
Index of 66; innovation in 50; international 68; large-scale monocrop 1; livelihoods in 153; mechanized 4, 55, 79; mediumand large-scale mechanized 52; minister of 69; model of 1; modernize 67; monocrop 56; outside 134n10; penetration of 6, 9 –10, 12, 15, 26, 54; plantation 2; resources 7, 74, 167, 186; separate(s) trade 23; small-scale 52, 103; standardize 4; subsistence 47, 89; traditional 75, 80n6; World Bank’s 4 agro-chemicals 4; inputs 6; intensive use of 176; toxic 22; usage 4 agro-commodity flex crops 13, 22 agro-ecological agave 179 –80 agro-ecological methods 3 agro-export model 28 agro-exports 28, 35, 48, 58, 111 agro-extractive dynamics 56 –9 agro-extractivism 7 –8, 166 –71; see also forestry extractivism agro-food system 3, 13 agro-fuels, sugarcane production 117 –34 agro-industrial inputs 50, 120; economic dependency on 126 –8 agro-industrial model 3, 46, 127 agronomic risk 34 agro-strategies 68 –9, 79 Alcohol Producers Association of Ecuador (APALE) 123 Alier, M. 168 Almaraz, A. 47 Almeida, A.W.B. 68 Alonso-Fradejas, A. 11, 14 – 15, 48, 56, 86, 186, 190 –1 Añez, J. 57 –8 appropriationism 4, 54 Arauz, L.F. 99 Araya, L. 10, 93 Argentina 49; 61 GM seed varieties 21; agrarian extractivism expands in 22; agro-chemicals 49; Biotechnological Agrarian Model (BAM) in 9, 21 –40; Clarin (largest newspaper) 21; crossed national borders 33; frontiers 33; global markets 34; herbicides 22; Miami Group 33; organizations in 33 Argentina Chamber of Agricultural, Livestock and Fertilizer Sanitation (CASAFE) 34 Argentine Association of Direct Sowing Producers (AAPRESID) 29 Argentine Association of Regional Consortiums for Agricultural Experimentation (AACREA) 29
209
Argentinian Biotechnological Forum 29 Autodefensas Unidas de Colombia (AUC) 88 autonomy, farmer 2, 5 AZTRA (Azucarera Tropicana Americana) 121 Bernstein, H. 5 biodiesel 51, 57 biodiversity loss 1, 176 biofuels 8, 57, 167 Biotechnological Agrarian Model (BAM): agrarian extractivism 9, 22, 27; in Argentina 9, 21 –40; consolidation stages 28 –39; extractive capitalism phase 23 –8; first period (1991–1997) 28 –32; fourth phase (2008–2019) 36 –9; overview 21 –3; second period (1998–2002) 32 –4; shared conceptions 32; third period (2002–2009) 34 –6 biotechnological closed package 26 Bolivia: agrarian change in 45 – 60; agrarian extractivism in 48; agribusiness 45; GM soybeans in 4; industrial agriculture in 2, 46; landholding structure 47; soybean in 54; soy complex in 48 Bolsonaro administration 65 –6, 69, 78 border restrictions 1 Borras, S.M., Jr. 119 Brazil 35, 41n17, 49; agrarian extractivism in 64 –80; industrial agriculture in 2; Land Pastoral Commission 55; “Miracle of the Cerrado” 4 Brazilian Agricultural Research Corporation 72 Brazilian Forest Code 75, 77 Bunker, S.G. 7, 54 Camacho, J. 131, 133 Canada 29, 168, 189 capital: accumulation 4, 8 –9, 11, 13, 22 –6, 28, 39 –40, 53, 87, 94, 102; within agrarian chain 25 –6; existence 23; expansion and accumulation 23; extractive 25, 27; extractivism 23; labour and nature 23; manifestations 23; penetration of 26; rhythms of 26 Capital (Marx) 24, 86 capital-intensive model 5, 120 capitalist agriculture 47 Carrasco, A. 36 – 7 “Carrasco effect” 36 Cartagena Protocol on Biosafety (CPB) 33 cattle ranching 22, 52, 87, 91 centre–periphery relations 10, 101 Cerrado 70 –1; biome 65
210
Index
