Explaining Human Actions and Environmental Changes

EXPLAINING HUMAN ACTIONS AND ENVIRONMENTAL CHANGES

ANDREW P. VAYDA

EXPLAINING HUMAN ACTIONS AND ENVIRONMENTAL CHANGES

EXPLAINING HUMAN ACTIONS AND ENVIRONMENTAL CHANGES Andrew P. Vayda

A Division of Rowman & Littlefield Publishers, Inc. Lanham • New York • Toronto • Plymouth, UK

AltaMira Press A division of Rowman & Littlefield Publishers, Inc. A wholly owned subsidiary of The Rowman & Littlefield Publishing Group, Inc. 4501 Forbes Boulevard, Suite 200 Lanham, MD 20706 www.altamirapress.com Estover Road Plymouth PL6 7PY United Kingdom Copyright © 2009 by AltaMira Press All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher. British Library Cataloguing in Publication Information Available Library of Congress Cataloging-in-Publication Data Vayda, Andrew Peter. Explaining human actions and environmental changes / Andrew P. Vayda. p. cm. Includes bibliographical references and index. ISBN 978-0-7591-0323-8 (cloth : alk. paper) — ISBN 978-0-7591-1900-0 (electronic) 1. Human ecology. 2. Human beings—Effect of environment on. 3. Ethnology. I. Title. GF51.V36 2009 304.2—dc22

2008055699

Printed in the United States of America

⬁ ™ The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI/NISO Z39.48-1992.

For my granddaughters, because they ask questions

Contents

Preface

ix

Acknowledgments

xi

1 2 3

4

Causal Explanation as a Research Goal: Dos and Don’ts

1

Both Ends of the Firestick: Causal Explanation of Indonesian Forest Fires

49

On Knowing What Not to Know about Knowing: A Critical View of Local Knowledge Studies (with Bradley B. Walters and Indah Setyawati)

79

Do We Need an Anthropological Perspective on Tropical Deforestation?

109

5

Seeing Nature’s Complexity but Not People’s

123

6

Against Political Ecology (with Bradley B. Walters)

129

7

Failures of Explanation in Darwinian Ecological Anthropology

143

Concepts of Process in Social Science Explanations (with Bonnie J. McCay and Cristina Eghenter)

191

8

vii

viii / Contents

9 10

Explaining Why Marings Fought: Different Questions, Different Answers

205

The Anthropology of War: Polemics and Confusion

229

Bibliography

239

Index

291

About the Author

303

Preface

The essays collected here are united by their concern with explanation and with methods of explanation-oriented research. An original stimulus for some of the essays was my desire to get at the causes of particular phenomena, like intergroup fighting in the mountains of New Guinea and extensive fires in the tropical moist forests of Indonesia. For other essays, the original stimulus was more my being dissatisfied—on logical, empirical, or pragmatic grounds—with research methods widely used or kinds of explanations commonly made in such fields or subfields as political ecology, Darwinian human behavioral ecology, and local knowledge studies. Whatever the stimulus, going beyond my criticisms of the work of others and achieving better explanations and identifying ways of achieving them were among the positive goals I set for myself. The essays are presented in mostly, but not entirely, the reverse chronological order of publication of the articles on which the essays are based. The first chapter is the most general, setting forth my understanding of causal explanation and of research with such explanation as a goal. The chapter includes brief illustrations of dos and don’ts in explanation and explanation-oriented research. A reason that many of the brief illustrations are from my own work is that the chapter was originally written in response to comments on my work by contributors to a festschrift. ix

x / Preface

In subsequent chapters, more detailed illustrations are given and particular points or arguments are elaborated. The topics treated include, among others, confirmation bias, naïve functionalism and adaptationism, the misdirection of interdisciplinary research, the practical defects of holistic approaches, the reification of processes or systems, the importance of clarity about our objects of explanation, and the advantages of having events as those objects. Some of the issues I am dealing with are complicated and have complicated histories involving disciplines ranging from anthropology and geography to ecology and philosophy. However, I have tried hard to keep my discussions of all issues clear and as free as possible of disciplinary jargon. Relevance to methodological and explanatory issues still with us today was a criterion for selection of essays for inclusion. Accordingly, hardly any changes needed to be made in their texts. Some words have been added, mostly in footnotes, for cross-referencing between chapters and, in chapter 1, for citing the aforementioned festschrift and clarifying a few points. Also, there are some new footnotes referring to especially relevant recent literature. In some cases, as in chapter 7 on Darwinian human behavioral ecology, the citation of certain recent works is intended in particular to underscore the present-day pertinence of criticisms I first voiced years ago.

Acknowledgments

In either writing some of the essays or doing the research for them, I received help in various forms from many people. Their contributions are gratefully acknowledged either at the end of each of those essays or in the notes to them. Three of the included essays are coauthored, and I give thanks to Brad Walters, Indah Setyawati, Bonnie McCay, and Cristina Eghenter not only for their original collaboration but also for their cooperation in making those essays ready for the present volume. Advice or other help useful for preparation of the book as a whole was received from Brad Walters, Andi Vaida, Seymour Weingarten, Margaret Nolan, and Audrey Charlton. Brad Walters’s contributions to my thinking and writing deserve special mention. He and I have been frequently, happily, and effectively bouncing words and ideas off each other since 1993 and are planning future collaborations. For permission to reprint material, I thank the publishers of the articles from which the present essays have been drawn. The following are the original titles and other publication details for each chapter: Chapter 1: Vayda, A. P., 2008. Causal explanation as a research goal: A pragmatic view. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda

xi

xii / Acknowledgments

Tradition in Human Ecology and Ecological Anthropology, Lanham, MD: AltaMira Press, pp. 317–367. Chapter 2: Vayda, A. P., 2006. Causal explanation of Indonesian forest fires: Concepts, applications, and research priorities. Human Ecology 34: 615–635. Chapter 3: Vayda, A. P., B. B. Walters, and I. Setyawati, 2004. Doing and knowing: Questions about studies of local knowledge. In A. J. Bicker, P. Sillitoe, and J. Pottier, eds., Investigating Local Knowledge: New Directions, New Approaches. London: Ashgate Publishing, pp. 35–58. Chapter 4: Vayda, A. P., 1998. Anthropological perspectives on tropical deforestation? A review article. Anthropos 93: 573–579. Chapter 5: Vayda, A. P., 1991. Review of D. B. Botkin, Discordant Harmonies: A New Ecology for the Twenty-First Century. Human Ecology 19: 423–427. Chapter 6: Vayda, A. P., and B. B. Walters, 1999. Against political ecology. Human Ecology 27: 167–179. Chapter 7: Vayda, A. P., 1995. Failures of explanation in Darwinian ecological anthropology: Parts I and II. Philosophy of the Social Sciences 25: 218–248, 360–375. Chapter 8: Vayda, A. P., B. J. McCay, and C. Eghenter, 1991. Concepts of process in social science explanations. Philosophy of the Social Sciences 21: 318–331. Chapter 9: Vayda, A. P., 1989. Explaining why Marings fought. Journal of Anthropological Research 45: 159–177. Chapter 10: Vayda, A. P., 1992. Review of J. Haas, ed., The Anthropology of War. Anthropos 87: 268–271.

1 Causal Explanation as a Research Goal: Dos and Don’ts

What does it mean to engage in causal explanation and make it a goal of our research? And how should we do these things? These are questions addressed in this chapter. Most of the chapter will be devoted to description and analysis of what I regard, from a pragmatic standpoint, as the most important dos and don’ts of engaging in causal explanation and explanation-oriented studies of human actions and environmental changes. Although brief illustrations of both dos and don’ts will be given, extended or more detailed illustrations will be left to the other chapters. Among the topics I will consider here are the use and misuse of generalizations, the use and misuse of evidence, prioritizing in research, moving from one level of analysis to another, and the uses made of statistics. First, however, I must describe what I mean by causal explanation, and I must discuss what makes it important. This will require devoting parts of the next few pages to consideration of certain philosophical questions about causation. To readers usually not engaged by such questions, I recommend going first to the summary at the end of the chapter for perspective on the issues.

1

2 / Chapter 1

Causes without Metaphysics Must the nature of causality itself be well characterized and explicated before causal explanation can proceed? This is a view held by some both within and outside philosophy (see Salmon 1990: 114–115 and passim and Kuznar and Long 2008). However, from a pragmatic standpoint, no antecedent immersion in what philosophers refer to as the metaphysics of causation is required for our getting on, as I have, with the practical business of causal explanation, that is, with trying to account for the occurrence of either particular events, like specific movements of people or changes in land use by them (Vayda and Sahur 1996: 5 ff.), or else, in a more general way, kinds of events, like fires in tropical moist forests (see chap. 2).1 Still, the fact of conflicting views on this makes it appropriate to mention some issues in the metaphysics of causation and their implications, if any, for the practical business of causal explanation. For those interested in going more deeply, but still not too deeply, into the issues, the one article I would recommend is Schaffer’s (2003). Mentioned at the outset by him are questions about the nature of the entities in causal relationships and the nature of their relatedness. More in the language of philosophy, the questions may be said to be about the nature of the causal relata (what they are, where they are, how finely grained they are, and how many there are) and about the metaphysics of the causal relation (e.g., the difference between causally related and causally unrelated sequences). However, any expectation that philosophers can help us nonphilosophers do causal explanations properly by providing us, once and for all, with the right, agreed upon answers to these metaphysical questions is quickly dispelled by Schaffer. Philosophers, he states, have disagreed 1 I will not dwell here on comparing, as have Driscoll and Stich (2008), my view of causal explanation with that of others. Let it be noted, however, that my emphasis on explanations as practical answers to questions makes it appropriate to describe my view as not only causal but pragmatic as well and, in contradistinction to some other pragmatic and/or causal views, also erotetic or question-based (cf. Rosenberg 2002: 155–156 and Risjord 2005: 8–9 on erotetic explanations). Something else that I will not explicitly dwell on are differences between making pragmatically oriented causal explanation the goal of our research and making, as many in various disciplines still do, theory construction or theory validation the goal (cf. Baker 1996: 205–206 and passim on this).

Causal Explanation as a Research Goal: Dos and Don’ts / 3

about all of them. For example, after noting the standard view that the causal relata are events and that they are necessarily two in number (cause and effect), he states that this has been disputed on all counts. Thus, he refers to such other proposed causal relata as facts and features. He notes also disagreements over what such entities as events and facts are.2 All this may seem like bad news for those interested in being guided in research and analysis by the goal of causal explanation. However, I do not take it that way. On the contrary, I regard it as supporting decisions I’ve already made in research projects to be flexible and, for example, to let the causal relata depend on the context of inquiry or on the question being asked. Thus, in projects to explain certain changes in land use or agricultural practice by Indonesian villagers, my causal relata, like those in many other studies,3 clearly were events, including such causative events as payments of compensation to farmers displaced from their land (Vayda and Sahur 1996) and the elimination of subsidies to farmers for the purchase of agricultural pesticides (Vayda and Setyawati 1998). Moreover, the events I was dealing with and considering as causal relata were ones I could identify and describe well enough for project purposes without having to delve deeply into the problems that exercise metaphysicians with regard to event ontology, identity, concreteness, and individuation (see, for example, Casati and Varzi 1996, 1997, 2006, and Laurence and Macdonald 1998: Part V)—even if many of the metaphysicians go along, as I do, with a simple, initial definition of an event as something that happens somewhere during a particular interval of time (see Lombard 1991, 1995). Events as causal relata served also in early stages of my review of research on the causal explanation of Indonesian forest fires (chap. 2) and led me to agree with others who had concluded that so-called predatory logging, leaving

2 More on the disagreements may be found among the works cited in Schaffer’s fivepage bibliography. For a clear and concise account of issues concerning causal relata, see Crane 1995: 191–194. On what events are, see Casati and Varzi 1997, with its 331-page annotated bibliography. For introductory essays on metaphysics, see the introduction by the editors and van Inwagen’s chapter 1 in Laurence and Macdonald 1998. 3 Examples in Walters et al. 2008 include the chapters by Haaland, McGuire, Setyawati, Vondal, and Walters.

4 / Chapter 1

flammable deadwood residues and creating canopy openings that change forest microclimates, causes fires in tropical moist forests. This explanation suffices for some broad analytic and policy purposes and is supported by remote-sensing evidence that fire reached 59 percent of East Kalimantan’s logged forest area and only 5.7 percent of its undisturbed forest area in 1997–1998 (Siegert et al. 2001: 439). However, when I turned to such questions as what, if any, kind of less predatory or more controlled selective logging would not have resulted in similar fire damage, I had to ask about the fire causes or fireinducing mechanisms present in 59 percent of the logged area and absent from the other 41 percent. This meant asking how fire behavior is affected by specific features or properties of logging, such as the amount of residues it leaves and the size of the canopy openings it creates, and by specific features of the environments in which the logging has occurred. Thus I was making my questions more specific, much as Bart (2008: esp. 106), in an analysis of biological invasion, has done in asking why ditching sometimes leads to establishment of Phragmites (Common Reed) in tidal marshes of the east and gulf coasts of the United States and sometimes not. Does this mean that I was regarding the specific features or properties and not logging per se as causal relata and that Bart was doing something similar in his Phragmites studies? Certainly that’s a possible interpretation, but it’s also possible to continue, if we wish, to regard only events as the relata if the events are appropriately redescribed, for example, if we refer no longer to logging per se in our explanations but rather to logging with certain properties, such as those related to the deadwood-residue and canopy-opening variables.4 Another alternative is not to explicitly redescribe the events but still to specify what Braun (1995) calls their “causally relevant properties,” whether hypothe4 Cf. Anscombe 1969: 155 and Emmet 1985: 30 on “explanatory force” or “causal relevance” as a basis for redescription. For one philosopher’s argument that flexibility in redescribing phenomena in the pursuit of explanations is the foundation for the success of the natural sciences and should be adopted by the social sciences, see McIntyre 2004.

Causal Explanation as a Research Goal: Dos and Don’ts / 5

sized or demonstrated. Here is one of Braun’s examples: When we say that someone’s singing caused a nearby glass to vibrate and shatter, the high pitch and amplitude of her singing are causally relevant properties of the shattering but the words of her song are not. If we were to opt for explicit redescription, we would say that singing at a high pitch and amplitude caused the glass to shatter. A causally relevant property of ditching in Bart’s studies is the burial of large rhizome fragments of Phragmites, but ditching without this property also occurs, just as does singing without glass-shattering pitch and amplitude. In line with my pragmatic orientation, it’s important to note that it doesn’t matter whether we choose for the purpose of causal explanation to redescribe events or simply to specify their causally relevant properties. Either of these courses may be followed as long as just what we’re talking about and how it relates to the causal questions we’re asking and answering are clear to ourselves and can be made clear to others at any given point in our inquiries. As Schaffer notes, the bulk of the vast literature on causation concerns the question of connection or, a little more precisely, the nature of, or metaphysical basis for, causal connection. Among the many accounts that have been given in answer to the question are some, with such labels as nomological, statistical, counterfactual, and agential, focusing on the role of causal connections in increasing the probability of some things happening, and others, with such labels as process and transference, focusing, more deterministically, on the role of the connections in actually producing the occurrences. However interesting philosophically I may find these various accounts, I must reiterate that the practical business of causal explanation can proceed without our having to commit ourselves to one or another of the accounts (or to some combination of them) or, indeed, without our having to address at all the metaphysical question of causal connection. This, rather than being simply a nonphilosopher’s naïve view, is a view variously held or implied by some, even if by no means all (see Salmon 1990: 114–115 and passim),

6 / Chapter 1

reputable philosophers as well.5 Some of these philosophers (e.g., Anscombe 1981: 135 ff., Hart and Honore 1985: chap. 1; Lipton 1991: 3–4, 33; McCullagh 1984: 171–172; Scriven 1966: 251), looking at the role of causation in explanation and noting that causal attributions and causal reasoning are fundamental or indispensable not only for much of science, philosophy, historiography, and/or jurisprudence but also for everyday life, say more or less directly that we can talk about causes and recognize them—as, in fact, we are continually doing—without our having any final or even merely tentative answers to the metaphysical questions. Some, like Lipton (1991: 33), express the hope that those final answers will eventually be found and will be congruent with their analyses of causal explanation (cf. Lewis 1986: 216–217), while others, like Hart and Honore, are more skeptical and accept a possibility suggested to them by John Stuart Mill (1886: Bk. III, chap. 5), namely, that causation varies from context to context, that there are different types of causal inquiry, and that there isn’t a single concept of causation to be found but rather a cluster of related concepts (Hart and Honore 1985: 18–19).6

Giving Causal Explanations What then do I mean by “the practical business of causal explanation”? I definitely do not mean only the kind of purportedly rigorous “laws-and-causes social physics” (Geertz 1983: 3) which was 5 Here as elsewhere, I am citing certain philosophers not for the reason imputed to me by Driscoll and Stich (2008: 180–181). Contrary to their misreading of parts of my critique of Darwinian ecological anthropology or, as it is now usually called, “human behavioral ecology” (HBE) (Vayda 1995a and chap. 7 of this volume), I do not claim that the view supported by those I cite is necessarily the only correct one. Instead I cite them to show that there are alternatives to a view held by others, be it human behavioral ecologists’ professedly nomological view that explanation is achieved by subsuming observed phenomena under generalizations or Kuznar and Long’s (2008) apparent view that analyzing the nature of causation or causal connection is a prerequisite to engaging in causal explanation. Note that one philosopher of science, evidently disagreeing with Driscoll and Stich’s characterization of their discipline as descriptive rather than prescriptive, sees as “something of a scandal” the fact that scientific explanation is still widely believed to consist of subsumption under law (Cummins 2000: 117; cf. Rosenberg 2005: 22 and Thagard 1999: 87, both also arguing for prescription as well as description in philosophy of science). 6 Cf. Cartwright (1999: 119): “there is no such thing as the causal relation. . . . One factor can contribute to the production or prevention of another in a great variety of ways. There are standing conditions, auxiliary conditions, precipitating conditions, agents, interventions, modifications, contributory factors, enhancements, inhibitions . . . etc.” For elaboration of Cartwright’s view, see Cartwright 2007.

Causal Explanation as a Research Goal: Dos and Don’ts / 7

rejected decades ago by so-called interpretive or “anti-positivist” anthropologists (e.g., Geertz 1983: 3, 34, and passim; Latour 1988; Rosaldo 1989: 37 ff.) and other social scientists and was, quite recently, still being conjured up for them by the word “explanation” (cf. Risjord 2000: 61; Somers 1996: 57–59).7 Instead the causal explanations I have in mind encompass not only those made by many of us in our academic pursuits but also those we are continually seeking and giving for such commonplace questions as why a train is late, why a friend hasn’t called, why a faucet leaks, why someone falls ill, why a car breaks down, why the food cooked from a recommended recipe tastes bad, why a child suddenly cries, why a promise is not kept, why the lines are so long at the checkout counter, why the price of gasoline or orange juice or whatever is up, and so forth. As for the answers given in response to the why-questions in such homely examples as well as in the cases that I and other academics in various disciplines deal with, an important point in accord with some pragmatic views of causal explanation is that the explanations given or accepted depend on what the explanation-givers or explanation-seekers already know and what they still want to know (Scriven 1975; Gärdenfors 1980, 1990; Lewis 1986: chap. 22; Hilton 1995; Vayda 1996: 23–24; cf. Passmore 1965). In one explication of such a pragmatic view, the philosopher Lewis (1986: chap. 22) has characterized the information sought and given in causal explanations of events as information about their causal histories. This is a characterization that I have been guided by in my own research.8 But what, more precisely, does it mean to give information about an event’s causal history? Lewis (1986: 218–220) observes that such information, which he designates as explanatory information, comes in “many shapes and sizes” and ranges from the very specific to the more general. In imparting explanatory information,

7 The rejection of “laws-and-causes” explanations by social scientists may have involved, inter alia, their having too rigid a view of the character of the laws actually being applied in explanations in physics (see, for example, Cartwright’s 1999 book, arguing at length that the laws hold only ceteris paribus). For discussion of some aspects of the “interpretive turn” in American social science, see Trencher 2000: 160 ff. 8 Since some such critics of my work as Padoch et al. (2008) have, rather confusingly, conflated the prediction of future events with the explanation of past ones, it may be in order here to state explicitly that this chapter and subsequent ones focus mainly on the problems of explaining and not of predicting (cf. Cleland 2001: 989 and 2002: 487 ff., citing Lewis 1979, on the temporal asymmetry of causation). For a brief reply to the criticisms by Padoch et al., see Vayda 2008: 363, note 23.

8 / Chapter 1

an explainer might specify only a single event as the cause of the event that is the explanandum or event to be explained. (“The car crash occurred because the driver had drunk too much.”) Or he might specify several events as causes. Or he might specify many. He might trace a single causal chain or a number of independent ones converging on the explanandum or else a more complicated structure of intersecting causal chains. (“Not only had the driver drunk too much but also his tires were bald and he had had an unhappy childhood which made him reckless about both driving while drunk and driving with bald tires.”) The explainer might keep causal chains short or make them long. If he has a theoretical bent, he might especially emphasize certain structures or patterns that he discerns in the causal histories—negative feedback structures or patterns of dialectical opposition, for example. On the other hand, he might include nothing like these in his explanatory information. Similarly, he might or might not point to certain correspondences between known, generalized causal processes or mechanisms—like oxidation or natural selection or, to take a sociological example, chain migration—and particular sequences of events antecedent to the event he wants to explain. Even negative information may qualify as explanatory. An example that Lewis (1986: 220) gives: If a question is raised about the causal significance of the presence of an agent of a foreign government when His Excellency dropped dead and the answer turns out to be that it was just a coincidence, this is negative information to the effect that “a certain sort of pattern of events” (that is, a plot) does not necessarily figure in the causal history. So there is much latitude in giving explanatory information. The foregoing is concerned mostly with explanations of particular events. Although there are some who think otherwise, such explanations can stand on their own.9 However, they can also be readily extended to explanations of what Lewis (1986: 225) refers to as “events of a given kind.” Even if the events thus designated have their separate causal histories, the very fact that 9 Among those thinking otherwise and regarding explanations as general by definition are the Darwinian human behavioral ecologists discussed in chapter 7. See esp. p. 176.

Causal Explanation as a Research Goal: Dos and Don’ts / 9

they are thus designated makes it likely that certain shared features characterize the causal histories of at least some and possibly most or even all of them. To identify some such features—and there is latitude about which of them and how many of them—is to give what Lewis calls general explanatory information about events of a given kind. Thus, to take one of his examples, regardless of whether we are explaining why struck matches light in particular cases or in general, we would be apt to cite causal histories involving, inter alia, friction, small hotspots, and liberation of oxygen from a compound decomposing as it becomes hot. For another example, consider Sen’s famine studies (1981). In these, actual deaths from starvation were regarded as the events of a given kind to be explained. Investigating the causal histories of those deaths meant asking who the victims were and what changes had occurred in their situations to make them starve while others in the same society still had enough to eat. General explanatory information is then constituted by shared features found in the causal histories. Thus, in the case of the Great Bengal Famine of 1943 in which about three million people died, the causal histories of many deaths in the countryside were found by Sen to have in common that the victims were landless poor (agricultural laborers, artisans, fishermen, and others) whose wages and incomes had not kept pace with the inflated prices of the staple grains during a wartime economic boom. Often, whether for the purpose of comprehension, policymaking, or action, general explanatory information rather than only the explanation of particular events is indeed what we are seeking. Many of the examples cited later in this and subsequent chapters are, in fact, illustrations of attempts to explain such events of a given kind as fires in tropical moist forests, deaths from diseases like lung cancer and cholera, migrations of would-be petty entrepreneurs to foreign locations, prey selection by hunter-gatherers, mangrove planting by local people in the Philippines, and intergroup fighting in the mountains of New Guinea. Regardless of whether the explananda are particular events or events of a given kind, explanatory information consists of causes falling into just two broad types: those involving physical

10 / Chapter 1

pressure being exerted by one object on another and those involving the intentions of agents who are living beings. The first type, in its simplest form, is the billiard-ball causation long featured in discussions by philosophers and in studies by developmental psychologists, and the second type is sometimes labeled mental causation.10 There is no evidence of any societies/cultures whose members do not recognize and cite both types in explanations. Indeed there also is evidence of both types in cognition by nonhuman primates, even if they may be less adept than humans in inferring the intentions of other individuals to explain their behavior (see Kummer 1995 and Alvard 2003: 143 and the references cited therein). Some recent experiments even suggest that rats engage in causal reasoning (Blaisdell et al. 2006). Perhaps needless to say, all this causal explanation or causal reasoning is not just mental recreation. Much of it is applied, often with critical effect, to the practical business of living. Of course, members of many societies/cultures, including our own, engage also in other than the two main types of causal explanation, for example, by attributing occurrences to mystical or magical forces. There may also be substantial intercultural variation in the kinds of intentions that lead to actions and are invoked to explain them (cf. Morris et al. 1995). None of this, however, invalidates assertions or evidence of the generality of the two main types of causal explanation. A relevant and classic ethnographic case is Evans-Pritchard’s account (1937: chap. 4) of the Zande people’s knowing about physical causation and using their knowledge in explanations and everyday life despite their readily attributing untoward events to witchcraft. An initial point to be made from the above examples of causal explanation and from my references to causal explanation or causal reasoning in all societies and among different species is about the ubiquity of causal explanation and of behavior based upon it. A follow-up point is about the mistakenness and waywardness of social scientists who, finding nomological, hypothetico-deductive, or so-called positivist causal explanation 10 The philosophical literature on both types is vast. For a brief introduction to some of the issues that have occupied philosophers, see Searle 1998: 58–62.

Causal Explanation as a Research Goal: Dos and Don’ts / 11

insufficiently accommodating of their research interests, made professions of turning away from causal explanation altogether and went about propagating the non sequitur that the meaningfulness of human actions precludes them from being subjected to causal inquiry.11 All this was done with virtually no recognition or acknowledgment that less exigent forms of causal explanation than that represented for them by the problematic, rejected ones had long been successfully applied in such fields as law, medicine, and history (see the discussion and references below) and might well be used to guide social science inquiry as well. As far as the non sequitur is concerned, there also was no explicit recognition or acknowledgment of various causal questions that can be usefully asked about meaningful human actions, for example, about what causes the occurrence of actions to which the actors attach some particular meanings, what caused those meanings to be attached to those actions, whether or to what extent the meanings themselves are causes of the actions, what other causes than the meanings are there for the actions, and so forth.12

Obtaining Explanatory Information Not only is there much latitude in giving explanatory information but, even if it’s definitely not the case that “anything goes,” there is latitude also in the means that may be used in research to obtain such information (cf. chap. 10). In line with this, I do not advocate, or try to identify, a set of methods as constituting scientific explanation and therefore being something special and 11 By saying that “professions of turning away” were made, I am alluding to the fact that, as Somers (1996: 59) observes, “implicit and ad hoc ‘causes’ inevitably sneak into arguments” made by interpretive social scientists. Among examples are the therapeutic or order-producing efficacy claimed or suggested for Balinese cockfights in Geertz’s famous article (Geertz 1973: chap. 15; Rabinow and Sullivan 1987: 26) and for Nepalese Sherpa rituals in Ortner’s 1978 book. Being inexplicit about their causal analyses provides, in effect, cover to the authors for doing them badly, for example, with only impressions and anecdotes as evidence and no consideration of alternative causal possibilities. For fuller accounts of these failures, see Martin 1993: 277–285 on Geertz and Jacobson 1991: 54–66 on Ortner. 12 For insightful discussions related to interpretive social scientists’ obliviousness to most such questions qua explicitly causal questions, see Greenwood 1990; Bunzl 1993: chapter 8; Roscoe 1995; Kincaid 1996: 205 ff.; and Roth 2003.

12 / Chapter 1

right. Instead I agree with two conclusions appropriately drawn by Roth (1996: 61, 67) from his review of issues concerning socalled scientific methods and explanations: 1) that we had best give up the notion that some particular methods and norms constitute science and “can be used to demarcate legitimate practices from the rest” (cf. Cleland 2001: 987; Kincaid 2000a); and 2) that giving this up would have the constructive effect of making us “explicate and assess each putative explanatory practice in terms other than those drawn from some abstract characterization of what it is to be a science.” Indeed I will shortly discuss some other terms or considerations useful for judging explanations to be good, bad, or something in between. Also in line with latitude in methods for obtaining explanatory information, I do not advocate any detailed methodology as the basis for new fields or subfields, deserving their own distinctive names of “event ecology” and “event social science” or something similar.13 In fact, when I first came up with the label of “event ecology” (in Vayda 1997a), it was with the idea not of starting a new field or subfield but only of rhetorically underscoring that research intended to explain environmental change should begin with a focus on the environmental events constituting such change rather than on only one set or kind of possible explanantia or explanatory factors, political or whatever, privileged in advance. More will be said below about the methodological point I was trying to make, since it still is, in my view, one of several basic precepts to be generally followed in research with pragmatically oriented causal explanation as its goal. Suffice it to say here that the precept is compatible with more research methods and procedures than even those that I and collaborators like Walters have highlighted in some of our articles (e.g., Vayda 1996; Walters 2008; Vayda and Walters 1999). 13 On the difference between “event ecology” and “event social science,” see Vayda 2000. The former label was meant to be applied only to studies, like Bart’s on Phragmites invasions, in which the explanandum events are environmental. For studies with human actions as the basic explananda, “event social science” would be the more appropriate label. In terms of this categorization, my forest-fire research (chap. 2) could be called “event ecology,” but many of my earlier studies (e.g., chap. 9 and Vayda 1980; Vayda and Sahur 1985, 1996; Vayda and Setyawati 1998), with explananda consisting of human actions like fighting and migration, were “event social science,” even if environmental causes of the actions were or could be found.

Causal Explanation as a Research Goal: Dos and Don’ts / 13

Consider, for example, a recommendation I made perhaps most explicitly in 1996 (even if not for the first time then), namely, that the threads of causal influence on concrete actions or events should be traced “outward in space” (Vayda 1996: 1 and passim). This is cited, mostly approvingly, by a number of contributors (e.g., Walters, Nyerges, McGuire, Eghenter, and Kuznar and Long) to the festschrift for me (Walters et al. 2008) and appears to have been used to good effect in some of their case studies. My making the recommendation in 1996 was probably a reflection of specific research I had done or was doing on such matters as Bugis migrants’ impermanent use of frontier land for growing commercially valuable perennial crops in Sumatra and Kalimantan (Vayda 1980, Vayda and Sahur 1985, 1996). That research did identify some causes far from the local explanandum events. For example, it allowed me to cite, inter alia, data on how Bugis practices and beliefs concerning mobility and migration to frontiers had been promoted or reinforced by geographically extensive Bugis social networks along which information about opportunities in other lands flowed back to would-be migrants in settlements either in the South Sulawesi Bugis homelands or elsewhere (cf. Vayda 1996: 21–22). In my subsequent forest fire research (chap. 2), I was at first still going outward in space in following chains of event causation backward in time from the extensive forest fires that occurred in Indonesia in 1997–1998. As a result, I joined others (e.g., Barber and Schweithelm 2000 and Dauvergne 1998) in concluding that the Suharto government’s forestry policies, having led to the kind of predatory or uncontrolled logging that makes tropical moist forests more fire-susceptible, were underlying causes of the fires. However, as already noted above, when I turned to such questions as what, if any, kind of less predatory or more controlled selective logging would not have resulted in such extensive fire damage, I had to ask how fire behavior is affected by specific features or properties of logging, such as the amount of deadwood residues it leaves and the size of the canopy openings it creates, and by specific features of the environments in which the logging has occurred. In other words, in trying to explain fire occurrence in some logged forest areas and not others, I was going inward rather than outward in space. I

14 / Chapter 1

should have paid more attention in 1996 to the fact that this is a not uncommon direction for causal inquiries to take. One area in which examples abound is research on causes of disease occurrence. To take just one example, consider studies of the causes of lung cancer (Thagard 2000: 257–260, 265, citing Hennekens and Buring 1987). A strong case for smoking as a cause was first indicated by studies that not only showed its high correlation with the disease but also ruled out various sources of observation bias and controlled for confounding factors.14 These studies were followed by more fine-grained or “inward-in-space” research leading to the identification, first, of many specific carcinogens in cigarette smoke and then, in the mid-1990s, of Benzo[a]pyrene as a particular component of the smoke whereby a pattern of DNA damage in the p53 tumor suppressor gene of lung cells is produced (Denissenko et al. 1996). Research is continuing on other agents that may similarly affect the p53 gene (Feng et al. 2006; Hecht 2006). As implied by the preceding paragraph, latitude in methods for obtaining explanatory information includes latitude regarding the level, grain, or scale at which research is conducted.15 Thus, we can feel free to move to a lower level or finer grain when we have reason to think that doing so will produce not only more specific or more testable but also more useful answers to specific causal questions, like those above concerning forest fires, Phragmites invasions, and lung cancer.16 However, contrary to the prescriptions of some so-called reductionist programs, we can also feel free to refrain from moving to a lower level when we have no reasonable expectation of finding useful explanatory

14 On the strength of this case compared to others in which statistical models have been used for drawing causal inferences, see Freedman 1999: 252–254 and the references below to critiques that Freedman and others have made of much weaker cases. 15 Latitude about this sometimes means disregarding disciplinary boundaries and conventions. This is discussed more in chapters 2, 3, and 4. 16 Other illustrations that I have given of this elsewhere (Vayda 1996: 18 ff.) include Sen’s study (1981), identifying causes of famine by shifting the focus from whole populations to the individuals who actually starved to death, and our early Kalimantan studies (Kartawinata and Vayda 1984), identifying causes of deforestation by focusing on those who actually cut down trees. Later in this chapter in my discussion of evidence, I will say more about increasing the testability of causal claims by moving to a lower level or finer grain of research.

Causal Explanation as a Research Goal: Dos and Don’ts / 15

information there.17 An illustration given by Driscoll and Stich (2008: 179–180) is that of not expecting everyday explanations of the behavior of cue balls in billiards to be improved by going into details of the interaction of individual molecules within the cue balls and the cue sticks. Another simple example is Hart and Honore’s observation (1985: 12) that, in an investigation of a man’s being shot to death, it would “usually be stupid or inappropriate” to take the trouble to establish that the man died because his blood cells were deprived of oxygen. Despite the predilection that many have for some particular statistical methods, this is something else about which there is latitude as far as obtaining explanatory information is concerned. One option is simply to make little or no use of any statistical methods. An example of this to which I shall refer again below is Haaland’s research (2008) to explain Nepalese domination of the tailoring business in a tourist area of Southern Thailand. His methods appear to have been close to the approach that Walters and I (Vayda and Walters 1999, Walters and Vayda 2009, Walters 2008), among many others (e.g., Griffin 1993, Hawthorn 1991, Weber 1949: 165 ff.), have recommended and used. This consists of putting together explanatory information in the form of causal histories and doing so by means of: 1) identifying antecedent events that we have reason to regard as possible causes of the events we want to explain; and 2) asking factual and counterfactual questions about those events in order to decide whether or not to include them in our causal histories. Besides Haaland, many others have used the 17 Advocates of these reductionist programs expect their desideratum of more fundamental explanations to be achieved by reducing the theories, presumed laws, units, and even empirical facts of one field or domain to those of another, for example, either those of social science to those of biology or those of organismic biology to those of chemistry and physics. Much has been written on such programs. Critical discussions that I have found to be illuminating include Sarkar 1998, Orr 1998, Mayr 2004: chapter 4, and various chapters in Van Regenmortel and Hull 2002. It should be noted that taking research and explanation to a lower level presents more options now than it did when relatively simple “wedding-cake” hierarchies of levels (Hull 2002: 162–163, citing Rose 1998: 10) figured in debates about reductionism. Thus, in the current quest for lower-level explanations of disease occurrence, investigations are planned or proceeding on the roles of myriad human intestinal microbiota (Gill et al. 2006; Dethlefsen et al. 2007) as well as, for example, on not only gene but also nongene DNA sequence variations that may affect disease susceptibility and disease occurrence (Couzin and Kaiser 2007, ENCODE Project Consortium 2007, Greally 2007). With regard to environmental influences on gene expression, there is research on such matters as how microscopic particles in diesel exhaust may combine with cholesterol to synergistically activate genes that trigger hardening of the arteries (Gong et al. 2007).

16 / Chapter 1

approach without much recourse to statistical methods. In fact, the methods are just about absent from Hawthorn’s 1991 work, which is the one book that I would recommend for case studies employing the approach in exemplary and highly illuminating fashion.18 However, the approach does allow for the use of quantitative data for deciding which events to consider including in our causal histories and which not. Thus, when Sahur and I were trying to explain why some but not all rice fields in a Bugis settlement studied by us in 1996 were converted to mandarin orange groves and why they were converted in 1991 and not before, simple statistical analysis of the data we had collected on sources of capital for the conversions made us realize that receiving substantial monetary compensation for land taken over by an expanding fertilizer factory was likely to have been a key event in the causal histories of all or most of the conversions (Vayda and Sahur 1996). What may seem the other extreme from making little or no use of statistical methods for obtaining explanatory information is represented by the work of social scientists, especially sociologists and economists, who have been partial to regression models and even more elaborate statistical methods for inferring causation from correlation. These methods have their uses for summarizing data and making some predictions. However, lest it be thought that there must be a superior, more rigorous, more scientific way of getting at causation, it is important to note that many of them have been criticized by both sociologists and statisticians for, among other things, the circularity of causal arguments based on significance tests and regression and the failure to truly control for confounding variables (Freedman 1991, 1997, 2005a; McKim 1997; Turner 1997; cf. Evans 1999).19 Interestingly and in line with what 18 The very disparate subject matters or explananda of Hawthorn’s three case studies include the political decisions resulting in the division of the Korean peninsula in 1945; the rise and fall of bubonic plague in Europe from the fourteenth to the eighteenth centuries; and the question of why an altarpiece for a cathedral in Siena was painted as it was by the artist Duccio in the early fourteenth century. 19 For a review of the critical literature on using statistical modeling to disentangle complexities of causation, see Freedman 2005b: 195–200. For a succinct account of problems with trying to infer causation specifically from significance tests, see Kincaid 2000b: S674–S676. Cohen (1994: 997) notes that attacks on significance testing go back as far as 1938, while Rozeboom (1997: 335) refers to it as “surely the most bone-headedly misguided procedure ever institutionalized in the rote training of science students” (cf. Gigerenzer 2004). That some who have been relying on statistical models for causal analysis recognize now some of the problems and uncertainties arising from doing so is indicated in Young et al. 2006.

Causal Explanation as a Research Goal: Dos and Don’ts / 17

I’ve been saying about multiple methods for obtaining explanatory information, Freedman, an applied statistician and perhaps the strongest critic of how regression models have been used by sociologists, holds up as an alternative to them the kind of work done famously and emulably by John Snow on the causal explanation of the 1853–1854 cholera outbreak in London. This work was marked by nothing like regression modeling but instead by detective work, logical analysis, counterfactual reasoning, and the use of considerable shoe leather to obtain the detailed quantitative and qualitative data whereby the area within which the disease was concentrated could be demarcated and the source from which the disease had spread (the Broad Street water pump, as it turned out) could be pinpointed. As part of the work, Snow also took the trouble to investigate and explain the anomalies within the Broad Street area, like a poorhouse with only a few cholera cases (it got water from its own well) and a brewery with no cases at all (the workers drank malt liquor, not water) (Freedman 1991, 1999; cf. Turner 1997, Evans 1999, Hempel 2006, Johnson 2006, and Snow’s original papers, reprinted in Snow 1936 and later editions). There is latitude also about the kinds of generalizations that may be used for obtaining explanatory information. The importance of using generalizations is perhaps underestimated by Kuznar and Long (2008: 164), who say that I seem to “admit only a secondary role for generalizations.” They may be alluding to the fact that for me, as for some others (see note 5), explanation cannot be said to have been achieved simply because observed phenomena have been subsumed under generalizations. However, with regard to using generalizations for obtaining explanatory information, my view remains as stated in the 1995 article criticized by Kuznar and Long, namely, that, on the one hand, our being able to draw on a stock of generalizations gives us head starts in developing causal explanations of the cases confronting us and that, on the other hand, finding that some of those generalizations do indeed apply in those cases enhances our confidence in our explanations by helping to assure us that we have not made ad hoc ascriptions of illusory causal connections unlike causal connections that have been discerned anywhere else (chap. 7 and Vayda 1997b: 3). The point in the present context, similar to points that have been made by others (e.g., Scriven 1959; Hart and Honore 1985:

18 / Chapter 1

15, 22, 31, 49; Elster 1989b: 8–10), is that the generalizations that we thus deploy need not be firmly established and exceptionless laws of nature or strict statements of invariant causal sequences. Instead, we may, in effect, simply be asking whether our knowledge of causal connections or sequences that have been identified or posited in certain situations or events have their parallels in what seem in some respects to be similar situations or events. Here are just four examples of research questions illustrating this with regard to diverse explananda: Asking whether the etiology of modern-day anorexia nervosa has parallels in the causes of holy women’s self-starvation in medieval times (Bell 1985); asking whether generalizations, derived from cases extending from the 1990s back to the French Revolution, about how war, sectarianism, and terrorism can be caused by the adoption of electoral politics in countries before various other democratic institutions have been established are applicable to contemporary Iraq (Mansfield and Snyder 2005, 2005/2006); asking whether causal models of soil degradation that have been found to apply in the Sahel are applicable also in other regions where rural impoverishment may be causing farmers to overuse marginal lands (Lüdeke et al. 1999); and asking whether causal models of forest fire propagation that have been developed on the basis of fire-behavior data from the Amazon can be applied to guide research in Indonesian tropical moist forests, where fire-behavior data have hardly been collected yet (chap. 2). Far from taking the form of strict statements of invariant sequences, the generalizations deployed by us may refer to sequences seen as recurrent but not necessarily invariant and may be encapsulated for us not in formal propositions but rather in platitudes, proverbs, and homely phrases or sayings. An example is “sour grapes,” encapsulating the denial that a goal which one has unsuccessfully sought was worth seeking. Elster (1983b: 110–111), who has translated “sour grapes” into psychological jargon as “adaptive preference formation,” has noted that its opposite, counteradaptive preference formation, has also found expression in homely sayings, like “Forbidden fruit tastes sweet.”

Causal Explanation as a Research Goal: Dos and Don’ts / 19

Two points are appropriate to make here about thus forming preferences and either acting or forbearing to act on the basis of them: 1) they are causal stories (or “causal mechanisms,” as Elster and others have called them) recurring in different times and places and with different actors (see Elster 1993: 2–3 and passim; also, Elster as quoted in Swedberg 1990: 247); and 2) the fact of the stories having been encapsulated in proverbs and sayings may reasonably be regarded as helping us to remember them, to recognize their possible manifestations, and accordingly to draw on them for the purpose of causal explanation. Similarly, when the generalizations we draw on are designated by somewhat more high-flown language (e.g., “the tyranny of sunk costs,” referring to persistence in some costly but unrewarding activity because of previous costly investments in it) or else, with greater scientific cachet, are designated as models or theories, the very fact that they are named and known is, as Schelling (1978: 90–91) has noted, likely to give us head starts in developing causal explanations of cases at hand. The commons model, one of Schelling’s illustrations of this (idem, chap. 3), is discussed below. Regardless of whether the generalizations we draw on are encapsulated as models, theories, or sayings, or else are perhaps not encapsulated at all, it is still the case that they need not be, actually or potentially, exceptionless laws of nature or strict statements of invariant causal sequences. Thus I refer in chapter 7 to natural selection itself as a causal story, that is, a sequence of changes whereby a trait spreads within a population by virtue of the greater reproductive success it confers on individuals who possess it. At the same time, I note that, as Darwin himself recognized (see the citations in Gould and Lewontin 1979: 589–590, Kitcher 1985a: 141–142, 179), other mechanisms than natural selection may sometimes be responsible for the spread of traits. Similarly, in cases of recognition by individuals that they have no efficacious practical means of achieving some goal, we may find a turning to magical means rather than to a “sour grapes” response. Consider also the commons model cited by Schelling. It has not only been applied to many, by-now commonplace cases of the

20 / Chapter 1

self-interested, ultimately destructive use of such common property resources as fish and game, but it has also given researchers a head start in knowing what to look for in causal explanations of some not so familiar and superficially quite different cases, such as the breakdown of a traditional Javanese communal rice-harvesting system (Sturgess and Wijaya 1983). It is, however, increasingly being recognized that there also are many cases of well-regulated use of common-property resources and that to these cases, contrary to some expectations when the commons model first became widely known in the late 1960s and early 1970s, the commons model does not apply (Feeny et al. 1990: 10–11).20 If we use models or generalizations as just described, we are expecting less from them in terms of generality than what Kuznar and Long (2008) regard as appropriate and defensible in the “youth of a research program.” Not believing that the explanatory utility of the generalizations that we deploy to particular cases depends on how general they are, we can accept from the outset that, like the commons model, they will apply in some cases and not in others and that our objective of causal explanation may be pragmatically pursued without our having to discover, develop, or employ models or theories of wider applicability.

Judging Explanations and Explanation-Oriented Research Notwithstanding all I’ve been saying about latitude, we still need to have criteria enabling us to recognize how well the goal of causal explanation has been, or is being, achieved, whether by our own research or that of others. Accordingly this section and the following ones will be devoted mostly to some important, mainly pragmatic considerations for judging explanations either good, bad, or something in between. Also to be discussed are 20 On exceptions to the tyranny of sunk costs, another of the generalizations mentioned above (although only parenthetically), see, for example, the psychological research described in Stanovich 1999: 109–115. Stanovich’s book as a whole, with its emphasis on variations, effectively counters current and recent excesses in either looking to reason, however defined, as the basis for human action (as in rational-choice theory) or looking to emotion as the basis and virtually ignoring reason.

Causal Explanation as a Research Goal: Dos and Don’ts / 21

related considerations for making judgments about explanationoriented research. If we are genuinely guided in our research by the goal of causal explanation, our decisions about what to study and when to study it should be made so as to further and facilitate our efforts to obtain evidence for or against various causal possibilities with respect to our explananda. To bring out this point, some discussion in this and the next two sections will be devoted to failures to make decisions in this way and to causes and consequences of such failures. A main concern in judging explanations has to be the evidence supporting them. There is no need to discuss explanations that fail because they are shown by evidence to be false, but explanations for which the relevant evidence is insufficient will be discussed later in this section. However, there also are explanations which, even if based on factually correct information, are unsatisfactory because they either do not address the whyquestion we’re asking and/or address it but don’t tell us anything we don’t already know. The latter, following Lewis (1986: 227), can be called “stale news” explanations. A simple illustration: If we ask why Jane is sick with the flu, we’re not going to be satisfied with the answer that she was exposed to the flu virus. That’s something we can take for granted and, presumably, we’re asking the question because we want such other things explained as how she could have got the flu when it’s not the flu season or how she could have got it despite having had a flu shot. Similarly, if we ask why Jill is crying, we’re not going to be satisfied with the answer that something made her unhappy. We’d like to know what that something is. Another stale-news explanation, for me and, no doubt, for others who have long been concerned with violence and war, is that scarcity of renewable resources “can contribute” to violent conflict (Homer-Dixon 1999: 177).21 As a critic of some of my own writing on war has said: “No one would suggest that, in some cases, 21 This cautious statement is a main conclusion from two large, multinational research projects involving some hundred researchers and advisors (Homer-Dixon 1999: xv). The projects and the publications resulting from them have been fiercely attacked by some political ecologists who have unjustifiably treated explanations produced by the projects as being more deterministic than the circumspect conclusion cited here (see Peluso and Watts 2001, 2003; Hartmann 2001; Homer-Dixon 2003; Kahl 2002).

22 / Chapter 1

people may not fight over land, or anything else which they fancy and which is in short supply.” (Hallpike 1973: 459). As for explanations unsatisfactory because of not addressing the why-question the explanation-seeker is asking, consider first a matter I address more fully in chapter 9, namely, that the explanationgiver and explanation-seeker may have different understandings of what the why-question is. An apocryphal anecdote sometimes used to illustrate this (e.g., in chap. 9 and Garfinkel 1981: 21 ff., Putnam 1978: 42–44) concerns bank robber Willie Sutton’s alleged reply to a rehabilitation-minded prison priest who asked him why he robbed banks. “Because that’s where the money is” answers a question much narrower in scope than intended by the priest, whose concern was why Sutton robbed at all instead of leading an honest life and not why he robbed banks instead of newsstands.22 Similar differences in how a why-question was understood resulted in some acrimonious exchanges between the anthropologist C. R. Hallpike and me (e.g., Hallpike 1973; Vayda 1979a) about the causes of intergroup fighting in New Guinea and, sad to say, it took us almost two decades to realize that we disagreed much less than we had thought: I was seeking explanations of particular outbreaks of fighting and therefore focused on such possible factors as land shortage at particular times in particular places, while Hallpike, more interested in such factors as an ethos of aggression and destruction among the New Guinea people he studied, was seeking explanations of the high frequency of outbreaks, whatever their causes in particular instances (chap. 9, and Hallpike 1990, pers. comm. [letter dated July 23]). More generally, the same kind of misunderstanding has occurred among participants in other debates about causes of war and has led them to accept or reject explanations as if they all were answers to the same question when, in fact, the answers were to such different questions as why some particular people fight at one time rather than another, why they fight at all rather than never fight, why they fight frequently rather than infrequently, why some particular conscious goals rather than others are held by those fighting, and why their fighting has certain consequences rather than not having them (see chap. 10 for more discussion of this). 22 On the frequent recounting of this anecdote in various contexts and Sutton’s own denial of its veracity, see Cocheo 1997.

Causal Explanation as a Research Goal: Dos and Don’ts / 23

Let’s recall also some of the examples I have already given in the preceding section about explanations either too fine-grained or not fine-grained enough to be appropriate answers to the why-questions being asked. Thus, information about molecular interactions within cue balls and cue sticks is too fine-grained to be likely to satisfy as an answer to the question of why John beat Jack in a game of billiards, while information about Suharto-era forestry policies as causes of predatory logging which altered forest microclimates cannot be fine-grained enough to explain why fire occurred in some Indonesian logged forest areas and not others in 1997–1998. Another possibility, noted by Lewis (1986: 227), is that explanation-seekers especially interested in certain parts of causal histories will not be satisfied with explanations that fail to deal with those parts. Statements made by some political ecologists seem to be expressions of this and may be accepted as such, even if some of the statements may strike us as needlessly dismissive or pejorative about explanations dealing only with other parts of causal histories. Thus, Watts and Bohle have been dissatisfied with Sen’s pathbreaking and generally well regarded famine analyses (Sen 1981, cited above) because, in highlighting the proximate mechanisms (“price movements, speculation, drought”) that trigger shifts in people’s entitlements to food, Sen neglects events further back in causal histories, that is, “the longterm structural and historical processes by which specific patterns of entitlements and property rights come to be distributed” (Watts and Bohle 1993: 48; cf. Watts 1991: 21). Robbins (2004: 38–40), referring critically to one of my articles (Vayda 1983), has gone so far as to designate such proximate-cause analyses as only descriptive and not explanatory (see Vayda 2008: 361–362, note 13, on misinterpretation of the 1983 article).23 Other such peremptory pronouncements have been made by some political

23 Yet other statements in the political ecology literature seem to ask for complete causal histories, something which, practically speaking, can never be provided. Thus, Gardner (2003: 274–275; 2005: 89–91), criticizing my and Walters’s event ecology, seems to have set himself the impossible task of providing a full account of the complex and interconnected strands of causation affecting the lives and environments of the Bedouin Arabs that he studied. If taken literally, such appeals for completeness imply that all our explanations need to go back to at least the Big Bang. See the discussion of this in chapter 3 and note the citation there from Carlyle’s classic essay, “On History” (1899, orig. 1830), on the interrelatedness of all events.

24 / Chapter 1

ecologists and will be noted in the course of my discussion of using evidence as a criterion for judging explanatory information. Turning to that discussion and continuing in a pragmatic vein, I must deal especially with the matter of judging how, or how well, any evidence presented in support of an explanation bears on the specific why-question being asked or the specific causal claim being made. Something that Walters and I have not emphasized enough until recently (e.g., in Vayda 2004 and Walters and Vayda 2009) is that an important reason for our sometimes advocating event-by-event construction of causal chains is that, problematic as the concept of events may be (see Casati and Varzi 1997, as cited in note 2), causal claims about them can still be more readily assessed in terms of evidence than can causal claims involving larger posited units or entities (e.g., “processes” or “systems”) regarded as wholes.24 In line with this point are my warnings in (chapters 8 and 9) against process reification and some telling observations made by Lewontin, namely, that the more specific a causal claim, the more susceptible it is to contrary evidence and that, by contrast, causal claims become more and more impervious to evidence as they are made about larger and larger domains of phenomena and greater numbers of complexly interacting causes (Lewontin 1994: 482–485). To make these evidential issues more concrete, consider Haaland’s recent article (2008), already mentioned above. His attribution of greater importance to “interaction systems” and their causal influence than simply to events or event chains and their influence may seem to be at odds with my own arguments. However, although he speaks of “‘systemic’ interdependencies that connect events at present” (italics mine), Haaland’s case studies seem to conform to many of my own in making events the explananda, looking for their explanations in not concurrent but antecedent events, and recognizing, even if not emphasizing, the need for evidence of the events we include in causal histories. To 24 Extreme current examples of claims of the latter kind are the many claims involving “globalization,” which, as noted by Rosenberg (2000: 3 ff. and passim), has transmogrified from being a label for certain modern-world changes that call for explanation to being freely invoked as the process to which the changes are attributed.

Causal Explanation as a Research Goal: Dos and Don’ts / 25

show, as Haaland wants to do, that certain interconnections of events are sufficiently nonarbitrary or noncontingent for them to be regarded as systemic also requires evidence of events. Indeed, if anything, it requires more of it. Besides the evidence for causal histories going back from the explanandum events under immediate consideration (e.g., the success of some particular Nepalese as tailors in Thailand in the case of Haaland’s studies) to the interconnected events of interest, evidence is needed to justify the “interaction system” label for those events. This could consist, for example, of evidence that interconnections of events of the kind in question occurred with some degree of regularity or consistency and with effects on other explanandum events (e.g., as Haaland suggests, the success of other Nepalese in other enterprises abroad).25 I myself have found evidence for “interaction systems” in this sense in the course of constructing causal histories of various explanandum events and have at times, contrary to the impression possibly conveyed by some of Haaland’s remarks, been as interested as he in describing the systems and their causal roles.26 An illustration mentioned earlier in this chapter is my explaining particular cases of Bugis migration and frontier land use by referring, inter alia, to the historically developed, geographically extensive social networks along which information about opportunities in other lands flowed back to would-be migrants in settlements either in the South Sulawesi Bugis homelands or elsewhere (see p. 13). Together with Sahur, I also made ethnographic studies of the events constituting migrant-recruiting systems in Sulawesi and migrant-receiving systems in Kalimantan at a particular time 25 Cf. Oyama 2001: 186 ff. and Davies 2001: chapter 4 on problems of system definition and dangers of promiscuity in ascription of system effects or functions. There are similar issues concerning the concept of process. On using such criteria as regularity and recurrence for designating event sequences or causal chains as “processes,” see the discussion and examples in chapter 8 and note also my references above to “generalized” causal processes or mechanisms, including natural selection. On methods for distinguishing between necessary and contingent outcomes of event sequences or causal chains, see Ben-Menahem 1997. As implied here and stated explicitly elsewhere (Vayda 2008: 363, notes 23, 24), I do not regard it useful to designate all event sequences as “processes.” 26 Haaland’s remarks that I have in mind are those invalidly suggesting an opposition between my turning to antecedent events for explanatory information and his turning to interaction systems (see, for example, Haaland 2008: 61–64).

26 / Chapter 1

(1980) with a view to using data from the studies to guide and illuminate our inquiries about possible earlier operation of the same systems or earlier forms of them as causes of Bugis settlement in our Kalimantan study area (Vayda and Sahur 1985). We did indeed find migrants in Kalimantan who had made use of such systems long before 1980. However, it is worth noting that we also found other Bugis migrants whose settlement in Kalimantan happened in a variety of other ways (Vayda and Sahur 1996). This underscores points that will continue to be emphasized in the remainder of this section, namely, the need for evidence and the need to consider alternative causal possibilities, points perhaps insufficiently acknowledged by Haaland because of his enthusiasm for interaction systems rather than simply events as causes of change.27 A credential of evidence that has been treated in some detail by specialists in evidence (e.g., Schum 1994: chap. 3 and 2003: 11, 20; Walton 2004) is relevance. Indeed, it can be said that ostensible evidence that is not relevant is not evidence at all. Accordingly, in order to use the goal of causal explanation for practical guidance in prioritizing our efforts in collecting and analyzing data, we need to make careful judgments of how relevant to the explanations we seek are (or will be) specific data that we can turn to. One way in which the failure to make such judgments and the explanatory consequences of the failure can happen is worth noting. It involves, first, decisions by investigators to accord priority not to obtaining data connecting their chosen explanandum events to possible causes but rather to obtaining data on causes of one or another of those possible causes and then, wittingly or unwittingly, misrepresenting the explanatory relevance of the data they have obtained. For a brief, concrete illustration, I draw here first on my analysis of Indonesian forest-fire research (see chap. 2 for a detailed account). 27 For caveats somewhat similar to mine here about Haaland’s predilection for interaction systems, see Hawthorn (1991: 30–31 and passim) on explanatory use of the concept of structure by some sociologists and by French historians of the Annales school and on the concomitant short shrift given by them to contingently decided outcomes. I myself have previously made similar criticisms of how Sahlins (1981, 1985) and other anthropologists have used the concept of structure (Vayda 1994: 324–325).

Causal Explanation as a Research Goal: Dos and Don’ts / 27

Something ignored by most fire researchers in Indonesia is important to note, namely, that data are lacking on which particular ignitions or types of ignitions have actually led to forest wildfires, like the extensive 1997–1998 fires that affected some six million hectares of tropical rain forest. Despite this lack, many researchers concentrated on collecting and analyzing data to explain particular types of ignitions as if success in that would, ipso facto, constitute explanation of the stipulated explanandum events, which were the forest fires themselves. Indeed the conclusions from such misdirected efforts have been used for making broad claims (e.g., in Dennis et al. 2005) about land-use or resource-access conflicts as proximate causes of arson and therefore underlying causes of the 1997–1998 fires. This is not at all warranted, especially in light of the fact that the sites for the ignition studies had not been selected on the basis of being located where extensive forest fires had occurred. The most that can be justifiably said is that land-use or resourceaccess conflicts were underlying causes of some ignitions. For that conclusion, there is evidence. Making the same evidence relevant to explaining the forest wildfires requires showing that those ignitions were causes of the fires. This simply has not been done yet. The same kinds of failures occur in political ecology when, as discussed in chapter 6, investigators with a professed interest in explaining environmental change or conservation, give priority in particular cases not to obtaining evidence connecting environmental change or conservation to its possible causes but rather to obtaining evidence on causes of just one of those possible causes, especially change in access to environmental resources. The fact that change in such access does at times (but not always) either cause or constrain environmental change is no warrant for assuming that it does so in cases at hand and for regarding evidence of its causes as, ipso facto, evidence of the causes of environmental change. Nor then should it be an excuse for failing, as have some political ecologists (e.g., those cited in chap. 6 and in Vayda 1997b: 7–10), to accord priority to obtaining relevant evidence in the form of data indicating case-specific causal interconnections of access change and change or conservation in

28 / Chapter 1

environments.28 It is on these data that the evidential relevance of data on access change depends. Even when evidence on causes of possible causes is not misrepresented as being more or less immediately relevant to explaining our designated explanandum events, a reason sometimes either stated or implied for obtaining such evidence is that eventually it will be relevant.29 A familiar way of expressing this is that it doesn’t matter where we start as long as we get to where we want to go.30 An appropriate response here: “Yes, but we can’t know in advance that we’ll get there within a reasonable period of time—or even at all.” From a pragmatic standpoint, this means that obtaining the kind of evidence indicated at the end of the preceding paragraph warrants higher priority in research to explain designated explanandum events than does obtaining evidence on causes of what are only possible and not actual or confirmed causes of such events. Yet, for the most part, such prioritizing has not been occurring. In discourse studies, for example, there has been a flood of works describing, analyzing, and explaining environmental discourse, while studies showing actual, case-specific environmental impacts of discourse remain few and far between. Consider in this regard a volume of studies of discourse contestation about India’s forests. As good as the studies are, the editors have had to acknowledge a lack of data on what differences for the forests themselves are made by all the words, changes in words, and control of words (Sundar and Jeffery 1999: 53; cf. Walters 2005c: 287 on the lack of evidence of environmental impacts or “material consequences” in Greenough and Tsing 2003, another volume of case studies centered on environmental discourse). In a more general way, a recent review article on environmental discourses makes their actual impact on 28 Conservation “events” include deliberately not acting in certain environmentally destructive ways. When not doing something is deliberate, it may call for explanation just as much as doing something does (cf. Vermazen 1985 on “negative acts” and Zimmerman 1984: 166–169 on “intentional omissions”). 29 There even have been calls for laborious, time-consuming research that takes on large domains within which causes of possible causes might be found. Examples are the calls by Sillitoe (1996, 1998a) and Ellen (1998) for prioritizing investigation of “whole” local knowledge systems in their local cultural contexts. For analysis and criticism of such calls, see chapter 3 and Eghenter 2008. 30 Note, for example, the following remark by Todorov (1993: xiii): “To take the route of discourse to gain access to the world is perhaps to take an indirect route, but it gets us there nevertheless.”

Causal Explanation as a Research Goal: Dos and Don’ts / 29

environments a main “future issue to be resolved” (Mühlhäusler and Peace 2006: 473; italics mine). Research confined to such domains as discourse, local or indigenous knowledge, and political economy can of course be valuable and important even if it produces evidence used only to explain events within its own domain and not evidence that accounts for explanandum events like the forest fires and other environmental changes mentioned above. If, however, explaining environmental changes is an important concern of ours, some alterations in the allocation and prioritizing of research effort are clearly to be hoped for.31

The Problem of Confirmation Bias Another failure, related to those just discussed and evinced by the work of some political ecologists, is perhaps the most common explanatory failure related to evidence in numerous areas of investigation concerned with accounting for human actions and/or environmental change. This is the failure to judge the evidence for any one explanation in relation to the support that either the same or other evidence provides for alternative explanations. A point important for being guided in research by the goal of causal explanation is that there can be, as a few examples will shortly show, a synergism between this failure and the failure to adhere to explanatorily appropriate and efficient data-collection priorities. Both these failures may be averted by following the explanatory strategy that Walters and I briefly described but, regrettably, insufficiently explicated in 1999 (see chap. 6 based on Vayda and Walters 1999).32 It is a strategy calling for explaining events

31 In other words, the hopes and recommendations that I am expressing here for allocating and prioritizing research effort are intended for the achievement of specific research goals involving specific explananda and are not intended to rule out or disparage research directed toward explanation of events or changes that are different—changes, for example, in discourse or politics and not, sooner or later, changes in environments. Good examples of research on causes of changes in discourse related to forest management and use may be found in Jeffery and Sundar 1999. The results of research on intragovernmental bureaucratic conflict in particular as a cause of simplification of forest-related discourse are reported in Saberwal 2000. 32 One result of the insufficient explication is that two of our critics (Watts and Peet 2004: 18, 19), evidently unfamiliar with abduction as described below, have seen our strategy as simply one of induction and “a rather crude empiricism.”

30 / Chapter 1

more by progressively eliminating possible causes than by single-mindedly applying particular generalizations, theories, or laws.33 A prerequisite to data collection when this strategy is used is identifying possible or plausible causes worthy of empirical scrutiny. The process whereby this is done, drawing on our antecedent knowledge of possibilities and involving reasoning from effects to causes, is what Charles Peirce (e.g., Peirce 1932: 495–499 and passim) and many others after him have often, although not always, called “abduction.”34 The explanatorily critical importance of the process is increasingly being recognized, especially in such fields as artificial intelligence (see, for example, Thagard and Shelley 1997), even if not yet in anthropology and geography. Once abduction in the specified sense has been achieved, data collection to obtain evidence for or against the various causal possibilities may proceed. This, for some, is part of abduction (e.g., Ahn and Kalish 2000: 209–210; Josephson 2000), while others, including Peirce, have sometimes designated it as “retroduction” (see Rescher 1978: 3, 8, 51, 65 and the citations there from Peirce). To avoid confusion and to emphasize the need to consider evidence not just for one explanation but also for alternatives, I will use the titular words of Chamberlin’s important but widely ignored 1890 article and speak of data collection guided by the “method of multiple working hypotheses.”35 Using charmingly old-fashioned language, Chamberlin, a distinguished American geologist, had put forward his method as a

33 See Scriven 1966: 250 for a similar brief characterization of such strategies. For a longer discussion with illustrations, see Woodward 1990: 232 ff. According to GodfreySmith (2003: 212), philosophers’ recognition of the importance of eliminative strategies of explanation is growing. For a concise account of John Stuart Mill’s “eliminative methods,” see Scarre 2000: 295–296. 34 On inconsistencies, including Peirce’s own, in use of the term, see, for example, Aliseda 2000: 47–49. 35 Discovery of Chamberlin’s article by many in the mid-1960s was prompted by its being prominently and appreciatively mentioned in Platt 1964 and its being reprinted in Science in 1965. After suffering some subsequent neglect, it is again being discovered and applied, even if not as widely as it deserves to be (see, for example, Railsback 2004; Pigliucci 2004: 157–161; Morad 2004: 665; and Conroy et al. 2006: 4–5). I agree with Platt (1964: 350) on the desirability of making the article required reading for every graduate student and every professor. My citations here are from the 1965 reprint in Science. For perspective on the pragmatic scientific tradition represented by Chamberlin, see Baker 1996.

Causal Explanation as a Research Goal: Dos and Don’ts / 31

way of circumventing the “habit of precipitate explanation” and “the dangers of parental affection for a favorite theory.” Here is what he says the method entails and achieves: The effort is to bring up into view every rational explanation of new phenomena, and to develop every tenable hypothesis respecting their cause and history. The investigator thus becomes the parent of a family of hypotheses: and, by his parental relation to all, he is forbidden to fasten his affections unduly upon any one. (Chamberlin 1965: 756 [orig. 1890])

For good illustrations of abduction and putting multiple hypotheses to empirical test en route to causal explanations, I would recommend, inter alia, accounts and analyses of criminal investigations and jurisprudence (Schum 2001; cf. Hart and Honore 1985: chap. 15; Walton 2004) and of medical diagnosis, especially by expert physicians (Patel et al. 2005; Magnani 2001: chap. 4). It’s true that real-life practitioners in law and medicine, unlike such master fictional abductive reasoners and hypothesis-testers as Edgar Allan Poe’s Auguste Dupin and Arthur Conan Doyle’s Sherlock Holmes (see Harrowitz 1983 and Caprettini 1983), do get it wrong at times, maybe often, because of overlooking some possible causes, like those different from the ones either usually or most recently encountered.36 Pressures, psychological as well as practical, to reach conclusions quickly must also be taken into account (cf. Ask and Granhag 2005 on the need for cognitive closure as a source of confirmation bias in criminal investigations and Groopman 2007b: 169 ff. on the occurrence of this in medical diagnoses). However, errors or bad explanations, as well as the sources of the errors in the failure to obtain and consider the evidence for additional causal possibilities, are recognized in these fields and remedies are sought. Thus there are calls for incorporating reasoning processes in medical training in order to reduce diagnostic errors resulting from heuristic biases (see, for example, Croskerry 2002, 2003; Graber et al. 2002; and 36 See Groopman 2007a, b, giving examples of such diagnostic errors by physicians and noting the correspondence of the errors to heuristics identified, analyzed, and labeled in cognitive psychology’s extensive “heuristics-and-biases” literature, well represented in the collections edited by Kahneman et al. 1982 and Gilovich et al. 2002.

32 / Chapter 1

Groopman 2007a: 41) and there are projects (e.g., Keppens and Zeleznikow 2003) to develop software with a still accessible but much larger “knowledge base” of possible causes of suspicious deaths than detectives would otherwise draw on in their investigations. By contrast, in some of the academic fields or subfields dealing with human behavior and the environment, heuristic biases leading to explanations poorly supported by evidence are encouraged and even, at times, presented as slogans or rallying cries for new theories or paradigms.37 Consider confirmation bias in particular. Centuries before Chamberlin wrote about possible ill effects of keeping one’s affections confined to a single theory, confirmation bias had already been identified by Francis Bacon, who warned against it as follows: The human understanding when it has once adopted an opinion (either as being the received opinion or as being agreeable to itself) draws all things else to support and agree with it. And though there be a greater number and weight of instances to be found on the other side, yet these it either neglects and despises, or else by some distinction sets aside and rejects; in order that by this great and pernicious predetermination the authority of its former conclusions may remain inviolate. And therefore it was a good answer that was made by one who when they showed him hanging in a temple a picture of those who had paid their vows as having escaped shipwreck, and would have him say whether he did not now acknowledge the power of the gods,—“Aye,” asked he again, “but where are they painted that were drowned after their vows?” (Bacon 1620 [Book One, Aphorism 46], cited in Nickerson 1998: 176)

Far from heeding the warnings sounded long ago by Bacon and Chamberlin and more recently by others in such fields as psychology (see the review by Nickerson 1998), new fields or subfields in recent environment-related social science have been proclaimed and propagated either in ignorance of the warnings or in defiance of them. Walters and I tried to call attention to this 37 A political ecology example of a bias as a rallying cry: Put politics first, that is, accord it “analytical pride of place” (Bryant and Bailey 1997: 5; cf. de Jong et al. 2003: 4).

