Against Better Judgment: Akrasia in Anthropological Perspectives 9781805390015

Table of contents :
Contents
Introduction
1. Trigger Warnings: Danger, Desire and Declensions of the Will in Eating Disorders Treatment
2. Three Problems with the Addiction as Akrasia Thesis That Ethnography Can Solve
3. To Live Like ‘People’: Drinking and Weakness of Will among the Runa of the Ecuadorian Amazon
4. Prayer, Demons and Akratic Sublation
5. Troubleshooting Humans: Modelling the Pathways to Inertia, Backsliding and Moral Transgression on Indonesia’s Hypnotherapy Circuit
6. The ‘Replication’ of Caste as a Form of Collective Akrasia
7. Is Grit Irrational for Akratic Agents?
8. Relational Akrasia: Care and the Distribution of Action
Afterword. Akrasia in Its Social Context
Index

Citation preview

Against Better Judgment

WYSE Series in Social Anthropology Editors: James Laidlaw, William Wyse Professor of Social Anthropology, University of Cambridge, and Fellow of King’s College, Cambridge Joel Robbins, Sigrid Rausing Professor of Social Anthropology, University of Cambridge, and Fellow of Trinity College, Cambridge Social Anthropology is a vibrant discipline of relevance to many a­ reas – e­ conomics, politics, business, humanities, health and public policy. This series, published in association with the Cambridge William Wyse Chair in Social Anthropology, focuses on key interventions in Social Anthropology, based on innovative theory and research of relevance to contemporary social issues and debates. Former holders of the William Wyse Chair have included Meyer Fortes, Jack Goody, Ernest Gellner and Marilyn Strathern, all of whom have advanced the frontiers of the discipline. This series intends to develop and foster that tradition. Recent titles: Volume 14 Against Better Judgment: Akrasia in Anthropological Perspectives Edited by Patrick McKearney and Nicholas H.A. Evans Volume 13 New Perspectives on Moral Change: Anthropologists and Philosophers Engage with Transformations of Life Worlds Edited by Cecilie Eriksen and Nora Hämäläinen Volume 12 Where Is the Good in the World? Ethical Life between Social Theory and Philosophy Edited by David Henig, Anna Strhan and Joel Robbins Volume 11 Making Better Lives: Hope, Freedom and Home-Making among People Sleeping Rough in Paris Johannes Lenhard

Volume 10 Selfishness and Selflessness: New Approaches to Understanding Morality Edited by Linda L. Layne Volume 9 Becoming Vaishnava in an Ideal Vedic City John Fahy Volume 8 It Happens Among People: Resonances and Extensions of the Work of Fredrik Barth Edited by Keping Wu and Robert P. Weller Volume 7 Indeterminacy: Waste, Value, and the Imagination Edited by Catherine Alexander and Andrew Sanchez Volume 6 After Difference: Queer Activism in Italy and Anthropological Theory Paolo Heywood

For a full volume listing, please see the series page on our website: https://www.berghahnbooks.com​ /series/wyse

Against Better Judgment Akrasia in Anthropological Perspectives

Edited by Patrick McKearney and Nicholas H.A. Evans

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2023 by Berghahn Books www.berghahnbooks.com © 2023 Patrick McKearney and Nicholas H.A. Evans All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. Library of Congress Cataloging-in-Publication Data Names: McKearney, Patrick, editor. | Evans, Nicholas H. A., editor. Title: Against better judgment : akrasia in anthropological perspectives / edited by Patrick McKearney and Nicholas H. A. Evans. Description: New York : Berghahn Books, 2023. | Series: Wyse series in social anthropology ; 14 | Includes bibliographical references and index. Identifiers: LCCN 2023000713 (print) | LCCN 2023000714 (ebook) | ISBN 9781805390008 (hardback) | ISBN 9781805390015 (ebook) Subjects: LCSH: Judgment. | Akrasia. | Ethics. Classification: LCC BF447 .A44 2023 (print) | LCC BF447 (ebook) | DDC 153.4/6--dc23/eng/20230405 LC record available at https://lccn.loc.gov/2023000713 LC ebook record available at https://lccn.loc.gov/2023000714 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978-1-80539-000-8 hardback ISBN 978-1-80539-001-5 ebook https://doi.org/10.3167/9781805390008

Contents

Introduction1 Patrick McKearney and Nicholas H.A. Evans 1. Trigger Warnings: Danger, Desire and Declensions of the Will in Eating Disorders Treatment34 Rebecca J. Lester 2. Three Problems with the Addiction as Akrasia Thesis That Ethnography Can Solve50 Darin Weinberg 3. To Live Like ‘People’: Drinking and Weakness of Will among the Runa of the Ecuadorian Amazon70 Francesca Mezzenzana 4. Prayer, Demons and Akratic Sublation89 Jon Bialecki 5. Troubleshooting Humans: Modelling the Pathways to Inertia, Backsliding and Moral Transgression on Indonesia’s Hypnotherapy Circuit102 Nicholas J. Long 6. The ‘Replication’ of Caste as a Form of Collective Akrasia126 Ivan Deschenaux 7. Is Grit Irrational for Akratic Agents?146 Lubomira Radoilska 8. Relational Akrasia: Care and the Distribution of Action169 Patrick McKearney

vi

Contents

Afterword. Akrasia in Its Social Context189 Richard Holton Index196

Introduction Patrick McKearney and Nicholas H.A. Evans

A man in the United States resolves to overcome his addiction to drugs yet returns to doing them almost immediately (Weinberg, this volume). Members of a low-ranking caste in Nepal believe caste discrimination to be wrong in all instances, but nevertheless mete it out on the caste closest to them (Descheneaux, this volume). A Muslim woman in Indonesia asks a male hypnotherapist to help her become more pious, but when she wakes from the treatment, she makes sexual advances to him (Long, this volume). A man in the Ecuadorian Amazon declares drinking alcohol to be immoral yet hosts the whole village for a drinking party the next day and becomes stupefyingly inebriated (Mezzenzana, this volume). A young woman in the United States diagnosed as having an eating disorder resolves to maintain her health, then refrains from eating for days (Lester, this volume). Do we think consistently about what it is best to do? Do we always resolve to do what we think it is best to do? Do we always act in line with our resolutions? Philosophers have long debated such questions through the concept of akrasia, an ancient Greek word often glossed into modern English as ‘weakness of will’, but also sometimes translated as ‘moral incontinence’. This ‘weakness of will’ can refer both to the way in which a person behaves, or to the aspect of character that caused them to behave in that way. For the ancient Greeks, an akratic person acted in a way that was contrary to their own best judgment. They either acted contrary to what they thought was best, or they lacked the self-control (the continence) to act on their better judgment. It is for this reason that the akratic person is often described as ‘morally incontinent’. At its heart, the philosophical debate about akrasia is about the nature of the inconsistencies between a person’s judgments, intentions and actions. It is about whether people can act in ways that are contrary to their better judgments. In this introduction, we show that social scientists ask questions that parallel this philosophical debate. Social scientists have sought to understand why people believe in inconsistent things, why they resolve to act in ways that

2

Patrick McKearney and Nicholas H.A. Evans

are contrary to their best interests, and why they fail to act in line with their resolutions. Nonetheless, we also show that in pursuing these questions, social scientists have only rarely followed in the footsteps of philosophers to examine what it means for there to be inconsistencies between peoples’ judgments, intentions and actions. The result, we argue, is a dominant trend of assuming that human beings mostly (if not always) act according to what they think it is best to do. This might seem like an innocent assumption to make, but as we show in the rest of this introduction, it has had serious consequences for social scientists’ abilities to understand the social worlds that they study. A key aim of this volume is to show that the above applies just as much to anthropology as it does to other disciplines. The dominant intellectual direction of anthropology has been to assume that the thing worthy of study is the variation between people. Because anthropologists have tended to assume that people act in ways that are coherent with their broader socio-cultural position, the actual relationship between people’s judgments about how they should act and their actions has not been taken as a subject of ethnographic study. Coupled with a desire to show that people act ‘rationally’ within their social worlds, this has led to a widespread ethnographic neglect of situations in which those people feel that their own judgments, intentions and actions are inconsistent. Indeed, we argue that it has led to an implicit denial that inconsistency between action and judgment is really possible. This denial directly echoes a controversial philosophical p ­ osition – fi ­ rst expressed by Plato’s ­Socrates – ­that akrasia does not exist. Socrates argued that no one truly acts against what they think is best because people always pursue what they see to be the good. Anthropologists’ commitment to explaining what people do in terms of their own specific reasons unwittingly tethers the discipline to this contested argument. Moreover, it rules out further investigation into how humans bring together judgments and actions, and whether this might be culturally variable. This ‘Socratic’ approach to a­ krasia – e­ specially when unacknowledged, unexamined, and ­uncontested – ­renders the discipline less able to grasp the complexities of inner life. The discipline is theoretically and empirically impoverished as a result. In this volume, we challenge the model of human action that has been baked into our discipline by our neglect of the question of akrasia. We directly confront anthropology’s position with the philosophical challenges to the Socratic argument. For over two millennia, philosophers have contended that the human psyche is more complex than Plato’s Socrates maintained, and that there is a less straightforward relationship between our judgments, intentions and actions than he suggested. We pose these same challenges to anthropology and to the social sciences more broadly. What would it take for anthropology to consider the possibility that the relationships between judgment, intention and action regularly break down? How would this change the practice of anthropology? And what enduring ethnographic and theoretical puzzles would it allow us to solve?



Introduction

3

Together with our contributors, we ask these questions ethnographically. We describe situations where people seem to act contrary to their own judgments, and we attend to how they and others evaluate their actions. We do not resort to the normal anthropological trick of using contextual evidence to reinforce our assumption that people never act against their own better judgments. Instead, we raise questions about whether the people in question really might be acting inconsistently, and whether (as Lubomira Radoilska argues in her chapter) they understand themselves to be acting inconsistently. Do people act contrary to their better judgments? What are the important differences between the various ways in which they might do that? And, what does this mean for our understanding of the human? Asking these questions is essential to building an anthropology able to research, debate and understand the terrain of inner struggle, contradiction and inconsistency. This, in turn, gives us new ways to approach central problems in social science. We contend that developing anthropological attention to akrasia also offers something distinctive back to the philosophical debate. As Lubomira Radoilska and Richard Holton both demonstrate in this volume, this kind of ethnographic investigation confronts the philosophical discussion with the richness, complexity and diversity of actual individuals’ interior and social lives, as well as the full range of their variation across cultures. The philosophical debate is too important to be conducted without detailed ethnographic evidence of how people judge, how that relates to what they do, how they understand seeming discrepancies between their judgments and their actions, and the extent to which they take apparent inconsistencies to be a moral or practical problem at all. The ethnographic evidence we present in this volume suggests that these issues are even more complex in practice than the already subtle debate has considered, and that they vary across the world in ways that a culturally specific philosophical discussion has not yet comprehended. We do not, however, exclusively offer this evidence as empirical information to be incorporated within existing philosophical theories; we also suggest that trying to construct a universal theory about the psychological properties of individual judgment and action may be the wrong way to go altogether. Philosophy’s way of debating akrasia without considering cultural variation ignores the possibility that akrasia is relational and social to the point that its content, form and even existence could vary between relationships and across societies. Our aim, in this introduction, is to repair the anthropological neglect of akrasia. We outline the assumed explanatory model of human action that produces this neglect, locate it as a particular position within the akrasia debate, and bring it into the full light of the philosophical debate. We start by detailing anthropology’s approach to a related question: how rational is human judgment in the first place? This question is always in the background of the akrasia debate and of central importance to closely related debates in philosophy. It is a question to which anthropology has clear answers because

4

Patrick McKearney and Nicholas H.A. Evans

it has played a very important part in the discipline’s history. As we show, this explicit anthropological argument against irrationality within people’s thinking has also led to an implicit neglect, within the discipline, of the possibility that their judgments, intentions and actions do not line up.

Consistently Judging Well Do humans think and judge consistently? In this section, we demonstrate how and why anthropology answers this question affirmatively by contrasting it with a certain kind of moral philosophy. We argue that this answer makes anthropology a ‘science of consistency’ and that this has important implications for its capacity to understand and investigate akrasia. It is ordinary for many people in many societies to eat meat. Many people see it as perfectly legitimate to kill animals such as pigs for the purpose of consuming them. Yet, it is also common for such meat-eaters to treat other animals quite differently. In Europe and North America, for instance, many people who eat pork also keep pets such as cats and dogs. They typically experience a genuine sense of loss when their dogs die and are morally and physically repulsed by the thought of killing or eating them. It is perfectly ordinary to combine these two ways of treating animals. Such people are lovers of meat and lovers of dogs. But they are not lovers of dog meat. Moral philosopher Peter Singer takes this activity as irrational in that it is inconsistent in two senses: externally and internally (Singer 2011; for more on the distinctions see Wilson 1974; Tambiah 1990). First, externally, it does not correspond with the way the moral world actually is. For Singer, the morally correct thing to do is that which prevents the most suffering. Greater mental capacity, he argues, enables living beings to suffer more (Singer 2010). Both pigs and dogs, as intelligent animals, deserve our r­ espect – i­t is cruel to kill both species. For Singer, the judgment that it is morally legitimate to eat animals is irrational because it is just not true. It is inconsistent with the objective moral truth. Such people have judged incorrectly. However, Singer also takes this activity as internally inconsistent. Regardless of the objective moral validity of eating meat and keeping pets, they are incompatible to Singer. How is it consistent, in one moment, to treat killing a dog as immoral, and yet, in the next, to slaughter a pig for meat? There is, for Singer, a logical contradiction between these two ways of behaving. If you are concerned about your dog’s potential to suffer, you are, by implication, committed to being concerned about the capacity of other living beings of equal or greater mental ability to suffer when they are killed for meat. Singer sees such people as not only having judged incorrectly in relation to how the world is, but as judging badly in that their judgments do not properly relate to one another. Both of these judgments are related to Singer’s view that moral life has objective rules that are out there in the world, irrespective of the kind of social



Introduction

5

relations one has, and that all these rules add up so that individual agents can follow them in a logical way. Thinking morally thus requires us to step out of our conventional ways of engaging with the world to see things as they truly ­are – a­ process that necessarily involves making our moral thinking more internally consistent too. As a utilitarian, Singer’s way of distancing himself from the biases of people’s ordinary moral thinking takes the form of trying objectively to calculate how much suffering is involved. The famous utilitarian judgment, for instance, that it is irrational to save your mother, rather than an eminent reforming bishop, from a burning building, relies on abstracting morality into the general principle of maximizing utility. The impersonal mathematical simplicity provides a yardstick against which to judge, from the outside, the contradictions of people’s supposedly unsystematized everyday judgments. Singer’s utilitarianism is not the only show in town within Anglophone moral philosophy. But the other ones replicate this practice of standing back from everyday practice to judge its validity and address its internal ­contradictions – a­ mode of philosophizing which is normal and widespread within this tradition (Banner 2014). Utilitarianism is just one of the tools that can be used to perform this operation, with Kantian categorical imperatives being another example. Kantianism is also an example of another dominant theory in this tradition that similarly holds that there are no real contradictions between moral values. This kind of moral philosophy thus takes it as a given that moral obligations and moral truths exist objectively, that they are consistent with one another, and that the task of philosophy is to identify and criticize the errors and inconsistencies of everyday thought and practice. Our job, as moral agents, is to submit ourselves to this kind of rational inquiry so that we can make judgments that are more consistent with the way the world is and with one another. This is a distinct way of seeing ethics and the task of philosophy. Moral philosophers from other traditions do not share this view, and the tradition from which contemporary Anglophone moral philosophy stems has not always shared it either. There are also other versions of contemporary Anglophone moral p ­ hilosophy – ­clustered around the tradition of virtue ethics, but not confined to ­it – t­ hat contest the idea that moral life has objective rules, that all moral values cohere, and that the task of philosophy is to root out inconsistency in everyday thought and practice (Anscombe 1958; MacIntyre 2007; Nussbaum 1986, 1990; Williams 2011, 1993b, 1981; Laidlaw and McKearney 2023). Other branches of Anglophone philosophy also regularly operate in a far less judgmental and much more descriptive mode. The akrasia debate itself takes place between the philosophies of ethics, mind and action. The aim of the debate is to understand what humans actually think and ­do – ­principally, how consistent the relationship between judgment and action actually is in human l­ife – r­ ather than to pronounce upon how consistently they should think and act.

6

Patrick McKearney and Nicholas H.A. Evans

Anthropology is a natural ally of this latter kind of investigation, into how humans actually think and behave in their ethical lives, for two reasons: first, because of its long tradition of opposing external judgments; and second because of its distinctive interest in how people themselves understand and engage in their social worlds. The discipline of anthropology emerged from colonial encounters in which Euro-Americans often judged the thought and practice of subject populations as objectively wrong. Europeans confronted ‘tribal’ societies with beliefs or practices (witchcraft, animism, cannibalism, magic) that they then assessed as obviously mistaken or morally wrong. Early ethnographic evidence reinforced the assumption that such populations arrived at wrong conclusions and did so through irrational thinking. The fact that logical thinking was taken within European societies to be definitional of a functioning mind, and indeed of humanity altogether, made these judgments central to racist arguments about the superiority of white and European races and the nonhuman status of others (Larsen and King 2018). Some early anthropologists reproduced these judgments and tried to explain why people in such societies thought in this way. They argued that societies differed in how evolved they ­were – ­contemporary European civilisations having reached a higher state after having passed through the earlier stages these other societies were still stuck in. They argued that the state of evolution a society was in directly shaped the way that the people in it thought. At earlier stages in evolution, people were unable to think logically, consistently, and scientifically to the point that they were unable to understand the relationship between cause and effect, or the difference between fact and fiction. Anthropology as we know it today was founded on the resistance to these explanations and the judgments of inconsistency they relied upon. Anthropologists argued that such beliefs were not, as they seemed, evidence that people were making mistakes, nor that they thought in ways that were less consistent and more ‘primitive’ than Euro-Americans. British anthropologist E.E. Evans-Pritchard (1976), for instance, studied the witchcraft beliefs of the Azande of Southern Sudan in the early twentieth century. It seemed impossible to deny that the Azande belief that witches caused misfortunes, illnesses and deaths was an error born of illogical thinking. The Azande maintained that when someone’s toe became infected, a witch had done it; that when a granary collapsed and killed the people sleeping under it, it must be witchcraft; and that when anyone died, a witch must be responsible. The Azande did seem to know that natural causes could explain how a toe could become infected, a granary could collapse and people could die. They obviously understood that in each of these cases, someone had stubbed their toe, termites had weakened the wood and the person had died from illness. However, the fact that the Azande also maintained, in the face of that empirical evidence, their belief that witchcraft caused these events suggested that



Introduction

7

they were incapable of systematic scientific thinking about cause and effect, and that they could not identify or reconcile inconsistencies between their judgments. Evans-Pritchard demonstrated that this external assessment of the Azande was empirically wrong. He showed that the Azande’s witchcraft beliefs and practices did not explain gaps in their empirical understanding of cause and effect, which were perfectly well developed and not compromised by their belief that witchcraft was also at play. This was because witchcraft was not an answer to how this event happened that contradicted or replaced the naturalistic one. It answered a different question; namely, ‘why did it happen to these people now?’ motivated by the further question, ‘who should be held responsible?’ What looked to be two inconsistent beliefs turned out to be quite consistent in practice. It is unlikely there can be a scientific justification for asking such a ‘why?’ question as the Azande did, or for when and how we hold others responsible. And that may speak as much to the limits of science’s capacity to address the full range of questions raised by individual and social existence, as it does to any supposed irrationality on the Azande’s part. Furthermore, the fact, and the way, that the Azande asked that why question made it hard to maintain they were in any way less intelligent or consistent in their thinking. When pressed by Evans-Pritchard, the Azande articulated sophisticated arguments about why this question was worth asking: because people sleep under granaries all the time without dying, and granaries collapse all the time without killing people. There was something to explain, they contended, that EvansPritchard’s way of thinking offered no answer to. Indeed, the Azande turned out to be far more systematic and curious than their western counterparts in pursuing these why questions. A European might have initially imagined that witchcraft beliefs were a way for the Azande to hold up their hands, in the face of the limitations of their scientific explanations for things like buildings collapsing, and say, ‘It just happens, we cannot know why!’ But Evans-Pritchard demonstrated that it was ironically the Europeans who, in the face of the question as to why the building collapsed when it did, had no explanations to o ­ ffer – a­ nd it was the so-called ‘primitives’ who rigorously questioned, searched for evidence and produced complex, worked out explanations. This is one among many ways in which anthropologists find rationality, meaningfulness and consistency in what people think and do. Later anthropologists, for example, were concerned to understand the effects of modern capitalist transformations on the populations they studied without reducing them to mere bystanders or by-products of these processes. They brought anthropology’s tradition of making sense of strange beliefs and practices together with a broadly Marxist framework in order to read beliefs in supernatural dangers (such as witchcraft) as ways of noticing, articulating and countering the violence, strangeness, and distorted nature of the structural market

8

Patrick McKearney and Nicholas H.A. Evans

forces such people face (Taussig 1977; Moore and Sanders 2002; Comaroff and Comaroff 1999; Graeber 2011). These anthropologists thus averred that such beliefs were perfectly consistent and coherent because they rationally comprehended a very real vulnerability in the face of capitalism. Anthropology’s tradition of showing consistency underneath seeming inconsistencies goes hand in hand with the a­ ssumption – ­in stark contrast to that of S ­ inger – t­ hat social life has patterns within it that can be uncovered. This is how anthropologists make sense of seeming inconsistencies or irrationalities as actually coherent, sensible and culturally particular thoughts and actions. Anthropologists, for instance, do not conclude that treating your mother differently from a bishop is evidence of thinking inconsistently. Instead, they demonstrate that ‘mother’ is an important category within a broader kinship system through which relations in a society are differentiated and structured. It is thus only inconsistent to save the mother from the burning building when you apply an external measurement that deliberately removes these patterns. When we leave the patterns in, saving the mother is a perfectly coherent thing to do that is consistent with how one lives out one’s obligations to differentiated relations. Your job, as an anthropologist, is to understand the context enough for you to see the consistency. Judgments of inconsistency very rarely acquire authority within anthropological d ­ iscussion – ­whereas demonstrations of the consistency of other people’s thoughts very frequently do.1 Let us apply this logic to the example of the pet-keeping meat-eaters. Singer judges this as evidence of inconsistent thinking. Another way of thinking, however, is to show that it only seems that there is inconsistency. In fact, people have a consistent judgment that just does not show itself immediately. Anthropologists have shown that societies classify animals into different categories, and that different societies undertake that classificatory work very differently (White and Candea 2018; Douglas 2008). The English, for example, make a distinction between animals that can be companions and w ­ orkers – ­such as dogs and ­horses – ­and those they raise for consumption. The way a society makes its divisions can also be used to draw further divisions between a society and another, such as between French people who eat horse and British people who do not consider it meat. The assumption, in Singer’s case, is that intelligent animals should be treated well and that, given dogs and pigs are both intelligent animals, people are being inconsistent in treating them differently. For Singer, when people pet their dogs, they are committing to treating intelligent animals well; a commitment they betray when they eat pork. Edmund Leach (1989), by contrast, argued that there is a more important contrast at work in English social life: between humans and food. Many animals fall into the cultural category of legitimate food. But, as ‘man’s best friend’, a dog is regarded as closer to humans than to edible animals. Thus, despite the fact that dogs are literally edible, they cannot be food for English people any more than other humans