chapadas 76 chaqueo 57 Chaves, A. 99 – 100 chemical agriculture 74 China 8, 49 class 104, 149, 156; active agribusiness 70; antagonism 24, 39; hegemonic 120; middle 110; power 159n13; social 16; structures 7; struggle 35 climate-smart agriculture 1 climate stewardship 3, 15, 157 Colombia 85 –95 Colque, G. 9 commodification 6 –7, 15, 21, 23 –4, 26, 36, 39 commodified forces 2 Commodities Consensus 8, 17n3, 141, 166 commodity 3, 5, 7, 11 –13, 23 –5, 34, 50, 143, 145 –6; boom 68 78, 166; international trade of food 3; resignification of 24; seeds 26 competitive advantages 107 Consortium of International Agricultural Research Centres (CGIAR) 6 constrained economy 153 –4 contract 174, 177; annual harvest 126; arrangements of debt and dependency 14; farming 3 –5, 126 corporate agriculture 8 corporate-led, external-input plantation agriculture (CEPA) 1 –2; extractivist features of 6; industrial agriculture/agroindustry 3 corporate mega-mergers 3 Costa Rica 99 –105 COVID-19 pandemic 1, 27, 58 Dasgupta, A. 119 debt 4, 11, 14, 88, 118, 126, 149, 171 deforestation 1, 27, 51 –3, 57, 66, 71, 100, 102, 107, 196 Delgado, G. 66, 69 –70 dependency: on agribusiness 4; on agroindustrial inputs 126 –8, 133; debt and 11, 27, 59; economic 120; on male partners 90; on money 93; theory 10, 101 desarrollo sostenible a la tica, geographies of 102 –5 dispossession 8 –9; accumulation by 22 –4, 27, 64 –5, 69; analysis of 86; of Brazilian countryside 70; capitalist 16; centrality of 86; dynamics of 23, 87, 93, 110, 112; embodied dimension of 89; gendered experiences of 94; legitimizing/naturalizing forms of 15;
of local populations 88; within research spaces 25; of resources 133; of rural populations 101; sexual violence 89; subtle mechanisms of 27; of women’s bodies 94 Ecuador 102, 117 –34 eco-efficiency/ecological efficiency 117, 181 efficiency 4; ecological 181; gains 4; monetary 4; technical 2, 4, 66; see also eco-efficiency/ecological efficiency Ehrnström-Fuentes, M. 12, 14 –15 El Niño–Southern Oscillation (ENSO) 53 Eloy, L. 75 –6 environmental assessment 34 environmental change 3 environmental degradation, high intensity of 51 –5 environmental governance 69 environmentalism: for profit 110; politics 109 –11 environmental quality, deterioration of 130 –3 Etchevere, L. 21 ethanol extraction 121 –5 ethanol production 125 –6 exploitation 86, 90, 92; dynamics of 22, 24; economic 39; of labour 8, 10, 13, 54 expropriation 9, 14, 22 –7, 31, 36, 40, 54, 56, 72 –3, 90 –4 extraction, modes of 7 extractive agave: agro-ecological agave 179 –80; concept of (agro)extractivism 166 –71; foreign takeover 171 –5; history of 166 –71; overview 165 –6; social and environmental impacts of industrialized agave 175 –9; tequila production in Jalisco, Mexico 165 –81; traditional mezcal production 179 –80 extractive agriculture 56 extractive capital 25, 27 extractive capitalism phase, BAM 23 –8 extractive economies, internal dynamics 7 extractive imperialism 167 extractivism 166 –71, 186; accumulation of capital 25; constitutes, defined 85; as mode of appropriation 6 –7; social consequences 7 Ezquerro-Cañete, A. 86, 166 farmers: autonomy 2, 5; dependent on GM seeds 4; dispossessed 22; large 75; production costs for 4; small-scale family 3; soybean 51 –2
Index farm size and productivity, relationship between 3 Favareto, A. 72 Federici, S. 87 Figueres, J.M. 104 Financial, Industrial, Commercial and Agricultural Chambers of Guatemala (CACIF) 153 flex crops 8, 13, 22, 119, 139–40, 166, 169 flex (sugar)cane 150–3 food: availability 1; imports 47; international trade of 3; security 2, 47; shortages 1; sovereignty 2, 3, 47; systems 3 forest fires 56–7 forestry 1; agro-ecological management 179; agro-extractivism 187–92; corporations 15; dynamics 12; enclosures 195–6; extractivism 12, 15, 186–204; kind of agro-extractivism 187–92; long-term setting-up through stages 193–5; modern apparatus 14; overview 186–7; particularities 192–7; political ecology of 197–200; political ontology of 200–2; post mill-construction phase 195; pulp mill planning and construction