Causal Explanation as a Research Goal: Dos and Don’ts / 33

previously in “Against Political Ecology” in 1999 (see chap. 6) by making the general point that fields or subfields should be distinguished on the basis of what is to be explained and not on the basis of a priori judgments, theories, or biases about what will do the explaining. However, although the point was made fairly conspicuously, it was not even mentioned by political ecologists responding to some of our other criticisms. In the case of some of the responders, like Watts and Peet (2004: 18), it may be that the gulf between their view of explanation, whereby deployment of theory is the key, and ours, rejecting that view, is simply too great for our point to have been taken seriously or even understood.38 Indeed, distinguishing fields or subfields on the basis of whatever theories or programmatic statements are thought to be magic bullets for knowing the world and acting on it is a wellestablished practice, resulting every decade or two in the ascendance of a new field to deal with factors neglected by previous ones. Thus there were discourse studies in the 1980s and political ecology in the 1990s, while in the new millennium we’ve been seeing the emergence of spiritual ecology, all of them havens for confirmation bias.39 In the case of political ecology, its champions did adopt some elements of discourse studies, but mainly they were reacting against what the political ecologist Neumann (2005: 41) has summarily described as “neo-Malthusian formulations, neo-classical economic models emphasizing the individual rational actor and technocentric approaches.” In so reacting, they, like champions of other new fields, fell into what the sociologist 38 Although I have not made it the subject of any explicit, extended discussion here, my rejection of the theory-deploying view is implied throughout this chapter. Proffering theorized explanations without making serious efforts to consider the evidence for alternatives has certain advantages—rather like what Bertrand Russell (1919: 71), in a somewhat different context, referred to as “the advantages of theft over honest toil.” Cf. Hawthorn 1991, wrapping up his estimable book of case studies with the statement that “understanding human affairs . . . turns on what is causally and practically possible . . . and cannot simply consist in deploying a theory” (p. 187). 39 Insofar as these fields or subfields have antecedents in earlier decades and persistence in later ones, my decadal distinctions here may be fairly regarded as an expository simplification. On spiritual ecology as a “new transdisciplinary field,” see Sponsel’s manifesto (2002) and responses to it by others and myself on the EANTH-L (ecological anthropology) listserv (Vayda et al. 2002). Perhaps needless to say, I am not denying that bringing previously neglected factors to the fore is a useful service. There is no problem as long as previously neglected causal possibilities are proffered as additions to previously emphasized ones and not as magic-bullet replacements for them.

34 / Chapter 1

Skocpol (1977: 1089) some years ago characterized as a “mirror image trap” whereby attempts to create something new through direct, polemic opposition to something old results in perpetuating the faulty or counterproductive methodological assumptions on which both the old and the new rest.40 The new in this case was no more conducive than the old to abductive reasoning, the method of multiple working hypotheses, and an eliminative strategy of explanation. Consider the studies in a 2001 political ecology volume on connections between violence and the environment (Peluso and Watts 2001). In their introductory chapter, the editors identify the main objects of their polemic opposition as claims of “automatic, simplistic linkages” between increases in demographically induced environmental or resource scarcity and in violent intrastate conflict (pp. 5, 13).41 To counter such claims, the volume presents studies intended to show that site-specific violence has connections to “larger processes of material transformation and power relations” as well as roots in local histories and social relations (p. 5). Accordingly the studies focus on political, economic, and discursive practices and structures regarded as constituting specific environments and different forms or patterns of violence. However, all but one of the studies present only evidence and arguments in line with the political ecology perspective. This is unabashed confirmation bias, without, moreover, any inkling from the authors

40 See also Greenwald et al. (1986: esp. 223) on other factors contributing to the perpetuation of confirmation-biased research and explanation. 41 That Peluso and Watts have made me an incidental target of their polemics (pp. 13–14 and note 17) is worth noting here insofar as it reflects the secondary role that facts and evidence play in their arguments and explanations. The statements for which I am criticized were made in the 1960s and 1970s but were recanted at length in 1989 (see chap. 9 based on Vayda 1989). That recantation is ignored by Peluso and Watts. Moreover, they attribute to me claims I never made. An example is the claim that the Marings of New Guinea were driven to war by scarcity of meat. For this, Peluso and Watts cite Vayda 1961a, which, written before I ever went to New Guinea, does not even mention Marings. In later publications (e.g., Vayda 1976: 39–40), I did say that shortage of primary forests may have driven some Maring groups to land encroachments and aggressions against other groups, but I described such forests as resource areas important for much more than only meat from hunted animals (see also chap. 9, p. 214). On Peluso and Watts’s misrepresentation of claims by Homer-Dixon (1999), who is a main target of their polemics, see note 21, above, and Kahl 2002 and Homer-Dixon 2003.

Causal Explanation as a Research Goal: Dos and Don’ts / 35

that there might be anything wrong with that. As Kahl (2002: 142) has noted in an evenhanded and methodologically astute review of the Peluso/Watts volume, the studies include few or no data on demographic and environmental changes and thus cannot—and do not even try to—assess the causal weight of political and economic structures and processes, culture, and discourse as compared to demographic pressures and environmental degradation. Similarly they cannot, and do not, give any serious consideration to possible interactions among the different sets of factors. Comparable confirmation bias is manifest when proponents of spiritual ecology (e.g., Sponsel 2001a and 2001b) and likeminded conservationists (e.g., in Ramakrishnan et al. 1998 and Lee and Schaaf 2003) attribute the relatively high biodiversity of particular forest sites to their sacredness or sacralization but fail to consider other possible causes, such as the remoteness of the sites from roads and human settlements.42 It’s likewise manifest when proponents of the Darwinian ecological anthropology or human behavioral ecology (HBE) discussed below in chapter 7 do not simply ask what factors have selected for behavior X but, in accord with their a priori “key assumption” of “ecological selectionism,” ask instead what ecological forces (resource density, competitor frequency, etc.) have selected for it (Smith 2000: 29; Smith and Winterhalder 2003: 378; italics mine). It’s manifest also when these proponents use simple optimality models (e.g., of foraging for resources) to try to answer their questions without considering that the evidence produced by testing predictions from those models is equally congruent with optimizing

42 An account of sacred groves in India’s Karnataka state (Kalam 1998) indicates that precisely their remoteness from roads and settlements best explains the relatively high biodiversity still to be found in a few groves. In one district for which there are data for an eighty-year period, the majority of sacred groves have disappeared altogether, while others have been degraded “in various ways and to different degrees” by human encroachments (cf. Bhandary and Chandrashekar 2003 on the decline of sacred groves in other Karnataka districts). In the sacred groves visited by me in eastern Indonesia, northern Thailand, and China’s Yunnan province, there was clear evidence that spiritual factors had not kept deleterious environmental changes from happening for economic and/or political reasons and even for such recreational ones as needing the site of a sacred grove for a soccer field.

36 / Chapter 1

behavior and satisficing behavior vis-à-vis the hypothesized selecting forces.43 Moreover, far from being warned against, confirmation bias is definitely present as well in the textbooks to which students and junior researchers are wont to turn for guidance on research and explanation. For example, textbooks in political ecology (Neumann 2005: 57–58, 151–152; Robbins 2004: 67, 111–113; Forsyth 2003: 34–35) approvingly cite Fairhead and Leach (1996, 1998) as having shown how afforestation and reforestation and not the deforestation alleged by government officials and forestry experts had been occurring at the hands of local people in the West African savanna-rain forest transition zone. However, the textbooks fail to mention the criticisms that Nyerges and others (e.g., Nyerges 1998; Nyerges and Green 2000) have made of Fairhead and Leach’s selective use of evidence and their failure to consider alternative causal scenarios more congruent with the evidence available (see Nyerges 2008 for more detail on this). Also not mentioned in these texts is McGuire’s 1997 analysis of evidence on the causes and context of the decline of the North Atlantic cod, an analysis that is made the basis for a cogent and incisive critique of political ecology.44 43 This point has been made by Ward (1992: 313) in criticizing optimal foraging studies by biologists. In elaborating the point, he discusses how the equal congruence with optimizing and satisficing behavior results from the failure of investigators to seek evidence of causation and from their being satisfied instead with simply modifying their models (but still calling them “optimal”) until they produce predictions congruent with actual foraging behavior (cf. Ward 1993). Although this is not the place for rehearsing the many criticisms that have been made of optimality models, it may be noted that one of the main criticisms in the past (see, for example, Hailman 1988) was the lack of adequate criteria of optimality and that, notwithstanding the arguments made by Driscoll and Stich (2008), this criticism seems to me to be as applicable to claims of optimality for human behavior, however engendered, as to claims of optimality for the behavior of other animals. For an alternative to testing predictions from optimality models as a way of obtaining explanatory information about human foraging, see chapter 7, p. 174. 44 McGuire shows, inter alia, that a consequential misreading of environmental reality (the actual state of the cod fishery) by Canadian fisheries scientists and policymakers was, most likely, a result of reasonable people’s making reasonable use of the limited information available and not, as others had claimed, a product of such factors of political economy as mechanisms for extracting surpluses from producers and the state’s favoring of certain classes of resource users over others. Relevant here is an aphorism usually attributed to Napoleon: “Do not ascribe to conspiracy that which can be explained by incompetence” (cited, for example, in Harford 2006: 191). Of course, we should not be surprised to find both incompetence and conspiracy in some of the cases to which we apply the method of multiple working hypotheses.

Causal Explanation as a Research Goal: Dos and Don’ts / 37

Confirmation Bias in Consequence Explanations In light of problematic defenses that have been proffered for using optimality models to explain human behavior (see Driscoll and Stich 2008), it’s also worth noting that there is confirmation bias when any researchers, not only the human behavioral ecologists mentioned above, take lightly what I have referred to as “the need to deal with the by-product possibility” in certain cases of explaining behavior by highlighting one or another—or some combination—of its beneficial consequences (Vayda 1995a: 227; see also chap. 9). Such explanations, called “consequence explanations” by Elster (1983b: 101 ff.; 1986a: 202), are questionable insofar as a designated beneficial consequence may be a chance effect or a by-product of behavior that occurs because of other actual or intended consequences it has.45 Driscoll and Stich’s note (2008: 190, note 5) about my being “somewhat unfair” in characterizing HBE explanations as consequence explanations misses the point that the basis for the characterization is not that human behavioral ecologists regard history per se to be irrelevant but rather that they regard historical research, including research on alternative causal scenarios, to be irrelevant. They so regard it insofar as they assume that purportedly showing beneficial fitness consequences of behavior suffices by itself to justify regarding the behavior as having been historically or evolutionarily selected to produce those consequences. On the other hand, explanations highlighting particular beneficial consequences are warranted if, like the causal explanations discussed throughout this chapter, they are supported by explanatory information in the form of causal histories. For example, the specified beneficial consequences may either be shown, or else may reasonably be assumed, to have been brought about by actions intended—by the actors themselves or by others shown to have influenced or manipulated the actors—to bring them about. In the 45 For somewhat fuller discussions than can be presented here of explanatory problems and fallacies associated with consequence explanations (including so-called functional explanations as one variety of them), see Elster 1983: 101–108 and 1986a. For my own fuller discussions of these issues, see chapters 7 and 9 and Vayda 1986: 304–305 and 1987a: esp. 502–506.

38 / Chapter 1

case of behavior that is not a one-shot event, the explanations may be warranted also if evidence is cited that the recurrent behavior has beneficial consequences because it has been fashioned by one or more selection mechanisms or processes to produce them. Preferably the cited evidence would include causal histories of the fashioning of the behavior, whether by natural selection, Driscoll and Stich’s rational practical reasoning, a combination of trial-anderror learning by some actors and imitation by others, or some other selection mechanism or combination of mechanisms. Not taking lightly the by-product possibility means making serious efforts to obtain such evidence (cf. note 51, below). Even in the absence of more direct evidence of causal histories, so-called design analysis may sometimes be used to infer that a beneficial consequence of some behavior is not just a by-product or an effect of chance but rather a result of a different kind of causal history. This is what I intended to illustrate with the New Guinea mound-building example cited by Driscoll and Stich (see my chaps. 3 and 7). The geographer Waddell had taken temperature readings from unmounded ground and from the upper parts of the large plano-convex mulch mounds constructed by New Guinea’s Fringe Enga people and had obtained such other data as measurements showing that the height of mounds and the minimum height at which sweet potatoes were planted increased with altitude (Waddell 1973: 36; 1975: 255). By means of such research, the construction and use of mounds were, I argued, shown to be such efficient and precise answers to the people’s need to protect their sweet potato crop from radiation frost—so apparently “well designed,” in other words, to deal with that problem—as to make it unlikely that mounding afforded protection from frost just by coincidence and not as a targeted result of the past operation of some unidentified selection mechanism or process.46

46 Design analysis identifying the benefit or function that traits were selected for is mistakenly equated by Driscoll and Stich (2008: 184–188) with identifying traits selected because they are optimal for producing some particular benefit or function. An illustration of my point here: We can identify a toothbrush as having been designed for the brushing of teeth without our having to claim that it was optimally designed for that function or that no better ways of cleaning teeth are conceivable. Driscoll and Stich fail also to grasp that my call for dealing with the by-product possibility applies not directly to behaviors that produce benefits (whether optimally or otherwise) but rather to particular beneficial consequences of behaviors.

Causal Explanation as a Research Goal: Dos and Don’ts / 39

In contrast to engaging in serious efforts either to make the kind of design analysis exemplified by Waddell’s work or to obtain as much useful causal-history information as possible about the behavior producing the beneficial consequences in question, those who resort to consequence explanations but are cavalier about the by-product possibility open themselves to confirmation bias by assuming that whatever beneficial consequences they have used their theories and/or methods to identify must be the consequences that the behavior was selected for. Even if they may think they are doing design analyses or obtaining necessary causal-history information, they engage in these, at best, in cursory or otherwise inadequate fashion and without duly considering alternative causal possibilities. Among once prominent cases are the many consequence explanations that Marvin Harris, with his notions of “infrastructural determinism” (Harris 1987a: 110–111 and 1994: 68 ff.; Kuznar and Sanderson 2007: 4–7), put forward some decades ago on a wide range of subjects. The subjects included any number of food taboos and preferences, which Harris professed to explain by virtue of his having found some material benefits that accrued from the practices in question— such practices, for example, as cannibalism in Aztec Mexico and aversions to beef-eating in India, pork-eating in the Middle East, and dog-eating in Europe and North America (Harris 1966, 1974, 1985). Characteristic of the explanations was Harris’s lack of serious and sustained concern with showing the causal chains or mechanisms whereby particular benefits of so-called foodways and of other practices affect their adoption, spread, or persistence and thus make those benefits other than mere by-products or chance effects (see the critiques in Vayda 1987a, b).47 47 To be noted in fairness to Harris is his eventual acknowledgment that evidence drawn from research by him and his congeners supports only a conclusion much more modest than his theorized and sweeping basic explanation of foodways as having been fashioned by selection mechanisms or processes to produce favorable balances of practical benefits over costs. The much more modest conclusion is simply that some practical benefits, like the dairy products, traction, and dung for fuel and fertilizer provided by India’s sacred cattle, do accrue from foodways regarded by other anthropologists as harmful and practically useless products of largely arbitrary cultural beliefs (Harris 1987a: 112–115 and 1987b: 62–65). This is the kind of conclusion to which some of my own studies in the 1960s led (e.g., Vayda 1961b, 1967; Vayda et al. 1961). Rappaport’s celebrated Pigs for the Ancestors (1968) was perhaps the apogee of such studies, to which the label of “neofunctionalist ecological anthropology” is sometimes attached (e.g., in Orlove 1980: 240–245).

40 / Chapter 1

A more recent example, taken from Steve Lansing’s widely acclaimed, prize-winning research on Balinese agroecology, is his claim that the centuries-long evolution of a temple-associated system whereby irrigation schedules were coordinated over wide areas is explained by beneficial consequences consisting of an optimum balance between agricultural pest control and equitable upstream/downstream water allocation in rice fields (Lansing 1991; 2000a: 210 ff.; 2000b: 311–314; 2002: 287; 2006: chap. 3; Lansing and Kremer 1993, 1996; Lansing et al. 2006: 330 ff.).48 This balance is described as one whereby, on the one hand, the neighboring farmer groups following the same schedules are enough in number to produce uniform fallow periods in large blocks of rice terraces and thus to reduce the numbers of pests by depriving them of habitats and, on the other hand, those following different schedules (downstream groups and upstream groups, for example) are enough in number to prevent simultaneous surges in the demand for the limited amount of water available for irrigation. The evidence supporting Lansing’s consequence explanation consists mainly of the following: computer simulation models showing that the system could have evolved to produce the aforesaid beneficial consequences; the results of a single entomological study in Bali (Aryawan et al. 1993), indicating that synchronous cropping and fallowing can be effective for controlling the numbers of the green leafhoppers that are efficient vectors of rice tungro virus disease (RTVD); and the results of a 1998 survey in which upstream, midstream, and downstream Balinese farmers were asked the following simple question, “Which problem is worse, damage from pests or irrigation water shortages?” Pursuant to confirmation bias, Lansing has ignored counterevidence and alternative causal possibilities. Most important is evidence suggesting that both green leafhoppers and brown planthoppers, the two main insect pests of Balinese rice fields after the Green Revolution of the 1960s, were much less a threat to the fields in earlier times, making it unlikely that the system of coordinated irrigation schedules already was, centuries ago, directed to controlling these pests. In the case of the brown planthoppers, the 48 Lansing’s 1991 book was awarded the J. I. Staley prize of the School of American Research, and his 2006 book won the 2007 Julian Steward Book Award of the Anthropology and Environment Section of the American Anthropological Association.

Causal Explanation as a Research Goal: Dos and Don’ts / 41

first major outbreaks in Indonesia resulted from the Green Revolution adoption of broad-spectrum insecticides that did a better job of eliminating the pests’ predators or natural enemies than of eliminating the pests themselves (Gallagher et al. 1994, Heinrichs 1994, and Settle et al. 1996, all cited in Vayda 1997c and Falvo 2000 but ignored by Lansing).49 As for the green leafhoppers, the one entomological study cited by Lansing notes that it was after the implementation of the Green Revolution rice intensification programs that RTVD became “one of the most serious rice diseases” (Aryawan et al. 1993: 39, citing Lim 1972 and Ou 1985; cf. Siwi 1987: 65), presumably because the programs involved staggered planting and more frequent cropping per year (cf. Chancellor et al. 1999: 121).50 The mere fact that the pest problems abated after the programs were cut back is not, as Lansing may think, evidence that the old, temple-associated system had evolved to provide an optimal solution to pest problems but, instead, only evidence that those problems had been exacerbated by the intensification programs. As for alternative causal possibilities to help account for whatever balance (not necessarily optimal) was achieved between water allocation and pest control either before or after the intensified rice cropping of the 1970s and 1980s, the possibilities mentioned by other critics of Lansing’s work include force or the threat of force by larger and stronger downstream populations in order to get water from those upstream (Falvo 2000: 645–646) and farmers’ eschewing continuous cropping of rice not simply or necessarily for the sake of pest control but, perhaps as much or more, for the sake of growing commercially profitable crops like chili peppers, garlic, and shallots after rice harvests (Helmreich 2000: 323–324, citing Nakatani 1999: 206–207). Lansing has not responded to the challenges that such suggestions present to the consequence explanation he favors.

49 Vayda 1997c was presented as the discussant’s comments in a symposium in which Lansing took part. 50 While there is hardly evidence of particular agricultural insect pests as the centuriesold problem that Lansing assumes them to be, pre-1960s occurrences of serious rice-crop damage from RTVD cannot be ruled out. In line with this, it should be noted that a rice disease called mentek, first reported in the mid-nineteenth century from Java and subsequently from Bali and other Indonesian islands (van der Vecht 1953: 48, 80), may have been RTVD (Ou 1985: 34). Available reports do not discuss whether planting was synchronous or asynchronous where mentek occurred.

42 / Chapter 1

In light of Harris’s and Lansing’s studies, as well as many HBE studies in which identifying some current utility of behavior and assuming that utility to be fitness-enhancing are unjustifiably regarded as sufficing to establish the evolutionary raison dêtre of the behavior (see p. 37 and note 45, above, and chap. 7, below), it is important to warn against being misled by a case like the New Guinea mound-building one. In contrast to it, there are many cases in which design analysis by itself does not warrant the inference that a particular beneficial consequence of some behavior occurs because the behavior has been fashioned by one or more selection mechanisms or processes to produce it and that the consequence is therefore not simply a by-product. What Orr (1996a: 468; 1996b), an evolutionary biologist, has said about design analysis in his own field can probably be said equally well about its equivalent in social and behavioral studies. One point is that, in both areas, there are cases of apparent “design” so striking as to rule out the by-product possibility. The case of butterfly mimicry, whereby a species tasty to predators closely resembles an unrelated, unpalatable species, is comparable to Waddell’s mound-building case in that, even without knowing in detail the course of selection producing the mimicry, we can rule out the by-product possibility because it is inconceivable to us that the mimicry affords protection from predators just by coincidence. So far, so good. However, design analysis has become problematic in evolutionary biology because of a penchant there for concocting adaptationist, consequence explanations of features whose alleged design is not at all conspicuous and may be illusory.51 The glory, as Orr says, lies in this and not in putting forth adaptationist explanations of obvious cases like mimicry in 51 Orr (1996a: 469 and 1996b) also notes, as have others (e.g., Pierce and Ollason 1987: 113; Fodor 1996: 252–255), that there is no good reason to expect optimal design to characterize all or most traits of organisms, since natural selection lifts overall fitness without being in the business of enhancing the design of isolated traits. Cf. Kitcher 1985a: 142, 179, notes 44, 45, cited in Vayda 1995b: 370, on Darwin’s having anticipated that unrestrained adaptationism will lead to explanations unwarranted by available evidence. For arguments to the effect that explaining traits as adaptations (i.e., as having been selected for specified beneficial consequences) calls for historical evidence of one kind or another (e.g., from comparative phylogenetic research), see Griffiths 1996 and Richardson 2007: esp. 150 ff. (but note the caveats in Davies 2001: 110–111). For a concise review of issues about adaptations and adaptationism, see Griffiths 1999.

Causal Explanation as a Research Goal: Dos and Don’ts / 43

butterflies. The same kind of point can probably be made about social and behavioral scientists’ consequence explanations of human actions. The kudos, like those which Lansing has received, come from using consequence explanations to purportedly show that actions of interest have causes other than the readily discernible ones.52 But it is precisely with regard to such explanations that we need to be most suspicious or skeptical and need most to insist on more causal-history information about the behavior in question. And should the required causal-history information be provided, the explanation ceases of course to be a consequence explanation and becomes like other causal explanations. So the message here goes beyond needing to be alert to confirmation bias in consequence explanations. With exceptions being made for those special cases where design is precise and unmistakable enough to preclude, by itself, the by-product possibility, the larger message is simply that the temptation to indulge in consequence explanations should be resisted.53

Conclusion What does it mean to engage in causal explanation and make it a goal of our research? And how should we do these things when we’re trying to explain human actions and environmental changes? These are the questions I posed at the beginning of this chapter. Let me now summarize and highlight some of the answers I have given in the preceding sections. 52 Cf. Gellner (1958: 183): “An anthropologist naturally has a greater sense of achievement if he succeeds in showing that something which seems to perform one role in fact or primarily performs another, or that something which seems to perform no role at all or one that makes sense only in the idiom of the society concerned . . . does perform an intelligible role after all.” Indeed, Harris’s reputation as “one of the most prominent contributors to 20th-century anthropological theory” (Margolis and Kottak 2003: 685) was based largely on his claims of an objective, material basis for practices that others saw only as culturally determined (see note 47, on how inadequate was the evidence cited by Harris to support his claims). 53 That temptation may be strong, presenting again the advantages of theft over honest toil that are referred to in note 38 and are cited by Brandon and Rausher 1996: 200 at the end of their critique of history-free adaptationist explanations.

44 / Chapter 1

Throughout the chapter, I emphasized pragmatism in being guided in research and analysis by the goal of causal explanation. Accordingly, I argued at the outset that addressing metaphysical questions of causal connection is not a prerequisite to going about the practical business of causal explanation. Emphasized as components of that practical business were: 1) asking why-questions about events; and 2) answering by imparting some—but not “complete”—information about the causal histories of those events. Regarding the question of determinants of what information or how much information is sought and given, I noted the salience of what the explanation-givers or explanation-seekers already know and what they still want to know. In the sense explicated in the chapter, causal explanation, whether involving the physical pressure of objects on one another or the intentions of agents, is ubiquitous in everyday life, pervasively implicated in everyday behavior, widely and importantly employed in such fields as law, medicine, and history, and applicable as well in other fields of natural and social science. The many and diverse illustrations cited in the chapter include causal explanations of events ranging from human migrations, plant invasions, forest fires, disease occurrence and spread, land-use and land-cover changes, and fish stock declines to intergroup fighting, criminal acts, automobile crashes, foraging behavior, famines, and the occurrence and prevention of insect-pest outbreaks. There is latitude not only about what constitutes explanatory information and how much of it is needed but also about the means used in research to obtain the information. This includes latitude about changing the level, grain, or scale at which research is conducted and latitude about using particular statistical methods or using statistical methods at all. And there is latitude too about the kinds of generalizations usable as sources of ideas or hypotheses about causal connections in cases at hand. Instead of being limited to established laws of nature or strict statements of invariant causal sequences, the usable generalizations encompass recurrent but not necessarily invariant causal sequences and are encapsulated for us sometimes in formal models or theories and sometimes only in platitudes, proverbs, and homely phrases or sayings.

Causal Explanation as a Research Goal: Dos and Don’ts / 45

Latitude in such matters does not, however, mean that anything goes. Being guided in our research by the goal of causal explanation means that our decisions about what to study and when to study it should be made so as to further and facilitate our efforts to obtain evidence for or against various causal possibilities with respect to whatever we are trying to explain. Accordingly, research to obtain evidence whereby some possible causes of our explananda are ruled out and others are not should have priority over research on causes or possible causes. Such prioritizing, in accord with abductive logic, Chamberlin’s method of multiple working hypotheses, and an eliminative strategy of explanation, requires serious research attention to various possible causes rather than just to those we are predisposed to favor because of our theories, heuristic biases, a priori assumptions, pressures to come up quickly with answers, or whatever. Absence of appropriate prioritizing is widespread. Some of the fields from which I cited examples are political, spiritual, human behavioral, and forest-fire ecology. Not only is research on various possible causes or causal alternatives not consistently prioritized. Sometimes—possibly often—it does not take place at all. A result is what I referred to as perhaps the most common explanatory failure related to evidence in numerous areas of investigation concerned with accounting for human actions and/or environmental changes: the failure to judge the evidence for any one explanation in relation to the support that either the same or other evidence provides for alternative explanations. At least in some cases and perhaps in many, this failure is rooted in the problematic view that the key to explanation is deploying theory or generalizations rather than considering and progressively eliminating possible causes. A corollary of this has been the view that finding data consistent with one’s favored theory is sufficient empirical support for an explanation even if little or no attempt has been made to obtain and consider the empirical support for alternative explanations. Although warnings against such confirmation bias have been sounded for centuries, explanations in new fields or subfields in recent environment-related social science—political ecology and spiritual ecology, for example—have been promulgated and

46 / Chapter 1

enthusiastically received either in ignorance of the warnings or in defiance of them. This is a contrast to other fields, like criminal investigations and medical diagnoses, where measures are taken to reduce bias-induced errors or bad explanations. Also noted as unsatisfactory because of failing to consider alternative causal possibilities were certain cases of explaining behavior by highlighting one or another—or some combination—of its beneficial consequences. So-called consequence explanations are acceptable when the behavior or actions we want to explain result in so efficient and precise a solution to a problem faced by the actors as to make it inconceivable that the behavior has not been fashioned by some selection mechanism or process to produce the beneficial consequence. In such cases, we may be justified in ruling out the possibility that the consequence is simply a chance effect or a by-product of behavior that occurs because of other actual or intended consequences it has. However, in many cases where the by-product possibility should have been taken seriously, it has not been: the researchers, conforming to confirmation bias, have simply assumed that whatever beneficial consequences they have used their theories and/or methods to identify must be the consequences that the behavior was selected for. The examples I discussed include explanations of food taboos and preferences and Balinese irrigation scheduling. In these and many other cases, the temptation to obtain easy answers by indulging in consequence explanations should be resisted and pertinent explanatory information in the form of causal histories needs to be sought. Other ways in which explanations fail or, at least from a pragmatic standpoint, are unsatisfactory were noted and exemplified. Thus they may be unsatisfactory because of either not addressing the why-question being asked or addressing it but without telling us anything we don’t already know. The former sometimes occurs when the explanation-giver and explanationseeker have different understandings of what the why-question is. This underscores the importance of being clear about our questions, both in giving explanations and, earlier, in conducting research to answer them.

Causal Explanation as a Research Goal: Dos and Don’ts / 47

Related to this and also noted were important reasons for presenting explanations in terms of events and their connections rather than making causal claims involving larger posited units or entities (e.g., “processes” or “systems”) regarded as wholes. Among the reasons, well stated in an article by Lewontin, are that the more specific a causal claim, the more susceptible it is to contrary evidence, and that, by contrast, causal claims become more and more impervious to evidence as they are made about larger and larger domains of phenomena and greater numbers of complexly interacting causes. A corollary is that making a causal claim more specific makes clearer what studies are needed to produce evidence either confirming or disconfirming it. The account given of causal explanation, both in the preceding paragraphs of this summary and in earlier sections, makes it very different from the purportedly rigorous “laws-and-causes social physics” mentioned earlier and rejected by many social scientists. The pragmatically oriented causal explanation of events described here and elaborated and illustrated in subsequent chapters can, as a goal, be an effective, user-friendly guide to productive research and sound analysis advancing our understanding of human actions and environmental change. I close here with the hope that this will be more widely recognized and acted on.

Acknowledgment For advice helpful specifically for my deciding what to change and what to leave unchanged in drawing on Vayda 2008 for preparation of the present chapter, I thank Brad Walters, Harold Kincaid, Paul Roth, Ed Carr, Gene Anderson, Annette MuehligHofmann, Raj Puri, and Susan Lees.

2 Both Ends of the Firestick: Causal Explanation of Indonesian Forest Fires

My object in this chapter is not so much to present causal explanations of Indonesian forest fires as to discuss some things wrong with such explanations in the past and some ways in which they might, in the future, be made better as explanations and, concomitantly, more useful for fire management. I will give examples not only of problematic explanations but also of fire-research and fire-management recommendations and programs made problematic by faulty or unsubstantiated causal assumptions or explanations. Because the tropical moist forests that are repositories of exceptional biodiversity are thought to be in danger of disappearing, they are the forests I will focus on. The research I will draw on for illustrations will consist of both my literature searches on the fires and on their explanations and, to a more limited extent, the fieldwork I conducted on these subjects in the province of East Kalimantan in collaboration with Ahmad Sahur of South Sulawesi’s Hasanuddin University in 1998, 2000, and 2001. The main earlier publication of my findings was in my 1999 World Wide Fund for Nature (WWF) report (Vayda 1999) on causes of the 1997–1998 fires that affected some 6 million ha of tropical

49

50 / Chapter 2

rainforest.54 Although my findings since 1999 were in accord with my earlier ones, my interpretation of the findings changed in certain important ways. Accordingly my discussion here makes certain conceptual and methodological distinctions in line with the goal of causal explanation—distinctions which, regrettably, I either did not make at all or made only inadequately in my 1999 report. The changed interpretation of my findings will be used to support, illustrate, and elaborate arguments I have been making (e.g., in chaps. 1, 3, and 4 in this book and in Vayda, 1996, 1997b) in favor of research guided by the goal of causal explanation of concrete events.55 I will be especially concerned here with explicating and illustrating what being so guided implies for setting research priorities, collecting evidence, and crossing disciplinary boundaries. As well, I will be concerned with showing that the results of the kind of research I am advocating can be drawn on to increase the effectiveness of policies and programs for fire management. Systematically showing the generalizability of my methodological arguments is beyond the scope of the present chapter, but the arguments are indeed intended to be generalizable and, as such, should be of interest even to human ecologists and other social and ecological scientists who are not especially interested in causal explanation of forest fires. While eschewing in this chapter any extended philosophical discussions of the more general methodological issues and ways of dealing with them, I do cite some references to such discussions, especially in my footnotes. As in my 1999 report (Vayda 1999: 2–3), my focus here will be on causes of fires in either primary or selectively logged but still presumably biodiversity-rich tropical moist Indonesian forests for

54 Estimates of the affected area vary, depending, at least partly, on varying definitions of forests. See the discussion of estimates in Tacconi, 2003: 2–5. Factors probably contributing to the confusion are that smoke haze and its economic and health consequences were emphasized in media and Internet coverage of the fires occurring during the El Niño drought in 1997–1998 and that the facile assumption made by many journalists, government officials, and the lay public seems to have been that whatever produces smoke haze must be a forest fire. In fact, much of the smoke haze—possibly most of it (Tacconi, 2003: 5)—came not from the burning of forest trees but rather from the burning of peat. 55 “Concrete events” in my usage here, as elsewhere in the volume, include human actions as well as such events as forest fires.

Both Ends of the Firestick / 51

which there remains some hope of conservation. Occurring on mineral soils, most of these are what are generally designated as dipterocarp or mixed dipterocarp forests. My focus will preclude much attention to not only the sources of the serious problem of smoke haze in 1997–1998 but also to fires in Indonesian forests, whether presumably biodiversity-rich or not, already officially zoned and sanctioned for conversion to oil-palm and other plantations. Nor, for the most part, will I be concerned with causes of fire in the peat-swamp forests that, along with the peat itself, burned in parts of Kalimantan and Sumatra in 1997–1998. Failing to make these kinds of distinctions (see, for example, Stolle et al. 2003) obscures or relegates the more specific questions that need to be dealt with for explaining and controlling wildfires in the forests on which I am focusing (cf. Tacconi 2003, discussing the different policy problems associated with fires under different conditions).56

Ignition Studies: Problems, Needs, and Applications Arguably the most consequential distortions of research priorities in attempts to explain Indonesian forest fires have been the undue attention and the wrong kind of attention accorded to the study of ignition events and their causes. Many studies of the 1997–1998 fires simply did not make the crucial distinction between explanations of the start of fires and explanations of their spread and, accordingly, made no attempt to identify and prioritize research needed for the latter in particular. These failures are reflected in the jumble of factors listed together as causes of fires in reports resulting from some of these studies, including, I am somewhat embarrassed to admit, my own WWF report (Vayda 1999), as well as articles and reports resulting from a project whose progress I followed during several stays at the Center for International Forestry Research (CIFOR) headquarters in Bogor, namely, the CIFOR/ICRAF/USDA Forest 56 For calculation of the forest remaining in three classes (high-, medium-, and lowdensity canopy) in Kalimantan, Sumatra, and Sulawesi as of 2003, see Tacconi and Kurniawan n.d.).

52 / Chapter 2

Service project on “Underlying Causes and Impacts of Fires in South-East Asia” (e.g., Applegate et al. 2001; Colfer 2002; Dennis et al. 2005; Suyanto 2000; Tomich et al. 1998, 2004).57 Such listed causes as arson, the facilitation of access to resources, and the clearing of land for swiddens and plantations clearly pertain to what fires are started for. They are causes that explain ignition events. On the other hand, other listed causes, such as changes in forest microclimates and the build-up of fuel loads and both intensive logging and specific forestry policies as causes of the microclimate and fuel-load changes, clearly pertain to the spread of fires. These reports do at least take note of causes of fire spread, even if they indiscriminately lump them with causes of ignitions as explanations of forest fires. There have also been a few studies (e.g., Steenis and Fogarty 2001) explicitly concerned with the problem of explaining fire spread in Indonesian forests as a different problem than simply explaining ignitions. Further reference to such studies will be made in due course. However, some other reports and analyses (e.g., Abberger et al. 2002; Varma 2003) consider only causes of ignitions and evidently assume that the understanding or explanations achieved by so doing suffice for making recommendations for controlling forest fires. Far from being peculiar to Indonesian forest-fire studies and policymaking, such assumptions underlie fire-management strategies set forth as general recommendations by leading conservation organizations (e.g., IUCN and WWF 2001). What then, more precisely, is problematic about the fire research and the fire-management recommendations I am referring to? If what we want to explain and manage are indeed forest fires and not simply ignition events, certain points need to be kept in mind: 1. Not all ignitions lead to forest wildfires, defined here as uncontrolled burning in forest areas.

57 ICRAF is the acronym for International Centre for Research in Agroforestry, now called the World Agroforestry Centre.

Both Ends of the Firestick / 53

2. It follows that, as far as the objective of preventing or limiting forest damage or destruction from wildfires is concerned, our focus needs to be on forest fires and not ignitions per se as our primary objects of explanation and control.58 3. In order to make ignition studies more relevant to explaining and controlling forest fires, we need: a) studies reconstructing the ignition events that have led to particular forest wildfires; and b) analyses of findings from such studies in order to ascertain whether forest wildfires are more likely to result from ignitions either for some particular purposes or by some particular types of actors. 4. If and when the possible ignition sources are seen to be many and/or difficult to monitor and control, causal explanation and control of wildfires in tropical moist forests may require priority attention not to ignitions but to changes in fuel loads, decreases in moisture, and similar factors affecting forest flammability, notwithstanding that a convention followed by many fire scientists (e.g., Stolle et al. 2003: 278–279; Stolle and Lambin 2003: 376) is to designate these not as causes of forest fires but as “predisposing” factors or conditions.59 58 Methodological injunctions here are ones generally applicable in research guided by the goal of causal explanation. Fundamental is the injunction to be clear and focused about one’s explananda, that is, about what one is trying to explain. This is discussed explicitly in Roberts 1996: 105 ff., as well as being noted or implied in other chapters of the present volume. A corollary is not to be sidetracked in one’s research and analysis from trying to explain the designated explanandum events, like forest fires, to trying to explain other events, like particular ignitions or types of ignitions, before their causal significance with respect to the designated explananda has been seriously or sufficiently probed. Although I have criticized others for losing sight of this corollary in their attempts to explain environmental change (see especially chaps. 1 and 6 in this book), I myself, as suggested in the mea culpa above, lost sight of it at times in the course of my 1998–1999 research and writing on causes of forest fires. 59 This convention, which I do not regard as analytically useful, may reflect an everyday view of explanation whereby particular triggering events are favored over particular conditions, or changes in conditions, as causes and explanatory factors even if, as will be discussed later, counterfactual analysis indicates that occurrence of the explanandum events in question (e.g., forest fires) depends more on the particular conditions than on particular triggers. The methodological and semantic issues raised here are discussed in a general but pragmatic way by Hart and Honore 1985: 71–73 and Miller 1987: 60–61. In both works, the example of fires as explanandum events is cited, and counterfactual reasoning is used to support arguments in favor of sometimes including conditions as causes in our explanations.

54 / Chapter 2

The above points may be discussed seriatim. The first of them may seem so obvious as to require little further consideration or documentation. Still it may be noted that an analysis of data on East Kalimantan hotspots (Steenis and Fogarty 2001)60 suggested that only a small proportion of the presumably very large number of small and isolated fires lit before and during the “major outbreaks of 1998” led to large and damaging wildfires (Fogarty et al. 2001: 2).61 This would seem to argue that, if ignition events are to be studied, priority should be given to those from which wildfires are known to have or are at least likely to have eventuated. However, despite the fact that the impetus for the CIFOR/ICRAF project and its funding was clearly the disastrous 1997–1998 fires and the disabling and financially costly smoke haze associated with them (see Dennis et al. 2005: 466 ff.), project participants are explicit in stating that their eight research sites in Kalimantan and Sumatra “were not selected to document causes of the most extensive fires of the late 1990s, or the fires that generated the most smoke.” Instead, the rationale for site selection was as follows: Due to the importance to this study of information drawn from interviews and narratives concerning sensitive topics related to fire, and on land use mapping by members of local communities, our selection of sites prioritized locales where we believed credible information would be available from research partners. (Dennis et al. 2005: 472)

In other words, as far as the information to be obtained was concerned, its likely availability is what counted for site selection, while its likely explanatory import for the 1997–1998 wildfires or its pertinence to ascertaining significant causes of those fires did not count. This goes against the second of my four points above, and it should be no surprise that, for explanatory

60 Hotspots are “High Temperature Events” (HTE) detected by NOAA satellites and indicating fire activity. Although fire detection by such means is far from foolproof (see Flasse and Ceccato 1996), hotspot data can still be used to identify significant spatial patterns in fire occurrence and spread. For a fuller description of the East Kalimantan hotspot data used in their analysis, see Steenis and Fogarty 2001: 5. 61 But see below on the lack of quantitative data concerning ignitions.

Both Ends of the Firestick / 55

purposes, the information obtained at some of the sites suffered from the sorts of limitations to be discussed later in this section. It’s all very well to decry, as have Dennis et al. (2005: 468, citing Harwell 2000), the analysis of fire hotspots in isolation and without “a textured understanding of social landscapes and the role they play in creating fire hazards.” Such understanding should certainly be sought. But it’s no advance to seek it without first trying to establish, through hotspot analyses or other means, whether there is some correspondence between the social landscapes selected for study and zones of both ignition and spread for the extensive wildfires that we want to explain—and, in the future, to prevent. Perhaps fortuitously, the sites chosen for study in the CIFOR/ ICRAF project included a Sumatran one where a wildfire in October 1997 had indeed burned between 100 and 500 ha according to the disparate estimates of officials and local people. However, investigations such as those described in the following paragraphs were not conducted at the site and, accordingly, simply asking about ignition sources for the wildfire proved inconclusive. As often happens in such cases, officials blamed local farmers’ slash-and-burn land-clearing methods, while the farmers themselves mostly blamed illegal loggers’ unextinguished campfires and cigarette butts (Suyanto 2000). What then are better means for identifying ignition sources and ignition zones? This question takes us to the third of the above points. Important means are the forensic fire-scene investigations routinely undertaken in some countries, like Australia, after forest fires. Such investigations were, in fact, strongly recommended for Indonesia in an unfortunately little publicized 1998 Working Paper of the Bappenas Planning for Fire Prevention and Drought Management Project (Bappenas [National Development Planning Agency of Indonesia] 1998: 36): To build a data base of fire causes [i.e., causes of ignition events leading to forest fires] requires some form of investigation into the circumstances surrounding individual fires, including on site investigation if practicable and assignation of causes. Investigation needs to commence as soon as

56 / Chapter 2

possible after ignition to ensure that clues and pointers to ignition sources and the reasons behind those sources are preserved. Clearly, such a course of action is impractical for the 1997/98 Indonesian fire event. There was no formal fire investigation process or structure evident, apart from the observation of fire on particular lands either by direct observation or by use of remotely sensed images. The fire path and pattern of fire on a particular parcel of land may give some clue as to the precise point of origin of the fire and even to the base cause. Unless those clues are supported by detailed evidence arising from investigation, admission of deliberate lighting by the perpetrator of the fire, or by eyewitness accounts, it is very difficult to accurately ascribe cause. . . . Because individuals or organisations who have a vested interest in lighting fire are almost always unwilling to admit causing them, the only way to establish cause accurately is to investigate as many fires as possible, while the fire activity is current. That is the only way to obtain proof.

April 1998, when this Bappenas paper was produced, was the last month of the 1997–1998 fires. However, the fires of subsequent years, although not as severe and widespread, still provided opportunities for putting into effect the fire-scene investigations recommended in the Bappenas paper. By and large, the opportunities were not taken. Cost and personnel considerations would no doubt have precluded carrying out these investigations for a large number of Indonesian forest fires. However, in 2006, all of Indonesia was reported to have a grand total of only two individuals who carried them out at all (Chew 2006). In my discussions at such organizations as CIFOR and WWF, forestfire researchers expressed little or no interest in either pursuing fire-scene investigations themselves or otherwise finding or providing support for them. The impression I got is that many of the researchers either simply accepted the old ways of making surmises about how fires start or else were not aware of alternatives to them, despite the existence of a substantial professional literature on the conduct of forensic fire-scene investigations and

Both Ends of the Firestick / 57

fire-path reconstructions.62 This overlooked literature includes detailed discussions pertaining to forest fires in particular and goes into such matters as how the directions from which particular fires have spread may be determined from the greater depth of charring of tree trunks and the greater heat discoloration of rocks on one side than the other and also what to look for as evidence of arson once the ignition sites have been ascertained (DeHaan 2002: chap. 8; Ford 1987, 1995). Once the decision had been made in projects like the CIFOR/ ICRAF one to study causes of extensive forest fires without prioritizing research sites where such fires had occurred, efforts could still be made to seek data on whether uncontrolled fires, extensive or not, resulted from identified ignition events at the chosen sites. The CIFOR/ICRAF project did, in fact, obtain local people’s reports and some other evidence of escaped fires in study sites other than the Sumatran one noted above. However, the more reliable reports referring to arson as a weapon in conflicts over land use or resource access were concerned with the spread of fires to areas either already set aside or already actually used for oil-palm or timber plantations. Accordingly, these reports are limited in their explanatory import for the fires on which I and others are focusing, that is, the extensive 1997–1998 fires in tropical moist forests which there is still some hope of conserving for, inter alia, their biodiversity value. Using such reports, as has been done in the CIFOR/ICRAF project, to make broad claims about land-use or resource-access conflicts as underlying causes of the extensive fires is not warranted. Even if it can be justifiably said that such conflicts were underlying causes of some ignitions, whether those ignitions contributed to the extensive forest fires of explanatory interest to us has simply not been shown yet. A methodological argument supported by the 62 My 1999 report (p. 30) does refer to forest rangers going on motorcycle patrols no more than 6–7 km from their guard post in an East Kalimantan nature reserve and then simply noting where fires had occurred and, on the basis of quick visual inspection of the sites, deciding what preconceived category of causes, like unextinguished cigarette butts, to assign the fires to. Such cursory exercises do not constitute bona fide forensic fire-scene investigations.

58 / Chapter 2

failure in this regard is that, for causal explanation of wildfires and not just of ignitions, it would have been more efficient and productive to begin research where extensive forest fires had recently occurred and to work backwards in time to seek data on where, how, and why the fires had started.63 As far as controlling and not simply explaining forest fires are concerned, it might be thought that research such as the CIFOR/ ICRAF project would still have some useful effect in pointing to the number and variety of ignition sources that may need to be dealt with. The sad truth, however, is that research in general seems, at least sometimes, to have been accorded low priority in decisions about recommending fire-management programs and putting them in place. Thus, recommendations have been made for solving Indonesia’s forest-fire problem by curtailing or even eliminating land-clearing fires (e.g., Varma 2003, criticized by Tacconi and Vayda 2006; Guyon and Simorangkir 2002) as if it had been well established that ignitions other than those for clearing land are insignificant or inconsequential. It may be true that research on other possible ignition sources has not firmly demonstrated their role, but that’s true also of research on the land-clearing fires as a source. Clearly what need to be prioritized then in ignition studies for fire management are the kinds of studies and analyses recommended in point 3 above. Of course, in Indonesia as elsewhere, there often is pressure to put management programs into operation without waiting for research to produce results or even to be undertaken. No doubt this happened when fire was high on both political and donor agendas during and right after such fires as those of 1997–1998 in Indonesia. There were what in other contexts have been called privileged solutions (Adams and Hulme 2001: 18–21; Moris 1987: 99f.) which, because of the gravity of the problem at hand, are thought to require only the resources and will to act. In the case of the Indonesian forest fires, curtailing land-clearing fires and promoting community fire management are examples of such solutions. The latter merits some discussion in order to 63 For comparison of “forward” and “backward” causal inquiry, see Einhorn and Hogarth 1987.

Both Ends of the Firestick / 59

show that a privileged solution may be substantially wide of the mark and that decisions on the amount of resources expended on it might have been quite different if higher priority had been accorded to getting answers to certain research questions. Community fire management was already prominent in plans formulated in the wake of extended fire and smoke episodes in Indonesia in the 1980s and 1991 (Goldammer et al. 2004: 383–384). By 2002, more than sixty villages in East Kalimantan had received “basic fire management training” under a project designed by the German Agency for Technical Cooperation (GTZ) (Abberger et al. 2002: 57).64 For some villages, training was carried further and, with a focus on institutionalization of fire management at the village level, included the development of volunteer village fire brigades or fire-management crews. These crews were charged with preventing and suppressing wildfires in the village and promoting “safe burning practices in slash-and-burn agriculture” (idem). All this appears to have been undertaken and achieved without any empirical justification for pushing fire-management efforts much more in villages than elsewhere. In fact, certain conclusions from the previously cited analysis of data on East Kalimantan hotspots (Steenis and Fogarty 2001) should have pushed efforts away from the villages. Part of the analysis pertained to a main 5.3 million-ha study area, for which 48,300 hotspots were recorded in the 129 days from January 6 through May 15, 1998.65 According to Steenis and Fogarty (2001: 7), this area was “systematically and manually scanned through visual interpretation” to find clusters of equal-dated hotspots constituting zones with spatial and temporal connectivity. Some of these could be identified on the basis of temporal considerations as “fire ignition zones” (FIZ) within which the fires started, while others could be identified as zones to which the fires spread during the 129 days

64 This “Integrated Forest Fire Management Project” (IFFM) was implemented jointly by GTZ and various Indonesian government agencies and services. 65 The hotspot database was built by Anja Hoffmann and Lenny Christy of the GTZ– IFFM project. January 6 was the day the 1998 fires started in East Kalimantan.

60 / Chapter 2

analyzed. The following are some of the conclusions reached about FIZ (Steenis and Fogarty 2001: 19, 27): 1. The greatest number of FIZ was in heavily logged over forest, which also comprised the largest area within the FIZ. 2. Logging roads were present in 76 percent of the 258 FIZ identified in the main study area. 3. More than 50 percent (possibly as high as 75 percent) of settlements or villages were more than 5 km away from FIZ. 4. Since hotspot analysis could identify only the zones where fires started and not who started them, it remains possible that some of the fires in FIZ more than 5 km from villages were started by village-based forest users; however, the distances involved make it unlikely that starting many of the fires in these FIZ was directly connected to routine village agricultural activities. Some caveats are in order. The first is that I am arguing not against community fire management per se but only against privileging it as a solution without having evidence from such fire-path and fire-source investigations as I have been recommending and without paying attention to such ignition-zone studies as Steenis and Fogarty’s. Another caveat is one which Steenis and Fogarty themselves sound in their report (pp. 2–3), namely, a warning against extrapolating the results of hotspot analyses from one region to another, for example, assuming, as many did during the 1997–1998 fires, that what was found in parts of Sumatra with respect to the concentration of hotspots in areas allocated for oil-palm plantations (see Bowen et al. 2001) must be true also in East Kalimantan, notwithstanding substantial differences in land-use patterns between East Kalimantan and Sumatra at the time of the fires. Similarly, even with respect to specific regions or areas, we must beware of extrapolating findings from one time to another without considering the possibility that things have changed. For example, with the spread of oil-palm plantations in East Kalimantan, future hotspot analy-

Both Ends of the Firestick / 61

ses for East Kalimantan might indicate clusters more like those found for parts of Sumatra at the time of the 1997–1998 fires than like those found by Steenis and Fogarty. As contexts or conditions change, so do fire ignition zones and causes of wildfires. Such caveats are consistent with my calls for research guided by the goal of causal explanation and should, in fact, serve to reinforce them. Consider then what might have been appropriate ways to proceed with the planning and implementation of fire-management programs for the times and locations to which Steenis and Fogarty’s analysis pertains. Their final conclusion (p. 19) is a recommendation that those involved in various ignition-producing activities in identified FIZ should be especially targeted in firemanagement programs, but they say nothing about how that should be done. I would amend their recommendation to say that assessments of the costs and feasibility of reaching particular categories of people (those whom officials regard as illegal loggers, for example) involved in ignition-producing activities would need to be made before deciding to target them or not. However, I certainly agree with a main implication of Steenis and Fogarty’s report: that targeting decisions should be based on more and better research. In fact, despite large gaps in research, there has been no lack of information indicating that such FIZ as have been identified by Steenis and Fogarty could have been the sites of a variety of ignition-producing activities other than agricultural landclearing by community-based small farmers. An admittedly partial list of other possible causes of ignition includes: arson; the facilitation of access to resources; loggers’ and other forestusers’ need for fire for cooking, insect repulsion, and nighttime warmth; long-burning underground coal seams extending to or near ground surface; and loggers’ use of equipment subject to accidental ignitions, for example, engine exhaust or friction heating fires in bulldozers, trucks, chain saws, and other logging machinery and also ignitions of gasoline spillage or overflow when chain saws are refueled. Cigarette smoking is another possible cause that cannot be completely ruled out, even though claims for its efficacy in causing forest wildfires have little empirical

62 / Chapter 2

support.66 These various causes of ignition are among those noted in Steenis and Fogarty’s report (p. 23) and a number of other reports, including my own in 1999. Not only could all of these ignition sources have been present in Steenis and Fogarty’s FIZ, but there were reasons for the presence of some of them in areas of logged forest rather than elsewhere. Obvious examples are accidental ignitions involving logging equipment. Among other good examples are ignitions for loggers’ cooking fires and campfires and for burning slashed vegetation to facilitate illegal loggers’ access to relatively fire-resistant and commercially valuable Borneo ironwood trees. As for arson, there have been plausible, even if unproved reports that it has been committed in logged forest areas by agents of timber concessions or plantation companies seeking government approval either for converting the areas to plantations or for moving on to new areas for timber harvesting. There also have been reports of arson by illegal loggers for the purpose of either destroying evidence of their activities or diverting attention from them. And even if land-clearing for relatively small-scale agricultural operations was among the possible causes of ignitions in Steenis and Fogarty’s FIZ, the clearing would not necessarily have been done by community-based small farmers who would have been reached by projects like the aforementioned GTZ one. Instead it could have been done by hired hands or would-be sharecroppers new to the area and recruited by land speculators intending to claim the land by virtue of having had it cleared. In what is now East Kalimantan’s East Kutai district, Sahur and I met and interviewed a number of such entrepreneurs in 1998 in an area where land speculation was rife because of rumors that a 66 Neither I nor any of the numerous fire specialists I have consulted have succeeded in finding any reliable studies or good data showing cigarette butts as ignition sources for actual forest fires either in Indonesia or anywhere else. According to one authority (DeHaan 2002: 139, 527), cigarettes have “been blamed in many more instances than they should” as sources of ignition even for fires in buildings. However, an experiment conducted in 1964 (Ford 1995: 105–106, 166) and another conducted in response to my inquiries in 2004 (Gönner, pers. comm. [2004]) indicated that unextinguished cigarettes may ignite wildland fires if certain special conditions are met, for example, if relative humidity is below 18 to 22 percent and if at least one third of the smoldering cigarette’s surface is in direct contact with fine fuels.

Both Ends of the Firestick / 63

large coal-mining company was planning to expand to new locations and would pay compensation to those with claims to the land there. When we returned in 2000, we found also that some of the entrepreneurs were establishing official “Farmer Groups” (Kelompok Tani), made up mostly but not entirely of fictitious members, to gain access to parcels of forest land ranging in size from 20 ha to 100 ha. The entrepreneurs were intent mostly on making speculative gains from the parcels and obtaining credit and subsidies from the government, although one of them did profess a desire also to enhance his status as a traditional patron by allocating a part of an 80-ha farmer-group parcel to those who had paid some small membership fees to join the group on the promise of profit from land speculation. In any event, neither the entrepreneurs, living in villages or small towns at least 20 km from the forest land being converted to farmland, nor the noncommunity based hired hands or would-be sharecroppers using fire to clear the forest land, would have been reached by community fire-management projects.67 Whether practices like these occurred widely in Steenis and Fogarty’s FIZ is not known. However, simply making the point that they occurred, or could have occurred, in some of them suffices to bolster my arguments about needing more research on ignition sources in FIZ before community management aimed at controlling fires in village farms and settlements is privileged as a solution to Indonesia’s forest-fire problems.68 Ideally there would be forensic fire-scene investigations to provide guidance on who should be, and practically could be, targeted in fire-management programs. If such investigations were too costly or otherwise too difficult to be carried out, priority research could consist of hotspot analyses to identify FIZ, followed by systematic consideration of possible ignition causes in the FIZ. 67 In Vayda 2000a, I give a more detailed account of speculative forest-clearing and the Kelompok Tani. 68 For a historical parallel, consider what happened in the wake of the Great Fire of 1871 in Michigan: Remedies were ill-conceived, focusing on fire prevention in villages and towns rather than in the surrounding timberlands where the fire had actually originated, having been caused by “unsafe lumbering practices” (Kreger 1998).

64 / Chapter 2

Studying Fire Susceptibility and Fire Behavior: Problems, Needs, and Applications The preceding section was concerned mostly with priorities among ignition studies. However, for causally explaining fires in tropical moist forests and for making the explanations practically useful, the question of the priority to be accorded to certain nonignition studies is at least as important. My previously stated point 4 concerns priority attention not to ignitions but to fuel loads, decreases in moisture, and other so-called predisposing conditions if we are trying to explain and control forest wildfires. Jeffrey Sayer, former director general of CIFOR, responding by e-mail in 2001 to a short piece about my forest-fire research (Vayda 2001), argued that, if people are about, there will always be enough ignition events to cause forests to burn if the forest vegetation has become susceptible to fires.69 It follows from this that, at least for Sayer in 2001, the causes or causal histories most important for purposes of explanation, and of making explanation useful for policymaking for fire-management, were the causes or causal histories of changes in forest flammability (see Toma 1999; Toma et al. 2000; Siegert et al. 2001, on some such changes) rather than the kinds of triggering events and their antecedents considered as causes in the preceding section concerning ignition studies.70 The counterfactual reasoning here is that if the burning of susceptible forests does not result from some particular ignition events, it will result from others, whereas there would be no forest fires 69 Cf. Swetnam and Baisan 1996: 29 for a somewhat similar view of ignitions in forests of the American Southwest prior to 1900. However, these ignitions often were caused by lightning strikes, whereas in tropical moist forests the heavy rainfall usually associated with lightning strikes limits their efficacy for ignitions. Regarding Asia’s tropical lowland deciduous forests, Stott et al. (1990: 35) say: “The sheer number and variety of people crossing the forests . . . clearly provoke a high probability for accidental wildfires.” 70 See note 59, the bias in favor of triggering events as causes in everyday explanation. My using the terms “cause” and “causal history” interchangeably here is in accord with the view that causal explanation of events (like forest fires) consists of obtaining or imparting information about their causal histories (see chap. 1 of this book and Lewis 1986: chap. 22). How much of such information is included in any particular explanation is a pragmatic matter, depending, inter alia, on what the explanation-seeker already knows and still wants to know (see chap. 1 of this book and Vayda 1996: 23 and note 13 and references cited therein).

Both Ends of the Firestick / 65

at all if the changes in forest flammability had not occurred; therefore those changes have greater explanatory import than do any particular ignition events (see Ben-Menahem 1997 for a philosophical discussion distinguishing between necessary and contingent outcomes). A policy implication is that forest-fire management may require devoting more resources to monitoring and somehow limiting the flammability changes than to monitoring and limiting the ignition events. With Sayer’s permission, I forwarded his comment to colleagues who have been engaged in forest-fire research and/or management projects. While their responses varied in particulars, they agreed in emphasizing that fire management requires continuing with efforts to control ignitions. This might seem to be stating the obvious, since it is a simple fact that there cannot be forest fires without ignitions. However, the responses referred only in general terms to ignition control, that is, by means of more training and education and more effective regulation and enforcement. There was no concrete and practical discussion of how and where and at what cost would such means need to be deployed so as to answer the practical questions that Sayer’s view raised about allocating fire-management resources between dealing with ignitions and dealing with flammability changes. Certainly it is not the case that preventing extensive forest wildfires like those in East Kalimantan in 1998 would have required bringing every single ignition under control. As Mark Cochrane, a participant in the discussion, stated, even when tropical moist forests become flammable, they are “not powderkegs that will explode at the first spark.” Instead, said Cochrane, who has done extensive research on the causes and ecology of forest fires in the Brazilian Amazon (see, for example, Cochrane and Schulze 1999; Cochrane et al. 1999; Cochrane and Laurance 2002; Cochrane et al. 2004), the “sustained presence of large numbers of ignitions across a wide region of fire susceptibility” is necessary for extensive forest fires like those of East Kalimantan or of Roraima in Brazil. This is because, at least according to extrapolation from Cochrane’s research in the eastern Amazon, the fires generally move slowly (0.25-0.5m/min.) (Cochrane et al. 1999: table 1) and, even in a few months of extreme drought whereby large expanses of forest become flammable,

66 / Chapter 2

could not have spread over millions of hectares in East Kalimantan in 1998 had there not been thousands of ignitions.71 In the case of less severe droughts making for discontinuous distribution of forest fuels, even more ignitions would, in Cochrane’s view, be needed in order to enable fire to spread from one patch of readily flammable forest (e.g., a patch with a sunny treefall gap) to another (Cochrane, pers. comm. [2001, 2002, 2005]; cf. Cochrane 2002: 21–22). A necessary caveat here is that there may be significant differences between fire-spread rates in forests in Indonesia and in the Amazon, but, unfortunately, no data are available on this.72 In a later contribution to the discussion, Sayer (pers. comm. [2005]), modifying his 2001 view that, as long as people are about, they will produce enough ignitions to cause flammable forests to burn, conceded that it may be necessary to consider the number of people who are about. He even suggested that population densities in and around forested areas may be some measure of the likelihood of fire spreading through those areas when their forests are flammable. I have been unable to obtain data to test this suggestion. Thus, neither the data on logged forest affected by the 1997–1998 fires in East Kalimantan as a whole (see Siegert et al. 2001: 439), nor whatever data there are on population density in the province in 1997–1998 have been disaggregated and made available for units small enough for the kind of analysis needed (Rücker, pers. comm., 2006). But whether, in the absence of clear-cut evidence, we incline to the view that the ignitions necessary for East Kalimantan’s extensive forest fires in 1998 were many (“thousands”) and widely distributed or that they were not so many and not so widely distributed, the difficulties of monitoring and controlling those ignitions in forest areas might still seem to favor arguments for redirecting research efforts to focus not only or primarily on ignitions but also on flammability changes. There might, however,

71 Even slower fire-spread rates have been reported from the moister forests of the Venezuelan Amazon (Uhl et al. 1988: 180). 72 On the general absence of quantitative information on fire behavior in Indonesian and other Southeast Asian forests, see Shields and Moore, n.d., p. 8.

Both Ends of the Firestick / 67

be demurrals on the grounds that enough about such changes is already known. For example, it is already generally accepted that logging can change the nearly fire-immune state of tropical moist forests and that this is so because logging leaves deadwood residues as potential fuel on the forest floor and opens canopies so that temperature is increased, humidity is decreased, and, accordingly, leaf litter, wood waste, and understory vegetation become drier and more fire-susceptible. The problem here is that these are matters accepted or known in only a fairly general way. Available data on fire damage to East Kalimantan’s forests in 1997–1998 suffice to support the generalization that recently logged forests are likely to be hit harder by fire than are “undisturbed or partially recovered forests” (Siegert et al. 2001: 440; see Cochrane et al. 2004 for a review of Amazon data supporting this generalization and Stolle and Lambin 2003: 384 for Sumatran exceptions to it). At a somewhat more specific level, data from the Amazon have led, inter alia, to the recognition of “linkages between forest structure, understory microclimate and the moisture content of fine fuels” (Ray et al. 2005: 1665), as well as to the formulation of “rules of thumb” (Williamson and Mesquita 2001: 325) about differences among forest types (primary, logged, and second-growth) (Uhl and Kauffman 1990) and between large- and medium-sized logging gaps (Holdsworth and Uhl 1997) in the number of rainless days required for sustained combustion to become possible (Kauffman and Uhl 1990: 124). However, just about absent from forest-fire studies in Indonesia (Shields and Moore, n.d.) and still in its infancy elsewhere (Cochrane 2003: 916; Nepstad 2002) has been finer-grained research digging beneath gross correlations, rough rules of thumb, and broad patterns and seeking to identify causal pathways or mechanisms that need to be known for greater success in modeling, explaining, and predicting forest-fire behavior. In Cochrane’s words (2003: 916), current formulations related to fire susceptibility fail to provide the “mechanistic understanding” of forest moisture dynamics that is needed to explain observed forest-fire behavior. Ray et al. (2005: 1676) similarly refer to the need for mechanistic understanding and note that we don’t obtain it by simply identifying land-use features commonly associated with forest fires.

68 / Chapter 2

In other words, while findings like the one cited above about the association between logging and forest fires in East Kalimantan may be important for causal explanation (perhaps in the way that the early findings of an association between smoking and lung cancer or between bacterial infection and duodenal ulcers were important), they can be regarded as only a starting point for research to ascertain how specific characteristics or components of the land use in question affect fire behavior and flammability under specific conditions.73 If, as indicated by remote-sensing data, fire reached 59 percent of East Kalimantan’s logged forest area and only 5.7 percent of its undisturbed forest area in 1997–1998 (Siegert et al. 2001: 439), then concluding from this that logging causes forest fires and letting it go at that may suffice for some broad analytic and policy purposes but leaves us with unanswered questions about the fire causes or fire-inducing mechanisms present in 59 percent of the logged area and absent from the other 41 percent. Similarly, there may be some justification for the claim that Suharto-era forestry policies, having led to so-called predatory or uncontrolled logging that made forests more fire-susceptible, were underlying causes of the 1997–1998 fires (see especially Barber and Schweithelm 2000 and Dauvergne 1998), but the claim still leaves open the questions of just how the flammability changes were caused by the uncontrolled logging and what, if any, kind of controlled logging would not have had similar effects. So there is again indicated the need for more research if mechanistic or finer-grained causal understanding is to be attained.

73 In the finer-grained research being recommended here, we would still be trying to explain extensive forest fires, albeit, in effect, regarding as our immediate explananda such forest-fire “sub-events” as certain changes in fire’s direction, speed, or height and some reignitions (cf. Gruner 1969: 148–150 on “events” and “sub-events” in historical analyses). Illuminating observations on the progression from correlation to causation by moving to finer-grained research to account for lung cancer, duodenal ulcers, and other diseases may be found in Thagard 2000: 256 ff. My arguments here, as in chapter 1, are intended not as a categorical endorsement of finer-grained research but rather an endorsement of it only when it may be expected to yield theoretically or practically significant answers to our questions about causes (cf. Vayda 1996: 17 and note 9). When I object, as I do above, to detailed ignition studies in the absence of studies more or less systematically connecting the ignitions to forest fires, I am, in effect, arguing against according high priority to finer-grained research which is, by itself, of not much use for explaining forest fires.

Both Ends of the Firestick / 69

The appeals that both Cochrane and Ray et al. make for this may, in fact, be subsumed under calls like my own for research guided by the goal of causal explanation.74 The further work to be done is challenging and likely to require better and more sustained funding than has been accorded to past forest-fire research projects in Indonesia. Thus there will need to be more research like the work in the Amazon by Cochrane et al. (1999), consisting of direct observation of fires and measurement or estimation of their rate of spread, the heights and depths of their flames, and flame residence times at particular points. So that observations may be made under a sufficient range of conditions in Indonesian forests, it may also be necessary to set and observe controlled experimental fires, like the three hundred set by Ray et al. (2005) in an Amazon project in 2002. Many observations and measurements of a large number of fires occurring under, and moving through, different conditions, combined with biophysical research to characterize those conditions, should make it possible to construct more useful, empirically supported models and explanations of fire behavior and flammability insofar as due consideration will have been given to how fire behavior and flammability may be affected by variations in not only ambient weather and type and structure of forest but also, for example, the ability of trees to tap into deep soil moisture, as well as by variations in forest fuels in structure, chemical composition, and response to changes in atmospheric humidity (on these variables, see Cochrane 2003: 916–917 and the references cited therein, and also Ray et al. 2005: 1676). At this point, a fairly clear and simple statement may be made about why higher priority should be accorded to research on fire behavior and forest flammability than to the kinds of ignition studies and training programs that have thus far been characteristic of Indonesian forest-fire research. If explaining and controlling forest fires are indeed the goals, there’s simply 74 For a discussion concisely distinguishing explanations based on identifying causal pathways or mechanisms from explanations emphasizing the subsumption of observable phenomena under generalizations, see Brandon 1990: 159–161. The distinction is one that I myself have found useful to make in various places (see especially Vayda 1995a: 223 ff. and chapter 7). It may be noted that some philosophers (e.g., Cummins 2000) say categorically that subsumption under generalizations is not explanation.

70 / Chapter 2

little sense in expending resources on studying and controlling those ignitions which, by and large, are precluded by biophysical and spatial factors from leading to forest fires. So the priority needs to be on establishing the conditions under which fires can happen in tropical moist forests and on the factors that either create those conditions or keep them from occurring. Once these are known, programs to monitor, study, and control ignitions would be very much in order where and when the ignitions would, according to our prior research, have the potential to lead to forest fires.

Interdisciplinary Research and the Human Dimension A failure of perspective that fire historian Stephen Pyne (2000) has attributed to anthropologists and some other social scientists is having their focus so much on seeing fire through people’s eyes as to fail to “pick up the other end of the firestick” and see people “through fire’s eyes.” Whether or not Pyne is right in saying that this failure is “exceedingly common,” social scientists in such projects as the CIFOR/ICRAF one have indeed made their objects of explanation not so much forest fires as people’s use of fire. Moreover, conforming to Pyne’s charge, they have applied themselves to studying such use without duly considering whether it was occurring under conditions that could lead to forest fires. Thus in many cases there has been no special attention paid to distinguishing people’s use of fire under nondrought conditions and under drought conditions or away from flammable forests and in them or near them. It should be no surprise then if the findings produced by social scientists in such cases are more on how fire helps or hinders people in pursuing their ends (“fire through people’s eyes”) than on how people, pursuing their ends, do things that either help fire to spread in forests or else hinder it from spreading there (“people through fire’s eyes”). The case for considering the human use of fire especially under drought conditions nicely illustrates the need to combine investigations from the people viewpoint and the fire viewpoint.