Introduction

9

can (cannibalism and dog-eating both being considered barbarous, although probably not equally). It is not like the Jewish case where pork is considered food but prohibited as not kosher, but rather that dogs are not taken as food at all (Leach 1989, 154). Petting a dog is thus based on a separation of dogs from edible animals. This separation is culturally distinct and thus contingent, but no less internally consistent for that. It is not that people are incapable of seeing the connection Singer wants to draw, but that they categorize the world differently. English people generally love their dogs and eat their pig meat, without ever mixing up the two practices. This is, indeed, not consistent with the specific proposition that Singer has imposed: to treat intelligent animals well. But that does not mean that there is actually an inconsistency here in people’s own beliefs or practices here. They seem inconsistent only when one does not understand the patterns of their thought. This practice of finding consistency was developed, in anthropology, alongside the ‘doctrine of the psychic unity of mankind’: the idea that all humans share the same cognitive capacities. This theory was originally developed to attack the idea that different ‘races’ had different levels of intelligence that stemmed from genuine biological distinctions (Larsen and King 2018). The side of psychic unity won the day in anthropology, and the doctrine has become central to anthropology’s disciplinary identity ever since. Anthropologists assume that the people they study are not stupid, irrational, or unreflective and that Euro-Americans of any stripe, even if they are philosophers, scientists or anthropologists, are not able to think more logically or consistently (Shore 2000; Geertz 1975; see also McKearney and Zoanni 2018). This commitment to the equality of human intelligence became closely tied to the demonstration of consistency, as can be seen by the fact that alternative ways of arguing for equality were not pursued. French anthropologist Lucien Lévy-Bruhl (2015), for instance, proposed a way of comprehending the thought of ‘primitive’ populations different from Evans-Pritchard’s. He argued that ‘primitive’ thought was nonlogical and inconsistent but not because it was a bad or incomplete version of scientific thought. Rather, it was something quite ­different – c­ loser to art and ritual than to empirical explanation (Tambiah 1990). Read sympathetically, Lévy-Bruhl was looking for a different way to resist the argument about the superiority of white races and EuroAmerican t­ hought – ­one that enabled us to see the value and sense of other ways of thinking, without having to accept the value that Euro-Americans placed on rationality. Evans-Pritchard (1934; 1965), however, famously responded that this argument did not do justice to the intelligence, practicality, and consistency evident in the thought of these supposedly ‘primitive’ people. Many contemporaries were won over by Evans-Pritchard’s argument that Lévy-Bruhl sailed far too close to racist waters by conceding too much to the idea that nonEuropeans might not display logic. For Evans-Pritchard, the best way to resist

10

Patrick McKearney and Nicholas H.A. Evans

external judgments of seemingly exotic non-European thought and practice was through an explicitly ‘intellectualist’ approach that emphasized the rationality of these populations (1933). That said, most anthropologists do not generally treat humans as exhibiting consistency in the rationalistic and propositional way that a philosopher might imagine, or that the term ‘intellectualist’ might suggest. Leach’s demonstration that English people are quite consistent in their relationships with animals, for instance, does not rely on those people stating consistent philosophical arguments about the relationship between different beliefs. Anthropologists generally investigate what people think in a more holistic way than through attention solely to the kind of propositional statements they make in response to direct questions. They tend to rely only very partially on interviews, and almost never on questionnaire responses to questions of belief. Leach found consistency, instead, in the thoughts embedded in the different things people do, the kind of contextual claims they made in those diverse practices, and in the typically implicit relationship between them. Evans-Pritchard’s demonstration of the Azande’s coherence similarly only worked because it attended to the social contexts in which the Azande’s claims were made. It did not demonstrate that the Azande thought consistently in exactly the same way as a Euro-American scientist might. It undermined the idea that the Azande should think like that when they were handling practical difficulties, conflicts with their neighbours, and the loss of their kin. The ‘rationality debate’ about Evans-Pritchard’s Azande material further took aim at the idealized picture of European thought that was implicit in the judgment that they were thinking inconsistently. Scholars pointed out that the comparison between European scientific thought and Azande everyday interactions with their neighbours, local medical treatments, and ritual practices was bound to make Europeans seem more rational (Evans-Pritchard 1965; Wilson 1974; Tambiah 1990). By contrast, an anthropological approach to actual European thought would reveal that it also did not accord to the standards by which the Azande were being j­udged – ­and thus, that the flaw was in the measuring device, not the thing being measured. Fellow British anthropologist Mary Douglas (1980) argued that EvansPritchard treated thought, above all, as fundamentally ‘institutional’. She developed this into the claim that all societies must, by necessity, invent ­‘entities’  to  distribute responsibility. European civilisations, for example, created the eminently contingent and debatable idea of IQ to explain and regulate misfortunes, incapacities and violations through a particular kind of population classification. This is just one demonstration of the general point that all societies have a cosmology or a theology implicit in their thought and practice and that all social action is value-laden. Such values do not and cannot have the kind of scientific objectivity and neutrality that ideals of rationality typically measure us against. Similarly, no society escapes the kinds of questions about misfortune, and the potential for others to cause it



Introduction

11

through ill-will, that the Azande used witchcraft to answer (Douglas 1973; 2008). Indeed, subsequent anthropologists demonstrated that such questions about who is to be held responsible for misfortunes are always contested within societies, and that they also vary across historical periods and social contexts (Gluckman 1955; Douglas 1980; 2006; Laidlaw 2013; Danziger 2006; Evans 2016; McKearney 2022). These arguments are just some of the ways that anthropologists have softened a certain strong and intellectual interpretation of the commitment to consistency. Indeed, the history of anthropology can be read as an attempt to incorporate more of the messiness and inconsistency of individual and social life into its models, without risking the fundamental commitment to equal intelligence in the way that Lévy-Bruhl was seen to do. The once dominant schools of structural-functionalism and structuralism, for instance, lost ground when others argued that social life was not nearly so socially cohesive or symbolically coherent as these schools suggested. Many anthropologists have since adopted a more processual picture of social life as something that people do, rather than something merely imposed upon them (Laidlaw 2013). And this goes with a greater attention to particular individuals; the ways in which they develop and change, how they experience social life as fragmented and contradictory, and the role of their own character and decisions in shaping the lives they lead (Briggs 1999; Humphrey 2008; Bourdieu 1998). The anthropological focus on the realities of everyday life directs us to focus on the ways in which humans are not perfectly rational beings, but instead creatures of desires, fragmented between the pull of multiple moral registers and multiple competing obligations (Briggs 1999; Humphrey 2008; Schielke 2009a; Mayblin 2017; Stevenson 2014; Robbins 2013b). There is an important history of debate around precisely this theme, and there is a significant, important and growing diversity of movements in contemporary anthropology that attempt to do just this (Berlant 2012; Laidlaw 2013; Das 2007; Mattingly 2014; Garcia 2014). Some scholars take this emphasis yet further by arguing that certain contemporary social conditions lead people’s own subjectivities to be fragmentary and disordered (Biehl, Good and Kleinman 2007; Berlant 2012). The result is that there is no disciplinary consensus in contemporary anthropology around what kind of consistency we should look for in the lives of our informants so as to dignify their intelligence and make sense of their thought and behaviour (Evans 2020). Anthropology now holds its commitment to finding consistency alongside an intellectual interest in going in precisely the opposite direction. It is, however, notable that few anthropologists have used this seriously to attack the idea that humans judge consistently. And the practice of dignifying informants by demonstrating how their seemingly contradictory judgments actually hang together remains just as s­ trong – ­most strikingly in those intellectual traditions most inclined to emphasize fragmentation and disorder (e.g. Biehl 2005). Anthropology is explicitly committed to

12

Patrick McKearney and Nicholas H.A. Evans

avoiding the idea that people make poor judgments, and this has only become more important with successive attempts to decolonize our discipline and recognize the inequalities that structure social life. None of this interest in messiness, therefore, makes anthropology any more disposed to external judgments about the inconsistency and irrationality of the people they work with. Making an external judgment of other people’s logical and moral thought assumes that correct logical and moral judgment is always the same everywhere, that one knows what correct moral thinking is, and that one knows enough about how other people think and behave to judge it to fall short of these standards. Anthropologists, by contrast, continue to challenge the idea that any individual or society might have somehow overcome culturally particular thought patterns to the point that they might think in a way that is more neutral, logical, or consistent than others. Anthropological work still consistently undermines the superiority and distance of the position from which one could judge externally, or the judgment that another pattern of thought and practice is wrong and inconsistent (McKearney 2016; Robbins 2020b). And so anthropology’s foundational assumption that all humans are of equal intelligence continues to make it a science of consistency to its core (Robbins 2007).

Consistently Acting on Judgments The anthropological commitment to finding consistency within people’s judgments has strong echoes of Socrates’ argument that there is also consistency between people’s judgments and their actions. Socrates denied that akrasia exists, in much the same way as anthropology has repeatedly aimed to demonstrate that people do not think inconsistently. In this section, we develop this analogy into the claim that anthropology implicitly (and largely unwittingly) follows Socrates’ denial of a­ krasia – ­and that this places significant limitations on its capacity to understand human thought and action. The Socratic argument that akrasia does not occur in human life relies upon the idea that we act consistently with our judgments. Suppose you are presented with an everyday decision such as whether or not to eat a burger. These are mutually incompatible courses of action. Many different things might incline you to refuse the burger: reducing the effect of livestock farming on the environment, fitting in with a vegetarian group of friends, wanting to lose weight. Many other things might incline you to eat it: the taste of it, your own need for protein and iron, deliberately showing your vegetarian friends you are not like them. But you must do one of the two: you cannot, logically, do both. And so you reconcile the inconsistencies between your different judgments to produce a single ‘best judgment’ about what to do. It is important to clarify, at this stage, the meaning of ‘best judgment’ within the akrasia debate. The term does not refer to a judgment that anyone



Introduction

13

else externally judges to be objectively correct. It simply refers to what the person judges, all things considered, to be the best thing to do. It is important that ‘best’ here does not have the specifically moral sense as implied by the word ‘good’. It might well be, for example, that I decide the thing for me to do is give moral considerations far less weight and prioritize my own desires: thus, I judge it is best for me to eat the burger. That is why Socrates, according to what we know of his thought through Plato’s writings, assumes that once we have arrived at such a judgment, we will follow it in our actions. When there is consistency among a person’s judgments, he maintained, there will be a consistency between judgment and action. ‘Nobody’, he famously said in the Protagoras, ‘errs willingly’ (Bobonich and Destrée 2007, xvi). People, Socrates claimed, act upon their perceptions of the good. To do something is to be internally committed to thinking it is good: that is a necessary condition for acting. Put differently, actions are the external expression of our judgments. A bad act can only ‘be done in ignorance, under the false belief that it is for the best’ (Lear 1988, 175). Let us imagine, for instance, that we go on not to eat the burger, even though we said to ourselves that it would be best if we did. This action seems akratic. But, according to Socrates, this is not actually possible. Actions are the external outworking of our desires. And we cannot really desire anything if we do not think that it is good. Our decision not to eat the hamburger might seem to go against what we judged best. But Socrates would argue that this can only be because that is not what we actually judged. Perhaps the moral considerations were much more important to us than we realized, and we judged deeper down that not eating the burger was the thing to do. No action, he claims, is possible unless we actually believe it worth d ­ oing – b ­ elieve it the best thing to do. So it must be that we only seemed to be convinced by the reasons to eat the burger, but we were still governed by another judgment. This amounts to a denial that akrasia exists. Socrates, and many who follow him in denying akrasia’s possibility, hold an ‘internalist’ position on the relationship between judgment and action (Bratman 1979; Stroud and Tappolet 2003; Stroud 2014). What this means is that they hold there is a necessary rather than contingent connection between practical judgments and action. The position of the modern philosopher R.M. Hare, for example, was that when a person makes a moral judgment that they ought to follow a particular path of action, they are in effect assenting to do so (Stroud and Tappolet 2003, 2). That is, Hare thought that if a person seriously says to themselves, ‘I ought to do x’, then they are in effect making a resolution to actually do that thing (Stroud 2014). Social scientists have not taken up explicit positions on these issues in the same way these philosophers have done. The idea that humans do not always judge consistently has always been a live one in the social sciences and in anthropology. That is why there have always been such strong arguments in favour of the idea that humans think consistently, as well as where there are

14

Patrick McKearney and Nicholas H.A. Evans

nuanced debates about this topic within contemporary anthropology. But the idea that humans do not do what they judge best has never even been acknowledged, let alone properly considered, within the social sciences and anthropology. Most of these disciplines have little to say about akrasia and show little prospect of developing a more complex picture of the relationship between judgment and action. This does not, however, mean that the social sciences make no assumptions about this aspect of human life. More often, social scientists have attempted to explain away seeming inconsistencies between judgment and actions, thus betraying an implicitly ‘internalist’ take that akrasia is not possible. This is particularly evident in two of the most influential explanatory paradigms in the social sciences. The first a­ pproach – ­dominant in certain kinds of economic, psychological and quantitative ­analysis – ­assumes that people are individual rational agents pursuing their own good. Such a view basically follows a Socratic line of reasoning that people ultimately do what they judge to be in their best interests. The second approach emphasizes the force of structural factors over individual lives. Sometimes this type of explanation makes no reference to people’s own understanding of their actions at all but shows, instead, how action is effectively determined by their social location and formation. (This approach often aims to show, by contrast, what people’s best interests really are by showing how their actions are determined by something other than their own free reflection). Neither of these two broad approaches investigates whether there is anything culturally variable about people’s interior lives, nor anything more complex about the relationship between the way they judge a situation and the way they act within it (Laidlaw 2013). These two explanatory paradigms are simply two sides of an opposition between the freedom of the rational agent and structural imposition. In fact, these paradigms can be brought together in the same model. Marxist and Peasant studies scholarship from the 1970s onwards, for example, combined these poles in the same way to explain why people do things that seem to go against their own i­nterests – ­such as supporting social systems that seem to oppress them. This work argued that people often intelligently resist their predicaments, only in ways that are not obvious or comprehensible from the outset when we do not understand either the structural forces or the subtle forms of resistance to them. The sociologist Pierre Bourdieu’s work represented a particularly developed form of this argument. His theories did not assume an equal playing field in the way economics does, but rather showed how people’s decisions are strategies for navigating situations in which their choices are confined and limited by hierarchical social games. This approach explicitly recognized how structural and cultural factors shape and limit the decisions people have available to them, while it also treated these people as rationally choosing between available options in the way an economist might imagine (Bourdieu 1984).



Introduction

15

Marxist scholars have also sometimes proposed variations upon the ‘false consciousness’ theme; that is, that people’s social circumstances have forced or indoctrinated them into a viewpoint that is directly against (what the analyst thinks is) their real interests (Godelier 1977). This explanation seems to seriously consider the idea that social forces might influence people’s interior lives; however, it does so only by showing people thinking rationally with the information given to them and doing what they think is rational to do. Marxist approaches thus preserve the same Socratic commitment to seeing people’s actions as pursuing what they see as good, but they do so by turning people into suffering ‘victims’ of larger forces (Robbins 2013a; Wilk 2001). Marxist theories thus continue to assume that all action is internally consistent with what people think is in their best interests, while simultaneously maintaining the authority of the analyst to point out that this is not what is really best for them. In this volume, Darin Weinberg demonstrates, for instance, that the social sciences have interpreted addiction as either a purely rational choice, or as behaviour that is not self-directed but rather determined by social forces outside an individual’s control. This simplistic duality is not confined to the study of addiction. It is not uncommon for a social scientist to try to rescue people from their own sense that they are acting in ways that do not express what they think they ought to do by showing that their actions are either rational or not actually free. The social sciences have traditionally focused on these two options. The reason for this, we argue, is because most social scientists implicitly adopt a Socratic position on akrasia that people act in line with their judgments of what is best, even if the actual question of akrasia is never made explicit. More often than not, social theory works only because it is premised on an implicit denial that akrasia might be up for ­debate – ­by excluding it as a possibility from the off. Anthropology’s traditional interest in the diverse ways that people actually experience their own worlds should have led its practitioners to be more interested in the quotidian complexities of human life. Its method of participantobservation is even centred around paying careful attention to the differences between what people do, what they feel they should do, and what they say they do (Malinowski 1987). In these conditions, an ethnographic and theoretical exploration of akrasia might well have taken off. We propose that the reason it did not is because of most anthropologists’ deep commitment to consistency as a way of understanding social life and as a way of dignifying the people they study. This neglect of akrasia as a possibility is even true for what might be the most concerted effort to recognize the complexities of human ethical life. The anthropology of ethics challenges the trends in social science that we have identified as preventing a deeper investigation of the complexities of people’s interior lives (Laidlaw 2002; Zigon 2009; Mattingly 2014; Faubion 2001). Central to this movement is a rethinking of the nature of agency itself. Many

16

Patrick McKearney and Nicholas H.A. Evans

anthropologists within the movement reconceptualize freedom not as the use of rational choice, in opposition to cultural limitations, but rather as a capacity for reflection developed in and through one’s social relations (Faubion 2011; Mahmood 2001; Laidlaw 2010; see also Foucault 1992). Social relations, in this model, not only shape the information we have or the situations we act within, but also the way we reflect ethically upon them (Heywood 2015). Yet, this idea is not intended to diminish but rather expand the role of the individual within our analyses: giving more space to the freedom our complex interior processes afford us. The point is that people’s behaviours do not need to be rationalized as either functional economic responses or as coming from false consciousness (Keane 2015). We can search for explanations in terms of how they distinctively and freely evaluate their relational worlds (Robbins 2013a; Lambek 2000; Laidlaw 2013). The anthropology of ethics represents an important break from other versions of social science. But the demonstration of how people are not dupes, but rather reflective, intelligent, and conscious, has also been argued to deepen the idea of humans as consistent and rational animals (Mittermaier 2012; Das 2014; Englund 2008; Kapferer and Gold 2018). This is more obviously true of the scholars that follow their informants’ own focus on consistency through keeping to resolutions (for more on this argument, see Evans 2016, 2017, 2020). But it is also true of those arguments that explicitly focus on ambivalence, failure and fragmentation. This scholarship also treats people’s difficulties as coherent and reflective responses to the diverse social patterns and moral imaginations they live within (e.g. Schielke 2009a, 2009b; see also Laidlaw 2013; Mayblin and Malara 2018). These arguments thus end up implying, in sharp contrast to Radoilska’s attention to similar instances in this volume, that these are not really failures to be good, but evidence of belief in a diverse set of values. If work in the anthropology of ethics shows us how diverse people’s judgments of the good are, it also still assumes that people, nevertheless, continue to act in line with those judgments; that is, that their action is rationally related to their ethical thinking. Many such arguments deliberately avoid concluding that something has gone wrong with people’s will by instead demonstrating how their actions make sense in a more complex social context. Even attempts to explicitly foreground people’s moral struggles unintentionally reproduce Socrates’ denial of akrasia, and thus foreclose rather than open up questions about the relationship between judgment, intention and action. When, how and why do people judge themselves and others as judging poorly, failing to act on good intentions, and failing to formulate the good intentions their judgments would merit? Anthropologists do not know because we have been so concerned not to impose a judgment upon them that we have never asked. We have relied so heavily on understanding social behaviour as having meaningfully linked and consistent patterns that we have not seriously inquired into whether human beings themselves consistently



Introduction

17

link their own judgments to their actions. The result is not simply that we have an embarrassing paucity of empirical material on this question, but also that a discipline that aims to be empirically and theoretically open to the full diversity of human experience has actually taken up a highly particular and highly controversial position on it. Anthropology is a science of consistency that neglects the possibility that there might be, in human life, a complex relationship between judgment, intention and action.

Philosophical Inconsistencies Unlike anthropologists, philosophers have debated the questions around akrasia for over two millennia. Indeed, few philosophers have been content to simply conclude the discussion of akrasia with Socrates’ simple denial of its possibility. In doing so, they have developed a wide variety of sophisticated resources for handling the topic’s intricate complexities. Within philosophy, the push back against Socrates began immediately, by some accounts, within Plato’s own corpus. How to interpret the development of Socrates’ words within Plato is a subject of debate. Some see Plato’s thought as evolving so that he came to view emotions and appetites as things that prevent us from judging, rather than reflective judgments in and of themselves (Nussbaum 1986). Thus, the conflict between a judgment and an appetite becomes a conflict between two unlike things. When we eat the burger, on this interpretation, we do so because we are captive to our unthinking hunger, rather than consciously choosing to go against our better judgment.2 On this interpretation, Plato did not end up making any more room for akrasia than Socrates. In other interpretations, however, Plato’s position moved away from Socrates so significantly that, by the time he was writing the Republic, he could argue that akrasia did, in fact, emerge out of the presence of competing motivations underlying human action (Bobonich and Destrée 2007, xvii). Aristotle, more explicitly than Plato, rejected the Socratic position with a detailed description of akrasia in the Nicomachean Ethics, in which he described it as a practical failure of character (Lear 1988, 174–75). For Aristotle, the truly virtuous person was one who was not conflicted by opposing desires, but instead experienced harmony and unity (Lear 1988, 167– 68). Aristotle’s vision of human flourishing was of a man who had managed, through the inculcation of habit, to align his desires in a single direction in which happiness and goodness were one and the same. Such a man was not akratic, for he felt none of the conflict that would typically lead to akrasia, and he felt his desires to always align with his judgments. For Aristotle, such a man possessed complete self-control (enkrateia). At the other end of the scale, the ignorant and the u ­ nfree – w ­ ho, for Aristotle, included slaves and w ­ omen – ­were also incapable of akrasia for different reasons.

18

Patrick McKearney and Nicholas H.A. Evans

Aristotle’s philosophical position differed with respect to different groups of people. Aristotle held a fully internalist view of the relationship between judgment and action for the truly virtuous. They always did what they judged best, and this is why akrasia was not possible for them. Aristotle’s view of women and slaves was, however, entirely externalist in that he thought their actions were not governed in any way, shape or form by their judgments. To Aristotle, these latter populations thus could not be akratic for entirely the opposite reason to the fully virtuous citizen. Aristotle’s ethics, however, was directed to an intermediate category: free male citizens who had yet to achieve virtue. Though they had the potential to reach it, their actions were not yet ruled by their judgments about what was best in as much as they gave in to passion and temptation. Therefore, if they were to cultivate themselves properly, the relationship between their actions and judgments would become more and more internalist over time. Aristotle sees these individuals as having both internalist and externalist possibilities. His thought emphasizes that the prospect of reconciling their judgments and actions, as well as the reality of ‘the divergences that can result between an agent’s evaluation of her options and her motivation to act’ (Stroud 2014). It is precisely this combination of this possible relationship between judgment and action, alongside the possibility of its breakage, that makes akrasia possible for such individuals in Aristotle’s thought (Mele 1987, 97). For these citizens, akrasia is a possibility because judgments do have some relationship to their actions, just not a relationship of perfect identification. In other words, for Aristotle, it was precisely those qualities that enabled a person to envision a coherent ethical end that also made them vulnerable to being akratic. In the memorable phrasing of Amélie Rorty, ‘Akrasia is a disease that only the strong can suffer’ (Rorty 1983, 176). Aristotle’s argument takes akrasia as the defect of the potentially, but not yet fully, virtuous. It thus relies on the idea that a truly ordered human existence would not feature akrasia. For some, this sidesteps the fundamental philosophical challenge of akrasia, for it does not show how incontinence in the very strict sense is possible (Lear 1988, 181). That is, Aristotle does not show that a person capable of arriving at all things considered best judgments is also capable of acting contrary to those judgments. Aristotle can thus still leave his readers with the fundamental question of whether it is possible to provide an explanation of akrasia that shows how a person might act inconsistently with what they believe and judge to be true. Efforts to provide an answer to this question continue to dominate the modern philosophical debate about akrasia, with varying degrees of success. Most philosophers agree that we should be able to demonstrate that the strong version of akrasia is possible. But they do not agree on how. The modern philosophical debate took on a new urgency in 1970 when Donald Davidson published ‘How Is Weakness of the Will Possible?’ (2001). In very broad terms, Davidson agreed with Socrates that the idea akrasia exists



Introduction

19

poses a genuine philosophical problem we cannot neatly resolve by simply pointing to instances of people acting strangely. Like Socrates, he felt that if a person truly judges one course of action to be better than another, then that person will want to carry out the better action. And like Socrates, he assumed that free humans will choose to do what they want to do. His solution to explaining how akrasia is, then, actually possible thus involved distinguishing between different kinds of judgment. Davidson argued that sometimes people act on partial judgments, rather than on their all-out judgment of what the right thing to do might be. This raised an important question about just what good judgment looks like, and whether humans always have such a thing as a ‘best judgment’. We pick up this question from an anthropological perspective in this volume by exploring societies in which it is not assumed that humans judge consistently, and where there is consequently little social expectation that people will have such a thing as a ‘best judgment’ in the first place. We see these as social realisations of Davidson’s argument, i.e. it is only if we have ‘best judgments’ in the first place that the philosophical problem of akratic inconsistency can exist. Nonetheless, many commentators have felt that Davidson, like Aristotle, sidesteps the actual problem of akrasia, because he shows only how humans can be akratic in limited and irrational ways (i.e. when they act on something less than their ‘best judgment’). Davidson does not actually explain how it is possible for us to freely, reflectively and intentionally act against our best (all things considered) judgments (Stroud 2014). Philosopher Richard Holton generated a new wave of discussion on akrasia by distinguishing between akrasia and weakness of will (Holton 1999). The distinction is as follows: Akrasia is when someone does not do what they judge best to do, or does do something they judge it is best not to do. Weakness of will is, by contrast, when someone does not do what they intended to do, or when do something that they intended not to do. Both of these involve a breakdown between one’s reasoning processes and one’s actions, but they are of different kinds. Let us return to our meat-eating example. Weakness of will occurs if someone breaks their intention not to eat meat because their emotions disabled their capacity to judge, or swayed them to make a new judgment that eating meat was OK. Neither of these would, however, constitute akrasia. If an emotion or appetite effectively forced one to act in a certain way by disabling one’s capacity to judge, then one would not be acting freely, consciously and willingly. And if one now had a new ‘best judgment’ that eating meat was OK, then one would not be acting against it. Both of these cases, however, would still involve a contradiction of a previous intention, and thus constitute weakness of will. Holton’s focus on weakness of will enabled him to account for our repeated failure to do what we judge, at a certain point in time, to be the best thing to do without needing to prove the philosophical possibility of akrasia.