phase 194–5; specific laws 193; strategic phase 193–4; trade deals 193; tree planting phase 194 fossil fuels 1, 5 França, F.C. 77 Fraser, N. 86, 90 Gago, V. 102 García, N.L. 10 gender 90–4, 156; -based violence 10, 87–90; division of labour 128–30; inclusion 117 – 35; unequal 2, 10–11, 14 genetically modified organisms (GMOs) 28–34, 46, 49, 58–9 genetically modified (GM) seeds 4, 21, 26–8, 31, 38, 45 geographies of desarrollo sostenible a la tica 102 – 5 Germany 147 Giarracca, N. 7, 167 global agri-food system 5 global commodity chains 3 global grain trade 5 global hunger 1 global resource rush 140–2 GM flex crops 13 GM Roundup Ready soybean 32 GM soybeans, Bolivia 4
211
Goodman, D. 4 grabbing: land 56, 64, 69–75, 77–8, 88, 93, 120, 139–40, 155, 168; water 74–5, 78, 93 GRAIN 140 Gramsci, A. 109 grano de oro (golden grain) 45 Green Revolution 65, 67, 70, 91 “grilagem” 64 Grosso, M. 72 Guarayos 52 Guatemala 139–58 Gudynas, E. 7, 27–8, 47–8, 101, 118, 141–2, 166, 186, 187–8 Guerrero, R. 120 Harvey, D. 23 Herrera-Rodríguez, M. 103 Hoekstra, A.Y. 54 Holguín, F. 121 hybrid seeds 4 impairing destruction 140 independent farming 5 industrial agriculture 2–3, 5–6, 8, 22, 46, 48, 54, 79; in Bolivia 2; in Brazil 2; CEPA model 3; defined 3; development 46 industrial capital: agriculture 4; appropriationism 4; penetration 2; transformation of agriculture 3–6; transformation of raw materials 101 industrialization of agriculture 66 inefficiency 124; see also efficiency intellectual property rights 25 International Assessment of Agricultural Knowledge, Science and Technology for Development (IAASTD) 3 Jakobsen, J. 198 Jalisco Environmental Protection Agency (PROEPA) 179 Karla, R.A. 10 Kloppenburg, J.R. 5–6, 26 Kröger, M. 12, 14–15 labour 2, 4–16, 23–4, 48–50, 55–6, 87, 89, 94, 101, 109, 120, 122, 128, 142–6, 149–154, 156–7, 177; and socio-ecological reproduction questions 146–9; opportunities and conditions, deterioration of 55–6; regimes 143; reserve 8 Lago Poopó 55 Lajmanivich, R. 22
212
Index
land: grabbing 56, 64, 69 –75, 77 –8, 88, 93, 120, 139 –40, 155, 168; regimes 73; -use change 1 Lander, E. 26 Landless Rural Worker’s Movement 2 Land Pastoral Commission, Brazil 55 large-scale 12, 47 –8; agribusiness enterprises 65, 70, 73, 92; farmers occupy 53; industrial agriculture 8; landowners 64, 77; mechanized agriculture 52; monocultures 51 –2, 54; monocrop agriculture 1; production of soybeans 78 Latin America: capital reproduction in 28; case studies from 9; chronic child malnutrition 160n30; development trajectory 67; Economic Commission (website) 181n1; expropriating territories 23; extractive capital 25; extractive frontier in 168; extractivism 78, 101; fisheries extractivism in 17n2; intellectual property rights in 27; natural resources 23; new extractivism in 8; palm oil expansion in 193; production of pineapples 105; rapid expansion in 167; rate of extracting materials from 168; seed laws in 27; social movement 85; societies 67; socioecological formations in 2; soybean plantations in 7; sugar mills 167 Levins, R. 25 Lewontin, R. 25 life purging agrarian extractivism 139 –58; countryside in name of sustainable development 153 –7; Guatemalan flex cane 146 –9; labour and socio-ecological reproduction questions 142 –50; latest global resource rush 140 –2; overview 139 –40; palm complexes 146 –9; predatory agro-extractivism 150 –3; resource extractivism 142 –6 livelihoods: in agriculture 153; communities 78; farmer 1; forests 52; monetization of 94; moral economies 92; Oaxaca 179; peasant 121; quilombola leaders 75; rural 4, 16, 56, 118; of small producers 127 lockdowns 1 Lomerío 52 López, H. 177 Lucio, C. 11, 13 –15 Lula da Silva, L.I. 2, 66 Luxemburg, R. 43 “map of death” 36 Marino, D. 22 market: application 31; oligopolies 5