Both Ends of the Firestick / 71

Too much of the social science contribution to tropical forest-fire research has consisted of studies of the uses that people make of fire under non-drought conditions. This was so at some sites of the CIFOR/ICRAF project and appears to have been so in some Amazon research as well (see, for example, Sorrensen 2000, 2004). Failures to obtain data on the extent to which any of the fire uses studied in such projects do actually lead to forest fires have already been discussed. In addition, a critical limitation of studies pertaining to nondrought uses of fire arises from the fact that there are people who use fire differently (and with a fair likelihood of adverse effects on forests) when drought presents them with special opportunities or needs. Thus, South Sumatran rice-growers take advantage of drought to practice deliberately uncontrolled burning of marshland vegetation in the hope that some of the burnt area, otherwise unusable for agriculture, will have soil properties making it good for cultivation (State Ministry for Environment 1998: I, 73; Ramon and Wall 1998; Bompard and Guizol 1999; Chokkalingam et al. 2007; cf. Vayda 1999: 14–15, 26–28 on this and, along with Chokkalingam et al. 2005, also on drought-specific use of fire to facilitate access to commercially valuable turtles in East Kalimantan’s Middle Mahakam peatlands). In an East Kalimantan area of lowland dipterocarp forest as well, Sahur and I found that some of the farmers whose cocoa and/or banana plantations were destroyed fairly early in the 1997–1998 drought tried, before the drought’s end, to make up for their losses either by using fire to clear forest for emergency swiddens or by illegal logging, which, at a time of heightened wildfire danger, involved making cooking fires and campfires, using potentially fire-igniting chain saws, and producing flammable logging residues in the forest. If time and funds were unlimited, we could perhaps go along with appeals that have been made (e.g., in Sorrensen 2000) for tropical forest-fire projects that include investigation of all land-use practices involving fire. However, real-world limitations on the resources available for research make it important to combine the people viewpoint and the fire viewpoint and to prioritize accordingly. For both explaining and managing tropical forest fires, the priority must then be higher for obtaining

72 / Chapter 2

data on how people use fire in times of drought than for obtaining data on all the different ways that they use it at any time at all. As indicated above, it would be desirable to go further and to prioritize forensic fire-scene investigations and/or hotspot analyses identifying FIZ so that there would be findings to guide decisions about locations for surveys or other studies of people’s use of fire. Still, even when the most logical sequential ordering of research may have to be ruled out for practical reasons, why has the study of how people use fire not been guided more by considered judgments about where and when forests are more fire-susceptible? The fault may lie, at least partly, in the rationale, organization, and conduct of so-called interdisciplinary research, whether on forest fires or more generally. According to some recent writing on this in the context of programs of natural resource management (Sayer and Campbell 2004: 50, 158–159), the goal of causal explanation cannot serve to guide interdisciplinary research because the study of causal relations, requiring attention to fine levels of detail, is necessarily a disciplinary or even subdisciplinary specialization. I have, implicitly or explicitly, argued against this view both here and elsewhere (Vayda 1998a and Vayda et al. 2004 [in chaps. 3 and 4]). Bolstering my arguments is the fact that the crossing of old disciplinary boundaries in a quest for understanding of causes or causal mechanisms is often what brings into being productive and exciting new areas of research. A current example is “social neuroscience” (Cacioppo and Berntson 2004; Harmon-Jones and Winkielman 2007), which has developed dramatically since elimination of the old boundary between studies and explanations of social behavior and studies and explanations of brain mechanisms (cf. Azar 2002; Ochsner and Lieberman 2001). Nor can I accept the view that joining interdisciplinary or multidisciplinary projects entails an obligation to serve, first and foremost, as representatives of our disciplines and the perspectives or foci associated with them. According to this view, if one is, for example, an anthropologist, the obligation is “to place people ahead of plants, animals, and soil” (Kottak 1999: 33). In the same sectarian spirit, social scientists have opted for a sepa-

Both Ends of the Firestick / 73

rate section for themselves within the Society for Conservation Biology (Mascia et al. 2003; Mascia 2005) rather than advocating mechanisms for closer interaction with biological and physical scientists and for more effective integration of their respective research. Here again, in more general form, is the kind of failure of perspective about which Pyne has complained in the context of studies of people and fire. Moreover, ignition events, characterized earlier as receiving “undue attention” in forest-fire research, have received just about all of the attention of some social scientists claiming to deal, as in Simmons et al. 2004, with “the human dimensions” of forest fires. This has not necessarily been simply a choice on the part of the social scientists. In my own experience in tropical forest-fire research, biological and physical scientists have often regarded only household surveys and community studies of fire use as the jobs of their social science colleagues. Fine-grained research on possibly important anthropogenic influences on fire behavior and fire susceptibility has been almost completely neglected. Consider hypothetically that, as suggested in some of the discussion above, fine-grained fire-ecology research were to show the following to be important as factors affecting fire susceptibility and fire propagation in logged over forests: the amounts and spatial distribution of deadwood logging residues and the size and spatial distribution of canopy openings resulting from both timber-cutting and extraction pathways. Explaining and controlling fires would then call for data on the human actions—and the causes of the human actions—producing the outcomes of interest. Such data would probably be forthcoming if the organization and work of an interdisciplinary research team were guided by the common goal of causally explaining forest fires but would not be forthcoming if researchers, whether team members or not, were left to pursue discipline-based interests and priorities. Important work has indeed been done on the ethnography of logging in Kalimantan and Sumatra (e.g., Casson and Obidzinski 2002; McCarthy 2000; Obidzinski et al. 2001; Ravenel 2004; Wadley and Eilenberg 2005), showing, inter alia, the proliferation of small-scale, illegal logging in place

74 / Chapter 2

of the large-scale concession logging of the first decades of the Suharto-era timber boom. Since, however, this research has been guided by such goals as analyzing political and economic causes of the changes and not at all by the goal of causally explaining forest fires, it should be no surprise that the research has not included fine-grained observations on how or to what extent the reported changes have affected the logging-residue and canopyopening variables noted at the beginning of this paragraph. In fact, for straightforward illustration of the kinds of “humandimension” observations that could be useful for explaining whatever causally significant variation is found in these variables, the report of a 1992 Amazon study is more useful than any reports from Kalimantan or Sumatra: Tree-felling decisions are made by chain sawyers (contract workers) as they walk the forest in search of potential timber trees. The chain sawyers have little formal training in tree felling and no training in forest management or silviculture and they are paid according to the volume of timber they fell per day. Thus, rapid sawing is better rewarded than careful sawing. Log skidding by bulldozer occurs several days after tree felling. Although chain sawyers and bulldozer drivers live in the same camp, there seems to be little communication among them regarding the locations of the felled trees. To find felled trees, bulldozer operators drive their tractors towards openings in the forest canopy. When a log is found, it is skidded back to the log landing, but not necessarily by retracing the path used to arrive at the log. The result of this unplanned searching and skidding is a criss-crossing network of skid trails, some of which lead to natural forest gaps in which no timber tree was felled. (Gerwing et al. 1996)

The report also describes poor felling techniques and poor skid-trail layout as causes of collateral forest damage. Two hundred trees per hectare with more than 10 cm dbh were reckoned to have been damaged, even though the number of commercial trees cut per hectare was only five to six. In the context of forest-fire research, part of the significance of such damage is that the increased tree mortality resulting from it (see Sist and Nguyen-Thé 2002 for Kalimantan studies of this) would contribute to fire susceptibility.

Both Ends of the Firestick / 75

It is of some interest to note that the cited study and report were done by ecologists exploring possibilities for forest conservation by means of reducing waste during logging and log processing. However, making the kinds of observations cited from the study and recognizing the task of making them as part of dealing with the human dimensions of forest fires are more important than the matter of the disciplinary origins of the observers, especially in light of what I have argued above in favor of crossing or ignoring disciplinary boundaries in seeking causal explanations. It is of course possible that if the investigators in the Amazon study had had more social science training they would have been concerned with the need for appropriate sampling of logging teams and situations as well as with other methodological guidelines to be followed in dealing with anthropogenic influences on fire susceptibility and fire behavior. However, such guidelines should be easy enough to communicate to non–social science participants in forest-fire research projects. No doubt there are conceptual or epistemological barriers between social and biophysical disciplines, but recent writing on this (e.g., Minnegal 2005; Newell et al. 2005) has exaggerated the extent to which such barriers must be articulated and overcome before interdisciplinary cooperation can be achieved in projects concerned with environmental conservation and change. In the case of forest-fire research, if agreement can be reached among investigators from different disciplines on being guided in their research by the concrete goal of causally explaining forest fires and on not losing sight of what they are trying to explain, that may suffice for productive interdisciplinary interactions doing justice to seeing people through fire’s eyes as well as fire through people’s eyes.

Conclusion Much of the discussion to which my distinctions and arguments have led here has been, explicitly or implicitly, about the inadequacy of evidence to support claims being made about the causes of forest fires. For example, I noted that whatever evidence there may be on causes of ignition events, there is

76 / Chapter 2

very little evidence on the spread of fires from particular ignitions or types of ignitions; that whatever evidence there may be on human use of fire in general, there is, with a few exceptions, very little systematic evidence on its use specifically when and where forests are susceptible to wildfires; that however much fire-management programs emphasize control of ignitions, evidence is lacking to show that the programs are effective and deserve a higher priority than programs emphasizing the control of changes in forest flammability; and that however strong the evidence, mostly from remote-sensing, that uncontrolled logging increases forest flammability and hence fire propagation, there have not yet been on-the-ground fire-behavior studies to provide the necessary evidence for deciding what, if any, kind of controlled logging would not have similar effects. The kinds of questions I have raised are important in making us see more clearly that we don’t know some things we need to know for advancing our causal explanations and for basing fire-management policies and programs on them, as well as for setting priorities for research when the time, money, and other resources available for it are limited. Accordingly, for the purpose of explaining and controlling the spread of fires in tropical moist forests, it may fairly be said that lower priority should be assigned to studies of all the causes of ignition events or all the ways in which fire is used than to the more specific or sharply focused studies that could provide the kind of needed evidence I have discussed. Among examples I have given are studies reconstructing the paths and ignition sources of particular forest fires; studies of fire use in or near forests during times of drought specifically; fine-grained research on fire behavior and fire susceptibility under varying conditions of fuel availability and moisture; and systematic research on human actions affecting those conditions.

Acknowledgments My fieldwork on forest fires in East Kalimantan was part of the WWF-Indonesia Forest Fires Project in 1998 and was supported

Both Ends of the Firestick / 77

in 2000 and 2001 by a grant from the National Oceanic and Atmospheric Administration (NOAA) to CIFOR, where I am a senior research associate. For valuable comments on earlier drafts or other useful advice, I thank Mark Cochrane, Jeff Sayer, Neil Byron, Pietro Ceccato, Unna Chokkalingam, Christian Gönner, Martin Hardiono, Anja Hoffmann, Tim Jessup, Peter Moore, Margaret Nolan, Gernot Rücker, Brett Shields, and three anonymous reviewers. I thank Margaret Nolan also for help with library research, and I thank Unna Chokkalingam, Dan Nepstad, Brett Shields, Luca Tacconi, and especially Anja Hoffmann for preprints or otherwise hard-to-find copies of articles or reports. For collaboration and camaraderie in the field, I must once again thank Ahmad Sahur, who first joined me in research in East Kalimantan in 1980. An earlier version of this paper was presented at the CIFOR Workshop on “Remote Sensing and Forest Governance in Indonesia,” in Bogor, June 22, 2004.

3 On Knowing What Not to Know about Knowing: A Critical View of Local Knowledge Studies (with Bradley B. Walters and Indah Setyawati) Claims made by anthropologists and others concerning indigenous or local knowledge tend to exaggerate the cultural mystique of so-called knowledge systems and, accordantly, the difficulties associated with rendering local knowledge accessible to outsiders and with ascertaining its utility for initiatives in economic development and environmental conservation.75 Related to this, researchers have been preoccupied with understanding knowledge per se rather than with understanding what knowledge actually influences human behavior in different situations or contexts and how and why it does so. This chapter proposes an alternative approach which begins by making specific events or actions and not purported knowledge systems our objects of study and explanation and then goes on to consider what actors know mainly insofar as that can help us to make sense of 75 Sillitoe (1998a: 223, note 2), conceding that it is “difficult to draw lines between indigenous knowledge, local knowledge, popular knowledge, folk knowledge, and so on,” opts for the term “indigenous knowledge” because of its wider currency than the other terms in development discourse (cf. Purcell 1998: 259). All of the aforementioned terms are, however, problematic in one way or another. In this chapter, “local knowledge” is used as a somewhat less incongruous label than “indigenous knowledge” for some of the recently acquired knowledge considered in the case-study sections. As will be apparent, not all of the knowledge to be discussed has what some anthropologists engaged in studies of indigenous or local knowledge include among its defining characteristics, that is, its being clearly local, culturally embedded, time-tested, and intergenerationally transmitted (Berkes 1999: 5–7; Ellen 1998: 238; Hunn 1993).

79

80 / Chapter 3

the events or actions of interest.76 Such an approach demystifies ideas about local knowledge and makes productive investigations possible without always requiring laborious investments in ethnographic fieldwork. The approach accords in various ways with the concerns and beliefs of other researchers who are advocates of greater involvement of anthropologists in studies of local knowledge relevant to new initiatives in economic development and environmental conservation (see especially Sillitoe 1998a, 1998c). Thus, shared here with these advocates are concern with the actions that people take in using and managing their environments or environmental resources; concern with the knowledge their taking those actions and not taking certain others is based on; and concern with the causes of changes in the actions and their knowledge bases. Presumably also shared with the advocates is the belief that studies of these matters can be important for the success of the initiatives in economic development and environmental conservation. Participation in the studies to be described below was motivated, inter alia, by this belief. Not shared, however, with at least some of the advocates are the positions to be set forth in this chapter concerning certain methodological issues related to practically relevant studies of doing and knowing. Experience in research projects in Indonesia (Vayda and Setyawati) and in the Philippines (Walters) will be cited to show the rationale for methods that differ in practically advantageous and consequential ways from methods that have been employed or advocated in other studies of indigenous or local knowledge.77 In brief, our view is that anthropologists can deal more effectively and expeditiously with the matters specified in the second paragraph above if they do not commit themselves to so-called holistic studies of necessarily shared or socially or culturally em-

76

On problems of system definition and reification, see chapter 1, pp. 24–25. A note on pronominal usage: In the remainder of this chapter, the first person plural will be used to refer to Vayda and either or both of the chapter’s other authors. However, in the next section, it will be used to refer exclusively to Vayda and Setyawati, who, without Walters, engaged in the pest-management research described in the section. 77

On Knowing What Not to Know about Knowing / 81

bedded local knowledge. In support of this view, we will present methodological arguments echoing those we have previously made elsewhere (especially in Vayda and Walters 1999; see also Vayda 1996, 1997b, 1998a; Vayda and Sahur 1996; and Vayda and Setyawati 1998). On the one hand, these are arguments in favor of being guided in our research by questions about the causes of outcomes of interest. On the other hand, they are arguments against limiting those questions to how-questions, concerned with how factors privileged in advance by the investigator influence such outcomes as environmental changes and the actions that people take in using and managing their environmental resources. These arguments were made in some of our previous publications (especially in Vayda and Walters 1999) against privileging certain kinds of political factors in advance, but here we will make them against privileging cultural knowledge and so-called indigenous knowledge systems.78 And we will further argue here that eschewing such privileging is important for making our studies bear more effectively and expeditiously on development and conservation initiatives. Our arguments for studying local practice and knowledge without necessarily engaging in long-term, holistic ethnographic research should, however, not be construed as opposition to all in-depth, ethnographic fieldwork in relation to development and conservation initiatives. In fact, as the case studies in the following two sections should make clear, the amount of time and effort invested in fieldwork will need to vary, depending on a variety of pragmatic considerations. Clearly there will be cases where in-depth ethnographic studies are needed. But the fact that the results from such studies sometimes justify the time, effort, and expense put into them cannot be used to argue that the studies must always be undertaken. By the same kind of illogic, the fact that the studies are sometimes unproductive could just as well be used to argue that they should never be undertaken.79

78 In chapter 1, p. 35, the same arguments are made against so privileging certain religious or ecological factors. 79 Cf. Huntington 2000: 1273, arguing against either always including or always excluding local knowledge components in environmental management projects.

82 / Chapter 3

There have indeed been unproductive studies, that is, research leading to the discovery of local knowledge and practice that turn out on analysis to be “useless, harmful, or otherwise scientifically indefensible” (Lees 2000, citing Hess 1997). But, as Hess (1997: 79) has noted, such cases are, by and large, omitted from the literature on local or indigenous knowledge. In her report on her own research on local knowledge and practice related to sheep management in an Andean community in Ecuador, Hess does ask for granting the same validity to our own and indigenous views of reality, including the different views of the etiology of disease in sheep. But, despite this appeal, she brings forth no local knowledge and practice on which development initiatives might build. On the contrary, her account indicates that local knowledge and practice have resulted in sick and unproductive sheep and impoverished people. As has already been stated, it would be illogical to use such cases to argue against ever undertaking in-depth, holistic studies of local knowledge. The cases do, however, support our view about the need for making local knowledge studies bear more effectively and expeditiously on development concerns and initiatives. In light of the possibility that there will be no significant payoff from some such studies regardless of how much time, effort, and resources are invested in them, it becomes all the more important to find and adopt ways of zeroing in on the local knowledge and practice which do have significant and positive practical relevance to development and conservation initiatives. Our arguments here are in accord with certain points that others have made in favor of “rapid rural appraisal” and similar shortcut, rapid research methods in development studies. In particular, we agree with points made by Chambers (1991: 522) to the effect that it is important to know what is not worth knowing and to abstain from trying to find it out. Regrettably, however, such points are left virtually as slogans by Chambers and like-minded advocates in the development studies field. Unlike us, they are not concerned with explicating procedures in research guided by questions about the causes of outcomes of interest. Accordingly, no clear advice is to be found in their writings about how, in the course of such research, decisions are to be made about

On Knowing What Not to Know about Knowing / 83

what is worth knowing and what is not. Thus, notwithstanding all their discussion of surveys, checklists, flow charts, diagrams, and other methods for obtaining and recording locally available information (see, for example, Chambers 1991 and Mikkelsen 1994), what is obtained and recorded by their methods is, both in our view and in that of some other critics (e.g., Pelkey 1995), too often only background information for the more sharply focused inquiries needed to produce usable evidence for or against particular causal possibilities in the kind of research that we are advocating (cf. Walters et al. 1999: 209–210). Our own research procedures will be explicated mainly in the context of the case studies presented in the next two sections. The following key features of our approach are, however, worth noting here: 1. Identifying at the outset certain environment-related or resource-related actions as objects of study on the basis of their relevance to development and/or conservation concerns or programs rather than on the basis of their meeting conventional criteria for anthropological subject matter. 2. Not undertaking studies of knowledge per se and not singling out shared or so-called cultural knowledge for investigation but, instead, trying only to ascertain any local knowledge likely to be useful to us for deciding about focusing on and explaining particular actions because of their relevance to development and/or conservation concerns.80 3. Not assuming that the practices and knowledge behind them which are of interest to us are embedded in a “whole system” (Sillitoe 1998a: 247) or “encompassing cultural matrix” (Ellen 1998: 238) which must be elucidated or comprehended in toto if we are to understand the practices and knowledge well enough to use them effectively to meet development and/or conservation goals. Instead, 80 An illustration from the Javanese pest-management studies described in the next section are the actions of impaling crabs on sticks and then planting the sticks at the edge of rice fields. Learning from farmers that they based these actions on their knowledge that the crabs would attract rice seed bugs that could then be set fire to, we looked more closely at the actions and used the farmers’ knowledge to help explain them.

84 / Chapter 3

subscribing on these matters to the views of such philosophers as Lewis (1986) and Rorty (e.g., 1989: chap. 1) and such social scientists as Hawthorn (1991: 173–174, 180, and passim), we assume the following: a) That understanding or explanation of anything that people do or know can be based on seeing or showing its connections to any number of other things or events, whether within an encompassing cultural matrix or not. b) That partial explanations, indicating only some connections and missing others, are useful and, practically speaking, necessary.81 c) That our decisions about which connections to pay more or earlier attention to may be made on pragmatic grounds, such as those discussed in the next two sections and in our earlier publications (e.g., Vayda 1983 and 1996: 9–16, 22–24), rather than on the basis of theories or discipline-rooted biases about the kinds of systems within which connections with explanatory import must be sought. Before proceeding to the case studies, we must make clear that we have been referring to cultural embeddedness only because of the importance assigned to it by others presenting anthropological views of indigenous or local knowledge (e.g., Sillitoe 1998a; Ellen 1998). Because we share the view of some critics (e.g., Sunley 1996: 345–346; Portes and Sensenbrenner 1993) who argue that concepts of cultural or social embeddedness generally suffer from vagueness and that their significance or utility remains uncertain, we do not employ any concept of embeddedness when we report on our own research. 81 A classic statement of the complexity of event causation, clearly implying the need for partial explanations, is the following from Carlyle’s 1830 essay entitled “On History”: “ actual events are nowise so simply related to each other as parent and offspring are; every single event is the offspring not of one, but of all other events . . . and will in its turn combine with all others to give birth to new.” (Carlyle 1899: 89). Referring again to the pest-management studies, we may reasonably claim that useful explanations of such actions as the spraying of a particular Javanese rice field with a particular insecticide at a particular time are possible without our considering all causal antecedents of the actions in chains of events extending back at least as far as the Big Bang.

On Knowing What Not to Know about Knowing / 85

There is, however, something to be said about a special, relatively clear sense in which a concept of embeddedness is sometimes used. In this sense, embedded knowledge refers to knowledge that local people have gained from their forbears about ways of doing things but without knowledge of why those ways work. What is known is simply that those ways must be used if the crops are to grow well or if the prey is to be captured (see, for example, the discussion in Alcorn 1989: 65–66). Although not encountered in the course of the Indonesian and Philippine research from which our main illustrations have been drawn for this chapter, such knowledge might still seem to some anthropologists to need to be considered here. This is because it might seem to them that, on the one hand, our less opaque examples of knowledge from Indonesia and the Philippines have led us to overstate our arguments against committing ourselves to holistic studies of cultural systems and that, on the other hand, such holistic studies are still needed to elucidate local knowledge which is embedded in the special sense indicated in this paragraph. Accordingly, an example of such knowledge from the New Guinea highlands will be briefly considered after the Indonesian and Philippine case studies presented in the next two sections.

Studies of Knowledge and Practice Related to Insect Pest Management in Indonesia As part of a nationwide Indonesian program intended to train 2.5 million rice-growers in integrated pest management (IPM), an invitation was extended at the beginning of 1990 to Vayda and a number of University of Indonesia faculty members and students, including Setyawati, to engage in research on variation and change in agricultural pest management as practiced by Central Javanese rice farmers. The questions which entomologists and other scientists in the program put to us as anthropologists concerned not only the farmers’ actual practices and the ideas on which these were based, but also the likelihood of change in the practices with the acquisition of new knowledge

86 / Chapter 3

such as the IPM training was intended to help develop. Examples of this included knowledge about interactions between insect pests and their predators, about damage caused by different pests at different stages in the rice plants’ growth, and about the dangers of resurgence of pests following indiscriminate use of broad-spectrum pesticides which do a better job of eliminating the pests’ natural enemies than eliminating the pests themselves. The hope was that farmers who had presumably been overusing pesticides would become ecologically informed decision-makers and actors in their rice fields. This, suggested one of the entomologists, would mean a “paradigm shift” among the farmers. Coming to the project as ecological anthropologists but regarding as its essential subject matter the interrelations of knowledge and action or of cognitive and agroecological change, we sought to make up for our own lack of training in cognitive anthropology by reviewing potentially relevant literature in that area. The discussion in this section, incorporating much of a previously published article (Vayda and Setyawati 1998), will be concerned with why that review was disappointing to us and how it led us to see that anthropologists’ assumptions about their proper subject matter were keeping them from making their studies more effectively and expeditiously relevant to programs of economic development or environmental management. There will also be further explication of the research procedures that we favor for achieving such relevance. That much of cognitive anthropology has been little concerned with practical activities or behavior is well-known. It has been noted both by critics of the field (e.g., Harris 1968: chap. 20 and others cited in Johnson 1974: 199) and by some of its practitioners (e.g., Gatewood 1985; Johnson 1974; Lave 1988; Quinn and Holland 1987; and, more recently, Hutchins 1995: xi–xii and passim; Keller and Keller 1993) and was known to us before we began our literature review. Even so, we hoped to find more than we actually did concerning possible knowledge bases of pest management practices. Thinking, for example, that such bases might include specific knowledge of insect reproduction and behavior, including predator-prey interactions, we hoped to find some good descriptions of people’s knowledge, as well as identification of their ignorance, concerning such matters.

On Knowing What Not to Know about Knowing / 87

Instead we experienced disappointment similar to that reported by Bentley, another anthropologist involved in IPM-related research. What he found was that too many accounts ostensibly treating indigenous knowledge of the natural world emphasized linguistic distinctions with little practical relevance. He also found that even as detailed and ethnographically far-ranging a work as Hunn’s on zoological classifications by Tzeltal-speaking Indians in Chiapas (1977) was deficient in describing aspects of the people’s knowledge and ignorance which could be important for our understanding their actual or potential pest management practices (Bentley and Andrews 1991: 117).82 Our disappointment did not, however, end with accounts in which behavior was, at best, a secondary concern. It extended to studies and methodological prescriptions by some in cognitive anthropology who were vocal in decrying inattention to behavior. The problem for us here was somewhat different from the already mentioned problems with studying knowledge prior to ascertaining that it lies behind some actions or behavior of actual or potential practical significance. The problem here lay in the restrictive nature of these cognitive anthropologists’ definition of both their behavioral and cognitive subject matter. In line with Goodenough’s original mandate (1957) to cognitive anthropology to study whatever one must know to behave in a culturally appropriate or societally acceptable manner, the knowledge featured in their studies and programmatic statements was shared, so-called cultural knowledge, while the featured behavior was behavior that is (or could be presumed to be) culturally appropriate or societally acceptable or, in some important way, influenced by culture-specific models of the world (see, for example, Quinn and Holland 1987; Keller and Keller 1993; Goodenough 1994). Even when “ecological effectiveness” was explicitly a concern of the analyst, studying how such effectiveness is achieved through the pursuit of culturally defined goals was the program recommended (Hunn 1989: 145).

82 On the “reluctance in ethnography to record what people do not know,” see Last 1981. But see also Kiros-Meles and Abang 2008 as an example of a recent study presenting data on both farmer knowledge and farmer ignorance regarding crop diseases and control strategies in northern Ethiopia.

88 / Chapter 3

We realized in due course that such programs were at odds with our desire, supported by our entomological colleagues, to be expeditious in making our studies relevant to Indonesia’s IPM program. In light of this, our bases for deciding what actions and knowledge to pay attention to and investigate had to be different from the culture-related considerations used by cognitive anthropologists whose work we had reviewed. Our criterion had to be whether the behavior and its knowledge bases were, or might be, significant for (or as) actual or potential pest management, regardless of what we could see or hope to see about its being shared, culturally appropriate, culturally influenced, or societally acceptable. This criterion was geared to our project’s practical goal of change in pest management by Indonesian farmers. However, the criterion also was compatible with more academic goals which we wished to pursue in the project, for example, the goal of further developing research procedures and explanatory models concerning interrelations of actions, the actors’ reasons for them, and the contexts in which they occur. As indicated elsewhere (e.g., Vayda 1983, 1994, 1996, 1998a; Vayda and Walters 1999) and as, in effect, is illustrated in the present chapter, the actions to which these procedures and models apply are whatever actions occur rather than just the kinds of standardized actions on which some cognitive anthropologists have been wont to focus. Having said all this, we have to make clear also that, in the IPM project, we were simply not privileging rather than shunning more traditionally anthropological lines of inquiry (cf. Barth 1994: 358 on focusing on acts and events rather than privileging “culture”). Thus, we certainly were not uninterested in the influence of shared or cultural knowledge and beliefs on the behavior we were studying, and we were aware that some novel or idiosyncratic pest management practices and ideas that we were finding might constitute culture in the making. But the fact remains that behavior did not have to pass some kind of cultural test in order to become a focus of our investigations and analyses. What then was some of the behavior on which we concentrated our attention? What knowledge and ideas as bases for the behavior did we find among the farmers? What relevance to the

On Knowing What Not to Know about Knowing / 89

IPM program did our findings have? And are there more widely applicable lessons here for conducting practically relevant studies of interrelations of knowledge and action and for making distinctively anthropological contributions to such studies? These are some of the questions to which the remainder of this section will be devoted. Our examples of behavior are drawn mainly from five villages in which Setyawati and four of her fellow students from the University of Indonesia conducted field research for seven months during 1990.83 The villages are located in the four regencies of the Special Region of Yogyakarta. Weekly morninglong IPM training sessions, including both field and classroom exercises, were held for approximately twenty-five farmers in each of the villages for ten weeks. Each farmer received 1,000 rupiahs (about US$0.55 in 1990) and a snack for each session attended. Research involved observation of these sessions (the “field schools”), as well as other participant observation and interviewing in the villages. Behavior learned in the field schools was certainly interesting to us, and, for making assessments of needs and prospects for IPM, we were especially interested in seeing whether or how or to what extent such behavior was continued outside the schools (cf. Lave 1988: chap.2 and passim on learning-transfer research). An example concerns making observations of the numbers of insect pests and predators in the fields in order to provide a basis for ecologically informed decisions about pest management. The specific procedures taught in the schools for making such observations involved counting insect pests and their predators in sample rice plants along a transect in a rice field in order to decide whether to spray the field with a chemical pesticide. Of interest to us were not only these procedures but also any alternatives to them developed by the farmers themselves.

83 The four other students were Rama Chandra, Indrati, Dian Rosdiana, and Bambang Setiawan. We are grateful to them for data that they made available to us. The students were supervised in the research by Vayda and by Iwan Tjitradjaja and Anto Achadiyat of the University of Indonesia. The program in which our project was included was called “Training and Development of Integrated Pest Management in Rice-Based Cropping Systems in Indonesia” and was funded mainly by the Food and Agriculture Organization of the United Nations (FAO).

90 / Chapter 3

We found that many of the farmers succeeded in learning the prescribed sampling and counting procedures and in putting them into practice in the fields rented for use in the school sessions. However, although the farmers had been told in the schools to follow the same procedures in the fields in which they were growing their own crops, we did not find any farmers doing this. Going with the farmers to their fields, we sought therefore to identify what they did instead in order to decide whether or not to spray. We found some walking through their fields, usually near the edges, and looking out for the more visible pests such as grasshoppers and rice seed bugs (Leptocorisa). Sometimes, as they walked, they also stretched their arms out sideways and passed them back and forth over the tops of the rice plants in order to see how many or how dense were the stemborers (Scirpophaga), rice seed bugs, and possibly other insect pests flying forth. Mostly the farmers did not walk into the fields at all but instead inspected them for pests from the bunds. These inspections, whether by walking through the fields or only along the bunds, were for the pests alone and not for their predators. The observed differences between the farmers’ practice in the schools and outside them led to interviews which indicated to us that evidence of whether the paradigm shift hoped for in the IPM program was being achieved had to consist of more than seeing whether the procedures of Western scientists were being copied. As already noted, there indeed were farmers who could and did learn rules of sampling and counting procedures; however, they regarded these simply as rules to be followed in the schools, and they saw no good reason for applying them to their own farming activities. Neither the logic behind sampling, nor predator-prey dynamics among insects (rather than among larger creatures), were understood well enough by the farmers in 1990 for them to devote—or probably to even consider devoting—the extra time and effort which would have been required to monitor the fields as prescribed at the time by the IPM program. That they had in fact decreased their use of pesticides was explained by some farmers by saying that pesticides had become too expensive. The rise in pesticide prices had occurred with the gradual elimination

On Knowing What Not to Know about Knowing / 91

of subsidies after IPM had become the national pest management strategy by presidential decree in 1986. We were able to return to the study area for only a brief stay in 1992, that is, just long enough for single-day visits to two villages where field schools had been in operation for two years. These visits sufficed, however, to impress upon us that some farmers attending the schools had become quite knowledgeable about predator-prey interactions affecting insect pests. They told us that their knowledge of these insect interactions had been gained in the schools, although some referred as well to their preschool knowledge of noninsect predator-prey relations, including knowledge of predation by snakes on rodent pests in rice fields (cf. van de Fliert 1993: 97).84 The newly knowledgeable farmers told us too that they had cut their use of pesticides because they were making their decisions about pest management on the basis of their observation of numbers of both insect pests and predators in their fields. This impressed us as being in accord with the paradigm shift hoped for by our entomological colleagues. However, the farmers were not following the sampling and counting procedures prescribed in the field schools at the time of our original research. In fact, after we had made our observations of differences between what farmers did in the schools and out of them in 1990, IPM training had been modified so that it was, in 1992, regarded sufficient for farmers, as distinct from IPM researchers, to make reasonably accurate estimates rather than exact counts (Gallagher n.d.). One farmer that we talked to had, on his own, hit upon the technique of monitoring stemborers by means of seeing what densities of stemborer moths were attracted to lights that he flashed in his field at night. It was by our criterion of practical relevance that such methods 84 Why these farmers previously lacked knowledge of insect predator/prey relations may be related not only to the difficulty of making observations of insect predation with the unaided eye but also to the probable absence of strong incentives for paying special attention to insect pests prior to the first major outbreaks of such pests as brown planthoppers in irrigated rice fields in the mid-1970s and mid-1980s. These outbreaks were, in large part, a result of inadvertent elimination of the pests’ natural enemies by means of sprayed broad-spectrum insecticides, which had only recently been added to Javanese rice farmers’ tool kits as part of the Green Revolution (cf. Kenmore 1991; Gallagher, Kenmore, and Sogawa 1994; Settle et al. 1996; van de Fliert 1993: 94).

92 / Chapter 3

of estimating pests and predators clearly merited our attention, but, as will be discussed shortly, the methods are anthropologically interesting as well because of the questions they raise about alternatives to their being interpreted or explained as culturespecific practices or developments. During our original research, the other behavior to which we paid attention included, at least initially, whatever farmers did or had done with pest management as their goal. Our thinking on this was that, once any such practices had been identified, we would look at them more closely if it seemed to us and to our entomological colleagues that the practices could actually achieve, or be adapted to achieve, the pest management intended by the farmers. One example of such a practice is using crabs impaled on sticks, which were planted at the edge of the rice fields, to attract rice seed bugs so that they could be set fire to. Among other examples are practices that are the same as those recommended by Western scientists, for example, flooding rice fields to drown mole crickets (Gryllotalpa) and draining the fields to rid them of nematodes (see the recommendations in Grist 1986: 345–347). As already noted, it did not matter to us whether these and other practices that we were finding, and whatever knowledge we were finding them to be based on, were already widely shared or evidently culturally embedded or whether, on the contrary, they were novel or idiosyncratic. If we were to find the latter to be the case with respect to something actually or potentially effective for pest management, we hoped we could proceed, in follow-up research, first to discover factors, including cognitive ones, contributing to its being practiced by some and not by others and then to assess the possibilities for its being more widely adopted. As for old and disused practices that were still remembered, we were interested in pursuing inquiries about them too but, in line with our criterion of practical relevance, not if they clearly held no promise of being revived and being efficacious for pest management in the future. An example of such an unpromising practice about which a few informants told us consists of catching a certain uncommon species of grasshopper in the bush or forest and then carrying it around the rice fields and chanting to the rice seed bugs, “This is your father [or, according to one informant, your husband]; go with him; your home is not here.”

On Knowing What Not to Know about Knowing / 93

Of interest to us too were certain instances of deliberately either not taking action or ceasing to take action against pests. When not doing something is deliberate, it may call for explanation just as much as doing something does.85 In our particular project, it was relevant for pragmatic reasons to ascertain whether inaction might have causes that could keep farmers from being or becoming ecologically informed pest managers. The cases in question concerned certain armyworm (Spodoptera) and brown planthopper (Nilapavarta lugens) attacks that farmers told us about. They said that, in the early stages of rice field infestation, they acted on the basis of practical ideas about chemical control of the pests. However, as infestation became more severe and the measures taken proved unavailing, they decided they could do nothing about it. Some said that they resigned themselves to loss of their crop and sought temporary employment as pedicab-drivers or construction workers in the city of Yogyakarta. Reasons for inaction were found by some farmers in traditional knowledge of the supernatural. To these farmers, the failure of insecticides in these cases was proof of two things: 1) that the pests in question had been magically brought forth from her realm by Nyai Loro Kidul, the goddess of the Southern Ocean; and 2) that the pests would not leave before sating themselves on the crop. Indeed, to some farmers, any major pest outbreaks were attributable to Nyai Loro Kidul. With respect, however, to the brown planthopper outbreaks, other farmers drew on other knowledge to contest this view. Specifically, they argued against it by noting the following: the absence of such outbreaks in former times when the pests that the goddess sent forth were rats (cf. Becht 1939); the fact that the planthoppers did succumb sometimes to pesticides; and the fact that visitations by them, unlike those by armyworms, did not occur suddenly and as if by magic but rather were preceded by such warning signs as strong winds or heavy rain. In the view of some farmers, planthoppers were actually brought from the north by wind and rain. Our conclusion from these cases, as

85 This “passive counterpart” of action is called “forbearance” by von Wright (1971: 170–171), but others, like Rheingold (1988: 173–174), have felt that there is a need in English for another word for “conscious nonaction.” Rheingold refers to the term wei-wu-wei as meeting the need in Chinese. Cf. note 28 in chap. 1

94 / Chapter 3

well as our overall impression from observations made in the course of the project, was that mystical knowledge or magical beliefs did not keep farmers from putting practical measures for pest control into effect. Such knowledge or beliefs were invoked by some farmers to justify inaction when the practical measures that they had taken did not work, while there were other farmers who found reasons not to use the same knowledge or beliefs even in these circumstances. As one of us has suggested elsewhere (Vayda 1996: 6), recourse to such ideas at some times and not others by the same farmers dealing with the same pests shows that we need to pay attention to contextdependent variability in whether or not particular knowledge or ideas or cognitive attitudes are efficacious in guiding actions (cf. Bratman 1992; Thomason 1987). In other words, there is an argument here not only for focusing first on actions rather than studying knowledge per se but also for focusing, as we had, on the contexts in which the actions occur in order to come closer to pinpointing the knowledge on which the actions are based. A more general argument may be set forth here as well in order to close this section. It is worth noting that once the criterion of actual or potential relevance to pest management had been used to make certain actions our objects of study in the IPM project, we could proceed in what Barth (1987: 24) has described as “the tradition of the wondering naturalist.” That is to say, we could look closely at particular situations in the search for whatever factors may have been operating to produce those actions, without a priori theory-based or discipline-rooted constraints on the factors we could consider or accord priority to.86 Thus, 86 In line with this are the appeals in Vayda 1998a: 578 for “clear-eyed eclecticism in pursuit of answers to questions about the causes of . . . concrete actions” and in Vayda and Walters 1999: 171 for not confining our consideration of situation-specific causal possibilities to “those prescribed by any single or simple agenda or theory.” Also in line with this are arguments set forth a long time ago by the philosopher Charles Peirce (e.g., Peirce 1932: 495–499 and passim) and more recently by many others, especially in such fields as artificial intelligence (see, for example, Thagard and Shelley 1997), even if not yet in anthropology, concerning the explanatorily critical importance of abduction. This, as distinct from deduction and induction, refers to ways of identifying or creating explanatory or causal hypotheses for whatever is to be explained and then deciding how worthy they are of further empirical attention. Related discussions in this volume include those in chapter 1 on abduction and in chapters 2 and 4 on following chains of event causation without regard to disciplinary boundaries.

On Knowing What Not to Know about Knowing / 95

when we focused on cultural factors, whether such Javanese mystical knowledge or magical beliefs as have been mentioned or more practical cultural knowledge, it was not because such factors are what anthropologists must focus on. Rather it was because it seemed to us, at particular times in the course of our research, that particular cultural factors were likely to have affected behavior that we needed to explain. For example, when it was found that some farmers had gained impressive knowledge of insect predator-prey relations and were applying it by making field observations of insect pests and predators to provide a basis for pest management decisions, our attention was drawn to the fact of widely shared preexisting knowledge concerning noninsect predator-prey relations, including the knowledge that many farmers had of predation by snakes on rodent pests in rice fields. We proceeded then to ask whether farmers were drawing on this knowledge in making their decisions to commit themselves to field observations of insect pests and predators. However, at other times in the course of the research, we were just as ready to focus on factors other than culture-specific ones to explain behavior. A case in point is the farmers’ making only visual estimates of insect densities and usually doing so by inspecting fields from the edges rather than following prescribed sampling or counting procedures. We recognized this to be like behavior observed by others in similar situations outside Java (e.g., in Honduras as described by Bentley and Andrews 1991: 118 and in the Philippines as described by Kenmore et al. 1987: 106). Accordingly we sought to understand and explain the behavior less as the product of culture-specific influences than as the product of such factors as time constraints, lack of practical experience of sampling, and habits of reliance on frequency estimates and statistical judgments like those which, according to some psychologists (e.g., Gigerenzer 1991, 1992; Cosmides and Tooby 1996), are intuitively made by people everywhere and are often adequate for the objectives at hand. Anthropologists intent on making their work as practically relevant as possible will need sometimes to pay close attention to such factors instead of feeling that they must always concentrate on factors to which they are directed by anthropological theories or by norms about anthropological subject matter.

96 / Chapter 3

Studies of Knowledge and Practice Related to Mangrove Tree Planting in the Philippines Illustrations from the Philippines are drawn from field research by Walters in 1997 on the causes and consequences of mangrove tree planting in Bais Bay, Negros Oriental (Walters 1998b, 2000a, 2000b). As in the IPM research, investigations of mangrove planting led to the finding that practically relevant knowledge was often not widely shared in the community. Specific to this case, however, was the added discovery that knowledge factors were, in general, not always important for our understanding or explaining actions or behaviors of interest. Mangrove forests provide important ecological services and are commonly exploited by coastal communities for firewood, construction materials, and various fish, crustaceans, and shellfish (DeWalt et al. 1996; Diop 1993; Hamilton et al. 1989; Kunstadter et al. 1986; Lacerda 1993; Macnae 1968; FAO 1994). Seventy percent of mangroves in the Philippines have been cleared since 1940 to make way for brackish water aquaculture, residential settlement, and various public and commercial infrastructure developments (Baconguis et al. 1990; Primavera 1995). Moreover, most remaining forests are degraded as a result of past and continued wood harvesting. This state of affairs has prompted a variety of policies and programs to protect and restore existing mangrove sites (Calumpong 1994; DENR 1990; Walters 1995). In particular, the deliberate planting of mangroves is now enthusiastically promoted by governments, NGOs, and aid agencies in the Philippines and elsewhere as a means to restore degraded ecosystems while enhancing livelihood options for povertystricken, coastal communities (e.g., Kaly and Jones 1998; Lewis 1990; Pomeroy et al. 1996; Primavera and Agbayani 1996; Thorhaug 1990; Sukardjo and Yamada 1992; Saenger and Siddiqi 1993; Van Speybroeck 1992). This broader context made significant the discovery of cases where local fisherfolk and fishpond owners have been plant-

On Knowing What Not to Know about Knowing / 97

ing and managing mangrove forests for decades under their own initiatives (e.g., Yao and Nanagas 1984; Cabahug et al. 1986; Walters 1995, 1997).87 The reasons for planting are varied and include the desire to provide a ready supply of construction materials and firewood, to increase tenure security, and to protect seaside homes and fishpond dikes from storm damage (Walters 1997, 2000a, 2000b). The extent to which these existing management systems might serve as a model or offer lessons to wider programs in mangrove restoration was an important consideration for Walters in undertaking the study. Accordingly, he sought specifically to understand the origins and spread of this “indigenous” planting. The possible importance of knowledge factors in its spread was recognized from the outset, and such factors were given due attention in the course of the research whenever they were seen to be relevant to explaining mangrove planting. Investigations led Walters to discover, first, that planters are many and diverse in the few villages where mangrove planting does occur. It is typically a private, household-based activity, but participation does not otherwise strongly correlate with education, income, or other socioeconomic characteristics. The ubiquity of mangrove planting in these sites led Walters initially to speculate that there must be a well-developed, widespread, and systematic knowledge base for planting in communities where it is common. However, investigations in villages in Bais Bay—where most of the fieldwork was done— revealed much variation in knowledge between planters and little evidence of widely held, systematic knowledge about mangroves and mangrove planting. In fact, many planters in Bais did not give much thought to their planting activities. Often, when made impatient or annoyed by questions intended to

87 Local mangrove planting has similarly been documented in Indonesia (Weinstock 1994). Extensive private mangrove plantations in Manila Bay also were described early in the last century by Brown and Fischer (1918). As far as we know, all of these have since been cleared, and the areas developed into fish ponds or reclaimed for residential housing and urban-industrial infrastructure.

98 / Chapter 3

elicit their knowledge of the subject, they insisted that planting is “easy—you just get the mangrove and stick it in the mud and it grows!” That these persons were able to plant at all reflects the fact that the mangrove planting is indeed not a basically complex task and requires little technical knowledge to accomplish. More specifically, it may be noted that Rhizophora species are the preferred and most widely planted mangrove trees (Walters 2000b). Rhizophora are viviparous, meaning that seeds begin to grow and elongate into stems while still attached to the parent plant (Tomlinson 1986). To plant, one simply collects the elongated seedlings, called “tawin,” from the parent tree when ripe—a condition easy to assess—and places them 1/4 to 1/3 their length deep in mud. If environmental conditions are suitable and the young plants are not damaged by disturbance (see below), they will sprout leaves and grow. These very basic planting techniques are straightforward and widely practiced in Bais. However, planters were found to vary considerably in terms of the deployment of other, more specific practices. Such variation was found, in cases, to reflect genuine differences in knowledge between planters and, either demonstrably or else possibly, to have important consequences for planting success. The remaining discussion will describe three such practices: spacing, a method of controlling shell infestation, and test planting. Something shown to have important consequences for subsequent survival and growth of plantation trees is the initial spacing distance applied between planted seedlings (Shepherd 1986; Wadsworth 1997). In Bais, 47 percent of the planters in Walters’s sample (n=91) used 30–60 cm spacing, and 41 percent used spacing of only 10–20 cm. The more experienced and knowledgeable planters almost always employed wider spacings (30–60 cm or more) with a clear understanding of their benefits, that is, compensation for a degree of post-planting mortality and facilitation of the rapid growth of straight stems, most of which are subsequently harvested for fish-corral con-

On Knowing What Not to Know about Knowing / 99

struction.88 For example, experienced planters tend to use closer spacing (30–40 cm) in areas where they anticipate, based on prior planting experience, relatively high post-planting mortality as a result of wave damage or other disturbance (see below). In sites where post-planting mortality is anticipated to be minimal, wider spacing (40–60 cm) is typically used. Other factors were sometimes found to be taken into consideration by planters when deciding on the appropriate spacing to use. For example, knowledge of local fishing, shell collecting, or boat passage practices led some planters to increase their spacing distances so as to enable those engaging in the practices to pass through planted areas without damaging young seedlings. Some planters were also found to initially use wider spacings (around 1.0 m) so that they might quickly establish claims over planting sites. In such cases, these same planters might later return and inter-plant to establish more desirable, higher densities. What of the 41 percent of planters in Bais who use spacing of only 10-20 cm? Spacing this close is costly in terms of added planting expense (halving the spacing distance quadruples the number of seedlings required). It also creates extremely high stocking densities which, in the absence of very high seedling mortality, almost certainly leads to crowding, reduced stem 88 Foresters typically recommend initial spacing of 1.5 - 3.0 m for most upland tree species (Shepherd 1986; Wadsworth 1997) and 1.0 m or more for mangroves (e.g., Khoon and Eong 1995; Siddiqi and Khan 1990). Researchers who first documented indigenous mangrove plantations in the Philippines were, in fact, initially perplexed by the widespread use of close spacing (e.g., Cabahug et al. 1986; Emma Melana and Calixto Yao, personal communication). They responded by conducting formal field trials comparing tree growth and survival of planted mangroves at different spacings (0.5x0.5 m, 1x1 m, 2x2 m and 3x3 m). In general, these studies show that close spacing of 0.5 m does not appreciably reduce growth in the early years and is practical, in particular, if one’s goal is to grow relatively small, straight stems for use in construction or as fuel wood or if one is planting in areas where waves and other forms of environmental disturbance exact a significant toll on survival (Yao 1996; Pedro Balagas and Emma Melana, personal communication). The dominant and most highly valued use of planted mangroves in Bais is for posts used in the construction of traditional fish corrals, called bunsod. For this purpose, tree stems must be straight and are typically harvested when small, that is, between 2.5 and 5.0 cm dbh (Walters 2000a). If spacing is too close, trees will become crowded early and grow slowly. If spacing is too wide, trees will manifest bushy growth and are more likely to have crooked stems.

100 / Chapter 3

growth and subsequently high tree mortality (Shepherd 1986; Wadsworth 1997). Interviews of planters who used very close spacing like this confirmed Walters’s suspicion that such practices were often ill-advised. Planters who used very close spacing were, with few exceptions, less experienced and often had small plantations.89 When asked to talk about the very close spacing they used, they often displayed a lack of awareness of the relationship between spacing and growth and/or they cited reasons for using close spacing that were suspect. For example, a number of planters explained that they had to compensate for having such a small area by planting more trees on that area. In short, many had planted without giving much thought at all as to the consequences of specific spacing practices, a fact less surprising given that most of these planters fish for a living and so lack appreciable horticultural experience. A second practice of interest—the application of used engine oil to control shell infestations—came to Walters’s attention during investigation of a particular site in Bais where only one person had managed to plant mangroves successfully. Upon questioning residents in the area, Walters was surprised to learn that many in this particular area had also planted mangroves, but these had all been killed by shell infestations.90 Further inquiry revealed that the lone, successful planter had repeatedly brushed used engine oil onto the stems of the young plants in order to prevent such infestations. The planter said he had learned this technique during a casual conversation with a man whose house he had been hired to build some years previously and who, by coincidence, was a teacher at the nearby fisheries college and a former project manager for an aid-financed mangrove reforestation project. In his view, the technique was effective by 89 In fact, there is a highly significant positive correlation between plantation size and spacing used (R = 0.478, F = 26.02, p < 0.001, df = 89). 90 Newly planted mangroves can be killed by barnacles and oyster spat which attach and grow on stems, often in such densities that young stems are either physically toppled by the weight of the growing shells or are asphyxiated by having photosynthetic surfaces on their stems blocked by the growing layer of shells (Cabahug et al. 1986; DENR 1994). Vulnerability to damage from shell infestations decreases rapidly as the stem strengthens with age and grows leaves. Planted trees are typically most vulnerable in their first six months.

On Knowing What Not to Know about Knowing / 101

virtue of making stem surfaces slippery and thus keeping young barnacles and oyster shells from attaching. However, the oil may also have some toxic effect whereby young shells and barnacles are killed directly and/or barnacle larvae are repelled from the stem surfaces. We are not aware of any rigorous tests of the efficacy of the technique. It is nevertheless interesting that interviews with the successful planter’s neighbors revealed that none were aware of his method for addressing the problem of shell infestation. Even though they had wondered about their neighbor’s success in planting, they had not actually asked him how he had achieved it while they had failed. Instead, they simply gave up, citing explanations of varying degrees of plausibility for their failures. The successful planter likewise had not been motivated to offer to share his trade secret with his neighbors; he cited eja-ejas (“each to his own”) as the prevailing ethos in such affairs. More will be said about this later. The third practice of interest was the frequent use of testplanting, in which persons plant prospective sites with small numbers of seedlings in order to evaluate the suitability of those sites for more extensive, subsequent planting. Test-planting was found to be commonly employed by the more ambitious, entrepreneurial, and typically prolific planters. These persons know that mangroves will grow only within a relatively narrow range of environmental conditions, that is, between the mean- and high-tide levels (Macnae 1968). But, based on their experience as fishermen and planters, they also know that intertidal lands are changing remarkably quickly in some parts of Bais Bay as a result of fishpond development, sediment deposition from rivers, and natural colonization and planting of mangroves (Walters 2000a).91 Such processes, sometimes creating and sometimes eliminating habitats suitable for mangroves, change local opportunities for planting. 91 Sediment deposition near the mouths of rivers in Bais is often substantial and results from upland soil erosion (Calumpong and Luchavez 1997). Near the mouth of the Tamugong River in Bais Bay, for example, sedimentation during the past four decades has raised the topographic level of offshore lands sufficiently to expand the habitat suitable for mangroves by nearly 200 ha (Walters 2000a).

102 / Chapter 3

As well, planters in Bais face an onslaught of environmental events—including storms, shell infestations, entanglement by floating seaweeds, and anthropogenic disturbance92—that can quickly destroy a young plantation (Vayda and Walters 1999; Walters 2000a). Experienced planters recognize that success depends to a large degree on chance events (e.g., storms and certain anthropogenic disturbances) and factors that vary greatly over space and change quickly in time (e.g., shell infestation and sedimentation) and so are difficult to predict. They thus treat each planting as a distinct, trial-and-error experiment with the objective of evaluating site suitability. They similarly will observe the plantings of others as tests, even though they rarely actively share what they learn with one another, presumably because good planting sites are rare and an underlying objective of much test-planting is to find these sites before others do. Unlike many of the less experienced or less ambitious planters, these entrepreneurs are not easily discouraged by failure. In some cases, thinking another test is called for, they will replant at a later date; in other cases, efforts at particular sites are chalked up to experience and the planters try elsewhere. As noted above, these planters may vary the spacing or the depth of sowing used, either just to see what happens or because they believe that closer spacing and deeper placement may improve seedling survival in the face of certain forms of disturbance (e.g., waves, floating seaweed, rambunctious dogs). Assessing environmental conditions remains, however, at the forefront of most test-planting efforts. In summary then, Walters’s research led to findings about local knowledge that may be of value for mangrove restoration efforts elsewhere. Thus he learned that basic planting practices are straightforward and easy to learn and apply and so should be easy to introduce into novel settings. However, that these practices are being applied in a given community does not mean that there also is a well-developed, widespread, and systematic

92 Persons fishing, gleaning, or passing their boats along the shore often destroy young trees. In most cases, such damage is unintended, but some deliberate destruction of planted seedlings in protest to having common areas planted was also found to have occurred (Walters 1998b).

On Knowing What Not to Know about Knowing / 103

knowledge base there for planting. In fact, it was found that many planters gave little thought to what they were doing and, as a result, demonstrably or possibly important techniques, such as optimal spacing and using waste oil to control shell infestation, were not widely practiced. Walters also learned that environmental events of various kinds often preclude successful planting, regardless of the level of technical knowledge or expertise of the planters. Considerable spatial and temporal variability in the occurrence of such events make it difficult to predict where and when they will constrain planting. In addition to leading to technical insights about planting, the research also led to valuable discoveries about contextual factors and social processes that influence the spread of knowledge. It was found, for example, that active sharing of knowledge was uncommon among planters in Bais and that, while this partly reflected the often large and amorphous character of villages there,93 it was also fostered by the competitive nature of planting in a context where suitable planting sites are scarce. In Bais, knowledge of planting is learned by observing one’s kin and neighbors and through direct planting experience (Walters 2000a). The more knowledgeable planters are typically opportunistic and curious in their approach to planting: most of what they know has been learned from ongoing trial and error and from observing others from a distance. However, with so little active knowledge-sharing and so few opportunities to learn directly from experience (because good planting sites are scarce and because, being fisherfolk, most planters have little prior experience of growing other kinds of trees), it is not surprising that many do their planting without drawing on a substantial knowledge base for doing it. Instead they simply observe and copy the basic planting techniques of kin or neighbors. In line with this, one finds little evidence of a complex and widely shared knowledge system to guide the planting of mangroves in Bais. This is consonant also with the further fact that successful planting 93 In addition to expanding populations of fisherfolk, landless families from surrounding areas now live along the shoreline around Bais Bay. Former village boundaries have been blurred in many areas as a result of increase in the density of houses.

104 / Chapter 3

depends, to a large degree, on environmental events largely out of the control of individual planters. In fact, the growing realization that knowledge factors were of only limited relevance to explaining the origins and spread of planting led Walters during the course of his research to enlarge the investigation to consider the influence of other factors already alluded to, including the role of environmental constraints and tenure (Walters 1998b, 2000a; Vayda and Walters 1999). Only about one-quarter of the nine consecutive months spent in the field was ultimately devoted to investigations of knowledgerelated matters. The decision to thus extend investigations beyond knowledge as a subject reflected an appraisal of the relative importance of various factors and also the pragmatic consideration that the factors besides knowledge found to be important for explaining planting success and failure in Bais would need to be taken into account in mangrove planting projects elsewhere.

A Study of Local Knowledge in the New Guinea Highlands Is there more of an a priori case to be made for holistic studies of cultural systems when knowledge, unlike that in our less opaque Indonesian and Philippine examples, is “embedded” in the special sense referred to at the end of our introduction, that is, when it is knowledge which local people have gained from their forbears about ways of doing things but without knowledge of why those ways work? Being confronted with such knowledge may well be a challenge to the researcher, but what we wish to emphasize here is that it is a challenge not necessarily to be met by holistic studies of cultural systems. A good illustration, cited by one of us elsewhere (Vayda 1995a: 227–228 [see chap. 7) and worth briefly repeating here, is provided by Fringe Enga shifting cultivators in New Guinea’s central highlands. Conveyed by them to the geographer Waddell (1972, 1973, 1975) was their knowledge that their sweet potatoes

On Knowing What Not to Know about Knowing / 105

would not grow well unless planted in carefully constructed, large plano-convex mulch mounds, more than a half meter in height and at least three meters in diameter. Why this was a requirement is something which Waddell’s Enga informants were “not, strictly speaking, able to say” (Waddell 1972: 136). Observations and tests by Waddell (1972: 159–161) and others (e.g., Sillitoe 1998b: 130–132) have indicated that the mounds, by virtue of the mulch in them, contribute to soil fertility (see also the sources cited on this in Sillitoe 1996: 382–383 and 1998b: 130). However, at the time of his field research, it was left to Waddell to surmise that the specific construction and design of mounds at the highest altitudes for growing sweet potatoes could be adaptations to protect the crop from frost. He then proceeded to make observations and tests to confirm this possibility. These further investigations by Waddell, far from being holistic studies of cultural systems, involved, inter alia, taking temperature readings from unmounded ground and from the upper parts of mounds and making measurements which showed that the height of mounds and the minimum height above ground at which sweet potatoes were planted increased with altitude, which generally correlated with the intensity and frequency of the frost hazard (Waddell 1973: 36; 1975: 255). As remarked in chapter 7, the thrust of the data obtained by Waddell was to show that the construction and use of mounds were so efficient and precise solutions to the frost problem—so “well designed,” in other words, to deal with it—as to make it unlikely that mounding by the Fringe Enga afforded protection from frost just by coincidence rather than as a result of such causes as long-ago trial-and-error learning by some, followed by imitation by others. However, no data were (or could be) obtained on the actual cause-and-effect sequences whereby knowledge of mound

94 See chapters 1 and 7 in this volume for further discussion of when and how design analysis and comparative studies may be used, in the absence of direct historical evidence, to distinguish those beneficial consequences that are by-products or the effects of chance from those that occur because particular behavioral traits have been fashioned by some selective process, such as trial and error, to produce them.

106 / Chapter 3

construction and of its importance for agricultural success became established or “embedded” in Fringe Enga culture.94 In this chapter, there are two points to be made then with an illustration such as Waddell’s case. One is that practices supported by opaque professions of knowledge, like those of the Fringe Enga about mounding, may well be important for development, resource management, or conservation and may need to be taken into account in new programs in these areas. The other point is that, with respect to these practices as with respect to some of those considered in the IPM and mangrove-planting cases, studying and explaining the practices so that their practical significance can be recognized and built upon for development or conservation initiatives call for something other than poorly focused inquiries thought to be justified by the mantra of holistic study of cultural systems. In our view, the need is for knowing about and testing situation-specific causal possibilities, such as those related to the New Guinea highland frost hazard as well as those referred to earlier.95

Conclusion The foregoing illustrations should suffice as supports for our arguments against using culture-related considerations, including assumptions of embeddedness in cultural matrices or systems, for the purpose, on the one hand, of delimiting the local knowledge and practice to be deemed appropriate for us to study and, on the other hand, of arguing for long-term, in-depth, holistic sociocultural studies as a prerequisite for recognizing, explaining, and applying local knowledge. Eschewing such considerations for designating what we must study, not only can we still contribute both to practical action programs such as IPM or mangrove restoration and to theoretically meaningful research on the interrelations of cognition and action but also we can, in 95 For a discussion of both the value and the limitations of our having substantial antecedent knowledge of a local people’s culture when we are trying to make causal connections among events in which they are involved, see Vayda and Sahur 1996: 51.

On Knowing What Not to Know about Knowing / 107

our view, contribute better and often more expeditiously. We can do this by virtue of being guided in our research more by open questions about why and with what knowledge do people do what they do than by restrictive questions about how actions and knowledge are affected by factors privileged in advance by us because they are cultural and/or assumed by us to be part of (or embedded in) cultural systems that must be elucidated.96 We can, in other words, do better work and often faster work by not tying our hands in order to make the work distinctively anthropological. However, as our case studies should have made clear, our saying this does not constitute an endorsement of “rapid rural appraisal” and similar shortcut, rapid research methods mentioned in our introduction. As stated there, what is obtained and recorded by the methods is too often only background information for the more sharply focused inquiries needed to produce usable evidence for or against particular, situation-specific causal possibilities in the kind of research that we are advocating on the causes of practically relevant actions. We believe that those who have turned to rapid appraisal methods, as much as those still committed to holistic ethnography, can make their research more useful by letting it be guided more, in the ways discussed and illustrated in this chapter, by clear questions about the causes of concrete actions or events relevant to development and/or conservation concerns.

96 It may be noted incidentally that our arguments here are in opposition also to Sahlins’s decades-long insistence that the cultural must be the object of the “anthropological project” (see, for example, Sahlins 1976a, 1995, and 1997).

4 Do We Need an Anthropological Perspective on Tropical Deforestation?

This chapter is a review of Tropical Deforestation: The Human Dimension, edited by Leslie E. Sponsel, Thomas N. Headland, and Robert C. Bailey. (New York: Columbia University Press, 1996, pp. xxviii + 365.)

Whenever an issue gains national or international prominence, it is a fair bet that it won’t be long before enterprising anthropologists will convene meetings and produce publications purporting to put the issue in what will be claimed to be an anthropological perspective. Accordingly, with deforestation very much a global issue, it should be no surprise to us that there has appeared a volume, Tropical Deforestation: The Human Dimension, with the aim, as professed by its editors, of showing that anthropology has a “special contribution” to make in helping to understand and resolve what they refer to as the deforestation crisis. This crisis is constituted for the editors by tropical deforestation rates which, in estimates they cite, reached 142,200 square kilometers per year by 1989 and subsequently exceeded 0.4 hectares per second and caused the extinction of as many as 10,000 species annually (p. 3). It should be also no surprise to us that the sources of the special contribution to be made by anthropology are seen by the

109

110 / Chapter 4

editors, in their preface, to lie in its “unique” fieldwork methods and in the special attention which anthropology, in comparison with other disciplines dealing with deforestation, pays to “more immediate human contexts and aspects of the phenomena and processes of deforestation at the local community level” (p. xix). The editors also make the claim that anthropologists are unique in their ability to combine such different research approaches as the ethnographic, the comparative, and the diachronic (p. 29). Some of the chapters do, in fact, reach conclusions based on the analysis of data produced by the different approaches, but, as I will suggest again later, the point that anthropologists have a unique ability in this regard remains arguable. More generally, I will suggest at the end of this review that anthropologists can contribute better or more if they give up trying to make their contributions distinctively anthropological. The thirteen chapters that follow the editors’ preface and introductory chapter were written mostly by anthropologists and contain various useful insights and items of information, as well as, in some cases, valuable historical reviews and analyses of the course of change in forest use and deforestation in particular countries or in particular regions. There also are two chapters on pre-European deforestation: one, citing detailed archaeological data, on the role of deforestation in the Classic Maya collapse, and the other, by an ethnobotanist, on the removal of forest cover for agriculture and the other changes that pre-European Polynesians, either wittingly or unwittingly, effected in forest ecosystems. As the editors recognize, an important point emerging from these two chapters is that “anthropogenic deforestation is nothing new” (p. 53). Still, if one considers the volume as a whole, what is perhaps most striking is how little it delivers on what the editors’ preface, in effect, promises, namely, the use of data from intensive, locallevel anthropological fieldwork to produce special insights and analyses concerning the causes and/or possible control of deforestation. On the one hand, mostly obvious, banal, problematic, or only tangentially relevant conclusions about deforestation are drawn in two of the three chapters that rely most heavily or most

Do We Need an Anthropological Perspective? / 111

explicitly on ethnographic fieldwork data. On the other hand, the more interesting, well-supported, and relevant conclusions presented in other chapters are evidently based, in the main, on other kinds of data. Consider first the chapters in which the most use appears to be made of ethnographic data, including data collected by the authors themselves. Carolyn Cook’s chapter includes a case study of the Amung-me people of Indonesian New Guinea. The data for the case study were gathered over a period of fifteen years, including eight years spent by Cook living in the Amungme homeland. A question that she devotes several pages to answering is how the Amung-me of one particular valley were able to live successfully there without destroying their forest environment. Her answer consists of giving yet another example of the kind of now well-known “established integral” swidden system described many decades ago by Conklin (1957) in his classic study of Hanunoo agriculture on Mindoro Island in the Philippines and subsequently reported by many other investigators from other tropical forest areas. The problematic conclusion to which Cook is led by this unsurprising answer is that the traditional system provides a foundation for future forest conservation and economic stability. Echoing the green romanticism of recommendations that other authors represented in the volume (e.g., Alcorn and Molnar, p. 116, and Bailey, pp. 322 ff.) make for community management of forests, this conclusion ignores the likelihood that the viability of established integral swidden systems depends on certain characteristics shared by all tropical forest areas from which such systems have been reported. These characteristics include: 1) population densities low enough for the cultivators to avoid clearing any appreciable primary forest annually and instead to return to previous swidden sites where forest regeneration has occurred after long fallows; and 2) the absence not only of such enterprises as logging, oil-drilling, mining, and large-scale pasturing or tree-crop planting by nonindigenous companies but also of opportunities for the indigenous people themselves to engage in such enterprises, either directly or through contracts with others.

112 / Chapter 4

Indeed, according to reports from the Brazilian Amazon, when the opportunities have been available, some indigenous people have seized them. Thus, in 1989, Guajajara Indians took hostages in order to force FUNAI, the government Indian agency, to give them logging permits (Redford 1990: 28; Anon. 1993). Kayapo Indians, through their leaders, granted concessions to timber companies for the logging of large tracts of virgin mahogany and other tropical hardwoods. The most that some of the Kayapo leaders did to try to stop deforestation was to go to Brasilia in 1993 and to offer to stop it if an indemnification of $800,000 per month for timber sales foregone were to be received. With the offer rejected, the trees continued to fall (Conklin and Graham 1995: 703; and Anon. 1993, which refers also to similar stories about Nambakwara and Kaxarari Indians). Instead of being “ecologically noble savages,” these indigenous people were, as Conklin and Graham (1995: 703) have noted, like other poor, rural Amazonians in needing cash and having few options for earning it. There is no good basis for assuming that there was anything in the culture of the native Amazonians to make them favor long-term environmental conservation over short-term profits. Moreover, from such other tropical forest areas as those in Indonesian Borneo (Dove and Nugroho 1994, cited in Vayda 1997b: 11), there have been reports of attempts by indigenous elites to reinterpret traditional institutions so as to justify timber concessions for themselves as they learn more about the world’s broader monetary and consumption-oriented economy and seek more rewards from it. In other words, although the editors do not make a point of it, contemporary indigenous people are not exempt from being motivated by what the editors themselves (pp. 14, 19) identify as “elemental factors” in deforestation: need and greed. Notwithstanding the two chapters on pre-European deforestation and a reference in David Wilkie’s chapter on “Logging in the Congo” to poor prospects for sustainable forest management because of the short planning horizons of rural forest landowners and poor forest farmers (pp. 245–246), the kinds of contemporary developments that I have noted in the preceding paragraph are, in fact, not mentioned at all in Tropical Deforestation. In the

Do We Need an Anthropological Perspective? / 113

editors’ introductory chapter, a paragraph arguing that indigenous people are “effective conservationists” is followed by this single sentence: “Others have marshaled arguments and evidence against a conservation ethic or sustainable practices for some indigenous groups” (p. 23). Not only is there no subsequent refutation of the arguments and evidence referred to, but there is not even a statement of what they are. One searches in vain in the volume for the merest allusion to even the simple but powerful argument that the persistence of forest environments inhabited by indigenous farmers and foragers, instead of constituting clear evidence that the people are conservationists imbued with a conservation ethic, may often be construed to result from their not having the numbers, technological capability, and market incentives to destroy their forests. Eder’s is another chapter drawing heavily on ethnographic data. These data were collected by Eder and others on Palawan Island in the Philippines. As far as showing how anthropology can make a special contribution to dealing with the deforestation crisis is concerned, the problem with the chapter is its being focused more on the aftermath of deforestation than on its occurrence. Eder does describe forest clearance in Palawan by migrants from other islands, but his main concern is to argue that the objectives of some of the migrants— even if only a minority—are to establish stable and productive upland farms and that the government can help them to achieve these objectives by making their land tenure more secure and by otherwise facilitating their adoption of purportedly sustainable agroforestry practices. One of the chapters written by anthropologists and drawing on data from their own field research does do a good job of showing what the editors refer to as deforestation from the people’s viewpoint (p. 26). This is Susan Stonich and Billie DeWalt’s chapter on Honduran deforestation. It comes closer than any other chapter in the volume to delivering on the editors’ promise of special insights and analyses on the basis of intensive, locallevel anthropological fieldwork on the causes of deforestation. Although, on methodological and empirical grounds (see Vayda 1997b and cf. Vayda 1996: 22–23), I reject Stonich and DeWalt’s

114 / Chapter 4

claims for the general or universal utility and applicability of a so-called political ecology approach whereby a priori judgments are made about the importance of interactions of government and market as factors in contemporary environmental change, their chapter leaves little doubt about the importance of these factors in Honduran deforestation.97 Specifically, they argue, with appropriate supporting evidence, that Honduras’s economic development strategy, oriented to production of beef and other commodities for export and abetted by bilateral and multilateral aid and lending organizations, intensified deforestation and other “resource abuse” by virtue of giving priority to short-term income generation and exacerbating inequalities in access to resources. It is in showing the effects of these increased inequalities that findings from Stonich and DeWalt’s local-level, anthropological fieldwork are tellingly brought into play. They describe, for example, the extension of farming to steep, forested slopes by smallholders and farm laborers no longer able to gain their livelihoods from better agricultural land which medium and large landowners appropriated, with the help of foreign aid channeled through government loans, for cattle ranching and other forms of labor-displacing commercial agriculture. Stonich and DeWalt also describe a specific mechanism whereby the new inequalities in access to land were used by large landowners to cheaply convert forest to pasture: forested hillside land was rented to poor peasants for planting maize and sorghum as subsistence crops and then, as yields declined in the second or third year of cultivation, the peasants were required by the terms of their rental agreements to sow pasture grasses between the rows of maize and sorghum before new forested land would be rented to them for clearing, planting, and eventual conversion to pasture. Is it the case that what Stonich and DeWalt found in Honduras was, as they suggest, “representative of processes occurring throughout the Central American isthmus” (p. 188)? One might hope to begin to get an answer to this question from the chapter that immediately follows theirs in the volume inasmuch as the

97

For further criticism of such political ecology approaches, see chapters 1 and 6.