20

Patrick McKearney and Nicholas H.A. Evans

Holton’s focus on the intentions that make weakness of will possible provided a way of seeing abstract ‘best judgments’ in people’s actions. The intentions Holton was interested in are resolutions to act in a certain way in the future based on evaluations made in the present, such as resolving to get up in the morning when one knows one will be tired. Such intentions save us from having to consider all things all the time, and instead allow us to arrive at a decision that will guide future action. Holton focused, in particular, on ‘contrary-inclination-defeating intentions’ that are specifically designed to counter our temptations to act in a way that goes against our current reasoning (such as deciding to be vegetarian in order to resist the opposing impulse to eat meat). Intentions, in this way, are ways in which we concretize our ‘best judgment’ in a given moment, and they attempt to mediate that judgment into a future situation. They are technologies for breaching temporal gaps between present judgments and future actions. Because Holton focuses on the observable things that people do, his concept has obvious resonances with anthropology, which is why many of the contributors to this volume have found it productive to work with. Amélie Rorty expanded the debate in another direction by identifying the possibility of what she calls akratic breaks in a wide variety of places. Traditionally, the akrasia debate has focused on the possibility of a broken relationship between judgment and action. While Holton directs us to focus on the more specific relationship between intention and action, Rorty demonstrated that there are even more places when an akratic break between these can occur, such as when we hold that something is best to do, but fail even to form an intention to do it. Alternatively, she shows how people can assess themselves as having morally erred even by failing to form the right kind of ­judgment – a­ s in cases of implicit racism, where people have yet failed to translate some of their principles into their ways of seeing the world. This is a possibility that Radoilska explores at length in this volume, and that anthropology’s tradition of attending to the relationship between people’s judgments can help shed even further light on. In conclusion, throughout the history of philosophy, it has proven surprisingly difficult to actually show that people can be akratic. This is despite the fact that it seems, intuitively, to be the case that akrasia is a possibility within human life. At the very least, there is no universally agreed upon explanation of how akrasia is possible, and the problem continues to give rise to new and rich debates. Even if very few people fully agree with Socrates anymore, his basic contention that action follows judgment continues to haunt the philosophical debate. We thus find philosophy in a situation analogous to anthropology: torn between the competing inclinations to emphasize the consistency of the human subject, and to attend to the mess and disorder of human life. But philosophy, in stark contrast to anthropology, has a developed debate about these topics, with sophisticated resources for analysing the terrain of



Introduction

21

inner struggle, in which scholars propose complex intellectual solutions. In particular, philosophers have sufficiently challenged the Socratic argument such that it is impossible to assume its model of human judgment and action is self-evident. This philosophical resistance challenges the implicit anthropological assumption that people’s actions are always explicable by reference to their way of seeing and judging the world.

An Anthropology of Akrasia In this volume, we devote anthropological attention to instances when people feel their own actions to be in conflict with their judgments. In doing so, we focus on the kind of cases that social scientists have traditionally sought to tidy up through neat explanations that show that people are really more consistent than they seem. We consider whether human action might have a much more complex relationship to judgment than anthropology has yet considered. Do people always act as they judge it is best to? And do people always intend to do what they think it is best to do? Do people act, in other words, in line with how they think? We want to challenge the assumption that akratic inconsistencies exist primarily in the eye of the observer (and can thus, through analysis, be expelled). We do so not from the naïve conviction that we could simply prove empirically that akrasia does exist, or even to argue for any particular side in the ongoing philosophical debate; we do so, instead, to open up a crucial ethnographic perspective on an issue to which anthropologists have never attended. Our challenge, in other words, is not just to countenance the idea that people may be in some kind of inner turmoil, but to confront analytical interpretations of that turmoil so as to understand why it is so often explained away. We use the resources of the philosophical debate to develop our anthropological capacity to attend to akrasia in ways that do not push it away from view, but rather allow us to reflect more deeply on the complexity of the relationship between judgment, intention and action in human life. The philosophical debate demonstrates that describing situations that look like akrasia is never straightforward or simple. Even trying to state what is going on in the mind of an akratic person involves going deep into contested philosophical questions (Williams 1993a). As anthropologists, we use the debates about whether or not akrasia exists to question actions that look like akrasia and to ask whether the relationship between judgment and action has broken down in any given instance or whether it only seems to have done. We draw upon the distinctions between akrasia and other forms of inner struggle to help us distinguish the kinds of breakdown that we observe and that our interlocutors describe. More broadly, we use the contentious nature of the philosophical debate to create breaks in the tight relationship we assume

22

Patrick McKearney and Nicholas H.A. Evans

between judgment, intention and action. Doing so opens up new ways of researching, describing and analysing human behaviour. The chapters offer different examples of what an anthropology that takes akrasia seriously might involve, and they say that better than we could for them. Here, we lay out just two of the wider issues that these chapters raise and that an anthropology of akrasia might lead to.

Issue 1: Can Akrasia Have Its Origins in Collective Life? This question has often been overlooked in the most well-known philosophical debates about akrasia, which has a resolutely individualistic focus. But those philosophers who have considered it have produced a body of work around the concept of ‘endemic akrasia’ that speaks directly to (though not always in agreement with) anthropology’s own traditions of thinking about the effect of social life on human judgment and action. Philip Pettit, for example, has written about whether there are kinds of collectives that might closely enough resemble a unified and rational agent such that they are capable of akrasia (Pettit 2003). In our current age, for example, we might want to talk about the akrasia of polluting corporations who may be explicitly committed to sustainability while routinely acting in direct contrast to this aim. We might want to think of whether committees who fail to motivate themselves to do what they all agree is right are ‘akratic’. In this volume, Ivan Deschenaux takes up a similar question in relation to caste in Nepal, by developing an ethnographic way to take this possibility seriously in relation to a pressing question of social science. A slightly different perspective on this question can be found in the work of Amélie Rorty, who argues that we should understand the manifestation of individual akrasia from an epidemiological and demographic perspective (Rorty 1997, 649). ‘Just as a disposition to chronic bronchitis may indicate a toxic environment,’ Rorty explains, ‘so individual akrasia may indicate social disorder’ (Rorty 1997, 649). For Rorty, a pattern of individual akrasia across a group can be understood as an endemic condition, born of particular social structures, institutional frameworks, and political discourses. As such, she argues the correct response to endemic akrasia is not to be undertaken at the level of the individual, but rather at the level of political and economic reform (Rorty 1997, 657). This work has resonances with contemporary work in anthropology on the shaping of psychic life by social and political conditions. We know, for example, that certain social conditions exacerbate the perils of substance addiction or schizophrenia (Marrow and Luhrmann 2017). Social scientists also frequently examine how far violence, precarity and poverty can shape people’s capacities to construct coherent lives. In short, we know there are situations in which people are pushed by social forces towards acting against



Introduction

23

their own judgments about what it is best for them to do morally and for their own interests. Can the idea of endemic akrasia add something to these debates in social science? The idea of endemic akrasia foregrounds the wilful, free nature of the individual, while also pointing toward the way that structures of power might constrain him or her. As an analytical frame, it keeps front and centre the idea of the person who freely goes against their better judgment while complicating the idea of that freedom. It is, we believe, a helpful complication of the old sociological debate between structure and agency, and it refuses to allow easy dissolution of the question into either extreme. Some philosophers even argue that diagnosing people as akratic can be a better way to maintain our focus on their rationality and coherency than the other option of treating them as simply determined in their actions (one classic social science approach), or more generally irrational (an option that anthropology will always be inclined to reject, and that has shown to be particularly problematic when analysing the decisions of people in poverty). Part of the challenge of endemic akrasia is that it forces us to think of coherence, continence and strength of will (in short, all those things that enable a person to resist the social forces that might induce endemic akrasia) as goods whose availability depends upon a person’s social position.3 We should be wary of the ways this idea has been pursued, from Aristotle’s argument that only free men could be properly strong-willed, to the conclusion that poor people are the authors of their own poverty. But the idea of strength of will might also provide new routes away from those old dangers, while enabling us to take into account more of the complexity of their relational and interior lives. An anthropology that is open to the idea of endemic akrasia might therefore be one that is able to explore why and how people can have their possibilities for action curtailed without reducing them to the level of rule-following automatons.4 As Richard Holton argues in this volume, taking this possibility seriously may well enable anthropology to realize its full potential to speak back to the individualistic focus on akrasia that is dominant in both the philosophical debate and in understandings of human action more broadly.

Issue 2: How Can We Analyse and Account for Variation in the Ways in Which People Experience and Respond to Inconsistency between Judgments and Actions? Philosophers are not the only people interested in the question of whether a person’s action is linked to their evaluative judgment. Indeed, humans everywhere make judgments about the relationship between other people’s actions and judgments (Austin 1956; Strawson 1962; Hughes et al. 2019; Lambek 2015, 2010). ‘He doesn’t think about what he does’; ‘she’s careless’; ‘she’s very deliberate.’ These are statements we hear all the time. All of them are comments

24

Patrick McKearney and Nicholas H.A. Evans

about whether a person’s judgments are informing and guiding their actions. They thus model different understandings of how judgment, intention and action can break down within the human psyche. The contributors to this volume show that these understandings are not everywhere the same. There is, instead, a staggering variation in the ways that people treat what might look like moments of inconsistency: from a whole host of ways in which they struggle with a feeling of having acted out of line with strong resolutions, to societies in which no-one seems in the slightest bit concerned about the idea that they might act incontinently. This raises the possibility that seemingly purely psychological relationships between judgment and action might also depend on social differences. We propose we can use the philosophical distinction we outlined earlier to contrast those societies that have a more ‘internalist’ interpretation of the mind, with others that tend to follow an ‘externalist’ reading of what people do. Societies that tend toward ‘internalism’ are those in which a strong social emphasis is placed on the idea that action is a deliberate emanation of a person’s decisions, and these societies are consequently likely to feature elaborate discourses about what to do when people’s decisions do not translate into action. Contemporary work in the anthropology of ethics has showcased a range of ethical projects that involve an active attempt to cultivate a correspondence between desire, judgment and action, particularly Islamic reform movements in which people seek to make their outward behaviour correspond directly with their inner desires (Laidlaw 1995; Mahmood 2012; Evans 2017, 2020; Deeb 2006). These ethical projects are similar to the Aristotelian tradition that we have engaged with in this introduction, in that they see akrasia as a basic problem of human nature that requires work to be fixed. In the case of Islamic reform movements, this is perhaps of no great surprise given the common genealogy they share with Western philosophy that goes back to ancient Greek thought. Often people living in such traditions want to cultivate themselves away from akratic action, and they may have developed sophisticated techniques of the self to do this. The ‘intentions’ Holton outlines are one such ethical technology for ensuring that future actions conform to judgments and elaborate variations upon the intention are common within many internalist societies. Christian confession, for instance, renders even lapses in behaviour as an occasion to reinforce the normativity of avoiding sin (Robbins 2004; see also Mayblin and Malara 2018). Whether people see it as possible to eradicate akrasia altogether will depend on the particular conception they have of human nature. A good example of this is the Christian conception described by Bialecki in this volume, which is strongly shaped by Augustin and sees action as having a normative relationship to judgment that is always breaking down. Other traditions problematize akrasia as something that is produced out of particular social conditions or relationships. Here, the work to cultivate



Introduction

25

oneself out of akrasia is likely to take a very different form, for it will involve not only working on the self, but also its environment. Indeed, anthropologically speaking, the intentions Holton outlines are a more individualistic kind of ethical technology in comparison to more social ones that can create more public and relational links between judgment and action. There is a long anthropological tradition that looks at religious and political rituals as tools to bind action more tightly to intentions (e.g. Bloch 1974; Rappaport 1999; Robbins 2015).5 Other forms of contemporary governance, particularly in the form of psychological management, similarly make such connections normative through the regulation of everyday life, examples of which can be found in the chapters by Lester and McKearney (see also Davis 2012; Lester 2019; Weinberg 2005). Several chapters in this volume nonetheless demonstrate sharper differences still in that they represent societies that have a more ‘externalist’ theory of mind and action. Within these traditions, there will be far less of a concern to link actions to judgments, decisions and desires. In these settings, akrasia appears very differently, and may not be problematized at all. People pay less attention to inconsistencies between judgment and action in these settings, and when these inconsistencies are recognized, there may be little attempt to rectify them. These traditions offer the sharpest challenge to anthropological theory, for they allow ways of being a human subject that do not sit easily with our discipline’s need to make action sensible in light of, and consistent with, people’s judgments. Francesca Mezzenzana’s chapter on the drinking habits of the Runa of the Ecuadorian Amazon, for instance, offers a rich ethnography in which we can explore such challenging possibilities. The idea of a society with a fully ‘externalist’ theory of mind might at first sound strange, but we think that there are examples with which most anthropologists will be familiar. Take, for example, those Melanesian societies in which people stick resolutely to a doctrine of the ‘opacity of minds’, refusing out of principle (at least in public) to reflect upon what others are thinking (Robbins and Rumsey 2008; Stasch 2008; Robbins 2008, 2020a). In those, and analogous societies, no clear link between evaluative judgment and action is likely to be drawn for anybody (see also Danziger 2006; Carey 2017; Mezzenzana 2020). This makes it practically impossible to hold people to account for their previous intentions or declarations and often means that people do not make them in the first place. Ethnographers, for instance, have attested to an almost total absence of the promise: a highly internalist ethical technology inasmuch as it creates a normative social link between a statement of what someone will do and their future actions (Robbins 2001, 2008, n.d.; Carey 2017). In practice, few societies exhibit such extreme features. Rather, people are likely to variously evaluate the actions of their fellow humans according to either internalist or externalist principles in a more variegated way, depending on the kind of social relations and contexts they are in. This, in itself, raises

26

Patrick McKearney and Nicholas H.A. Evans

new questions for an ethnographer. What kinds of action receive internal and which external forms of accountability? Are people in different social situations, and differing social positions, held responsible differently according to varying understandings of how their judgments relate to their actions? We raise these questions because we believe that the chapters in this volume challenge the basic philosophical position, often assumed since Aristotle, that akrasia is a state that most people would seek to overcome. Instead, some of these chapters raise the idea of akrasia b ­ eing – f­or some people at l­east – a­ n unremarkable possibility or even a desired end itself. And Radoilska’s chapter reflects on more recent philosophical debates about these issues. This all leads to a question that challenges the very foundations of the akrasia debate, namely, are there versions of human life in which there is so little expectation that an individual’s actions and best judgments will be co-ordinated that akrasia itself cannot even be considered a philosophical puzzle?

Conclusion Taking akrasia seriously does not mean adopting a philosophical certainty about it. The idea of akrasia poses a question and not an answer.6 The philosophical debate has not arrived at any consensus about what akrasia is, how we can identify it, or whether it is even possible. And that is precisely our point. At the moment, these questions are not contentious within anthropology because they are not even acknowledged. Many of the issues are under the surface of our most important contemporary debates, and at stake in the ethnographic issues we consider. But we do not recognize the depth, complexity and contested nature of the philosophical questions they involve. Thus, we ignore important material and unwittingly make highly debatable assumptions in our interpretations of thought and behaviour­without even realising we are doing so. Taking akrasia seriously as a concept means opening ourselves to the questions it poses about human life and social action. The philosophical debate about akrasia offers an impressive array of resources for considering and confronting those questions. As the contributors to this volume show, many of those resources can be readily integrated into an ethnographic investigation of how different societies approach conflicts between judgment and action. Doing so immediately reveals the sheer variety of ways people conceive the self, its component parts and how they can relate to action. In the same way that the position philosophers take on akrasia reveals much about how they conceptualize humans as thinking and acting creatures, ethnographic attention to these topics uncovers new ways of seeing and analysing different social approaches to human ethical judgment and action. The contributors to this volume do more than just draw upon the akrasia debate to begin an ethnographic and analytical conversation about akrasia



Introduction

27

within anthropology. They also demonstrate a wide diversity of ways of understanding the person that do not fit neatly with any of the common assumptions underlying the various different positions in the philosophical akrasia debate. The philosophical debate, for example, gives pride of place to consistency, as seen through the implicit but pervasive assumption that its absence is a problem requiring not only moral wrangling but also philosophical analysis. These chapters ask whether that might be more of a culturally specific concern than philosophy recognizes. There are social conditions that reinforce the sense that inconsistency is a concern, but there are also those that render it unproblematic, uninteresting or unremarkable. What if many of our confusions and debates about akrasia stem from a highly particular conception of the role consistency must play in human life? What if it does not have to play that role at all? If analytically repurposed, the kind of deep ethnographic understanding that has kept anthropology out of the akrasia debate can offer something back to philosophy. When anthropologists attend to the intrapersonal complexities the akrasia debate concerns itself with, they expose connections between people’s inner terrain and the broader relational and social world they live in. Akrasia may be a complication of the individual, but that individual is a social actor, and their experience of akrasia cannot be understood without a grasp of what society means for them. The result of our ethnographic investigation is to reveal relationships people have with their own judgments and actions that depart from universal philosophical models of the person. And this raises questions about whether and how the very possibility of the internal complications of akrasia may depend, more than we had imagined, on our external relationships. If this is so, it is about time that ethnography enters the debate. In this volume, we have borrowed much from philosophy to form our own arguments. It is our hope that this idea can, in turn, offer philosophers new ways of discussing an ancient problem. And it is our hope that this will open up for anthropologists, too, new ways of conceiving and investigating the complexity of human beings in their social relations. Patrick McKearney is an Assistant Professor at the University of Amsterdam conducting research in the UK, India, and Italy. His recent articles on disability, care, ethics,= and religion include publications in Social Analysis, Ethnos and JRAI. He has also edited two special issues on cognitive disability in The Cambridge Journal of Anthropology and Medical Anthropology. Nicholas H.A. Evans is a visiting fellow at the London School of Economics and author of Far From the Caliph’s Gaze: Being Ahmadi Muslim in the Holy City of Qadian (Cornell University Press, 2020).

28

Patrick McKearney and Nicholas H.A. Evans

Notes   1. Insiders and outsiders to the discipline have sometimes articulated this position as a form of moral relativism (e.g. Geertz 1984). But this is just one of the ways in which the anthropological aversion to external judgments can be articulated without resorting to the shaky philosophical argument for relativism (e.g. Laidlaw 2013). Anthropological practice has never relied on a relativist philosophical argument about morality, and it is thus a mistake to think the discipline’s challenge to external judgments can be undermined by undermining moral relativism.   2. This argument parallels anthropological theories that attribute social forces with the ability to cloud people’s judgment, although Plato’s idea is that proper judgment is overrun not by forces in society, but by appetites within the person.   3. On strength of will, see Holton (2003).   4. Compare with Laidlaw (2013) on the ‘science of unfreedom’.   5. Though the potential for ritual to sever the links between intention and action are just as important (Humphrey and Laidlaw 1994).   6. We are indebted to Anastasia Piliavsky for this observation.

References Anscombe, G.E.M. 1958. ‘Modern Moral Philosophy’. Philosophy 33(124): 1–19. Austin, J.L. 1956. ‘A Plea for Excuses: The Presidential Address’. Proceedings of the Aristotelian Society 57: 1–30. Banner, Michael. 2014. The Ethics of Everyday Life: Moral Theology, Social Anthropology, and the Imagination of the Human. Oxford: Oxford University Press. Berlant, Lauren Gail. 2012. Cruel Optimism. Durham, NC; London: Duke University Press. Biehl, João. 2005. Vita: Life in a Zone of Social Abandonment. Berkeley: University of California Press. Biehl, João, Byron Good, and Arthur Kleinman. 2007. Subjectivity: Ethnographic Investigations. University of California Press. Bloch, Maurice. 1974. ‘Symbols, Song, Dance and Features of Articulation: Is Religion an Extreme Form of Traditional Authority?’ European Journal of Sociology / Archives Européennes de Sociologie 15(01): 54–81. Bobonich, Christopher, and Pierre Destrée (eds). 2007. Akrasia in Greek Philosophy: From Socrates to Plotinus. Philosophia Antiqua, v. 106. Leiden; Boston: Brill. Bourdieu, Pierre. 1984. Distinction: A Social Critique of the Judgment of Taste. Harvard University Press. . 1998. Acts of Resistance: Against the Tyranny of the Market. New Press. Bratman, Michael. 1979. ‘Practical Reasoning and Weakness of the Will’. Noûs 13(2): 153–71. https://doi.org/10.2307/2214395. Briggs, Jean L. 1999. Inuit Morality Play: The Emotional Education of a Three-YearOld. New Haven; London: Yale University Press. Carey, Matthew. 2017. Mistrust: An Ethnographic Theory. Chicago: Hau Books. Comaroff, Jean, and John L. Comaroff. 1999. ‘Occult Economies and the Violence of Abstraction: Notes from the South African Postcolony’. American Ethnologist 26(2): 279–303. Danziger, Eve. 2006. ‘The Thought That Counts’. In Roots of Human Sociality: Culture,



Introduction

29

Cognition and Interaction, edited by Stephen C Levinson and Nicholas J Enfield, pp. 259–78. Oxford: Berg Publishers. Das, Veena. 2007. Life and Words: Violence and the Descent Into the Ordinary. Berkeley: University of California Press. . 2014. ‘Ethics, the Householder’s Dilemma, and the Difficulty of Reality’. HAU: Journal of Ethnographic Theory 4(1): 487–95. Davidson, Donald. 2001. How Is Weakness of the Will Possible? Oxford University Press. http://www.oxfordscholarship.com/view/10.1093/0199246270.001.0001/acp​ rof-9780199246274-chapter-2. Davis, Elizabeth Anne. 2012. Bad Souls: Madness and Responsibility in Modern Greece. Durham, NC: Duke University Press. Deeb, Lara. 2006. An Enchanted Modern: Gender and Public Piety in Shi’i Lebanon. https://press.princeton.edu/books/paperback/9780691124216/an-enchanted-mo​ dern. Douglas, Mary. 1973. Natural Symbols. Middlesex, England: Penguin Books. . 1980. Edward Evans-Pritchard. New York: Viking Press. , ed. 2006. Witchcraft Confessions and Accusations. Reprinted. Witchcraft, Folklore and Mythology: In 6 Volumes 2. London: Routledge. . 2008. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. New York: Routledge. Englund, Harri. 2008. ‘Extreme Poverty and Existential Obligations: Beyond Morality in the Anthropology of Africa?’ Social Analysis 52(3): 33–50. Evans, Nicholas H.A. 2020. Far from the Caliph’s Gaze: Being Ahmadi Muslim in the Holy City of Qadian. Cornell University Press. https://www.jstor.org/stable/10.75​ 91/j.ctvq2w3b5. . 2016. ‘Witnessing a Potent Truth: Rethinking Responsibility in the Anthropology of Theisms’. Journal of the Royal Anthropological Institute 22(2): 356–72. . 2017. ‘Beyond Cultural Intimacy: The Tensions That Make Truth for India’s Ahmadi Muslims’. American Ethnologist 44(3): 490–502. Evans-Pritchard, Edward Evan. 1933. ‘The Intellectualist (English) Interpretation of Magic’. Bulletin of the Faculty of Arts (Alexandria) 1: 282–311. . 1934. ‘Lévy-Bruhl’s Theory of Primitive Mentality’. Bulletin of the Faculty of Arts, Egyptian University. . 1965. Theories of Primitive Religion. Oxford; New York: Clarendon Press; Oxford University Press. . 1976. Witchcraft, Oracles and Magic among the Azande. Oxford: Oxford University Press. Faubion, James D. 2001. ‘Toward an Anthropology of Ethics: Foucault and the Pedagogies of Autopoiesis’. Representations 74(1): 83–104. . 2011. An Anthropology of Ethics. Cambridge: Cambridge University Press. Foucault, Michel. 1992. The History of Sexuality, Vol. 2: The Use of Pleasure. London: Penguin Books. Garcia, Angela. 2014. ‘The Promise: On the Morality of the Marginal and the Illicit’. Ethos 42(1): 51–64. Geertz, Clifford. 1975. The Interpretation of Cultures: Selected Essays. London: Hutchinson & Co. . 1984. ‘Distinguished Lecture: Anti Anti-Relativism’. American Anthropologist 86(2): 263–78. Gluckman, Max. 1955. Custom and Conflict in Africa. Oxford: Blackwell.