Martínez-Alier, J. 160n23 Martins, J.S. 67 Marx, K. 8, 24, 50, 86, 167, 200 Matopiba, agrarian extractivism in 70 –8 McCulligh, C. 11, 13 –15 McKay, B.M. 9, 49, 86, 118 – 19, 186, 188 – 90 mechanization 5 mechanized agriculture 4, 55, 79 Mekonnen, M.M. 54 methodological notes, labour and socioecological reproduction questions 142 – 6 Mexican Institute of Water Technology (IMTA) 178 Mexico 29, 165 –81 mezcal production, traditional 179 –80 Mezzadra, S. 102 Mezzadri, A. 158n10 Miami Group 33 Mies, M. 86 Ministry for Industry and Productivity (MIPRO) 123 –4 Ministry of Agrarian Development (MDA) 72 Ministry of Social Development (MDS) 77 Modern Biotechnology Development and Production Law 36 modernization 24; agribusiness 66; agricultural 66 –67, 70; agro-industrial groups 121; of large properties 67; theories of 1, 46 Montes de María 87 –90 Moore, J.W. 86, 198 Morales, E. 47, 58 Mothers of Ituzaingo, The 36 Movimiento al Socialismo (Movement Towards Socialism, MAS) 47 Müller, R. 52 multi-stakeholder governance mechanisms 2 Naranjo, A. 127 natural disasters 51 nature: agrarian extractivism 54; appropriation of 10, 14 –15, 65 –9; conservation of 100, 104, 112; degradation of 103; environmentalism for 103; expropriates 2; external 15; importance in rural production 4; nonhuman 6 –7, 13; subsumption of 24 Neilson, B. 102 neo-institutional economy 4 neoliberal agrarian transformation 106 neo-Malthusian 1, 46 New Economic, Social, Communitarian, and Productive Model 47 new or neo-extractivism 117
Index
213
non-governmental organizations (NGOs) 29, 33 non-human: agency 142; agents 191; elements and processes 102; life 143, 156; nature 2, 6–7, 12, 13, 15, 140, 143; population 152; species 154 non-traditional agricultural exports (NTAE) 102 Nygren, A. 103
property 5, 7, 11, 14–15, 21, 24, 73, 143, 146, 148 Protected Areas Downgrading, Downsizing and Degazettement (PADDD) 76 public–private partnerships 25, 36 pulp mill planning and construction phase, forestry extractivism 194–5
O’Connor, J. 86, 158n6 O’Connor, M. 160n23 oil palm plantations 8, 87–90, 144 Ojeda, D. 10, 15 oligopoly 5–6, 59; capital 22; market 5–6, 27, 59, 121; power of sugarcane mills 119, 124 Operation Far West 64
reproduction 2, 5, 8, 14; capital 23, 25, 28; conditions 130–3; of seeds 39; social 10–11, 14–16, 85–95, 142–53; work 10, 129; unpaid 129 resistance 10, 24–5, 33, 35, 37, 66–8, 72–3, 200–2 resource extractivism 142–6, 146 Rosario Agrobiotechnology Institute (INDEAR) 27 Roundtable on Sustainable Palm Oil (RSPO) 1 Rousseff, D. 2, 65–6, 71–7
PAIS Alliance 117 – 18, 124, 127 palm complexes 150–3 Paraguay 49 Parnaíba River Spring National Park 66, 75–7 Pastoral Commission on Land (CPT) 74 patented seeds 6 payment for environmental services (PES) 102 Petras, J. 7, 168 Pineapple Development Company (PINDECO) 106 pineapple republic 105–9 Pineapple Research Institute (PRI) 106 plantation(s): agriculture 2, 12, 15; agrocommodity 10; banana 102; Caribbean 101; eucalyptus 189; large-scale 8; palm 10, 56, 87–90, 93–4, 151; pineapple 108–9; soybean 7; sugar cane 56, 101 post mill-construction phase, forestry extractivism 195 Poth, C. 9, 14, 22 predatory agro-extractivism 150–3; jobs and labour dynamics 150–2; palm commodity production 152–3; social metabolism of flex cane 152–3; value and capital flows, control over 150 Principles for Responsible Agricultural Investment (PRAI) 1 production–expropriation–appropriation of knowledge 9, 22, 24, 27, 31, 36, 40 productive and reproductive conditions 130–3 Program for Direct Support for Rural Areas (PROCAMPO) 177 progressive political cycle 69–70
quilombolas (Maroons) 72, 75