Do We Need an Anthropological Perspective? / 115

chapter is like Stonich and DeWalt’s in being concerned with contemporary Central America and in linking some local-level observations of deforestation to the production of nontimber export commodities. The chapter, however, is disappointing. Its author is tropical biologist John Vandermeer, whose observations were made while he was engaged not in intensive ethnographic research but rather in biological research spanning several decades. During this time, there were ebbs and flows in exportoriented banana production near the biological preserves of Costa Rica’s Sarapiqui Valley and, when the ebbs occurred, Vandermeer saw the creation of farm plots from forest by laid-off banana plantation workers who mainly were rural migrants. Most of Vandermeer’s chapter, however, is devoted to presenting not such observations but rather the argument that less deforestation in Nicaragua than Costa Rica during the 1980s was the result of Nicaragua’s having an agrarian reform program more successful than Costa Rica’s in eliminating the land hunger of peasants and their consequent conversion of forest to farmland along access roads constructed by companies engaged in selective logging. The argument may sound plausible, but Vandermeer does not support it adequately with statistics. On land ownership and land reform, only a few statistics from Nicaragua (p. 220) but nothing from Costa Rica are cited. On countrywide deforestation rates, Vandermeer’s only statement is that, according to the 1990 report of the World Resources Institute (WRI), Costa Rica had the highest rate in the world (p. 217). Nothing is said about what that rate was, how it was calculated, and how it compares with the Nicaraguan rate, whatever that may have been. Nor is it mentioned that serious doubts about the accuracy of WRI’s claims about Costa Rican deforestation have been raised, that the remote sensing data providing the basis for the claims pertain to a seven-year period ending in 1983 and thus cannot be fairly used for comparing Costa Rican and Nicaraguan deforestation during the entire 1980s decade, and that, in fact, there was for much of the decade a crisis in the Costa Rican cattle industry and, according to some estimates, a consequently steeply declining rate of deforestation (Edelman 1995: 33–35; Lehmann 1992: 67, citing Dudenhoeffer 1990). Interestingly, Stonich and DeWalt

116 / Chapter 4

and also the anthropologists whose chapters are discussed in the paragraph after next do much more than Vandermeer to support their arguments and analyses with macro-level data. Although I am not yet ready to believe that anthropologists in general have the unique ability referred to earlier with respect to integrating different research approaches, some of those represented in this volume certainly do a better job than Vandermeer in this regard. One can only wonder why the volume includes his chapter in its present form. The one other chapter in which the main conclusions drawn are clearly based on local-level fieldwork is the aforementioned chapter by Wilkie on the Congo. In order to ascertain effects of logging on indigenous farmers and foragers, Wilkie, although not an anthropologist but a conservation ecologist, devoted himself over a thirty-nine-day period in 1989–1990 to taking censuses and conducting interviews in fifteen villages within a Congolese timber concession that had been the site of selective logging since 1985. As a result of this research, he is able to describe changes in living styles and standards because of the increased reliance of local people either on wage labor for the timber company or on the sale of game meat, fish, and agricultural products to timber-company employees. He is also able to raise some troubling questions about what the lives of the local people will be like after the timber company moves on. The chapter constitutes exactly the kind of report that needs very much to be brought to the attention of timber company executives and government planners and policymakers. Accordingly Wilkie deserves credit for having written it and for having done the research on which it is based. But he can hardly be said to have provided any examples of deforestation-related studies employing anthropology’s unique fieldwork methods (involving, as stated by the editors themselves [p. 26], participant observation and more than just a few days in each community studied) or any demonstration of anthropological contributions to dealing with the deforestation crisis. Consider now some of the chapters which, although written (or cowritten) by anthropologists, draw not mainly on anthropological fieldwork data but rather on a variety of sources,

Do We Need an Anthropological Perspective? / 117

including government documents, to describe and explain the changing course of deforestation in particular countries or regions. Eduardo Bedoya and Lorien Klein’s chapter consists of a perceptive and edifying account of how deforestation in Peru resulted from peasant responses to factors that included, either continually or at one time or another since the 1940s, government road-building and drug-eradication programs, pressures from Shining Path terrorists, and market forces favoring illicit coca production on steep slopes over foodstuff production on land more suitable for agriculture. In Alfonso Peter Castro’s chapter, historical data on colonial reforestation schemes in Kenya are effectively used to explain the failure of some schemes (for example, because of British colonialist obliviousness to certain indigenous land rights and to the absence of traditions of forced labor and communal tree-planting) and the success of others (for example, because the particular tree species being promoted could be used for both domestic and commercial purposes and required little labor after planting). Emilio Moran’s chapter on the Brazilian Amazon ably reviews various ways in which either government action or inaction contributed to deforestation: not only through road-building programs as in the Peruvian case but also through subsidies to large-scale cattle ranchers (after the 1973 oil crisis changed the economics of colonizing the Amazon with small farmers) and through failing to collect the prescribed 25 percent capital gains tax on land sales and thus encouraging forest clearing for the purpose of the kind of speculating in “improved” land which, according to Moran, accounts for most of the deforestation in the state of Rondonia. All of the chapters noted in the preceding paragraph consist of solid and illuminating reviews and analyses, but, despite having been written by anthropologists, there is little that is distinctively anthropological about any of these chapters. For the most part, with their emphasis on the effects of government programs and policies and their reliance on historical and statistical data, they might just as well have been written by the economists and other nonanthropologists contributing their analyses to such valuable, edifying volumes as Repetto and Gillis’s Public Policies and the Misuse of Forest Resources (1988).

118 / Chapter 4

The same may be said of the chapter on India by ethnobotanist Janis Alcorn and natural resource management specialist Augusta Molnar. Although their chapter, more than many of the others, does deal with deforestation at the local community level, it is hardly a chapter relying on the traditional ethnographer’s stockin-trade of intensive local-level fieldwork. What it does do is to present a useful review of data from various sources on community forest management in precolonial India and on the acceleration of deforestation as, first under colonial and then under postcolonial rule, community rights of forest use were progressively reduced and there was increase in state regulation and management of forests in ways favorable to commercial interests and detrimental to the subsistence of large numbers of forest-dependent tribal people and rural poor. However, the last two pages of Alcorn and Molnar’s chapter are problematic, containing some dubious answers to the chapter’s subtitular question of “What Can We Learn from India?” In these pages, the authors extrapolate from their reading of Indian forest history to recommend community forest management for parts of the world where frontier conditions, probably long gone from most of India, are still present and provide, in this time of economic globalization, attractive opportunities for entrepreneurial and destructive use of forests not only by outsiders but also, as has already been discussed, by local communities or indigenous people (cf. Rudel and Roper 1997: 55).98 Perhaps more than anything else, Alcorn and Molnar’s extrapolative inclinations are what they share with Malinowski and other past ethnographers who spent much time in the study and analysis of a particular society and then saw it as a microcosm of the world. Is there a better case made for anthropology’s contributions in the two chapters that I have not yet mentioned? The answer, sad to say, is no. In the chapter coauthored by anthropologist Robert Sussman, sociologist Linda Sussman, and earth scientist Glen Green, satellite imagery and demographic data are used to reconstruct some of the history of deforestation in certain regions of Madagascar and there are conclusions reached about how the history was influenced by such factors as population density 98

On this, see also some of the references cited in chapter 6, note 108.

Do We Need an Anthropological Perspective? / 119

and the slope of the land. There also is, in a couple of pages of the twenty-page chapter, some citation of Linda Sussman’s findings from her interviews on the subject of why residents of hamlets near a small forest reserve in southern Madagascar engaged in forest clearance which, as revealed by a comparison of 1968 and 1987 aerial photographs, occurred not very long ago. These findings, pointing to the failure of wet-season crops and the economic impact of increased cattle rustling as causes of forest clearance, clearly support the authors’ plea for integrated projects in which there is on-the-ground, local-level collection of data to complement data from remote sensing. The fact remains, however, that the chapter says very little about such projects or about dealing with deforestation by using findings from them. The last chapter in the volume is coeditor Bailey’s. This does deal with people at the local level, but it does so without any explicit or systematic presentation of the results of intensive local-level fieldwork. Instead it is an extended plea for empowering local people as protectors of biodiversity and for giving them and their interests more adequate representation in conservation projects. These projects, according to Bailey, have been the products of planning and surveys by those—mainly biologists and representatives of government and international conservation and funding agencies—with priorities other than local people’s goals and needs. Incorporated in Bailey’s plea are what seem to be his personal observations of conservation projects in central Africa, where he has engaged in research in biological anthropology over a long period. These observations, about such matters as the top-down nature of project planning, the segregation and underfunding of the so-called human or sociological components of projects, and the facile references in project proposals and conservation management plans to generating revenue by means of making local populations the objects of tourism, all ring true to me and correspond to my own observations of various projects in southeast Asia in the 1980s and 1990s. For others, however, the accuracy of Bailey’s observations may be hard to judge, for he does not present them in any systematic fashion or as the products of any conscious, intensive study. One is left to suspect that the special deforestation-related

120 / Chapter 4

anthropological contribution which Bailey regards his chapter as illustrating is not the anthropologist’s distinctive research or its distinctive results but rather his advocacy on behalf of the people that he has studied. Indeed the introductory chapter by Bailey and his fellow editors refers favorably to such advocacy and to the presumed ability of anthropologists to help to empower local forest communities (p. 27). Nowhere in the volume, however, are there any questions raised about possible conflicts between the advocate’s role and the researcher’s. It is in accord with this omission that, as previously mentioned, there is also no explicit recognition or serious consideration of the already available evidence indicating that at least some communities, once empowered, become destructive exploiters rather than unflinching protectors of their forest environments. It might be thought that some of my criticism of Bailey’s chapter, as well as my somewhat similar criticism of the chapter by Alcorn and Molnar, amounts to saying that anthropologists may justifiably speak authoritatively only on those contemporary matters on which either they themselves, or others cited by them, have indeed done intensive, local-level research and that, on other matters, like that of forest management by indigenous communities within range of new economic incentives for forest exploitation under frontier conditions, the anthropologists should speak either tentatively or not at all. That this definitely is not what I am saying is something I must make clear before ending this review. Certainly I believe that intensive, local-level fieldwork can be invaluable for answering some important questions related to deforestation, and such questions include not only those already discussed here but also some others oddly neglected in Tropical Deforestation—for example, the question of what interplays of social, political, technological, economic, and environmental factors may be responsible for either controlling or not controlling the burning of rainforests at the time of such severe, El Niño-induced droughts as occurred in 1982–1983 and 1997–1998 (cf. Cohen et al. 1998; Mackie 1984).99 But the training

99

See chapter 2 on this.

Do We Need an Anthropological Perspective? / 121

and abilities that anthropologists have for doing fieldwork and for using its results as bases for answering important questions need not be regarded as in any way precluding their making substantial and effective use of other bases as well. This is something recognized by the editors but not especially emphasized by them (p. 29). If lessons about how anthropologists can contribute to dealing with deforestation are to be drawn from the chapters praised as well as from those criticized in this review, it is noteworthy that, apart from the chapters of Tropical Deforestation on prehistory, the volume’s most commendable chapters written by anthropologists are those whose authors, drawing as readily on historical and macro-level data as on the results of intensive local-level research, seem not to have worried about making special contributions from the standpoint of their discipline. Instead, in following the chains of event causation wherever they led, they ignored disciplinary boundaries and conventions and evidently felt free to resort to a variety of means to obtain data, much in the manner of a good, open-minded, knowledgeable, analytic historian or a resolute investigative journalist doing his best to adhere to high standards of verification and to limit being swayed by personal or political sympathies.100 Whether or not regarded as anthropological, clear-eyed eclecticism in pursuit of answers to questions about the causes of such concrete actions as cutting down trees or setting fire to them (cf. Vayda 1996: 18 ff. and Vayda 1997b: 14) may be the method most worth emulating by anthropologists concerned with issues like deforestation.

100 Cf. the discussions of doing interdisciplinary research and crossing or ignoring disciplinary boundaries in chapters 2 and 3

5 Seeing Nature’s Complexity but Not People’s

This chapter is a review of Discordant Harmonies: A New Ecology for the Twenty-First Century, by Daniel B. Botkin. (New York: Oxford University Press, 1990, pp. xii + 241.)

To those of us for whom much of the challenge and interest of ecological studies lies in dealing with change and contingency, Botkin’s book will be most welcome. Despite the reservations indicated at the end of this review, I have recommended the book to any number of students and social science colleagues in order to disabuse them of notions about the balance or stability of nature (or of ecosystems) and to make them see that ecological studies are not necessarily concerned with how that balance is or can be, despite disruptions, maintained. Botkin makes the need for a new image or perception of nature the theme of his book. He says in his preface that the original impetus for this was his trying to understand and explain certain contradictions which he first confronted in the mid-1970s. What he saw then was that ideas which were widely accepted in his own field of ecology and which were evidently the basis for decisions and policies about managing nature were clearly contrary to facts. Moreover, the facts had been collected by the ecologists themselves and were known, even if not used, by advocates of 123

124 / Chapter 5

the decisions and policies in question. The conclusion to which Botkin was led is that the ideas can best be understood as variants of myths and metaphors from the past, either from the machine age and its metaphor of nature as a machine or from earlier times when nature was first viewed as divine order and the earth as an organism. Accordingly, there is exposition in his book to show both that our ideas concerning the orderliness and constancy of nature are rooted in the past and that recent and current explanations, models, and policies which Botkin sees as being based on these ideas are either falsified or at least not well supported by the facts which we now have at hand. There also are arguments in favor of new ways of managing nature, based on a new image of it as “a nature of chance” in which many separate events and processes occur simultaneously with possible effects upon one another (pp. 129–130). The case of mule deer and mountain lions on the Kaibab Plateau of the Grand Canyon’s north rim, long regarded as showing the adverse effects of human interference with a natural balance between predators and prey, is one of Botkin’s examples of widely held but unsubstantiated explanations. The account put forward by the great conservationist, Aldo Leopold, in the 1940s and then repeated as true in prominent textbooks and treatises by such ecologists as Lack, Andrewartha and Birch, E. P. Odum, and Allee and his collaborators is that “predator control” programs from 1906 to 1931, involving the killing of hundreds of mountain lions, led to an irruption of the mule deer population, which then proceeded to overbrowse and destroy its food supply so that there ensued a population crash in which 50 percent of the deer starved to death in just two winters. But Caughley’s reexamination of the case, published in 1970, showed it to be not at all clear that whatever increase in deer numbers had occurred was great enough to be an irruption or that it had resulted from the elimination of predatory mountain lions. It could just as well have resulted from an increase in edible vegetation because of drastic reductions in the 1890s and early 1900s in the numbers of sheep and cattle grazed on the Kaibab Plateau or because of changes in the weather and in the frequency of fire and other disturbances (Caughley 1970: 54–56). There are thus

Seeing Nature’s Complexity but Not People’s / 125

various alternatives to seeing the changes in deer numbers as the effects of human interference with a natural balance. Botkin reasonably asks why, when the data are so ambiguous, should one particular scenario have been promulgated and accepted by usually careful and observant naturalists and respectable scholars. He answers by suggesting that the scenario expresses or is bolstered by “deep-seated beliefs about the necessity for the existence of predators as well as all other creatures on the Earth . . . with each species having its place in the great working of the whole system” (p. 80). Further, he suggests that these beliefs have descended to us in a line that includes eighteenth century theodicy and earlier Greco-Roman views of a divinely ordered great chain of being. Botkin himself has been a pioneer in the use of computer models to simulate ecological successions in forests. In the differences between his models and those based on systemsanalysis methods borrowed from engineering, he finds another illustration of his theme of the need for a new image of nature. The assumptions underlying the systems-analysis models were those of nature as a machine: that the forest or any other natural system under study has a steady-state condition and will, like a well-designed, well-made mechanical device, seek that condition of constancy and stability. The models assumed, moreover, that descriptions of the steady state and of the processes leading away from it or back to it could be in simple mechanical terms. It was because of these nature-as-a-machine assumptions that the models had little success in mimicking or predicting what happens in forests. As Botkin puts it, “When we make nature act like a machine in a computer program, it does so exactly and the results are quite unnatural.” (p. 120). By contrast, complexly changing interactions in nature were assumed for the models developed by Botkin and his colleagues. They incorporated ideas and observations about such matters as not only how the growth of trees of a particular species, size, and age changes with the amount of light received but also how this amount is affected by changes in the size and number of competing trees nearby. The operation of chance was also made part of the models. Thus, whether a tree dies would be

126 / Chapter 5

determined by the computer’s equivalent of rolling dice, albeit the tree’s growing less than some crucial minimum in a year would greatly increase its chance of dying when the dice are thrown. The computer-generated forest stands resulting from these models have been realistically informative, showing, for example, that some old forests could indeed have had certain characteristics (such as many spruce trees at low elevations in the case of a New England forest) which were accorded them in early surveyors’ descriptions but have been doubted by modern ecologists and foresters. As for predictive use of the models, this has included showing which species of trees would continue to grow in different forests after selective cutting. Policy concerns also are important to Botkin. To bring out differences between policies which he regards as being based on old and new perceptions of nature, he considers various controversies, including those related to using fire in the management of sequoia groves in California’s national parks, trying to establish a new sea otter population along the California coast in order to reduce the risk of sea otter extinction, and “harvesting” excess animals from the elephant population in Kenya’s extensive Tsavo National Park. In the Tsavo case, when increasing numbers of elephants, protected from poachers and provided with yearround water by dams and artesian wells, were seen in the 1960s to be destroying the park vegetation on which they depended, a hands-off policy, based on the assumption that the deaths of all the weaker elephants, culled by nature, would be followed by regeneration of vegetation to support the remaining population of healthy animals, won out over some experts’ recommendations that “about 3000 elephants should be shot to keep the population within its food supply and protect the game from the dangerous effects of its intrinsic capacities for growth” (p. 17). The result, when the severe drought of 1969–1970 hit, was that there was massive starvation of elephants (some six thousand deaths), and they destroyed so much more of the vegetation that the Tsavo landscape, a decade later, in contrast to the green land outside the park, still consisted of “desert-like dusty soil with a thin scattering of live and dead shrubs and trees, among which almost no game was visible” (p. 16). Seeing rampant, illegal elephant poaching in

Seeing Nature’s Complexity but Not People’s / 127

Africa at the time of writing his book, Botkin acknowledges the danger of so many elephants being killed that their population would be unable to recover. He emphasizes, however, that the problem lies in the rate of decline and not in harvesting per se and that wise management in the face of environmental variability rather than constancy can sometimes require harvesting excess elephant populations in parks or preserves. The lessons of the elephants for him are that a hands-off policy of leaving it to nature can lead to destructive and undesirable results and that, under some conditions, it makes sense for us to promote certain changes and rates of change (pp. 18–19, 156–157). Finding most ideas of balance or constancy in nature no more useful than Botkin does, I am glad to have his book with its welldescribed cases and perceptive arguments to show that change is intrinsic and natural at many scales of time and space in the biosphere. At the same time, I cannot end this review without noting certain misgivings that I have. These relate to Botkin’s having a much simpler view of cause and effect with respect to human actions than with respect to such matters as tree growth and other events and processes in nature. There are alternatives not considered by him. The book therefore leaves one wondering, for example, whether the acceptance of particular explanations, like Leopold’s concerning the Kaibab deer irruption, does indeed result from deep-seated beliefs about the orderliness of nature. Unsubstantiated explanations or ill-founded policies do not have to be based on deep-seated beliefs in order to be accepted. Their acceptance in particular cases might be due more to ad hoc notions about the mechanisms operating, as well as to economic and political considerations and to other factors, like the authority of an Aldo Leopold or whoever else promulgates the explanations or policies. Such possibilities, as well as that of complex interplays of factors, are ignored by Botkin.101 From the efficacy which he attributes to the deep-seated beliefs, it follows that replacing them with an appropriate new image of nature—an image of an infinitely complex, intricately 101 For a fairly recent and more detailed examination of Aldo Leopold’s role in spreading an unsubstantiated view of the Kaibab deer irruption, see Young 2002.

128 / Chapter 5

and dynamically reticulating nature characterized by chance and randomness—is seen by Botkin as essential for the wise and sustainable use of the resources of the biosphere in the twentyfirst century and for the triumph of the new ecology of his subtitle. Clearly, he is hopeful that the new image can prevail. This, however, is another matter about which skepticism is possible insofar as the psychological comfort and esthetic satisfaction of a sense (or illusion) of order and permanence will not reinforce the new image but probably have been, as Botkin himself suggests (pp. 127, 130, 188), important in reinforcing the old beliefs and the images which have prevailed in this and previous centuries. In this regard, it may be significant that views dissenting from that of nature as orderly and constant have emerged repeatedly through the centuries but have always been held, as Botkin notes (pp. 9, 88), only by a minority. Historical experience, in other words, hardly shows the dissenting views to have mass appeal. Of course, if one regards broad images of nature as less critical for ecological analyses, policies, and explanations in particular cases than does Botkin, it may not matter so much that there are reasons for doubting that the kind of new image which he sets forth will prevail. In any event, disagreements with him on the changes needed or in prospect with respect to images and ideas need not interfere with our appreciation of his treatment of other changes as natural and pervasive in the biosphere.

6 Against Political Ecology (with Bradley B. Walters)

Starting with a priori judgments, theories, or biases about the importance or even primacy of certain kinds of political factors in the explanation of environmental changes, self-styled political ecologists have focused their research on environmental or natural resource politics and have missed or scanted the complex and contingent interactions of factors whereby actual environmental changes often are produced. As an alternative to the present plethora of programmatic statements on behalf of political ecology, a proposal is presented here for what may be called evenemental or event ecology. Walters’s experience in applying an evenemental approach to research on mangrove forests of the Philippines will be drawn on for the purpose of illustration.102

Why Political Ecology Now and What It’s About In anthropology and related fields, the program or movement now being called “political ecology” appears to have begun as 102 Thanks to Simon Batterbury, Neil Byron, Tim Jessup, Susan Paulson, and Tom Rudel for reading one or another draft of the article (Vayda and Walters 1999) on which this chapter is based and for suggesting changes. An early version of the article, with an extended illustration from Borneo rather than from the Philippines, was presented at the 96th Annual Meeting of the American Anthropological Association, Washington, D.C., November 22, 1997 (Vayda 1997a). The original illustration, concerning forest-product exploitation in Borneo, is summarily mentioned later on.

129

130 / Chapter 6

a reaction to certain features of human ecology or ecological anthropology as it was practiced in the 1960s and early 1970s. In particular, there was reaction to the neglect of “the political dimensions of human/environment interactions”— a neglect that some have regarded as stemming from preoccupation with homeostatic or adaptive processes and, related to this, from the treatment of human communities as if they were fairly homogeneous, autonomous units involved in, or engaging in, those processes in relation to their biotic and abiotic environments (Durham 1995: 249; Moore 1996: 125).103 As a general rule, more attention to political influences on human/environment interactions and on environmental change itself is no doubt a good thing, since such influences are no doubt often important. Many self-styled political ecologists, however, go well beyond asking for or paying more attention to such influences. Problematically, they insist that political influences— especially political influences from the outside, from the so-called wider political-economic system—are always important, arguably more important than anything else, and should accordingly be given priority in research (see, for example, Bryant and Bailey 1997: 5–7 on “putting politics first”).104 This is a prescription for 103 For a recent but not very different summary statement noted in chapter 1, p. 33, see Neumann (2005: 41), who says that what he and other political ecologists were reacting against were “neo-Malthusian formulations, neo-classical economic models emphasizing the individual rational actor and technocentric approaches.” It should be noted that the “political ecology” label has been adopted also by some overtly political movements for their promotion of alternatives to what they regard as an inherently environmentally destructive capitalist system based on constantly expanding production (see, for example, Martinez Alier, 1995). The origins of these movements seem to lie more in Marxist tradition and green environmental politics than in reaction to anything in the human ecology of earlier decades. For the sake of keeping the focus on methodological issues which the research agenda and explanatory claims of political ecologists have raised about human/ environment interactions and environmental change as objects of study and explanation, these political movements are not treated in this chapter. 104 What may be operating here is a confusion of generality with priority. If so, then, as suggested by Ellis (1997: 61) in his comments on politicized literary criticism, the same illogic that is used for inferring from a presumed political dimension in all actions that “politics is the deepest and most important consideration in any situation” may be used for making the absurd inference that each of a dozen or more other kinds of factors—physical, chemical, economic, and psychological, for example—is likewise the most important insofar as it is also an aspect or feature of all actions. Note, however, that what is said to be prioritized in versions of political ecology favored by Watts and Peet (2004: 41, note 4) is not simply politics but rather “the broader political economy.”

Against Political Ecology / 131

question-begging research, that is, for concentrating on factors assumed in advance to be important and for thus missing both other factors and the complex and contingent interactions of factors whereby environmental changes often are produced. Moreover, some political ecologists do not even deal with literally the influence of politics in effecting environmental change but rather deal only with politics, albeit politics somehow related to the environment. Indeed, it may not be an exaggeration to say that overreaction to an earlier “ecology without politics” is resulting in a “politics without ecology,” which, in violation of truth in labeling, is still billing itself as “political ecology” (instead of “natural resource politics” or simply “political anthropology” or “political science”) and is still claiming, in at least some of the many programmatic statements made on behalf of political ecology, to be seeking understanding or explanations of “environmental change” (e.g., Bryant 1992: 13; cf. Bryant and Bailey 1997: 191 and passim) or “ecological processes” (e.g., Moore 1996: 125), even though what are actually studied are political controls or political contests over natural resources and not, or at least not to any significant extent, how the resources are affected by those controls or contests. If the object of explanation is truly to be environmental change, a programmatic alternative to political ecology is what may be called evenemental or event ecology. This does not prejudge the importance of political factors but is still duly attentive to any and all kinds of them whenever they are seen in the course of research to be interesting and relevant to explaining particular environmental or environment-related events.105 An effective method is to begin research with a focus on the environmental events or changes that we want to explain and then to work backward in time and outward in space so as to enable us to construct

105 A difference between evenemental or event ecology as we are advocating it here and what Lees and Bates cogently put forward in 1984 as an “event-focused approach to ecological study” is that the latter is more concerned with human responses to environmental events than with the causes (including human actions) of those events. Although much more cognizant of environmental change than were the homeostasis-oriented ecological anthropologists of the 1960s (as represented, for example, in Vayda 1969), Lees and Bates in 1984 were still like them in drawing on environmental data primarily for the purpose of elucidating human behavior.

132 / Chapter 6

chains of causes and effects leading to those events or changes.106 By contrast, the practice of many political ecologists, presumably regarding access to resources as always politically determined and as always important for understanding or explaining environmental change, is to focus their research on such access, or on change in such access, and to pay little or no attention to actually demonstrating environmental effects. For example, Gezon (1997), reporting on her research in Madagascar, designates her study as “political ecology” and focuses almost entirely on “how people interact to establish and contest access to resources.” In another study (described in Vayda 1997b: 7–10), social scientists with a political ecology view of access were interpreting the exclusion of outside forest-product collectors from certain village lands in Borneo as a conservation measure preventing the local extinction of the Aquilaria trees from which the valued aromatic forest product, gaharu (also called aloes wood or eaglewood), may be obtained. Little or no attempt was made, however, by these social scientists to obtain data on the actual status of Aquilaria populations in their study area, and they ignored biological studies suggesting that Aquilaria extinction is not likely to occur in an area before outside collectors abandon it, regardless of whether any action to exclude them is taken by local villagers.107 106 For arguments and illustrations in support of this method, see Vayda (1996, 1998a [chap. 5 in this book]) and Rudel (1999). Deciding not in advance of research but rather finding in the actual course of it that certain political factors are relevant to explaining particular environmental or environment-related events can be illustrated by research experiences in Indonesian Borneo in 1996 and southwest China in 1998. In the Borneo case, it was found that government success in relocating forest-cutting, pepper-farming Bugis settlers from a nature reserve despite earlier reports of the settlers’ resistance to relocation could be attributed less to the decline in yields from old pepper plantations and the related factors that were initially investigated than to little-publicized military intimidation of the settlers (Vayda and Sahur 1996: 43–47). Similarly, in the course of a research reconnaissance in southwest China, it was found that plans, formulated on the basis of other researchers’ projects and reports, for a study of increases in agricultural biodiversity as a result of farmers’ decisions and actions had to be scrapped in favor of developing a project concerned not only with biodiversity increases but also with biodiversity losses as a result of political pressures to produce particular crops (Vayda 1998b). 107 A political ecology view of access has been set forth more systematically by Ribot and Peluso (n.d.), who state the following: “in tracing out from the most local structures of forest control to larger social and political-economic relations . . . access analysis demonstrates how local physical access to forests is shaped by control over market access and labor controls embedded in relations of authority and social identity.” This view of access, presented also in Ribot (1998), is entirely consistent with the following characterization of political ecology by Peluso: “political ecology emphasizes the social relations within which actors are embedded and which affect the ways they use the environment . . . [and] political ecology assumes that larger social structures and political-economic processes will affect the actions of local resource users. . . . These structures and processes are examined in a more systematic manner, therefore.” (Peluso 1992: 51; italics hers).

Against Political Ecology / 133

Furthermore, some political ecologists have put their practice into the service of a populist political agenda and the “green romanticism” of thinking that devolving control over resources to local communities so as to mitigate certain influences from the wider political economic system must effect more sustainable use of the resources. For these political ecologists, changes resulting in greater community control over resources therefore rank high on both political and research agendas. These points will be discussed further in relation to the Philippine case to be described.108

The Rationale for Event Ecology as an Alternative Why is evenemental or event ecology being proposed here as an alternative to political ecology? For one thing, the proposal is consistent with both philosophical and practical arguments for letting research be guided more by open questions about why events occur than by restrictive questions about how they are affected by factors privileged in advance by the investigator. But it should be understood that being guided by open questions does not mean considering all conceivable causes or else having the license to consider any conceivable event or action as a cause, no matter how remote from the events to be explained. What it does mean is taking ourselves, either actually or by means of thought

108 Also bearing on these points is some Amazonian evidence of the failure of community regulation for sustainable use (e.g., Rudel 1995: 504, citing Coello Hinojosa 1992, on the Ecuadorian Amazon; Anon. 1993 and Conklin and Graham 1995: 703 on the Brazilian Amazon; cf. Vayda 1997b and in chapter 4, [based on Vayda 1998a], both also citing this evidence). That there are exceptions to the green romanticism of political ecologists with a populist political agenda is indicated by Li (1996: 501), who argues for consciously misrepresenting communities as “sites of consensus and sustainability” in order to “strengthen the property claims of potentially disadvantaged groups.” Contrary to what is implied by Watts and Peet (2004: 17–19), the article on which this chapter is based (Vayda and Walters 1999) contains no statement or argument to the effect that green romanticism and a populist political agenda are general or universal among political ecologists. Nor does it contain any statement or argument against a view increasingly favored in conservation circles, namely, that a more equitable distribution of conservation costs and benefits is needed to make community regulation effective for sustainable resource use in conservation programs (Eghenter 2006; cf. Brechin et al. 2003). For discussion of which attributes of resources and resource-users are more likely to lead to community regulation of forest resources for sustainable use and which attributes are less likely to do so, see Ostrom 1999 and cf. Pagdee et al. 2006.

134 / Chapter 6

experiments, to the time and place of those events and then asking ourselves what antecedent events occurring then and there could have brought about the outcomes of interest to us and could have kept things from turning out differently. In other words, the possibilities we consider should not be confined to those prescribed by any single or simple agenda or theory, but, at the same time, the causes we consider and seek evidence for should, as Moore (1978: 377) has suggested, be concrete events specific to concrete situations. There is a lot to be said in favor of being guided more by open questions in our research, and more will indeed be said in the course of the description and discussion of the illustrative research in the Philippines.109 There is, however, a second general point to be made about the proposal in light of all the hype and hoopla in social science about political ecology as an emerging field or subfield. The point is that the proposal, unlike political ecologists’ programmatic statements, conforms to the heuristically productive practice of distinguishing fields or subfields on the basis of what is to be explained (in this case, environmental events or changes) and not on the basis of a priori judgments, theories, or biases about what will do the explaining.

An Illustration from the Philippines This illustration is drawn from field research done by Walters in 1997 on the causes and consequences of mangrove forest planting and cutting in Bais Bay and Banacon Island, Philippines. Mangroves are a class of trees that grow in sheltered, intertidal areas throughout the tropics. Mangroves are recognized as ecologically important because they constitute fisheries and wildlife 109 More has been said also in such other publications as Vayda 1996, 1998a; Vayda and Sahur 1996: 50–51; Walters et al. 1999; and Walters and Vayda 2009. The kind of point made by Moore about the concreteness and situation-specificity of the causes appropriate for us to consider is discussed further by various authors concerned with the explanatory use of counterfactual analysis (see, for example, Fearon 1996: 66; Griffin 1993: 1101–1104; Hawthorn 1991: 107, 187, and passim; Tetlock and Belkin 1996: 7–8, 23–25; and Weber 1949: 165 ff.).

Against Political Ecology / 135

habitat and build land and protect shorelines from erosion. Because human settlements tend to concentrate along coastlines, mangrove forests are often harvested for wood and fish/shellfish products, serve as storm buffers to property, and are frequently cleared for settlement and aquaculture ponds (Baconguis et al. 1990; Primavera 1995). The widespread degradation and wholesale clearing of mangrove forests is commonly cited as one of the Philippine nation’s most pressing environmental concerns. Changes to forestry and coastal resources policy in the Philippines in the 1990s emphasized the need to devolve greater management authority from state agencies and large, commercial interests to local communities and households (DENR 1993, 1996; Gibbs et al. 1990; Pomeroy and Pido 1995). To achieve this, the Department of Environment and Natural Resources (DENR) turned to household and community-based tenurial instruments that give leaseholders usufruct control of forests, including mangroves, for stipulated periods of time (usually twenty-five years). Underlying this shift in policy emphasis to more decentralized management was the arguably important social goal of promoting greater equity in access to the nation’s natural resources. At the same time, it was assumed that devolution to the local level would lead to more effective, sustainable forest management. The twin plagues of Philippine development—inequality and environmental degradation—were thus seen as being addressed simultaneously. The political issue of decentralization of control over forest resources is certainly an important subject and presumably would be of considerable interest to persons who study power relations in the Philippines, especially as such relations influence the environment. For example, Broad and Cavanagh (1993: 74–80), in examining environmental changes in the Philippine provinces of Bataan and Negros Occidental, describe how elites, trying to further enrich themselves and supported by exportoriented national development policies, destroyed vast tracts of mangroves for the development of capital-intensive prawn farms. The authors suggest that the poor fisherfolk living in these coastal areas were more inclined to be concerned about the environment because of their direct reliance on coastal resources

136 / Chapter 6

for their livelihoods. Implicit in their analysis is the assumption that environmentally destructive outcomes are more likely to be averted if the local fisherfolk are endowed with more secure access control over mangroves and other coastal resources. As the following illustration will show, some of the sites studied by Walters were similarly characterized by highly unequal distribution of land, and this had interesting implications for the mangroves there. However, the actual environmental effects of this distribution were more complex than would be expected from the analysis by Broad and Cavanagh (1993), and were often inconsistent with their underlying assumptions about the causes of mangrove degradation and conservation. The impetus for Walters’s study came from observations of important environmental changes in mangroves in two specific sites in the Philippines: Bais Bay in Negros Oriental and Banacon Island in Bohol. Natural mangrove forests in both sites had been dramatically reduced in distribution since the Second World War, and remaining natural mangroves were highly degraded, apparently from human use (de Leon et al. 1991; Walters 1995, 1997). At the same time, mangroves in certain areas were actually showing evidence of expanded distribution, apparently the direct result of local peoples’ planting and protecting mangrove trees (Cabahug et al. 1986; Walters 1995). The research was thus guided by the desire, first, to assess more precisely the nature of the environmental changes in question and, second, to understand the actual causes of these changes. Answering these questions called for considering the potential causal influence of socioeconomic and political factors, including resource access and tenure. Unlike most studies in political ecology, however, the analysis began with a careful consideration of the actual environmental changes that required explanation and then worked outward in space and backward in time in search of relevant causal influences. For example, between the 1940s and 1970s, wealthy landowners with property adjacent to the coast, as well as several ambitious low- and middle-income entrepreneurs, cleared large tracts of mangrove forest to develop fish ponds to raise milkfish (Chanos chanos) for domestic markets. Some of these ponds were

Against Political Ecology / 137

later converted to prawn production during the prawn boom of the 1980s, though most have since reverted to milkfish. Thus, mangrove deforestation in Bais antedated the capital-intensive and export-oriented political economy of prawn farming. Furthermore, some of these elites subsequently invested heavily in planting and protection of significant stands of mangrove forest on the perimeters of their properties. In fact, with the exception of a local ecological reserve, the largest and arguably best protected stands of mangrove forests in North and South Bais Bay were managed by local elites and entrepreneurs who acknowledged the importance of having mangroves to protect their fishponds and lands and who had the means to plant and effectively protect large areas. Poor fisherfolk, many of them landless, also claimed small areas near their homes for planting mangroves for bunsod (fish corral) poles, storm protection, and tenure security. Government extension agents acting under the auspices of aforementioned policies facilitated this process by allocating mangrove stewardship tenure contracts to many such households in Bais and elsewhere. Planting in these areas was, nonetheless, very uneven in space and over time. For one, many sites were simply not ecologically suitable for mangroves, and repeated efforts to plant in these areas failed for this reason. The provision of tenurial instruments to fisherfolk in areas ecologically unsuitable for planting proved to be a largely futile exercise in mangrove management.110 The situation was further complicated by the fact that local peoples’ claims to mangroves were often more about claims to land than about the trees standing on them (Walters 1998b). For example, we documented numerous cases of mangrove sites being claimed and planted but subsequently cut by fisherfolk who had attained sufficient capital to convert their mangrove areas into what they viewed as economically more productive uses (fishponds, house sites, etc.). In at least one case, a small group of landless persons claimed a mangrove area, cut the trees there, 110 These findings mirror experiences from other sites in the Philippines where tenurial instruments were used as a basis for local mangrove reforestation and management (Primavera, pers. comm.).

138 / Chapter 6

developed the basic infrastructure on the site (dikes and drainage canals), and then sold the pond to a person who could afford the capital investments needed for a functioning aquaculture operation. In short, there is clear evidence that many local peoples’ planting and protection of mangroves, however apparently compatible with conservation in the short term, was often motivated mainly by other concerns, such as wanting to establish land claims and associated development rights that might otherwise be taken by neighbors or outsiders.111 Attaining security of tenure had important implications for mangroves in Bais, but by no means did it assure successful planting and conservation of mangroves by either poorer or richer households. So far the focus here has been on instances of the clearing or creation of entire stands of forest. The research also revealed that the composition and structure of many existing forest stands were being altered as a result of widespread cutting for fuelwood and construction materials. For example, ecological surveys of mangroves revealed the presence of as many as fifteen different species of trees, although sites differed dramatically in terms of species composition and only five or six species were ubiquitous across many sites. These same surveys also revealed that cutting of individual trees for fuelwood and construction materials was the most common form of micro-structural disturbance in both natural and planted forests and that, among other effects, this cutting had reduced the mean live basal area of forests by nearly half. Subsequent interviews of local mangrove users confirmed these general observations and showed that, while virtually all species were cut for fuelwood, only certain species were cut for use as construction materials. 111 Just as similar motivations, rather than conservation, were behind some attempts to establish the marine reserves—for example, in Papua New Guinea (Polunin 1984: 273)—that were prematurely hailed as Pacific Islanders’ “traditional conservation measures” (see, for example, Johannes 1978 and 1981: chap. 5). Similarly, in a consideration of whether conservation was practiced by certain Dayak villagers in Borneo rainforests, it was found that the people were restricting access to their territories not, as some observers have assumed, for the sake of conservation, but rather for the sake of reserving to themselves the profits from intensified exploitation of particular forest products in demand by traders at particular times (Vayda 1997b: 7; cf. the discussion above on restriction of access to village lands in Borneo).

Against Political Ecology / 139

The most economically valued uses of mangrove wood in both study areas were for posts in fish trap, fish corral (bunsod), and home construction. Bunsod, in particular, were abundant in both areas and there were active, albeit highly informal markets for buying and selling bunsod posts. In Banacon, mangrove posts were also bought and sold within and to neighboring islands for construction of homes, fences, and the like. For this use, unlike the use of wood for fuel, there was a clear preference for “bakau” (Rhizophora sp.) because the wood is particularly strong and durable and the trees tend to grow straighter than most other common species. One consequence of this was that many persons, having made claims to mangrove areas for the purpose of plantation establishment, “high-graded” them subsequently for bakau by cutting back and weeding out less valued species. As well, the highly unusual reproductive biology of bakau makes it relatively easy to propagate and plant compared to other common trees, including “pagatpat” (Sonneratia sp.), “piapi,” and “bungalon” (Avicennia sp.). Bakau are viviparous, which means that the seeds germinate and elongate into stems while still attached to the mother plant. These elongated propagules are easily detached when ripe and will quickly sprout roots and leaves when stuck several inches deep into mud in a suitable environment. R. mucronata, in particular, fruits profusely and so tends to be collected and planted in much larger quantities than R. apiculata. No doubt it was because of its economic value and ease of planting that bakau dominated most managed forest stands. In fact, most mangrove areas claimed, planted, and managed by local people tended to be virtual tree monocultures of R. mucronata, in contrast to even heavily cut, unmanaged mangrove areas, which tended to be characterized by three or more tree species (Walters 1998a). This finding has important ecological implications because it suggests that the proliferation of local planting and management may exact a cost in terms of reduced species diversity, even if it leads to local expansion of mangrove forest distribution. Equally interesting is the clear influence of ecological and biological factors—specifically, the varied distribution and reproductive biology of different species—on subsequent patterns of mangrove utilization, planting, and management.

140 / Chapter 6

In summary, research on mangroves in Bais Bay and Banacon Island has shown that human influence on these forests is substantial.112 Efforts to explain particular environmental changes— including specific cases of forest expansion or contraction and changes in forest structure and species composition—revealed the causal influence of a variety of events, including some to which political ecologists would have been likely to have been attracted for the purpose of explaining environmental changes. The expansion of aquaculture and the clearing of mangroves for settlement by landless are examples of such, although the case study should serve as a caveat to political ecologists who are inclined to make generalizations about the relationship between such factors as wealth and income distribution and environmental change. Other environmental changes of interest were found to be caused by events that political ecologists would quite likely have overlooked or ignored. For example, probably the most widespread class of events limiting the spread of mangrove planting in Bais Bay is biophysical in nature: young mangrove plantations were frequently destroyed by natural causes, including wave damage, pest infestations, and entanglement by floating seaweeds and other debris. Similarly, the explanation of other environmental changes, such as the establishment of many plantations that are monocultures of one tree species, calls for referring first to species-specific selection in planting and high-grading by mangrove planters, and then, to account for these practices, referring to decision-making on the basis of differences in not only the local market value but also the reproductive biology of tree species.

Conclusion As shown both by arguments above and by the Philippine case study and other illustrations, the concerns expressed in this 112 For recent articles presenting detailed findings from the research, see Walters 2003, 2004, 2005a, 2005b.

Against Political Ecology / 141

chapter regarding political ecology are not about whether political and political-economic events can be important causes of environmental change. Obviously they can and often are. Rather, criticism has been directed at those political ecologists who choose to privilege such factors in certain ways. One way consists of always attributing special causal significance to these factors and not admitting that other factors are, or may be, more important sometimes. A more extreme way consists of focusing research on political events (as in the access studies to which we have referred) and not at all or hardly at all on environmental ones and then, without further ado, using the research to make claims about the political events as causes of environmental events or changes, which, alas, are only assumed rather than demonstrated. The evenemental or event ecology proposed as an alternative to political ecology as a research approach offers two important advantages. First, it requires a more careful appraisal of actual environmental changes and thereby lessens the likelihood of inaccurate or erroneous claims about the environment on the basis of preconceived theoretical ideas or agendas. An example that has been given of such claims is Broad and Cavanagh’s (1993) about differences in wealth and power as factors in environmental destruction in the Philippines. As discussed, this claim was not consistently supported by the case-study evidence. Both rich and poor were involved, in complicated ways, in acts of both mangrove restoration and mangrove destruction. There is a second important advantage. Being guided more by open questions about why events occur than by restrictive questions about how they are affected by factors privileged in advance by the investigator, event ecology does not prejudge political factors to be the most important or even important at all in the case at hand, although it is still duly attentive to any and all kinds of political factors—just as it is attentive to any and all kinds of biological or physical factors—whenever they are seen in the course of research to be interesting and relevant to explaining particular environmental events. Thus, in the Philippine case study, there was due attention paid to political events, but biophysical events were found to be sufficient to explain why plantations had not

142 / Chapter 6

spread to many sites; biophysical events helped also to explain why plantations are so dominated by one species. Such findings are in line with the point made earlier about the heuristic advantages of distinguishing fields or subfields on the basis of what is to be explained (in this case, environmental events) and, contrary to the practice of political ecologists, not on the basis of a priori judgments, theories, or biases about what will do the explaining.

7 Failures of Explanation in Darwinian Ecological Anthropology

This chapter is a review of Evolutionary Ecology and Human Behavior, edited by Eric Alden Smith and Bruce Winterhalder. (New York: Aldine de Gruyter, 1992, pp. xv + 470.)

The volume entitled Evolutionary Ecology and Human Behavior, published in 1992, was intended as a “unified, coherent, and comprehensive” survey of theory and research in a field that the editors regard as having begun to coalesce in the latter part of the 1970s (pp. xiv–xv). That field is a particular version of the study of human social behavior “within a simultaneously evolutionary (selectionist) and ecological perspective” (p. xiii). It is a field in which the number of workers and the number of their projects and publications grew markedly after the 1970s, thus making appropriate not only such a survey volume as Evolutionary Ecology and Human Behavior and review articles written by those in the field (e.g., Borgerhoff Mulder 1991, Cronk 1991, Smith 1992a and 1992b) but also critical appraisals, like the one here, by those who are not in the field. My focus in this chapter will be mainly on methodological or, more specifically, explanatory issues, and I will rely heavily on the editors’ essays and commissioned articles in Evolutionary Ecology and Human Behavior, but I will refer as well to other works by contributors to the volume and their congeners. Focusing on these 143

144 / Chapter 7

issues with these particular works as my texts will entail addressing some issues of explanation in social and biological science in general, including issues still very much with us today concerning the explanatory use of theories, generalizations, models, predictions, causal statements, functionalist claims, and narratives.113 As indicated by the editors and other contributors to Evolutionary Ecology and Human Behavior, their having an evolutionary perspective means “fully embracing” what is referred to as “Darwin’s central tenet that differential reproduction is the primary force shaping biological adaptation and diversity” (p. xiii). Ostensibly consonant with this, their evolutionary perspective also means believing—making it “virtually axiomatic,” according to Foley (p. 137), one of the contributors—that human phenotypes, including behavior or behavioral traits, can be somehow accounted for in terms of their benefits and costs in fitness as understood by evolutionary biologists, that is, in terms of what the phenotypes contribute either to actual reproductive success or to the propensity to survive and leave offspring in a particular environment (see the editors’ discussion of fitness on p. 27, citing Mills and Beatty 1979; cf. Beatty 1993 on the pros and cons of the “propensity interpretation”). The editors, rather grandly, see themselves and their colleagues as “unleashing the power of Darwinism” to deal with these phenotypes (p. xiii). What makes their perspective ecological as well as evolutionary is their mindfulness of environmental variability and their consequent concern to relate environmental variations to expressions of phenotypic or behavioral plasticity. In other words, they are interested in including environmental variations in their explanations of what would simply be called “behavioral variations” by some less evolutionary-minded anthropologists. Research within the “simultaneously evolutionary and ecological” perspective typically proceeds at the hands

113 Treatment of explanatory issues in more recent review articles by those in the field (e.g., Smith 2000, Smith and Winterhalder 2003, both cited in chap. 1 of this book) is little changed from the treatment in the articles of the 1990s and includes no response to my critical comments in the two-part article (Vayda 1995a, 1995b) on which this chapter is based.

Failures of Explanation in Darwinian Ecological Anthropology / 145

of contributors to the volume by means of a “model-based methodology” (p. xiii), that is, resorting to so-called evolutionary ecology models for hypotheses or predictions to be tested. A typical model is concerned with what the editors refer to as a “particular behavior” (p. 7) and with how it should vary as a result of environmental variables that affect the fitness benefits and costs accruing from the behavior in question. In fact, the “particular behaviors” in models referred to in the volume are sometimes quite general classes or categories of behavior, as general, or almost as general, as those featured in the titles of some of the topical chapters (e.g., “Food Acquisition,” “Sharing and Collective Action,” and “Competition and Conflict”). For example, the “prey choice” model that will be discussed later is concerned with how food-getting behavior, as regards foods exploited and not exploited, should vary as a result of environmental variables which, in affecting food-getting efficiency, are assumed to affect fitness.

A Unified Field? The editors provided no concise, unambiguous name for their purported field in their 1992 book. It was only in later years that the label “human behavioral ecology” (HBE) became established. Accordingly, for the sake of convenience in the article on which this chapter is based, I adapted Symons’s usage (1989a, 1989b) in his critiques of what he called “Darwinian anthropology” and I used the label of “Darwinian ecological anthropology,” abbreviated to DEA, both for the field, as characterized by its practitioners and shown in their work, and for the practitioners themselves. I am using it again in this chapter and am doing so with full cognizance of two points to be taken up later, namely, that, as Symons (1989a, 1989b, 1992) also suggests, DEA methods of analysis and explanation (and the assumptions evidently underlying them) diverge significantly from those employed by Darwin himself and that contextual and historical/narrative approaches constitute other, less problematic ways of studying

146 / Chapter 7

human behavior with due regard for its evolution and for environmental factors.114 The methodological or, more specifically, explanatory issues that are the main focus of this review are my main grounds for considering alternatives to DEA and questioning whether it should be regarded a unified field that can make its own distinctive and important contributions to the analysis and explanation of environment-related human behavior. However, before those issues are taken on, it should be made clear that non-DEA human ecologists will find much that is informative and useful in Evolutionary Ecology and Human Behavior, especially in the nine topical chapters that follow the first three chapters in which consideration of methods and concepts is highlighted. Two of the topical chapters deal with primate social structure in relation to ecology and with the behavioral ecology of early hominids. In the remaining seven chapters, constituting more than half of the book, the topics treated are: food acquisition; time allocation; spatial organization and habitat use; sharing and collective action; competition, conflict, and development of social hierarchies; reproduction, parental care, and mating systems; and population dynamics in relation to change in resource availability. All nine of these topical chapters are valuable for their reviews of certain findings concerning their topics and also, but to a more limited extent than most of the authors seem to realize, for their reviews of models guiding the collection and interpretation of data on each topic. Contrary, however, to what DEAs might think, the utility of these reviews does not attest to (or derive from) the distinctiveness and unity of DEA or its having a distinctively and justifiably Darwinian explanatory framework. Similar charges of lack of unity have been leveled against human sociobiology. This broader, alleged field is distinguishable from DEA by virtue of its not necessarily making, as the editors put it, “environmental variation an essential element in 114 Although the “human behavioral ecology” label was already being used by some DEAs in the early 1990s (e.g., Borgerhoff Mulder 1991, Cronk 1991, Smith 1992a and 1992b), I did not then expect it to be ceded to DEAs by non-DEAs who, like myself, were also studying interrelations of environment and human behavior. Since, despite my remonstrations (Vayda 1995a: 244, note 1), it has been so ceded, I do use the HBE label in my chapter 1 of this book, even if not in the present chapter.

Failures of Explanation in Darwinian Ecological Anthropology / 147

the explanation of behavioral variation” (p. xiv). Still, much of what some prominent critics said some years ago about human sociobiology—for example, that it has no rigorous central core (Kitcher 1985c) and is united mainly by its practitioners’ common commitment to a flawed, cardboard Darwinism whereby some perceived current utility of behavior is declared its raison d’etre (Gould 1987: 31, 48)—might be thought, at least by some, to apply to DEA as well. It is somewhat surprising therefore that no direct response to these critics is to be found in Evolutionary Ecology and Human Behavior.115 Many or most DEAs would no doubt regard the aforementioned “Darwin’s central tenet” as constituting the core of their field and would see the field’s unity as deriving in large part from the consistent application of this tenet in the study and explanation of variations in human social behavior. Is it, however, really applicable in any interesting and important way to explaining particular behavioral variations? The answer must, I think, be “no,” for reasons to be discussed. First, however, it may be well to note that answering “no” does not entail a categorical exclusion of Darwinian selection from explanations of human behavior and of the mechanisms that produce it. More particularly, it does not mean denying the importance of certain evolved characteristics of the Homo sapiens brain, presumably the products of natural selection, for making possible our behavioral plasticity and for explaining the fact of plasticity itself. Nor does it require a rejection of the views of Symons and other self-styled evolutionary or Darwinian psychologists (DPs), who have been arguing that the human mind or brain is endowed not only with a few general-purpose mechanisms (e.g., for operant conditioning, social learning, and trial-and-error induction) but also with many more specialized mechanisms or modules (e.g., for acquiring grammar, for perceiving sexual attractiveness, for reasoning about social exchange, and for recording and predicting frequencies of events) and that this is a result of natural selection during and before the

115 In a volume on human behavioral ecology published in 2000, there was still no response but, instead, a celebration of the fact that such criticisms had abated (Irons and Cronk 2000: 12)—as if that showed that the critics had been won over and not that they recognized the futility of endlessly repeating unanswered criticisms.

148 / Chapter 7

two million years that our ancestors spent as Pleistocene huntergatherers (Cosmides and Tooby 1987, 1992, and n.d.; Cosmides, Tooby et al. 1992; Tooby and Cosmides 1990, 1992; Symons 1992). We might even consider that, as Ryan (1991: 22) has suggested, there was selection during our remote hunting-and-gathering past for mechanisms for following norms (cf. Richerson and Boyd 1989: 209 and their chapter in Evolutionary Ecology and Human Behavior, pp. 82–83, 86, 88) and that this might help to explain why, even in our contemporary world of heightened social and geographical mobility, people are still switching back and forth between actions based on rational calculation or narrowly construed self-interest and actions based on compliance with norms at odds with self-interest, as was found by Elster (1989a) in his examination of labor-management negotiations in Sweden. This of course is highly speculative, and certainly there is, as some DEAs have argued (e.g., Borgerhoff Mulder 1991: 97–98, and others cited in Smith 1992b), a need for more and better evidence to support some other DP claims.116 Moreover, when our aim is to explain present-day behavioral variations, it makes sense to appeal first (and sometimes only) to contextual and historical factors closer in time to our explananda and therefore, as a rule, better known than are the Pleistocene living conditions posited by DPs to have resulted in natural selection for the mechanisms that make possible the present-day behavioral variations. It should be understood therefore that my object in referring to DP at this point is to suggest not that it should always, or even usually, be included in our explanations of human behavior but only that not all claims of using Darwinism to explain such behavior are problematic in the same ways and to the same degree as are the claims made by DEAs. Later in this chapter (p. 171), more will be said about differences between DP and DEA. 116 For a recent powerful critique of DP on various grounds, including its undue speculativeness, see Richardson 2007. For criticism of DP claims of “massive modularity” in the human brain, see also Buller and Hardcastle 2005 and Fodor 2000. For recent economic experiments and ethnographic studies of the extent to which norms of fairness and reciprocity at odds with narrow self-interest are observed and sanctioned, see Henrich et al. 2004, Gintis et al. 2005, and, questioning the extent to which prosocial behavior in experiments reflects behavior in the “real world,” Gurven and Winking 2008.

Failures of Explanation in Darwinian Ecological Anthropology / 149

Nomological and Causal/Mechanical Explanations From consideration of DEAs’ proffered or implied explanations, we can see problems not only with their claims of using Darwinism to explain behavior but also with their more basic claims that behavior is indeed explained by their advocated procedures. When non-DEAs want to explain human behavior, they may try to ascertain some or all of the following and to use them as explanatory factors or explanantia: the intentions, purposes, knowledge, and beliefs on the basis of which particular actions are undertaken; the events and processes whereby those intentions, purposes, knowledge, and beliefs were formed; the ways in which actions and the ideas behind them are related to the contexts or environments in which they occur; and, of special interest to DPs, the psychological and neurophysiological mechanisms that trigger particular behavior in particular contexts. Apart, however, from having, as previously noted, an interest in covariations of environment and behavior, most DEAs are not interested in any of these possible explanantia. Indeed, as indicated in Evolutionary Ecology and Human Behavior in the first of two chapters by the editors (pp. 11, 23), some DEAs do not regard them as answering why-questions about behavior at all but regard them instead as providing, at best, answers to how-questions, that is, questions about how fitness-conferring behaviors, seen as adaptations, function. In the exclusionist view of these DEAs, the only why-questions are questions about the “evolutionary origins” of behavior and, paradoxically, these questions can be answered ahistorically, that is, by means of indicating fitness payoffs from the behavior in question and then taking these payoffs as somehow constituting evidence of the operation of natural selection. To other DEAs, inquiries concerned with the kinds of explanantia noted above are studies of the “proximate causes” of behavior. Towards such causes, these DEAs profess the same agnosticism as was set forth in the late 1970s in such works on human sociobiology as Alexander’s (1979) and Chagnon and Irons’s (1979). In a chapter on these works in his 1985 book, Kitcher first described reviewers’ puzzled reactions to this

150 / Chapter 7

agnosticism and then presented his own trenchant analysis to show that the agnosticism is unjustified (Kitcher 1985c: chap. 9).117 Coeditors Winterhalder and Smith, alone among the DEAs represented in Evolutionary Ecology and Human Behavior, do mention Kitcher’s criticisms (pp. 47–48), but they then go on to cite various authors, including Smith (1987) himself, in favor of focusing not on proximate causes and mechanisms but rather on fitness costs and benefits. The conjunction of DEAs’ professed goal of explaining behavior with their lack of interest in its so-called proximate causes and mechanisms must seem especially perverse to those who, like myself, are inclined toward what has been called the causal/mechanical view of explanation (Salmon 1990: 183; Brandon 1990: 160). According to this, explanation consists primarily of showing causes rather than, as was generally thought to be the case by philosophers of science some decades ago, citing laws or generalizations. No awareness of alternative explanatory approaches is evinced by DEAs in Evolutionary Ecology and Human Behavior. They either commit themselves unquestioningly to the old nomological approach of citing presumed laws or generalizations to subsume observable phenomena—for example, as will be further discussed later, treating natural selection as a law and invoking it to subsume observed cases of apparently optimizing or maximizing behavior (pp. 168, 237, 325)—or else they simply follow certain procedures, such as setting forth models and using cost/benefit data to test them, without much evident thought to the view of explanation that the procedures imply. The changing conceptions of explanation to which DEAs seem so oblivious are decidedly pertinent to assessing their claims and their work and therefore merit some further discussion here.118 One point to be noted is that, according to some phi-

117 An example of misplaced charity in this regard: The distinguished evolutionary biologist Maynard Smith’s statement that he found it “hard to believe” that the sociobiologists are right, but they are at least saying testable things (Maynard Smith 1985: 122; cf. Maynard Smith in Runciman 1996: 292–293). 118 On such obliviousness, see also chapter 1, note 5.

Failures of Explanation in Darwinian Ecological Anthropology / 151

losophers, the nomological and causal/mechanical approaches need not be regarded as mutually exclusive. Thus, Railton, while emphasizing the role of laws, turns to the actual explanatory practices of scientists and notes that the “best-developed explanations in physics or chemistry textbooks and monographs . . . typically include not only derivations of lower-level laws and generalizations from higher-level theory and facts, but also attempts to elucidate the mechanisms at work” (Railton 1981: 242; italics his; cf. Railton 1980, cited in Salmon 1990: 164). Salmon (1990: 180–186), Brandon (1990: 161), and Schwartz (1993: 284, note) all agree with Railton about the compatibility of the nomological and causal/mechanical approaches, but Schwartz (1993: 284, note, and 288) makes an additional point important for evolutionary and ecological studies, namely, that causal/mechanical explanations do not require laws—“which is convenient in social and biological science where we often do not have them” (cf. Irzik 1990; Bhargava 1992: 106–109). Although advocates of causal/mechanical explanation may not agree on its details, it may be generally characterized as involving the elucidation or specification of causal relations among events (including actions and mental events) leading to the explanandum.119 As already suggested, this means attention to precisely what DEAs ignore: proximate causes of events in causal chains. In the case of cost/benefit explanations, which are favored by DEAs and which have been called consequence explanations by Elster (1983b: 101ff.; 1986a: 202) and others, it means attention also to the puzzle of how the explanans can be preceded by the explanandum in time or, in other words, how something can be caused by its effects—how, for example, behavior can be caused by the balance of benefits and costs resulting from it. For adherents of causal/mechanical explanation, one possible solution to the puzzle consists of showing that the effects were brought about by actions intended—by the actors themselves or by others shown to have influenced or manipulated the actors—to bring about those effects. Recourse to this solution

119

See chapter 1 on other ways in which some causal explanations may be character-

152 / Chapter 7

means, in effect, treating intentions as proximate causes, a subject on which there is a burgeoning philosophical literature. (For an introduction to some of the issues, see Bratman 1987: chap. 1; Polkinghorne 1983: chap. 5; Searle 1984: chap. 4; and Searle 1991.)120 As I have discussed elsewhere (e.g., Vayda 1987a, 1987b, 1993), it also means reserving for ourselves the option of trying to determine causes of the intentions that we have identified. As Elster (1986a: 202f.) notes, another kind of solution to the puzzle may be applicable if the explanandum is not a oneshot event. Causal/mechanical explanation may then consist of showing how particular occurrences of certain behavior at particular points in time have (or had) consequences causing the same (or similar) behavior to occur at later points in time. Actually demonstrating (rather than simply assuming or asserting) the operation of certain causal mechanisms, or types of causal chains, such as natural selection and reinforcement or operant conditioning, with respect to the behavior to be explained would be examples of this kind of explanation. How strong, direct, or complete must be our evidence for causal chains extending from beneficial consequences as explanantia to behaviors as explananda in consequence explanations is a matter of some dispute among adherents and expositors of causal/mechanical explanation. Some (e.g., Kincaid 1990: 350–351; Schwartz 1993: 288, 289), charging Elster with having been, at times, unduly exigent in this regard, have suggested that it suffices to have only enough evidence to give us reasonable assurance that some causal chain or other does extend from the beneficial consequences to the behavioral explananda under consideration—evidence that would, in other words, assure us that the beneficial consequences are not mere by-products of behavior. Being less exigent than Elster does not, however, mean taking lightly what qualifies as such evidence and how difficult it might be to obtain. To show this, we cannot turn to DEA studies, since, in line with DEA agnosticism about proximate 120 For a more recent review of issues debated by philosophers under the heading of “mental causation,” see Robb and Heil 2008.

Failures of Explanation in Darwinian Ecological Anthropology / 153

causes, the need to deal with the by-product possibility when proffering consequence explanations of behavior has generally been disregarded by DEAs.121 The possibility is, however, one that has received some attention from non-DEA human ecologists, including myself (Vayda 1987a: 502–507 and 1989: 163–164, 170–171), and it may be worth noting that some have tried to deal with it by recourse to comparative studies and “design” analyses similar in some respects to the procedures which some evolutionary biologists have recommended and used for dealing with the problem of distinguishing beneficial consequences that are by-products or the effects of chance from those that occur because traits have been fashioned by natural selection to produce them (see, for example, the discussions in Curio 1973; Harvey and Pagel 1991: 13ff.; Mayr 1983: 329; Palmer 1991: 366; Thornhill 1990: 31–33; Tooby and Cosmides 1992: 61–63; WestEberhard 1993; Williams 1966: 9–10; and Williams 1992: 40–44, 101–105). For the purpose of illustration at this point, it will suffice to present a single, extended example in which the geographer Waddell (1972, 1973, 1975) uses the beneficial consequence of protection from frost to explain certain customary practices of some of the Enga people of New Guinea’s central highlands.122 The particular type of frost to which these people’s crops were vulnerable is radiation frost, consisting of a marked inversion of the temperature regime close to the ground’s surface when there is excessive outgoing radiation on clear, still nights; the practices which Waddell explains by referring, inter alia, to the protection they afforded from such frost are the specific construction, design, and use of large mulch mounds, more than a half meter in height and three meters in diameter, for cultivation of the people’s staple sweet potatoes at higher altitudes. The benefits of the 121 What may seem, at first, an exception occurs in an article by Caro and Borgerhoff Mulder (1987: 66). However, their recognition that “specific behavioral traits may have arisen through pleiotropy, drift, or cooption of function” is, rather perversely, made a basis for arguing not that the causal histories of behavioral traits should receive more attention but rather that they should be ignored because they may be complicated. For further discussion of Caro and Borgerhoff Mulder’s article, see pp. 159–160. 122 As an example useful for arguing various points, it is cited also in chapters 1 and 3 of this book.

154 / Chapter 7

practices for protection from frost were indicated to Waddell by such data as differences in temperature readings on unmounded ground and on the upper parts of mounds (Waddell 1975: 252). By virtue of the mulch in them, the mounds contributed also to maintaining soil fertility (Waddell 1972: 159–161), but Waddell’s informants referred neither to this, nor to protection from frost, as a reason for their mounding practices; although they believed that tubers would not develop in unmounded ground, they were “not, strictly speaking, able to say why sweet potatoes are only planted in mounds” (Waddell 1972: 136). This meant that referring to the informants’ intentions was of little avail in this case for disposing of the possibility that protection from frost was only a by-product of practices that had developed for maintaining soil fertility or for achieving some other effect unknown to us. Waddell was, however, able to obtain other data to support his inclusion of protection from frost in a consequence explanation of specific features of mounding at higher altitudes.123 Some of these were data from comparative studies, showing that the spatial distribution of mounding practices, both among Enga groups and among other New Guinea highland peoples, closely coincided with the distribution of the frost hazard (Waddell 1975: 254, citing Brookfield 1962: 250). The thrust of other data which Waddell obtained was to show that the construction, design, and use of mounds were so efficient and precise solutions to the frost problem—so “well designed,” in other words, to deal with it—as to make it unlikely that they afforded protection from frost just by coincidence. For example, the data obtained show that the height of Enga mounds and the minimum height above ground at which sweet potatoes were planted increased with altitude, which generally correlates with the intensity and frequency of the frost hazard (Waddell 1973: 36; 1975: 255).124 But even if we decide that Waddell’s data dispose of the by-product possibility and imply a causal history whereby certain features of mounding were developed specifically for frost 123 A possibility not to be ruled out, however, is that soil-fertility benefits were important in the development of mounding at all as distinct from development of some specific features of mounding on which Waddell focused. For more on the soil-fertility benefits, see Sillitoe 1996: chapter 13 and Sillitoe 1998b. 124 For other differences between mounding practices at higher and lower altitudes, see Waddell 1975: 255.

Failures of Explanation in Darwinian Ecological Anthropology / 155

protection, it must be kept in mind that there are no data on actual cause-and-effect sequences in that history—no data, for example, on whether there was trial-and-error learning by some, followed by imitation by others; whether greater reproductive success of those who practiced mounding in specific ways and their teaching it in those ways to their children were factors; or whether mounding at all originated in connection with protection from frost or else originated in connection with something else, like enhancing soil fertility, and was developed for protection from frost only subsequently. Among the lessons to be drawn then from a study like Waddell’s is that disposing of the byproduct possibility when direct evidence of the causal history of the behavioral trait in question is lacking can be an exacting task and that, moreover, accomplishing that task may still leave us a long way from being able to invoke natural selection as having any special explanatory import for the case at hand. Another possible lesson is that to insist, as Elster has at times done, on direct evidence of actual causal chains when consequence explanations like those of DEAs are proffered may be, as Elster (1986a: 205–206) himself has admitted in one article, “excessively purist.” Obtaining such evidence is clearly, however, a superior method of disposing of the by-product possibility.

Naïve Functionalism and Mysterious Mechanisms In any case, as far as DEA studies are concerned, it is not necessary to take sides in the dispute between Elster and those who, charging him with undue exigency, are satisfied with having reasonable assurance that some causal chain or other does extend from the beneficial consequences to the behavioral explananda under consideration. Either position provides a standpoint for criticizing what Elster (1986a: 203) and Schwartz (1993: 280) agree on calling “naive functionalism,” whereby finding beneficial consequences is tacitly regarded as sufficient for an explanation, with no allowance having to be made for the possibility that the benefits in question are the by-products of behavior not caused, either now or in the past, by the benefits themselves (cf. Vayda 1987a: 498–507; 1987b:

156 / Chapter 7

520).125 DEAs’ neglect of proximate causes and mechanisms lays them open to being charged with just such naïve functionalism in spite of their frequent and fervent invocations of Darwinism. It should be added that they may be charged with it in spite of some DEAs’ not being completely oblivious to the challenge that Elster and others have issued to practitioners of consequence explanations. Thus, the editors cite Elster in their first chapter in Evolutionary Ecology and Human Behavior (pp. 42–44) and also refer to others (Reynolds 1976, Kitcher 1985c) who “do not appear to deny Darwinism a limited explanatory role, but see the causal distance between genes and human action as a major impediment to evolutionary analysis” (p. 47). The editors’ inadequate response to such criticism will be noted shortly. Suffice it to say at this point that they do not seem to understand Elster’s challenge and do not meet it, that is, they do not discuss how causes or the operation of causal mechanisms are to be shown in particular cases of consequence explanation so as to assure us that the beneficial consequences identified in those cases have explanatory import for the behavior in question rather than being merely or mainly its by-products. It may be noted that Elster’s challenge would likewise not be met if the editors or other DEAs were simply to assume and assert—as, in effect, Alexander (1988: 324–325) has done in vehement response to a similar challenge from Kitcher (1985c: 289–290, 329, 402) and Maynard Smith (1985: 122)—that there are certain mechanisms which must be operating in particular cases because they are always operating. These, as far as can be made out from Kitcher’s original critique and Alexander’s fulminations in response (and elsewhere, e.g., Alexander 1989: 485ff. and 1990: 251), are mechanisms possessed by all human beings for somehow identifying the actions that would maximize their inclusive fitness and for acting accordingly. Although some would no doubt regard the very idea of Alexander’s mechanisms as preposterous (see, for example, Symons 1992: 139; Tooby and Cosmides 1990: 125 See also the section on “Confirmation Bias in Consequence Explanations” in chapter 1 of this book.