30

Patrick McKearney and Nicholas H.A. Evans

Godelier, Maurice. 1977. Perspectives in Marxist Anthropology. Cambridge University Press. Graeber, David. 2011. Debt: The First Five Thousand Years. New York: Melville House. Heywood, Paolo. 2015. ‘Freedom in the Code: The Anthropology of (Double) Morality’. Anthropological Theory 15(2): 200–217. Holton, Richard. 1999. ‘Intention and Weakness of Will’. The Journal of Philosophy 96(5): 241–62. https://doi.org/10.2307/2564667. . 2003. ‘How Is Strength of Will Possible?’ In Weakness of Will and Practical Irrationality, edited by Sarah Stroud and Christine Tappolet, pp. 39–67. Oxford: New York: Clarendon; Oxford University Press. Hughes, Geoffrey, Megnaa Mehtta, Chiara Bresciani, and Stuart Strange. 2019. ‘Introduction: Ugly Emotions and the Politics of Accusation’. The Cambridge Journal of Anthropology 37(2): 1–20. https://doi.org/10.3167/cja.2019.370202. Humphrey, Caroline. 2008. ‘Reassembling Individual Subjects: Events and Decisions in Troubled Times’. Anthropological Theory 8(4): 357–80. https://doi.org/10.1177​ /1463499608096644. Humphrey, Caroline, and James Laidlaw. 1994. The Archetypal Actions of Ritual: A Theory of Ritual Illustrated by the Jain Rite of Worship. Clarendon Press. Kapferer, Bruce, and Marina Gold. 2018. Moral Anthropology: A Critique. Keane, Webb. 2015. Ethical Life: Its Natural and Social Histories. Princeton: Princeton University Press. Laidlaw, James. 1995. Riches and Renunciation: Religion, Economy, and Society Among the Jains. Oxford: Clarendon Press. . 2002. ‘For an Anthropology of Ethics and Freedom’. The Journal of the Royal Anthropological Institute 8(2): 311–32. . 2010. ‘Agency and Responsibility: Perhaps You Can Have Too Much of a Good Thing’. In Ordinary Ethics: Anthropology, Language, and Action, pp. 143–64. New York: Fordham University Press. . 2013. The Subject of Virtue: An Anthropology of Ethics and Freedom. Cambridge: Cambridge University Press. Laidlaw, James, and Patrick McKearney. 2023. ‘Agonistic Liberals: Three Analytical Philosophers of Value Conflict’. In The Cambridge Handbook of the Anthropology of Ethics, edited by James Laidlaw. Lambek, Michael. 2000. ‘The Anthropology of Religion and the Quarrel Between Poetry and Philosophy’. Current Anthropology 41(3): 309–20. . 2010. Ordinary Ethics: Anthropology, Language, and Action. New York: Fordham University Press. . 2015. The Ethical Condition: Essays on Action, Person, and Value. Chicago: University of Chicago Press. Larsen, Timothy, and Daniel J. King. 2018. ‘The Dependence of Sociocultural Anthropology on Theological Anthropology’. In Theologically Engaged Anthropology, edited by J. Derrick Lemons, pp. 50–65. Oxford: Oxford University Press. Leach, Edmund. 1989. ‘Anthropological Aspects of Language: Animal Categories and Verbal Abuse’. Anthrozoös 2(3): 151–65. https://doi.org/10.2752/08927938978705​ 8055. Lear, Jonathan. 1988. Aristotle: The Desire to Understand. Cambridge: Cambridge University Press. Lester, Rebecca J. 2019. Famished: Eating Disorders and Failed Care in America. Oakland: University of California Press.



Introduction

31

Lévy-Bruhl, Lucien, and Lilian A. Clare. 2015. How Natives Think: (Les Fonctions Mentales Dans Les Sociétés Inférieures). MacIntyre, Alasdair. 2007. After Virtue: A Study in Moral Theory. London: Duckworth. Mahmood, Saba. 2001. ‘Feminist Theory, Embodiment, and the Docile Agent: Some Reflections on the Egyptian Islamic Revival’. Cultural Anthropology 16(2): 202–36. . 2012. Politics of Piety: The Islamic Revival and the Feminist Subject. Princeton: Princeton University Press. Malinowski, Bronislaw. 1987 [1922]. Argonauts of the Western Pacific. London: Routledge & Keagan Paul Limited. Marrow, Jocelyn, and Tanya Marie Luhrmann. 2017. Our Most Troubling Madness: Case Studies in Schizophrenia across Cultures. Oakland, California: University of California Press. Mattingly, Cheryl. 2014. Moral Laboratories: Family Peril and the Struggle for a Good Life. Berkeley: University of California Press. Mayblin, Maya. 2017. ‘The Lapsed and the Laity: Discipline and Lenience in the Study of Religion’. Journal of the Royal Anthropological Institute 23(3): 503–22. https://​ doi.org/10.1111/1467-9655.12650. Mayblin, Maya, and Diego Malara. 2018. ‘Introduction: Lenience in Systems of Religious Meaning and Practice’. Social Analysis 62(3): 1–20. https://doi.org/10.31​ 67/sa.2018.620301. McKearney, Patrick. 2016. ‘The Genre of Judgment’. Journal of Religious Ethics 44(3): 544–73. https://doi.org/10.1111/jore.12153. . 2022. ‘Disabling Violence: Intellectual Disability and the Limits of Ethical Engagement’. Journal of the Royal Anthropological Institute 28(3): 956–74. https://​ doi.org/10.1111/1467-9655.13773. McKearney, Patrick, and Tyler Zoanni. 2018. ‘Introduction: For an Anthropology of Cognitive Disability’. The Cambridge Journal of Anthropology 36(1): 1–22. https://​ doi.org/10.3167/cja.2018.360102. Mele, Alfred R. 1987. Irrationality: An Essay on Akrasia, Self-Deception and SelfControl. New York: Oxford University Press. Mezzenzana, Francesca. 2020. ‘Between Will and Thought: Individualism and Social Responsiveness in Amazonian Child-Rearing’. American Anthropologist 122(3): 540–53. Mittermaier, Amira. 2012. ‘Dreams from Elsewhere: Muslim Subjectivities beyond the Trope of Self-Cultivation’. Journal of the Royal Anthropological Institute 18(2): 247–65. Moore, Henrietta L, and Todd Sanders. 2002. Magical Interpretations, Material Realities: Modernity, Witchcraft, and the Occult in Postcolonial Africa. New York: Routledge. http://www.tandfebooks.com/isbn/9780203398258. Nussbaum, Martha Craven. 1986. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Cambridge: Cambridge University Press. . 1990. Love’s Knowledge: Essays on Philosophy and Literature. New York; Oxford: Oxford university press. Pettit, Philip. 2003. ‘Akrasia, Collective and Individual’. In Weakness of Will and Practical Irrationality, edited by Sarah Stroud and Christine Tappolet, pp. 68–96. Oxford: Clarendon Press. Rappaport, Roy A. 1999. Ritual and Religion in the Making of Humanity. Cambridge: Cambridge University Press.

32

Patrick McKearney and Nicholas H.A. Evans

Robbins, Joel. 2001. ‘God Is Nothing but Talk: Modernity, Language, and Prayer in a Papua New Guinea Society’. American Anthropologist 103(4): 901–12. https://doi​ .org/10.1525/aa.2001.103.4.901. . 2004. Becoming Sinners: Christianity and Moral Torment in a Papua New Guinea Society. Berkeley, CA.: University of California Press. . 2007. ‘Continuity Thinking and the Problem of Christian Culture: Belief, Time, and the Anthropology of Christianity’. Current Anthropology 48(1): 5–38. . 2008. ‘On Not Knowing Other Minds: Confession, Intention, and Linguistic Exchange in a Papua New Guinea Community’. Anthropological Quarterly 81(2): 421–29. . 2013a. ‘Beyond the Suffering Subject: Toward an Anthropology of the Good’. Journal of the Royal Anthropological Institute 19(3): 447–62. . 2013b. ‘Monism, Pluralism and the Structure of Value Relations: A Dumontian Contribution to the Contemporary Study of Value’. HAU: Journal of Ethnographic Theory 3(1): 99–115. . 2015. ‘Ritual, Value, and Example: On the Perfection of Cultural Representations’. Journal of the Royal Anthropological Institute 21(S1): 18–29. . 2020a. ‘Mental Opacity’. In The International Encyclopedia of Linguistic Anthropology, 1–6. Wiley Blackwell. https://doi.org/10.1002/9781118786093.iela​ 0257. . 2020b. Theology and the Anthropology of Christian Life. Oxford: Oxford University Press. http://search.ebscohost.com/login.aspx?direct=true&scope=site​ &db=nlebk&db=nlabk&AN=2532394. . n.d. ‘Opacity of Mind, Imagining Others, and the Coordination of Action: Melanesianist Reflections on the Ethics of Trust’. Unpublished Manuscript. Robbins, Joel, and Alan Rumsey. 2008. ‘Introduction: Cultural and Linguistic Anthropology and the Opacity of Other Minds’. Anthropological Quarterly 81(2): 407–20. Rorty, Amélie. 1983. ‘Akratic Believers’. American Philosophical Quarterly 20(2): 175–83. . 1997. ‘The Social and Political Sources of Akrasia’. Ethics 107(4): 644–57. Schielke, Samuli. 2009a. ‘Ambivalent Commitments: Troubles of Morality, Religiosity and Aspiration among Young Egyptians’. Journal of Religion in Africa 39(2): 158– 85. . 2009b. ‘Being Good in Ramadan: Ambivalence, Fragmentation, and the Moral Self in the Lives of Young Egyptians’. Journal of the Royal Anthropological Institute 15(s1): S24–40. Shore, Brad. 2000. ‘Human Diversity and Human Nature. The Life and Times of a False Dichotomy’. In Being Humans: Anthropological Universality and Particularity in Transdisciplinary Perspectives, edited by Neil Roughley, pp. 81–104. Berlin: De Gruyter. Singer, Peter. 2010. ‘Speciesism and Moral Status’. In Cognitive Disability and Its Challenge to Moral Philosophy, edited by Eva Feder Kittay and Licia Carlson, pp. 330–44. Chichester: Wiley-Blackwell. . 2011. Practical Ethics. Cambridge University Press. Stasch, Rupert. 2008. ‘Knowing Minds Is a Matter of Authority: Political Dimensions of Opacity Statements in Korowai Moral Psychology’. Anthropological Quarterly 81(2): 443–53. Stevenson, Lisa. 2014. Life Beside Itself: Imagining Care in the Canadian Arctic. Berkeley: University of California Press.



Introduction

33

Strawson, Peter F. 1962. ‘Freedom and Resentment’. Proceedings of the British Academy 48: 1–25. Stroud, Sarah. 2014. ‘Weakness of Will’. In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Spring 2014. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/spr2014/entries/weakness-will/. Stroud, Sarah, and Christine Tappolet. 2003. ‘Introduction’. In Weakness of Will and Practical Irrationality, edited by Sarah Stroud and Christine Tappolet, pp. 1–16. Oxford: Clarendon Press. Tambiah, Stanley Jeyaraja. 1990. Magic, Science, Religion, and the Scope of Rationality. Cambridge: Cambridge University Press. Taussig, Michael. 1977. ‘The Genesis of Capitalism amongst a South American Peasantry: Devil’s Labor and the Baptism of Money’. Comparative Studies in Society and History 19(2): 130–55. Velleman, J. David. 1992. ‘The Guise of the Good’. Noûs 26(1): 3–26. https://doi.org/10​ .2307/2215684. Weinberg, Darin. 2005. Of Others Inside: Insanity, Addiction, and Belonging in America. Philadelphia: Temple University Press. White, Thomas, and Matei Candea. 2018. ‘Animals’. Cambridge Encyclopedia of Anthropology, May. https://www.anthroencyclopedia.com/entry/animals. Wilk, Richard. 2001. ‘Consuming Morality’. Journal of Consumer Culture 1(2): 245–60. https://doi.org/10.1177/146954050100100211. Williams, Bernard. 1981. Moral Luck: Philosophical Papers 1973–1980. Cambridge: Cambridge University Press. . 1993a. Shame and Necessity. Berkeley: University of California Press. . 1993b. Morality: An Introduction to Ethics. Cambridge: Cambridge University Press. . 2011. Ethics and the Limits of Philosophy. Abingdon: Routledge. Wilson, Bryan R. 1974. Rationality. Oxford: Blackwell. Zigon, Jarrett. 2009. ‘Within a Range of Possibilities: Morality and Ethics in Social Life’. Ethnos 74(2): 251–76.

1 Trigger Warnings Danger, Desire and Declensions of the Will in Eating Disorder Treatment Rebecca J. Lester

In Wasted: A Memoir of Anorexia and Bulimia (2014), author Marya Hornbacher describes in vivid detail her excruciating descent into an eating disorder that almost took her life, and her slow and tortured climb back out. A few years ago, I casually mentioned to a group of eating disorder clinicians that I regularly use this book in one of my undergraduate anthropology classes. The reaction was swift, intense and unequivocal: they were appalled. ‘You can’t use that book in a class!’ one therapist exclaimed. ‘It’s so triggering!’ ‘It’s like an instruction book for how to be anorexic,’ another added. ‘I hope you’re using trigger warnings!’ I knew Wasted gave a stark and heart-wrenching sense of what it’s like to live with an eating ­disorder – ­that’s precisely why I had chosen it. But I had not considered the book to be dangerous. And although I knew what a ‘trigger’ was from my own clinical work with clients diagnosed with eating disorders, it was the first I had experienced an injunction that I use a trigger warning as a form of ethical practice. A ‘trigger warning’ is a statement cautioning that content (in a text, video or class) may be disturbing or upsetting to some people, especially if they have experienced something similar like sexual abuse, mental illness, self-injury or suicide. While trauma triggers have long been recognized in the domain of post-traumatic stress disorder (PTSD), the concept of a ‘trigger warning’ as a part of public discourse is a more recent, emerging phenomenon that has gained steam over the past decade. A Slate Magazine blog identified 2013 as ‘the year of the trigger warning’ (Marcotte 2013). Television shows, podcasts and social media posts now regularly provide trigger warnings, and most universities in the US (including my own) have since developed explicit policies about whether and how to give trigger warnings in the classroom.



Trigger Warnings

35

Here, I consider the concept of triggers and the increasingly pervasive ethical imperatives to warn people about them as a way of thinking through contemporary grapplings with issues of the will and its weaknesses, including notions of personal responsibility, attributions of moral agency, and the politics of blame. Specifically, I want to explore how such constructions of the will relate to the phenomenon of akrasia. Akrasia can be defined as a lack of selfcontrol or weakness of the will wherein one acts against one’s better judgment (Steward 1998). My interest in this chapter is less in exploring akrasia philosophically, and more in considering how debates about akrasia and speculations about why someone might act in apparently self-undermining ways can emerge as central vectors of meaning in an ethnographic sense, through which different stakeholders make various sorts of truth claims. I get at these issues through examining how ‘triggers’ are conceptualized and enacted within a North American eating disorder clinic, where they are seen as both the cause of dysfunction and the currency of recovery. While this may seem to be a rather circumscribed focus, the clinical setting offers a close-up view of some of the broader issues and tensions at hand.

Anthropology in the Clinic First, however, a few words about my orientation to these issues. I am an anthropologist specializing in medical and psychological anthropology, and I am also a licensed clinical social worker. I have worked both ethnographically and clinically with individuals diagnosed with eating disorders for the past twenty-five years. My most recent project (Lester 2019) is based on seven years of engagement as both a clinician and ethnographer at an eating disorder clinic in the American Midwest that I call Cedar Grove. It is a place where practices of care and harm are entwined, often becoming indistinguishable. In the face of what seem to be insurmountable ­odds – ­clients who often don’t want to get better, families who are in crisis yet resist change, an illness that kills more people than any other psychiatric condition, and a healthcare system that devalues the very sorts of clinical expertise that seem to ­help – p ­ atients are routinely set up to fail as economic, affective and relational dynamics generate paradoxes and double binds within which both patients and clinicians become ensnared. Clients and clinicians alike get caught up in these knots and must struggle to find a way forward, racing against the clock and the spectre of disappearing insurance benefits to access treatment that is complex, contradictory a­ nd – i­ n far too many ­cases – ­counterproductive. It is in this context that I became interested in how clients’ apparently selfdestructive or self-defeating behaviours were interpreted within the clinic. These interpretations had direct implications for how eating disorders and eating disorder clients were understood, and carried practical consequences

36

Rebecca J. Lester

for clients’ treatment trajectories (whether a client was allowed to use the restroom unsupervised, was to be granted a pass, or given permission to use the internet, for example, as well as how long a client would remain in treatment and at what level). The question of a client’s ability to ‘make healthy choices’ or ‘reach for recovery’ was paramount not only in clinical decision making about treatment, but also in how clients themselves were perceived as moral agents. This foregrounding of patient ‘choice’ as a moral issue is directly tied to the logics and economics of managed healthcare delivery in the contemporary United States, shorthanded as ‘managed care’. Managed care, which came to dominate the American healthcare landscape in the 1990s, is, first and foremost, an economic model: a system of organizing and rationing health care services within a capitalist system where market forces determine both the cost and value of those services. Built on a rational choice model of human action with profitability as the ultimate good, managed care is predicated on the standardization of a product (health care) across domains, a regulation of the provision of that product, and a rationing of the supply of the product in order to maximize economic profit and minimize loss. To do this, managed care organizations (MCOs), such as Blue Cross, Aetna, United Healthcare, and so forth, contract with hospitals and providers to offer services to subscribers at reduced costs; what Donald (2001) calls ‘the Wal-Marting of American psychiatry’. Importantly, almost all managed care companies are for-profit enterprises. This means they are ultimately beholden to their shareholders, not to the patients whose healthcare they manage. Profits come from taking in more healthcare premium dollars than are spent on healthcare treatments, incentivizing managed care companies to deny approvals for treatment unless deemed absolutely necessary (as determined by the managed care company, not by doctors). One reason for denying care is the suspicion that a patient might not fully comply with medical recommendations, or a belief that certain kinds of patients are not likely to get well. According to managed care logics, approving treatment in such cases would make poor economic sense. Against this backdrop, eating disorder clients are notorious for being ‘resistant’ to treatment and for undermining and undoing care, as many push back against clinical recommendations or even profess wanting to remain ill. Leaving aside for the moment the possible critiques of this perspective (see Lester 2019 for a fuller discussion), clinicians at Cedar Grove often found themselves confounded by the mismatch between insurance company expectations and what clients actually presented on the ground. In the US managed care system, insurance companies will only continue to pay for treatment if it achieves demonstrably positive results according to the insurance company’s own criteria. These criteria may or may not map onto clinical indicators of improvement or success. For example, an insurance company may measure treatment success by whether or not a client consistently gains weight at a significant and steady pace (e.g. two to three pounds per week). Yet, weight



Trigger Warnings

37

gain can be irregular, especially with lower weight clients, regardless of compliance with the program. And any clinical progress might be punctuated with lapses (bingeing on pass, self-harm, skipping a snack) or other disruptions, like conflicts with staff or not engaging actively in therapy. When these sorts of things occurred, clinicians were forced to develop ad hoc understandings of the subject and human motivation that could account for clients’ apparently akratic behaviour (a client who skips a snack knows she is ‘supposed’ to eat it and that doing so will help her get well; one who struggles with staff knows this will ultimately harm her recovery) while also maintaining that treatment was having a positive impact. Nowhere was this challenge more pronounced and paradoxical than in the case of triggers.

Triggers in Eating Disorders The notion of triggers is central to contemporary western understandings of eating disorders and is foundational to the treatment process at Cedar Grove. In the eating disorder world, a ‘trigger’ is anything that, as one Cedar Grove clinician put it, ‘activates the eating disorder’ or ‘makes someone want to engage with their eating disorder’. This could come in the form of imagery (photos in a magazine or online, for example), words (someone talking about their recent diet or how much weight they’ve been losing) or any sort of numbers (numbers on a scale, lab results, calorie counts listed in menus). It can come from activities like trying on clothes at a clothing store, going grocery shopping or looking in a mirror. A hallmark of being ­triggered – ­in an eating disorder or anything e­ lse – i­ s that the person feels out of control, compelled to act in ways they don’t want to, may actively resist, and may even find abhorrent. Triggers can even lead people to self-destruct. Because of this, triggers are generally thought of as something negative, something to be avoided, or perhaps even be protected ­from – ­hence the notion of trigger warnings. But there is another side to triggers when it comes to eating disorders. Unlike in something like PTSD, where people often go to great lengths to avoid sights, sounds or situations that might be potential triggers, people who are struggling with an eating disorder do not always avoid eating disorder triggers, and they may even seek them out specifically in order to activate or intensify the eating disorder. That is, they may purposefully trigger themselves in order to fuel the eating disorder and make it stronger. So compelling is this practice that the internet is populated with scores of pro-anorexia and probulimia websites with names like ‘Little Baby Nothing’ and ‘My Skinny Dream’ that are full of ‘thinspirational’ pictures and other self-consciously triggering material that exist precisely so that visitors can fuel their eating disordered urges. In accessing such sites, many people are specifically seeking to be triggered. They know they are doing something harmful according to dominant scales of value, but doing so also involves cultivating other kinds of value that

38

Rebecca J. Lester

are inseparable from these very registers of harm; it inches them toward particular kinds of g­ ood – t­ heir own deprivation and self-­destruction – t­ hat they value more than health (Lester 2014).

Triggers and Treatment As is common in eating disorders, many clients at Cedar Grove are ambivalent about recovery. They are therefore ambivalent about triggers, fearing them, yet sometimes seeking them as they struggle towards health. Kendra, for example, a twenty-year-old client diagnosed with bulimia, was regularly caught accessing ‘inappropriate’ material on her phone (mostly calorie lists and exercise videos) and Jenna, a thirty-two-year-old client diagnosed with anorexia, tried on her ‘anorexic clothes’ while home on a pass because she felt she was beginning to enjoy food too much. Clients’ love/hate relationship with eating disorders and the hair’s breadth that often separates these two experiences can make treatment dicey, in part because treatment at Cedar Grove and most similar clinics is deliberately built around triggers (or, at least, certain kinds of triggers) and shaping how clients responds to them. While in treatment, clients at Cedar Grove are largely protected from the outside world and are prevented from viewing certain images or media, knowing their weights or talking about stressful topics during meals. The idea is to make the clinic a safe space and to reduce the external stimuli that can trigger eating disorder urges. But, at the same time, most c­ linicians – a­ nd certainly those at Cedar ­Grove – ­would agree that a key component of treatment is to bring forward the affects, emotions and cognitions that fuel the eating disorder and to help clients develop more adaptive ways of managing them. To do this, the eating disorder must be brought front and centre. To this end, clients engage in four to six hours a day of group therapy sessions on topics such as body image, sexuality and femininity, where uncomfortable and triggering material is viewed and discussed, and clients are mentored in processing the feelings and emotions that arise. They also have individual therapy sessions three times a week where they talk about family dynamics, life stresses and intense affects like fear, shame, anger and loss while receiving individualized guidance in how to modulate and manage the feelings that arise. And they must face their ultimate t­ rigger – ­food – six times a day through structured meal and snack times, always accompanied by the encouragement to process what came up for them while eating. Through all of these mechanisms, a key function of treatment is, specifically, to trigger the kinds of feelings and emotions that fuel the eating disorder to ‘rile it up’ – but also to prevent the client from acting on these desires through eating disorder behaviours while they learn new ways of managing the resulting feelings and impulses. This is not always successful, and sometimes clients are catapulted back into problematic



Trigger Warnings

39

behaviours. Clinicians must then figure out how to make the case to the client’s insurance company that, despite appearances to the contrary, she is actually making strides towards health.