sacrifice zone 74 Sánchez, P. 51 Santisteban, R.S. 86 Sauer, S. 10, 77 Schumpeterian paradigm 25, 40n2, 159–60n21 sectoral disarticulation 49–51 seeds: commodity 26; genetically modified (GM) 4; hybrid 4; intellectual property of 22; patented 6 sexual violence 88 small-scale agave farmers, marginalization of 175 small-scale agriculture 52, 103 small-scale family farmers 3 social and environmental impacts of industrialized agave 175–9 social-environmental conflicts 74–8 social metabolic order 2 social metabolism 143 social reproduction 85–95; backstory of expropriation 90–4; defined 87; gender 90–4; gender-based violence 87–90; overview 85–7 social-reproductive regimes 143 socio-territorial networks 24 soil erosion 176 soils 4 Soluri, J. 101 Sorj, B. 4 soybean: agro-commodity flex crops 13; agro-extractivisms 196; Bolivia’s 54;
214
Index
cash crop 45; crop 45; cultivation 49, 52, 76; expansion 52; farmers 51 –2; global market 72; GM crops 58; harvested 54; jobs generated from 55; in Latin America 193; mechanized 55; noble 45; nutrients for 50; plantations 7, 45, 187, 194, 197; production 8, 50, 57 –8, 78; semi-processed 49 –50; surplus value 50; transgenic seeds 58; volume of 49 spatio-temporal fix 23 stages of consolidation, BAM 28 –39 strategic phase, forestry extractivism 193 –4 Striffler, S. 102 subordinated incorporation 108 subsistence 47, 89, 93, 107, 153, 156 –7, 170, 175 sugarcane mills 121 –3 sugarcane production 8; cultivated area 122; ethanol extraction 121 –5; gender inclusion 117 –34; overview 117 –20; small-scale producers’ integration 125 –33; from sugarcane to the ethanol 120 –1 surplus population 8, 156 surplus value 2, 118; appropriation 133, 142, 190; extraction of 5, 48, 50, 120; non-capitalist social strata 23; production 50; from production of ethanol 123; production of monocultures 79; resource extractivism through 143; sugarcane mills 124, 133; value-chain components 49 sustainability studies 2; transitions 2, 142, 196 sustainable development: 2030 Agenda for 139; agrarian extractivism and 99 –113; common sense 105, 107, 110; concept of 14; consensus 103; in Costa Rica 104 –5, 109; discourses of 11, 100 –2; ecologically efficient in 167; and economic revitalization 103; effects 144; framework of 100; geographical expression of 105; guise of 2; initiatives 157; promoting 15; purging countryside in name of 153 –7; silver bullet for 140; social-reproductive questions in 143; sociology of 142; sophisticated
legitimization campaigns 12; understandings of 103; version of 100 Sustainable Development Goals (SDGs) 1 Svampa, M. 8, 86, 101, 141, 166, 186, 191 –2 technical efficiency 2, 4, 66 Temer, M. 65 –6, 71 –2 tequila production 169 –71, 172; brands 173; foreign takeover 171 –5; history of 169 –71; in Jalisco, Mexico 165 –81; social and environmental impacts 175 –9 Tetreault, D. 11, 13 –15 Teubal, M. 7, 167 Tilzey, M. 49 “total market utopia” 26 toxic agro-chemicals 22 toxic landscapes 109, 154 –5 traditional agriculture 75, 80n6 traditional communities 76 traditional crops, displacement of 175 tree planting phase, forestry extractivism 194 Twente Water Centre 54 ultra-liberal economic narratives 69 United Nations: Bolivia 52; Costa Rica 111; Sustainable Development Goals (SDGs) 1 United States 29, 49 United States Department of Agriculture 6 unoccupied lands 67 Uruguay 186 –202 Vadillo, A. 52 value-chain concentration 49 –51 Vazquez, M.Á. 22 Veldwish, G.J. 75, 80n5 Veltmeyer, H. 7, 168 Verzeñazi, D. 22 Walter, M. 168 water grabbing 74, 75, 78, 93 Webber, J.R. 49 Weis, T. 4 White, B. 119 Wilkinson, J. 4 women, double burden on 128 –30 World Bank 3; agriculture for development 4 World Trade Organization (WTO) 33