Failures of Explanation in Darwinian Ecological Anthropology / 157

405; Kitcher 1986: 67) and, like Kitcher (1985c: 329), certainly would not feel that the evidence at hand gives us any assurance of their operation, the specific point that must be made in the present context is simply that Alexander and others (e.g., Hughes 1987: 421) who posit these mysterious mechanisms and assume them to be always operating are, in effect, denying the by-product possibility and are providing no help for making the distinctions that Elster is calling for between beneficial consequences that have explanatory import for behavior and those that do not. In the absence of explicit statements in Evolutionary Ecology and Human Behavior about Alexander’s mechanisms, it is not clear what view of them, if any, is held by the DEAs represented in the volume. It may, however, be surmised that probably closer to their views are the ideas of other human sociobiologists, such as Chagnon, to the effect that natural selection has shaped the human mind/brain or psyche so that it makes various benefit-cost calculations (not just calculations of fitness benefits and costs) and decisions to act on them in ways that are, in general, favorable to the reproductive success or inclusive fitness of individuals (Chagnon 1989: 567; italics mine; cf. Chagnon 1988: 985); Alexander himself seems at times to be saying something like this (e.g., Alexander 1989: 486–487). In line with these ideas are some statements made by the editors in their discussion of “Evolutionary Ecology and Intentionality,” namely, that humans do not seem to hold fitness maximization as a goal; that the achievement of other goals held by them may reasonably be expected to highly correlate in general with fitness; and that there may nevertheless be little or no correlation in specific cases (p. 49). In another publication, editor Smith’s reference to the human decision-making machinery produced by natural selection is followed by his explicit statement of the “hypothesis” that human behavior is “fitness enhancing on average” (Smith 1987: 86; italics mine). A similar “proposition” put forward in an article by Hawkes, another contributor to Evolutionary Ecology and Human Behavior, is “that human evolution has shaped people to assess and then adjust to a wide array of circumstances according to cues that have been generally associated with their own fitness” (Hawkes 1993: 342; italics mine).

158 / Chapter 7

With their agnosticism about proximate causes and mechanisms, what objectives might DEAs hope to realize by examining particular cases if they subscribe to hypotheses like Smith’s or propositions like Hawkes’s? No DEAs, as far as I know, have gone so far as to make the patently false claim that the cases they are considering are part of a rigorously chosen sample of human behavior in all times and places to show systematically that human behavior is, on average, fitness-enhancing. Nevertheless might the idea that they are helping (albeit unsystematically) to validate hypotheses like Smith’s be regarded by some DEAs as a raison d’etre for their case studies? This is suggested by certain DEAs’ remarks to the effect that the basic question their studies have been directed to answering is whether people behave adaptively, that is, so as to enhance fitness (Blurton Jones 1990; Caro and Borgerhoff Mulder 1987: 62–63; Hill, cited in Morell 1993: 1801; cf. the discussion of fitness payoffs in Kaplan and Hill’s chapter in Evolutionary Ecology and Human Behavior, p. 168). In other words, the possibility is suggested that some DEAs, starting with the assumption (or hypothesis) that most but not all human behavior is fitness-enhancing, are content to try to answer whether behavior in the case at hand, however it has been chosen for study and whatever its proximate causes may be, is fitness-enhancing or not. If, however, they are indeed content with this, it must be said that they are far from making good on DEAs’ grandiose claims regarding the power of Darwinism to explain human behavior. My impression is that most DEAs, including some referred to in this paragraph, are not content with this and that deeply implanted in them is the naive functionalist belief that they are actually explaining behavior in particular cases by merely indicating fitness benefits from it. However, consistent with DEAs’ general insensitivity to explanatory issues, clear statements of what they believe in this regard are not to be found in Evolutionary Ecology and Human Behavior. The closest to such a statement is the previously cited one by Foley: “It is virtually axiomatic to behavioral and evolutionary ecology that adaptive features of organisms are accounted for in terms of fitness costs and benefits” (p. 137). Something that will be discussed in the next section is how ideas about the explana-

Failures of Explanation in Darwinian Ecological Anthropology / 159

tory significance of fitness benefits may be reinforced for DEAs by their commitment to the hypothetico-deductive method and by their use of it to make sometimes successful predictions about consequences of behavior which are interpreted by them as fitness-enhancing. Another possibility that is appropriate to mention in this section relates to a useful distinction that Grafen (1988: 454–457) has made concerning studies in evolutionary biology. The distinction is between studies of “adaptations” in Williams’s sense of traits fashioned by past natural selection to achieve certain effects or to serve some function for organisms having the traits (Williams 1966) and studies of what Grafen refers to as “selection in progress,” which he characterizes as “gene frequencies changing now as a result of differences in design between genetically different individuals” (Grafen 1988: 455; italics mine). Traits may be studied as adaptations regardless of whether or not selection is currently operating on them, and Grafen suggests that most evolutionary biologists are more interested in explaining adaptations than in detecting selection in progress. He does, however, note the importance of the latter to so-called ecological geneticists (e.g., Ford 1975) and to animal breeders needing to keep track of their stocks. All this is worth noting in the present context because an article coauthored with Caro by Borgerhoff Mulder, one of the contributors to Evolutionary Ecology and Human Behavior, refers to Grafen’s distinction and then proceeds to make a similar one for anthropological studies in order to argue in favor of concentrating not on the causal antecedents of different human behavioral traits but rather on differences in their present reproductive consequences, without any assumptions about past functions of those traits or about their being products of natural selection (Caro and Borgerhoff Mulder 1987). Besides the difficulty of showing past natural selection and the past functions of traits (see note 121), one of the reasons that Caro and Borgerhoff Mulder give for concentrating on present reproductive consequences is that information obtained on them should make it possible to predict behavioral traits in future generations—provided that the spread of a behavior is “to some extent dependent on the differential ability of individuals to adopt

160 / Chapter 7

that behavior, and that such differential ability is inherited” (Caro and Borgerhoff Mulder 1987: 67). In effect then, with the help of their problematic proviso about the transmission of behavioral traits from one generation to another, one of Caro and Borgerhoff Mulder’s justifications of their research priorities is their vision of how causal chains beginning with currently studied behavioral traits and their reproductive consequences will extend into the future. The extent to which other DEAs share this vision and go along with this justification is not clear from Evolutionary Ecology and Human Behavior, but it can be said in any event that abjuring the study of causal antecedents of traits in favor of studies for which only unverified predictive value rather than any present explanatory value is claimed cannot but add to our doubts concerning DEAs’ grandiose claims about the power of Darwinism to explain human behavior. In actuality, despite professions of being uninterested in proximate causes of behavior, DEAs sometimes include them, implicitly or explicitly, in considering particular cases. They include them, however, without attaching much explanatory significance to them and without getting any help from them for avoiding naïve functionalism. Among examples that may be noted from Evolutionary Ecology and Human Behavior are several stating or suggesting certain intentions as proximate causes. Thus Borgerhoff Mulder, in a chapter entitled “Reproductive Decisions,” refers to her own data on women’s choices of husbands among the Kipsigis, polygynous Kenyan agropastoralists. As she notes, the data show that women preferred men who offered more land for them to settle on (pp. 356–357). To us who are non-DEAs, this rather unsurprising finding might suggest that intentions of securing economic and social benefits through marriage to those with more land are proximate causes of Kipsigis women’s choices of husbands. Some rational-choice theorists might simply take such intentions for granted, but we might also find additional data to verify them. We could then feel justified in using the economic and social benefits of having more land as explanantia in a consequence explanation of the women’s choices because we would have met Elster’s challenge by identifying a causal chain including the women’s intentions of gaining

Failures of Explanation in Darwinian Ecological Anthropology / 161

the benefits and their actions that result in their gaining them. Borgerhoff Mulder, however, does not take this tack. Instead the land available to new wives to settle on is simply equated by her with “breeding opportunities”; in a table on Kipsigis marriage, a column of figures on such land is even labeled “breeding opportunity (in acres).” Having thus satisfied herself with the likely fitness benefits of the women’s choices, Borgerhoff Mulder feels no need to deal with proximate causes of those choices. This, notwithstanding some words of homage in her chapter (p. 373) for the objective of explaining behavior and sociocultural diversity, is of course consistent with the arguments expressed in the article that she coauthored with Caro.126 In his chapter on “Competition, Conflict, and the Development of Social Hierarchies,” Boone is more explicit than Borgerhoff Mulder in referring to intentions as proximate causes of action. Thus, in sketching the process of group formation from a “cost-benefit point of view,” he assumes, rightly or wrongly, that, in general, individual acts of joining or leaving groups are the results of intentions that individuals form as they “weigh the costs of group affiliation (increased competition for resources, increased exposure to disease) versus the costs and benefits of leaving the group for a less competitive environment or affiliating with another group under more favorable terms” (p. 301). But, like other DEAs, Boone plays down the explanatory import of such proximate causes. Thus, he presents a graphic and clear three-page summary (pp. 325–327) of Barth’s study of competition by chiefs or patrons in Swat for land and followers and the concomitant competition by followers or clients for rewards from patrons. Although non-DEAs would find both Boone’s summary and Barth’s original account (1959) intelligible as explanations of actions in relation to the intentions, purposes, and beliefs behind them and in relation to the contexts or environments in which they occur, Boone wants to go further. He criticizes Barth for not having “incorporated an evolutionary 126 However, in fairness to Borgerhoff Mulder, it should be noted that, in a later work, she does express some interest in “understanding of mechanism” (Borgerhoff Mulder 1996: 218–219; cf. Borgerhoff Mulder et al. 1997: 275).

162 / Chapter 7

dynamic or causal force such as natural selection” (p. 325). How, in fact, this is to be incorporated in the particular case of Barth’s study is not explicitly stated. However, the impression given by statements elsewhere in Boone’s chapter (e.g., pp. 318, 323–324) is that he is among those DEAs who think they are giving evolutionary explanations of behavior (or showing its evolutionary causes) by simply identifying some economic or other benefits from it and then, as will be further discussed below, summarily equating those beneficial consequences with fitness gains. In other words, the impression we get is that indicating fitness gains in this manner is regarded by Boone as sufficient for incorporating “an evolutionary dynamic or causal force such as natural selection”; once such gains have been indicated, there is no need to elucidate causal mechanisms. Indeed, when he refers to mechanisms at all, it is not to elucidate them but rather to posit them, in Alexander’s fashion, to work in unspecified ways to achieve wondrous effects, such as making individuals “fight only for fitness gain” (p. 318). Evidently, unlike the critics referred to by the editors (as noted earlier), he does not “see the causal distance between genes and human action as a major impediment to evolutionary analysis.” Another example indicating DEAs’ explanatory prepossessions is provided by an article cited and summarized in Hames’s chapter on “Time Allocation” (p. 223). In this article, Hawkes et al. (1989) consider three hypotheses to account for the fact that postmenopausal women among Hadza huntergatherers in northern Tanzania spend more time than younger women in foraging for food. Two of the hypotheses might be regarded by non-DEAs as more or less plausible intentionalistic and/or contextual explanations, namely, that 1) the older women, less burdened by child care, have more time free to be devoted to foraging, and that 2) their efforts are directed towards obtaining extra vegetable food to be exchanged for meat from male hunters, since they are too old to exchange sexual favors for meat. However, the explanation favored by Hawkes and her colleagues is that set forth under the third hypothesis, namely, that the hard work by older Hadza women

Failures of Explanation in Darwinian Ecological Anthropology / 163

is “an optimal expenditure of reproductive effort” insofar as they are getting “fitness gains by reducing the work load of their younger female kin who can thus allocate more of their own resources to pregnancy and lactation” (Hawkes et al. 1989: 354, 355). Compatible with this hypothesis would be an intentionalistic one to the effect that the older women who worked hard did so with intentions of relieving daughters or other kinswomen from some of their foraging work so as to enable them to devote themselves more effectively to pregnancy and child care. Hawkes and her colleagues, however, make no mention of intentions in this regard. Instead, they propose to test their hypothesis by making various observations, including comparisons of ovulatory regularity between those young women who get help from older ones and those who do not. But none of the tests proposed, however ingenious they may be, relate to the possibility that any gains in reproductive success in this case are only by-products without explanatory import. In naïve functionalist fashion, Hawkes and her colleagues ignore this possibility and, accordingly, are not at all concerned to indicate, with evidence and arguments specifically applicable to the case at hand, a causal chain (or, without Elsterian stringency, at least what we non-DEAs could reasonably regard as the likelihood of some causal chain or other) from fitness gains to the behavior to be explained. In other words, from the standpoint of causal/mechanical explanation, it would not be enough for Hawkes and her colleagues to show us (as they propose to try to do) that women receiving help have—or are likely to have—more children and grandchildren. We would also need to be persuaded, by evidence or arguments, that somehow transmitted to these women’s female descendants is the trait of working hard (or the inclination to work hard) after menopause to help younger kinswomen. (Cf. Wahlsten’s critical comment [1993: 310] on Perusse’s study [1993] of the relation between cultural and mating success: “it is essential to show . . . that the frequencies of different manifestations of the behavior really do change over generations because of differential reproduction.”)

164 / Chapter 7

Proxy Currencies, Foraging Models, and the Hypothetico-Deductive Method With their proposals for ovulatory comparisons and for obtaining other data directly related to producing or being able to produce offspring, Hawkes and her colleagues are at least trying to find empirical evidence of the actual effects of behavior on reproductive success. In this respect, they differ from other DEAs, who are content to carry out their cost-benefit interpretations or analyses with so-called proxy currencies, especially foodgetting efficiency but also such more specific or ad hoc measures as the gains in land and followers referred to in Boone’s version of Barth’s findings in Swat. These other DEAs, including the editors in their second chapter (p. 55), give two main reasons for recourse to these currencies. One is that they are more amenable than fitness to empirical research. Thus, Hames (p. 204) states baldly that they are used because “fitness payoffs for alternative behaviors are practically impossible to measure.” The second reason, not surprisingly, is that the other currencies are thought by DEAs to correlate with fitness. However, the presumption of such correlation is either mentioned offhandedly by them in Evolutionary Ecology and Human Behavior or else is only implied rather than explicitly stated. Thus, Cashdan, in her chapter on “Spatial Organization and Habitat Use,” first declares her “underlying assumption” that how space is used is subject to natural selection, and then, without ado, she immediately follows this by noting that the models presented in her chapter mostly use energetic return rather than fitness as a currency (p. 237). The editors are at least explicit about presuming the correlation between fitness and other currencies, but, instead of giving us any extensive testatory discussion of it in their two chapters, they present only the modest claim that “we have reason to believe” that such a correlation either obtains now or has obtained in the evolutionary past (p. 55). Kaplan and Hill, in their chapter on food acquisition, and Hames, in his chapter on time allocation, go somewhat further insofar as they set forth different ways in which food-getting efficiency may affect fitness among human foragers, for example, by producing

Failures of Explanation in Darwinian Ecological Anthropology / 165

more food that can be used to promote survival and enhance reproductive success or else by freeing time from foraging for use in other, presumably fitness-enhancing activities, such as parental care (pp. 168, 206). Another possibility mentioned by Kaplan and Hill is that reducing the time spent in food-getting because of increase in its efficiency affects fitness by virtue of reducing the foragers’ exposure to dangerous predators, pathogens, or other adverse environmental factors (p. 168). It may be noted that the editors themselves, following the lead of Schoener (1971) and other biologists, have mentioned such possibilities in earlier publications (e.g., Smith 1979; Winterhalder 1981: 20). That these are only possibilities is glossed over by DEAs. It should, however, be underscored, and other possibilities should be noted, lest it be thought that DEAs have done enough to justify their recourse to proxy currencies.127 One notable example of other possibilities: time which has been freed from food-getting may be used for play and other not necessarily fitness-enhancing activities. In line with this, I have remarked elsewhere that what has been said of animals (e.g., by Elton 1927: 56 and Wiens 1984: 413) applies as much or more to people: “they are not always struggling for existence, and much of what they do and why they do it may have no appreciable relation to their fitness or reproductive success” (Vayda 1988: 2). However problematic DEA use of proxy currencies may be, it is consequential in ways that may provide DEAs with reinforcement for what they regard as a Darwinian approach and for faith in what they regard as its explanatory power. For one thing, the use of proxy currencies helps them to gain access to the “model-based methodology” which, for the editors and other DEAs, is the means of “unleashing the power of Darwinism to analyze complex, flexible phenotypes” (p. xiii). After noting examples of what DEAs do with the methodology, we will be able to return to considering why they think that it provides them with explanations of behavior and why, from the standpoint of causal/mechanical explanation, it does not.

127

Cf. Sterelny and Griffiths 1999: 323 on this.

166 / Chapter 7

Another effect of the use of proxy currencies is that it makes it easier for DEAs to appropriate models, questions, and findings from such fields or subfields as microeconomics and economic anthropology. Thus, in asking much the same questions as workers in these other fields might ask about economizing in the use of time and resources, DEAs regard themselves as not simply studying economic behavior but rather dealing with what the editors (p. 53) describe as the evolution of phenotypes with fitnessmaximizing attributes. In line with this, the prey choice models tested in DEA studies of foraging behavior (as illustrated below), although based on economists’ models of utility-maximizing behavior by consumers, are not simply economic models to DEAs. Nor are they simply ecological models. Rather, they are “evolutionary models,” because their being used to test whether foragers are maximizing their returns in energy or protein or some other “currency” is cavalierly equated by DEAs with testing for fitness maximization. And findings about prey choice are of course regarded as “evolutionary” findings, regardless, incidentally, of whether or not they bear out model predictions. Some more detailed illustrations are in order at this point. In Kaplan and Hill’s chapter on food acquisition, coeditor Smith’s own research (1991) on patch choice by Inuit Eskimos is noted. Unlike the animals to which biologists have applied patch choice models, Inuit foragers do not sequentially encounter different patches, each with a cluster of a particular resource type to be exploited or not exploited for a particular period of time; instead the foragers themselves choose which patches to visit each day. Because there were differences in the rates of caloric returns from terrestrial and marine patches in the Inuit environment and because the rates varied seasonally, Smith was able to develop models and test hypotheses whereby foragers should adjust the time spent in different habitats or patches according to their relative profitabilities in the different seasons. Observations made to test these hypotheses indicated that Inuit hunters spent more time in what were the more profitable patches in particular seasons but, contrary to some of the hypotheses, did not maximize their returns by limiting themselves to these patches. Further hypotheses to account for this finding, as well as for a similar

Failures of Explanation in Darwinian Ecological Anthropology / 167

finding by Beckerman (1983) concerning patch choice by Bari Indians in Colombia, are noted by Kaplan and Hill (p. 182), for example, that a period of nonexploitation allows patches to recover after having their returns dip below the seasonal average because of exploitation for several days or weeks in a row. Kaplan and Hill’s chapter also describes a prey choice model developed by evolutionary biologists. This predicts the food items that animals, assumed to be maximizing their long-term net rates of acquiring energy through foraging, will attempt to exploit and the foods that they will ignore in favor of continued search for other foods. To test this through ethnographic research on the Ache people of Paraguay’s subtropical forests, investigators, including Hill himself, followed individual Ache foragers to determine the time spent searching their environment for food items, the rates at which different types of food resource were encountered, and the time spent in pursuit and processing after each resource encounter. The investigators also weighed the food acquired in these encounters and converted it into caloric equivalents. All of this clearly was research guided by the prey choice model, and the results, whatever their explanatory import, were useful quantitative data on Ache foraging and its caloric consequences. The initial findings, from four months of field work, were consistent with the model’s predictions: the foragers followed had exploited only those resources whose caloric returns on encounter exceeded the average foraging return rate for all food types. The initial findings led to refinements taking into account variations other than strictly environmental ones. Thus, for the purpose of analysis, the investigators had originally aggregated all the foragers followed. In subsequent studies, certain distinctions were made, for example, between shotgun hunters and bow-and-arrow hunters and between men and women. Some results from these studies were again consistent with predictions from the prey choice model. Thus, on the basis of return rates for different species of birds, it was calculated that no birds weighing less than 1 kg should be taken by shotgun hunters, and none weighing less than 0.4 kg should be taken by bow hunters. The actual weights of the smallest birds shot by shotgun and bow

168 / Chapter 7

hunters, respectively, were 1.4 kg and 0.4 kg. Other results from the studies diverged, however, from the model’s predictions. Thus, separate consideration of men and women showed that men often ignored plant foods (especially exploitable palms) which would have substantially increased their average caloric return rate from foraging, while women generally avoided pursuing “highly profitable” game resources. A tentative conclusion drawn by Kaplan and Hill from these findings is the following: “Existing optimal foraging models are quite useful for predicting food choice among resources composed of similar macronutrients but may need to be modified to account for sensitivity to the nutrient constituents of foods” (p. 176). To indicate another direction in which modifications of the models may be called for, Kaplan and Hill refer also to further studies among the Ache by Hurtado and others, suggesting that child-care requirements and the effects that women’s pursuit of dangerous, mobile prey might have on child mortality are constraints on women’s foraging productivity. Some of the gender differences in Ache foraging are noted also in Hawkes’s chapter on sharing and collective action; she proposes the hypothesis that “mating advantages” enjoyed by men who provide collective goods from hunts are the incentives for them to hunt more instead of provisioning their families better with plant foods that would not have to be shared with those outside their families (pp. 295–296). Kaplan and Hill’s analyses of Ache data, and the modification of models on the basis of their analyses, are referred to again in Hames’s chapter on time allocation. Interested in how “local constraints identified in fieldwork” may be captured more realistically in models, Hames (p. 215) cites an article in which Hill and Kaplan (1988) first describe a simple diminishing-return model of hunting effort and then refer to their observations of differences among Ache men in hunting skill and in resultant benefits. On the basis of such observations, a modified model, featuring different rates of deceleration in the hunting effort of good and poor hunters, is proposed. But, as Hames notes, Hill and Kaplan (1988: 285–286) make a point of stating that Ache data, being a source of the modified model, cannot be used to test it. They recommend testing it in societies other than that

Failures of Explanation in Darwinian Ecological Anthropology / 169

of the Ache. This is consistent with Hill and Kaplan’s evidently having as their ultimate goal not the explanation of particular cases but rather, as stated in their chapter in Evolutionary Ecology and Human Behavior, the development of “comprehensive theory” (p. 195; cf. Hill 1988: 194). Nevertheless DEAs claim to explain behavior and think that they are indeed explaining it. What makes them think so? As already indicated, they themselves give no clear answers to this question, but, referring to the examples just given from studies of foraging behavior (described by Smith [1992a: 22] as “arguably the most highly developed” area within DEA), we can suggest some further answers. One possibility is that some DEAs were led to their view of explanation in much the same way as some economists had been. As well described by de Marchi (1988b), the development of a methodology of setting forth models and testing predictions from them began in economics in the 1950s. This was in reaction to the previously dominant mode of economic theorizing which, notwithstanding that it was intended to guide policy and presumably was accorded some explanatory value, was hardly subjected to any rigorous empirical testing. Proponents of the new methodology were influenced by Karl Popper and his ideas about making theories testable and using pointed and falsifiable predictions to test them, but there is nothing in de Marchi’s account to indicate that explanation per se was a major concern. Theories, rather than predictions, were what Popper himself (1965: 61, note) thought of as explanatory, and he warned against regarding theories simply as sources of predictions and reducing an interest in explanation to a “practical technological interest in the deduction of predictions.” Nonetheless, what Popper warned against seems to have occurred in economics and, as de Marchi (1988a: 7) notes, by the 1970s the formula, “substantive hypothesis followed by empirical test,” had become identified by many economists as Popper’s contribution; it had also become the standard for acceptable Ph.D. dissertations in economics (cf. Caldwell 1982: 199ff. on the “disarray in economics” concerning explanation). Has DEAs’ adoption of the hypothetico-deductive method advocated by Popper and others led some of them to a similar equation of explanation with

170 / Chapter 7

setting forth models, testing predictions from them, and obtaining confirmation of the predictions? That it has is suggested, for example, by Smith’s interchangeable references to “predictions” from models and “explanatory hypotheses” generated by them (Smith 1983: 27 and 1984: 61) and Kaplan and Hill’s statement in Evolutionary Ecology and Human Behavior (p. 169) that “the prey choice model is designed to explain which of several food resources the forager will eat” (italics mine). Thus, DEAs might think they are confirming an explanatory hypothesis by finding, for example, that the actual weights of the smallest birds shot by Ache shotgun and bow hunters corresponded to what had been predicted on the basis of calculations of the minimum weights that made birds worth pursuing, whereas non-DEAs seeking causal explanations might regard the finding as merely a more precise description of behavior still to be explained.128 At the same time, it should be remembered that some DEAs, including Smith as coeditor of Evolutionary Ecology and Human Behavior, affirm their model-based methodology to be their means of unleashing the power of Darwinism. It may be that for these DEAs their predictions have explanatory force because the predictions are regarded as deriving from Darwinian theory and because this is regarded by them as having explanatory force— even if, incidentally, it is not so regarded by some philosophers, including, at one time, Popper.129

128 Actually, in this particular case, many non-DEAs, if not given to methodological punctiliousness, might be prepared to infer from the correspondence between predicted and actual weights that hunters were intentionally aiming only for larger birds because they had learned from experience that the food value of the small ones was insufficient to justify expenditures of time, effort, and shotgun pellets. Whether we go along with such an inference or ask first for more evidence, a point to be made here about the inference is that, in treating intentions as causes and assuming them to have been generated by the hunters’ experience, it is specifying causal relations among events leading to the explanandum, whereas DEAs like Kaplan and Hill, with their agnosticism about proximate causes, may be assuming that they have explained the hunters’ behavior by simply showing it to have conformed to their predictions and to have resulted in caloric returns equatable, in their minds, with fitness benefits. Cf. Collingwood 1939: 127–132 on historians’ use of evidence to infer intentions. 129 In Popper’s view as expressed in his “intellectual autobiography” (1976: 167–180), Darwinism is a research program rather than a testable, explanatory theory like a theory in physics.

Failures of Explanation in Darwinian Ecological Anthropology / 171

But then Darwinian theory itself is, in effect, reduced by these DEAs to a simple prediction, namely, the prediction that, as a result of natural selection, human beings, like other organisms, will “maximize the net (i.e., benefits minus costs) fitness results of their possible behavioral options” (p. 168 of Kaplan and Hill’s chapter). Apt criticisms of the DEA procedure of reducing Darwinian theory to this prediction have been made by DPs and may be briefly considered here. The criticisms serve also to underscore differences between DEAs and DPs concerning what constitutes Darwinian explanation of human behavior. DPs, in contrast to DEAs, do not privilege fitness consequences as explanantia. Indeed, they make a point of saying that the DEA procedures are not causal explanation, and they assert that bringing Darwinian explanation to bear on present behavior calls for referring not to its present fitness consequences but rather to certain mechanisms, predispositions, capacities, or features of phenotypic design which, presumed by DPs to be outcomes of natural selection under ancestral conditions, make the present behavior possible (Tooby and Cosmides 1990: 381–382). Among such features that DPs might regard as relevant to explaining the foraging behavior studied by DEAs are panhuman capacities for planning, for mentally recording and predicting frequencies of events, and for allocating effort according to expected rewards. In the DP view, studying adaptation means focusing on such capacities and their evolutionary causes rather than on the highly variable, context-specific behavior that is made possible by our having the capacities. Believing, as DPs do, that the capacities are products of nonrandom differential reproduction does not imply to DPs, as it does somehow to many DEAs, that the behavior itself must represent or instantiate “some sort of generalized reproductive striving” (Symons 1992: 138). There is, as Symons (1990: 430) has noted, a non sequitur in DEA thinking about this. Moreover, again as Symons has noted, it is a non sequitur not to be found in what Darwin himself wrote on human affairs, a subject he was keenly interested in viewing in an evolutionary perspective. Citing a personal communication from Michael Ghiselin, Symons (1989a: 133) states that “nowhere in his published or unpublished writings does Darwin suggest that

172 / Chapter 7

his theory of evolution by natural selection implies that human behavior can be expected to maximize reproductive success.” DEAs, it seems, have reduced Darwinian theory to a Darwinian prediction that Darwin never made! So the chain of reasoning and unwarranted assumptions leading some DEAs to equate explanation with setting forth their models and testing predictions from them may run more or less as follows: 1) The Darwinian theory of evolution by natural selection is explanatory. 2) The prediction that the behavior of organisms, including human beings, promotes reproductive success or enhances fitness is an embodiment or derivation of this theory and is, ipso facto, an explanatory hypothesis. 3) Therefore, any behavior of human beings is explained by being shown to conform to this prediction, either through actual measurements of fitness or reproductive success or through measurements of proxy currencies. DEAs’ faith in their procedures is no doubt reinforced by the fact that some confirmation of various versions or instantiations of their basic prediction about behavior as fitness-enhancing is often forthcoming, but there is, in fact, nothing either surprising or necessarily explanatory about such confirmation. After all, many of the versions of the prediction tested in DEAs’ studies of foraging behavior, as illustrated above, amount to little more than predictions that people will economize in their use of time and/or energy or other resources in certain delimited economic activities with fairly well-defined goals. We know that people in general are (as some DPs remind us) capable of thus economizing, that their doing so is often encouraged by strong and conscious incentives (including, in recent times, monetary gain and, probably throughout human history, gains in time for rest and relaxation), and that they often actually do thus economize, thereby not only conforming to DEAs’ predictions but also meeting conventional and virtually axiomatic expectations of some economists and other rational-choice theorists in social science.130 130 I deliberately refer here to “economizing” rather than “optimizing” because of the oft-discussed difficulties of identifying what should count as optimizing behavior in socalled “optimality” studies by evolutionary behavioral ecologists (see, for example, the critiques by Kitcher 1987, Emlen 1987, Gray 1987, Hailman 1988, and Martin 1983). By ignoring or minimizing these difficulties, some DEAs can claim to be testing predictions of optimizing, but referring to these more modestly as predictions of economizing is more defensible and accurate.

Failures of Explanation in Darwinian Ecological Anthropology / 173

If DEAs’ faith in their approach and theorizing is reinforced by confirmation of their predictions, it is because, regarding the predictions as deriving from Darwinian theory, they feel licensed to conclude that confirmation shows the explanatory power of Darwinism as they understand it. But a fallacy in this lies in ignoring implications of the fact that predictions or hypotheses about economizing can be readily put forward without any help from DEAs’ version of Darwinian theory. Similar fallacies have been committed by other proponents of the hypothetico-deductive method who have similarly ignored something well known to philosophers, namely, that hypotheses predicting the same phenomena can often be derived from many different theories and that, when this is the case, successful predictions do not necessarily favor any particular theory over others (Kincaid 1995: 213).131

Generality and Causality in Models It should not be thought that my criticisms in the preceding section concerning DEA recourse to models for explanatory purposes bespeak a general antipathy to using models. However, while I disavow any such antipathy, my disagreements with DEAs concerning the role of models in research and explanation still need to be made more explicit. For one thing, contrary to DEAs, I regard that role as a sometimes useful one rather than an always necessary one. This means that before deciding to try to apply a model in a particular case and committing ourselves to research and analysis to test predictions from it, we need, as suggested by Schelling in a lucid and judicious discussion of models in the social sciences, to ask ourselves “whether we need the model—whether the

131 A philosopher’s illustration of the kind of fallacy committed by DEAs in claiming special support for their version of Darwinian theory from their successful predictions about foragers’ economizing in the use of time and/or resources is provided by Kincaid (1995: 213) as follows: “ if hypothesis H entails evidence E, then so does H conjoined with any other claim. So if general relativity entails the extent to which light bends around the sun, then so does [sic] general relativity and the proposition that Ronald Reagan is a transvestite.”

174 / Chapter 7

model gives us a head start in recognizing phenomena and the mechanisms that generate them and in knowing what to look for in the explanation of interesting phenomena” (Schelling 1978: 89). Assume, for example, that Smith (1991) and Beckerman (1983) were attracted to Inuit and Bari patch choices (see the illustrations of patch choice models in the previous section) as interesting phenomena which they genuinely wanted to explain rather than simply, or mainly, to use for testing Charnov’s Marginal Value Theorem and other models. In that case, their asking Schelling’s questions at the outset might have made them realize then what they appear to have realized, if at all, only much later, namely, that the behavior to be explained is too complex for the simple hypotheses derived from the models to be of much help (Smith 1991: 25; cf. Cronk 1991: 31) and that, in such cases, focusing investigations on how the foragers’ daily decisions and actions and their cognized bases for them relate to their past experience and present circumstances may serve more expeditiously than blinkered model-testing for ascertaining how and why the foragers choose one patch rather than another for exploitation.132 However, even in cases of my agreeing with DEAs that models have a useful role to play, there may still be disagreement about what that role is. Views of that role consistent with causal/mechanical explanation may be substantially different from the views of nomologically minded DEAs. We can see this perhaps most clearly in relation to models that have wide appli132 For a biologist’s appeal to optimal foraging theorists to avoid “infatuation with models of little utility to understanding the real world,” see Heinrich 1983: 279–280. An anthropological testament to the kind of methodological openendedness to which I am pointing here as sometimes preferable to model-testing is Ralph Bulmer’s distinguished body of work on New Guinea human ecology—a body of work well-provided with causal explanations. Describing his approach in a 1974 article, Bulmer said: “I have no personal inclination towards the kind of problem-oriented ethnography which records virtually nothing that is not directly relevant to the testing of hypotheses generated outside that particular study.” He said further that his cast of mind leads him instead to pick on particular questions or puzzles arising from his field research and to use a wide range of data and concepts to try to answer them (Bulmer 1974: 94; cf. Guelke 1982: 48–49 on using the “question-and-answer method” in empirical investigations and note my references in chap. 1, note 1, to erotetic or question-based views of research and explanation).

Failures of Explanation in Darwinian Ecological Anthropology / 175

cability. Adherents of causal/mechanical explanation need not scorn such models.133 Indeed, as Schelling (1978: 90–91) suggests, what alerts us to the possibility that certain models (or families of models) will give us head starts in developing causal explanations of cases at hand is the very fact that the models are named and known to be applicable to different problem areas or in different fields of inquiry. Among the examples that Schelling gives are criticalmass or threshold models, known to have application in fields or problem areas as diverse as nuclear engineering, epidemiology, and studies of species extinction, fashion, residential segregation, panic behavior, and political movements (Schelling 1978: 89–110; cf. Granovetter 1978 and also Richerson and Boyd’s mechanism of “frequency-dependent bias,” discussed later on [p. 182]). Another example is the commons model, known to have application to cases of congestion of places as various as freeways, cattle pastures, and recreational beaches and to cases of wasteful exploitation of open-access resources as various as fisheries, forests, and petroleum drilled by separate oil companies from a common pool (Schelling 1978: 110–115). Consider now an example of a model from DEA studies: the prey choice model. As discussed in Kaplan and Hill’s chapter in Evolutionary Ecology and Human Behavior and applied in their own field research (see the previous section), the model does appear to be widely applicable and to have met at least some of Schelling’s criteria of utility, that is, to have given Kaplan and Hill and other investigators a head start in identifying prey choice phenomena as

133 It may be noted, incidentally, that to Sahlins, as an early critic of human sociobiology, applying the same models to humans and to nonhumans is ipso facto reprehensible, to be decried as part of a stultifying process in which Western thinkers have been caught up since the seventeenth century, namely, a “vicious cycle, alternately applying the model of capitalist society to the animal kingdom, then reapplying this bourgeoisfied animal kingdom to the interpretation of human society” rather than seeking understanding of each—society and the organic world—in its own right (Sahlins 1976b: 101–105, citing Marx and Engels’s letters on Darwin in Schmidt 1971: 46–47). In their social-science isolationism and their intolerance of cross-species analogies as sources of insight, such critics as Sahlins surely go too far. As Gould (1979: 31) has commented, “Darwin may have cribbed the idea of natural selection from economists, but it may still be right.”

176 / Chapter 7

objects of investigation and in knowing at least some of the things, like caloric costs and benefits, to which they should pay attention for the purpose of explaining the phenomena. As noted by Winterhalder and Smith in their first chapter, most hunter-gatherer studies prior to anthropologists’ discovery of biologists’ foraging models assumed prey selection to be mainly a function of prey abundance; it was the prey choice model which made anthropologists realize that “other variables, such as the food value of prey types and their pursuit and handling costs, likely were equally or more important” (p. 17). There are those who would say that the anthropologists should have realized this before and that it does not reflect well on them that they needed biologists to tell them that how much food was obtainable and how much work was required to obtain it were apt to be important factors in decisionmaking about subsistence activities. Nevertheless, the prey choice model deserves credit, from adherents of causal/mechanical explanation as well as from adherents of nomological explanation, for setting hunter-gatherer research on a productive course. DEAs appear, however, to want to give it credit for more than that. They, or at least some of them, seem to think that the more general the models which they succeed in applying to the cases at hand, the better are their explanations of those cases, regardless of how much about the cases is left unexplained by the models. In this regard, a revealing passage in Evolutionary Ecology and Human Behavior is the following by the editors: Consider the example of hunter-gatherer prey choice. We could describe each instance of prey choice in terms particular to the specific group being studied. We might note that the prey chosen varied systematically, and perhaps isolate a statistical relationship between prey abundances and the size or caloric value of chosen prey. But these findings would be specific to that case and lack any broader relevance. We would have an observation or correlation in search of an explanation. (pp. 50–51; italics mine)

For contrast, Smith and Winterhalder go on to commend the prey choice model for applying not only to both seed-gathering Paiutes and seal-hunting Inuit but also to other species, not-

Failures of Explanation in Darwinian Ecological Anthropology / 177

withstanding that “the cognitive processes or behavioral tactics underlying prey choice in bees, birds, and humans . . . doubtless are quite diverse.” Similarly, Cashdan refers to the “power of using arguments that apply to a wide variety of species” (p. 266). Indeed both her chapter (pp. 252–253) and Borgerhoff Mulder’s (pp. 351–355) feature reviews of nonhuman animal studies, evidently with the hope, as yet largely unfulfilled, that they will find that the same cost/benefit models apply to humans and nonhumans in such matters as habitat selection and polygyny and that they will thereby have powerful explanations of these phenomena in human societies. In other words, regularities so general as to pertain not only to humans but also to other species are, ipso facto, the explanantia of choice for these DEAs, regardless of how little the regularities concern either the causal mechanisms triggering the behavior of interest or the causal chains (possibly several or many converging ones) producing the ability or disposition to engage in that behavior. DEAs are of course not the first to make much of generality in explanations and to regard it as justifying only a limited concern with causes. Before the recent turn in philosophy to causal/mechanical explanation, it was common among those espousing the nomological approach not only to regard scientific explanation as “global rather than local” (Kitcher 1985b: 636) but also to give short shrift to causes in such explanation (see the discussions in Salmon 1990: 24, 46, 106), although, as Salmon (1990: 106) notes, the idea of an intimate connection between causes and scientific explanation goes back at least as far as Aristotle. However, lest DEAs draw comfort from the fact that philosophers have in the past held views like theirs in highlighting regularities and downplaying or simply dispensing with causes in explanation, they should be reminded of a point made earlier in this essay, namely, that even nomologically minded philosophers now see that scientific explanations include the elucidation of causal mechanisms. Actually, a use of models that is consonant with both causal/ mechanical explanation and with Schelling’s suggestions as noted above is rather like what Elster (1989b: chap. 1) has called “explanation by mechanisms” and what I have previously referred to as

178 / Chapter 7

using “generic assertions of intelligible connection” in explanations (Vayda 1989: note 8; 1991: 317; 1994: 323).134 Elster’s mechanisms and my “generic assertions of intelligible connection” have in common that they are generalizations about causal connections, chains, or structures—they are causal stories that recur in different times and places (cf. Elster as quoted in Swedberg 1990: 247; also, Elster 1993: 2–3). Examples may range from such psychological ones affecting behavior as what Elster (1983b: 110–111) calls “adaptive preference formation” (the “sour grapes” phenomenon) and “counteradaptive preference formation” (the “grass is always greener on the other side” phenomenon) to natural selection itself as a causal story, that is, a sequence of changes whereby a trait spreads within a population by virtue of the greater reproductive success it confers on individuals who possess it (see Vayda 1987a: 501–502, 505 for discussions of natural selection as a mechanism). These are not generalizations necessarily applicable to all or even most of what are, in some sense or other, the same kinds of cases. Thus, in cases of recognition by individuals that they have no efficacious practical means of achieving some goal, we may find a turning to magical means rather than adaptive preference formation. And, as Darwin himself recognized (see the citations in Gould and Lewontin 1979: 589–590, Kitcher 1985a: 141–142, 179), other mechanisms than natural selection may be responsible for the spread of traits in some cases. Nevertheless, the generalizations in question have a twofold importance. On the one hand, they are important in that our being able to draw on a stock of them give us head starts in developing causal explanations of whatever cases are at hand. On the other hand, it is also important that finding generalizations that can be applied in those cases enhances our confidence in our explanations: the generalizations help to assure us that we have not made ad hoc ascriptions of illusory causal connections and that, on the contrary, the causal connections which we have identified conform to real-world causal connections that have been discerned elsewhere (cf. Vayda 1988: 134 The expression “generic assertions of intelligible connection” is taken from Martin 1977: 120. The discussion of generalizations and models in the section on “Obtaining Explanatory Information” in my chapter 1 draws heavily on the rest of the present section.

Failures of Explanation in Darwinian Ecological Anthropology / 179

4 on this and also Cartwright 1989: 2 on regarding regularities as one kind of “evidence . . . that certain kinds of singular causal fact have happened”). In light of the preceding paragraph, a point to be made is that models may be used for explanatory purposes in much the same way as the generalizations just discussed if the models encapsulate causal stories recurring in different times and places. Consider, as an example, the commons model. It has not only been applied to many, by-now commonplace cases of the selfinterested, ultimately destructive use of such common-property resources as fish and game, but it has also given researchers a head start in knowing what to look for in causal explanations of some not so familiar and superficially quite different cases, such as the breakdown of a traditional Javanese communal rice-harvesting system (Sturgess and Wijaya 1983). It is, however, increasingly being recognized that there also are many cases of well-regulated use of common-property resources and that to these cases, contrary to some expectations when the commons model first became widely known in the late 1960s and early 1970s, the commons model does not apply (Feeny et al. 1990: 10–11). If we use models or generalizations as just described, we are, it should be clear, expecting less from them in terms of generality than is expected from their models by DEAs, who have set for themselves such chimerical “ultimate” goals as developing a single model that can “predict and explain the foraging patterns for different groups of Homo sapiens across time and space” (Hill 1988: 194).135 Not believing that the explanatory import which our models have in particular cases depends on their generality, we can accept that, like the commons model, they will apply in some cases and not in others and that our objective of causal explanation may be pursued without our having to discover, develop, or employ models of wider applicability.

135 Comparably vast expectations have been not uncommon in the social sciences in general. Indeed, Elster has said that “the greatest problem in the social sciences today is that of unrealistic expectations” and that “more realistic expectations will generate better social science than the grand theoretical ambitions that have been around for so long” (cited in Swedberg 1990: 247–248).

180 / Chapter 7

What Lessons from Failed Predictions? While at least one critic (Martin 1983: 615–616) has noted that DEAs tend, at times, to interpret rather generously what constitutes confirmation of their predictions, this is a point not especially germane to the methodological and explanatory issues under discussion here. But DEAs must also reckon with the fact that, even by their own standards, confirmation of their predictions is often not forthcoming.136 In the present context, this is an important point, for we can further elucidate DEAs’ explanatory failures and limitations by considering the kinds of responses they make to nonconfirmations or disconfirmations of their predictions. What might be called particularizing and generalizing responses may be distinguished. The former, set forth, for example, by Borgerhoff Mulder (p. 373) and Kaplan and Hill (p. 169) in Evolutionary Ecology and Human Behavior and by Winterhalder (1981: 33 and 1987: 314) elsewhere, consist of looking for previously unrecognized local and historical “constraints” on behaving in accord with DEAs’ predictions. To illustrate such constraints, Borgerhoff Mulder, harking back to her chapter’s discussion of predictions about the distribution of polygynous systems among human societies (pp. 352–361), states the following: Thus polygyny is not simply the outcome of domestic calculations given certain resource and investment constraints, but can be affected by the proximity of a neighboring population from whom to raid wives, which may in itself be attributable to a particular historical migration. Similarly, religiously sanctioned monogamous marriage may exist as a result of a history of conquest. (p. 373)

It might be thought that the discovery of such constraints and recognition of their causal force would lead DEAs to reconsider their professions of agnosticism toward proximate causes and would make them ask themselves whether there might not be bet-

136 It may be of incidental interest that the same may be said of the predictions of evolutionary behavioral ecologists in biology (Gray 1987: 75–79 and 1988: 217), who have been objects of emulation for some DEAs.

Failures of Explanation in Darwinian Ecological Anthropology / 181

ter, more direct ways of arriving at explanations of behavior than their procedures of setting forth and testing predictions of fitness enhancement measured in proxy currencies. But there has been hardly any such reconsideration or questioning by DEAs. Instead, their reactions have been mostly those of upholders of a research tradition, much like members of those insulated research (or “thought”) collectives which Fleck (1979) has analyzed (cf. Boon and Smit 1989: 24). In line with this, Hill and Kaplan’s study of deceleration in Ache men’s hunting effort, as noted earlier, shows that discovering local constraints through fieldwork can cause DEAs to fall right back to their nomologically oriented modelbased methodology, that is, not to closer investigation of causal chains in the cases at hand but rather to modifications of models, recommendations for their being tested in other societies, and thus possibly new rounds of making predictions from models, discovering local constraints, and again modifying the models. Similarly, the impetus to search for constraints when predictions are not confirmed is presented by Winterhalder (1987: 314) as an argument for carrying on with established DEA procedures because of their heuristic value. Appeals to such value may of course always be made, regardless of how much or how little merit resides in the predictions that are not confirmed and in the methodology and theory behind them (cf. Pierce and Ollason 1987: 115, characterizing as “simply an excuse for failure” the appeals to heuristic value made by some evolutionary behavioral ecologists in biology).137 A different kind of response to nonconfirmation or disconfirmation of DEAs’ predictions of fitness enhancement may be labeled “generalizing” insofar as it consists of developing general, simple models of how behavior not enhancing fitness may be generated. This response is represented in Richerson and Boyd’s chapter on “Cultural Inheritance and Evolutionary Ecology” in Evolutionary Ecology and Human Behavior, as well as in much of their other work since the late 1970s (see especially Boyd and Richerson 1985: chaps. 6–9).138

137 Note also the previous discussion (pp. 35–36 and note 43) on modification of DEA models in light of local and historical constraints. 138 For a recent book representative of their approach and incorporating much of their earlier work, see Richerson and Boyd 2005.

182 / Chapter 7

They are especially concerned with the role of “nonparental transmission of cultural variation” in producing behavior which is “genetically maladaptive” in the sense of not promoting fitness. What they are referring to here is the familiar fact that such nonparental individuals as grandparents, teachers, priests, employers, and peers may be imitated in their beliefs, values, and behavior; if the characteristics that make such individuals successful in cultural transmission are at odds with those promoting success in becoming biological parents, imitation of the nonparental individuals may result in the spread of behavior that reduces rather than enhances fitness. According to Richerson and Boyd (p. 78), the spread of low-fertility norms and behavior in modern Europe would be an example if it were the case that a mechanism for their spread was basing reproductive decisions and behavior on the imitation of entrepreneurs, business managers, professional educators, and others who had attained positions of wealth, power, and/or influence partly because of having come from and having had small families characterized by increased parental investment per child. Richerson and Boyd (pp. 81–82) also set forth the possibility that imitation of those who have attained elite positions will sometimes be wholesale—perhaps because of difficulty in determining exactly what has contributed to their success—and will therefore sometimes involve taking over from them not only what has made them economically or politically successful but also maladaptive fads and fashions as well as religious beliefs and symbolism which do not necessarily promote either fitness or economic/political success. Moreover, once that many or most individuals in a population have adopted a nonadaptive or maladaptive trait, adoption by remaining individuals may be hastened by another mechanism, epitomized by the maxim about acting Roman when in Rome. This mechanism, whereby the observed frequency of a behavior in a population is the basis for deciding whether to imitate it, is labeled “frequencydependent bias” by Richerson and Boyd (p. 82) and is discussed by them in their chapter in Evolutionary Ecology and Human Behavior in relation to the question of group selection and the evolution of human cooperation (cf. Schelling’s discussion [1978: 91–110] of critical-mass models, noted previously).

Failures of Explanation in Darwinian Ecological Anthropology / 183

Unlike their discussion of mechanisms producing genetically maladaptive or nonadaptive behavior, much of Richerson and Boyd’s chapter, as, in fact, much of their previous writing and modeling, is closer to DEA orthodoxy and is directed toward showing that mechanisms of selective learning and imitation may orient cultural evolution to mimic genetic evolution in making individuals behave so as to enhance fitness. Even their discussion of the sources of maladaptive or nonadaptive behavior includes speculations, supported by simple models, to the effect that such behavior obtains because it is produced by mechanisms which have been selected for in the course of evolution by virtue of the fact that, more often than not, they result in fitness-enhancing behavior. More specifically, Richerson and Boyd argue that social learning per se, virtually equated by them with imitation, may be favored because, in populations in which there is individual learning as well, social learning allows many individuals to avoid the errors and the costs (in time and energy) that would be entailed in learning individually about environments that are variable but not randomly so (pp. 70–75). And the further costs that would be entailed by having to obtain information about which of alternative traits should be acquired by means of social learning or imitation are avoided by following culturally evolved rules about imitating high-status persons and about conformity (“doing as the Romans do”). Thus, suggest Richerson and Boyd, “these forms of biased transmission” have been selected for because they “allow individuals to better their chance of acquiring adaptive behavior,” even if it is genetically maladaptive or nonadaptive behavior that is sometimes acquired (pp. 81, 86). Limited as Richerson and Boyd’s departures from DEA orthodoxy may be, it must be said that they do not back off from causal explanation as do other DEAs responding to nonconfirmation or disconfirmation of predictions of fitness enhancement. It must also be said, however, that Richerson and Boyd’s responses to the failed predictions constitute, at best, rather special explanations that should be of more interest to DEAs than to most of the rest of us. Noteworthy, first of all, is that the explanations may be regarded as examples of what some philosophers and

184 / Chapter 7

biologists have called “how-possibly explanations” (Brandon 1990: 176–184; Dray 1957: chap. 6; Ereshefsky 1992: 83–92; O’Hara 1988: 146–150; cf. Boyd and Richerson 1992: 203–204). These present us with causal stories and models about how some events (including actions) could have come about, but they need not be construed as telling us how the events actually did come about. Nevertheless, as both Dray (1957: 162) and O’Hara (1988: 147) emphasize, the stories have explanatory value for us if we have harbored preconceptions or expectations making us puzzled or incredulous about the events in question—if, in effect, we have been asking ourselves, “How could such things have happened?”139 But it seems highly unlikely that most of us who study and try to explain human behavior have been asking this with respect to particular items of behavior because of having found them to be genetically nonadaptive or maladaptive. On the contrary, most of us probably have, as Kitcher (1985c: 287) also suggests, a priori expectations inclining us to think that what people do (including their imitation of others) is sometimes fitness-enhancing and sometimes not and that whether some behavior falls into one category or the other is mostly irrelevant both for explaining it and for deciding whether to devote attention to it as an explanandum. Since DEAs and like-minded sociobiologists have a priori expectations that incline them to think otherwise, it is for them and not for the rest of us that Richerson and Boyd’s how-possibly explanations of behaviors maladaptive or nonadaptive in terms of fitness should have special explanatory value and methodological implications. Sad to say, however,

139 In fact, alternative how-possibly explanations may sometimes be set forth for the same events and may serve more or less equally to make us less puzzled or incredulous about the occurrence of the events. Thus, if the question is how the ancestors of the Polynesians, believed to have only crude means of navigation, could have reached and settled remotely located islands, one how-possibly explanation consists of first citing evidence of accidental landfalls on various Polynesian islands by voyagers lost at sea in historic times and then arguing that such landfalls could have been the prehistoric means whereby the islands were peopled and a common culture was diffused (Sharp 1956); another and different how-possibly explanation in response to the same puzzle consists of showing that Polynesian means of navigation were not as crude as had been thought and that, accordingly, remote islands could have been settled as a result of voyages deliberately undertaken for exploration and colonization (Finney 1991). Even if we are unable to decide between the alternative how-possibly explanations proffered, we may feel satisfied that our original question about “how such things could have happened” has been answered.

Failures of Explanation in Darwinian Ecological Anthropology / 185

there is no evidence in Evolutionary Ecology and Human Behavior that Richerson and Boyd’s work had any significant effect on other DEAs contributing to the volume. None of them, for example, draw any lessons about needing to pay more attention to causal mechanisms (such as Richerson and Boyd’s “forms of biased transmission”) or, more basically, about needing to ask themselves whether it makes sense to continue concentrating on testing predictions of fitness enhancement when Richerson and Boyd have shown that maladaptive or nonadaptive as well as fitness-enhancing behaviors may be proliferated through the operation of certain pervasive mechanisms. Indeed, except for some elliptical references by the editors (pp. 31, 39, 48) and by Boone (p. 309) and the discussion in Richerson and Boyd’s own chapter, there is not a single reference in Evolutionary Ecology and Human Behavior to their explanations of maladaptive and nonadaptive behavior, although the explanations were substantially set forth as early as 1978 (Richerson and Boyd 1978) and subsequently in various places, including, most conspicuously, Boyd and Richerson’s 1985 book. Even in Rogers’s chapter on population dynamics in relation to change in resource availability, a chapter in which Rogers is sorely exercised by the apparent lack of correlation between fitness and socioeconomic success in modern societies, there is no heed to Richerson and Boyd’s explanations, despite their using as a prominent possible example the spread of low-fertility norms and behavior in modern Europe (discussed in their 1985 book [pp. 199–202] as well as in their chapter in Evolutionary Ecology and Human Behavior).140

What Lessons from Criticisms of Darwinian Ecological Anthropology? If Richerson and Boyd’s work, despite their DEA affiliations, had so little effect on the methods and thinking of other DEAs,

140 It should, however, be noted that, subsequent to the publication of both Evolutionary Ecology and Human Behavior and the review article (Vayda 1995a and 1995b) on which the present chapter is based, Richerson and Boyd’s explanations were at last given some consideration by at least some DEAs (see, for example, Luttbeg et al. 2000).

186 / Chapter 7

should we expect DEAs to have drawn any substantial lessons from the criticisms of outsiders like Gould, Kitcher, Symons, and me? We should not be surprised that they have hardly done so and that they persist in what I have criticized here: their confusions about what questions they are trying to answer; their neglect of proximate causes and causal mechanisms; their obliviousness to the importance of causes in emerging philosophical conceptions of explanation; their fallacy of inferring either from natural selection as a theory or from the role of natural selection in past evolution that the present behaviors of organisms, including human beings, necessarily or at least generally promote reproductive success or enhance fitness; their “naive functionalism” with respect to fitness benefits as explanations of behavior; their offhanded presumptions of correlation between fitness and so-called proxy currencies; their monistic, unquestioning reliance on the hypothetico-deductive method and the testing of simple, a priori models, regardless of what phenomenon is to be explained; their valuing generality over causality in their models and intended explanations; and their misconstruing confirmation of predictions from their models as constituting explanation and attesting to the power of their version of Darwinian theory. There is, however, another kind of criticism to be briefly noted, with the hope that DEAs will treat it less dismissively than they have treated other criticisms in the past. This relates to their claiming the cachets of evolutionary biology and Darwinism for their methods and ideas but insulating themselves from new developments in ecological and evolutionary biology and in Darwinian studies. While DEAs regard their field as having begun to coalesce in the late 1970s, to non-DEAs it may seem to have begun to petrify at that time and to have changed little since then. Thus, the ecologist Robert MacArthur, who died in 1972 after a career of less than two decades, is lauded by the editors of Evolutionary Ecology and Human Behavior (p. 5) for having “brought all the elements of evolutionary ecology together in a creative synthesis: definition of central topics [including feeding strategies] . . . , explicit Darwinian premises, hypothetico-deductive methods, and a reliance on simple mathematical models” (italics mine). Alone among contributors to Evolutionary Ecology and Human Behavior,

Failures of Explanation in Darwinian Ecological Anthropology / 187

the editors (p. 5) refer also to Kingsland’s 1985 book on Modeling Nature, but they may not have read beyond the chapter in which she discusses MacArthur’s work and influence to her final chapter, in which, after quoting Oster and Wilson (1978: 311) on the importance of attending to “the crucial fact that evolution is a history-dependent process,” she states the following: This critique is only one example of a generally greater awareness of history that recurs in the literature from the mid-1970s on, in response to the mathematical effusions of the previous decade and a half. Oster and Wilson realized that an awareness of history did not imply that everything was random, that there were no patterns, or that a historical discipline was not “scientific.” They only meant to suggest that mathematical models had to be used with caution as “provisional guides” to research. This cyclical shift—from a period of optimism in modeling, with an attendant belief that mathematics would raise the status of ecology, to a period of reappraisal characterized by an awareness of the historical dimensions in biology—is characteristic of ecological and evolutionary science. (Kingsland 1985: 209–210)141

While this is not the place for a comprehensive review of the newer biological and philosophical literature with respect to these issues, it may be useful to call attention to articles by O’Hara (1988), Taylor (1987), Gould (1986), Miles and Dunham (1993), Lauder et al. (1993), and Leroi et al. (1994) as a few examples of discussions of methodological or explanatory implications of the history-dependent character of biological evolution.142 In Evolutionary Ecology and Human Behavior, there are no such discussions cited, nor even just alluded to in any noticeable way. The evolutionary biology in the volume is, by and large, the evolutionary biology in vogue among animal ecologists in the 1960s and early 1970s. And, finally, what about the cachet of Darwinism? Should this be accorded to DEAs? As noted earlier, the consequential 141 142

Cf. the “Afterword” in Kingsland 1995. See also the more recent references cited in chapter 1, note 51.

188 / Chapter 7

DEA inference that human behavior can be expected to promote or enhance reproductive success is evidently not to be found in Darwin’s writings. Also, unlike DEAs who equate evolutionary explanation with finding (or inferring) fitness payoffs, Darwin was sensitive to the point that even apparently beneficial traits of contemporary organisms have not necessarily developed as targets of natural selection.143 Moreover, studies have shown that Darwin’s method of explanation was not like the hypotheticodeductive method cherished by DEAs but consisted rather of using evidence to construct and apply the kinds of causal stories that are often called narratives. Thus, in Kitcher’s 1985 account, Darwin is shown to have answered families of questions about the geographical distribution, anatomy, embryology, and adaptations of organisms by recourse to what Kitcher calls “Darwinian histories,” which are narratives tracing successive, intergenerational modifications in the organisms as a result mainly of natural selection but in response to other factors as well (Kitcher 1985a: 132–133, 139, 175 [note 11], and passim; cf. Kitcher 1993: 18–34, 263–272, and Gould 1986: 61, citing Kitcher’s 1985 article). Richards (1992) has presented cogent arguments and evidence for the view that narrative explanation, consisting of fixing events along a temporal dimension so that some events can be seen to have given rise to others, was not only Darwin’s mode of explanation, as evinced in complicated ways in The Origin of Species, but also is the mode that must ultimately be used for explanations in all such sciences as geology, paleontology, astrophysics, anthropology, and evolutionary biology, which are concerned with changes through time (cf. Gould 1988).144 143 On this, Darwin stated that “we may easily err in attributing importance to characters, and in believing that they have been developed through natural selection”; an illustration he gave was the following: “If green woodpeckers alone had existed, and we did not know that there were many black and pied kinds, I dare say that we should have thought that the green colour was a beautiful adaptation to conceal this tree-frequenting bird from its enemies; and consequently that it was a character of importance, and had been acquired through natural selection.” (Darwin 1876: 157–158). 144 As indicated in chapter 1, “causal-history explanations” is the designation I now prefer for what Richards calls narrative explanations. To the nomological argument that history can provide only “explanation sketches” (Hempel 1942), Richards (1992: 23) answers that this is getting it backwards and that nomological-deductive explanations are the sketches—skeleton explanations that must be fleshed out with narrative [or, as I would say, “causal-history”] sinew and muscle.

Failures of Explanation in Darwinian Ecological Anthropology / 189

If DEAs want the cachet of Darwinism, they will need to come to terms with the points made and the studies and arguments summarily indicated in the preceding paragraph, even if this will require their facing up to other criticisms noted at the beginning of this section. Otherwise, no matter how much DEAs pride themselves on unleashing the power of Darwinism for ecological anthropology, the Darwinism that they profess and practice, bypassing the sometimes hard work of causal explanation in favor of having it easy by simply assuming some perceived current utility of behavior to be its raison d’etre (as discussed earlier in this chapter), will seem to non-DEAs to be a flawed, cardboard Darwinism.

Acknowledgments My thanks to Wouter de Groot, Tom Rudel, and Brad Walters for their helpful comments on a draft of the article on which this chapter is based.

8 Concepts of Process in Social Science Explanations (with Bonnie J. McCay and Cristina Eghenter)

What we do not do is to go out into the world and find something called a process lying about with clear and definite boundaries. W. Harrison (1958: 250)

Social scientists refer to processes for explaining a great variety of phenomena and often make processes themselves the objects of explanation. In this essay, concepts of process are examined because of concern that, in using them, anthropologists and other social scientists are not giving enough weight and explanatory importance to the events and actions which, when linked in some intelligible fashion, constitute processes. Indeed, events have suffered from not only relative neglect but at times even outright scorn for being of only transitory significance. Braudel (1980: 27: 39), for example, described them as “brief flashes” giving off smoke which does not last. It is in response to Braudel that some historians have had to make explicit defenses of the “contextual study of short-term events” (Mayer 1988: viii). The view taken here is that the whole notion of opposing the study of processes to the study of events is misconceived insofar as it is by explaining events and connecting them to other events that we may discern the kind of causal sequence which some

191

192 / Chapter 8

philosophers of history (e.g., Walsh 1976: 24–25, 34) would designate a process.145 Many social scientists use the term “process” in a loose, unreflective fashion and are not concerned with underlying issues of methodology and explanation. For example, as Harrison (1958) noted five decades ago, references to the “political process” may signify simply the entire field of politics. More recently, some social scientists have flown the banner of process to show their position in the camp of those who see change and conflict rather than stasis and concordance as fundamental characteristics of social reality. Thus there are social scientists who have referred to society as “a set of interacting processes” (Pinxten et al. 1988: 4) and have meant by that to counter the persisting view of society as an enduring entity with definite and permanent boundaries (cf. Holy and Stuchlik 1983: 107; Moore 1975). Another quite common and loose practice consists of designating processes in terms of some recognizable outcome of events even if the events themselves and the linkages among them are little known or understood. In this mode, one speaks of “innovation processes” and “political centralization processes” more because innovation and central state control appear as outcomes than as a result of knowledge of what led to them. Those with a more restricted concept in mind sometimes use the term “process” to refer to a set of connected events occurring according to specifiable rules and within specifiable parameters of time (e.g., the “electoral process”). In addition, anthropologists and other social scientists often use the term to describe what are believed to be general types of these sequences. Thus Gluckman’s (1955) “judicial process” and Turner et al.’s (1969) “ritual process” imply, in a universalizing mode, that similar types of repeated unfoldings of events may occur in different societies or in different times (Moore 1975: 224–225). One problem with such usage is that by lending itself to a view of processes 145 There is also an important consideration emphasized in chapter 1, p. 24, namely, that claims about events as causes or effects can be more readily assessed in terms of evidence than can such claims about larger posited units or entities (e.g., “processes” or “systems”) regarded as wholes.

Concepts of Process in Social Science Explanations / 193

as unitary entities with causal powers in their own right, it has kept social scientists from adequately confronting the kinds of methodological and explanatory issues with which we are concerned in this essay. We begin to address these issues in the next section.

The Reification Problem Social scientists sometimes use process to explain reality as if processes had lives of their own, as if they existed independent of human agency and were regulated by some larger dynamic in history. This may be the result of what Thompson (1978: 75), commenting on a work in Parsonian sociology, has described as a “reification of process entailed by the very vocabulary of analysis.” Indeed, very often the word “change” would do as well, but, by using the term “process,” writers suggest that they have something in mind with greater regularity and different ontological status. Forbearance is called for if that is indeed their intention and if it is somehow justified. The danger, however, is that the reification will occur without intention and, more regrettably, without justification.146 This is a danger which those of us who were “systems oriented” social scientists in the 1960s and 1970s did not always avert. An example from explanations of New Guinea warfare is discussed in the next chapter (based on Vayda 1989). There was likewise a reification of process in McCay’s (1978) exploration of how people responded to resource decline in coastal Newfoundland in the 1960s and 1970s. The causally connected events that she observed included these: First, many people diversified their situations by relying on government welfare payments as well as wage labor and fishing; then, the magnitude of the increase in the welfare rolls was deemed unacceptable by government agents, who responded with policy changes and subsidies 146 Reification of processes and systems by the actors of concern to us is not germane to the discussion here. As Barth (1981: 80) puts it, “the fact that actors themselves often harbour . . . reifications must enter into our models at appropriate points, and not be regarded as a carte blanche for us to do the same” (cf. Elster 1986b: 23).

194 / Chapter 8

which, in combination with other events, led more people to increase their commitment to the fishery by participating in the subsidized development of an offshore fishing fleet. Involvement in offshore fishing reduced the time and resources people had to maintain diversified ways of making a living. Although McCay outlined these and other events as part of a critique of systems theorists for excessive reification and misplaced teleology, she inadvertently contributed to both practices in her use of the concept of response process. She borrowed the concept from writings on evolution by Bateson (1963) and Slobodkin and Rapoport (1974), who posited selection for systems that are designed for flexibility, thereby reserving more costly, “deeper” responses (i.e., intensification) for the longer term and more severe hazards that less costly responses (i.e., diversification) could not adequately meet. The choice of this model led her to regard the sequence of events she observed as the predictable behavior of a response process conforming to a certain logic (“the economics of flexibility”). Accordingly, despite the fact that her thesis was the need to avoid ascribing agency and purpose to systems instead of people, McCay ascribed both to a response process. The foregoing example shows that even when human agency is acknowledged, processes may be reified. Sometimes, this may be a result of an inability to accept fortuitousness or randomness and a corresponding insistence on finding meaning and direction in all events. Paradigmatic in this regard is Tolstoy (1982), saying on one hand, that every man acts to achieve personal aims and uses his freedom to live his own life and, on the other hand, that whatever one does has a predestined significance in history and belongs to an “inevitable march of events” (717–718, 886). As previously suggested elsewhere (Vayda 1986: 297, 300), the fact that human actions so often have unintended consequences may be a stimulus to reification by those to whom randomness is unacceptable. Seeing that the patterns we design and try to impose on the world do not obtain, they conclude there must be some other agency which is producing patterns and meaning (cf. Berlinski 1986: 6, 73; Elster 1983b: 101ff.).

Concepts of Process in Social Science Explanations / 195

In anthropology, even when the causal relevance of human agency is acknowledged, the language used is sometimes so ambiguous as to leave the readers uncertain whether the authors impute to process an ontological status and intrinsic properties that are neither empirically justified, nor theoretically useful. Thus Wolf (1982), in recounting diverse aspects of the slave trade and the impact it had on the economies of both Europe and Africa, concludes that “these different configurations cannot, therefore, be understood as typologically separable states or ‘tribes’ of people without history. They are the variable outcomes of a unitary historical process” (230–231). How are we to make sense of this statement? Is it a matter of unreflective use of language, or is it being suggested that the one historical stream produces the variety of occurrences and, moreover, regulates the overall unfolding of events in a meaningful manner? Other critics, such as Taussig (1987), have also noted that the very language used by Wolf (1982) grants a certain autonomy and vitality to processes. Vocabulary and style create a sense that the processes have causal powers and a reality in their own right. Consider the following excerpt from Wolf’s book: External warfare, trade, and internal consolidation created new states in Europe. . . . Agriculture ceased to grow, perhaps because the available technology reached the limits of its productivity. The climate worsened. (p. 108)

Commenting on these sentences, Taussig (1987: 165–166) said: Thus does this writing animate things and abstractions of things; war becomes a person, a spirit, or a god. It creates States. So does trade. . . . Internal consolidation also creates (States). Agriculture stops growing. Technology reaches. And all this spiritualization of things is on a par with natural process; “the climate worsened.”

Similarly, Comaroff (1982), in an otherwise excellent argument against views of local communities as mere products or pawns of external forces, often casts his description of the

196 / Chapter 8

economic transition in the Barolong area of Botswana in terms of an “unfolding of dialectical articulations” between the “systemic logic of local structures” and “external forces.” Again, here, because of the language that is being used, events seem to be conceptualized as part of a larger process, but we never learn how the component events are linked in a process or what are the conditions under which they become linked. The mechanisms of dialectical articulation remain a mystery. Smith (1984) has also written persuasively about the onesidedness of attempts to take into account the “larger or global processes” that affect the small communities studied by anthropologists (see also Mintz 1977; Comaroff 1984; Richards 1983; Hall 1986, among many others concerned about this “onesidedness”). In contrast to the view of peasants as passively caught up in an expanding and essentially autonomous capitalist world system, Smith argues, from her work in western Guatemala, that so too were “capitalist institutions changed by the response and resistance of Guatemalan and other peasants to them” (Smith 1984: 197). Like many of her anthropological contemporaries (see Ortner 1984), she emphasizes the role of human action in history.147 However, she translates her professed concern with peasants responding and resisting into a concern with one set of processes—the local ones—interacting with another set of processes—the more extensive ones—coming from the outside world. “Forces” become independent agents, and process seems to be conceptualized as an autonomous entity. So, in this passage criticizing those who focus on so-called global processes: rarely do they look at the way in which local systems affect the regional structures, economic and political, on which global forces play, which prevents them from describing the way in which global forces have adapted to local conditions. (p. 224)

Smith’s analysis and exposition would have been clearer if she had been both explicit and consistent about making events, 147 Assertiveness of anthropologists in the 1980s about human action was a reaction to its neglect in earlier decades. Human action had of course long been emphasized by various nonanthropologists, including Marx and certain Marxists (see Magarey 1987).