Refracted Selves, Distributed Wills To manage these ambivalences and risks, clinicians at Cedar Grove work with a model of the person that distinguishes the individual from her eating disorder and construes these two entities as diametrically opposed ‘wilful subjects’ (Ahmed 2014). This is a standard orientation in western eating disorders treatments, and Cedar Grove is by no means unusual in this regard. Where Cedar Grove does stand apart is the degree to which this bifurcation of ‘the person’ and ‘the eating disorder’ structures language, behaviour and experience in the clinic and how this becomes crucial to treatment management in the context of managed care. As clients progress in treatment at Cedar Grove, they are mentored in personifying their eating disorder as something separate from their true selves, a sub-personality with a will of its own, which (following Schaefer and Rutledge 2003) they call Ed (for Eating Disorder). They are taught to think of Ed as a partner in an abusive relationship, who is toxic, damaging and malignant, and who will necessarily lead them towards undermining their own best interests if left unchecked. At the same time, they learn (somewhat paradoxically) to view Ed as, in fact, trying to help them, even as he1 is leading them down a path of self-destruction. That is, they learn that eating disorders emerge as strategies for coping with things such as dysregulated emotions and affects, unhealthy interpersonal relationships, and feelings related to trauma, anger and grief. Over time, these strategies coalesce into practices that then become neurologically, physiologically, emotionally and interpersonally entrenched, and around which clients build a sense of identity. This makes ‘giving up’ an eating disorder something far more complicated and difficult than simply recognizing that it is harmful: because that is not all it is. It both helps and harms in different ways at different times, and clients often develop elaborate calculi to account for how engaging in the eating disorder ‘makes sense’ within their personal ethical weighting of goods and harms. Whether an eating disorder is understood as self-protective, self-destructive or both, the framing of Ed as a wilful subject separate from the ‘true’ self-complicates understandings of ‘the will’ and moral responsibility for disordered eating behaviours such as bingeing, purging, compulsive exercise or calorie restriction.2 Cedar Grove’s philosophy is that it is not really the client  herself who does these things: it is her eating disorder that does them. Clients learn that Ed is the one who has caused them to isolate from friends, obsessively calculate calories and cruise pro-anorexia websites. He is the one who pushes her to obsessively check her body in the mirror every

40

Rebecca J. Lester

morning for any sign of weight gain, to cut up her food into miniscule bites and to eat them at precise four-minute intervals. Ed makes people mistrust the dietician, keep secrets from their therapists and feel jealous of the new patient who is so thin and weak she needs a wheelchair. Through the language of Ed, clients learn that what they thought was willpower – obsessively counting calories, pacing bites of food at four-minute intervals during m ­ eals – i­s actually its opposite: a weakness of will where Ed overpowers and controls them, leading them to act against their own best interests, to painfully and thoroughly self-destruct. On the whole, clinicians and families seemed to have very positive orientations toward the idea of ‘separating out’ the person from the eating disorder and identifying the illness as a juxtaposition of separate wills. ‘It was so much easier to relate to her after we learned to think of it that way,’ said Betsy, referring to her daughter Miranda, who was in recovery from bulimia: An eating disorder is like a kidnapping. Seriously. Like someone has come and kidnapped your child. And you can’t negotiate with the kidnappers. There’s no point. You have to be absolute, and take action. That’s the way it is. It helped enormously to know our daughter was in there somewhere just being held, basically, against her will. That made it easier to deal with the way she was acting and all the anger and ugliness she was directing towards us.

Clients, however, had a different sort of experience. They often became frustrated at the way talk of Ed was at times deployed against them to delegitimize their claims either to agency or to self-knowledge (or both). Kendall, a twentyone-year-old woman diagnosed with anorexia, had quite a bit to say on this subject: I guess, a lot of times, I feel like they think we don’t have a mind of our own or that we’re only our eating disorders and we’re not. We’re each individual human beings who deserve to be treated as individuals … I’m not just my eating disorder, and just because I disagree with something it doesn’t mean it’s my eating disorder talking. Sometimes it is, but usually it’s actually not. But there’s no way to win that argument because the more you argue, the more they say it’s evidence that Ed is controlling things. It’s pretty infuriating.

Julie, a thirty-two-year-old woman diagnosed with bulimia agreed: Nobody’s listening to me. They say it’s my eating disorder talking. It feels like every time I try to claim my voice and make my own decisions, people just kind of shoot me down or doubt me or tell me that I’m malnourished. And I know I’m malnourished. But I’m still a person and these are still my feelings.

For Kendall and Julie (whose views were echoed by many other clients), having any of their thoughts, feelings or intentions that didn’t fit clinicians’ preferences dismissed as ‘part of the eating disorder’ led them to feel profoundly



Trigger Warnings

41

unseen. Clients generally felt there was no way to effectively speak back to treatment practices without it being dismissed as ‘coming from the eating disorder’. It often felt to them as if ‘Ed’ talk became a central vector of delegitimation and a subversion of their attempts to be recognized as full and legitimate subjects in their own right.

Akrasia and Situated Subjects Here, we see a key component in thinking about akrasia that such an ethnographic perspective can bring: not only is akrasia not always clear-cut due to competing systems of value, but it must also be considered from a temporal and developmental perspective. A practice such as an eating disorder that was experienced as self-protective at one time (or, at least, more self-protective than other things going on) might become self-defeating at a later point, or vice versa. We might say, then, that akrasia cannot be discerned from a observing a behaviour, or even from a person’s intentions, but must rather be considered within the larger context within which the behaviour and the intention co-emerge. This brings us to a key issue. Akrasia, in the classic sense, seems to presume that thought and rationality are the precursors to behaviour and that ‘best judgment’ is based on assessments of what will lead to the most benefit and the least harm. It further presumes that such assessments are self-evident, and that, once made, one has the capacity to act (or not) upon these assessments as one chooses. The notion of ‘best judgment’ then, and the possibility of akrasia, are both predicated on presumptions of clear and unfettered perception and a freedom of action unencumbered by structural limitations. These are problematic premises on a number of fronts. The first, and most obvious, is that we don’t live in a world where ‘best judgments’ are always simple or clear cut, nor is pursuing them always available to everyone in the same ways or with the same goods and risks at stake. In her book Choosing Unsafe Sex (1995), for example, anthropologist Elisa Sobo explores why some urban African American women in the US continued to have unprotected sex during the height of the AIDS crisis despite public awareness campaigns about how condoms can prevent HIV transmission. Public health officials insisted that these women clearly didn’t understand the risks involved and they needed more health education. What Sobo found, though, is that her interlocuters understood very well the risks of getting AIDS and that condoms could reduce these risks: they simply judged other risks to be more acute, and other benefits to be more imperative. Similarly, anthropologist Holly Wardlow (2006) sought to understand why increasing numbers of young women in New Guinea were leaving their families to become ‘passenger women’ (prostitutes), despite the dangers and intense stigma attached to that path. Her conclusion is that doing so enabled these women to reclaim a sense of ownership over their own

42

Rebecca J. Lester

sexualities and to escape networks of kin obligation that otherwise limited their life options. In other words, actions always involve moving both towards and away from potential risks as well as potential goods. This makes akrasia as much a feature of context or situation as a question of judgment per se. If we were to diagram this out visually, it would look more like a threedimensional, dynamically shifting web than a bifurcated path (i.e. acting either for or against one’s own interests). How actions sit in relation to one’s best judgment (and what that even means) are not only culturally and historically contingent, but deeply inflected by gender, race, class and other distinctions that affect both the contours of this web, as well as the degrees to which individuals can move within it and with what consequences. This, in turn, informs what kinds of goods are available, attainable, and sustainable for that person, and what kinds of risks they deem worth taking. This question of discernment brings us to another important point. While philosophical engagements with akrasia traditionally focus on things like interests and judgments, in real life, these calculi are not always (or only) experienced as cognitive or rational processes. Often, we ‘go with our gut’ or ‘do what feels right’, even if we can’t always articulate why a given choice is the one we selected, and even if it doesn’t seem to make logical sense. In other words, the ‘best judgment’ against which one may be acting in akrasia is not necessarily always cognitive or rational in the strict sense of the term, and may be derived from a whole host of cues, including sensory, phenomenological and interpersonal ones. But more than this, behaviour that looks akratic from the outside (or even from the inside) may be fully in line with one’s more embodied sense of the world and what kinds of values derive from that embodied experience. This is important because it complicates the question of judgment in akrasia. Akratic behaviour becomes somewhat less perplexing if we consider that there are multiple ways of knowing, multiple ways of discerning risks and rewards, and multiple scales of value simultaneously at play. Certainly, a feeling that one ‘can’t help it’ when engaging in self-destructive behaviour (in compulsions to self-harm, for example) can seem to suggest one’s judgment is faulty or has been subverted. But if we take seriously some of what I’ve just ­said – ­namely, that all actions entail a simultaneous moving towards various ‘goods’ and away from various risks across a range of scales of values; that akrasia is temporally, culturally and situationally relative; and that intention and action co-emerge within interpersonal, embodied ­worlds – ­then we might view issues of judgment in akrasia not so much as correct or incorrect, but rather, simply, multiplex. For example, consider the following scenario: Carrie, in treatment for anorexia, begins to eat all of her meals without protest and is advancing toward her weight goals. We could say that Carrie is performing the acts of recovery and is therefore authentically progressing in her treatment and acting in her best interests. But let us add this additional information: Carrie and Colleen,



Trigger Warnings

43

another patient diagnosed with anorexia, have made a pact to get out of treatment as quickly as possible, move into an apartment together, and jointly resume their eating disorders with a vengeance. How, then, should we understand Carrie’s apparent ‘compliance’? Is it still a step toward heath? Or, in this case, is performing health actually an akratic move? This calls to mind philosopher Harry Frankfurt’s distinction between what he calls the ‘willing’ and ‘unwilling’ addict (Frankfurt 1971). In Frankfurt’s terms, both willing and unwilling addicts have a first-order desire to use a drug, say, heroin. The willing addict, Frankfurt says, knows she wants to use heroin, and she wants to want to use it; that is, she embraces her desire for the drug. The unwilling addict wants to use heroin but does not want to want to use it. She feels compelled to use heroin due to whatever physiological, psychological or social reasons that fuel her addiction, but she does not want to have this d ­ esire – s­ he wants to be free of it. We have, then, two people engaging in the same action (using heroin) but with two very different orientations. In terms of eating disorders, Frankfurt would make a clear distinction between a person who restricts (or binges and/or purges), and wants to want to do so (in the clinic, they would say this person is ‘in’ her eating disorder) and one who engages in these behaviours but does not want to want to do so (in clinic terms, this person would be ‘struggling’ or ‘in recovery’). But Frankfurt’s model assumes a subject whose desires and their relation to her behaviours are relatively ­consistent – i­f I do not want my eating disorder now, I won’t want it an hour from now. And if I eat my 10:00 a.m. snack because I want to want to eat it, then eating my lunch indicates a similar investment. But intentions and behaviours are not always consistent, particularly for someone recovering from an eating disorder. First- and second-order desires can align, diverge and realign multiple times per day (or even per hour) as the person struggles through complex issues of being-in-the-world. What does this mean in terms of triggering, akrasia and questions of judgment? In thinking about this, I find Musolino and colleagues’ (2018) perspective to be useful. They consider triggering less as an on/off switch activating behaviour and more as a circulation of desire (Musolino, Warin and Gilchrist 2018, 534), and as a tool of igniting that desire (ibid., 535). But this desire, they say, imperfectly conforms to expectations, ‘produc[ing] movements of desire that are experienced as ambivalent and compelling, and rife with holes, gaps, and contradictions’ (Musolino, Warin and Gilchrist 2018, 535). As a result, argues Lavis (2013), ‘narratives of being ‘triggered’ [in eating disorders] draw attention to informants’ subjectivities and the many and ­continual – ­e ven ­daily – ­shifts in the relationship between personhood and anorexia’ (2013, 46). Triggering, then, might be less of an issue of a binary on/off issue of whether one acts according to one’s best judgment or against it, and more an occasion through which certain forms of ­desire – ­and concerns about that ­desire – ­come to the fore, and through which different kinds of realities are engendered

44

Rebecca J. Lester

and foreclosed. This then opens up lines of inquiry that can begin to account for the contextual dimensions of akrasia.

Triggering, Akrasia and Desire Suzanna A brief ethnographic example can be illustrative. At the time of these events, Suzanna was twenty years old. She had been transferred to Cedar Grove directly from the hospital where she had been treated for kidney failure because of her bulimia. Prior to the hospitalization, Suzanna had been bingeing and purging for up to ten hours a day. She spent hundreds of dollars a week on binge food and had even been engaging in sex work to get money for her binges. She also struggled with self-harm and had scars on both forearms from years of self-inflicted razor blade cuts. She had a long history of depression and frequently felt suicidal, though she had never actually attempted. By the time she made it to Cedar Grove, Suzanna was distraught. She had been in the hospital for two weeks straight and her urges to binge and purge were extremely high. She felt and acted like an addict desperate for a fix. She paced, she shook, she couldn’t concentrate because of intrusive thoughts about food. She was a wreck. Eventually, she figured out a way to purge in secret at the clinic and began to do so regularly. She was caught when Ziploc bags full of vomit were discovered hidden under her bed. Over the next several weeks, she continued to find new ways to purge, but began to come to staff directly afterwards to tell them what she had done and to process what had prompted the behaviour. She participated in groups, saw her therapist, and took her medications. Gradually, although she still struggled with strong urges, Suzanna’s episodes of purging began to decrease. She continued to be invested in treatment and gradually developed new strategies for modulating her urges. At one point, three months into her treatment, she went seventeen days without bingeing or purging, the longest she had gone in over six years. She began, for the first time, to feel optimistic about the possibility of recovery, even though it still seemed distant. It was at this point that Suzanna’s insurance company determined that her condition was no longer ‘acute’ and she should be discharged from treatment. When Suzanna heard the news, she panicked. ‘I can’t leave treatment!’ she told me, sobbing. ‘I’m not ready! If I go out there, I know things will go back to the way they were. I can’t go back to that life!’ The evening of this decision, Suzanna purged for the first time in over two weeks. The following day while on a pass, she spent the entire three hours binging and purging. She cut. She became suicidal. The Cedar Grove staff initiated three insurance appeals on Suzanna’s behalf, all of which were denied. When pressed by Dr. Casey, the clinical director, for an explanation, the insurance case manager said that



Trigger Warnings

45

Suzanna’s reaction following the initial denial led them to determine that Suzanna was ‘borderline’ (that is, she had borderline personality disorder, which is characterized by chronic self-destructive impulses) and, because, in his words, ‘you can’t treat borderlines’, they were no longer authorizing coverage. Here, the insurance case manager is making an interesting argument. He maintains that, because (prior to the insurance denial) Suzanna’s acute symptoms of binging and purging had abated, her ‘will’ had been effectively ­rehabilitated – s­ he had gained control of her self-destructive behaviours, and the cessation of symptoms marked the end of the present illness episode. She was therefore no longer eligible for care. But he was now faced with a client who was clearly spiralling and unsafe. How, then, to justify denying care in such a situation? He does this by reaching to a different understanding of the will. As I’ve noted, wanting recovery, or even wanting to want recovery, presumes a subject that is relatively stable. The hallmark of borderline personality disorder is precisely an instability or inconsistency of the subject, such that emotional and cognitive states are always in flux. By characterizing Suzanna as ‘borderline’, then, the case manager is signalling that he does not think Suzanna has the capacity to ‘will’ anything in particular, at least not for very long, and that she will therefore necessarily gravitate toward self-destruction. This akratic orientation is so profound and systemic, he suggests, that treatment is hopeless, similarly indicating for him that they should not pay for further care. The Cedar Grove clinicians strenuously disagreed with the insurance company’s assessment. The cessation (while in treatment) of Suzanna’s eating disorder symptoms did yet not represent, in their view, an authentic shift in Suzanna’s state of being. She was, simply, contained. If they could keep Suzanna engaged in recovery, they argued, she could continue to get better. But withdrawing treatment support prematurely and not treating Suzanna for these more chronic, underlying issues as the managed care company insisted doomed her to ­relapse – ­not because she is essentially akratic, but because she was ill. This is an important distinction because they figure the moral subject in very different ways: as either incapable of getting well, or as on the pathway to recovery. In the end, Suzanna was discharged, and within a month, was in full relapse. But because she had relapsed, the insurance company determined that she had ‘failed’ treatment and denied further care. What we see here are struggles over how to think about Suzanna’s will in relation to her eating disorder, as well as her willingness (or not) to get well. It seems clear that Suzanna’s desire for recovery was tentative at best and was easily co-opted by her eating disorder. Everyone agreed on that. Where the parties diverged was to what extent they considered Suzanna’s akrasia to be situational or constitutional, rehabilitable or not. And central to this was the degree to which they considered Suzanna to be a subject who had the capacity for forward-reaching action.

46

Rebecca J. Lester

Suzanna’s reactions to triggers was thought to be diagnostic of these issues. Clinicians saw Suzanna’s acting out after the insurance denial as clear evidence that she needed more treatment. The insurance company, on the other hand, viewed Suzanna’s acting out as precisely the opposite, as indicating a deeper and more intractable personality disorder where clients’ intentions were inconsistent, unreliable and not to be trusted. This was evidence for them that treatment should end. Appeals to different orders of ‘willing’ enabled clinicians and case managers to inhabit different positions at different times as they sought the particular outcomes they want to achieve. But this was not merely an intellectual exercise. In a context where a patient’s perceived willingness to strive for health and her capacity to successfully persist in this aim, despite challenges, are defining features of ‘medical compliance’, and therefore deservingness for care. Struggles like those Suzanna experienced become not only diagnostic but carry direct practical consequences for the ability to access care. In Suzanna’s view, the way to get more care was to demonstrate that she was still ill. What, in the end, signals her relapse as akratic behaviour is not so much her intention (which was oriented towards wanting to get help) or even whether she was acting according to her best judgment, but, rather, the ultimate outcome, whose determination rested in others’ hands.

Concluding Reflections So, what can these discussions about triggering in an eating disorders clinic contribute to considerations of akrasia? Here, I will gesture at two key points. First, it seems clear that people can have multiple ‘goods’ to which they aspire, which may or may not be consistent with one another, and that what constitutes someone’s best interests at any particular juncture is an interpretive determination, not an absolute one. It depends on who is asking and on the scale(s) of value, as well as the scope of inquiry. It also depends on the positioning of different moral actors in relation to one another, and who has the power to determine not only outcomes but truths. Second, how we think of the will and its weaknesses depends fundamentally on how we figure the relationships between subjects and worlds, and this is as much an ethnographic question as a philosophical or theoretical one. Contemporary notions of t­ riggering – ­including those at Cedar G ­ rove – ­figure a ‘trigger’ as something ‘out there’, separate from the subject; a specific something which is either avoided, tolerated, or reacted to. But what if, instead, the ‘trigger’ is so diffuse as to elude simple identification and bounding? What if the relationship between the subject and the trigger is similarly blurry? Or, what if the trigger is specifically that which defines a subject as a subject? What was ultimately the most triggering for clients at Cedar Grove was not the therapy, or the food, or the talk of body image and sexuality, but the way



Trigger Warnings

47

care itself was conceptualized and operationalized, and how this constituted them as ‘patients’ as a result. In such a context, what does it mean to reach for health, if in doing so you participate in your own existential erasure? And is it akratic to reach for that which harms you, if that also constitutes a pathway to recognition? This is kind of the flip side of what Lauren Berlant (2011) calls cruel optimism, or a state of affairs when the thing you desire is actually an obstacle to your flourishing, when you are induced to long for something that harms you. An attachment to an ­object – a­ person, a goal, a state of ­being – ­becomes cruel, Berlant says, ‘when the object that draws your attachment actively impedes the aim that brought you to it initially’ (2011, 1). In other words, cruel optimism persuades us to reach for something that, in the end, undoes its own impetus and causes us harm. This formulation, of course, problematizes the notion of akratic ­action – ­or, rather, embeds questions of akrasia within structures of inequality and raises questions of whether reaching toward ‘the good’ is always in our best interests, and whether reaching toward ‘harm’ is always necessary against those interests. This brings me back to the issue of using the book Wasted in the classroom. The clinicians who challenged this decision viewed the book itself as a trigger, given its explicit descriptions of eating disorder thoughts and behaviours. The trigger, for them, rests in the object itself (the book) and what kinds of akratic actions it might ignite in the reader. In my view, however, the more powerful triggers for eating disorders are shame, ignorance and misrecognition in the intersubjective sense, and reading a book like Wasted can help people struggling with disordered eating to feel seen and acknowledged in ways too often obscured by the very systems of care that purport to help them. Warning students about an impending trigger is not only about respecting and caring for vulnerable students. It also effectively transfers moral responsibility to the individual for modulating her or his own engagement with the material, raising the question of whether trigger warnings are primarily about preventing akrasia or if they are instead more about issues of legal liability. Regardless, faulting a book like Wasted, or celebrity culture, or proanorexia w ­ ebsites as triggers for apparently akratic action in eating disorders is admittedly easier than calling out and actively addressing things like everyday gender dynamics, structural and symbolic violence, and targeted erasures of minoritized bodies and subjects. But doing so cripples the capacity to engage proactively in the kinds of conversations that are essential for affecting change. I will teach this class again next spring, and Wasted is most definitely on tap. Rebecca J. Lester is Professor of Anthropology at Washington University in St. Louis and a licensed clinical social worker specializing in eating disorders,

48

Rebecca J. Lester

trauma, personality disorders, mood disorders and gender/sexuality issues. In 2010, she started Lumen Psychotherapy, a 501(c)(3) nonprofit organization that offers low-fee mental health services in the St. Louis area. She is the author of the award-winning book Jesus in Our Wombs (2005) as well as numerous academic articles. Her most recent book, Famished: Eating Disorders and Failed Care in America (2019), recognized with a Victor Turner Prize for Ethnographic Writing, presents a profound rethinking of eating disorders and examines how a system set up to treat these illnesses ends up perpetuating them.

Notes   1. Ed is always gendered male in this discourse. Other personifications of anorexia and bulimia as female (as Ana and Mia, respectively) also circulate, but are usually associated with pro-eating disorder orientations. When the eating disorder is conceptualized as something harmful, it is almost always gendered as male, at least for female sufferers. I am not aware of any work that has looked at how eating disorder personification might differ for male clients.   2. It is worth noting the striking similarities between this formulation of Ed and anthropological studies of spirit possession. Janice Boddy (1988), for example, proposes a model of ‘accommodating agency’ in Zar spirit possession wherein women’s relationships with their possessing spirits allow them creatively to engage with and work through competing models of who they are and who they should be and to come to largely individualized solutions. Through developing an agency predicated on accommodating (but not surrendering to) the cohabitating spirit, women learn how to live with their possessing spirits; to even collaborate with them. In so doing, they can paradoxically inhabit contradictory value systems as they carve out for themselves ways of being in the world otherwise foreclosed to them.   I find this line of analysis useful for thinking about what happens at Cedar Grove. Clients at Cedar Grove develop a form of accommodating agency with Ed akin to that described by Boddy. Both recovery at Cedar Grove and Zar possession entail the person identifying a force other than herself who has co-opted her will and bent her behavior to fit its own needs and wants. Both provide a language for talking about the relationship between the person and this entity and a framework for understanding a path to healing. And both hinge on the notion that agency is complex, situational and contested. Just as Boddy’s informants had to learn how to be possessed, clients at Cedar Grove must learn how to live in relationship with Ed. And Ed is thought to never entirely go away. Clients must learn how to live in an accommodating relationship with him for the rest of their lives.