Concepts of Process in Social Science Explanations / 197

the linkages among them, and their consequences the objects of explanation. For the most part, she has done this in her analysis of the relevant factors and circumstances that led to the economic and social transformation of western Guatemala, including local initiatives and arrangements that played a role in shaping the timing and nature of capitalist development there.148 But then she falls into the trap of substituting process for events and then attaching to it the labels of local, regional, and global, as if doing this made for greater explanatory force. It may be that Smith’s recourse to the vocabulary of process at local as well as regional and global levels was prompted by questions that had often been raised by others (e.g., Bennett 1967; DeWalt and Pelto 1985) concerning the relations between micro- and macro-processes. Such questions fade in significance if we recognize that conceptualizations of this kind only impose an artificial and misleading grid on reality and if we concentrate instead on causal connections among events at different points in space and time.149 A further point is that obviating the reification problem requires more than lip service to causal mechanisms linking actions or events. Evidence and arguments must be produced, and it has to be recognized that mechanisms which at first seem plausible may be shown to be otherwise on the basis of the closer examination that is necessary. Consider, for example, the possibility that the connection among a series of actions over a period of time is that they all are deliberate attempts of the actors to carry out a previously formulated program or policy. A historical process that may illustrate this is legal reform in early nineteenth-century Britain. But, as Walsh (1976) remarks, such examples seem to be the exception in history rather than the rule (60). In his view, it would be “absurd” to explain the rise of monopoly capitalism or any similar development as a

148 Local forces emphasized by Smith (1984) include the “parcialidad,” a social unit of land-holding and cooperation below the municipality; the organization of a rural marketing system; and forms of peasant resistance. 149 For similar ideas expressed recently in somewhat different language in an important article by geographers, see Marston et al. 2005 on “flat ontology.”

198 / Chapter 8

process consisting of a deliberately planned series of happenings, and yet, illustrating the point about the need for more than lip service concerning causal mechanisms, such explanations have been put forward and have been widely accepted in some fields. Thus archaeologists and others long regarded plant and animal domestication as processes which were not only adaptive, but also teleological, insofar as they were assumed to be directed from the start toward greater control over human subsistence (Rindos 1984). But this view faces the problem posed by Rindos (1984: 87): If we assume that people intentionally invented agricultural systems and fitted cultivated plants into that system, we must face a fundamental problem: the response of the plants would have been both unpredictable and slow. This follows because the plants with which any agricultural system would have had to start are wild and adapted to the conditions of survival in the wild: many of their traits are ill-suited to survival in an agroecology.

This rules out human precognition and thus argues against explaining domestication as a unitary, end-directed, intentional process.

Does an Event a Process Make? There has been, in some social science literature, an implicit or explicit recognition that processes are made up of human actions or of events involving human actions. In addition, “processual” social scientists, such as Moore, have emphasized the importance of focusing on events to better understand processes: “After all, events are to processes what categories are to structures” (Moore 1987: 736). But how do we justify regarding some particular events as constituting a process? There has not been enough systematic attention to this. Should we, for example, consider Brutus’s stabbing of Caesar, Washington’s crossing of the Delaware, and the first moon walk by astronauts as parts

Concepts of Process in Social Science Explanations / 199

of one process? If we were to work long enough and diligently enough at it, we might be able to show how even these events are connected, and the point can be made that almost any events might be ultimately connectable.150 However, if this is so, it underscores the need for methodological justification to regard some events as constituting a process and others as not. The absence of systematic attention to the nature and strength of the linkages among events which are regarded as constituting a process encourages a simple part-whole fallacy that we have found in numerous studies in which alleged processes are put forward as causes of behavioral, social, or biological effects. The fallacy is that of assuming that evidence of any of the events regarded as constituting a process is evidence that the process itself is in operation. Some anthropologists’ explanations of settlement patterns illustrate this fallacy. From the valid observation that proximity to resources is sometimes the basis on which people decide where to settle, extrapolations are made so that the relation between resources and location of settlements which is observable at a particular time is interpreted as a determinate outcome of a unitary process, designated as a “settlement strategy,” for distributing people in relation to resources (Jochim 1981: chap. 6). Ignored is the alternative possibility that a particular resource/ settlement relation may be the “not readily predictable outcome of various individual and collective decisions which people have made over time on the basis of various economic, social, and political considerations rather than solely or even primarily on the basis of resource distribution” (Vayda 1986: 305). Relevant here is Gould’s (1980) warning against confusing plausible storytelling with actual evidence that a process is in operation. Moore (1986, 1987) has argued strongly in favor of both so-called processual ethnography and ethnographic attention to events as being most appropriate to anthropology in a changing world. Regrettably, however, her arguments too are 150 On this point, see chapter 3, note 81, and note the citation there from Carlyle’s classic essay, “On History” (1899, orig. 1830), on the interrelatedness of all events.

200 / Chapter 8

methodologically deficient. They rest on the claim that a particular event (even “the least temporally stretched out” episode) can evince or indicate a process. The examples she gives from her research among the rural Chagga people of Tanzania are of events involving land acquisition that diverge from the normal and traditional Chagga conveyance from father to son. Without being entirely clear about it, Moore (1987: 729) suggests that these events are evidence of an “ongoing dismantling of structures.” If she means this, we must object. While the events may well be part of such a process, they also may not be. Indeed, when critics of Geertz’s work in Java cited similar events as evidence of capitalist transformation in the Javanese countryside, Geertz (1984: 519, 527, note 13) counters by referring to the same kinds of events earlier in Javanese history and arguing that the events led only to transitory changes, to “local spasms” or soon engulfed “bubbles in the stream,” rather than to enduring transformations. No more does an event, by itself, reveal a large-scale social process than does a grain or even bucketful of sand reveal the morphology of a coastline.151 Certainly a new and different anthropology is needed for today’s complex and changing world. In line with this, our notions of what can and should be done in fieldwork are being revised. For example, Moore (1987: 735) noted that anthropologists are replacing a conception of themselves as looking at whole cultures or whole societies with an acute consciousness of “observing part of the cultural construction of part of a society at a particular time.” There are shortcuts in fieldwork, but they can mislead us and take us down dangerous alleyways. If we are interested in long-term changes, we must do what we can to observe them over the long run—and, if we do, we may be surprised. Thus Scudder and Colson (1979) found, from recurrent field research among the Gwenbe Tonga of Zambia over a period of more than twenty years, that much of what they originally saw as evidence 151 It should be noted that no argument is intended here to the effect that a process must lead to transformation or even involve substantial change; it may instead entail the renewal or re-creation of existing conditions.

Concepts of Process in Social Science Explanations / 201

of stable adaptations and persistent processes were only ad hoc, transitory adjustments.

A Useful Concept: Two Examples If we abandon process reification and recognize that events sometimes connected in a sequence are not invariably thus connected, we still need not reject all concepts of process. Among the explanatorily significant possibilities which remain is that of discerning not invariant or inevitable but still recurrent sequences of causally connected actions and events. Boom-bust cycles in economics may be examples, and many other examples could be given from other fields. However, for the purpose of illustration it will suffice to give just two examples here (neither much associated with the other except in the context of this chapter): the so-called “natural cycle” of the American presidency (Steele 1980) and the process of chain migration. The presidential cycle produces what has been called the paradox of presidential power: One of the most powerful people in the Western world is perpetually obstructed in the exercise of his power. An analysis of five American presidencies led to the following depiction in 1980: The first year in office is a time of honeymoon. Laws are proposed. Stands are taken. By the second year, as a “coalition of minorities” begins to form against him and frustrate his intentions, the President turns inward, resentful of the media, angry with Congress, and suspicious of his own Cabinet. In the third year he turns to foreign affairs, where his freedom of action is greater. In the fourth year his attention snaps back to domestic considerations as the ticking of the electoral clock grows louder. (Steele 1980, reviewing Hodgson 1980; see Hess 1976 for the full analysis)

Although the model may have seemed a powerful predictor as it stood (Hess 1976 successfully predicted the course of the

202 / Chapter 8

Carter presidency), the process is not inevitable; its course may change as new and different conditions arise. For example, when this chapter was originally written early in 1989, the U.S. budget deficit and other concerns were making uncertain the duration of George H. Bush’s honeymoon. Contexts and the causally connected actions which occur in them may follow one another as before, but, as Hodgson (1980) warns, “there is nothing automatic or predestined about the cycle” (35). As it happened, George H. Bush’s honeymoon extended well beyond his first year in the presidency. And, in contexts of heightened political partisanship and fears of terrorist attacks, the course of George W. Bush’s presidency diverged substantially from the model. Our second example is chain migration, a familiar sequence in which an initial migration is followed by other movements of people along the same immigrant pathways. Initially some individuals leave their homeland, for one reason or another, and those who become successful encourage friends and relatives to join them to share in the new life and wealth. Eventually, others come, receiving help with jobs and housing, and a small immigrant community is formed, enriched if and when the more securely established send for or personally bring their wives and children. The process continues in ways too detailed to recount here (see Price 1963: 113ff.), with consequences for the nature of community at home and abroad and predictable changes in the nature of migrants and migrations. The process has been seen by some as self-regulating, in that it is very sensitive to economic booms and depressions in the country of settlement, so that, when labor is needed, migration increases, and when jobs are scarce, it decreases (Price 1963: 121). However, there is no need to postulate an autonomous process or system as doing the regulating. Those involved, through their own experiences or the news they receive from friends and relatives, assess the possibilities and act accordingly. Nor is there an inevitability or unidirectionality to parts of the process; phases of the process are contingent on what is happening in native and “new” worlds and much else besides. For example, the pioneer migrants are not always successful

Concepts of Process in Social Science Explanations / 203

enough in their new homes to attract other migrants; the chain may be short indeed.152

Conclusion Neglecting such questions as what constitutes a process and what causes it, anthropologists and other social scientists have assumed or implied that processes have causal force in their own right and that relations between them—for example, between micro- and macro-processes—can be systematically investigated. Linguistic usage itself may contribute to the notion of processes as unitary entities with teleological properties. A line of inquiry with greater explanatory promise and validity requires that we conceptualize processes as constituted of events and that we translate questions about short- and long-term processes into questions about events. Concomitantly, a causal explanation of processes and not a merely constitutive one must be sought with respect to the events themselves, the linkages among events, and the conditions under which the linkages do or do not obtain.153 It is as a result of such explanation that certain events may be said to relate to one another as causes and effects in intelligible sequences and thus to constitute what may be usefully labeled, as in the two examples given in the last section, as processes. These recommendations are made here with the recognition that, as literature in philosophy attests (e.g., Davidson 1980; LePore and McLaughlin 1985; Lombard 1986), the concept of events and of event causation (and composition) is also problematic (cf. Mink 1987: 198–201).154 Even so, it is much less problematic than concepts of process and more useful for explanatory purposes. 152 Price (1963: 124), whose example of chain migration is very clearly described, also shows that some of the southern European settlers in Australia could actually take advantage of a depression to buy bankrupt businesses and bring out their families, countering the tendency for migration to decline or cease when the receiving area is in economic or political trouble. 153 See Dray (1980: 53) on the distinction between causal and constitutive explanations. 154 For more recent references on problems with the concept of events (especially the problems that exercise metaphysicians), see chapter 1, p. 3 and note 2.

9 Explaining Why Marings Fought: Different Questions, Different Answers

In several of my publications in the 1980s (Vayda 1986, 1987a, 1987b, 1988), I took others to task for committing certain errors of reification and fallacies of functional explanation which occur also in some of my own studies that were published in the 1970s (Vayda 1970, 1971, 1974, 1975, 1976). Warfare was the subject of those studies, and I returned to it at the close of the 1980s partly to end the unseemliness of pointing the finger at the sins of others without acknowledging my own.155 However, my principal reason for returning to the subject was to deal with some explanatory issues raised but not resolved by certain criticisms of my earlier studies. Since the main target of attacks by my staunchest critic, C. R. Hallpike (1973, 1977b, 1986), was my analysis of warfare among the Marings of the Bismarck Mountains of Papua New Guinea (Vayda 1971, 1974, 1976), I used the Maring case in the 1989 article on which this chapter is based for illustrating both my own past errors and certain other explanatory problems. Those problems included ones related to a point made in chapter 1, namely, that a question, such as that of why Marings fought, can be given different correct answers, 155 That those who substantially change their views have an obligation to give testimony of their conversion is well argued by McKinley (1977: 436) in a review of Marshall Sahlins’s Culture and Practical Reason (1976a).

205

206 / Chapter 9

depending on our assumptions about what the question means. Also included were problems related to points made elsewhere (e.g., in Vayda 1994) about failures of explanation because of insufficient attention to variability and context-relatedness when purposeful human behavior, like fighting or deciding to fight, is being explained. After giving a summary description of Maring warfare, I will again use the Maring case here for elucidating explanatory problems related to these points, as well as to certain other points made in chapter 1 about reification (see also chapter 8 on this) and about consequence explanations.

The Maring Case156 My fieldwork among the Marings was conducted in the early and mid-1960s. At that time the total Maring population of seven thousand was unevenly distributed within a rugged forested area of 190 square miles; in the more densely settled parts, a dozen autonomous local groups ranged in size from about 200 to 850 people, but some smaller Maring groups inhabited the less densely settled lower altitudes of the area. Slash-and-burn cultivation of tuberous staples and other crops was the main subsistence activity of the people, who engaged also in pig husbandry, pandanus tree cultivation, gathering wild plant foods, and hunting feral pigs, small marsupials, and birds. In Maring wars the main belligerents were always autonomous local groups with adjacent territories. Most groups seem to have averaged one or two wars per generation. The last wars 156 This summary of Maring warfare is drawn mainly from Vayda 1974: 186–187 and 1976: 13, 22–23. Throughout this chapter, I use the English s for the plurals of New Guinea group names (e.g., “why Marings fought” rather than “why the Maring fought”) in order to avoid undue suggestions of uniformity in the ideas and behavior of actors. The fact that, in the accounts of wars later in this chapter, I nevertheless speak of Kundagais, Ambrakuis, and members of other Maring local groups as if all members of any particular group did and felt the same things at the same time (e.g., “the Ambrakuis were enraged” or “the Ambrakuis fled”) may reflect not so much actual uniformity in actions and feelings as my not having asked more about interindividual variation when I was receiving informants’ accounts of past wars. On the other hand, insofar as warfare is comprised of concerted actions, there are grounds for expecting greater unity of both action and purpose in war than in some other areas of human activity, even if the unity may be the product of considerable antecedent intragroup mobilization efforts and politicking.

Explaining Why Marings Fought / 207

were fought just a few years before my arrival among the Marings in 1962, and I was able to reconstruct the nature of the wars from informants’ accounts. The informants specified the antecedents of thirty-nine Maring wars for me, and, in almost every case, some offense by the members of one group against the members of another was involved. Murder or attempted murder was the most common offense, having occurred in twenty-two of the cases. Other offenses mentioned included poaching, theft of crops, and territorial encroachment; sorcery or being accused of sorcery; abducting women or receiving eloped ones; rape; and insults. Sometimes these other offenses led directly to war and sometimes first to homicide. At times offended groups committed homicides deliberately as, in effect, war declarations. The transition from peace to war was marked by the performance of prefight rituals, which included offerings to the spirits of ancestors who had died in previous wars. The first phase of hostilities in the wars often consisted of a series of “nothing fights,” daylong bow-and-arrow encounters at a prearranged battleground. This phase could continue for many days and even weeks. The succeeding phase consisted of “true fights,” in which the arms employed at the battleground were expanded to include weapons of close combat and the warriors sometimes made quick charges into the enemy lines. However, in most of the fighting of this phase, the combatants remained in static positions behind large shields, and, accordingly, engagements could take place day after day for weeks and even months without heavy casualties. Moreover, hostilities could, by mutual agreement, be suspended for a day or more during this phase in order to allow the combatants to repaint their shields, to attend to rituals in connection with casualties, to rest, or to attend to agricultural tasks. As a rule, mortality became heavy only with escalation to routing. In this the warriors of one side went to the enemy settlements, burned the houses there, killed indiscriminately any men, women, or children that they found, and, after having put the survivors to flight, destroyed gardens, fences, and pandanus groves and defiled the burial places. Escalations, it should be noted, were not inevitable in Maring wars. A return to peace was possible from nothing fights, from true fights, and from the raids which were alternative

208 / Chapter 9

antecedents to routing, especially in the low-density parts of the Maring area. Raids usually consisted of stealing in the night to the houses where the men of an enemy clan slept. At dawn the raiders would make fast the doors of as many of these houses as possible and then shoot arrows and poke long spears through the leaf-thatched walls at the men inside. If the latter succeeded in undoing the doors, they might be picked off by raiders waiting behind the house fences. After killing some men, the raiders were, as a rule, forced to retreat because of counterattacks by warriors from houses other than those raided. Decisions by both sides in Maring wars in favor of armistice were influenced by assessments of relative fighting strength, numbers of casualties, and the nature of previous relations between the antagonists. Even from the refuging that followed routing, there could be a return to the status quo ante bellum. It might be expected that the land of a routed group whose members had gone into refuge would have been immediately taken by the victorious warriors, and sometimes, as will be discussed in a later section, it was. Often, however, the victors were constrained by Maring notions about the continuing dangerousness of the ancestor spirits of displaced enemies. Accordingly, they would not attempt to move into enemy land until a later time, perhaps many years later, when they could count on the support of their own ancestor spirits because of having made appropriate sacrifices of pigs to them in a long sequence of ceremonies (described in detail in Rappaport 1968). If, however, the routed group succeeded in rehabilitating itself and returning to its old lands before any move by the victorious warriors, no territorial annexations would take place. In my publications in the 1970s, I argued that the features of Maring war summarized above should be regarded as a process consisting of recurrent, distinguishable phases. Making territorial conquests (“peace with land redistribution”) the final phase, I diagrammed the process (see Figure 1). My claim was that important understandings can result from viewing war not simply as something that either does or does not occur, but rather as a process (Vayda 1974: 185 and 1976: 2). Examples which I gave of such understandings, based at least partly

Figure 9.1.

210 / Chapter 9

on my analysis of Maring warfare, included statements that the causes of entry into war are not necessarily the same as the causes of moving from one phase of war to another; that, by escalating to territorial conquests, war processes can be effective in counteracting stresses associated with population pressure; and that, “even if territorial conquests had been only an infrequent rather than a regular aftermath of Maring warfare for a considerable time, the warfare remained the kind that could, through an already institutionalized systemic process, lead again to population dispersion and land redistribution whenever demographic and ecological conditions changed sufficiently to make it appropriate for this to happen” (Vayda 1976: 42, 103).

Recantation How had I gone wrong in analyzing Maring warfare? Like other anthropologists, I was using the concept of “process” without paying systematic attention to how regularly and in what ways the events regarded as constituting a process were linked.157 Accordingly, how the Maring war process culminates in territorial conquests could, as Hallpike has noted, be made to look “very impressive” in my diagram, but what is represented may in fact be nothing more than that a group, again as Hallpike has noted, “may, if sufficiently determined and strong, chase another group off its land permanently, instead of allowing them [sic] to return as is usually the case” (Hallpike 1986: 107–108). I would say now that using process diagrams and jargon, rather than plain English prose, to describe such possibilities not only does not advance analysis and understanding but also may often get in their way. A danger is that which Thompson (1978: 75), commenting on a work in Parsonian sociology, has described as a “reification of process entailed by the very vocabulary of analysis.” This is a danger I did not avert. As indicated in the preceding paragraph, I 157 The loose, unreflective use of the concept of process in social science has been examined by Harrison (1958: 243–252). In the present volume, arguments against such use are noted in chapter 1 and elaborated in chapter 8. See also Barth 1981: 77–79.

Explaining Why Marings Fought / 211

had the war process, not people, responding to demographic and ecological changes, escalating to territorial conquests, and counteracting stresses associated with population pressure. And by incorporating such events as territorial conquests in a reified process, I could give short shrift to human agency in treating them. If I appealed for more data, it was for data on the demographic and ecological changes assumed to trigger process responses (Vayda 1976: 41) and not for data on the actual motives and intentions of those who were taking land or on the social and political as well as ecological contexts in which they were doing so. If, with due attention to context and human agency, we now look more closely at the taking of enemy land, what, if anything, may we infer from it for causal explanation of Maring warfare? Contrary to the assumptions made by me in the 1970s and by other anthropologists subsequently (e.g., Ember 1982; Ferguson 1984: 30–31), the fact that victorious warriors sometimes take enemy land and benefit from doing so has, by itself, no necessary explanatory import. To think that it does is a fallacy, not because it involves putting consequences forth as causes but rather because it involves putting them forth without due concern for causal histories or selection mechanisms, as if the mere fact of their being beneficial automatically conferred causal efficacy upon the consequences. The mechanisms to which I am referring are not hypostatized ones in a reified process, but rather such mechanisms as intentional action, reinforcement, and natural selection, ontologically grounded in the actions, properties, and experience of individual human beings. Thus, land may be there for the taking as a result of fighting and may indeed be taken, but to claim territorial annexations as not a mere by-product of fighting requires showing in particular cases that those fighting were doing so with intentions of taking land or else that, even in the absence of such intentions, territorial annexations were affecting fighting through the operation of reinforcement, natural selection, or some other mechanism whereby beneficial consequences of behavior are causes of its repetition or spread. The next section will include further, more concrete discussion of these points in relation to particular Maring wars.

212 / Chapter 9

With respect to reinforcement as a mechanism, a point that I have made elsewhere is that its operation cannot be taken for granted but must rather be supported with evidence and argument (Vayda 1986: 304–305 and 1987a: 505). If, on the contrary, it is assumed that any rewarding outcome serves automatically to reinforce the behavior that led to it, we may be invoking reinforcement only in unmeditated, knee-jerk fashion and thus still committing the fallacy of imputing causal efficacy to certain consequences simply because they are beneficial (see Jochim 1981: 199–201 on warfare and reinforcement). We would likewise be committing the fallacy if natural selection were invoked as a causal mechanism every time we sought to explain items of behavior by their particular beneficial outcomes (cf. Gould 1987: 26–50 on “cardboard Darwinism”). Once we admit that benefits may, as often as not, be the products of happenstance rather than design, it cannot be taken for granted that a particular mechanism, whether reinforcement, natural selection, intentional action, or something else, is operating in a particular case of behavior with benefits.158 Although these points are clear to me now, they were not clear to me when I published my studies of war in the 1970s. Accordingly, adequate specification of causal histories or selection mechanisms to justify ascribing causal significance to territorial conquests was not a concern in those studies. I have no sure answer to why such unwarranted ascription should have ever seemed plausible or acceptable to me and others. It may be worth noting, however, that my earlier explanatory claims regarding territorial conquests exemplify a common, even if unsatisfactory, frictionless mode of functional explanation and that Elster (1983b: 101–105, 1985: 28) has speculated that this has its roots not only in individual psychology but also in the history of ideas, including notions about the goodness and omnipotence of God or Nature despite apparent evil, sin, suffering, and monstrosity in the world. Thus, if confronted by the

158 For more on the defects of some functional or consequence explanations in which such mechanisms are either not specified at all or are simply taken for granted, see the discussions of consequence explanations in chapters 1, 7, and 10 in this book and also Vayda 1987a: esp. 502–506 and Elster 1980: 125–128 and 1983a: chapter 2.

Explaining Why Marings Fought / 213

horrors of war, we may be predisposed to try to make sense of them by seeing them as necessary for one or another beneficial outcome.159 I erred not only in too readily ascribing causal efficacy to territorial conquests but also in assuming that population pressure or land shortage must be what made conquests beneficial. The acceptability or plausibility of this assumption may have lain, at least partly, in widespread and erroneous Malthusian notions that population pressure or the threat of it is a problem in all societies (see the references cited in Vayda 1976: 5) or that, as Jochim (1981: 1) puts it, current dangers of resource depletion and overpopulation “are but special and magnified variants of similar problems that have faced pygmies, peasants, pastoralists, and princes.” Actually, in the introduction to my 1976 book on war, I cited Cowgill’s arguments (1975) against taking population pressure for granted, and I then recommended the following: “Instead of starting with the assumption that all populations have to contend with the same problems (whether these be problems of overpopulation or whatever), we need to find out what problems particular populations actually are or have been confronted with” (Vayda 1976: 5–6; cf. Vayda et al. 1980: 186). However, as Hallpike (1977b: 557) remarks in a review of the book, I ignored my “own good advice” and without adequate evidence assumed that either actual or potential population pressure was a problem for all Maring groups and that relief from it was the critical benefit which warfare provided to them. I did acknowledge an absence of clear indications of pressure for all except two Maring groups, but still, as Sillitoe (1977: 74) notes, I begged the question by arguing that pressure was a threat to every group because of the continuing possibility of changes in demographic and ecological fortunes and that this possibility made the institutionalization of warfare advantageous or adaptive for Marings in general (Vayda 1971: 20–23 and 1976: 39–42). 159 A “promiscuous” inclination to functional or consequence explanations, observed in young children, has been thought to be increasingly restrained in us as our acquisition of causal mechanical ideas and knowledge proceeds. However, recent psychological research (Lombrozo et al. 2007) suggests that the inclination is never truly overcome and becomes salient again in patients with Alzheimer’s disease.

214 / Chapter 9

Land Shortage as a Cause of Fighting by Two Maring Groups Let us now consider actual, rather than potential, population pressure. By saying in his review that I had no evidence at all for the former, Hallpike was exaggerating. For two Maring groups, the Kauwatyis and the Kundagais, indications of past pressure (described in Vayda 1971: 20–21 and 1976: 39–40, and Lowman 1980: 139–149) included the fact that their territories, unlike those of other Maring groups, had extensive tracts of permanent grassland and degraded secondary forests. Furthermore, aerial photographs and/or informants’ statements indicated that, at the time of my field research, the two groups were short of primary forest, which was needed not only for new swiddens but also as foraging areas for pigs and as sources of game, firewood, building materials, and various wild food plants. The Kauwatyis had gone so far as to impose upon themselves a taboo against making swiddens in their primary forest. I stated in my book as well as earlier (Vayda 1971: 21 and 1976: 40) that shortage of primary forest may have been important in promoting land encroachments and aggressions by the Kauwatyis and Kundagais. In recent times, these were the only two groups committing land encroachments that led to warfare. I argued, moreover, that they were the only two large Maring groups which, if not for the intervention of the newly established Australian administration in 1956, might have succeeded in permanently displacing other groups and taking over all or part of their territory in the aftermath of the Maring wars of the 1950s. And I suggested that if the Kundagais and Kauwatyis had taken enemy land, it would have been because they had attained thresholds of population pressure not reached in recent times by other Maring groups (Vayda 1976: 40–41). With my changed ideas about explanation, I no longer would look for some regularity in the relationship between population pressure and actual or attempted territorial conquests and then expect to use that regularity mechanically to explain the course of wars like those of the Kundagais and Kauwatyis. However, I still regard population pressure or land shortage to have been a

Explaining Why Marings Fought / 215

factor in the Kundagai and Kauwatyi wars of the 1950s. Accordingly, a consideration of some details of two of those wars can be useful for bringing out differences between acceptable and unacceptable modes of explanation. The Kundagais fought their last war against the Ambrakuis, an enemy from ancestral times. Kundagai and Ambrakui informants agreed on the cause of the war. When Kundagais began to make swiddens on Ambrakui territory, the Ambrakuis were enraged and killed a Kundagai man who was hunting in forest belonging to the Ambrakuis. Then the Ambrakui warriors carried the corpse to their border with the Kundagais and shouted, “Why have you taken our land? We are not yet extinct. Our bellies are hot with anger. We killed your man in our forest and now you can come and get him and bury him.” As noted earlier, such proceedings constituted a declaration of war (cf. Rappaport 1968: 119). The Kundagais collected their corpse the next day and then, three days later, went off to the designated, borderland fight ground where the Ambrakui warriors were waiting for them. The ensuing “nothing fights” continued on a daily basis for several months without fatalities. However, with escalation, there came a day when the Ambrakuis killed four Kundagais and the Kundagais killed five Ambrakuis. Instead of going to the fight ground the next morning, the Kundagais rushed down upon the Ambrakui settlements and destroyed gardens and killed whom they could. The Ambrakuis fled across the Jimi River to take refuge with friends and affines among the Mimas, a non-Maring group. An Australian patrol officer and his police repatriated the Ambrakuis in the latter part of 1957, about three years after the people had gone into refuge. Similar in genesis to the last Kundagai-Ambrakui war was the war that was fought between the Kauwatyis and the Tyendas in 1955. For the second year in succession, Kauwatyi men had been making swiddens on Tyenda land. While they were cutting down trees, some Tyenda warriors came on the scene, angrily gave chase to the Kauwatyis, and caught one of them in the back with a fatal arrow. The Kauwatyis collected their corpse and wailed, and the Tyendas, shouting exultantly, returned to their houses. The passage of more than a year made the

216 / Chapter 9

Tyendas regard the issue as settled, but then some Kauwatyis suddenly appeared and butchered seven Tyendas (three men, three women, and a child) who were working in their swiddens. The next dawn, Kauwatyi warriors appeared en masse and fully armed at the Tyenda settlements and proceeded to wreak death and destruction. This was the only time in recent Maring military history that the fighting force of a large group from the more densely settled parts of the Maring area attacked its foes in their settlements without having engaged them previously in prearranged fight-ground battle. Tyenda houses were set afire and pigs were killed, as were twenty-three of the approximately three hundred Tyenda people. The rest fled for refuge, mostly to nearby Kundagai territory where affines and friends lent them land and planting material for making swiddens. As will be discussed below, the refuging Tyendas subsequently gave a substantial portion of their territory to these Kundagai friends, whose own land shortage can be seen to have been a factor in the last Kundagai-Ambrakui war. While in refuge, the Tyendas also continued to harvest crops from their own land, to which they returned permanently in 1956 after an Australian patrol officer and a complement of rifle-armed New Guinea policemen began to extend Pax Australiensis into the Jimi Valley. If not for the newly established Australian administration in the Jimi Valley, a consequence of the wars just described might have been the permanent annexation of enemy land by the Kundagais and Kauwatyis. Both groups, after routing their foes, continued to farm the land that had been the object of the encroachments leading to war. No fear of enemy ancestor spirits (see p. 208) was indicated by informants from either group. My Kauwatyi informants in 1963 justified their actions by saying simply that the land they took was good land and belonged to an enemy—as if taking such land were an appropriate venting of hostility. My Kundagai informants boasted that only they, and not their enemies, were capable of annexing another group’s land after warfare. One factor which might have deterred both the Ambrakuis and Tyendas from returning to their settlements and repossessing their land was numerical weakness. Both of them had enemies

Explaining Why Marings Fought / 217

more than twice as numerous as they: the Kundagai group had some 600 people and the Kauwatyis had 850, while there were approximately only 250 Ambrakuis and 300 Tyendas.160 Indeed, the Ambrakuis were skittish about being repatriated even with Australian help. In September 1956, some of them accompanied a patrol led by an Australian officer to their abandoned territory. The officer instructed the Ambrakuis to resettle there and left a few policemen in the general area to help in the work. He also held discussions with both Kundagais and Ambrakuis about their lands and then affirmed boundaries which, according to what I was told by Ambrakui informants in 1963, corresponded to the ancestral ones between the lands of the two groups. However, when the officer revisited the territory in June of 1957, he saw that resettlement was still far from complete: the men had simply been visiting their old lands, while the women, children, and pigs had, for the most part, remained in their Mima refuges. New gardens had not been made in the old lands. Only with increased police supervision and further warnings to the Kundagais was the full return of the Ambrakuis to their territory finally effected later in 1957. Less than five years later, the Kundagais made new land encroachments. In response to the complaints of the Ambrakuis, a new Australian officer sent policemen to destroy Kundagai plantings on Ambrakui land and to arrest leaders of the offending Kundagais and take them to jail at the Jimi River patrol post. How effectively such punishment would deter the Kundagais was not clear at the time of my field research. The Tyendas were less timid about being repatriated under Australian aegis. In mid-1957, when the Ambrakuis were still staying close to their Mima refuges, the Australian patrol officer found that the Tyendas were going ahead well with resettlement and had prepared extensive new swiddens in their own territory. This difference in the pace of resettlement may be related to the fact that the Tyendas, unlike the Ambrakuis, had support from

160 These figures are estimates based on a 1963 census. In the case of the Kauwatyis, the number of people living on their territory in 1955 may have been swelled to over 1,000 by an influx of war refugees (Rappaport 1968: 143). My account of Ambrakui repatriation here is taken from Vayda 1976: 28.

218 / Chapter 9

two powerful Maring groups. One of these groups, the Kundagais, had given refuge to the Tyendas. The other, the Manambans, a traditional foe of the Kauwatyis, had shown its support by its response to a Tyenda kinsman’s pleas for revenge against the Kauwatyis who had killed his children and brothers: the Manambans had gone en masse to challenge their old enemies by killing a Kauwatyi man. This killing led to the Kauwatyi-Manamban war of 1956. A reasonable speculation is that without Australian intervention the Ambrakuis would not have repossessed any of their territory, but the Tyendas would have repossessed at least some, and possibly most, of theirs.161 On the basis of the events just described, what may justifiably be said and what may not be said about land shortage as a cause of war among the Marings? Even if there is some warrant for saying that the Kauwatyi and Kundagai encroachments that led to war were a result of land shortage, I now agree with my critics (e.g., Sillitoe 1977: 74; Hallpike 1973: 458; King 1976: 313–314) about the fallaciousness of using these cases to support the claim that Maring groups in general were apt to suffer from land shortage and that this is what made institutionalized warfare adaptive or advantageous for all of them. It is, however, equally fallacious to argue, as Hallpike has seemed to at times, that land shortage cannot have been a factor in any Maring wars since it was, as I acknowledged, not directly a factor in many of them. Thus, in his 1973 paper Hallpike cited me on the absence of population pressure among Maring groups who fought just as frequently as the Kundagais and Kauwatyis, and he concluded from this that there was not “any significant relationship between aggression and land shortage” among the Marings. However, on the next page of the same paper, he conceded my having established that in some cases Marings fought over land and that warfare might be the means whereby groups short of primary forest for new swiddens acquired land from other groups (Hallpike 1973: 457–459). Committing fallacies of undue generalization may go hand in hand with inattention to agency and to the variability of the contexts in which agents act. In accord with persisting uniformist, typologizing, essentializing culture-as-norms tendencies in 161

This speculation was previously put forward in Vayda 1971: 16 and 1976: 29–30.

Explaining Why Marings Fought / 219

anthropology (Pelto and Pelto 1975; cf. Vayda 1994), attempts have been made in New Guinea studies to pigeonhole the warfare of whole regions like the New Guinea highlands (Berndt 1964) or, only a little more modestly, the warfare of ethnolinguistic units whose total populations comprised numerous warring local groups (Sillitoe 1977).162 A partial corrective to such endeavors and their neglect of context-dependent variations is the recognition of local differences in how and why fighting occurred among members of a single ethnolinguistic unit. But even when, in refutation of gross generalizations, some local differences are noted (as, for example, they have been by Meggitt [1977: 178–179] among the Engas and by Morren [1987: 484] among the Miyanmins), too much may be taken for granted concerning context and agency, and it may be too simply assumed that the variations worth noting are those which correspond to some broadbrush dichotomy, such as that between fringe or frontier areas and more densely settled central areas. These points are well illustrated by the feature that I have been discussing in this section: going to war because of being short of land. This has been claimed by Sillitoe (1977) to distinguish the warfare of an entire ethnolinguistic unit (Abelam warfare as compared with the warfare of other New Guinea peoples) and has been used by Meggitt (1977: 178–179) to make distinctions between Fringe Enga and Central Enga warfare. However, on the basis of the accounts given in this section, it could be argued that differences with respect to the feature may occur among the belligerents in a single war. In other words, the statement that they were fighting because they were short of land may be warranted about the Kundagais and Kauwatyis but not about their enemies, the Ambrakuis and Tyendas. There is no evidence that either of these latter groups was short of land. In fact, after their defeat at the hands of the Kauwatyis, the Tyendas gave about 30–40 percent (1.8 square miles) of their total territory to Kundagai friends, including those who had provided them with refuge (see Vayda 1976: 32 for details). Using a concept of “economic density” in

162 Some examples of the estimated total populations of ethnolinguistic units whose warfare has been typologized by Sillitoe (1977) are: Maring, 7,000; Abelam, 30,740; and Chimbu, 55,000.

220 / Chapter 9

my earlier publications, I defined as “more densely settled” those Maring groups that had either close to a hundred people or many more per square mile of land under cultivation or in secondary forest (Vayda 1971: note 23 and 1976: note 23). This is noteworthy because Tyenda density prior to the land transfers was only about 69 people per square mile of such land. This density was the lowest among the ten Maring groups whose territory was within the 55 square miles for which aerial photographs were available and whose density could accordingly be calculated (Lowman-Vayda 1968: 202–203). No comparable calculations could be made for the Ambrakuis, but Kundagai informants did state that the Ambrakuis had suffered considerable population decline prior to their last war. They were said to have been as big a group as the Kundagais at one time. Indeed, far from having land in short supply and fighting for it, the Ambrakuis and Tyendas might be seen as having retaliated against the Kundagais and Kauwatyis in order to show that their having more land than they could use did not mean that their old enemies could act against them with impunity. When groups became extinct, their land was in fact taken over by other groups among the Marings; the Ambrakuis, Tyendas, and Kundagais themselves had annexed the territories of groups that had become extinct in the more malarial, lower altitudes along the Jimi River. By fighting back, the Ambrakuis and Tyendas were telling their enemies that they were not dead yet and that their land was not there for the taking. They might well have been fighting more for honor and reputation than for the land which the Kundagais and Kauwatyis were trying to get from them. Honor and reputation were important. Small groups without men who, having distinguished themselves in warfare, could attract wives from other groups were disparaged by Marings as “vulnerable seeds, not deeply rooted trees.” By contrast, groups like the Kauwatyis commanded respect because of their size and military prowess, and their men were sought after as affines (Lowman 1980: 13, 182, and passim). If we are not blinkered by uniformist, culture-as-norms assumptions, we should hardly expect groups as different in size, prowess, and demographic situation as were either the Kauwatyis from the Tyendas

Explaining Why Marings Fought / 221

or the Kundagais from the Ambrakuis to have been fighting for the same cause. Indeed, a better guide than many anthropological theories and assumptions for explaining why particular people fought at particular times is an old anecdote about two men who asked each other why their respective groups were waging war. The man from the first group said, “We fight for plunder: we are poor and hope to get booty. And what do you fight for?” The man from the second group answered, “We fight for honor and glory.” The first man’s rejoinder to this was: “People fight for what they do not have.”163 The lesson to be drawn here is simple. By attending to contexts and their variability, we can see what things or properties are absent or in short supply for particular human beings at particular times. And by attending to agency, we can determine whether these are things or properties which those human beings are fighting for, even if their enemies should be fighting for something else. This lesson, interestingly enough, is not denied by Hallpike (1973: 459), who acknowledges that, at least in some cases, people “fight over land, or anything else which they fancy and which is in short supply.” Oddly, however, Hallpike goes on to say in his next sentence that this does not constitute an explanation of why fighting occurs. Somewhat similarly, Koch (1974: 8–9), although avowedly rejecting ecological explanations in which land shortage is emphasized, states that wars in the New Guinea highlands may result from “conflicting interests in land, especially in valleys without surplus of arable ground.” These apparently antithetical assertions will be dealt with below.

163 In the version of this anecdote which the Yugoslav writer, onetime political leader, and subsequent dissident Milovan Djilas (1962: 80) reports as having told to Stalin, the first man was, like Djilas himself, a Montenegrin, and the second was a Turk. I avoid ethnic labels in my own version because any suggestion of a single distinctive goal as characteristic of all the fighting of all the people belonging to a particular ethnic or ethnolinguistic unit is of course directly contrary to what I am arguing for here. The statement that people fight for what they do not have is an example of what I refer to in chapter 1 as generalizations referring to recurrent but not necessarily invariant cause-and-effect sequences encapsulated for us not in formal propositions but rather in platitudes, proverbs, and homely phrases or sayings.

222 / Chapter 9

But, first, the relation between land shortage as a cause of war and territorial annexation as a consequence of it will be considered to underscore points made in the preceding section. As stated earlier, explanatory import may, in particular cases, be claimed for such annexation if it can be shown either that those fighting were doing so with intentions of taking land or else that, even in the absence of such intentions, territorial annexations were contributing to fighting through the operation of reinforcement, natural selection, or some other feedback mechanism. In their last wars against the Tyendas and Ambrakuis, the Kauwatyis and Kundagais may be said to have been fighting for land insofar as intentions of taking it by force from their enemies may be attributed to them on the basis of the evidence of land shortage and encroachments and the statements which Kauwatyi and Kundagai informants made about why they had fought. Territorial conquest thus has explanatory import in these cases because it was a goal for which the Kauwatyis and Kundagais may be said to have consciously provoked and conducted war. A different scenario is suggested by the accounts which informants from other large Maring groups, the Tukumengas and Yombans, gave of two wars in which their side had been victorious. The offenses giving rise to these wars were unrelated to land shortage or encroachments. In both wars, weeks of fight-ground battle culminated in the killing of several enemy warriors by the ultimate victors, who thereupon decided to escalate to routing so as not to give their enemies a chance to even the score quickly. Routed and with their houses, swiddens, and pigs destroyed, the enemies could be kept from being a threat to the victors for some time. While the informants gave no indication that the routing had been undertaken with the intention of permanently displacing their enemies, they allowed for the possibility of territorial annexation if the enemies failed to return from refuge. Although the enemies did in fact return in these cases, it may be instructive to consider a hypothetical case conforming to the actual ones except in two respects: the enemies do not return and their land is annexed. In such a case, benefits accruing to the victors might include not only having a territorial buffer against

Explaining Why Marings Fought / 223

their enemies but also obviating population pressure and its stresses in the future, even if the victors suffer from no shortage of land at the time of the annexation.164 This hypothetical case is offered here as an illustration to underscore points made in the preceding section. Since those fighting are not doing so with intentions of gaining land, intentionality cannot be invoked here to justify ascribing causal significance to territorial conquests. And any inclination to invoke some other mechanism like natural selection to justify such ascription must be tempered by my earlier warning against invoking mechanisms without evidence and argument to support claims of their being in operation. In the absence of support for such claims, any benefits eventually gained by the victorious side as a result of taking enemy land in the hypothetical case would have to be regarded as by-products of fighting rather than as somehow their cause.

Explanatory Relativity As stated at the beginning of this chapter, a question such as why Marings fought can be given different correct answers, depending on our assumptions about what the question means. Divergent assumptions may be made, for example, about the scope of the question. Garfinkel (1981: 21ff.) illustrates this with the felonious Willie Sutton’s alleged answer in prison to a priest who was bent on reforming him and had asked him why he robbed banks. By saying, “Because that’s where the money is,” Sutton was answering a question much narrower in scope than that intended by the priest, whose concern was why Sutton robbed at all instead of leading an honest life and not why Sutton robbed banks instead of newsstands.165 On the other hand, an answer may be to a much broader question than that intended by the questioner. Garfinkel’s illustration of this is a 164 Costs as well as benefits might, of course, accrue to the victors. Their enlarged territory might, for example, make them, like the Tyendas and Ambrakuis, the target of the aggressions of land-hungry neighbors. 165 On the apocryphal character of this much-cited exchange, see chapter 1, note 22.

224 / Chapter 9

murder suspect’s evasive answer to the detective who has asked him why the victim has died. By saying that “everyone has to go sometime,” the suspect is answering not the detective’s intended question of why the victim died when he did instead of living to die at another time but rather the question of why the victim died at all instead of living forever (Garfinkel 1981: 22). These illustrations conform to Garfinkel’s general claim that explanations are both made and either accepted or not accepted with at least implicit reference to specific alternatives or contrasts. This claim is what Garfinkel (1981: 28–34) calls “explanatory relativity” (cf. Kincaid 1986: 505). Recognizing explanatory relativity helps to make sense of such initially puzzling assertions as Hallpike’s statement that the fact that Marings sometimes fought for land “does not explain why the Marings fought” (Hallpike 1973: 459). Evidently the question of why they fought is, for Hallpike, not something as narrow as the question of why some Marings, like the Kauwatyis or Kundagais, fought at a particular time and not at another time. To that question, Hallpike might accept, as at least a partial answer, that at that time the people in question needed more land and had weak neighbors from whom they could expect to take it by force. Just what broader question Hallpike has in mind is not made explicit in his published discussions of my New Guinea studies, but two possibilities may be noted. One is the question of why Marings or some other people fought at all (in contrast to never fighting), and the other is the question of why they fought frequently (in contrast to fighting infrequently). That the first is Hallpike’s true question is suggested in his review of Meggitt’s book (1977) on warfare among the Mae Enga people of the New Guinea highlands: the grounds for his objection to Meggitt’s explanation of Mae Enga fighting as responses to land shortage are the likelihood that the people “would still fight even if land shortage were not the principal occasion of conflict” (Hallpike 1977b: 556). Thus, he seems to be saying that land shortage can be accepted as an explanation of why Mae Engas fought only if without land shortage they did not fight at all (cf. Feil 1977). Other statements made by Hallpike suggest, however, that his real concern is explaining why people fight frequently. Thus,

Explaining Why Marings Fought / 225

in his book on social evolution, he refers to the “high levels of warfare” in New Guinea as a challenge to ecological explanations (Hallpike 1986: 106). In his earlier monograph on the Tauade people of New Guinea, their “large amount of violence” or “violence of considerable intensity” is explicitly the main object of explanation (Hallpike 1977a: 82, 275). Although thefts of pigs and other specific causes of hostilities are noted in Hallpike’s detailed accounts of Tauade warfare, he refers, in the concluding chapter of his monograph, not to these but rather to such purported factors as the Tauades’ Heraclitean mentality and their ethos of aggression and destruction. I have expressed skepticism about these factors elsewhere (Vayda 1979a), but my point here is simply that it makes sense to regard Hallpike’s reference to them as an attempted answer to the question of why Tauades fought frequently and not to the question of why they fought at some particular time. Indeed, in a gracious letter written to me after publication of the article on which this chapter is based, Hallpike (1990) referred to his interest in the “general level of warfare in a society” and joined me in regretting that differences in the questions we were asking had not been clear to us much earlier. It may well be that another anthropologist, Klaus Koch (1974), was also trying to answer the question of why fighting was frequent and not some narrower question that might be treated under the heading of the explanation of fighting or war. Thus when Koch deals with warfare among the Jalés and other peoples of the New Guinea highlands, he emphasizes such factors as “patterns of socialization that develop an aggressive and vindictive personality” and the purportedly general absence in the New Guinea highlands of institutionalized and traditional third-party authorities to keep intergroup quarrels from escalating to warfare. If Koch were also to make some explicit reference to the frequency of these quarrels, it would be quite clear that the question he is concerned with is why people fought frequently. The question is important, but of course this does not mean that questions about what people were fighting for in the case of particular wars are not important. Thus, if we wish to explain the frequency of fighting by referring, inter alia, to the frequency

226 / Chapter 9

of intergroup quarrels, then obviously it is important to have data showing that particular wars resulted from such quarrels (whether over land, pig thefts, or something else), rather than simply being, as has been suggested for some New Guinea fighting (Feil 1987: 68–69), expressions of pervasive and perpetual intergroup enmity which were not linked to specific incidents or provocations to fight. Furthermore, if a practical concern is to control or put an end to fighting, it can be important to know what its occasions are. Thus, if intergroup fighting is viewed as an expression of abiding, reasonless enmity, then governments, like the Australian administration confronted with resurgent warfare among some New Guinea highlanders in the late 1960s and early 1970s, may see no solutions other than the use of mobile riot squads and similar police measures; however, if specific intergroup quarrels are recognized as leading to war, then their resolution by such means as litigation may be encouraged, and if it is seen that some (even if not necessarily all or most) of the quarrels in question are over land and that the courts are inadequate to deal with them (as has been the case in some New Guinea highland areas since the 1960s), then consideration can be given to making more land available by such means as providing people with opportunities for resettlement (Meggitt 1977: chaps. 8–10; Vayda 1979b: 194–196; cf. Gordon and Meggitt 1985: 10 and passim). Another argument for the importance of the question of why people fought at particular times is one which I have, in effect, already presented, namely, the need to give the variability and context-relatedness of purposeful human behavior their due and to avoid the errors of uniformist, typologizing, culture-as-norms approaches whereby people like the Marings who sometimes fought for land are assumed, because of the categories in which they have been placed, never to have done that. In this connection I have indicated that the Kundagai and Kauwatyi actions described in the preceding section were of a kind which, according to Sillitoe’s typology (1977), occurred only among a single New Guinea people, the Abelams, and not at all among the Marings. Lesser (1968: 95), another of my critics, has assigned all New Guinea societies to the category of “primitive stateless so-

Explaining Why Marings Fought / 227

cieties” without “organized offensive warfare to conquer people or territory,” and this has made him deny that Marings or members of any other New Guinea societies ever fought for land. As Koch (1974: 8) has noted, errors like Lesser’s have been made by many writers committed to evolutionary typologies. These errors and the other ones discussed and illustrated in this chapter should of course be avoided. The more detailed prescriptions given in chapters 1 and 8 for dealing with errors of reification and fallacies of functional explanation in particular need not be repeated here. However, an admonition first made in chapter 1 does bear repeating in light of the account given here of the long and acrimonious debate that Hallpike and I could have avoided if it had been clearer to us that our questions about the causes of intragroup fighting in New Guinea were not the same. The admonition is simply that we should make as clear as possible, both to ourselves and to others, just what why-questions we are asking. Insofar as our understanding of why-questions depends, as discussed, on the specific contrasts or alternatives we have in mind (“robbing banks rather than robbing newsstands”), it is worth noting as well that becoming more explicit about the contrasts may be one way of achieving more of the needed clarity about the questions being asked.

10 The Anthropology of War: Polemics and Confusion

This chapter is a review of The Anthropology of War, edited by Jonathan Haas. (Cambridge: Cambridge University Press, 1990, pp. xiv + 242.)

The nine articles in The Anthropology of War provide useful data and references on war in various pre-state societies, argue sundry polemical points about how and why war and peace should be studied and explained, and raise certain methodological and explanatory issues which deserve discussion here even though no consensus on them, nor even any substantial clarification of them, appears to have been achieved at the weeklong 1986 seminar from which the articles have been drawn. The stated goal of the seminar was arriving at “a better understanding of the causes of both war and peace in pre-state societies and the impact of war on the evolution of those societies” (xi), and the articles contribute in varying degrees to meeting this goal, sometimes in ways different from those intended or claimed by the authors themselves. Thus Robert Carneiro, using only Williams (1870) and Trimborn (1949) as his sources, describes warfare in Fiji and Colombia’s Cauca Valley in his article. The description is useful for reminding us that pre-state warfare could be much more than low-casualty ritualized hostilities and revenge-motivated sneak attacks by 229

230 / Chapter 10

small warrior bands and, contrary to evolutionary typologies mentioned in the preceding chapter (p. 226–227), could instead feature fighting forces of hundreds and even thousands of men, fierce hand-to-hand combat, battles in which hundreds died, and territorial conquest as both an aim and a consequence of fighting. However, Carneiro gives no adequate evidence or justification for regarding the Fiji and Cauca cases as “reasonably typical of societies at the chiefdom level of political development” (p. 207), and so one is left doubting that the cases substantially strengthen his claims about chiefdoms in general as creatures of war and, even more sweepingly, about the centrality of war in political evolution. Significantly, in the article immediately preceding Carneiro’s in the volume, the conclusion reached by volume editor Jonathan Haas about how bands become tribes (regarded as preceding chiefdoms in his evolutionary typology) is that neither a model assigning a primary causal role to warfare, nor a model emphasizing consolidation in response to environmental stresses, is adequate by itself to account for the course of political development among the Anasazi Indians of northeast Arizona prior to A.D. 1300. The actual changes extended over centuries and occurred, according to Haas, in response to both warfare and environmental problems. An implication of Haas’s article, in line with my methodological arguments elsewhere in this volume, is that events rather than evolutionary types or long-term “processes” like tribalization should be our primary objects of causal analysis, but this is not explicitly stated by him. Napoleon Chagnon’s article is presented as the beginning of a synthesis of concepts and theory from sociobiology and from the cultural materialism championed by Marvin Harris (see the brief discussion in my chapter 1 of this book) in order to deal with the genesis and development of conflicts, including intergroup warfare, in band and village societies like that of the Yanomamö whom he has been studying for several decades. Thus he takes from cultural materialism the idea that conflicts are associated with securing or protecting material resources or the means of production, but then, drawing on sociobiology, he posits that they relate also to the means of reproduction. Indeed he suggests that so-called reproductive conflicts, related, for example,

The Anthropology of War / 231

to female abductions, rapes, seductions, defaults in bride-price payments, and insults to men who, in the sociobiological perspective, are “reproductive competitors,” predominate in band and tribal societies and diminish in relative importance only as the societies in question approach the limits of their territories’ carrying capacities. The distinction between the two kinds of conflicts is subsumed by Chagnon’s more general distinction between somatic and reproductive effort, the former referring to the activities whereby individuals achieve their bodily survival and the latter referring to the activities whereby they (or relatives with whom they share genes) achieve reproductive success. Notwithstanding the misleading use of the term “effort” (and, elsewhere in Chagnon’s article, “striving” and “strategy”), the distinction, it should be noted, is made not in terms of the conscious goals or intentions of the actors but rather in terms of the consequences (whether intended or not) which the behavior in question has for their survival and reproductive success. As for their actual and immediate goals and intentions, Chagnon, according to Clark McCauley’s introductory “overview” article, limited himself at the seminar to supposing that actors behave “as if” to maximize reproductive success. That actions or activities in war or related to it can have important consequences for both survival and reproduction can hardly be questioned, and there would be no problems if Chagnon were simply showing this systematically for the Yanomamö (a large task in itself) and asking other anthropologists to do likewise for other societies. However, like the sociobiologists whose work he draws on, Chagnon wants to use the reproductive consequences to explain behavior. Evidently this is to be done by using differential rates of reproductive success as one measure of the costs and benefits of fighting or other behavior (p. 79) and by assuming that there is selection for behavior from which more benefits accrue. In criticizing such “consequence explanations,” various scholars, including Elster (1983b: 101–108), Little (1991: 98–102), and me (Vayda 1987a, 1989), have argued that the fallacy lies not in putting consequences forth as causes but rather in putting them forth without due concern for how they affect behavior, as if the mere fact of their being beneficial justified

232 / Chapter 10

their being regarded as causes of the repetition or spread of the behavior in question.166 Chagnon follows sociobiologists in assuming natural selection to be the mechanism whereby behavior is affected, but the fallacy remains insofar as, quite unjustifiably, he takes the operation of natural selection for granted in all his consequence explanations rather than using evidence and argument to support the claim of its operation in particular cases. His explanatory procedures amount then to what Gould (1987) has called “cardboard Darwinism,” the facile equaling of some current benefits with causes or origins so that the complexity and diversity—the vicissitudinous twists and turns—of historical causation are missed. In another article in the volume, Brian Ferguson also considers consequences of war, both ecological and demographic ones. However, these are used by him not for what the title of his article promises, namely, “Explaining War,” but rather for making the point that mixtures of costs and benefits resulting from war entail difficulties for making judgments about its adaptiveness or maladaptiveness. Ferguson’s article is rich in references and valuable as a review of literature on causes and correlates as well as consequences of war. But, like other articles in the volume, it falls short of goals set by the author. For Ferguson the goal is nothing less than a synthesis of materialist theorizing for the purpose of providing “a comprehensive explanation of war in stateless societies” (p. 26). What he actually gives us in most of his article, however, is not any unified theory consisting of systematic propositions but rather advice, similar to that to be found in books like Notes and Queries in Anthropology, about the kinds of things and interrelations to which attention should be paid, for example, whether and how subsistence technology and skills and domestic authority patterns are carried over into combat (pp. 34, 37), how war and exchange are related (p. 39), how alliances are formed and how long they last (p. 43), and whether and how personal aggressiveness relates to decisions to fight (p. 44). For two propositions put forward by Ferguson in his arti166 See also the discussion of consequence explanations in chapters 1, 7, and 9 of this book.

The Anthropology of War / 233

cle, a claim of generality is made but hardly justified or supported adequately. One proposition is that material or infrastructural factors, such as shortages of essential material resources explain why war occurs, and the other is that “wars occur when those who make the decisions to fight estimate that it is in their material interest to do so” (p. 30). Apart from Ferguson’s fuzziness about what it means to explain why war occurs, a stumbling block for his propositions is the simple fact of many reports—including some by Chagnon in his already discussed article—of wars which occur not (or not primarily) on the basis of material factors or estimates of material interests. Indeed Ferguson himself does not rule out such wars but rather accords them “a low likelihood” (p. 30). Two of the articles in the volume are, in different degrees, throwbacks to the kind of simplistic, essentialist, idealistic anthropology to which the development of a simplistic, dogmatic cultural materialism was a polemic reaction. Thus Thomas Gregor’s article, noting the absence of warfare between the groups living in the Upper Xingu basin of central Brazil for the century for which records exist, poses the question of why Xinguanos are peaceful. Insofar as it refers to the Xinguanos’ antiviolent value system and “essentially antiviolent culture” (p. 119), Gregor’s answer harks back to Ruth Benedict’s broadbrush characterization of the Zuñi Indians as Apollonians in her Patterns of Culture (1934) and invites the same criticisms, that is, for using data selectively or interpreting them tendentiously. While most of his article is devoted to showing how Xinguanos “aim their behavior towards the ideal of peace” (p. 115) and how their ideas about violence, blood, and non-Xingu “wild” Indians remind them of what they must avoid, Gregor also refers to the aggressive hazing of resident husbands from other villages (involving such acts as submerging their canoes in distant swamps and filling their hammocks with ashes) and brutal witch executions, in which the attackers surprise their victim by setting upon him with arrows, machetes, and rifles. In addition to such intravillage violence, warfare is indicated as well. Gregor takes pains to emphasize that this consisted of strictly retaliatory raids against non-Xingu Indians and that Xinguanos regarded these as exceptions to the normal pattern of peacefulness.

234 / Chapter 10

Among these exceptions was a series of raids in which so many men, women, and children of the non-Xingu Txicão tribe were slaughtered as to leave only some desperate survivors, ready to live in peace with Xinguanos and to become assimilated to their culture. Information provided by Gregor himself thus negates his central question. His historical and other data do, however, begin to answer such empirically more defensible questions as why Xinguanos fought at some times and not at other times and why they fought some people but not other people. Unlike Gregor’s article, Thomas Gibson’s does not have the virtue of including data to counter the essentialist, uniformist, idealist characterizations set forth. The article compares three insular Southeast Asian peoples: the Buids of highland Mindoro, presented as highly valuing individual autonomy and being categorically averse to aggressive behavior; the Ilongots of northern Luzon, presented as valuing household rather than individual autonomy and regarding competition and aggression as characteristic of a stage in the life cycle of men; and the Ibans of Sarawak, presented as relating to the outside world by predation (hunting heads and chopping down virgin forests) and as making success in that the basis of intracommunity prestige. Neither the studies which Padoch (1982) and Cramb (1986, 1989) have made of variation and change among Ibans, nor any other works which might spoil his simple portraits, are cited by Gibson. Although he pays lip service to the importance of historical experience and remarks that it was different for each of his three peoples, he assumes it to have happened to each en bloc and to have thereby disposed each to develop its monolithic ideology. Thus Ibans as a whole had the experience of being predators and developed a predatory ideology, while Buids had the experience of being prey and developed an antiviolent ideology. Simple as this may sound, Gibson insists that he is not presenting a deterministic argument to explain ideologies: he notes as a possibility the failure to develop appropriate ideologies, and he attributes independent causal force to the ideologies themselves. This is exemplified in his claim, no doubt striking materialists as empty or tautological, that the “autonomous groups which have sur-

The Anthropology of War / 235

vived have done so because they possessed an ideology which rejected any form of dominance” (p. 143). Gibson’s article is followed in the volume by Neil Whitehead’s, which brings readers back into a real, complicated, and changing world. Making extensive use of unpublished archival material, Whitehead considers some three centuries (ca. 1500– 1820) of warfare among South American Caribs. He states at the outset that both Carib warfare as a distinct phenomenon and the possibility of an overarching causal explanation of it are illusory and that, moreover, only some Carib groups were persistently warlike and that, even among these, the causes and objectives of warfare varied considerably over time. The body of the article indicates substantial variation in war practices as well and shows the influence of changes in European colonial priorities. One conclusion reached is that conflict intensified as Europeans became more involved in it. Another conclusion is that no single factor is either an adequate explanation or a continuous determinant of war when seen in historical perspective. Also a far cry from Gibson and Gregor’s essentialist, uniformist assumptions are the arguments presented in Clayton Robarchek’s article, notwithstanding that all three authors, along with Whitehead, are identified in Haas’s preface (p. xii) as having represented the historical school of thought in opposition to the materialistic/ecological and biocultural theories of other participants in the 1986 seminar. After giving a graphic and detailed description of the nonviolent resolution of a territorial dispute among members of a Semai band in the mountains of central West Malaysia, Robarchek (p. 72) criticizes ecological and sociobiological approaches for a reification fallacy that is to be found equally in approaches like Gibson’s. The fallacy consists of relegating intentionality to culture, ascribing purposes to it, making it, in effect, think for the people. The examples which Robarchek gives of this are of decisions which, being “correct” by the criterion of reproductive success or ecological efficiency, have somehow been “institutionalized as part of the culture, as cow worship or female infanticide or polygyny or warfare” (p. 72) so that the actions of individuals can be regarded as their

236 / Chapter 10

simply following the dictates of their culture. Unlike those criticized by Robarchek, Gibson subscribes to no sociobiological or ecological criterion of correctness. However, like them, he lets culture, once it is established, think for the people, and he shows no concern for how they actually form intentions and make decisions in order to act. By contrast, Robarchek, while recognizing that culture, in the sense of shared values, beliefs, meanings, and assumptions, makes certain behavioral options more desirable than others and certain methods legitimate and others illegitimate means to ends, emphasizes that individuals must still decide whether, when, and how to put institutionalized ideas or norms into action. He illustrates this by referring to not only the Semai whom he studied but also Chagnon’s Yanomamö, about whom he asks why some men decide to attack a neighboring band on a particular day rather than another day and why that particular band rather than another. Decisions about such matters, insists Robarchek, are “not trivial or irrelevant or epiphenomenal” (p. 73). He suggests that, on the contrary, attention to them and data on them, including attention to their causes as well as effects, are indispensable for explaining what people actually do. We must surely agree here with Robarchek if we care about not committing either the fallacy of cultural reification or the previously discussed fallacy of consequence explanations like Chagnon’s. Robarchek’s questions about Yanomamö raiding, along with the distinctions which McCauley, in his overview article, makes between explaining the initiation of conflict and explaining its continuation, point also to the inadequacy of thinking simply that war is our object of explanation. Much of the polemics about explaining war may be related to this insofar as those involved, including participants in the 1986 seminar, mistakenly think, as did Hallpike and I in our prolonged debate,167 that their different explanations are answers to the same question when, in fact, they are answers to such different questions as why some particular people fight at one time rather than another, why they

167

See chapter 9 of this book

The Anthropology of War / 237

fight at all rather than never fight, why they fight frequently rather than infrequently, why some particular conscious goals or objectives rather than others are held by those fighting, and why their fighting has certain consequences rather than not having them (cf. my discussion in chap. 9). While differences among the authors of The Anthropology of War will no doubt remain, the volume does give hope that anthropologists concerned with war and peace will be able to set aside some of the futile polemics of the past in favor not of finding the grand theory to explain it all but rather of asking numerous more clearly defined questions and giving sharply focused answers to them. It remains to mention that the volume has many typographical errors. Most of these are inconsequential, but some, such as having “states” instead of “stages” three times (pp. 10–11) in a discussion of warfare in relation to evolutionary stages in the development of the state, are not.

Bibliography

Abberger, H. M., B. M. Sanders, and H. Dotzauer. 2002. The development of a community-based approach for an integrated forest fire management system in East Kalimantan, Indonesia. In P. Moore, D. Ganz, L. C. Tan, T. Enters, and P. B. Durst, eds., Communities in Flames: Proceedings of an International Conference on Community Involvement in Fire Management. Bangkok: FAO Regional Office for Asia and the Pacific, pp. 53–65. Adams, W., and D. Hulme. 2001. Conservation & community: Changing narratives, policies & practices in African conservation. In D. Hulme and M. Murphree, eds., African Wildlife & Livelihoods: The Promise and Performance of Community Conservation. Oxford: James Curry, pp. 9–23. Ahn, W., and C. W. Kalish. 2000. The role of mechanism beliefs in causal reasoning. In F. C. Keil and R. A. Wilson, eds., Explanation and Cognition. Cambridge: MIT Press, pp. 199–225. Alcorn, J. 1989. Process as resource: The traditional agricultural ideology of Bora and Huastec resource management and its implications for research. In D. A. Posey and W. Balee, eds., Resource Management in Amazonia: Indigenous and Folk Strategies Advances in Economic Botany. New York: The New York Botanical Garden, pp. 63–77. Alexander, R. D. 1979. Darwinism and Human Affairs. Seattle: University of Washington Press. Alexander, R. D. 1988. Evolutionary approaches to human behavior: What does the future hold? In L. Betzig, M. Borgerhoff Mulder, and

239

240 / Bibliography

P. Turke, eds., Human Reproductive Behavior: A Darwinian Perspective. Cambridge: Cambridge University Press, pp. 317–341. Alexander, R. D. 1989. Evolution of the human psyche. In P. Mellars and P. Stringer, eds., The Human Revolution. Princeton: Princeton University Press, pp. 455–513. Alexander, R. D. 1990. Epigenetic rules and Darwinian algorithms: The adaptive study of learning and development. Ethology and Sociobiology 11: 241–303. Aliseda, A. 2000. Abduction as epistemic change: A Peircean model in artificial intelligence. In P. A. Flach and A. C. Kakas, eds., Abduction and Induction: Essays on Their Relation and Integration. Dordrecht: Kluwer, pp. 45–58. Alvard, M. S. 2003. The adaptive nature of culture. Evolutionary Anthropology 12: 136–149. Anon. 1993. The savage can also be ignoble. Economist, June 12, p. 56. Anscombe, G. E. M. 1969. Causality and extensionality. Journal of Philosophy 66 (6): 152–159. Anscombe, G. E. M. 1981. Metaphysics and the Philosophy of Mind. Vol. 2 of The Collected Philosophical Papers of G. E. M. Anscombe. Minneapolis: University of Minnesota Press. Applegate, G., U. Chokkalingam, and S. Suyanto. 2001. The Underlying Causes and Impact of Fires in Southeast Asia, Final Report. Jakarta: Center for International Forestry Research (CIFOR). Aryawan, I. G. N., I. N. Widiarta, Y. Suzuki, and F. Nakasuji. 1993. Life table analysis of the green rice leafhopper, Nephotettix virescens (Distant) (Hemiptera: Cicadellidae), an efficient vector of rice tungro disease in asynchronous rice fields in Indonesia. Researches in Population Ecology 35: 31–43. Ask, K., and P. A. Granhag. 2005. Motivational sources of confirmation bias in criminal investigations: The need for cognitive closure. Journal of Investigative Psychology and Offender Profiling 2: 43–63. Azar, B. 2002. At the frontier of science [Electronic version]. Monitor on Psychology 33 (1) from http://www.apa.org/monitor/jan02/frontier. html. Last consulted: September 6, 2008. Bacon, F. 1620. Novum Organum. Excerpted in S. Hampshire, The Age of Reason: The Seventeenth Century Philosophers. New York: George Braziller, 1958, pp. 23–30. Baconguis, S. R., D. M. Cabahug, and S. N. Alonzo-Pasicolon. 1990. Identification and inventory of Philippine forested-wetland resource. Forest Ecology and Management 33/34: 21–44.

Bibliography / 241

Baker, V. R. 1996. The pragmatic roots of American Quaternary geology and geomorphology. Geomorphology 16: 197–215. Bappenas (National Development Planning Agency of Indonesia). 1998. Fire and Drought Management Issues. Planning for Fire Prevention and Drought Management Project, Working Paper 1. Jakarta: Bappenas. Barber, C. V., and J. Schweithelm. 2000. Trial by Fire: Forest Fires and Forestry Policy in Indonesia’s Era of Crisis and Reform. Washington, DC: World Resources Institute in collaboration with World Wide Fund for Nature—Indonesia Programme and Telapak Indonesia Foundation. Bart, D. J. 2008. Looking for cause with all the small changes: Using event ecology to find human causes of biological invasions. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 99–115. Barth, F. 1959. Political Leadership among Swat Pathans. London: Athlone Press. Barth, F. 1981. Process and Form in Social Life. London: Routledge & Kegan Paul. Barth, F. 1987. Cosmologies in the Making: A Generative Approach to Cultural Variation in Inner New Guinea. Cambridge: Cambridge University Press. Barth, F. 1994. A personal view of present tasks and priorities in cultural and social anthropology. In R. Borofsky, ed., Assessing Cultural Anthropology. New York: McGraw-Hill, pp. 349–361. Bassett, T. J. 1988. The political ecology of peasant-herder conflicts in the northern Ivory Coast. Annals of the Association of American Geographers 78: 453–472. Bateson G. 1963. The role of somatic change in evolution. Evolution 17: 529–539. Beatty, J. 1993. Fitness: Theoretical contexts. In E. G. Keller and E.A. Lloyd, eds., Keywords in Evolutionary Biology. Cambridge: Harvard University Press, pp. 115–119. Becht, C. J. G. 1939. Rattenplaag-Bezwering in Goenoengkidoel. Djawa 19: 258–259. Beckerman, S. 1983. Carpe diem: An optimal foraging approach to Bari fishing and hunting. In R. B. Hames and W. T. Vickers, eds., Adaptive Responses of Native Amazonians. New York: Academic Press, pp. 269–299.