References Ahmed, Sara. 2014. Wilful Subjects. Durham, NC: Duke University Press. Berlant, Lauren. 2011. Cruel Optimism. Durham, NC: Duke University Press. Boddy, Janice. 1988. ‘Spirits and Selves in Northern Sudan’. American Ethnologist 15(1): 4–27. Donald, Alasdair. 2001. ‘The Wal-marting of American Psychiatry: An Ethnography



Trigger Warnings

49

of Psychiatric Practice in the Late 20th Century’. Culture, Medicine, & Psychiatry 25(4): 427–439. Frankfurt, Harry. 1971. ‘Freedom of the Will and the Concept of a Person’. Journal of Philosophy 68(1): 5–20. Hornbacher, Marya. 2014. Wasted: A Memoir of Anorexia and Bulimia. New York, NY: Harper Perennial. Lavis, Anna. 2013. ‘The Substance of Absence: Exploring Eating and Anorexia’. In Why We Eat, How We Eat: Contemporary Encounters between Food and Bodies, edited by Emma-Jayne Abbot and Anna Lavis, pp. 35–52. New York, NY: Routledge. Lester, Rebecca. 2014. ‘Health as Moral Failing: Medication Restriction Among Women with Eating Disorders’. Anthropology & Medicine 21(2): 241–250 . 2019. Famished: Eating Disorders and Failed Care in America. Berkeley, CA: University of California Press. Marcotte, Amanda. 2013. ‘The Year of the Trigger Warning’ [blog]. Slate Magazine. Available at: https://https://slate.com/human-interest/2013/12/trigger-warningsfr​om-the-feminist-blogosphere-to-shonda-rhimes-in-2013.html [Accessed August 1, 2020]. Musolino, Connie, Megan Warin, and Peter Gilchrist. 2018. ‘Positioning Relapse and Recovery Through a Cultural Lens of Desire: A South Australian Case of Disordered Eating’. Transcultural Psychiatry 55(4): 534–550. Shaefer, Jenni, and Thom Rutledge. 2003. Life Without Ed: How One Woman Declared Independence from her Eating Disorder and How You Can, Too. New York, NY: McGraw-Hill Education. Sobo, Elisa. 1995. Choosing Unsafe Sex: AIDS-Risk Denial Among Disadvantaged Women. Philadelphia, PA: University of Pennsylvania Press. Steward, Helen. 1998. ‘Akrasia’. In Routledge Encyclopaedia of Philosophy. Taylor and Francis. https://www.rep.routledge.com/articles/thematic/akrasia/v-1. Wardlow, Holly. 2006. Wayward Women: Sexuality and Agency in a New Guinea Society. Berkeley, CA: University of California Press.

2 Three Problems with the Addiction as Akrasia Thesis That Ethnography Can Solve Darin Weinberg

In his justly famous intervention into the debate as to whether weakness of the will is possible, Donald Davidson (2001, 29) wrote, ‘There is no proving such actions exist; but it seems to me absolutely certain that they do’. I agree. The fact that there is no proving, in Davidson’s strict sense, such actions exist, though, advises a more inclusive, and pragmatic, assessment of the value of adopting this interpretation of action in any given instance. In what follows, I  take a broad look at the value of interpreting addiction, in particular, as a form of weakness of the will, and/or akrasia. I then consider three problems that arise from adopting the specific views defended by those who have explicitly made the case for this thesis, as well as some of their less explicit fellow travellers. To be clear, I am certain that some drug problems do in fact exhibit all of the characteristics I myself would associate with akrasia. However, the thesis that addictions ought to be generally understood as a form of akrasia seems to me to be vulnerable in certain important respects. The first section of the chapter provides an overview of the current state of debate surrounding the addiction as akrasia thesis. The work of the distinguished clinical psychologist Nick Heather is given pride of place, as it is he who has provided by far the most methodical and comprehensive defence of this thesis. However, the work of other important contributors to this and related debates in the addiction science literature is also critically considered. The chapter then elaborates on three problems with this thesis as it is currently formulated that I believe can be effectively addressed ethnographically. The first problem is that this thesis too often posits the rational unity of properly functioning or healthy self-control as an integrated source of evaluation and volition. There are very good reasons to believe to the contrary, though, that properly functioning healthy people exhibit varying degrees of rational



Three Problems with the Addiction as Akrasia Thesis

51

unity and disunity that are often explicable sociologically. The second problem is that this thesis too often posits self-control as invariably an exercise in emotional restraint, response inhibition or delayed gratification. Once again, however, there are very good reasons to believe that self-control is also exercised through self-discovery and self-actualisation, which are not so obviously opposed to emotional expression, disinhibition and personal gratification. Finally, the addiction as akrasia thesis tends to undertheorize the intrinsic relationship between experience, evaluation and volition and the social contexts within which these are shaped, stabilized, stimulated and sustained. The chapter concludes with some brief reflections on the ramifications of these arguments for addiction science and treatment.

The Addiction as Akrasia Thesis In a series of important essays, Nick Heather, sometimes with his colleague Gabriel Segal, has developed the argument that addiction is beneficially understood as a particularly extreme form of akrasia (cf. Heather and Segal 2013, 2015; Heather 2017, 2020). Drawing selectively from each of these essays, I outline both the arguments they have made and their specific intentions in doing so. Most fundamentally, these arguments have been developed to provide philosophical ‘clarification regarding the conceptual foundations on which a theory of addiction should be based’ (Heather and Segal 2013, 445). While the first two essays by Heather and Segal remain largely agnostic regarding the best theor(ies) with which to scientifically define and explain addiction, the later two essays, written by Heather alone, link the addiction as akrasia thesis to what are commonly called ‘dual process’ or ‘dual systems’ theories of addiction (Heather 2017, 2020). In this section, I attend to both the philosophical arguments concerning addiction as akrasia and the relationship Heather and others have drawn between these arguments and dual process and/or dual systems arguments concerning human decision making generally and in the case of addiction specifically. The core objective of these arguments is to develop a conceptual middle ground between the arguments one often finds, on the one hand, among brain disease theorists that addiction is an overwhelming or irresistible compulsion the addict has no ability to control and, on the other hand, arguments that addiction is no more than a myth because behaviour deemed addicted is, as it turns out, not scientifically distinguishable from ordinary freely chosen activities. Disease theorists can’t easily provide for the fact recurrently demonstrated in psychological and social scientific studies that most behaviour attributed to addictions is ‘incentive sensitive’ (consistent with rational costbenefit analysis) insofar as research subjects can be incentivized to abstain from the activities to which they are ostensibly addicted. Whereas those commonly known as ‘choice theorists’ can’t easily provide for the ubiquitous

52

Darin Weinberg

tenacity of addicted patterns of behaviour, even in the face of sometimes severely harmful consequences. Heather is persuaded by these critiques of both received disease and choice theories of addiction and has sought, in the concept of akrasia, an intellectual path to overcoming them. So what, then, is akrasia? Heather and Segal (2013, 447) write, ‘akrasia occurs when someone acts intentionally counter to his own best judgment; in these circumstances we often say that the person lacks the willpower to do what he knows or believes to be better to do’. They draw heavily upon Donald Davidson’s argument in favour of the logical possibility of akrasia. Davidson defines akrasia, or what he refers to as weakness of the will or incontinent action, as follows: In doing x an agent acts incontinently if and only if: (a) the agent does x intentionally; (b) the agent believes there is an alternative action y open to him and (c) the agent judges that, all things considered, it would be better to do y than to do x. (quoted from Heather and Segal 2013, 448)

Heather and Segal (2013) proceed to emphasize that, for Davidson, it is necessary that the agent continues to judge that it would be better to do y at the time s/he does x, and they follow Davidson in this requirement.1 They then go on to formulate the philosophical problem Davidson seeks to solve in terms of reconciling three propositions, P1 If an agent wants to do x more than he wants to do y and he believes himself free to do either x or y, then he will intentionally do x if he does either x or y intentionally. P2 If an agent judges that it would be better to do x than to do y, then he wants to do x more than he wants to do y. P3 There are incontinent actions.

Davidson’s solution is predicated on a distinction between the conditional judgment that it would be better to do x than to do y, all things c­ onsidered – ­that is, ‘relative to all facts, beliefs and values he thinks relevant to the decision’ (Heather and Segal 2013, 449), and the unconditional, or strictly logical, judgment that it would be better to do x than to do y, for which there can be no exceptions. Because incontinent agents predicate their actions on conditional rather than unconditional judgments, there is, philosophically speaking, no logical contradiction. Because the conditional judgment is not absolute (it may well ultimately prove wrong), it falls short of logically compelling the agent to act on its basis. However, Davidson argues, it remains the case that while the incontinent agent has not behaved illogically, s/he has nonetheless behaved irrationally. As Heather and Segal (2013, 450) put it, ‘[t]he irrationality stems from the fact that the agent’s reason for doing x is, as it were, insulated as a practical action



Three Problems with the Addiction as Akrasia Thesis

53

from the better reason for not doing x’. The upshot of these arguments is that the agent, as a rational evaluator of her own actions, cannot understand or rationally account for having chosen to do x. As Heather and Segal (2013, 2015) note, this condition of having acted contrary to one’s better judgment and of not understanding or being unable to rationally account for one’s actions appears deeply resonant with the reports of many putative addicts who tell us not of voluntarily ‘choosing’ the activities to which they are presumed addicted but of feeling alienated from, and confused by, their cravings to engage in these activities. This is vividly illustrated, for example, in the following statement I observed in a group therapy session while conducting ethnographic research in an addiction treatment programme, I’ve promised myself I wouldn’t use a thousand times and really meant it. And then I use. I mean it’s like there are two sides of me. The rational reasonable person who knows he’s gonna die if he keeps on living the way he is and the insane one who just doesn’t care. My reasonable side of me can be as sure as it wants to be but when those drugs appear in front of me the insane one takes over and all those reasons I had not to use are just gone. They just disappear. And I use. It’s like my mind just goes dead and my addiction takes over. I hate myself right afterwards and I’m completely confused by the fact that I just used. I didn’t want to but I did. It’s all well and good to say you need to make a commitment but for some of us that’s not enough. We need something more than that and it doesn’t help for people to be all smug about how we need to make a commitment and it’s all that simple.

Well-known writers on addiction like Stanton Peele (1989) and John Davies (1992) have insisted such accounts are not valid descriptions of the experiential reality of addiction but merely socially functional for those who provide and/or believe them. Cultures like our own, they argue, furnish people with narratives of addiction which they embrace and apply strategically to fulfil their own self-interests.2 I dispute neither the claim that these accounts are socially functional nor that they reflect conceptual commitments prevalent in the cultures to which putative addicts belong. What I do dispute is the claim that either of these facts necessarily forecloses on an account being also honest and/or descriptively valid (Gubrium and Holstein 2009; Haraway 1991). All accounts, including the most rigorous scientific accounts, are socially functional and reflect received conceptual commitments prevalent in the particular cultures in which they are made. Therefore, contra writers like Peele and Davies, pointing out that an account is socially functional and culturally embedded should not be confused with a decisive refutation of either its honesty or its descriptive validity. The addiction as akrasia thesis, and its philosophical provision for how it could possibly be that people might act intentionally counter to their own best judgment, serves better than do conventional disease theories to provide for the fact that addicted behaviour is intentional behaviour. People are often

54

Darin Weinberg

quite shrewd in their efforts to facilitate and fulfil their addictions. By orthodox neurological lights, it is hard to see how addictive drug using activities might be construed as symptomatic of, or causally determined by, a biomedical pathology precisely because they are so obviously intentional or purposive. However, unlike the addiction as ordinary rational choice thesis, the addiction as akrasia thesis also creates conceptual space for the claim that addiction is in some sense a genuine affliction over which the akratic sometimes struggles for control. As such, however, the addiction as akrasia thesis does not in its philosophical form flesh out scientifically or explain specifically what it is that subverts our better judgment, or so often makes addicted behaviour so unfathomable to people. Without abandoning the premise that addictive behaviour is freely chosen, the dual process or dual systems approach with which Heather (cf. 2017, 2020) and others link the addiction as akrasia thesis is intended to provide a more satisfactory scientific understanding of the phenomenology of addiction as affliction and to better explain the tenacity of addictions, despite some people’s concerted efforts to overcome them. This approach posits that human decision-making is based on two distinct forms of mental activity: one which is variously characterized as fast, automatic, habituated, directly responsive to external cues, impulsive, affective, unintentional, unconscious and energy efficient; and a second which is slower, deliberative, conscious, self-directed, rational, intentional, rule-based, decontextualized and energy inefficient. These systems are often referred to as system 1 and system 2, and I will follow this convention. The first system we are said to share with higher animals, the second is distinctly human (cf. Frankish and Evans 2009). Heather argues that addiction reflects one type of conflict that can emerge between these two systems (see also Holton and Berridge 2013; Levy 2013). It is, then, to be clear, a case of addiction residing in the first of these systems and the self that is afflicted invariably residing in the second. The reason addicts, as rational evaluators of their own actions, often cannot understand or rationally account for their addictions is that decisions based in the first system are in fact often comparatively insulated and comparatively immune to decisions based in the second. In addition to the studies conducted by self-proclaimed dual process and/ or dual systems theorists themselves (cf. Frankish and Evans 2009; Heather 2020; Henden 2017), there is a wealth of evidence to support this approach to be found in the clinical (cf. Garland and Howard 2018), behavioural economic (cf. Ainslie 1992), disinhibition (cf. Kallmen and Gustafson 1998), ego depletion (cf. Baumeister 2003), ethnographic (Weinberg 2000), neurological (Robinson and Berridge 2003) and philosophical (cf. Levy 2013) literatures on addiction as well. For example, a large and growing literature linking clinical recovery from addiction to various mindfulness techniques is broadly consistent with dual systems theories (cf. Garland and Howard 2018). The ethnographic data excerpt cited above seems to support something like a dual



Three Problems with the Addiction as Akrasia Thesis

55

systems approach, as does much of the ethnographic literature on addiction and recovery (cf. Weinberg 2000). Space constraints do not allow me to do full justice to these literatures or to the various ways in which support for a dual systems understanding of addicted decision making can be found in them. Suffice to say that I fully acknowledge that much robust support can be found in these literatures. However, as I have said, there are also some important vulnerabilities to this approach that I believe require highlighting. It is to these that I now turn.

Is Self-Control Necessarily Rationally Coherent? Though the addiction as akrasia thesis as supplemented by dual systems theories divides decision making processes into two systems, it nonetheless tends to locate processes of self-control specifically in only one of these ­systems – ­system 2, the rational and deliberative one. Moreover, the rational and deliberative decision making system is pervasively cast as capable of ‘all things considered’ judgments predicated on conscious, decontextualized, rule-based deliberations (cf. Henden 2017, 127). In some cases, this is cast in terms of executive function and survival-relevant behaviour. Sometimes, it is cast somewhat less narrowly in terms of people’s long-term interests or preferences and short-term temptations, but the locus of self-control (as opposed to mere volitional choice) is invariably specified in terms of the more or less rationally coherent, holistic and deliberative decision making system. These arguments are often vague about exactly what expressions like ‘all things considered’, ‘decontextualized’, or ‘rational’ actually mean in any given case and/ or simply inconsistent with much of the scholarly literature. They often do not provide sufficiently for the manner in which type 2 decision making processes are invariably based on foundations provided by type 1 processes (cf. Evans 2009; Frankish 2009; Saunders and Over 2009). And, because they tend to be formulated in etic rather emic terms, they usually result in experience distant specifications of addiction that can seem rather remote from the exigencies of particular people’s addictions and the provision of therapeutic care. As I show in this section, ethnography can provide substantial contributions to clarifying these matters. Let us begin with some rather high-profile, first-hand accounts of the experience of addiction, those to be found in the ‘Big Book’ of Alcoholics Anonymous (Alcoholics Anonymous 2001).3 In addition to outlining general principles, the Big Book presents several dozen personal stories of alcohol addiction. One thing that is immediately evident about these various accounts is that in fact they are so various. The personal narratives supplied in the Big Book are not easily subsumed under a singular generic account of the reasons for people’s alcohol use or the eventual problems that ensued for these people. More generally, the truly vast diversity of ways in which addictions arise in

56

Darin Weinberg

people’s lives suggests it may be wise to sometimes leave aside the effort to learn the universal, or etic, essence that unites them all and to conceive of the problems of addicts in terms of what the philosopher Ludwig Wittgenstein called ‘family resemblances’.4 By these more emic lights, we cease to search for the essence of the a­ bstraction – a­ ddiction – that all addicts share in common. Instead, we turn ethnographically to the local details of practice within which people variously construe events as evidence of addiction. Focus shifts from addiction as such to how particular people’s specific addictions are configured. As ethnographers, we must recognize not only the commonplace, in that different cultural backgrounds may very well dictate different orientations to whether given activities, including drug-using activities, are or are not problematic, but also that particular people may find themselves torn between conflicting values with respect to this question (cf. Ray 1961; Room 1976). This holds not only for the first order values by which we normally conduct ourselves, but also the second order values, or what Charles Taylor (1989) called the ‘hyper-goods’ we use as standards against which to compare and coordinate our first order values. The ethnographic literature on morality and ethics is replete with evidence that many, or perhaps most of us do not live in a coherent evaluative universe wherein ‘all things considered’ judgments can be achieved through rationally coherent assessments or result in coherent, consistent or stable preferences (cf. Laidlaw 2014). This literature suggests addicts are by no means the only people who suffer chronic struggles to rationally reconcile their preferences, to set their priorities or, more colloquially, to achieve balance in their lives. And if most of us suffer from chronic challenges in this department, perhaps it is time to refine and disaggregate our understanding of the characteristics that define particular people’s selves and the specific addictions held to afflict them in light of these more ubiquitous challenges. One element of this refinement will entail a greater recognition of the fact that how we contemplate our preferences and our ‘all things considered’ judgments of them usually involves a complex and often fluid personal mix of emotive and intellectual considerations, only sometimes more and sometimes less amenable to propositional expression, comparison and rational reconciliation. Not only are we sometimes prone to fetishizing or repressing our beliefs on the kinds of psycho-emotional grounds studied by psychoanalysts, but we may also distrust or dismiss the machinations of propositional rationality on other grounds as well. Indeed, as Wittgenstein (1953) taught us, conscious, rule-based deliberations are even themselves invariably grounded in prediscursive dispositional competences forged in what he called particular forms of life; that is, ecologically bounded fields of activity. This insight suggests that even type 2 mental processes cannot be fully ‘decontextualized’ or disengaged from our prediscursive dispositional habits and the habitats that shape them (cf. Brubaker 1993; Taylor 1993).5 These culturally diverse forms of dispositional competence ground different forms of deliberation and evaluative priorities depending on the practical



Three Problems with the Addiction as Akrasia Thesis

57

fields in and for which they are provoked and engaged. These deliberations and evaluative priorities may sometimes, among some people, be subject to more general meta-level evaluations if people become, for some reason, invested in the production of a more coherent ethic, lifestyle or life story, but the more usual pattern is that unless they overtly, persistently and consequentially clash, they are left to their benignly mutual irrelevance. I possess, for example, dispositional competencies that facilitate my navigating road traffic in a car and for composing academic papers that ground different forms of rational deliberation and evaluative priorities. But I have never felt compelled to rationally reconcile them to one another or to ask whether I self-identify, all things considered, more fundamentally as a driver or an academic. I’m not sure I would know how to answer such questions even if I was so disposed. In short, our dispositional competences are multiple, heterogeneous and, depending upon several factors to be considered below, more or less rationally coherent. While they need not necessarily do so, dispositional competences may also extend beyond merely instrumental considerations of how best to succeed in particular practical contexts and into various forms and degrees of self-identification with the practices in question. Hence, deciding which bundle of such competences is best seen to be the locus of a person’s self-control can very often be a quite labile and dynamic matter that, in any case, indispensably requires getting to know them personally. Generic differentiations of type 1 from type 2 decision making processes and their concomitant identification of self-control uniformly with the later are simply too blunt to get us to the ethnographic diversity of self-control in what Hutchins (1995) called ‘the wild’.

Is Self-Control Necessarily Emotionally Repressive? As noted in the last section, the addiction as akrasia thesis, as currently articulated, tends to overwhelmingly equate self-control with executive function and/or sticking to one’s long term interests or preferences. System 2 decisionmaking processes are here construed as sometimes vulnerable to irrational short-term temptations introduced by system 1 that at times threaten, undermine or even, as in the case of addiction, derail our self-control. By these lights, our fears, desires, ambitions, aspirations, hopes and other emotional attitudes figure as motivations that, in order to be made rational and expressive of our self-control, must be rationalized and brought to heal before our survival needs and/or long-term plans. To use the term Heather (2020) now favours, they must be regulated. This orientation to self-control is evident throughout the dual process/systems literature and in some of the most influential writings on addiction more broadly (cf. Ainslie 1992; Baumeister 2003; Kalivas and O’Brien, 2008; Levy 2005). There can be no serious doubt that self-control is often exercised in just this way, and that this manner of self-control is often evidently compromised

58

Darin Weinberg

among people with addictions. However, as the ethnographic record amply demonstrates, the argument that people universally equate their own or one another’s self-control or self-government with long-term planning, impulse control or delayed gratification is undeniably false. Not only do we freely throw caution to the wind on occasion, but so too do we sometimes equate our most authentic values and self-realization with our gut instincts, spontaneous desires, and other emotional impulses and, indeed, equate the kinds of cognitive processes associated with executive function or delayed gratification with alienation from our true selves, freedom and authentic self-­control (cf. Alasuutari 1992; Hochschild 2012; Laidlaw 2014; Turner 1976). On these accounts, our autonomy, freedom or self-control is undermined rather than facilitated by the inhibitions imposed by executive functions. Indeed, a vivid and pertinent illustration of this is the extensive therapeutic emphasis drug rehabilitation programmes like Alcoholics Anonymous themselves place not on the executive repression of spontaneous emotions but encouraging their free and open expression in the interest of facilitating people’s better self-understanding and self-control (Carr 2011; Garriott and Raikhel 2015; Valverde 1998; Weinberg 2013). If we are to more fully understand how self-control can be lost to addictions, we must seek to more fully appreciate how addictions usually arise out of nonpathological habits that are initially, and perhaps for a considerable period of time, experienced as enjoyable, life-enhancing and self-expressive rather than painful, life-diminishing or self-destructive. The Big Book presents a multitude of accounts testifying to this. For instance, When I was drinking I was ok. I understood. Everything made sense. I could dance, talk and enjoy being in my own skin. It was as if I was an unfinished jigsaw puzzle with one piece missing; as soon as I took a drink, the last piece instantly and effortlessly snapped into place. (Alcoholics Anonymous 2001, 320)

It must always be remembered that drugs are, for most people, including those who eventually develop addictions, enjoyable and performance-­ enhancing rather than painful and performance-debilitating. Indeed, it must also be remembered that a significant majority of those who take up drug use and other potentially addictive activities never become self-destructive, and that when they do, this is always a matter of degree. It is also a matter about which people often feel deeply uncertain and ambivalent. Given the extent to which expert discourse on drug use and other potentially addictive activities is skewed toward correctional and/or therapeutic agendas, it should not be surprising to anyone that this discourse tends to minimize, and sometimes completely overlook, the fact that self-destructive engagement in such activities is, by a considerable margin, the exception rather than the rule. But as John Davies (1992, xi) quite properly insisted, ‘Most people who use drugs do so for their own reasons, on purpose, because they like it, and because they find no adequate reason for not doing so’. If freedom, autonomy, or