242 / Bibliography

Bell, R. M. 1985. Holy Anorexia. Chicago: University of Chicago Press. Benedict, R. 1934. Patterns of Culture. Boston: Houghton Mifflin. Ben-Menahem, Y. 1997. Historical contingency. Ratio 10: 99–107. Bennett, J. W. 1967. Microcosm-macrocosm relationships in North American agrarian society. American Anthropologist 69: 441–454. Bentley, J. W., and K. L. Andrews. 1991. Pests, peasants, and publications: Anthropological and entomological views of an integrated pest management program for small-scale Honduran farmers. Human Organization 50: 113–124. Berkes, F. 1999. Sacred Ecology: Traditional Ecological Knowledge and Resource Management. Philadelphia: Taylor & Francis. Berlinski, D. 1986. Black Mischief: The Mechanics of Modern Science. New York: Morrow. Berndt, R. M. 1964. Warfare in the New Guinea highlands. In J. B. Watson, ed., New Guinea: The Central Highlands. American Anthropologist Special Publication. Menasha: American Anthropological Association, pp. 183–203. Bhandary, M. J., and K. R. Chandrashekar. 2003. Sacred groves of Dakshina Kannada and Udupi districts of Karnataka. Current Science 85: 1655–1656. Bhargava, R. 1992. Individualism in Social Science: Forms and Limits of a Methodology. Oxford: Clarendon Press. Blaisdell, A. P., K. Sawa, K. J. Leising, and M. R. Waldmann. 2006. Causal reasoning in rats. Science 311: 1020–1022. Blurton Jones, N. G. 1990. Three sensible paradigms for research on evolution and human behavior? Ethology and Sociobiology 11: 353–359. Bompard, J. M., and P. Guizol. 1999. Land Management in the Province of South Sumatra, Indonesia. Fanning the Flames: The Institutional Causes of Vegetation Fires. Palembang: Forest Fire Prevention and Control Project, European Union and Ministry of Forestry and Estate Crops. Boon, L., and H. Smit. 1989. Research styles and the reception of sociobiology. Philosophy of the Social Sciences 19: 19–40. Borgerhoff Mulder, M. 1991. Human behavioural ecology. In J. R. Krebs and N. B. Davies, Behavioural Ecology: An Evolutionary Approach, 3rd edition. Oxford: Blackwell, pp. 69–98. Borgerhoff Mulder, M. 1996. Responses to environmental novelty: Changes in men’s marriage strategies in a rural Kenyan community. In W. G. Runciman, J. Maynard Smith, and R. I. M. Dunbar, eds., Evolution of Social Behaviour Patterns in Primates and Man. Proceedings of the British Academy 88. Oxford: Oxford University Press for the British Academy, pp. 203–222.

Bibliography / 243

Borgerhoff Mulder, M., P. J . Richerson, N. W. Thornhill, and E. Voland. 1997. The place of behavioral ecological anthropology in evolutionary social science. In P. Weingart, S. D. Mitchell, P. J. Richerson, and S. Maasen, eds., Human by Nature: Between Biology and the Social Sciences. Mahwah, NJ: Erlbaum, pp. 253–282. Bowen, M. R., J. M. Bompard, I. P. Anderson, P. Guizol, and A. Gouyon. 2001. Anthropogenic fires in Indonesia: A view from Sumatra. In P. Eaton and M. Radojevic, eds., Forest Fires and Regional Haze in Southeast Asia. Huntington: Nova Science, pp. 41–66. Boyd, R., and P. J. Richerson. 1985. Culture and the Evolutionary Process. Chicago: University of Chicago Press. Boyd, R., and P. J. Richerson. 1992. How microevolutionary processes give rise to history. In M. H. Nitecki and D. V. Nitecki, eds., History and Evolution. Albany: State University of New York Press, pp. 179–209. Brandon, R. N. 1990. Adaptation and Environment. Princeton: Princeton University Press. Brandon, R. N., and M. D. Rausher. 1996. Testing adaptationism: A comment on Orzack and Sober. American Naturalist 148: 189–201. Bratman, M. E. 1987. Intention, Plans, and Practical Reason. Cambridge: Harvard University Press. Bratman, M. E. 1992. Practical reasoning and acceptance in a context. Mind 101: 1–15. Braudel, F. 1980. On History. Translated by S. Matthews. Chicago: University of Chicago Press. Braun, D. 1995. Causally relevant properties. Philosophical Perspectives 9: 447–475. Brechin, S. R., P. R. Wilshusen, C. L. Fortwangler, and P. C. West, eds. 2003. Contested Nature: Promoting International Biodiversity Conservation with Social Justice in the Twenty-First Century. Albany: State University of New York Press. Broad, R., and J. Cavanagh. 1993. Plundering Paradise: The Struggle for the Environment of the Philippines. Berkeley: University of California Press. Brookfield, H. C. 1962. Local study and comparative method: An example from central New Guinea. Annals of the Association of American Geographers 52: 242–254. Brown, W. H., and A. F. Fischer. 1918. Philippine Mangrove Swamps. Bulletin No. 17. Manila: Bureau of Forestry, Department of Agriculture and Natural Resources. Bryant, R. L. 1992. Political ecology: An emerging research agenda in Third-World studies. Political Geography 11: 12–36.

244 / Bibliography

Bryant, R. L., and S. Bailey. 1997. Third World Political Ecology. London: Routledge. Buller, D. J., and V. G. Hardcastle. 2005. Modularity. In D. J. Buller, Adapting Minds: Evolutionary Psychology and the Persistent Quest for Human Nature. Cambridge: MIT Press, pp. 127–200. Bulmer, R. N. H. 1974. Memoirs of a small game hunter: On the track of unknown animal categories in New Guinea. Journal d’Agriculture Tropicale et de Botanique Appliquée 21: 79–99. Bunzl, M. 1993. The Context of Explanation. Dordrecht: Kluwer. Cabahug, D. M., F. M. Ambi, S. O. Nisperos, and N. C. Truzan. 1986. Impact of community-based mangrove forestation to mangrove dependent families and to nearby coastal areas in the Central Visayas: A case study. In Mangroves of Asia and the Pacific: Status and Management. Quezon City: Natural Resources Management Center and National Mangrove Committee, Ministry of Natural Resources, pp. 441–466. Cacioppo, J. T., and G. G. Berntson, eds. 2004. Essays in Social Neuroscience. Cambridge: MIT Press. Caldwell, B. J. 1982. Beyond Positivism: Economic Methodology in the Twentieth Century. London: Allen & Unwin. Calumpong, H. P. 1994. Status of mangrove resources in the Philippines. In C. Wilkinson, S. Sudara, and C. L. Ming, eds., Proceedings of the Third ASEAN-Australia Symposium on Living Coastal Resources, Vol. 1. Townsville: Australian Agency for International Development (AUSAID) and Australian Institute of Marine Science, pp. 215–229. Calumpong, H. P., and J. A. Luchavez. 1997. Profile of the Bais Bay Basin. Silliman Journal 37(3/4): 1–27. Caprettini, G. P. 1983. Peirce, Holmes, Popper. In U. Eco and T. A. Sebeok, eds., The Sign of Three: Dupin, Holmes, Peirce. Bloomington: Indiana University Press, pp. 135–153. Carlyle, T. 1899. Critical and Miscellaneous Essays, Vol. 2. London: Chapman and Hall. Caro, T. M., and M. Borgerhoff Mulder. 1987. The problem of adaptation in the study of human behavior. Ethology and Sociobiology 8: 61–72. Cartwright, N. 1989. Nature’s Capacities and Their Measurement. Oxford: Clarendon Press. Cartwright, N. 1999. The Dappled World: A Study of the Boundaries of Science. Cambridge: Cambridge University Press. Cartwright, N. 2007. Hunting Causes and Using Them: Approaches in Philosophy and Economics. Cambridge: Cambridge University Press. Casati, R., and A. C. Varzi, eds. 1996. Events. Aldershot: Dartmouth Publishing.

Bibliography / 245

Casati, R., and A. C. Varzi. 1997. 50 Years of Events: An Annotated Bibliography 1947 to 1997. Bowling Green: Philosophy Documentation Center, Bowling Green State University. Casati, R., and A. C. Varzi. 2006. Events. The Stanford Encyclopedia of Philosophy (Summer 2006 Edition) from http://plato.stanford.edu/ archives/sum2006/entries/events/. Last consulted: August 13, 2008. Casson, A., and K. Obidzinski. 2002. From new order to regional autonomy: Shifting dynamics of “illegal” logging in Kalimantan, Indonesia. World Development 30: 2133–2151. Caughley, G. 1970. Eruption of ungulate populations, with an emphasis on Himalayan thar in New Zealand. Ecology 51: 53–72. Chagnon, N. A. 1988. Life histories, blood revenge, and warfare in a tribal population. Science 239: 985–992. Chagnon, N. A. 1989. Response to Ferguson. American Ethnologist 16: 565–570. Chagnon, N. A., and W. G. Irons, eds. 1979. Evolutionary Biology and Human Social Behavior: An Anthropological Perspective. North Scituate: Duxbury. Chamberlin, T. C. 1890. The method of multiple working hypotheses. Science (old series) 15: 92–96 (reprinted in Science, 1965, Vol. 148, pp. 754–759). Chambers, R. 1991. Shortcut and participatory methods for gathering social information for projects. In M. Cernea, ed., Putting People First: Sociological Variables in Rural Development, 2nd edition. New York: Oxford University Press, pp. 515–537. Chancellor, T. C. B., E. R. Tiongco, J. Holt, S. Villareal, and P. S. Teng. 1999. The influence of varietal resistance and synchrony on tungro incidence in irrigated rice ecosystems in the Philippines. In T. C. B. Chancellor, O. Azzam, and K. L. Heong, eds., Rice Tungro Disease Management. Makati City: International Rice Research Institute, pp. 121–127. Chew, A. 2006. Opinion: Few answers to the burning question. New Straits Times, Nov. 23. Chokkalingam, U., I. Kurniawan, and Y. Ruchiat. 2005. Fire, livelihoods, and environmental change in the Middle Mahakam peatlands, East Kalimantan. Ecology and Society 10: 26–42. Chokkalingam, U., S. Suyanto, R. P. Permana, I. Kurniawan, J. Mannes, A. Darmawan, N. Khususyiah, and R. H. Susanto. 2007. Community fire use, resource change and livelihood impacts: The downward spiral in the wetlands of southern Sumatra. Mitigation and Adaptation Strategies for Global Change 12: 75–100.

246 / Bibliography

Cleland, C.E. 2001. Historical science, experimental science, and the scientific method. Geology 29: 987–990. Cleland, C. E. 2002. Methodological and epistemic differences between historical science and experimental science. Philosophy of Science 69: 474–496. Cocheo, S. 1997. The bank robber, THE QUOTE, and the final irony. ABA Banking Journal, March from http://www.banking.com/ABA/ profile_0397.htm. Last consulted: August 13, 2008. Cochrane, M. A. 2001, 2002, 2005. Personal communications. Cochrane, M. A. 2002. Spreading like Wildfire: Tropical Forest Fires in Latin America and the Caribbean: Prevention, Assessment and Early Warning. Mexico City: Regional Office for Latin America and the Caribbean, United Nations Environment Programme (UNEP). Cochrane, M. A. 2003. Fire science for rainforests. Nature 421: 913–919. Cochrane, M. A., A. Alencar, M. D. Schulze, C. M. J. Souza, D. C. Nepstad, P. Lefebvre, and E. A. Davidson. 1999. Positive feedbacks in the fire dynamic of closed canopy tropical forests. Science 284: 1832–1835. Cochrane, M. A., and W. F. Laurance. 2002. Fire as a large-scale edge effect in Amazonian forests. Journal of Tropical Ecology 18: 311–325. Cochrane, M. A., and M. D. Schulze. 1999. Fire as a recurrent event in tropical forests of the eastern Amazon: Effects on forest structure, biomass, and species composition. Biotropica 31: 2–16. Cochrane, M. A., D. L. Skole, E. A. T. Matricardi, C. Barber, and W. Chomentowski. 2004. Selective logging, forest fragmentation, and fire disturbance: Implications of interaction and synergy. In D. J. Zarin, J. R. R. Alavalapati, F. E. Putz, and M. Schmink, eds., Working Forests in the Tropics: Conservation through Sustainable Management? New York: Columbia University Press, pp. 310–324. Coello Hinojosa, F. 1992. The Cuyabeno Wildlife Production Reserve: Human needs and natural resource conservation in the Ecuadorean Amazon. In K. Redford and C. Padoch, eds., Conservation of Neotropical Forests: Working from Traditional Resource Use. New York: Columbia University Press, pp. 245–258. Cohen, J. 1994. The earth is round (p < .05). American Psychologist 49: 997–1003. Cohen, M., with B. Dolven and M. Hiebert. 1998. Yes, again: And this time Indonesia’s fires are harder to stop. Far Eastern Economic Review, March 19, pp. 22–23. Colfer, C. J. P. 2002. Ten propositions to explain Kalimantan’s fires— A view from the field. In C. J. P. Colfer and I. A. P. Resosudarmo,

Bibliography / 247

eds., Which Way Forward? Forests, Policy and People in Indonesia. Washington, DC, and Bogor: Resources for the Future, An RFF Book, pp. 309–324. Collingwood, R. G. 1939. An Autobiography. Oxford: Oxford University Press. Comaroff, J. L. 1982. Dialectical systems, history and anthropology. Units of study and questions of theory. Journal of Southern African Studies 8: 143–172. Comaroff, J. L. 1984. The closed society and its critics: Historical transformations in African ethnography. American Ethnologist 11: 571–83. Conklin, B. A., and L. R. Graham. 1995. The shifting middle ground: Amazonian Indians and eco-politics. American Anthropologist 97: 695–710. Conklin, H. C. 1957. Hanunóo Agriculture in the Philippines. FAO Forestry Development Paper No. 12. Rome: Food and Agriculture Organization of the United Nations. Conroy, M. J., P. Beier, H. Quigley, and M. R. Vaughan. 2006. Improving the use of science in conservation: Lessons from the Florida panther. Journal of Wildlife Management 70 (1): 1–7. Cosmides, L., and J. Tooby. 1987. From evolution to behavior: Evolutionary psychology as the missing link. In J. Dupré, ed., The Latest on the Best: Essays on Evolution and Optimality. Cambridge: MIT Press, pp. 277–306. Cosmides, L., and J. Tooby. n.d. Are Humans Good Intuitive Statisticians after All? Rethinking Some Conclusions from the Literature on Judgment under Uncertainty. Preprint Series of the Research Group on Biological Foundations of Human Culture (1991/92) at the Center for Interdisciplinary Research, No. 3/92. Bielefeld: University of Bielefeld. Cosmides, L., and J. Tooby. 1992. Cognitive adaptations for social exchange. In J. H. Barkow, L. Cosmides, and J. Tooby, eds., The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press, pp. 163–228. Cosmides, L., J. Tooby, and J. H. Barkow. 1992. Introduction: Evolutionary psychology and conceptual integration. In J. H. Barkow, L. Cosmides, and J. Tooby, eds., The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press, pp. 3–15. Cosmides, L., and J. Tooby. 1996. Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment under uncertainty. Cognition 58: 1–73. Couzin, J., and J. Kaiser. 2007. Closing the net on common disease genes. Science 316: 820–822.

248 / Bibliography

Cowgill, G. L. 1975. On causes and consequences of ancient and modern population changes. American Anthropologist 77: 505–525. Cramb, R. A. 1986. The evolution of Iban land tenure. Monash University Department of Economics Working Papers 39. Cramb, R. A. 1989. Explaining variations in Bornean land tenure. The Iban case. Ethnology 28: 277–300. Crane, T. 1995. Causation. In A. C. Grayling, ed., Philosophy: A Guide Through the Subject. Oxford: Oxford University Press, pp. 184–194. Cronk, L. 1991. Human behavioral ecology. Annual Review of Anthropology 20: 25–53. Croskerry, P. 2002. Achieving quality in clinical decision making: Cognitive strategies and detection of bias. Academic Emergency Medicine 9: 1184–1204. Croskerry, P. 2003. The importance of cognitive errors in diagnosis and strategies to minimize them. Academic Medicine 78: 775–780. Cummins, R. 2000. “How does it work?” versus “What are the laws?”: Two conceptions of psychological explanation. In F. C. Keil and R. A. Wilson, eds., Explanation and Cognition. Cambridge: MIT Press, pp. 117–144. Curio, E. B. 1973. Towards a methodology of teleonomy. Experientia 29: 1045–1058. Darwin, C. 1876. The Origin of Species by Means of Natural Selection, 6th edition. London: Murray. Darwin, F., ed. 1903. More Letters of Charles Darwin, Vol. 1. London: Murray. Dauvergne, P. 1998. The political economy of Indonesia’s 1997 forest fires. Australian Journal of International Affairs 52: 13–17. Davidson, D. 1980. Essays on Actions and Events. Oxford: Clarendon Press. Davies, P. S. 2001. Norms of Nature: Naturalism and the Nature of Functions. Cambridge: MIT Press. DeHaan, J. D. 2002. Kirk’s Fire Investigation. Upper Saddle River, NJ: Prentice Hall. de Jong, W., L. Tuck-Po, and A. Ken-Ichi. 2003. The political ecology of tropical forests in Southeast Asia: Historical roots of modern problems. In L. Tuck-Po, W. de Jong, and A. Ken-ichi, eds., The Political Ecology of Tropical Forests in Southeast Asia: Historical Perspectives. Kyoto: Kyoto University Press, pp. 1–28. De Leon, R. O. D., A. C. Alcala, and R. J. Raymundo. 1991. Potential levels of primary production and community structure of the Talabong Mangrove Forest, Negros Oriental, Philippines. In A. C.

Bibliography / 249

Alcala, ed., Proceedings of the Regional Symposium on Living Resources in Coastal Areas. Quezon City: Marine Science Institute, University of the Philippines, pp. 453–458. de Marchi, N. 1988a. Introduction. In N. De Marchi, ed., The Popperian Legacy in Economics. Cambridge: Cambridge University Press, pp. 1–15. de Marchi, N. 1988b. Popper and the LSE economists. In N. De Marchi, ed., The Popperian Legacy in Economics. Cambridge: Cambridge University Press, pp. 139–166. Denissenko, M. F., A. Pao, M.-s. Tang, and G. P. Pfeifer. 1996. Preferential formation of benzo[a]pyrene adducts at lung cancer mutational hotspots in P53. Science 274: 430–432. Dennis, R. A., J. Mayer, G. Applegate, U. Chokkalingam, C. J. P. Colfer, I. Kurniawan, H. Lachowski, P. Maus, R. P. Permana, Y. Ruchiat, F. Stolle, S. Suyanto, and T. P. Tomich. 2005. Fire, people and pixels: Linking social science and remote sensing to understand underlying causes and impacts of fires in Indonesia. Human Ecology 33: 465–504. DENR (Department of Environment and Natural Resources). 1990. Compilation of Mangrove Regulations. Quezon City: Coastal Resources Management Committee, Republic of the Philippines. DENR (Department of Environment and Natural Resources). 1993. Establishing the Coastal Environment Program (CEP) and Providing Funds Thereof. Administration Order No. 93–19. Quezon City: Department of Environment and Natural Resources. DENR (Department of Environment and Natural Resources). 1994. Mangrove Regeneration and Management. Quezon City: Mangrove Technical Review Committee, Fisheries Sector Program, Republic of the Philippines. DENR (Department of Environment and Natural Resources). 1996. Rules and Regulations for the Implementation of Executive Order 263, Otherwise Known as the Community-based Forest Management Strategy (CBFMS). Administration Order No 96-29. Quezon City: Department of Environment and Natural Resources. Dethlefsen, L., M. McFall-Ngai, and D. A. Relman. 2007. An ecological and evolutionary perspective on human–microbe mutualism and disease. Nature 449: 811–818. DeWalt, B. R., and P. J. Pelto, eds. 1985. Micro and Macro Levels of Analysis in Anthropology. Boulder, CO: Westview Press. DeWalt, B. R., P. Vergne, and M. Hardin. 1996. Shrimp aquaculture development and the environment: People, mangroves and fisheries on the Gulf of Fonseca, Honduras. World Development 24: 1193–1208.

250 / Bibliography

Diop, E. S., ed. 1993. Conservation and Sustainable Utilization of Mangrove Forests in Latin America and African Regions, Part 2: Africa. Mangrove Ecosystem Technical Reports 3. Okinawa: International Society for Mangrove Ecosystems and International Tropical Timber Organization. Djilas, M. 1962. Conversations with Stalin. New York: Harcourt, Brace and World. Dove, M. R., and T. Nugroho. 1994. Review of “Culture and Conservation” 1991–1994: A Sub-Project, Funded by the Ford Foundation, of the World Wide Fund for Nature’s Kayan Mentarang Nature Reserve Project in Kalimantan, Indonesia. December 31. Typescript. Dray, W. 1957. Laws and Explanation in History. Oxford: Clarendon Press. Dray, W. 1980. Perspectives on History. London: Routledge & Kegan Paul. Driscoll, C., and S. Stich. 2008. Vayda blues: Explanation in Darwinian ecological anthropology. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 175–191. Dudenhoeffer, D. 1990. Out-of-date data sparks doubts. Tico Times, August 17, p. 2. Durham, W. H. 1995. Political ecology and environmental destruction in Latin America. In M. Painter and W. H. Durham, eds., The Social Causes of Environmental Destruction in Latin America. Ann Arbor: University of Michigan Press, pp. 249–264. Edelman, M. 1995. Rethinking the hamburger thesis: Deforestation and the crisis of Central America’s beef exports. In M. Painter and W. H. Durham, eds., The Social Causes of Environmental Destruction in Latin America. Ann Arbor: University of Michigan Press, pp. 25–62. Eghenter, C. 2006. Equitable distribution of costs and benefits: Outlining a new framework for equitable alternatives in conservation. Unpublished working paper for the Learning & Action Network (LAN) of the World Wide Fund for Nature (WWF)—Asia. Eghenter, C. 2008. What kind of anthropology for successful conservation management and development? In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 195–205. Einhorn, H. J., and R. M. Hogarth. 1987. Decision making: Going forward in reverse. Harvard Business Review 65: 66–70.

Bibliography / 251

Ellen, R. 1998. Comment [on Sillitoe 1998a]. Current Anthropology 39: 238–239. Ellis, J. M. 1997. Literature Lost: Social Agendas and the Corruption of the Humanities. New Haven: Yale University Press. Elster, J. 1980. Cohen on Marx’s theory of history. Political Studies 28: 121–128. Elster, J. 1983a. Explaining Technical Change. Cambridge: Cambridge University Press. Elster, J. 1983b. Sour Grapes: Studies in the Subversion of Rationality. Cambridge: Cambridge University Press. Elster, J. 1985. Making Sense of Marx. Cambridge: Cambridge University Press. Elster, J. 1986a. Further thoughts on Marxism, functionalism, and game theory. In J. Roemer, ed., Analytical Marxism. Cambridge: Cambridge University Press, pp. 202–220. Elster, J. 1986b. An Introduction to Karl Marx. Cambridge: Cambridge University Press. Elster, J. 1989a. The Cement of Society. Cambridge: Cambridge University Press. Elster, J. 1989b. Nuts and Bolts for the Social Sciences. Cambridge: Cambridge University Press. Elster, J. 1993. Political Psychology. Cambridge: Cambridge University Press. Elton, C. 1927. Animal Ecology. New York: Macmillan. Ember, M. 1982. Statistical evidence for an ecological explanation of warfare. American Anthropologist 84: 645–649. Emlen, J. M. 1987. Evolutionary ecology and the optimality assumption. In J. Dupré, ed., The Latest on the Best: Essays on Evolution and Optimality. Cambridge: MIT Press, pp. 163–177. Emmet, D. 1985. The Effectiveness of Causes. Albany: State University of New York Press. ENCODE Project Consortium. 2007. Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature 447: 799–816. Ereshefsky, M. 1992. The historical nature of evolutionary theory. In M. H. Nitecki and D. V. Nitecki, eds., History and Evolution. Albany: State University of New York Press, pp. 81–99. Evans, R. 1999. A cause to believe in? [review of McKim and Turner 1997]. Social Studies of Science 29: 145–150. Evans-Pritchard, E. E. 1937. Witchcraft, Oracles and Magic among the Azande. Oxford: Clarendon Press.

252 / Bibliography

Fairhead, J., and M. Leach. 1996. Misreading the African Landscape: Society and Ecology in the Forest Savanna Mosaic. Cambridge: Cambridge University Press. Fairhead, J., and M. Leach. 1998. Reframing Deforestation: Global Analyses and Local Realities: Studies in West Africa. London: Routledge. Falvo, D. J. 2000. On modeling Balinese water temple networks as complex adaptive systems. Human Ecology 28: 641–649. Fearon, J. D. 1996. Causes and counterfactuals in social science: Exploring an analogy between cellular automata and historical processes. In P. E. Tetlock and A. Belkin, eds., Counterfactual Thought Experiments in World Politics: Logical, Methodological, and Psychological Perspectives. Princeton: Princeton University Press, pp. 39–67. Feeny, D., F. Berkes, B. J. McCay, and J. M. Acheson. 1990. The tragedy of the commons: Twenty-two years later. Human Ecology 18: 1–19. Feil, D. K. 1977. Review of Mervyn Meggitt’s Blood Is Their Argument. Mankind 11: 156–157. Feil, D. K. 1987. The Evolution of Highland Papua New Guinea Societies. Cambridge: Cambridge University Press. Feng, Z., W. Hu, Y. Hu, and M.-s. Tang. 2006. Acrolein is a major cigarette-related lung cancer agent: Preferential binding at p53 mutational hotspots and inhibition of DNA repair. Proceedings of the National Academy of Sciences 103: 15404–15409. Ferguson, R. B. 1984. Introduction: Studying war. In R.B. Ferguson, ed., Warfare, Culture, and Environment. Orlando: Academic Press, pp. 1–81. Finney, B. 1991. Myth, experiment, and the reinvention of Polynesian voyaging. American Anthropologist 93: 383–404. Flasse, S. P., and P. Ceccato. 1996. A contextual algorithm for AVHRR fire detection. International Journal of Remote Sensing 17: 419–424. Fleck, L. 1979. Genesis and Development of a Scientific Fact, translated by F. Bradley and T. J. Trenn. Chicago: University of Chicago Press. Fodor, J. 1996. Deconstructing Dennett’s Darwin. Mind & Language 11: 246–262. Fodor, J. 2000. The Mind Doesn’t Work That Way: The Scope and Limits of Computational Psychology. Cambridge: MIT Press. Fogarty, L., M. Santoso, and S. Widodo. 2001. Fire Detection Options for PT Inhutani I. Berau Forest Management Project. Jakarta: European Union and Ministry of Forestry and Estate Crops. Food and Agriculture Organization of the United Nations (FAO). 1994. Mangrove Forest Management Guidelines. FAO Forestry Paper 117. Rome: Food and Agriculture Organization of the United Nations.

Bibliography / 253

Ford, E. B. 1975. Ecological Genetics, 4th edition. London: Chapman and Hall. Ford, R. T. 1987. Investigation of Vegetation Fires. Fresno: Fire Scene Investigations. Ford, R. T. 1995. Investigation of Wildfires. Sunriver: R.T. Ford. Forsyth, T. 2003. Critical Political Ecology: The Politics of Environmental Science. London: Routledge. Freedman, D. A. 1991. Statistical models and shoe leather. In P. V. Marsden, ed., Sociological Methodology, Vol. 21, pp. 291–313. Oxford: Basil Blackwell for the American Sociological Association. Freedman, D. A. 1997. From association to causation via regression. In V. R. McKim and S. P. Turner, eds., Causality in Crisis? Statistical Methods and the Search for Causal Knowledge in the Social Sciences. Notre Dame: University of Notre Dame Press, pp. 113–161. Freedman, D. A. 1999. From association to causation: Some remarks on the history of statistics. Statistical Science 14: 243–258. Freedman, D. A. 2005a. Linear statistical models for causation: A critical review. In B. Everitt and D. Howell, eds., Encyclopedia of Statistics in Behavioral Science, Vol. 2. Chichester: Wiley. Freedman, D. A. 2005b. Statistical Models: Theory and Practice. Cambridge: Cambridge University Press. Gallagher, K. n.d. Training Model of the Farmers’ Field Guide. Jakarta and Yogyakarta: Indonesian National Integrated Pest Management Programme. Gallagher, K. D., P. E. Kenmore, and K. Sogawa. 1994. Judicial use of insecticides deters planthopper outbreaks and extends the life of resistant varieties in Southeast Asian rice. In R.F. Denno and T.J. Perfect, eds., Planthoppers: Their Ecology and Management. London: Chapman & Hall, pp. 599–614. Gärdenfors, P. 1980. A pragmatic approach to explanations. Philosophy of Science 47: 404–423. Gärdenfors, P. 1990. An epistemic analysis of explanations and causal beliefs. Topoi 9: 109–124. Gardner, A. 2003. The new calculus of Bedouin pastoralism in the kingdom of Saudi Arabia. Human Organization 62: 267–276. Gardner, A. 2005. The new calculus of Bedouin pastoralism in the kingdom of Saudi Arabia. In S. Paulson and L. L. Gezon, eds., Political Ecology across Spaces, Scales, and Social Groups. New Brunswick, NJ: Rutgers University Press, pp. 76–93. Garfinkel, A. 1981. Forms of Explanation: Rethinking the Questions in Social Theory. New Haven: Yale University Press.

254 / Bibliography

Gatewood, J. B. 1985. Actions speak louder than words. In J. W. D. Dougherty, ed., Directions in Cognitive Anthropology. Urbana: University of Illinois Press, pp. 199–219. Geertz, C. 1973. The Interpretation of Cultures. New York: Basic Books. Geertz, C. 1983. Local Knowledge: Further Essays in Interpretive Anthropology. New York: Basic Books. Geertz, C. 1984. Culture and social change: The Indonesian case. Man 19: 511–532. Gellner, E. 1958. Time and theory in social anthropology. Mind 67: 182–202. Gerwing, J. J., J. S. Johns, and E. Vidal. 1996. Reducing waste during logging and log processing: Forest conservation in eastern Amazonia. Unasylva 47: 17–25. Gezon, L. L. 1997. Political ecology and conflict in Ankarana, Madagascar. Ethnology 36: 85–100. Gibbs, C., E. Payuan, and R. del Castillo. 1990. The growth of the Philippine social forestry program. In M. Poffenberger, ed., Keepers of the Forest: Land Management Alternatives in Southeast Asia. West Hartford: Kumarian Press, pp. 253–265. Gigerenzer, G. 1991. How to make cognitive illusions disappear: Beyond “heuristics and biases.” European Review of Social Psychology 2: 83–115. Gigerenzer, G. 1992. Cognitive illusions illusory? Rethinking judgment under uncertainty. Preprint Series of the Research Group on Biological Foundations of Human Culture (No. 25/92, 1991/92). Bielefeld: Center for Interdisciplinary Research, University of Bielefeld, Germany. Gigerenzer, G. 2004. Mindless statistics. Journal of Socio-Economics 33: 587–606. Gill, S. R., et al. (9 others). 2006. Metagenomic analysis of the human distal gut microbiome. Science 312: 1355–1359. Gilovich, T., D. Griffin, and D. Kahneman, eds. 2002. Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge: Cambridge University Press. Gintis, H. M., S. Bowles, R. T. Boyd, and E. Fehr, eds. 2005. Moral Sentiments and Material Interests: The Foundations of Cooperation in Economic Life. Cambridge: MIT Press. Gluckman, M. 1955. The Judicial Process among the Barotse. Manchester: Manchester University Press. Godfrey-Smith, P. 2003. Theory and Reality: An Introduction to the Philosophy of Science. Chicago: University of Chicago Press.

Bibliography / 255

Goldammer, J. G., P. G. H. Frost, M. Jurvélius, E. M. Kamminga, and T. Kruger. 2004. Community participation in integrated forest fire management: Some experiences from Africa (Annex I). In J. G. Goldammer and C. de Ronde, eds., Wildland Fire Management Handbook for Sub-Sahara Africa. Freiburg, Germany: Global Fire Monitoring Center (GFMC), pp. 382–402. Gong, K. W., et al. (9 others). 2007. Air pollutant chemicals and oxidized lipids exhibit genome wide synergistic effects on endothelial cells. Genome Biology 8 (7), published online, July 26, from http://genome biology.com/2007/8/7/R149. Last consulted: July 19, 2008. Gönner, C. 2004. Personal communication. Goodenough, W. H. 1957. Cultural anthropology and linguistics. In P. L. Garvin, ed., Report of the Seventh Annual Round Table Meeting on Linguistics and Language Study. Washington, DC: Georgetown University Press, pp. 167–177. Goodenough, W. H. 1994. Toward a working theory of culture. In R. Borofsky, ed., Assessing Cultural Anthropology. New York: McGrawHill, pp. 262–275. Gordon, R. J., and M. J. Meggitt. 1985. Law and Order in the New Guinea Highlands: Encounters with Enga. Hanover: University Press of New England. Gould, S. J. 1979. Darwin’s middle road. Natural History 88 (10): 27–31. Gould, S. J. 1980. Sociobiology and the theory of natural selection. In G. W. Barlow and J. Silverberg, eds., Sociobiology: Beyond Nature/Nurture. Boulder, CO: Westview, pp. 57–69. Gould, S. J. 1986. Evolution and the triumph of homology, or why history matters. American Scientist 74: 60–69. Gould, S. J. 1987. An Urchin in the Storm: Essays about Books and Ideas. New York: Norton. Gould, S. J. 1988. Mighty Manchester. New York Review of Books, October 27, pp. 32–35. Gould, S. J., and R. C. Lewontin. 1979. The spandrels of San Marco and the panglossian paradigm: A critique of the adaptationist programme. Proceedings of the Royal Society of London, Series B 205: 581–598. Graber, M., R. Gordon, and N. Franklin 2002. Reducing diagnostic errors in medicine: What’s the goal? Academic Medicine 77: 981–992. Grafen, A. 1988. On the uses of data on lifetime reproductive success. In T. H. Clutton-Brock, ed., Reproductive Success: Studies of Individual Variation in Contrasting Breeding Systems. Chicago: University of Chicago Press, pp. 454–519.

256 / Bibliography

Granovetter, M. 1978. Threshold models of collective behavior. American Journal of Sociology 83: 1420–1443. Gray, R. D. 1987. Faith and foraging: A critique of the “paradigm argument from design.” In A. C. Kamil, J. R. Krebs, and H. R. Pulliam, eds., Foraging Behavior. New York: Plenum, pp. 69–140. Gray, R. D. 1988. Metaphors and methods: Behavioural ecology, panbiogeography and the evolving synthesis. In M.-W. Ho and S. W. Fox, eds., Evolutionary Processes and Metaphors. Chichester: Wiley, pp. 209–242. Greally, J. M. 2007. Encyclopaedia of humble DNA. Nature 447: 782– 783. Greenough, P. R., and A. L. Tsing, eds. 2003. Nature in the Global South: Environmental Projects in South and Southeast Asia. Durham, NC: Duke University Press. Greenwald, A. G., A. R. Pratkanis, M. R. Leippe, and M. H. Baumgardner. 1986. Under what conditions does theory obstruct research progress? Psychological Review 93: 216–229. Greenwood, J. D. 1990. The social constitution of action: Objectivity and explanation. Philosophy of the Social Sciences 20: 195–207. Griffin, L. J. 1993. Narrative, event-structure analysis, and causal interpretation in historical sociology. American Journal of Sociology 98: 1094–1133. Griffiths, P. E. 1996. The historical turn in the study of adaptation. British Journal for the Philosophy of Science 47: 511–532. Griffiths. P. E. 1999. Adaptation and adaptationism. In R. A. Wilson and F. C. Keil, eds., The MIT Encyclopedia of the Cognitive Sciences. Cambridge: MIT Press, pp. 3–4. Grist, D. H. 1986. Rice, 6th edition. London: Longman. Groopman, J. 2007a. What’s the trouble? How doctors think. New Yorker, January 29, pp. 36–41. Groopman, J. 2007b. How Doctors Think. Boston: Houghton Mifflin. Gruner, R. 1969. The notion of an historical event, I. Aristotelian Society Supplementary Volume 43: 141–152. Guelke, L. 1982. Historical Understanding in Geography. Cambridge: Cambridge University Press. Gurven, M., and J. Winking. 2008. Collective action in action: Prosocial behavior in and out of the laboratory. American Anthropologist 110: 179–190. Guyon, A., and D. Simorangkir. 2002. The Economics of Fire Use in Agriculture and Forestry: A Preliminary Review for Indonesia. Jakarta: Project FireFight South East Asia. Haaland, G. 2008. Explaining causes in evolving contexts: From Nepali hill farmers to business managers in Thailand. In B. B. Walters, B.

Bibliography / 257

J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 43–66. Hailman, J. P. 1988. Operationalism, optimality and optimism: Suitabilities versus adaptations of organisms. In M.-W. Ho and S. W. Fox, eds., Evolutionary Processes and Metaphors. Chichester: Wiley, pp. 85–116. Hall, T. D. 1986. Incorporation in the world system: Toward a critique. American Sociological Review 51: 390–402. Hallpike, C. R. 1973. Functionalist interpretations of primitive warfare. Man 8: 451–470. Hallpike, C. R. 1977a. Bloodshed and Vengeance in the Papuan Mountains. Oxford: Oxford University Press. Hallpike, C. R. 1977b. Review of Mervyn Meggitt’s Blood Is Their Argument and Andrew P. Vayda’s War in Ecological Perspective. Man 12: 555–557. Hallpike, C. R. 1986. The Principles of Social Evolution. Oxford: Oxford University Press. Hallpike, C. R., 1990. Personal communication [letter dated July 23, 1990]. Hamilton, L. S., J. A. Dixon, and G. O. Miller. 1989. Mangrove forests: An undervalued resource of the land and of the sea. In E. M. Borgese, N. Ginsburg, and J. R. Morgan, eds., Ocean Yearbook 8. Chicago: University of Chicago Press, pp. 254–288. Harford, T. 2006. The Undercover Economist: Exposing Why the Rich Are Rich, the Poor Are Poor—And Why You Can Never Buy A Decent Used Car! New York: Oxford University Press. Harmon-Jones, E., and P. Winkielman, eds. 2007. Social Neuroscience: Integrating Biological and Psychological Explanations of Social Behavior. New York: Guilford Press. Harris, M. 1966. The cultural ecology of India’s sacred cattle. Current Anthropology 7: 51–66. Harris, M. 1968. The Rise of Anthropological Theory. New York: Crowell. Harris, M. 1974. Cows, Pigs, Wars & Witches: The Riddles of Culture. New York: Random House. Harris, M. 1985. Good to Eat: Riddles of Food and Culture. New York: Simon and Schuster. Harris, M. 1987a. Cultural materialism: Alarums and excursions. In K. Moore, ed., Waymarks: The Notre Dame Inaugural Lectures in Anthropology. Notre Dame: University of Notre Dame Press, pp. 107–126. Harris, M. 1987b. Foodways: Historical overview and theoretical prolegomenon. In M. Harris and E. B. Ross, eds., Food and Evolution: Toward a Theory of Human Food Habits. Philadelphia: Temple University Press, pp. 57–90.

258 / Bibliography

Harris, M. 1994. Cultural materialism is alive and well and won’t go away until something better comes along. In R. Borofsky, ed., Assessing Cultural Anthropology. New York: McGraw-Hill, pp. 62–76. Harrison, W. 1958. Political processes. Political Studies 6: 234–252. Harrowitz, N. 1983. The body of the detective model: Charles S. Peirce and Edgar Allan Poe. In U. Eco and T. A. Sebeok, eds., The Sign of Three: Dupin, Holmes, Peirce. Bloomington: Indiana University Press, pp. 179–197. Hart, H. L. A., and T. Honore. 1985. Causation in the Law, 2nd edition. Oxford: Clarendon Press. Hartmann, B. 2001. Will the circle be unbroken? A critique of the project on environment, population, and security. In N. Peluso and M. Watts, eds., Violent Environments. Ithaca, NY: Cornell University Press, pp. 39–62. Harvey, P. H., and M. D. Pagel. 1991. The Comparative Method in Evolutionary Biology. Oxford: Oxford University Press. Harwell, E. E. 2000. Remote sensibilities: Discourses of technology and the making of Indonesia’s natural disaster. Development and Change 31: 307–340. Hawkes, K. 1993. Why hunter-gatherers work: An ancient version of the problem of public goods. Current Anthropology 34: 341–361. Hawkes, K., J. F. O’Connell, and N. G. Blurton Jones. 1989. Hardworking Hadza grandmothers. In V. Standen and R. A. Foley, eds., Comparative Socioecology: The Behavioural Ecology of Humans and Other Mammals. Oxford: Blackwell, pp. 341–366. Hawthorn, G. 1991. Plausible Worlds: Possibility and Understanding in History and the Social Sciences. Cambridge: Cambridge University Press. Hecht, S. S. 2006. Smoking and lung cancer—a new role for an old toxicant? Proceedings of the National Academy of Sciences 103: 15725– 15726. Heinrich, B. 1983. Do bumblebees forage optimally, and does it matter? American Zoologist 23: 273–281. Heinrichs, E. A. 1994. Impact of insecticides on the resistance and resurgence of rice planthoppers. In R. F. Denno and T. J. Perfect, eds., Planthoppers: Their Ecology and Management. London: Chapman & Hall, pp. 571–598. Helmreich, S. 2000. Power/networks: A rejoinder to Lansing. Critique of Anthropology 20: 319–327. Hempel, C. 1942. The function of general laws in history. Journal of Philosophy 39: 35–48.

Bibliography / 259

Hempel, S. 2006. The Medical Detective: John Snow and the Mystery of Cholera. London: Granta Books. Hennekens, C. H., and J. E. Buring. 1987. Epidemiology in Medicine. Boston: Little, Brown. Henrich, J., R. Boyd, S. Bowles, C. Camerer, E. Fehr, and H. Gintis, eds. 2004. Foundations of Human Sociality: Economic Experiments and Ethnographic Evidence from Fifteen Small-Scale Societies. Oxford: Oxford University Press. Hershkovitz, L. 1993. Political ecology and environmental management in the Loess Plateau, China. Human Ecology 21: 327–353. Hess, C. 1997. Hungry for Hope: On the Cultural and Communicative Dimensions of Development in Highland Ecuador. London: Intermediate Technology Publications. Hess, S. 1976. Organizing the Presidency. Washington, DC: Brookings Institution. Hill, K. 1988. Macronutrient modifications of optimal foraging theory: An approach using indifference curves applied to some modern foragers. Human Ecology 16: 157–194. Hill, K., and H. Kaplan. 1988. Tradeoffs in male and female reproductive strategies among the Ache: Part 1. In L. Betzig, M. Borgerhoff Mulder, and P. Turke, eds., Human Reproductive Behavior: A Darwinian Perspective. Cambridge: Cambridge University Press, pp. 277–289. Hilton, D. J. 1995. Logic and language in causal explanation. In D. Sperber, D. Premack, and A. J. Premack, eds., Causal Cognition: A Multidisciplinary Debate. Oxford: Clarendon Press, pp. 495–529. Hodgson, G. 1980. All Things to All Men: The False Promise of the American Presidency. New York: Simon & Schuster. Holdsworth, A. R., and C. Uhl. 1997. Fire in Amazonian selectively logged rain forest and the potential for fire reduction. Ecological Applications 7: 712–725. Holy, L., and M. Stuchlik. 1983. Actions, Norms, and Representations: Foundations of Anthropological Theory. Cambridge: Cambridge University Press. Homer-Dixon, T. 1999. Environment, Scarcity, and Violence. Princeton: Princeton University Press. Homer-Dixon, T. 2003. Debating Violent Environments. ECSP (Environmental Change and Security Program) Report 9, pp. 89–93. Hughes, A. L. 1987. Social and antisocial behavior. Quarterly Review of Biology 62: 415–421. Hull, D. L. 2002. Varieties of reductionism: Derivation and gene selection. In M. H. V. Van Regenmortel and D. L. Hull, eds., Promises and

260 / Bibliography

Limits of Reductionism in the Biomedical Sciences. Chichester: Wiley, pp. 161–177. Hunn, E. 1977. Tzeltal Folk Zoology. New York: Academic Press. Hunn, E. 1989. Ethnoecology: The relevance of cognitive anthropology for human ecology. In M. Freilich, ed., The Relevance of Culture. New York: Bergin & Garvey, pp. 143–160. Hunn, E. 1993. What is traditional ecological knowledge? In N. M. Williams and G. Baines, eds., Traditional Ecological Knowledge: Wisdom for Sustainable Development. Canberra: Centre for Resource and Environmental Studies, Australian National University, pp. 16–20. Huntington, H. P. 2000. Using traditional ecological knowledge in science: Methods and applications. Ecological Applications 10: 1270– 1274. Hutchins, E. 1995. Cognition in the Wild. Cambridge: MIT Press. Irons, W., and L. Cronk. 2000. Two decades of a new paradigm. In L. Cronk, N. Chagnon, and W. Irons, eds., Adaptation and Human Behavior: An Anthropological Perspective. New York: Aldine de Gruyter, pp. 3–26. Irzik, G. 1990. Singular causation and law. PSA 1990 (Proceedings of the 1990 Biennial Meeting of the Philosophy of Science Association) 1: 537–543. IUCN (World Conservation Union) and WWF (World Wide Fund for Nature) 2001. Forest fires position statement. December, 4 pp. Jacobson, D. 1991. Reading Ethnography. Albany: State University of New York Press. Jeffery, R., and N. Sundar, eds. 1999. A New Moral Economy for India’s Forests? Discourses of Community and Participation. New Delhi: Sage. Jochim, M. A. 1981. Strategies for Survival. New York: Academic Press. Johannes, R. E. 1978. Traditional marine conservation methods in Oceania and their demise. Annual Review of Ecology and Systematics 9: 349–364. Johannes, R. E. 1981. Words of the Lagoon: Fishing and Marine Lore in the Palau District of Micronesia. Berkeley: University of California Press. Johnson, A. 1974. Ethnoecology and planting practices in a swidden agricultural system. American Ethnologist 1: 209–223. Johnson, S. 2006. The Ghost Map: The Story of London’s Most Terrifying Epidemic—and How It Changed Science, Cities, and the Modern World. New York: Riverhead Books. Josephson, J. R. 2000. Smart inductive generalizations are abductions. In P. A. Flach and A. C. Kakas, eds., Abduction and Induction: Essays on Their Relation and Integration. Dordrecht: Kluwer, pp. 31–44.

Bibliography / 261

Kahl, C. 2002. Review of Peluso and Watts 2001. ECSP (Environmental Change and Security Program) Report 8, pp. 133–143. Kahneman, D., P. Slovic, and A. Tversky, eds. 1982. Judgment under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press. Kalam, M. A. 1998. Sacred groves in Coorg, Karnataka. In R. Jeffery, ed., The Social Construction of Indian Forests. New Delhi and Edinburgh: Manohar Publishers and Centre for South Asian Studies, pp. 39–54. Kaly, U. L., and G. P. Jones. 1998. Mangrove restoration: A potential tool for coastal management in tropical developing countries. Ambio 27: 656–661. Kartawinata, K., and A. P. Vayda. 1984. Forest conversion in East Kalimantan, Indonesia: The activities and impact of timber companies, shifting cultivators, migrant pepper farmers, and others. In F. di Castri, F. W. G. Baker, and M. Hadley, eds., Ecology in Practice. Dublin: Tycooly, Pt. 1, pp. 98–126. Kauffman, J. B., and C. Uhl. 1990. Interactions of anthropogenic activities, fire, and rain forests in the Amazon Basin. In J. G. Goldammer, ed., Fire in the Tropical Biota: Ecosystem Processes and Global Challenges. Berlin: Springer-Verlag, pp. 117–134. Keller, C., and J. D. Keller. 1993. Thinking and acting with iron. In S. Chaiklin and J. Lave, eds., Understanding Practice: Perspectives on Activity and Context. Cambridge: Cambridge University Press, pp. 125–143. Kenmore, P. E. 1991. Indonesia’s Integrated Pest Management—A Model for Asia. Manila: FAO Rice IPC (Insect Pest Control) Programme. Kenmore, P. E., J. A. Litsinger, J. P. Bandong, A. C. Santiago, and M. M. Salac. 1987. Philippine rice farmers and insecticides: Thirty years of growing dependency and new options for change. In J. Tait and B. Napompeth, eds., Management of Pests and Pesticides: Farmers’ Perceptions and Practices. Boulder, CO: Westview Press, pp. 98–108. Keppens, J., and J. Zeleznikow. 2003. A model based reasoning approach for generating plausible crime scenarios from evidence. Proceedings of the 9th International Conference on Artificial Intelligence and Law, pp. 51–59. Khoon, G. W., and O. J. Eong. 1995. The use of demographic studies in mangrove silviculture. Hydrobiologia 295: 255–261. Kincaid, H. 1986. Reduction, explanation, and individualism. Philosophy of Science 53: 492–513.

262 / Bibliography

Kincaid, H. 1990. Assessing functional explanations in the social sciences. PSA 1990 (Proceedings of the 1990 Biennial Meeting of the Philosophy of Science Association) 1: 341–354. Kincaid, H. 1995. Optimality arguments and the theory of the firm. In D. Little, ed., On the Reliability of Economic Models: Essays in the Philosophy of Economics. Dordrecht: Kluwer, pp. 211–242. Kincaid, H. 1996. Philosophical Foundations of the Social Sciences: Analyzing Controversies in Social Research. Cambridge: Cambridge University Press. Kincaid, H. 2000a. Formal rationality and its pernicious effects on the social sciences. Philosophy of the Social Sciences 30: 67–88. Kincaid, H. 2000b. Global arguments and local realism about the social sciences. Philosophy of Science 67: S667–S678. King, V. T. 1976. Migration, warfare, and culture contact in Borneo: A critique of ecological analysis. Oceania 46: 306–327. Kingsland, S. E. 1985. Modeling Nature: Episodes in the History of Population Ecology. Chicago: University of Chicago Press. Kingsland, S. E. 1995. Modeling Nature: Episodes in the History of Population Ecology, 2nd edition. Chicago: University of Chicago Press. Kiros-Meles, A., and M. M. Abang. 2008. Farmers’ knowledge of crop diseases and control strategies in the Regional State of Tigrai, northern Ethiopia: Implications for farmer-researcher collaboration in disease management. Agriculture and Human Values 25: 433–452. Kitcher, P. 1985a. Darwin’s achievement. In N. Rescher, ed., Reason and Rationality in Natural Science. Lanham, MD: University Press of America, pp. 127–189. Kitcher, P. 1985b. Two approaches to explanation. Journal of Philosophy 82: 632–639. Kitcher, P. 1985c. Vaulting Ambition: Sociobiology and the Quest for Human Nature. Cambridge: MIT Press. Kitcher, P. 1986. The transformation of human sociobiology. PSA 1986 (Proceedings of the 1986 Biennial Meeting of the Philosophy of Science Association) 2: 63–74. Kitcher, P. 1987. Why not the best? In J.Dupré, ed., The Latest on the Best: Essays on Evolution and Optimality. Cambridge: MIT Press, pp. 77–102. Kitcher, P. 1993. The Advancement of Science. New York: Oxford University Press. Koch, K.-F. 1974. The Anthropology of Warfare. Addison-Wesley Module in Anthropology no. 52. Reading, MA: Addison-Wesley.

Bibliography / 263

Kottak, C. P. 1999. The new ecological anthropology. American Anthropologist 101: 23–35. Kreger, J. 1998. Fire and failure. Michigan History Magazine, January/ February, pp. 15–20. Kummer, H. 1995. Causal knowledge in animals. In D. Sperber, D. Premack, and A. J. Premack, eds., Causal Cognition: A Multidisciplinary Debate. Oxford: Clarendon Press, pp. 26–39. Kunstadter, P., E. C. F. Bird, and S. Sabhasri, eds. 1986. Man in the Mangroves. Tokyo: United Nations University. Kuznar, L. A., and K. Long. 2008. Deductive-nomological vs. causalmechanical explanation: Relative strengths and weaknesses in anthropological explanation. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 159–173. Kuznar, L. A., and S. K. Sanderson. 2007. Introduction: The potentials and challenges of cultural materialism. In L. A. Kuznar and S. K. Sanderson, eds., Studying Societies and Cultures: Marvin Harris’s Cultural Materialism and Its Legacy. Boulder, CO: Paradigm Publishers, pp. 1–17. Lacerda, L. D., ed. 1993. Conservation and Sustainable Utilization of Mangrove Forests in Latin America and African Regions (Part 1: Latin America). Mangrove Ecosystem Technical Reports 2. Okinawa: International Society for Mangrove Ecosystems and International Tropical Timber Organization. Lansing, J. S. 1991. Priests and Programmers: Technologies of Power in the Engineered Landscape of Bali. Princeton: Princeton University Press. Lansing, J. S. 2000a. Anti-chaos, common property, and the emergence of cooperation. In T. Kohler and G. J. Gumerman, eds., Dynamics in Human and Primate Societies: Agent-Based Modeling of Social and Spatial Processes. New York: Oxford University Press, pp. 207–223. Lansing, J. S. 2000b. Foucault and the water temples: A reply to Helmreich. Critique of Anthropology 20: 309–318. Lansing, J. S. 2002. “Artificial societies”and the social sciences. Artificial Life 8: 279–292. Lansing, J. S. 2006. Perfect Order: Recognizing Complexity in Bali. Princeton: Princeton University Press. Lansing, J. S., and J. N. Kremer. 1993. Emergent properties of Balinese water temple networks: Coadaptation on a rugged fitness landscape. American Anthropologist 95: 97–114. Lansing, J. S., and J. N. Kremer. 1996. The goddess and the Green Revolution. Encyclopedia Brittanica Yearbook of Science and Technology and the Future, Chicago, pp. 196–219.

264 / Bibliography

Lansing, J. S., J. Schoenfelder, and V. Scarborough. 2006. Rappaport’s rose: Structure, agency, and historical contingency in ecological anthropology. In A. Biersack and J. B. Greenberg, eds., Reimagining Political Ecology. Durham, NC: Duke University Press, pp. 325–357. Last, M. 1981. The importance of knowing about not knowing. Social Science and Medicine 15B: 387–392. Latour, B. 1988. The politics of explanation: An alternative. In S. Woolgar, ed., Knowledge and Reflexivity: New Frontiers in the Sociology of Knowledge. London: Sage, pp. 155–176. Lauder, G. V., A. M. Leroi, and M. R. Rose. 1993. Adaptations and history. Tree 8: 294–297. Laurence, S., and C. Macdonald, eds. 1998. Contemporary Readings in the Foundations of Metaphysics. Oxford: Blackwell. Lave, J. 1988. Cognition in Practice. Cambridge: Cambridge University Press. Lee, C., and T. Schaaf, eds. 2003. The Importance of Sacred Natural Sites for Biodiversity Conservation. Proceedings of the International Workshop held in Kunming and Xishuangbanna Biosphere Reserve, People’s Republic of China, 17–20 February 2003. Paris: UNESCO. Lees, S. H. 2000. Review of Hess 1997. American Ethnologist 27: 503– 504. Lees, S. H., and D. G. Bates. 1984. Environmental events and the ecology of cumulative change. In E. F. Moran, ed., The Ecosystem Concept in Anthropology. Boulder, CO: Westview Press, pp. 133–159. Lehmann, M. P. 1992. Deforestation and changing land-use patterns in Costa Rica. In H. K. Steen and R. P. Tucker, eds., Changing Tropical Forests: Historical Perspectives on Today’s Challenges in Central & South America. Durham: Forest History Society, pp. 58–76. LePore, E., and B. P. McLaughlin, eds. 1985. Actions and Events. New York: Blackwell. Leroi, A. M., M. R. Rose, and G. V. Lauder. 1994. What does the comparative method reveal about adaptation? American Naturalist 143: 381–402. Lesser, A. 1968. War and the state. In M. Fried, M. Harris, and R. Murphy, eds., War: The Anthropology of Armed Conflict and Aggression. Garden City, NY: Natural History Press, pp. 92–96. Lewis, D. 1979. Counterfactual dependence and time’s arrow. Noûs 13: 455–476. Lewis, D. 1986. Philosophical Papers, Vol. 2. New York/Oxford: Oxford University Press.

Bibliography / 265

Lewis, R. R. 1990. Creation and restoration of coastal plain wetlands in Florida. In J. A. Kusler and M. E. Kentula, eds., Wetland Creation and Restoration: The Status of the Science. Washington, DC: Island Press, pp. 73–101. Lewontin, R. C. 1994. Facts and the factitious in natural science. In J. Chandler, A. I. Davidson, and H. Harootunian, eds., Questions of Evidence. Chicago: University of Chicago Press, pp. 478–491. Li, T. M. 1996. Images of community: Discourse and strategy in property relations. Development and Change 27: 501–527. Lim, G. S. 1972. Studies on penyakit merah disease of rice, 3: Factors contributing to an epidemic in North Krian, Malaysia. Malaysian Agricultural Journal 48: 278–294. Lipton, P. 1991. Inference to the Best Explanation. London: Routledge. Little, D. 1991. Varieties of Social Explanation. Boulder, CO: Westview Press. Lombard, L. B. 1986. Events: A Metaphysical Study. London: Routledge & Kegan Paul. Lombard, L. B. 1991. Events. In H. Burkhardt and B. Smith, eds., Handbook of Metaphysics and Ontology, Vol. 1. Munich: Philosophia Verlag, pp. 256–259. Lombard, L. B. 1995. Event theory. In J. Kim and E. Sosa, eds., A Companion to Metaphysics. Oxford: Blackwell, pp. 140–144. Lombrozo, T., D. Kelemen, and D. Zaitchik. 2007. Inferring design: Evidence of a preference for teleological explanations in patients with Alzheimer’s disease. Psychological Science 18: 999–1006. Lowman, C. 1980. Environment, Society and Health: Ecological Bases of Community Growth and Decline in the Maring Region of Papua New Guinea. Ph.D. diss., Columbia University, New York. Lowman-Vayda, C. 1968. Maring Big Men. Anthropological Forum 2: 199–243. Lüdeke, M. K. B., O. Moldenhauer, and G. Petschel-Held. 1999. Rural poverty driven soil degradation under climate change: The sensitivity of the disposition towards the Sahel Syndrome with respect to climate. Environmental Modeling and Assessment 4: 315–326. Luttbeg, B., M. Borgerhoff Mulder, and M. Mangel. 2000. To marry again or not: A dynamic model for demographic transition. In L. Cronk, N. Chagnon, and W. Irons, eds., Adaptation and Human Behavior: An Anthropological Perspective. New York: Aldine de Gruyter, pp. 345–368. Mackie, C. 1984. The lessons behind East Kalimantan’s forest fires. Borneo Research Bulletin 16: 63–74.

266 / Bibliography

Macnae, W. 1968. A general account of the fauna and flora of mangrove swamps and forests in the Indo-West-Pacific region. Advances in Marine Biology 6: 73–270. Magarey, S. 1987. That hoary old chestnut, free will and determinism: Culture vs. structure, or history vs. theory in Britain (Review article). Comparative Studies in Society and History 27: 626–639. Magnani, L. 2001. Abduction, Reason, and Science: Processes of Discovery and Explanation. New York: Kluwer. Mansfield, E. D., and J. Snyder. 2005. Electing to Fight: Why Emerging Democracies Go to War. Cambridge: MIT Press. Mansfield, E. D., and J. Snyder. 2005/06. Prone to violence: The paradox of the democratic peace. The National Interest 82 (Winter): 39–45. Margolis, M. L., and C. P. Kottak. 2003. Marvin Harris (1927–2001) [obituary]. American Anthropologist 105: 685–688. Marston, S. A., J. P. Jones III, and K. Woodward. 2005. Human geography without scale. Transactions of the Institute of British Geographers 30: 416–432. Martin, J. F. 1983. Optimal foraging theory: A review of some models and their applications. American Anthropologist 85: 612–629. Martin, M. 1993. Geertz and the interpretive approach in anthropology. Synthese 97: 269–286. Martin, R. 1977. Historical Explanation: Re-enactment and Practical Inference. Ithaca, NY: Cornell University Press. Martinez Alier, J. 1995. De la Economía Ecológica al Ecologismo Popular, 3rd edition. Barcelona: Icaria Editorial. Mascia, M. B. 2005. SCB 2006. EANTH-L (Ecological Anthropology listserv) archives, September 9, from http://www.listserv.uga.edu/ cgi-bin/wa?A1=ind0509&L=eanth-l. Last consulted: September 12, 2008. Mascia, M. B., J. P. Brosius, T. A. Dobson, B. C. Forbes, L. Horowitz, M. A. McKean, and N. J. Turner. 2003. Editorial: Conservation and the social sciences. Conservation Biology 17: 649–650. Mayer, A. J. 1988. Why Did the Heavens Not Darken? The “Final Solution” in History. New York: Pantheon. Maynard Smith, J. 1985. Biology and the behaviour of man. Nature 318: 121–122. Maynard Smith, J. 1996. Conclusions. In W. G. Runciman, J. Maynard Smith, and R. I. M. Dunbar, eds., Evolution of Social Behaviour Patterns in Primates and Man. Proceedings of the British Academy 88. Oxford: Oxford University Press for the British Academy, pp. 291–297.

Bibliography / 267

Mayr, E. 1983. How to carry out the adaptationist program. American Naturalist 121: 321–334. Mayr, E. 2004. What Makes Biology Unique? Considerations on the Autonomy of a Scientific Discipline. Cambridge: Cambridge University Press. McCarthy, J. F. 2000. ‘Wild Logging’: The Rise and Fall of Logging Networks and Biodiversity Conservation Projects on Sumatra’s Rainforest Frontier. CIFOR Occasional Paper No. 31. Bogor: Center for International Forestry Research. McCay, B. J. 1978. Systems ecology, people ecology, and the anthropology of fishing communities. Human Ecology 6: 397–422. McCullagh, C. B. 1984. Justifying Historical Descriptions. Cambridge: Cambridge University Press. McGuire, T. R. 1997. The last northern cod. Journal of Political Ecology 4: 41–54 from jpe.library.arizona.edu/volume_4/7MCGUIRE.PDF. Last consulted: July 19, 2008. McGuire, T. R. 2008. Shell games on the water bottoms of Louisiana: Investigative journalism and anthropological inquiry. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 117–134. McIntyre, L. C. 2004. Redescription and descriptivism in the social sciences. Behavior and Philosophy 32: 453–464. McKim, V. R., 1997. Introduction. In V. R. McKim and S. P. Turner, eds., Causality in Crisis? Statistical Methods and the Search for Causal Knowledge in the Social Sciences. Notre Dame: University of Notre Dame Press, pp. 1–19. McKinley, R. 1977. The place of the symbol in the practice of anthropological reasoning. Reviews in Anthropology 4: 414–439. Meggitt, M. 1977. Blood is Their Argument: Warfare among the Mae Enga Tribesmen of the New Guinea Highlands. Palo Alto, CA: Mayfield. Mikkelsen, B. 1994. Methods for Development Work and Research: A Guide for Practitioners. London: Sage. Miles, D. B., and A. E. Dunham. 1993. Historical perspectives in ecology and evolutionary biology: The use of phylogenetic comparative analyses. Annual Review of Ecology and Systematics 24: 587–619. Mill, J. S. 1886. A System of Logic Ratiocinative and Inductive, 8th edition. London: Longmans, Green. Miller, R. W. 1987. Fact and Method: Explanation, Confirmation and Reality in the Natural and the Social Sciences. Princeton: Princeton University Press.

268 / Bibliography

Mills, S., and J. Beatty. 1979. The propensity interpretation of fitness. Philosophy of Science 46: 263–286. Mink, L. O. 1987. Historical Understanding. Ithaca, NY: Cornell University Press. Minnegal, M., ed. 2005. Sustainable Environments, Sustainable Communities: Potential Dialogues between Anthropologists, Scientists and Managers. SAGES Research Paper 21. Victoria: School of Anthropology, Geography and Environmental Studies, The University of Melbourne. Mintz, S. W. 1977. The so-called world system: Local initiative and local response. Dialectical Anthropology 2: 253–70. Moore, B., Jr. 1978. Injustice: The Social Bases of Obedience and Revolt. White Plains, NY: M. E. Sharpe. Moore, D. S. 1996. Marxism, culture, and political ecology: Environmental struggles in Zimbabwe’s eastern highlands. In R. Peet and M. Watts, eds., Liberation Ecologies: Environment, Development, Social Movements. London: Routledge, pp. 125–147. Moore, S. F. 1975. Epilogue: Uncertainties in situations, indeterminacies in culture. In S. F. Moore and B. G. Myerhoff, eds., Symbol and Politics in Communal Ideology. Ithaca, NY: Cornell University Press, pp. 210–239. Moore, S. F. 1986. Social Facts and Fabrications. Cambridge: Cambridge University Press. Moore, S. F. 1987. Explaining the present: Theoretical dilemmas in processual ethnography. American Ethnologist 14: 727–736. Morad, M. 2004. Talking hypothetically: The Duhem-Quine thesis, multiple hypotheses and the demise of hypothetico-deductivism. Geoforum 35: 661–668. Moris, J. 1987. Irrigation as a privileged solution in African development. Development Policy Review 5: 99–123. Morell, V. 1993. Anthropology: Nature-culture battleground. Science 261: 1798–1802. Morren, G. E. B., Jr. 1987. Comment on Bruce Knauft’s “Reconsidering Violence in Simple Human Societies.” Current Anthropology 28: 483–484. Morris, M. W., R. E. Nisbett, and K. Peng. 1995. Causal attribution across domains and cultures. In D. Sperber, D. Premack, and A. J. Premack, eds., Causal Cognition: A Multidisciplinary Debate. Oxford: Clarendon Press, pp. 577–614. Mühlhäusler, P., and A. Peace. 2006. Environmental discourses. Annual Review of Anthropology 35: 457–479.

Bibliography / 269

Nakatani, A. 1999. “Eating threads”: Brocades as cash crop for weaving mothers and daughters in Bali. In R. Rubinstein and L. H. Connor, eds., Staying Local in the Global Village: Bali in the Twentieth Century. Honolulu: University of Hawaii Press, pp. 203–229. Nepstad, D. 2002. A heated fight against devastation. Nature 415: 476. Neumann, R. P. 2005. Making Political Ecology. London: Hodder Arnold. Newell, B., C. L. Crumley, N. Hassan, E. F. Lambin, C. Pahl-Wostl, A. Underdal, and R. Wasson. 2005. A conceptual template for integrative human-environment research. Global Environmental Change 15: 299–307. Nickerson, R. S. 1998. Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology 2 (2): 175–220. Nyerges, A. E. 1998. Review of Fairhead and Leach 1996. Journal of the Royal Anthropological Institute 4: 390–391. Nyerges, A. E. 2008. Orthodoxy and revision in West African Guinea savanna ecology. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 81–98. Nyerges, A. E., and G. M. Green. 2000. The ethnography of landscape: GIS and remote sensing in the study of forest change in West African Guinea savanna. American Anthropologist 102: 271–289. Obidzinski, K., I. Suramenggala, and P. Levang. 2001. L’exploitation forestière illégale en Indonésie: Un inquiétant processus de légalisation. Bois et forêts de tropiques 55/270: 85–97. Ochsner, K. N., and M. D. Lieberman. 2001. The emergence of social cognitive neuroscience. American Psychologist 56: 717–734. O’Hara, R. J. 1988. Homage to Clio, or, toward an historical philosophy for evolutionary biology. Systematic Zoology 37: 142–155. Orlove, B. S. 1980. Ecological anthropology. Annual Review of Anthropology 9: 235–273. Orr, H. A. 1996a. Dennett’s dangerous idea. Evolution 50: 467–472. Orr, H. A. 1996b. Dennett’s strange idea [review of D. C. Dennett’s Darwin’s Dangerous Idea]. Boston Review, Summer, 28–32. Orr, H. A. 1998. The big picture [review of E. O. Wilson’s Consilience: The Unity of Knowledge]. Boston Review, October/November, 42–45. Ortner, S. B. 1978. Sherpas through Their Rituals. Cambridge: Cambridge University Press. Ortner, S. B. 1984. Theory in anthropology since the sixties. Comparative Studies in Society and History 26: 126–166. Oster, G. F., and E. O. Wilson. 1978. Caste and Ecology in the Social Insects. Princeton: Princeton University Press.

270 / Bibliography

Ostrom, E. 1999. Self-governance and Forest Resources. CIFOR Occasional Paper No. 20. Jakarta: Center for International Forestry Research (CIFOR). Ou, S. H. 1985. Rice Diseases, 2nd edition. Slough, UK: CAB (Commonwealth Agricultural Bureaux) Commonwealth Mycological Institute. Oyama, S. 2001. Terms in tension: What do you do when all the good words are taken? In S. Oyama, P. E. Griffiths, and R. Gray, eds., Cycles of Contingency: Developmental Systems and Evolution. Cambridge: MIT Press, pp. 177–193. Padoch, C. 1982. Migration and Its Alternatives among the Iban of Sarawak. Verhandelingen van het Koninklijk Instituut voor Taal-, Land- en Volkenkunde 98. The Hague: Martinus Nijhoff. Padoch, C., M. Pinedo-Vasquez, and A. Roberts. 2008. Process in an eventful environment. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 135–143. Pagdee, A., Y.-S. Kim, and P. J. Daugherty. 2006. What makes community forest management successful: A meta-study from community forests throughout the world. Society and Natural Resources 19: 33–52. Palmer, C. T. 1991. Human rape: Adaptation or by-product? Journal of Sex Research 28: 365–386. Passmore, J. 1965. Explanation in everyday life, in science, and in history. In G. H. Nadel, ed., Studies in the Philosophy of History. Selected Essays from History and Theory. New York: Harper & Row, pp. 16–34. Patel, V. L., J. F. Arocha, and J. Zhang. 2005. Thinking and reasoning in medicine. In K. J. Holyoak and R. J. Morrison, eds., Cambridge Handbook of Thinking and Reasoning. Cambridge: Cambridge University Press, pp. 727–750. Peirce, C. S. 1932. Collected Papers of Charles Sanders Peirce (C. Hartshorne and P. Weiss, eds.), Vol. 2. Cambridge: Harvard University Press. Pelkey, N. 1995. Please stop the PRA RRA rah. Out of the Shell (Coastal Resources Research Network Newsletter) 5(1): 17–24. Pelto, P. J., and G. H. Pelto. 1975. Intra-cultural diversity: Some theoretical issues. American Ethnologist 2: 1–18. Peluso, N. L. 1992. The political ecology of extraction and extractive reserves in East Kalimantan, Indonesia. Development and Change 49 (4): 49–74. Peluso, N. L., and M. Watts, eds. 2001. Violent Environments. Ithaca, NY: Cornell University Press.