Three Problems with the Addiction as Akrasia Thesis

59

self-control, is invariably a matter of self-regulation, or the ‘all things considered’ rational repression of putatively irrational temptations, it is not obvious how potentially (but not yet) addictive activities could figure freely, enjoyably, life-enhancingly, self-expressively and on purpose in the lives of most people who use them, including most of those who eventually develop addictions. By exploring the myriad ways in which drugs and alcohol figure enjoyably, sometimes self-expressively or indeed even productively and, at any rate, unproblematically in the lives of people from a range of cultures and in a range of specific cultural contexts (cf. Douglas, 1987; Heath, 2012; Moore 2004; Room, 1984; Singer, 2012; Waldorf, Reinarman and Murphy 1991), the ethnographic literature provides a valuable compliment to the bulk of addiction research that focuses only on self-destructive activities. This literature highlights the various ways and reasons people come to self-identify, in the Frankfurtian sense, with their drug using practices, and how these practices are inextricably bound up with a multitude of features of the specific cultural contexts within which they develop (cf. Weinberg 2013). In other words, they show us that drug using practices are not usually experienced as threatening to our self-control but, very much on the contrary, as entirely consistent with, and as manifestly enhancing of, the satisfaction with which people conduct their lives. A better ethnographic appreciation and further investigation of this reality will provide us with a wealth of specific insights as to why particular people often seem so deeply reluctant to forego these practices, and hence why their addictions are often so tenacious. To conclude this section, and in order to flesh out this account of selfidentifying with one’s drug using practices in a bit more detail, it will be useful to briefly consider Frankfurt’s specific arguments on self-identification. According to Frankfurt, we self-identify with an element of our mental life when we are ‘wholehearted’ with respect to it. He defines this as being ‘satisfied’ with that element, which he further specifies as follows, To be satisfied with something does not require that a person have any particular belief about it, nor any particular feeling or attitude or intention. It does not require, for instance, that he regard it as satisfactory, or that he accede to it with approval, or that he intend to leave it as it stands. There is nothing that he needs to think, or to adopt, or to accept; it is not necessary for him to do anything at all … Satisfaction with one’s self requires, then, no adoption of any cognitive, attitudinal, affective, or intentional stance. It does not require the performance of a particular act; and it also does not require any deliberate abstention. Satisfaction is a state of the entire psychic ­system – ­a state constituted just by the absence of any tendency or inclination to alter its condition. (Frankfurt 1999, 104)

As Frankfurt (1999, 105) further argues, ‘satisfaction with one psychic element does not require satisfaction with any other,’ so his is emphatically not a theory that would seem to require a rational coherence or consistency of all those psychic elements with which we self-identify. So, what then, might he mean

60

Darin Weinberg

when he argues that despite its not requiring rational coherence or the conscious, let alone deliberate commission or omission of any active performance, that satisfaction must nonetheless be ‘a state of the entire psychic system’? Or that ‘insofar as his desires are entirely unreflective, he is to that extent not a person at all. He is a wanton’ (Frankfurt 1999, 105–6)? This puzzle is best solved by clarifying what Frankfurt means by ‘unreflective’. He means by this not inactive or unconscious, both of which self-identification may very well be, but rather, reason insensitive. That is, in order to retain one’s status as a person (as opposed to a mere wanton), one must remain to be seen to be amenable, at least in principle, to reasonable persuasion with respect to the psychic elements with which one self-identifies or, for that matter, with which one does not. One need not actually put these into question, but one must, at least, be amenable to their reasonable interrogation. So, what does this have to do with the present argument? It suggests in the first instance that self-identification, or distinguishing those psychic elements Frankfurt calls ‘internal’ to the self from those he calls ‘external’ – those with which we self-identify and from which flow our own self-willed actions from those we find alien and from which flow constraints on, or afflictions of, our ­autonomy – ­will invariably be a matter of particular selves in all of their specificity rather than the self as such as a generic locus of belief and volition. Secondly, one’s self is here identified as those elements of our psychic life that are held expressive of our character, whether they are emotionally repressive in the name of our long-term interests or preferences or emotionally expressive of our natural and often spontaneous inclinations, including our inclinations toward the people or things we love or which give us pleasure (Moran 2002). While Frankfurt (1999) insists we must be wholehearted in our identification of those elements internal to ourselves, this is not necessarily a matter of ‘all things considered’ rational evaluations, but may also include, much more modestly, efforts to merely minimize ambivalence. Ambivalence only arises when we actually become aware of conflicts between elements of our psychic life. There is, then, no evidence that Frankfurt believes wholeheartedness requires a comprehensive self-legislative inventory. Because on Frankfurt’s account, so much of that with which we are satisfied to self-identify has not been deliberatively endorsed or even consciously considered; hence, it is implausible to attribute to him the kinds of rational coherence or consistency requirements for autonomy and selfhood defended by others (cf. Ekstrom 2005). The most tenable interpretation of the requirement to avoid wantonness, then, is only a minimalist requirement of reason sensitivity rather than a comprehensive rational coherence requirement. Autonomy, will, or selfcontrol is exercised both through emotionally repressive self-regulation and emotionally expressive self-revelation. Therapeutic self-exploration, discovery, actualisation and empowerment overwhelmingly attend to both of these dimensions of who we are and what we do.



Three Problems with the Addiction as Akrasia Thesis

61

Self-Control and Social Context There is a ubiquitous tendency among defenders of the addiction as akrasia thesis, as well as many of their fellow travellers in addiction science, to assume the existence of an ‘ideal rationality,’ the normative principles of which hold universally; that is, regardless of social context. By these lights, rationality and, hence, self-control invariably consists in honouring or being guided by these principles. Whenever or wherever someone seems not to be guided by them we are presented with the challenge of explaining their evident irrationality and a loss of self-control. Indeed, the scholarly importance of akrasia and/or weakness of the will arguments are often explicitly predicated on precisely this assumption. While there remain lively controversies in the philosophical, psychological and economic literatures pertaining to the specific identity of these principles, the details of these debates need not concern us here. Suffice to say these debates are largely about the causes, characteristics and consequences of this ideal rationality and much less concerned with whether we ought to think of it as universal. The principles of rationality I have in mind include, but are not limited to, the normative requirements that our beliefs, intentions and actions be coherent, consistent and stable. It is often said that Donald Davidson himself held this view of rationality and that it undergirds his famous arguments concerning mental holism, the principle of charity and, in turn, via the notion of ‘all things considered’ judgments, weakness of the will. There is certainly textual evidence for the truth of this claim. However, I think it is in need of important qualification. In his essay ‘Paradoxes of Irrationality,’ Davidson (1982) goes beyond his earlier descriptive account of akrasia to develop a prospective explanation for how it could come about. He writes, Mental phenomena may cause other mental phenomena without being reasons for them, then, and still keep their character as mental, provided cause and effect are adequately segregated … To constitute a structure of the required sort, a part of the mind must show a larger degree of consistency or rationality than is attributed to the whole … The idea is that if parts of the mind are to some degree independent, we can understand how they are able to harbor inconsistencies, and to interact on a causal level … Recall the analysis of akrasia … the way could be cleared for explanation if we were to suppose two semi-autonomous departments of the mind, one that finds a certain course of action to be, all things considered, best, and another that prompts another course of action. On each side, the side of sober judgment and the side of incontinent intent and action, there is a supporting structure of reasons, of interlocking beliefs, expectations, assumptions, attitudes and desires. (Davidson 1982, 300)

In this essay, presumably because its central task was to engage with Freudian thought and, more specifically, to explain the possibility of irrationality, Davidson seems to focus on formulating this partitioning of the mind as a

62

Darin Weinberg

segregation of its rational and irrational components (though he explicitly distinguishes this division from that between Freud’s conscious and unconscious mind). If indeed his intention was confined to partitioning a singular ‘sober’ agent from an ‘incontinent’ one, we might very well want to insist that with his philosophy of mental holism, Davidson invariably identifies self-control with that mental structure exhibiting the greatest conformity with a universal concept of ideal rationality. However, in a footnote he provides the following further specification of his position, Here as elsewhere my highly abstract account of the partitioning of the mind deviates from Freud’s. In particular, I have nothing to say about the number or nature of divisions of the mind, their permanence or aetiology. I am solely concerned to defend the idea of mental compartmentalization, and to argue that it is necessary if we are to explain a common form of irrationality. I should perhaps emphasize that phrases like ‘partition of the mind’, ‘part of the mind’, ‘segment’ etc. are misleading if they suggest that what belongs to one division of the mind cannot belong to another. The picture I want is of overlapping territories. (Davidson 1982, 300)

The facts that Davidson appears agnostic as to the ‘number or nature of divisions of the mind,’ that he casts compartmentalisation as necessary to explaining a common form of irrationality but not obviously confined to the explanation of irrationality, and that these territories need not be mutually exclusive but are instead ‘overlapping,’ combine to suggest Davidson was at least open to an occasional degree of vagueness, equivocality and/or instability in our judgments as to which territory was best understood as the locus of self-control and/or all things considered judgments in any given case. In his highly Davidsonian discussion of Freud and moral reflection, Richard Rorty (1991, 151) quotes Philip Rieff ’s reference to ‘Freud’s egalitarian revision of the traditional idea of a hierarchical human nature’ wherein self-control is invariably identified with our ‘higher’ rational nature and set against our passions. Rorty argues Freud ‘gave us a vocabulary that lets us describe all the various parts of the soul, conscious and unconscious alike … as equally plausible candidates for “the true self ”’ (ibid., p 152), and goes on to write, I want to focus on the way in which Freud, by helping us see ourselves as centerless, as random assemblages of contingent and idiosyncratic needs, rather than as more or less adequate exemplifications of a common human essence, opened up new possibilities for the aesthetic life. He helped us become increasingly ironic, playful, free, and inventive in our choice of self-descriptions. (ibid., 155)

Rorty’s attention to the rich variety of possibilities we enjoy in our choice of self-descriptions allows us to better appreciate the implications of Davidson’s views on mental compartmentalization. By Rorty’s reading, Davidson probably didn’t (and would have been wrong to) believe that rational requirements like coherence, consistency and stability were universal criteria with which to identify selves and self-control generically. It should also be recalled here



Three Problems with the Addiction as Akrasia Thesis

63

that Davidson’s principle of charity was plainly not a principle he thought was applied exhaustively and acontextually, for its own sake, to all of the mental contents with which people self-identify. Instead, its application was driven in the first instance by an applicant’s desire to get on with their lives in ways they find satisfying in the Frankfurtian sense and in ways that allow them to cooperate with others in their lives with whom they feel they must, should or would like to cooperate. The coherence we find in our own and each other’s minds, then, tends to be anchored ­contextually – ­that is, in the various forms of life within which we participate and which variously (and, to a considerable extent, discontinuously) make salient particular beliefs, intentions, values and the practical dispositions or competences upon which these are grounded. When we speak of holism in this context, then, it is crucial that we recognize we are usually referring to a holism bounded by inductive considerations of what might conceivably be relevant to particular, collectively orchestrated practical projects and decidedly not the much more inclusive holism suggested by invocations of type 2 decision making ­processes – ­that is, a universal, wholly decontextualized, rule-based and consciously deliberated holism. By these lights, the partitions Davidson wrote of insulating various regions of our mental activity should not be understood as confined to partitions between the rational and irrational. Instead, they also extend to the porous boundaries we tacitly experience between the ecological contexts within which we acquire and incorporate our own particular region-specific repertoires of dispositional competences. In short, whether or not the relations between elements of our mental life figure as reasons or as unreasoned causes of what we do is not only a matter of distinguishing rational from irrational motivations but of distinguishing context-specific rationalities as well. This can be seen in the following ethnographic fieldnote excerpt taken in a staff meeting of an out-patient recovery programme mandated to serve homeless and precariously housed clients suffering from addiction and a concurrent major mental illness. In this case, a client, Layla,6 had dropped out of the programme and was accumulating insurmountable debts using crack cocaine. She was contacted by a counsellor but resisted this counsellor’s efforts to place her into an inpatient recovery setting. In the absence of mitigating factors, counsellors normally regarded such resistance as irrational and evidence, more specifically, of the insurmountable hold clients’ addictions had over them. However, in this case there were mitigating circumstances and, despite the putative severity of Layla’s relapse, her resistance was interpreted as a reasoned rational choice rather than a causal effect of her crack addiction: She’s resisting it because she doesn’t want to lose her apartment. You know she moved into that apartment next to her parents’ place, and it’s kind of a good situation for her. If she goes into inpatient treatment she’s afraid she’ll lose her place. I can understand how she feels. She’s in a tough situation.

64

Darin Weinberg

In this excerpt, knowledge of Layla’s practical circumstances beyond the realm of treatment and recovery led a counsellor to rationalize her resistance to treatment as strategically motivated by practical circumstances, rather than as evidence of the causal force of her addiction. However, more commonly, the immediate practical demands of participating in a recovery programme overrode whatever judgments might be based on the practical circumstances clients faced outside the programme. Thus, a counsellor in the same programme as that noted above, in connection with the Layla excerpt, elaborated on her difficulty getting Del to improve his personal hygiene and Neal to obtain false teeth: People may be coming regularly, but if they continue to look like they’re homeless and don’t really care about their appearance, I can’t help thinking they aren’t bringing a lot of what they get in the programme out with them when they leave. I’m working with Neal to get his teeth again for the same reason. He looks awful without his teeth. I don’t think looking like that can do anything for his self-image and self-esteem.

Del, though, claimed to regard his lax personal hygiene as an exercise in personal competence and a survival tactic while living homeless, mentioning to me later that he was more likely to be robbed if he looked better. Neal said essentially the same thing; that he fit into the homeless street scene better without his dentures and this kept him out of trouble. We can see in these examples the contingency, fluidity and, indeed, contestation that is often evident in determinations of whether particular people’s actions are reasoned and self-governed or comparatively unreasoned and, hence, quite possibly evidence of the causal force of their addictions. In Layla’s case, we might want to say Davidson’s principle of charity was very obviously at work. However, the cases of Del and Neil are more complicated. While their counsellor was evidently devoted to helping them to elevate their self-image, self-esteem and, by implication, their self-control; they themselves felt this effort to empower them, guided by a well-known standard of recovery culture (foster clients’ selfrespect), could not be reconciled with the tactics they considered necessary to managing life on the streets. Here, the question of which rationality should be seen to be the true locus of their self-control was not nearly so clear. There seems to have been a clash between the standards of rationality characteristic of the recovery programme and the standards conducive to remaining safe on the streets.

Concluding Remarks According to the addiction as akrasia thesis as it is currently formulated, addiction science must necessarily aspire to produce universal theoretical findings regarding such things as the self, self-control, the will, autonomy,



Three Problems with the Addiction as Akrasia Thesis

65

freedom and addiction itself. Whatever value addiction science may bring to the field of therapeutics is, in this light, cast uniformly in terms of the aptness of these universal models to the needs of caregivers and receivers. The thesis of this chapter is that we might instead begin with ethnographic research on the diverse and particular needs of caregivers and receivers in the myriad settings in which recovery is a concern. The insights so gleaned might then be used to interrogate the limitations of received universal models and thereby enrich our scientific understanding of addiction and recovery through a refinement and disaggregation of our knowledge of the various ways that these become manifest in the wild. This is by no means a rejection of the aspirations of universalist addiction science as such, but, rather, a call to recognize that ethnography provides a distinctively robust, but particularist alternative to the conduct of addiction science that can incisively supplement universalism. Philosophers the likes of Ludwig Wittgenstein, Donald Davidson, Harry Frankfurt and Richard Rorty have given us powerful philosophical justifications and philosophical resources with which to clarify and accomplish this kind of particularist inquiry. More specifically, the preceding arguments are entirely consistent with Davidson’s views on both akrasia and mental compartmentalization. Ethnography is perhaps the best, if not the only, method of empirical inquiry for illuminating the reality of just how people in practice collectively orchestrate determinations that they themselves, and one another, have acted for reasons of their own, as a matter of self-government, or have been caused to do something by an addiction that exists external to that self (cf. Weinberg 2005, 2019). It is for this reason I have cast this intervention not as a critique of the addiction as akrasia thesis but rather as an effort to bring ethnography to bear on the solution of three specific limitations and/or vulnerabilities one now finds in the literature. From these philosophers and a wealth of important ethnographic work on drug use, addiction and everyday ethical practices, we have learned that self-control is not only a matter of rational self-regulation but also self-expression, and that it extends to the exercise of freedom in all of its guises. If we are to more fully understand addiction, we must understand that it poses not only a threat to our self-regulation but to our freedom more generally conceived. Darin Weinberg is a Professor in the Department of Sociology, Cambridge University, and a Professorial Fellow of King’s College, Cambridge. His research focuses primarily on the practical purposes to which concepts of addiction, mental disorder and learning disability are applied in various historical and contemporary contexts. His books include Contemporary Social Constructionism: Key Themes and Of Others Inside: Insanity, Addiction, and Belonging in America, both of which won the Melvin Pollner Prize, Intoxication and Society (edited with Jonathan Herring, Ciaran Regan and Phil Withington), Talk and Interaction in Social Research Methods (edited with Paul Drew and Geoffrey Raymond), and Qualitative Research Methods (edited).

66

Darin Weinberg

Notes   1. This is not a position uniformly adhered to in the literature concerning the addiction as akrasia thesis. Some prefer to see addiction as preference instability rather than a failure of one’s actions to conform to stable preferences. This debate, while interesting, is not relevant to the arguments I am making in this chapter.   2. Peele’s and Davies’ arguments in this regard are confluent with a wider range of important research that has effectively changed the subject of addiction science away from the experience of affliction or the loss of self-control, as such, to the ways in which claims regarding the dangers of addiction have figured in campaigns of social control. As vitally important as it is, this largely social constructionist research tends either, like Peele and Davies, to completely reject the claim that some people really do sometimes lose control of their use of drugs and ⁄ or beg the question of how such a loss of control is best understood. By orthodox social constructionist lights, we may sometimes suffer from being defined in one way or another. But the notion that something other than human subjects, as defining agents, might be the source of suffering has been more difficult to formulate within the social constructionist frame. My own research has for some time been concerned to demonstrate that this need not be the case (cf. Weinberg 2005, 2013, 2019).   3. The Big Book was compiled by one of the co-founders of Alcoholics Anonymous, Bill W., and published for the first time in 1939 by Alcoholics Anonymous World Services. Its purpose was to outline the recovery programme of Alcoholics Anonymous. It inaugurated the ‘Twelve Step Movement’, which has arguably had more influence on thought and practice concerning recovery from addiction than any other popular or scientific movement in history. The Big Book has since 1939 passed through three additional editions, sold over 30 million copies and been translated into over 70 languages. In 2011, Time magazine listed it as one of the “best and most influential” 100 non-fiction books published in English since its own founding in 1923. In 2012, the Big Book was named by the Library of Congress as one of the 88 ‘Books that Shaped America’.   4. In his Blue Book, Wittgenstein (1958, 17) wrote of a pervasive ‘craving for generality’‘ among both philosophers and scientists that, he argued, hailed from a number of sources, one of which was,

The tendency to look for something in common to all the entities which we commonly subsume under a general term … We are inclined to think that there must be something in common to all games, say, and that this common property is the justification for applying the general term ‘game’ to the various games; whereas games form a family the members of which have family likenesses. Some of them have the same nose, others the same eyebrows and others again the same way of walking; and these likenesses overlap.

  5. The celebrated neurologist of addiction Marc Lewis (2017, 187) has noted in this respect that habit learning originates piecemeal in setting-specific kinds of ways, but that eventually, because brains tend to conserve structure and resources, habits, or acquired ‘synaptic networks,’ often converge and become mutually reinforcing. However, he notes these processes need not converge on a wholly unified subject, or rationally integrated hub for all deliberation and volition. Instead, ‘alternative synaptic networks can compete with each other … this is the case when addiction arises in development, but also when it dissipates, replaced by the desire for and belief in alternative outcomes’.   6. All names are pseudonyms.



Three Problems with the Addiction as Akrasia Thesis

67

References Ainslie, George. 1992. Picoeconomics. Cambridge: Cambridge University Press. Alasuutari, Perti. 1992. Desire and Craving. Stony Brook, NY: SUNY Press. Alcoholics Anonymous. 2001. Alcoholics Anonymous. New York: Alcoholics Anonymous World Services. Baumeister, Roy F. 2003. ‘Ego Depletion and Self Regulation Failure: A Resource Model of Self-Control’. Alcoholism: Clinical and Experimental Research 27(2): 281–4. Brubaker, Rogers. 1993. ‘Social Theory as Habitus’. In Bourdieu: Critical Perspectives, edited by Craig Calhoun, Edward LiPuma and Moishe Postone, pp. 212–34. Chicago: University of Chicago Press. Carr, E. Summerson. 2011. Scripting Addiction. Princeton, NJ: Princeton University Press. Davidson, Donald. 1982. ‘Paradoxes of Irrationality’. In Philosophical Essays on Freud, edited by Richard Wollheim and James Hopkins, pp. 289–305. Cambridge: Cambridge University Press.. . 2001. Essays on Actions and Events. Oxford: Oxford University Press. Davies, John Booth. 1992. The Myth of Addiction. Amsterdam, Netherlands: Harwood. Douglas, Mary, ed. 1987. Constructive Drinking. Cambridge: Cambridge University Press. Ekstrom, Laura Wadell. 2005. ‘Alienation, Autonomy, and the Self ’. Midwest Studies in Philosophy 24: 45–67. Evans, Jonathan. 2009. ‘How Many Dual-Process Theories Do We Need? One, Two, or Many?’ in In Two Minds: Dual Processes and Beyond, edited by Jonathan Evans and Keith Frankish, pp. 33–54. Oxford: Oxford University Press. Evans, Jonathan, and Keith Frankish (eds). 2009. In Two Minds: Dual Processes and Beyond. Oxford: Oxford University Press. Frankfurt, Harry G. 1999. Necessity, Volition, and Love. Cambridge: Cambridge University Press. Frankish, Keith. 2009. ‘Systems and Levels: Dual System Theories and the PersonalSubpersonal Distinction’. In In Two Minds: Dual Processes and Beyond, edited by Jonathan Evans and Keith Frankish, pp. 89–108. Oxford: Oxford University Press. Garland, Eric L., and Matthew O. Howard. 2018. ‘Mindfulness-Based Treatment of Addiction: Current State of the Field and Envisioning the Next Wave of Research’. Addiction Science & Clinical Practice 13(1): 14. Garriot, William, and Raikhel, Eugene. 2015. ‘Addiction in the Making’. Annual Review of Anthropology 44: 477–91. Gubrium, Jaber F. and James A. Holstein. 2009. ‘The Everyday Work and Auspices of Authenticity’, in Authenticity in Culture, Self and Society, edited by Phillip Vannini and J. Patrick Williams, pp. 121–38. Burlington, VT: Ashgate. Haraway, Donna J. 1991. Symians, Cyborgs and Women. New York: Routledge. Heath, Dwight. 2012. Drinking Occasions. London: Routledge. Heather, Nick. 2017. ‘Addiction as a Form of Akrasia’. In Addiction & Choice: Rethinking the Relationship, edited by Nick Heather and Gabriel Segal, pp. 133–50. Oxford: Oxford University Press. . 2020. ‘The Concept of Akrasia as the Foundation for a Dual Systems Theory of Addiction’. Behavioural Brain Research 390, 15 July 2020: 112666. Heather, Nick, and Gabriel Segal. 2013. ‘Understanding Addiction: Donald Davidson and the Problem of Akrasia’. Addiction Research and Theory 21(6): 445–52.