Bibliography / 271

Peluso, N. L., and M. Watts. 2003. Violent Environments: Responses. ECSP (Environmental Change and Security Program) Report 9, pp. 93–96. Perusse, D. 1993. Cultural and reproductive success in industrial societies: Testing the relationship at the proximate and ultimate levels. Behavioral and Brain Sciences 16: 267–322. Pierce, G. J., and J. G. Ollason. 1987. Eight reasons why optimal foraging theory is a complete waste of time. Oikos 49: 111–125. Pigliucci, M. 2004. Studying the plasticity of phenotypic integration in a model organism. In M. Pigliucci and K. Preston, eds., Phenotypic Integration: Studying the Ecology and Evolution of Complex Phenotypes. Oxford: Oxford University Press, pp. 50–79. Pinxten, R., Balu, E. Soberon, D. Verboven, and K. Snoeck. 1988. Cultural dynamics: A vision and a perspective. Cultural Dynamics 1: 1–28. Platt, J. R. 1964. Strong inference: Certain systematic methods of scientific thinking may produce much more rapid progress than others. Science 146: 347–353. Polkinghorne, D. 1983. Methodology for the Human Sciences. Albany: State University of New York Press. Polunin, N. V. C. 1984. Do traditional marine “reserves” conserve? A view of Indonesian and New Guinean evidence. In K. Ruddle and T. Akimichi, eds., Maritime Institutions in the Western Pacific. Senri Ethnological Studies No. 17. Osaka: National Museum of Ethnology, pp. 267–283. Pomeroy, R. S., and M. D. Pido. 1995. Initiatives towards fisheries comanagement in the Philippines: The case of San Miguel Bay. Marine Policy 19: 213–226. Pomeroy, R. S., R. B. Pollnac, C. D. Predo, and B. M. Katon. 1996. Impact Evaluation of Community-based Coastal Resource Management Projects in the Philippines. Makati City: International Center for Living Aquatic Resources Management (ICLARM). Popper, K. 1965. The Logic of Scientific Discovery. New York: Harper Torchbooks. Popper, K. 1976. Unended Quest: An Intellectual Autobiography. La Salle: Open Court. Portes, A., and J. Sensenbrenner. 1993. Embeddedness and immigration: Notes on the social determinants of economic action. American Journal of Sociology 98: 1320–1350. Price, C. A. 1963. Southern Europeans in Australia. Melbourne: Oxford University Press.

272 / Bibliography

Primavera, J. H. 1995. Mangroves and brackish water pond culture in the Philippines. Hydrobiologia 295: 303–309. Primavera, J. H., and R. F. Agbayani. 1996. Comparative strategies in community-based mangrove rehabilitation programs in the Philippines. Paper presented at the ECOTONE V conference: Community Participation in Conservation, Sustainable Use and Rehabilitation of Mangroves in Southeast Asia. Ho Chi Minh City, Vietnam, January 8–12. Purcell, T. W. 1998. Indigenous knowledge and applied anthropology: Questions of definition and direction. Human Organization 57: 258–272. Putnam, H. 1978. Meaning and the Moral Sciences. London: Routledge & Kegan Paul. Pyne, S. J. 2000. Review of Robert Boyd, ed., Indians, Fire, and the Land in the Pacific Northwest. H-AmIndian, H-Net Reviews, February, from http://www.h-net.org/reviews/showrev.php?id=3779. Last consulted: September 13, 2008. Quinn, N., and D. Holland. 1987. Culture and cognition. In D. Holland and N. Quinn, eds., Cultural Models in Language and Thought. New York: Cambridge University Press, pp. 3–40. Rabinow, P., and W. M. Sullivan. 1987. The interpretive turn: A second look. In P. Rabinow and W. M. Sullivan, eds., Interpretive Social Science: A Second Look. Berkeley: University of California Press, pp. 1–30. Railsback, L. B. 2004. T. C. Chamberlin’s “Method of Multiple Working Hypotheses”: An encapsulation for modern students. Houston Geological Society Bulletin 47 (2): 68–69. Railton, P. 1980. Explaining Explanation. Unpublished Ph.D. dissertation in philosophy, Princeton University. Railton, P. 1981. Probability, explanation, and information. Synthese 48: 233–256. Ramakrishnan, P. S., K. G. Saxena, and U. M. Chandrashekara, eds. 1998. Conserving the Sacred for Biodiversity Management. Enfield, NH: Science Publishers. Ramon, J., and D. Wall. 1998. Fire and Smoke Occurrence in Relation to Vegetation and Land Use in South Sumatra Province, Indonesia with Particular Reference to 1997, Report No. 47 of the European Commission’s Forest Fire Prevention and Control Project, Palembang, Indonesia. Rappaport, R. A. 1968. Pigs for the Ancestors: Ritual in the Ecology of a New Guinea People. New Haven: Yale University Press.

Bibliography / 273

Ravenel, R. M. 2004. Community-based logging and de facto decentralization: Illegal logging in the Gunung Palung area of West Kalimantan, Indonesia. Journal of Sustainable Forestry 19: 213–237. Ray, D., D. Nepstad, and P. Moutinho. 2005. Micrometeorological and canopy controls of fire susceptibility in mature and disturbed forests of an East-Central Amazon landscape. Ecological Applications 15: 1664–1678. Redford, K. 1990. The ecologically noble savage. Orion 9 (3): 24–29. Repetto, R., and M. Gillis, eds. 1988. Public Policies and the Misuse of Forest Resources. Cambridge: Cambridge University Press. Rescher, N. 1978. Peirce’s Philosophy of Science: Critical Studies in His Theory of Induction and Scientific Method. Notre Dame: University of Notre Dame Press. Reynolds, V. 1976. The Biology of Human Action. San Francisco: Freeman. Rheingold, H. 1988. They Have a Word for It. Los Angeles: J. P. Tarcher. Ribot, J. C. 1998. Theorizing access: Forest profits along Senegal’s charcoal commodity chain. Development and Change 29: 307–341. Ribot, J. C., and N. L. Peluso. n.d. A theory of access: Putting property and tenure in place. Typescript (published in a revised version in Rural Sociology 68: 153–181 [2003]). Richards, P. 1983. Ecological change and the politics of African land use. African Studies Review 26: 1–72. Richards, R. J. 1992. The structure of narrative explanation in history and biology. In M. H. Nitecki and D.V. Nitecki, eds., History and Evolution. Albany: State University of New York Press, pp. 19–53. Richardson, R. C. 2007. Evolutionary Psychology as Maladapted Psychology. Cambridge: MIT Press. Richerson, P. J., and R. Boyd. 1978. A dual inheritance model of the human evolutionary process I: Basic postulates and a simple model. Journal of Social and Biological Structures 1: 127–154. Richerson, P. J., and R. Boyd. 1989. The role of evolved predispositions in cultural evolution: Or, human sociobiology meets Pascal’s wager. Ethology and Sociobiology 10: 195–219. Richerson, P. J., and R. Boyd. 2005. Not by Genes Alone: How Culture Transformed Human Evolution. Chicago: University of Chicago Press. Rindos, D. 1984. The Origins of Agriculture: An Evolutionary Perspective. Orlando: Academic Press. Risjord, M. W. 2000. Woodcutters and Witchcraft: Rationality and Interpretive Change in the Social Sciences. Albany: State University of New York Press.

274 / Bibliography

Risjord, M. W. 2005. Reasons, causes, and action explanation. Philosophy of the Social Sciences 35: 1–13. Robb, D., and J. Heil. 2008. Mental causation. Stanford Encyclopedia of Philosophy (Fall 2008 Edition). Forthcoming from http://plato.stanford .edu/archives/fall2008/entries/mental-causation/. Last consulted: September 2, 2008. Robbins, P. 2004. Political Ecology: A Critical Introduction. Malden, MA: Blackwell. Roberts, C. 1996. The Logic of Historical Explanation. University Park: Pennsylvania State University Press. Rorty, R. 1989. Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press. Rosaldo, R. 1989. Culture & Truth: The Remaking of Social Analysis. Boston: Beacon Press. Roscoe, P. B. 1995. The perils of ‘positivism’ in cultural anthropology. American Anthropologist 97: 492–504. Rose, S. 1998. Lifelines: Biology beyond Determinism. Oxford: Oxford University Press. Rosenberg, A. 2002. Reductionism in an historical science. In M. H. V. Van Regenmortel and D. L. Hull, eds., Promises and Limits of Reductionism in the Biomedical Sciences. Chichester: Wiley, pp. 125–160. Rosenberg, A. 2005. Philosophy of Science: A Contemporary Introduction. 2nd edition. New York: Routledge. Rosenberg, J. 2000. The Follies of Globalisation Theory: Polemical Essays. London: Verso. Roth, P. A. 1996. Will the real scientists please stand up? Dead ends and live issues in the explanation of scientific knowledge. Studies in the History and Philosophy of Science 27: 43–68. Roth, P. A. 2003. Beyond understanding: The career of the concept of understanding in the human sciences. In S. P. Turner and P. A. Roth, eds., Blackwell Guide to the Philosophy of the Social Sciences. Oxford: Blackwell, pp. 311–333. Rozeboom, W. W. 1997. Good science is abductive, not hypotheticodeductive. In L. L. Harlow, S. A. Mulaik, and J. H. Steiger, eds., What If There Were No Significance Tests? Mahwah, NJ: Erlbaum, pp. 335–391. Rücker, G. 2006. Personal communication. Rudel, T. K. 1995. Did TVA make a difference? An organizational dilemma and reforestation in the southern Appalachians. Society and Natural Resources 8: 493–508.

Bibliography / 275

Rudel, T. K. 1999. Critical regions, ecosystem management, and human ecosystem research. Society and Natural Resources 12: 257–264. Rudel, T., and J. Roper. 1997. The paths to rain forest destruction: Crossnational patterns of tropical deforestation, 1975–90. World Development 25: 53–65. Russell, B. 1919. Introduction to Mathematical Philosophy. London: Allen & Unwin. Ryan, A. 1991. When it’s rational to be irrational. New York Review of Books, October 10, pp. 19–22. Saberwal, V. K. 2000. Environmental alarm and institutionalized conservation in Himachal Pradesh, 1865–1994. In A. Agrawal and K. Sivaramakrishnan, eds., Agrarian Environments: Resources, Representations, and Rule in India. Durham, NC: Duke University Press, pp. 68–85. Saenger, P., and N. A. Siddiqi. 1993. Land from the sea: The mangrove afforestation program in Bangladesh. Ocean and Coastal Management 20: 23–39. Sahlins, M. 1976a. Culture and Practical Reason. Chicago: University of Chicago Press. Sahlins, M. 1976b. The Use and Abuse of Biology: An Anthropological Critique of Sociobiology. Ann Arbor: University of Michigan Press. Sahlins, M. 1977. The state of the art in social/cultural anthropology: The search for an object. In A. F. C. Wallace, J. L. Angel, R. Fox, S. McLendon, R. Sady, and R. Sharer, eds., Perspectives on Anthropology 1976. Special Publication No.10. Washington, DC: American Anthropological Association. Sahlins, M. 1981. Historical Metaphors and Mythical Realities: Structure in the Early History of the Sandwich Islands Kingdom. Ann Arbor: University of Michigan Press. Sahlins, M. 1985. Islands of History. Chicago: University of Chicago Press. Sahlins, M. 1995. How “Natives” Think: About Captain Cook, for Example. Chicago: University of Chicago Press. Salmon, W. C. 1990. Four Decades of Scientific Explanation. Minneapolis: University of Minnesota Press. Sarkar, S. 1998. Genetics and Reductionism. Cambridge: Cambridge University Press. Sayer, J. 2001, 2004, 2005. Personal communications. Sayer, J., and B. M. Campbell. 2004. The Science of Sustainable Development: Local Livelihoods and the Global Environment. Cambridge: Cambridge University Press.

276 / Bibliography

Scarre, G. 2000. Mill. In W. H. Newton-Smith, ed., A Companion to the Philosophy of Science. Oxford: Blackwell, pp. 293–298. Schaffer, J. 2003. The metaphysics of causation. Stanford Encyclopedia of Philosophy (Spring 2003 Edition) from http://plato.stanford.edu/ archives/spr2003/entries/causation-metaphysics/. Last consulted: July 22, 2007. Schelling, T. C. 1978. Micromotives and Macrobehavior. New York: Norton. Schmidt, A. 1971. The Concept of Nature in Marx. Translated by B. Fowkes. London: NLB. Schoener, T. W. 1971. Theory of feeding strategies. Annual Review of Ecology and Systematics 2: 369–404. Schum, D. A. 1994. The Evidential Foundations of Probabilistic Reasoning. New York: Wiley. Schum, D. A. 2001. Species of abductive reasoning in fact investigation in law. Cardozo Law Review 22: 1645–1681. Schum, D. A. 2003. Evidence and inferences about past events: An overview of six case studies. In W. Twining and I. Hampsher-Monk, eds., Evidence and Inference in History and Law: Interdisciplinary Dialogues. Evanston, IL: Northwestern University Press, pp. 9–62. Schwartz, J. 1993. Functional explanation and metaphysical individualism. Philosophy of Science 60: 278–301. Scriven, M. 1959. Truisms as the grounds for historical explanations. In P. Gardner, ed., Theories of History. New York: Free Press, pp. 443–475. Scriven, M. 1966. Causes, connections and conditions in history. In W. H. Dray, ed., Philosophical Analysis and History. New York: Harper & Row, pp. 238–264. Scriven, M. 1975. Causation as explanation. Noûs 9: 3–16. Scudder, T., and E. Colson. 1979. Long-term research in Gwembe Valley, Zambia. In G. M. Foster et al., eds., Long-Term Field Research in Social Anthropology. New York: Academic Press, pp. 227–254. Searle, J. 1984. Minds, Brains and Science. Cambridge: Harvard University Press. Searle, J. 1991. Intentionalistic explanations in the social sciences. Philosophy of the Social Sciences 21: 332–344. Searle, J. R. 1998. Mind, Language and Society: Philosophy in the Real World. New York: Basic Books. Sen, A. 1981. Poverty and Famines. Oxford: Clarendon Press. Settle, W. H., H. Ariawan, E. T. Astuti, W. Cahyana, A. L. Hakim, D. Hindayana, A. S. Lestari, and Pajarningsih. 1996. Managing tropical

Bibliography / 277

rice pests through conservation of generalist natural enemies and alternative prey. Ecology 77: 1975–1988. Setyawati, I. 2008. Indigenous knowledge and in situ crop germplasm conservation. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 307–313. Sharp, A. 1956. Ancient Voyagers in the Pacific. Polynesian Society Memoir No. 32. Wellington: Polynesian Society. Shepherd, K. R. 1986. Plantation Silviculture. Boston: Martinus Nijhoff (Kluwer). Shields, B., and P. Moore. n.d. Fire Behaviour in the Tropics: A Review of Fire Behaviour in Tropical Regions for the Gesellschaft für Technische Zusammenarbeit (GTZ) Integrated Forest Fire Management Project, East Kalimantan. Unpublished report. Bowral, Australia: Metis Associates. Siddiqi, N. A., and M. A. S. Khan. 1990. Growth Performance of Mangrove Trees along the Coastal Belt of Bangladesh. Mangrove Ecosystems Occasional Papers No. 8 (1), UNESCO/UNDP/COMAR. Siegert, F., G. Ruecker, A. Hinrichs, and A. A. Hoffmann. 2001. Increased damage from fires in logged forests during droughts caused by El Niño. Nature 414: 437–440. Sillitoe, P. 1977. Land shortage and war in New Guinea. Ethnology 16: 71–81. Sillitoe, P. 1996. A Place Against Time: Land and Environment in the Papua New Guinea Highlands. Amsterdam: Harwood. Sillitoe, P. 1998a. The development of indigenous knowledge: A new applied anthropology. Current Anthropology 39: 223–252. Sillitoe, P. 1998b. It’s all in the mound: Fertility management under stationary shifting cultivation in the Papua New Guinea highlands. Mountain Research and Development 18: 123–134. Sillitoe, P. 1998c. What, know natives? Local knowledge in development. Social Anthropology 6: 203–220. Simmons, C. S., R. T. Walker, C. H. Wood, E. Arima, and M. A. Cochrane. 2004. Wildfires in Amazonia: A pilot study examining the role of farming systems, social capital, and fire contagion. Journal of Latin American Geography 3: 81–95. Sist, P., and N. Nguyen-Thé. 2002. Logging damage and the subsequent dynamics of a dipterocarp forest in East Kalimantan (1990–1996). Forest Ecology and Management 165: 85–103. Siwi, S. 1987. On the occurrence, morphology and biology of Nephotettix species in Indonesia. In M. R. Wilson and L. R. Nault, eds., Proceedings,

278 / Bibliography

2nd International Workshop on Leafhoppers and. Planthoppers of Economic Importance, held Provo, Utah, USA, 28th July-1st August 1986. London: CAB International Institute of Entomology, pp. 65–80. Skocpol, T. 1977. Wallerstein’s world capitalist system: A theoretical and historical critique. American Journal of Sociology 82: 1075–1090. Slobodkin, L. B., and A. Rapoport. 1974. An optimal strategy of evolution. Quarterly Review of Biology 49: 181–200. Smith, C. A. 1984. Local history in global context: Social and economic transitions in western Guatemala. Comparative Studies in Society and History 26: 193–228. Smith, E. A. 1979. Human adaptation and energetic efficiency. Human Ecology 7: 53–74. Smith, E. A. 1983. Evolutionary ecology and the analysis of human social behavior. In R. Dyson-Hudson and M. A. Little, eds., Rethinking Adaptation: Biological and Cultural Models. Boulder, CO: Westview Press, pp. 23–40. Smith, E. A. 1984. Anthropology, evolutionary ecology, and the explanatory limitations of the ecosystem concept. In E. F. Moran, ed., The Ecosytem Concept in Anthropology. Boulder, CO: Westview Press, pp. 51–85. Smith, E. A. 1987. Folk psychology versus pop sociobiology. Behavioral and Brain Sciences 10: 85–86. Smith, E. A. 1991. Inujjuamiut Foraging Strategies: Evolutionary Ecology of an Arctic Hunting Economy. New York: Aldine de Gruyter. Smith, E. A. 1992a. Human behavioral ecology: I. Evolutionary Anthropology 1: 20–25. Smith, E. A. 1992b. Human behavioral ecology: II. Evolutionary Anthropology 1: 50–55. Smith, E. A. 2000. Three styles in the evolutionary study of human behavior. In L. Cronk, N. Chagnon, and W. Irons, eds., Adaptation and Human Behavior: An Anthropological Perspective. Hawthorne: Aldine de Gruyter, pp. 27–46. Smith, E. A., and B. Winterhalder, eds. 1992. Evolutionary Ecology and Human Behavior. New York: Aldine de Gruyter. Smith, E. A., and B. Winterhalder. 2003. Human behavioral ecology. In L. Nadel, ed., Encyclopedia of Cognitive Science, Vol. 2. London: Nature Publishing Group, pp. 377–385. Snow, J. 1936. Snow on Cholera, Being a Reprint of Two Papers by John Snow, M.D. New York: Oxford University Press for The Commonwealth Fund. Somers, M. R. 1996. Where is sociology after the historic turn? Knowledge cultures, narrativity, and historical epistemologies. In T. J.

Bibliography / 279

McDonald, ed., The Historic Turn in the Human Sciences. Ann Arbor: University of Michigan Press, pp. 53–89. Sorrensen, C. L. 2000. Linking smallholder land use and fire activity: Examining biomass burning in the Brazilian Lower Amazon. Forest Ecology and Management 128: 11–25. Sorrensen, C. L. 2004. Contributions of fire use study to land use/cover change frameworks: Understanding landscape change in agricultural frontiers. Human Ecology 32: 395–421. Sponsel, L. E. 2001a. Do anthropologists need religion, and vice versa? Adventures and dangers in spiritual ecology. In C. L. Crumley, ed., New Directions in Anthropology and Environment: Intersections. Walnut Creek, CA: AltaMira Press, pp. 177–200. Sponsel, L. E. 2001b. Is indigenous spiritual ecology a new fad? Reflections from the historical and spiritual ecology of Hawai`i. In J. A. Grim, ed., Indigenous Traditions and Ecology: The Interbeing of Cosmology and Community. Cambridge: Harvard University Center for the Study of World Religions, pp. 159–174. Sponsel, L. E. 2002. New spiritual ecology concentration . . . EANTH-L archives—December 2002, from http://www.listserv.uga.edu/cgibin/wa?A1=ind0212&L=eanth-l#36. Last consulted: July 22, 2007. Stanovich, K. E. 1999. Who Is Rational? Studies of Individual Differences in Reasoning. Mahwah, NJ: Erlbaum. State Ministry for Environment. 1998. Forest and Land Fires in Indonesia. 2 vols. Jakarta: State Ministry for Environment Republic of Indonesia and United Nations Development Programme (UNDP). Steele, J. 1980. Tribal magic. Review of G. Hodgson’s All Things to All Men. Manchester Guardian Weekly, 23 November, p. 22. Steenis, M. Z., and L. G. Fogarty. 2001. Determining Spatial Factors Associated with Fire Ignition Zones. Hotspot Analyses for East Kalimantan. Berau Forest Management Project. Jakarta: European Union and Ministry of Forestry and Estate Crops. Sterelny, K., and P. E. Griffiths. 1999. Sex and Death: An Introduction to the Philosophy of Biology. Chicago: University of Chicago Press. Stolle, F., and E. F. Lambin. 2003. Interprovincial and interannual differences in the causes of land-use fires in Sumatra, Indonesia. Environmental Conservation 30: 375–387. Stolle, F., K. M. Chomitz, E. F. Lambin, and T. P. Tomich. 2003. Land use and vegetation fires in Jambi Province, Sumatra, Indonesia. Forest Ecology and Management 179: 277–292. Stott, P. A., J. G. Goldammer, and W. L. Werner. 1990. The role of fire in the tropical lowland deciduous forests of Asia. In J. G. Goldammer,

280 / Bibliography

ed., Fire in the Tropical Biota: Ecosystem Processes and Global Challenges. Berlin: Springer-Verlag, pp. 32–44. Sturgess, N. H., and H. Wijaya. 1983. Rice harvesting: A view from the theory of common property. Bulletin of Indonesian Economic Studies 19 (2): 27–45. Sukardjo, S., and I. Yamada. 1992. Biomass and productivity of a Rhizophora mucronata Lamarck plantation in Tritih, central Java, Indonesia. Forest Ecology and Management 49: 195–209. Sundar, N., and R. Jeffery. 1999. Introduction. In R. Jeffery and N. Sundar, eds., A New Moral Economy for India’s Forests? Discourses of Community and Participation. New Delhi: Sage, pp. 15–54. Sunley, P. 1996. Context in economic geography: The relevance of pragmatism. Progress in Human Geography 20: 338–355. Suyanto, S. 2000. Fire, deforestation and land tenure in the north-eastern fringes of Bukit Barisan Selatan National Park, Lampung. Bogor: Center for International Forestry Research (CIFOR) from http://www.cifor. cgiar.org/fire/pdf/pdf71.pdf. Last consulted: September 13, 2008. Swedberg, R. 1990. Economics and Sociology, Redefining Their Boundaries: Conversations with Economists and Sociologists. Princeton: Princeton University Press. Swetnam, T. W., and C. H. Baisan. 1996. Historical fire regime patterns in the southwestern United States since AD 1700. In C. D. Allen, ed., Fire Effects in Southwestern Forests. Proceedings of the Second La Mesa Fire Symposium. Fort Collins: USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, pp. 11–32. Symons, D. 1989a. A critique of Darwinian anthropology. Ethology and Sociobiology 10: 131–144. Symons, D. 1989b. Review of L. Betzig et al., eds., Human Reproductive Behavior: A Darwinian Perspective. Man 24: 528–529. Symons, D. 1990. Adaptiveness and adaptation. Ethology and Sociobiology 11: 427–444. Symons, D. 1992. On the use and misuse of Darwinism in the study of human behavior. In J. H. Barkow, L. Cosmides, and J. Tooby, eds., The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press, pp. 137–159. Tacconi, L. 2003. Fires in Indonesia: Causes, Costs, and Policy Implications. CIFOR Occasional Paper No. 38. Jakarta: Center for International Forestry Research (CIFOR). Tacconi, L., and I. Kurniawan. n.d. Deforestation and Forest Degradation in Indonesia. CIFOR Draft Working Paper. Bogor: Center for International Forestry Research (CIFOR).

Bibliography / 281

Tacconi, L., and A. P. Vayda. 2006. Slash and burn and fires in Indonesia: A comment. Ecological Economics 56: 1–4. Taussig, M. 1987. History as commodity in some recent American (anthropological) literature. Food and Foodways 2: 151–169. Taylor, P. J. 1987. Historical versus selectionist explanations in evolutionary biology. Cladistics 3 (1): 1–13. Tetlock, P. E., and A. Belkin. 1996. Counterfactual thought experiments in world politics: Logical, methodological, and psychological perspectives. In P. E. Tetlock and A. Belkin, eds., Counterfactual Thought Experiments in World Politics: Logical, Methodological, and Psychological Perspectives. Princeton: Princeton University Press, pp. 1–38. Thagard, P. 1999. How Scientists Explain Disease. Princeton: Princeton University Press. Thagard, P. 2000. Explaining disease: Correlations, causes, and mechanisms. In F. C. Keil and R. A. Wilson, eds., Explanation and Cognition. Cambridge: MIT Press, pp. 255–276. Thagard, P., and C. Shelley. 1997. Abductive reasoning: Logic, visual thinking, and coherence. In M. L. Dalla Chiara, K. Doets, D. Mundici, and J. van Benthem, eds., Logic and Scientific Methods. Dordrecht: Kluwer, pp. 413–427. Thomason, R. H. 1987. The context-sensitivity of belief and desire. In M. P. Georgeff and A. L. Lansky, eds., Reasoning about Actions & Plans. Los Altos: Morgan Kaufmann, pp. 341–360. Thompson, E. P. 1978. The poverty of theory & other essays. New York: Monthly Review Press. Thorhaug, A. 1990. Restoration of mangroves and seagrasses—economic benefits for fisheries and mariculture. In J. J. Berger, ed., Environmental Restoration: Science and Strategies for Restoring the Earth. Washington, DC: Island Press, pp. 265–281. Thornhill, R. 1990. The study of adaptation. In M. Bekoff and D. Jamieson, eds., Interpretation and Explanation in the Study of Animal Behavior, Vol. 2. Boulder, CO: Westview Press, pp. 31–62. Todorov, T. 1993. On Human Diversity: Nationalism, Racism, and Exoticism in French Thought. Translated by C. Porter. Cambridge: Harvard University Press. Tolstoy, L. N. 1982 [orig. 1869]. War and Peace. Translated by R. Edmonds. Harmondsworth: Penguin. Toma, T. 1999. Exceptional droughts and forest fires in eastern part of Borneo Island. Tropics 9: 55–72. Toma, T., P. Matius, Hastaniah, Y. Kiyono, R. Watanabe, and Y. Okimori. 2000. Dynamics of burned lowland dipterocarp forest stands

282 / Bibliography

in Bukit Soeharto, East Kalimantan. In E. Guhardja, M. Fatawi, M. Sutisna, T. Mori and S. Ohta, eds., Rainforest Ecosystems of East Kalimantan. Tokyo: Springer, pp. 107–119. Tomich, T. P., A. M. Fagi, H. de Foresta, G. Michon, D. Murdiyarso, F. Stolle, and M. van Noordwijk. 1998. Indonesia’s fires: Smoke as a problem, smoke as a symptom. Agroforestry Today, January–March, pp. 4–7. Tomich, T. P., D. E. Thomas, and M. van Noordwijk. 2004. Environmental services and land use change in Southeast Asia: From recognition to regulation or reward? Agriculture, Ecosystems and Environment 104: 229–244. Tomlinson, P. B. 1986. The Botany of Mangroves. London: Cambridge University Press. Tooby, J., and L. Cosmides. 1990. The past explains the present: Emotional adaptations and the structure of ancestral environments. Ethology and Sociobiology 11: 375–424. Tooby, J., and L. Cosmides. 1992. The psychological foundations of culture. In J. H. Barkow, L. Cosmides, and J. Tooby, eds., The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press, pp. 19–136. Trencher, S. R. 2000. Mirrored Images: American Anthropology and American Culture, 1960–1980. Westport, CT: Bergin & Garvey. Trimborn, H. 1949. Senorió y Barbarie en el Valle del Cauca. Madrid: Instituto Gonzalo Fernández de Oviedo. Turner, S. 1997. “Net effects”: A short history. In V. R. McKim and S. P. Turner, eds., Causality in Crisis? Statistical Methods and the Search for Causal Knowledge in the Social Sciences. Notre Dame: University of Notre Dame Press, pp. 23–45. Turner, V. W., S. F. Moore, and B. G. Myerhoff. 1969. The Ritual Process: Structure and Anti-Structure. Chicago: Aldine. Uhl, C., and J. B. Kauffman. 1990. Deforestation, fire susceptibility, and potential tree responses to fire in the Eastern Amazon. Ecology 7: 437–449. Uhl, C., J. B. Kauffman, and D. L. Cummings. 1988. Fire in the Venezuelan Amazon 2: Environmental conditions necessary for forest fires in the evergreen rainforest of Venezuela. Oikos 53: 176–184. Van de Fliert, E. 1993. Integrated Pest Management: Farmer Field Schools Generate Sustainable Practices. A Case Study in Central Java Evaluating IPM Training. Wageningen Agricultural University Papers No. 93–3. Wageningen, The Netherlands: Wageningen Agricultural University.

Bibliography / 283

Van der Vecht, J. 1953. The problem of the mentek disease of rice in Java. Landbouw 25: 45–130. Van Regenmortel, M. H. V., and D. L. Hull, eds. 2002. Promises and Limits of Reductionism in the Biomedical Sciences. Chichester: Wiley. Van Speybroeck, D. 1992. Regeneration strategy of mangroves along the Kenya coast: A first approach. Hydrobiologia 247: 243–251. Varma, A. 2003. The economics of slash and burn: A case study of the 1997–1998 Indonesian forest fires. Ecological Economics 46: 159–171. Vayda, A. P. 1961a. Expansion and warfare among swidden agriculturalists. American Anthropologist 63: 346–358. Vayda, A. P. 1961b. A re-examination of Northwest Coast economic systems. Transactions of the New York Academy of Sciences, 2d ser., 23: 618–624. Vayda, A. P. 1967. Les fêtes commerciales des Indiens Pomo. Économies et Sociétés, Série V, 10: 237–244. Vayda, A. P., ed. 1969. Environment and Cultural Behavior: Ecological Studies in Cultural Anthropology. Garden City, NY: Natural History Press. Vayda, A. P. 1970. Maoris and muskets in New Zealand: Disruption of a war system. Political Science Quarterly 85: 560–584. Vayda, A. P. 1971. Phases of the process of war and peace among the Marings of New Guinea. Oceania 42: 1–24. Vayda, A. P. 1974. Warfare in ecological perspective. Annual Review of Ecology and Systematics 5: 183–193. Vayda, A. P. 1975. Headhunting near and far: Antecedents and effects of coastal raiding by Iban in the nineteenth century. Sarawak Museum Journal 23: 111–138. Vayda, A. P. 1976. War in Ecological Perspective: Persistence, Change, and Adaptive Processes in Three Oceanian Societies. New York: Plenum Press. Vayda, A. P. 1979a. Review of C. R. Hallpike’s Bloodshed and Vengeance in the Papuan Mountains. American Anthropologist 81: 424–425. Vayda, A. P. 1979b. War and Coping. Reviews in Anthropology 6: 191– 198. Vayda, A. P. 1980. Buginese colonization of Sumatra’s coastal swamplands and its significance for development planning. In E. C. F. Bird and A. Soegiarto, eds., Proceedings of the Jakarta Workshop on Coastal Resources Management. Tokyo: United Nations University, pp. 80–87. Vayda, A. P. 1983. Progressive contextualization: Methods for research in human ecology. Human Ecology 11: 265–281.

284 / Bibliography

Vayda, A. P. 1986. Holism and individualism in ecological anthropology. Reviews in Anthropology 13: 295–313. Vayda, A. P. 1987a. Explaining what people eat: A review article. Human Ecology 15: 493–510. Vayda, A. P. 1987b. Reply to Harris. Human Ecology 15: 519–521. Vayda, A. P. 1988. Actions and consequences as objects of explanation in human ecology. Environment, Technology, and Society 51: 2–7. (also in Borden, R. J., et al., eds. 1988. Human Ecology: Research and Applications. College Park, MD: Society for Human Ecology, pp. 9–18.) Vayda, A. P. 1989. Explaining why Marings fought. Journal of Anthropological Research 45: 159–177. Vayda, A. P. 1991. Current issues in social science explanation: An introduction. Philosophy of the Social Sciences 21: 317. Vayda, A. P. 1993. Ecosystems and human actions. In M. J. McDonnell and S. T. A. Pickett, eds., Humans as Components of Ecosystems. New York: Springer-Verlag, pp. 61–71. Vayda, A. P. 1994. Actions, variations, and change: The emerging anti-essentialist view in anthropology. In R. Borofsky, ed., Assessing Cultural Anthropology. New York: McGraw-Hill, pp. 320–330. Vayda, A. P. 1995a. Failures of explanation in Darwinian ecological anthropology: Part I. Philosophy of the Social Sciences 25: 219–249. Vayda, A. P. 1995b. Failures of explanation in Darwinian ecological anthropology: Part II. Philosophy of the Social Sciences 25: 360–375. Vayda, A. P. 1996. Methods and Explanations in the Study of Human Actions and Their Environmental Effects. Special Publication. Jakarta: Center for International Forestry Research (CIFOR) and World Wide Fund for Nature (WWF). Vayda, A. P. 1997a. Against political ecology. Paper presented at the 96th Annual Meeting of the American Anthropological Association, Washington, DC, November 22. Vayda, A. P. 1997b. Managing forests and improving the livelihoods of forest-dependent people: Reflections on CIFOR’s social science research in relation to its mandate for generalisable strategic research. CIFOR Working Paper No. 16. Jakarta: Center for International Forestry Research (CIFOR). Vayda, A. P. 1997c. Rappaport and causal explanation of events. Discussant’s comments, presented in the session on “Culture, Power, History, Nature: Papers in Honor of Roy A. Rappaport” at the 96th Annual Meeting of the American Anthropological Association, Washington, DC, November 21.

Bibliography / 285

Vayda, A. P. 1998a. Anthropological perspectives on tropical deforestation? A review article. Anthropos 93: 573–579. Vayda, A. P. 1998b. Persistence and change in agricultural and forest biodiversity in Yunnan Province, China: An event-focused approach. Draft proposal for a collaborative project with the Chinese Academy of Sciences/Kunming. Typescript. Vayda, A. P. 1999. Finding Causes of the 1997-98 Indonesian Forest Fires: Problems and Possibilities. WWF Indonesia Forest Fires Project. Jakarta: World Wide Fund for Nature (WWF) Indonesia. Vayda, A. P. 2000a. Forest-fire research in Borneo: Findings and implications. Paper presented at the American Anthropological Association Annual Meeting, San Francisco, November 15. Vayda, A. P. 2000b. Political ecology, event ecology, and event social science: Some methodological issues. Discussant’s comments, presented in the session on “Political Ecology at a Crossroads: Methods, Analyses, and Applications” at the 99th Annual Meeting of the American Anthropological Association, San Francisco, November 18. Vayda, A. P. 2001. Forest-fire research in Indonesia, II. Anthropology News, January, p. 45. Vayda, A. P. 2004. Panelist’s remarks, presented in “Cultural and Political Ecology at the AAG Century Mark,” Session II, at the Centennial Meeting of the Association of American Geographers, Philadelphia, March 16, from http://www2.stetson.edu/cape/ Features/capecentury/vayda.htm. Last consulted: August 13, 2008. Vayda, A. P. 2008. Causal explanation as a research goal: A pragmatic view. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 317–367. Vayda, A. P., K. Alley, and J. Hess. 2002. Messages on “Spiritual Ecology and Methodological Issues.” EANTH-L archives—December 2002, from http://www.listserv.uga.edu/cgi-bin/wa?A1=ind0212&L=eanthl#36. Last consulted: July 22, 2007. Vayda, A. P., C. J. P. Colfer, and M. Brotokusumo. 1980. Interactions between people and forests in East Kalimantan. Impact of Science on Society 30: 179–190. Vayda, A. P., A. Leeds, and D. Smith. 1961. The place of pigs in Melanesian subsistence. In V. Garfield, ed., Proceedings of the 1961 Annual Spring Meeting of the American Ethnological Society. Seattle: University of Washington Press, pp. 69–77.

286 / Bibliography

Vayda, A. P., and A. Sahur. 1985. Forest clearing and pepper farming by Bugis migrants in East Kalimantan: Antecedents and impact. Indonesia 39: 93–110. Vayda, A. P., and A. Sahur. 1996. Bugis Settlers in East Kalimantan’s Kutai National Park: Their Past and Present and Some Possibilities for Their Future. A CIFOR Special Publication. Jakarta: Center for International Forestry Research (CIFOR). Vayda, A. P., and I. Setyawati. 1998. Questions about culture-related considerations in research on cognition and agro-ecological change: Illustrations from studies of agricultural pest management in Java. Antropologi Indonesia (Indonesian Journal of Social and Cultural Anthropology) 55: 44–52. Vayda, A. P., and B. B. Walters. 1999. Against political ecology. Human Ecology 27: 167–179. Vayda, A. P., B. B. Walters, and I. Setyawati. 2004. Doing and knowing: Questions about studies of local knowledge. In A. J. Bicker, P. Sillitoe, and J. Pottier, eds., Investigating Local Knowledge: New Directions, New Approaches. London: Ashgate Publishing, pp. 35–58. Vermazen, B. 1985. Negative acts. In B. Vermazen and M. B. Hintikka, eds., Essays on Davidson: Actions and Events. Oxford: Clarendon Press, pp. 93–104. Vondal, P. J. 2008. Using progressive contextualization to understand a rural development success story in Indonesian Borneo. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 255–271. von Wright, G. H. 1971. Explanation and Understanding. Ithaca, NY: Cornell University Press. Waddell, E. 1972. The Mound Builders: Agricultural Practices, Environment, and Society in the Central Highlands of New Guinea. Seattle: University of Washington Press. Waddell, E. 1973. Raiapu Enga adaptive strategies: Structure and general implications. In H. C. Brookfield, ed., The Pacific in Transition: Geographical Perspectives on Adaptation and Change. New York: St. Martin’s Press, pp. 25–54. Waddell, E. 1975. How the Enga cope with frost: Responses to climatic perturbations in the Central Highlands of New Guinea. Human Ecology 3: 249–273. Wadley, R. L., and M. Eilenberg. 2005. Autonomy, identity, and “illegal” logging in the borderland of West Kalimantan, Indonesia. Asia Pacific Journal of Anthropology 66: 19–34.

Bibliography / 287

Wadsworth, F. H. 1997. Forest Production for Tropical America. Agricultural Handbook 710. Washington, DC: U.S. Department of Agriculture (Forest Service). Wahlsten, D. 1993. Sociobiology flops again. Behavioral and Brain Sciences 16: 310–311. Walsh, W. H. 1976. An Introduction to Philosophy of History. Atlantic Highlands: Humanities Press. Walters, B. B. 1995. People, policies and resources: Mangrove restoration and conservation in the Bais Bay Basin, Negros Oriental, and wider Philippine context. In M. A. Junio-Menez and G. F. Newkirk, eds., Philippine Coastal Resources under Stress. Halifax and Quezon City: Coastal Resources Research Network, Dalhousie University and the Marine Science Institute, University of the Philippines, pp. 151–165. Walters, B. B. 1997. Human ecological questions for tropical forest restoration: Experiences from planting native upland trees and mangroves in the Philippines. Forest Ecology and Management 99: 275–290. Walters, B. B. 1998a. Cutting, clearing, planting and protecting: Human influences on the ecology of Philippine mangrove forests. Paper presented at the Annual Meeting of the Ecological Society of America, Baltimore, MD, August 3–6. Walters, B. B. 1998b. Muddy mangroves and murky common property theories. Paper presented at the meeting of the International Association for the Study of Common Property, Vancouver, Canada, June 6–9. Walters, B. B. 2000a. Event Ecology in the Philippines: Explaining Mangrove Cutting and Planting and Their Environmental Effects. Ph.D. dissertation, Rutgers University, New Brunswick, NJ. Walters, B. B. 2000b. Local mangrove planting in the Philippines: Are fisherfolk and fish pond owners effective restorationists? Restoration Ecology 8(3): 237–246. Walters, B. B. 2003. People and mangroves in the Philippines: Fifty years of coastal environmental change. Environmental Conservation 30: 293–303. Walters, B. B. 2004. Local management of mangrove forests: Successful conservation or efficient resource exploitation? Human Ecology 32: 177–195. Walters, B. B. 2005a. Ecological effects of small-scale cutting of Philippine mangrove forests. Forest Ecology and Management 206: 331–348. Walters, B. B. 2005b. Patterns of local wood use and cutting of Philippine mangrove forests. Economic Botany 59: 66–76.

288 / Bibliography

Walters, B. B. 2005c. Review of Greenough and Tsing 2003. American Anthropologist 107: 286–287. Walters, B. B. 2008. Events, politics, and environmental change. In B. B. Walters, B. J. McCay, P. West, and S. Lees, eds., Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press, pp. 67–79. Walters, B. B., A. Cadelina, A. Cardano, and E. Visitacion. 1999. Community history and rural development: Why some farmers participate more readily than others. Agricultural Systems 59: 193–214. Walters, B. B., B. J. McCay, P. West, and S. Lees, eds. 2008. Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology. Lanham, MD: AltaMira Press. Walters, B. B., and A. P. Vayda. 2009. Event ecology, causal historical analysis and human-environment research. Annals of the Association of American Geographers, in press. Walton, D. 2004. Relevance in Argumentation. Mahwah, NJ: Erlbaum Ward, D. 1992. The role of satisficing in foraging theory. Oikos 63: 312–317. Ward, D. 1993. Foraging theory, like all other fields of science, needs multiple working hypotheses. Oikos 67: 376–378. Watts, M. 1991. Entitlements or empowerment? Famine and starvation in Africa. Review of African Political Economy 51: 9–26. Watts, M., and H. G. Bohle. 1993. The space of vulnerability: The causal structure of hunger and famine. Progress in Human Geography 17: 43–67. Watts, M., and R. Peet. 2004. Liberating political ecology. In R. Peet and M. Watts, eds., Liberation Ecologies, Second Edition: Environment, Development, Social Movements. London: Routledge, pp. 3–47. Weber, M. 1949. The Methodology of the Social Sciences. Translated and edited by E. A. Shils and H. A. Finch. Glencoe: Free Press. Weinstock, J. A. 1994. Rhizophora mangrove agroforestry. Economic Botany 48: 210–213. West-Eberhard, M. J. 1993. Adaptation: Current usages. In E. F. Keller and E. A. Lloyd, eds., Keywords in Evolutionary Biology. Cambridge: Harvard University Press, pp. 13–18. Wiens, J. A. 1984. Resource systems, populations, and communities. In P. W. Price, C. N. Slobodchikoff, and W. S. Gaud, eds., A New Ecology: Novel Approaches to Interactive Systems. New York: Wiley, pp. 397–436. Williams, G. C. 1966. Adaptation and Natural Selection: A Critique of Some Current Evolutionary Thought. Princeton: Princeton University Press. Williams, G. C. 1992. Natural Selection: Domains, Levels, and Challenges. New York: Oxford University Press.

Bibliography / 289

Williams, T. 1870. Fiji and the Fijians, 3rd edition. London: Hodder and Stoughton. Williamson, G. B., and R. C. G. Mesquita. 2001. Effects of fire on rainforest regeneration in the Amazon Basin. In R. O. Bierregaard Jr., C. Gascon, T. E. Lovejoy, and R. C. G. Mesquita, eds., Lessons From Amazonia: The Ecology and Conservation of a Fragmented Forest. New Haven: Yale University Press, pp. 325–334. Winterhalder, B. 1981. Optimal foraging strategies and hunter-gatherer research in anthropology: Theory and models. In B. Winterhalder and E. A. Smith, eds., Hunter-Gatherer Foraging Strategies: Ethnographic and Archeological Analyses. Chicago: University of Chicago Press, pp. 13–35. Winterhalder, B. 1987. The analysis of hunter-gatherer diets: Stalking an optimal foraging model. In M. Harris and E. B. Ross, eds., Food and Evolution: Toward a Theory of Human Food Habits. Philadelphia: Temple University Press, pp. 311–339. Wolf, E. R. 1982. Europe and the People without History. Berkeley: University of California Press. Woodward, J. 1990. Supervenience and singular causal statements. In D. Knowles, ed., Explanation and Its Limits. Royal Institute of Philosophy Lecture Series 27. Cambridge: Cambridge University Press, pp. 211–246. WRI, UNEP, and UNDP (World Resources Institute, United Nations Environment Programme, and United Nations Development Programme). 1990. World Resources 1990–91. New York: Oxford University Press. Yao, C. E. 1996. Mangrove reforestation in the Central Visayas. Greenfields 24 (5/6): 29–31, 34–37. Yao, C. E., and F. Nanagas. 1984. Banacon Island: Biggest bakauan plantations in Central Visayas. Canopy International 10: 1, 4, 5, 9. Young, C. C. 2002. In the Absence of Predators: Conservation and Controversy on the Kaibab Plateau. Lincoln: University of Nebraska Press. Young, O. R., E. F. Lambin, F. Alcock, H. Haberl, S. I. Karlsson, W. J. McConnell, T. Myint, C. Pahl-Wostl, C. Polsky, P. S. Ramakrishnan, H. Schroeder, M. Scouvart, and P. H. Verburg. 2006. A portfolio approach to analyzing complex human-environment interactions: Institutions and land change. Ecology and Society 11 (2), Article 31, from http:// www.ecologyandsociety.org/vol11/iss2/art31/. Last consulted: July 19, 2008. Zimmerman, M. J. 1984. An Essay on Human Action. New York: Peter Lang.

Index

abduction: achievement of, 30–31; female, 231; illustrations of, 31; Peirce on, 30 abuse, resource, 114 adaptation: DPs and, 171; studying, 171–72 adaptive preference formation, 178 agency. See human agency agriculture, slash-and-burn, 59 Alcorn, Janis, 118 Amazon forest, 67–68 Amung-me, 111 analysis: DEA methods of, 145; design, 38 anorexia nervosa, 18 anthropology, 85; cognitive, 86–87; human agency in, 195; new/different, 200–201; research approaches in, 110; tropical deforestation and perspective of, 109–21; of war, 229–37. See also Darwinian ecological anthropology

The Anthropology of War (Haas): review of, 229–37; typographical errors in, 237 Aquilaria trees, 132 arson, 52, 57; loggers and, 62 assumptions, studies and, 86–87 Bacon, Francis, confirmation bias as identified by, 32 balance: as maintained, 123; in nature, 123, 125, 127 Bari Indians, 167, 174 Barth, F., 161–62 Bedoya, Eduardo, 117 behavior, 37; DEAs claim to explain, 169; evolutionary origins of, 149; explanations of, 149, 181; factors contributing to, 95; of fires, 64–70; fitnessenhancing, 158; knowledge bases of, 88; maladaptive/ nonadaptive, 185; nonDEA explanations of, 149; prediction of, 172; proximate causes of, 149–50; reproductive

291

292 / Index

success and effect of, 164; shared knowledge’s influence on, 88; spread of, 159–60, 232; studying, 145–46; traits of, 155, 159–60. See also human behavioral ecology beliefs: Botkin on, 125, 128; deepseated, 127–28 Benedict, Ruth, 233 Bentley, J. W., on indigenous knowledge, 87 bias: confirmation, 34–35; heuristic, 32. See also confirmation bias biodiversity, 119, 132n106 biology, evolutionary, 42, 153, 186 Boone, J. L., 161–62 Borgerhoff Mulder, Monique, 159–61, 177, 180 Botkin, Daniel B.: on beliefs, 125, 128; computer models and, 125; on elephants, 126–27; on perception of nature, 123–24; on policy concerns, 126; on predators/prey, 124–25 Braun, D. on causally relevant properties, 4–5 Bush, George H., 202 Bush, George W., 202 butterfly mimicry, 42 Carneiro, Robert, 229–30 Castro, Alfonso Peter, 117 causal connections, 178–79; evidence of, 38; nature of, 5–6; probability and, 5 causal explanation(s), 2n1, 183; account given by, 47; of cholera outbreak, 17; of concrete events, 50, 50n55; engaging in, 43; giving, 6–11; goal of, 3,

20–21, 26, 44, 50; of Indonesian forest fires, 49–76; practical business of, 2; of processes, 203; questions and, 7; as research goal, 1–47, 53n58; turning away from, 11, 11n11; types of, 10; as ubiquitous in everyday life, 44 causal histories: events included in, 24–25; explanations, 188n144; explanatory information put together in form of, 15–16; features characterizing, 9; information given about events, 7–8, 43; investigating, 9 causality: in models, 173–79; nature of, 2 causal relata: events as, 3–4; standard view of, 3 causation: billiard-ball, 10; event, 84n81, 203; as inferred from correlation, 16; literature on, 5–6; mental, 10; metaphysics of, 2; explanatory role of, 5–6 causes: behavior’s proximate, 149–50; of deforestation, 113– 15; of disease occurrence, 14, 15n17; eliminating possible, 29–30, 30n33; events as, 8; explanatory information consisting of, 9–10; of fires, 52, 61; intentions as proximate, 152; without metaphysics, 2– 11. See also causal explanation Center for International Forestry Research (CIFOR), 51–52 Chagnon, Napolean, 230–31 chain migration, 202–3, 203n152 Chamberlin, T. C., 30n35; on data collection, 30–31; multiple working hypotheses of, 45

Index / 293

chance, 125–28 change: environmental, 129–32, 135–36, 140; knowledge and practice, 85–86; in pest management practices, 88 cholera, 17 CIFOR. See Center for International Forestry Research cigarettes, 61, 62n66 claims, causal, 24 Cochrane, Mark, 65–66 community: control, 133; empowering local forest, 120; fire management, 63, 69–61; forest management, 111; nature of, 202; regulation, 133n108; view of, 195–96 complexity, seeing nature’s, 123–28 computers, models of forest successions on, 125–26 conclusions, pressures to reach, 31 confirmation bias, 40; Bacon as identifying, 32; in consequence explanations, 37–43; problem with, 29–37; in textbooks, 36 conflicts: reproductive, 230–31; types of, 231 confusion, polemics and, 229–37 connections. See causal connections consequence explanations: as acceptable, 46; confirmation bias in, 37–43; Harris’, 39, 39n47; as questionable, 37 consequences, beneficial, 37–38 conservation: economic profits v. long-term, 112; environmental, 79; forest, 75; initiatives, 81; short-term, 138

contexts, 202, 211 control: community, 133; ignition, 65; over forests’ resources, 135–36; predators and, 124 cooperation, evolution of human, 182 correctness, 236 Costa Rica, deforestation in, 115–16 costs, sunk, 20n20 crabs, used in pest management, 92 currencies: fitness and, 164–65; proxy, 164–73 Darwin, Charles, 19; central tenent of work by, 146–47; on differential reproduction, 144; human affairs and, 171–72; on trait development, 188 Darwinian ecological anthropology (DEA): agnosticism of, 152; analytic methods of, 145–46; behavior as explained by, 169; as characterized, 145; costbenefit interpretations and, 164; criticisms of, 185–89; Darwinian theory as reduced to predictions in, 171; Darwinism and, 189; on explanation, 172; explanations of, 149; explanatory issues and, 158–59; failures of explanation in, 143–89; goal of, 150; hypothetico-deductive method as adopted by, 169–70; predictions in, 171–72, 172n130, 180–81; procedures, 181; proxy currencies as used in, 165–66; studies, 152–53

294 / Index

Darwinian histories, 188 Darwinian psychologists (DPs): adaptation as studied and, 171; arguments of, 147–48; critique of, 148n116; Pleistocene living conditions posited by, 148 data collection, 45, 54, 164, 211; on forest fire damage, 67; guided by method of multiple working hypotheses, 30–31 DEA. See Darwinian ecological anthropology deforestation: acceleration, 118; anthropological perspective on tropical, 109–21; causes of, 113–15; changing course of, 117; conclusions about, 110–11; in Costa Rica, 115–16; elemental factors in, 112; as global issue, 109; Honduran, 114; from people’s viewpoint, 113–14; in Peru, 117; processes on local level of, 110; rates of tropical, 109, 115–16 DENR. See Department of Environmental and Natural Resources Department of Environmental and Natural Resources (DENR), 135 design analysis, 38–39, 38n46, 42; in evolutionary biology, 42 diagnosis, medical, 31 Discordant Harmonies: A New Ecology for the Twenty-First Century (Botkin), 123 disease, causes of occurrence of, 14, 15n17 Doyle, Arthur Conan, 31 DPs. See Darwinian psychologists

droughts, 71, 120–21, 126–27; fire as used in times of, 72 ecology: event, 12, 12n13, 131–32, 131n105, 133–34, 141; evolutionary, 144–45; spiritual, 33, 33n39, 35, 45–46. See also Evolutionary Ecology and Human Behavior (Smith); human behavioral ecology; political ecology economic development, 79 effectiveness, ecological, 87 elephants, 126–27 Elster, J., 179n135, 231; criticisms of, 155; on explanation by mechanisms, 177–76; on naïve functionalism, 155–56 embeddedness, cultural, 84–85, 104, 106–7 environment: change in, 129–32, 135–36, 140; degradation of, 135–36; violence and, 34–35 erosion, 135 ethnography: holistic, 107; of logging, 73–75; studies and, 81–82 Evans-Pritchard, E. E., on Zande people, 10 event(s): in causal histories, 24–25; as causal relata, 3–4; causally relevant properties of, 4–5; causation, 84n81, 203; as causes, 8; concept of, 24, 203; connecting, 199; as explained through strategy eliminating possible causes of, 29–30, 30n33; explanations of particular, 8–9; ignition, 73–74; interconnections of, 25–26;

Index / 295

process as made by, 198–201, 210; substituting process for, 197. See also event ecology event ecology, 12, 12n13, 131–32, 131n105, 141; rationale for alternative of, 133–34 evidence: of causal histories, 38; explanations supported by, 21–22; failure to judge, 29–30 evolution: genetic, 183; as history-dependent process, 187–88; of human cooperation, 182; social, 225 evolutionary biology, 153, 159, 186; design analysis in, 42 Evolutionary Ecology and Human Behavior (Smith), 148–49, 156–58, 164; chapters of, 146; comprehensive theory development in, 169; field of study in, 143; prey choice in, 176–77 explanations: of behavior, 149, 181; beneficial consequences highlighted by, 37–38; causal histories, 188n144; causal/mechanical, 149–55, 177; causation’s role in, 5–6; concepts of process in social science, 191–203; of DEA, 149; DEA and failures in, 143–89; DEAs on, 172; eliminative strategy of, 34; evidence supporting, 21–22; as failing, 46; failure to judge evidence for, 29–30; functional, 205–6; generality in, 177; giver/seeker’s different understandings of, 22; HBE, 37; how-possibly,

183–84, 184n139; judging, 20–29; laws-and-causes, 7n7; by mechanisms, 177–78; narrative, 188; nomological, 149–55; object of, 131–32, 197; of particular events, 8–9; as presented, 47; scientific, 11–12; stale news, 21–22. See also causal explanation; consequence explanations explanatory information, 7–8; causal histories form of, 15–16; consisting of causes, 9–10; generalizations used for obtaining, 17–18; latitude in methods for obtaining, 14, 14n15; obtaining, 11–20; statistical methods for obtaining, 16–17 explanatory relativity, 223–27; recognizing, 224 faith, 173 families, 182 Ferguson, Brian, 232–33 fire(s): behavior of, 64–70; causes of, 52; cigarettes and, 61; direct observation of, 69; droughts and, 72; hazards, 55; human use of, 70–71; management, 49–50, 52–53, 58–61, 64–65; nonignition studies on, 64–65; policy-making for management of, 64; research, 52–53; site of, 57; slash-andburn agriculture and, 50; susceptibility, 64–70. See also forest fires fire ignition zones (FIZ), 59; conclusions reached about,

296 / Index

60; ignition sources in, 63; in logged forest areas, 62 fitness, 145; currencies and, 164– 65; enhancement, 158, 181, 183; explanatory significance of, 159; among human foragers, 164–65; maximization, 157–58; phenotypes and, 166 FIZ. See fire ignition zones food acquisition, 164, 166–69; time spent on, 165 foraging, 164–67, 174; behavior, 169; gender differences in, 168; optimal, 174n132 forest fires, 27; causal explanation of Indonesian, 49–76; causes of, 52, 61; data on damages by, 67; distortions of research priorities and Indonesian, 51–52; explanations of start v. spread of, 51–52; forensic firescene investigations routinely undertaken after, 55–56; ignition studies and, 51–63; in Indonesia, 13–14, 27, 49–76; land-use features associated with, 67, 71–72; literature on, 57; logging as related to, 68; management of, 65; preventing, 53, 65; research, 13–14, 64–70, 68n73, 73; spread of, 51–52, 69, 76; types of, 67 forests: characteristics shared by areas of tropical, 111; community management of, 111; conservation, 75; control over resources of, 135; dipterocarp, 51; ecological successions in, 125; fire immune state of, 67; flammability of, 69–70; Indian,

28–29, 118; microclimates as changing of, 52; sustainable management of, 112–13; tropical moist, 49–50, 67. See also deforestation formation. See adaptive preference formation frost: protection from, 153–54; radiation, 153 gender, in foraging, 168 generality: in explanations, 177; in models, 173–79 generalizations, 19; as applicable, 178–79; explanatory utility of, 20; misuse of, 1; for obtaining explanatory information, 17–18 geneticists, ecological, 159 Gezon, L. L., on political ecology, 132 Ghiselin, Michael, 171–72 Gibson, Thomas, 234 globalization, 24n24 goal(s): of causal explanation, 3, 20–21, 26, 50; causal explanation as research, 1–47, 53n58; culturally defined, 87; ultimate, 179 Gould, S. J., on storytelling, 199–200 Grand Canyon, 124 Great Bengal Famine of 1943, 9 greed, 112 Green Revolution, 41 Guajajara Indians, logging permits and, 112 Haaland, G.: on interaction systems, 24–25; on Nepalese tailoring in Thailand, 15–16

Index / 297

Haas, Jonathan, 230, 235 Hallpike, C. R., 22, 205–6, 221, 224–25, 227 Harris, Marvin: consequence explanations of, 39, 39n47; cultural materialism championed by, 230 Hawkes, K., 162–63; behavior’s effect on reproductive success and, 164 HBE. See human behavioral ecology historiography, 6 history. See causal histories; Darwinian histories Honduras, deforestation in, 114 human agency, 193–95, 211; attending to, 221 human behavioral ecology (HBE): characterizing explanations of, 37; Darwinian, 35; as established, 145; as manifesting confirmation, 35–36; studies, 42; use of, 146, 146n114 hunting, 167–68, 170n128, 181 hypothetico-deductive method, 169–73 ignition(s): accidental, 62; causes of, 61–62; control, 65; events, 73–74; sites of, 57; sources in FIZ, 63. See also fire ignition zones ignition studies: for fire management, 58; forest fires and, 51–63; ignition sources/ zones identified by, 55; priority of events in, 54; as relevant, 53; site selection for, 54 India, 35n42; forests in, 28–29, 118

indigenous people: as effective conservationists, 113; logging and, 116 Indonesia: change in pest management by farmers in, 88; forest fires in, 13–14, 27, 49–76; IPM in, 85–89; IPM training in, 89; knowledge studies/ practice related to insect pest management in, 85–95; logging in, 3–4, 55; management programs in, 58–59 inequality, 135 information. See explanatory information inquiry, causal, 11 insect pest management: farmers’ practices for, 40, 92; knowledge bases of practices in, 86–87; knowledge studies/ practice related to Indonesian, 85–95. See also integrated pest management insect pests: damage caused by, 86; deliberate nonaction toward, 93–94, 93n85; field observations of, 95; infestation of, 93; inspecting for, 90; interactions between predators and, 86; predatory-prey interactions affecting, 91, 91n84; visible, 90 integrated pest management (IPM): cultural factors and, 95; farmers in training v. practice of, 90–91; in Indonesia, 85–89; modification of training in, 91– 92; objects of study in project, 94–95; training, 86, 89 intentions, 163; as proximate causes, 152

298 / Index

interaction systems, importance of, 24, 130 interdisciplinary research, human dimension and, 70–76 IPM. See integrated pest management Klein, Lorien, 117 knowledge: applying local, 106– 7; bases in practices of insect pest management, 86–87; behavior as influenced by shared, 88; behavior’s bases of, 88; cultural embeddedness and local, 84; demystifying ideas about local, 80; embedded, 85, 104; inaction and, 93–94; indigenous, 79n75, 87; insect pest management related to studies of practice and, 85–95; mangrove trees and local, 102– 3; New Guinea Highlands’ local, 104–6; practice changes with new, 85–86; results of local, 82; spread of, 103–4; studies on local, 79–107; studies on practice and, 96– 104; systems, 79–80 Koch, Klaus, 225 land: as annexed following war, 216–17; forest fires and use of, 67, 71–72; shortages, 214–24; taking enemy, 211–12 Lansing, Steve, 40–41 laws, 7n7; causal mechanisms and, 151 learning: selective, 183; social, 183 legal reform, 197–98 Leopold, Aldo, 124, 127

Lewis, D.: on explanationseekers, 23–24; on explanatory information, 7–8 literature: on causation, 5–6; on forest fires, 57 logging, 62; arson and, 62; ethnography of, 73–75; fireimmune state of forests and, 67; forest fires’ relation to, 68; indigenous farmers as affected by, 116; in Indonesia, 3–4, 55; permits, 112; predatory, 3–4; selective, 12; uncontrolled, 68 lung cancer, 14–15 MacArthur, Robert, 186–87 management: community fire, 59–61, 63; fire, 49–50, 52–53, 58–61, 64–65; forest, 111; mangrove trees and systems of, 97; programs in Indonesia, 58–59; resources in forest fire, 65; sustainable forest, 112–13. See also insect pest management; integrated pest management mangrove trees: ability for planting, 98; bakau, 139; claiming area of, 137–38; clearings of, 96, 136–37; control over resources of, 136; distribution of, 136; as ecologically important, 134–35; ecological services provided by forests of, 96; ecological surveys of, 138; environmental events and, 102; erosion and, 135; fisherfolk planting under own initiatives, 96–97; human influence on forests of, 140;

Index / 299

local knowledge’s value for restoration elsewhere of, 102– 3; management systems for planting, 97; in Philippines, 96–104, 134–40; as planted, 139–40; protecting, 136; shell infestations of, 100–101, 100n90; spacing of, 99–100, 99n88; species of, 98, 139; techniques for planting, 98–99; test-planting and, 101; used engine oil used for, 100–101; uses of wood from, 139; wood harvesting of, 96 Marginal Value Theorem, 174 Marings: fieldwork among, 206–10; land shortage as cause of fighting by, 214–24; population pressure and, 214; territorial conquests and, 210, 212–13; warfare, 206–8, 206n156, 214–18 materialism, cultural, 230 mechanisms: causal, 151; explanation by, 176–78; mysterious, 155–63 metaphysics: of causation, 2; causes without, 2–11 method(s): data collection, 30– 31; for obtaining explanatory information, 14, 14n15, 16–17; hypothetico-deductive, 169– 73; research, 82; statistical, 15–17 migration, chain, 202–3, 203n152 Mill, John Stuart, 6 mimicry, 43; butterfly, 42 mind, DPs on, 147 mirror image trap, 34 models: applicability of, 174–75, 175n133, 179;

applying, 173–74; causality in, 173–79; commons, 175, 179; computer, 125–26; critical-mass, 175; developing methodology for, 169; for explanatory purpose, 179; foraging, 164–69; generality in, 173–79; prey choice, 167, 170, 175–76; research role of, 173–74; system-analysis, 125; threshold, 175. See also optimality models; regression models Molnar, Augusta, 118 Moran, Emilio, 117 mounding practices, 154 mountain lions, 124 mule deer, human interference with nature and, 124–25 myths, 124 naïve functionalism: Elster on, 155–56; mysterious mechanisms and, 155–63 narratives, 188 natural selection, 153, 157, 159, 162, 164, 172, 211; as causal story, 19, 178 nature: balance of, 123, 125, 127; of causal connections, 5–6; of causality, 2; of community, 202; complexity of, 123–28; human interference with, 124–25; interactions in, 125–26; managing, 124; myths about, 124; orderliness of, 124; perception of, 123–24 Neumann, R. P., 33–34 New Guinea highlands, 224–25; local knowledge studies of, 104–6

300 / Index

optimality models, for explaining human behavior, 37 peace, 207, 209 Peirce, Charles, on abduction, 30 perspective: anthropological, 109; ecological, 144–45; evolutionary, 150; fire’s, 70 Peru, deforestation in, 117 pesticides, 86, 90–91 pests. See insect pests Philippines, 85, 142; environmental change in, 135– 36; forestry/coastal resources policy in, 135; illustrations from, 96, 134–40; knowledge/ practice studies on mangrove tree planting in, 96–104, 134– 40; plagues of development for, 135 philosophy, 6 planting: of mangrove trees, 96– 104; success of, 104–5; test, 101 Poe, Edgar Allan, 31 polemics, confusion and, 229–37 policy, 193; Botkin on concerns of, 126; in Philippines, 135 policy making: Botkin and, 126; for fire-management, 64 political ecology, ix, 23n23, 36, 36n44, 45–46, 114; against, 129–42; applications of, 114; beginnings of, 129–30, 130n103; champions of, 33–34; criticisms of, 141; event ecology as alternative of, 133–34, 141; failures occurring in, 27–28; now, 129–33; programmatic alternative to, 131–32; studies on, 34

politics, environmental change as affected by, 131 polygynous systems, 180 Popper, Karl, 169 population pressure, 213–15 power, paradox of presidential, 201–2 pragmatism, 44 prawns, farming, 137 precognition, 197–98 predators: control, 124; prey and, 124 predictions, 170–72; of behavior, 172; confirmation of, 180; in DEA, 171–72, 172n130, 180–81; failed, 180–85 presidency, cycle of, 201–2 prey: choice model, 167, 170, 175–76; predators and, 124 prioritizing, 45 process(es): autonomy of, 195; causal explanation of, 203; concept of, 210; deforestation and, 110; designating, 192; events and making, 198–201, 210; evolution as historydependent, 187–88; global, 196; as independent of human agency, 193; innovation, 192; opposing study of, 191–92; phases of, 202–3; political, 192; reification, 201; reification problem and, 193–98, 193n146; response, 194; social science explanations and concepts of, 191–203; as substituted for events, 197; term as used, 192; vocabulary of, 197 properties, causally relevant, 4–5 proxy currencies, 164; DEA use of, 165–66

Index / 301

psychologists. See Darwinian psychologists Pyne, Stephen, on failure of perspective, 70 reasoning, rational practice, 38 regression models, sociologists’ use of, 17 regulation, community, 133n108 reification: excessive, 194; problem, 193–98, 193n146; process, 201 reinforcement, as mechanism, 212 reproductive success, 164, 188, 231 research: causal explanation as goal of, 1–47, 53n58; developing procedures for, 88; explanationoriented, 20–29; finer-grained forest fire, 68, 68n73, 73; fire, 52–53; forest-fire, 13–14, 51–52, 64–70, 68n73, 73; guided by, 45; Indonesian forest fires and distortions of priorities in, 51– 52; methods in developmental studies, 82; models’ role in, 173–74; procedures, 83–84; social science contribution to, 71; surveys used in, 83. See also interdisciplinary research; studies resolution, nonviolent, 235 resource(s): abuse, 114; common property, 179; control over forests’, 135–36; in forest fire management, 65; policy, 135. See also Department of Environmental and Natural Resources; World Resources Institute

rice fields: flooding, 92; pest infestation in, 93 Robarchek, Clayton, 235 Sayer, Jeffrey, on forest-fire research, 64 Schelling, T. C., 19, 174; causal model cited by, 19–20 schools, field, 89–91 science, 6, 12 sea otters, 126 settlement strategy, 199 significance, causal, 8 Snow, John, 17 social science: process concepts in explanations of, 191–203; research contribution by, 71 Society for the Conservation Biology, 73 sociobiology, 230; human, 146–47 sociology, 17; Parsonian, 210–11 statistical methods for, 15–17 stories: causal, 18–19, 178, 184; explanatory value of, 184 storytelling, 199–200 studies: assumptions as affecting, 86–87; of cultural systems, 104–5; developmental, 82; ethnographic, 81–82; HBE, 42; holistic, 80–82; local knowledge, 79–107; on political ecology, 34; unproductive, 81. See also ignition studies studying: adaptation, 171–72; humans, 145–46 success. See reproductive success surveys, 83 Sutton, Willie, 22, 223 Symons, D., 147; on Darwin and human affairs, 171–72;

302 / Index

Darwinian anthropology critiques by, 145–46 territorial conquests, 210, 212–13 textbooks, 36 theories: development of, 169; as explanatory, 169–70; as not the keys to explanation, 33, 33n38; as studied, 169 traits, 155, 159–60; cause of spread of, 178; development of, 188 tribes, 230 Tropical Deforestation: The Human Dimension (Sponsel), 109–21 unified field, 145–48 Vandermeer, John, 115 violence, environment and, 34–35 Waddell, E., 38, 105, 154–55 Walters, Bradley B.: on mangrove tree planting, 97–98; political ecology and, 129–42

war(s): actions in/related to, 231–32; anthropology of, 229–37; consequences of, 216; descriptions of, 229–30; enemy land as taken in, 211–12; entry into, 210; escalations and, 207– 8; explaining, 232; frequency of, 224–25; horrors of, 212–13; Kundagai-Ambrakui, 215–16; land as annexed following, 216–17; land shortages as cause of, 214–23; Maring, 206– 8, 206n156, 214–18; time of, 226–27; transition from peace to, 207; as viewed, 208, 210. See also conflicts Whitehead, Neil, 235 wildfires. See forest fires World Resources Institute (WRI), 115 World Wide Fund for Nature (WWF), 49 WRI. See World Resources Institute WWF. See World Wide Fund for Nature

About the Author

Andrew P. Vayda is professor emeritus of anthropology and ecology at Rutgers University. Formerly a professor at Columbia University, he has also taught at the University of Indonesia and other Indonesian universities and at the University of British Columbia. His publications include some hundred articles and several books. His research, often crossing disciplinary boundaries, has focused both on philosophical issues and on subjects ranging from warfare and migration to forest fires and insect pest management, and he has directed and participated in numerous research projects on people’s interactions with forests in Indonesia and Papua New Guinea. For much of his career, he has specialized in methodology and explanation at the interface between social and ecological science. He founded the journal Human Ecology and was its editor for five years. A festschrift in his honor, Against the Grain: The Vayda Tradition in Human Ecology and Ecological Anthropology was published in 2008.

303