68

Darin Weinberg

. 2015. ‘Is Addiction a Myth? Donald Davidson’s Solution to the Problem of Akrasia Says Not’. International Journal of Alcohol and Drug Research 4(1): 77–83. Henden, Edmund. 2017. ‘Addiction, Compulsion, and Weakness of the Will: A DualProcess Perspective’. In Addiction & Choice: Rethinking the Relationship, edited by Nick Heather and Gabriel Segal, pp. 116–32. Oxford: Oxford University Press. Hochschild, Arlie R. 2012. The Managed Heart (3rd ed.). Berkeley: University of California Press. Holton, Richard, and Kent Berridge. 2013. ‘Addiction between Compulsion and Choice’. In Addiction and Self Control, edited by Neil Levy, pp. 239–68. Oxford: Oxford University Press. Hutchins, Edwin. 1995. Cognition in the Wild. Cambridge, MA: MIT Press. Kalivas, Peter W. and Charles C. O’Brien. 2008. ‘Drug Addiction as a Pathology of Staged Neuroplasticity’. Neuropsychopharmacology 33: 166–80. Kallmen, H., and R. Gustafson. 1998. ‘Alcohol and Disinhibition’. European Addiction Research 4: 15–62. Laidlaw, James. 2014. The Subject of Virtue: An Anthropology of Ethics and Freedom. Cambridge: Cambridge University Press. Levy, Neil. 2005. ‘Autonomy and Addiction’. Canadian Journal of Philosophy 36(3): 427–48. . (ed.). 2013. Addiction and Self Control. Oxford: Oxford University Press. Lewis, Marc. 2017. ‘Choice Isn’t Simple: Reply to Pickard’. Neuroethics 10: 181–3. Moore, David. 2004. ‘Beyond ‘Subculture’ in the Ethnography of Illicit Drug Use’. Contemporary Drug Problems 31(2): 181–212. Moran, Richard. 2002. ‘Frankfurt on Identification: Ambiguities of Activity in Mental Life’. In Contours of Agency, edited by Sarah Buss and Lee Overton, pp. 189–217. Cambridge, MA: MIT Press. Peele, Stanton. 1989. Diseasing of America. Lexington, MA: Lexington Books. Ray, Marsh. 1961. ‘The Cycle of Abstinence and Relapse Among Heroin Addicts’. Social Problems 9: 132–40. Robinson, Terry E., and Kent C. Berridge. 2003. ‘Addiction’. Annual Review of Psychology 54: 25–53. Room, Robin. 1976. ‘Ambivalence as a Sociological Explanation: The Case of Cultural Explanations of Alcohol Problems’. American Sociological Review 41: 1047–65. . 1984. ‘Alcohol and Ethnography: A Case of Problem Deflation?’ Current Anthropology 25: 169–78. Rorty, Richard. 1991. Objectivity, Relativism and Truth. Cambridge: Cambridge University Press. Saunders, Clare, and David E. Over. 2009. ‘In Two Minds about Rationality?’ In In Two Minds: Dual Processes and Beyond, edited by Jonathan Evans and Keith Frankish, pp. 317–34. Oxford: Oxford University Press. Singer, Merrill. 2012. ‘Anthropology and Addiction: An Historical Review’. Addiction 107(10): 1747–55. Taylor, Charles. 1989. Sources of the Self. Cambridge, MA: Harvard University Press . 1993. ‘To Follow a Rule…’ In Bourdieu: Critical Perspectives, edited by Craig Calhoun, Edward LiPuma and Moishe Postone, pp. 45–60. Chicago: University of Chicago Press. Turner, Ralph. (1976). ‘The Real Self: from Institution to Impulse’. American Journal of Sociology 81: 989–101. Valverde, Mariana. 1998. Diseases of the Will. Cambridge: Cambridge University Press.



Three Problems with the Addiction as Akrasia Thesis

69

Waldorf, D., Reinarman, C., and Murphy, S. 1991. Cocaine Changes. Philadelphia, PA: Temple University Press. Weinberg, Darin. 2000. ‘“Out There”: The Ecology of Addiction in Drug Abuse Treatment Discourse.’ Social Problems 47(4): 606–21. . 2005. Of Others Inside: Insanity, Addiction and Belonging in America. Philadelphia, PA: Temple University Press. . 2013. ‘Post-humanism, Addiction and the Loss of Self Control: Reflections on the Missing Core in Addiction Science.’ International Journal of Drug Policy 24: 173–81. . 2019. ‘Psychosomatic Subjects and the Agencies of Addiction.’ BMJ Medical Humanities 45: 162–68. Wittgenstein, Ludwig. 1953. Philosophical Investigations. Oxford: Blackwell. . 1958. The Blue and Brown Books. Oxford: Blackwell.

3 To Live Like ‘People’ Drinking and Weakness of Will among the Runa of the Ecuadorian Amazon Francesca Mezzenzana

We, people from the forest, are born to drink, to get drunk, to live without obstacles: you won’t be able to change our custom —F. Pierre, Viaje de Exploración al Oriente Ecuatoriano (1889)

This is what an unnamed Runa man told Françoise Pierre, the Dominican missionary who travelled to the Bobonaza region in the Ecuadorian Amazon between 1887 and 1888, and then recounted his adventures in a manuscript titled Viaje de Exploración al Oriente Ecuatoriano. A large portion of the book is dedicated to the description of the drinking excesses of the indigenous peoples Father Pierre encountered in his trip along the Bobonaza river and to his rather frustrated attempts at convincing them to quit their habits. The words ‘drunken’, ‘drinking’ and ‘manioc beer’ recur frequently throughout the book, which witnesses the missionary repeatedly confronting people who seem to be spending all their time drinking and getting drunk. Father Pierre was not alone in lamenting the drinking excesses of local indigenous people. In a report by a rubber merchant who owned plantations in the nearby region of Napo, he described local Runa people as indulging in drinking for several days a week, especially over Saturdays and Sundays (in Porras 1979). Both Father Pierre and the rubber boss identified drinking as a major impediment to the moral conversion of the Runa, who were represented as weak-willed and lazy. Reducing the amount of drinking was thus a major goal of missionaries and rubber barons who deployed a variety of strategies in order to accomplish this objective. The rubber boss, for instance, recounts how another rubber lord, in despair because the Runa did not work as much as he wished, recurred to a s­ ubterfuge – h ­ e pretended that the Church, in



Drinking and Weakness of Will

71

punishment for the Indians’ excesses, had abolished holy Sundays and hence the drinking that took place on that day. This allowed him to regulate the workers’ drinking habits for ‘three months’ (in Porras 1979, 34). Despite this escamotage, the Runa soon resumed their drinking parties. These relapses were taken as evidence for the Indians’ weakness of will and inability to control their impulses. Over a century later, the opinions of most nonindigenous people about the Runa have not changed much. In contemporary Ecuador, white and urban mixed race people living in the region of Pastaza often portray the Runa as untrustworthy and lazy people lacking in self-discipline. Excessive drinking is often brought up as evidence of Runa people’s lack of self-control. ‘They cannot control their instincts. When there is chicha (manioc beer), forget about work, forget about plans! They will not do anything but sit and drink,’ a mestizo (mixed race) engineer exclaimed to me once. He was sweaty and furious: he had flown from the city to a rural Runa village to discuss with the community about a project for the construction of a bridge and he was outraged when people failed to show up at the meeting. He had found out later that they hadn’t come because they went to a drinking party somewhere in the village. The engineer’s description is by no means exceptional: one of the major complaints about the Runa is that they are unreliable alcoholics. ‘Work gets delayed because everyone is drunk, children are left without food because parents are away drinking,’ complained a white governmental officer who has worked in the area for some years. ‘Alcoholism here is a major problem. If that problem isn’t eradicated, we will never accomplish anything,’ the officer told me in the aftermath of a meeting with Runa school teachers. Excessive, unregulated drinking is seen by nonindigenous Ecuadorians as a major impediment to the socioeconomic development of the Runa, who are perceived as poor and lacking in educational, social and economic resources (Viteri 2003). The nonindigenous discourse around Runa d ­ rinking – a­ nd the portrayal of the Runa as people who cannot control t­ hemselves – i­ s also increasingly being adopted by some urban Runa people across different ages and socioeconomic status. Today you can even hear some Runa people, both in urban and rural contexts, complaining that the greatest problem in the villages is alcoholism (alcoholismo in Spanish). With alcoholism, people generally refer to the protracted consumption of aguardiente or manioc beer that takes place over days in occasion of collective work parties known as minga. Women complain that such prolonged drinking leads to an increase in marital violence, and men often comment that drinking too much on these occasions is the cause of physical fights and accidents. People also often mention the health risks associated with drinking, such as gastritis, cirrhosis and hepatitis, the last two being some of the most common causes of death among the Runa in rural settings. Throughout my fieldwork, I witnessed the conflicts, reflections and conversations that took place around the issue of drinking first-hand. Runa people

72

Francesca Mezzenzana

were often critical of the prolonged drinking that took place during collective parties and also expressed their wish to change their drinking habits. During community meetings or small social gatherings, they spoke about the need to do something about the drinking and pointed out the causal link between drunkenness and physical violence. Despite their proclaimed intentions, however, most people never actually do anything to change the state of things. A few years ago, in one well-known village along the Bobonaza, the community president declared the local consumption of any industrially produced alcohol illegal. People who contravened had to pay a fine of $50. The proposal was passed, amidst cheers of approval, with the majority of votes. However, a couple of months later, people decided to call elections to change the president. They were extremely unhappy with the prohibition to consume alcohol. ‘It was a good idea, but it didn’t work at the end. People want to follow their minds,’ my friend Jairo commented with a shrug. He was amongst those who, at the time, wanted the president to be removed. They did not succeed in removing him, but he had to put an end to the prohibition. While many Runa people explicitly recognize the downsides of drinking and complain about those, they hardly ever do something about it. Take, for instance, Renan, a young man in his twenties. He is married to Anita with whom he has three children. He works in a small ecotourism business in the city of Puyo and earns good money. Every weekend he returns to his village and sponsors a work party. His wife prepares manioc beer, and he brings plenty of aguardiente, a liqueur derived from sugar cane. His parties are notorious in the village: they usually end up with a drunken fight between him and his wife or between him and his brothers, or him and his guests. Renan’s parties always provide juicy gossip for everyone. Whenever a party ended badly, Renan would always come to the house of my host family to complain. He got along very well with my host and often sat with him, drinking manioc beer and promising to himself that he would never throw a party again. ‘Enough of parties!’ he complained. ‘Too much money and too many problems!’ He used to swear that he would stop drinking altogether. His w ­ ife – ­another frequent visitor of my host ­house – a­ lso complained. She described how Renan wasted large sums of money to buy aguardiente, leaving her and her children with no money to buy clothes or candles. She also said that drinking was the main cause of their fights. Renan, too, seemed very upset about the recurrent fights that took place during his parties. And yet, despite this, the parties continued, and so did the fights and the problems. In the many years I have known him, nothing has changed. Despite the speeches, despite the good intentions and promises, Renan’s resolution to reduce his drinking had very little effect on his subsequent behaviour. The obvious interpretation of Renan’s behaviour is that it is a classic example of weakness of will. Richard Holton has defined weakness of will as instances in which an agent who has made a resolution, revises it too readily and thus fails to act as previously intended. In recognition of the extensive



Drinking and Weakness of Will

73

literature debating this definition of weakness of will (Dodd 2009), I  adopt Holton’s definition in this chapter since it resonates with the depiction that most non-Runa give of the Runa: as people who are irresolute, unreliable, unable to keep up with what they have previously resolved. In this paper, I explore some basic ethnographic questions. Why does Renan not quit drinking if he says it is detrimental to his life? Why do Runa people continue to engage in long sessions of drinking despite their claims that these bring problems to their health and overall wellbeing? Does Renan’s behaviour really signal an incapacity to carry on with one’s previously agreed resolutions? What is the value of resolutions like ‘I want to quit drinking’ in a context such as that of the Runa? In order to address these questions, I explore the nature of Runa drinking. In particular, I investigate the kind of moral project that underpins the extensive drinking practices that take place among the Runa in collective work parties. The first question to address then is: why do the Runa drink? After addressing this question, I then offer different ethnographic explanations as to why behaviours such as Renan’s are not examples of weakness of will. The relationship between drinking and weakness of will is particularly tricky. In Euro–American thought, drinking alcohol conjures up different and contrasting ideas about individual freedom and will. For instance, on the one hand, there is a social stigma attached to alcoholics because they are seen as people who are unable to control their appetites. From this perspective, alcoholism is understood as a form of weakness of will whereby a subject retains the freedom to engage in addictive behaviour (Heather 2016). On the other hand, however, there is a parallel narrative that sees alcoholics as overpowered by the chemical effects of alcohol and thus not entirely responsible for their actions. This view, emphasizing the material capacities of alcoholism, suggests that people might not be free to quit drinking as they wish. Neither of the two interpretations of alcoholism make sense in the Runa context. My claim is simple: while the conspicuous amount of alcohol consumed in Runa villages might lead external observers to classify the phenomenon as alcoholism, the impression is misleading. Even those Runa who describe their own practices as encouraging ‘alcoholismo’ do not refer to alcoholism in the same way we understand it. After having explored the value of drinking in Runa villages, I examine whether the discrepancy between what people propose to do (quit or reduce drinking) and what they effectively do can be considered as an instance of weakness of will. I do not provide any definitive answer to this question, but I bring evidence to suggest that, for Runa, such a discrepancy is not an issue of great moral relevance. What seems to us as an evident failure to act following one’s previous decisions is a phenomenon which leaves the Runa untroubled. No social stigma or shame is attached to instances in which people fail to perform according to what they had previously proclaimed they wanted to do. Instead, there is a widespread acceptance of the idea that one might not live up to resolutions taken beforehand.

74

Francesca Mezzenzana

People can change their minds without that being taken as evidence of moral failure. The materials discussed in this paper come from over thirty months of fieldwork in Runa communities along the Bobonaza river, in the region of Pastaza. The observations I make here are valid for most of the rural communities in the region, where the consumption of alcohol follows very specific rules. My argument, however, might not apply to the phenomenon of drinking in urban communities where the drinking behaviour, the norms and even the types of alcohol consumed starkly differ from those of rural forest villages. Before I delve into the ethnography of drinking, however, I wish to give a brief historical and cultural background to the Runa.

The Runa: Between Independence and Care Runa people have traditionally inhabited a large portion of what is today known as the region of Pastaza in Ecuador. People in rural communities along the Bobonaza river in the region of P ­ astaza – w ­ hich is the area where I have conducted most of my fi ­ eldwork – l­ ive mostly off subsistence agriculture, fishing, hunting and some informal labour. Villages in the Bobonaza area were established by Dominican missionaries in the 18th century. In their early history and throughout the rubber boom that took place in the Amazon between 1890 and 1910, such communities were extremely volatile. Most Runa families lived dispersed in their hunting sites and only came to the ­village – ­which consisted of a few houses, a school, and a ­church – ­in order to attend community festivals or seek medical help. It was only during the 1970s, with the rise of the national indigenous movement and the creation of more schools that people abandoned their hunting sites and came to live permanently in the villages. Today, most Runa communities are organized around an airstrip, a school, and a health centre. Bilingual in Spanish and Kichwa, Runa people have today entered, if only peripherally, the market economy. From hunting to making pottery, the Runa staunchly defend the uniqueness of their way of doing things: it is rare for someone to ask for help or spontaneously offer advice on other people’s work. Ideally, a family unit should be able to be completely autonomous, with men hunting, fishing and building houses and women tending swidden gardens and making manioc beer. Being able to exert one’s choice without any interference is considered to be a fundamental right. Among the Runa, I have often been struck at how individual choices, from basic to important ones, are mostly taken individually and rarely discussed with others. In everyday interactions, people do not give advice, hardly make attempts to change someone else’s mind and prefer to verbally articulate their lack of approval in the absence of the concerned person. To the Runa, each person is the unique owner of his will. The Runa call this will munay or munana, which can be translated respectively as ‘desire’ or ‘to want/to desire’.



Drinking and Weakness of Will

75

When people refer to munay, they do not talk only about the impulses to fulfil one’s basic ­needs – s­ uch as that for warmth and food in the case of babies. They also use the term to refer to a broader range of ideas about volition, curiosity, interest; anything that, broadly speaking, could also be referred to as someone’s will or intent. For the Runa, being the owner of one’s will means that one is held to be entirely in charge of one’s decisions. This does not mean that others’ actions cannot affect you, but rather than what one chooses to do is seen as your own, individual responsibility. This also applies to categories of ­people – ­such as children or mentally ill ­people – ­whom we would not normally think as being able to make such decisions but who, among the Runa, are regarded as having an individual munay that must be respected. While, on the one hand, Runa people consider their material and emotional independence a virtue, on the other hand, to them ‘living well’ (sumaclla causana) consists of living together with others. Living well means caring for each other, through acts of feeding, sharing and working together among members of the extended family and community (ayllu) (Uzendoski 2004). A typical Runa ­family – ­comprising parents, grandparents and other ­relatives – l­ ives in a household often at some distance from other neighbours. But everyday life is punctuated by frequent social visits. Privacy is virtually absent in a Runa household, where the flow of visitors coming and going to eat, sleep or chat, is constant and any condition of loneliness is seen with suspicion and worry. To be lazy (quilla) and stingy (mitsa) are two of the worst insults a Runa person could receive and one which generate great shame. To call someone lazy or stingy is to imply he/she is not willing to give things to others and, as a consequence, that she is unable to enter into meaningful social relationships with other members of the community. To be a ‘person’ (Runa) means to be able to enter into productive relationships with others. This entails the capacity and the will to both works autonomously (to be self-sufficient) and to care for others (to be socially responsible). Care is often framed in the idiom of food: to care for others is mainly manifested in pragmatic terms rather than in affective psychological terms. As anthropologists have demonstrated in other regions in the Amazon, kinship ties among the Runa are based not so much on blood relatedness but rather on co-residence. Those who drink and eat together are considered to be ayllu (kin). All acts of sociality involve the consumption of manioc ­beer – ­a drink made of fermented ­manioc – ­considered as the key substance of any Runa household (Uzendoski 2004). Manioc ­beer – a­ quintessential female product, always made by and associated with w ­ omen – ­is a drink made out of boiled manioc which then is smashed and chewed for a couple of hours. Enzymes in saliva break down the starch contained in manioc and turn it into sugar. Sugars aid the fermentation process that can take up to five days, depending on the desired alcoholic content. As among other Amazonian indigenous people (Sztutman 2008), women make different types of manioc beer which are used for different purposes. The first is a thick drink with very low alcoholic content and it

76

Francesca Mezzenzana

is mainly made for household consumption. This latter is a thick nutritious beverage that is often consumed in lieu of food. The second most common type of manioc beer is much more liquid and higher in alcoholic content: this is made for events such as work parties, birthdays and ceremonial parties and is consumed with non-kin. The two beers serve two diametrically opposed purposes: the first is to feed people, to make them feel full; the second is to make people drunk and to empty their stomachs through vomiting (see also Viveiros de Castro 1992). The ways in which beer is served also changes depending on whether drinking takes place in the domestic realm or in the context of a public gathering where non-kin are invited. While the first takes place within a relaxed, laidback atmosphere, drinking at larger social gatherings has a distinctive character of intense sociality and competitiveness. Drinking with non-kin happens most frequently in occasion of collective work parties (minga). These are a form of collective work very popular throughout South American indigenous societies (Harris 2007). Such parties are sponsored by a household who invites kin and non-kin to help with tasks that require extensive cooperation (e.g. the building of a house). The party ‘owners’ or sponsors, prepare abundant manioc beer for their guests. After a couple of hours of collective work, the party sponsors invite everyone to return to the house to eat and drink. From then on, people drink until the manioc beer is over, which often, given the large quantities produced, takes a day or two of almost uninterrupted drinking. Because of its protracted nature, it is this type of social drinking that is often understood by non-Runa people as a practice that encourages alcoholism. Runa people themselves mention these occasions as the origin of marital fights, violence and other disruptions. I will now turn to examine more closely the dynamics of drinking during Runa work parties.

Making People Drunk When guests arrive to the host house, Susana, the female head of the household and the sponsor of the drinking party, promptly serves manioc beer to women and men, who sit separately on the benches in the large oval house. Susana starts with Juanita, a young woman who comes from a different hamlet and is unrelated to her. Susana holds the drinking bowl close to the Juanita’s mouth. As she holds the bowl, she laughs and make jokes. Juanita sips just a little bit. Then she moves the bowl away. Susana gently pushes the bowl back and presses it against Juanita’s lips again. ‘You haven’t drunk anything. Drink!’ she says. She inclines the bowl to force Juanita to drink all the beer. Juanita drinks. Rivulets of beer run from the angles of her mouth, across her chin and neck. While holding the bowl, Susana looks away. When Juanita finishes, the pushes away the bowl and wipes her mouth and neck with a hand. Susana moves on to the next guest but, behind her, arrives her son, who quickly offers



Drinking and Weakness of Will

77

Juanita another bowl full of beer. Each person seated on the bench receives multiple rounds of manioc beer by multiple people. Since everyone in the host ­house – ­children and a­ dults – p ­ articipates in the serving, there is very little time between one bowl and the other. The quick turnaround of bowls results in people getting drunk very fast. After a few rounds in this house, people then proceed to visit another house. The drinking continues from house to house until late at night. What I have just described is a typical scene from a collective work party. The etiquette of Runa drinking on these occasions has one very simple objective: to make people drunk. To have drunk guests is the joy and pride of any household, and failing to do so gives birth to criticisms and gossip. A common complaint of guests, for instance, is that a specific host has not prepared enough manioc beer or that the beer is not fermented enough to make people drunk. Merely giving beer to guests is not enough: getting them drunk is what really counts. That is why, when the hosts see their guests arriving from a distance, they start frantically to fill bowls with beer and then stand in a corner of the house, bowls in their hands, as if ready to attack. Serving beer to guests is in fact conceived as a kind of playful attack. Hosts force beer down people’s throats, even at a point in the drinking when those people cannot even walk. This aggressive stance towards serving manioc beer was perhaps one of the most difficult lessons I had to learn during my first period of fieldwork. My friend Rosa repeatedly explained to me that, when visitors came to our house, I had to make them drink forcefully (sinchi). I could not have pity for them. ‘Make them drink until the bowl is empty,’ she told me. The first times, however, I was too shy to follow her instructions: I feared that pressing the bowl to someone’s lips, inclining the bowl and literally pouring the beer down their throats was too much of an imposition. People often implored me to stop, others pushed away the bowl, and a few refused to drink if I didn’t take a sip first. Women told me that in this last ­case – ­if someone repeatedly refused to drink unless the host drinks ­first – ­I should empty the bowl on their heads. ‘Drink all of it or I am pouring it on your head’ is a common admonishment made by hosts to their reluctant guests. I could not bring myself to do it; hence, I myself drank first, and often ended up completely drunk within a few hours. Rosa was extremely disappointed. She told me to be strong and make people drink, whether they liked it or not. We ought to notice two things about this kind of drinking. First of all, it is extremely frequent. Families hold work parties every two weeks or so: this means that, within the village, many parties go on simultaneously. This also means that the party sponsors are often themselves invited to drink in someone else’s house where another work party is taking place. This creates a kind of reciprocity in drinking: I invite you to drink, and I know that you will also invite me to drink at your house. People thus move from one house to another to drink, then come back to one’s own house and serve manioc beer to their guests. It is this reciprocity that gives a distinctive playful character

78

Francesca Mezzenzana

to drinking. You know that the person you made drunk will make you drunk later. The second obvious fact about this drinking is its ritual nature: it is limited to a very specific time (work party) place (the oval house) and people (nonkin). It also follows a strict etiquette. The desired outcome of this communal drunkenness is to engender a state of happiness (cushi) amongst unrelated members of the community. Other anthropologists have noticed how, among Amazonian indigenous people, the consumption of manioc beer is a fundamental act through which unfamiliar people are transformed into kin. Through communal drinking, non-kin are incorporated, albeit momentarily, to the host group. In writing about Trio beer festivals in Guayana, anthropologist Vanessa Grotti (2012) points out that it is not a coincidence that drinking festivals intensified with processes of sedentarization and colonization. Grotti suggests that the intensification of beer drinking in contexts of interethnic contact is linked to the transformative power of drinking: by sharing s­ ubstances – i­ n this case ­beer – u ­ nrelated foreign people are actively transformed into kin. She also notices a contradiction between the normal everyday ethos of the ­Trio – w ­ ho, like the Runa, are people who greatly value individual a­ utonomy – ­and the practices that instead take place during Trio beer festivals. During such events, she explains, leaders summon people to undertake the work needed for the festival and give advice in the form of ritual speeches throughout the drinking sessions. Grotti notices how these practices of persuasion starkly contrast with the usual forms of Trio sociality, which centre around autonomy and independence and where nobody tells the other what to do or how to behave. The same happens during Runa drinking parties. On occasions of collective work parties, a leader (the party sponsor) organizes collective work and gives indications to his helpers. Throughout the day, the same person repeatedly gives speech to encourage his guests to live well and in peace, without fighting or arguing. Constant admonitions punctuate the drinking sessions. There is yet another striking way in which the values of autonomy and independence that are central to Runa social life are challenged during drinking parties. The challenge lies in the very act of drinking. I have already mentioned how Runa people feel it is inconceivable to force someone to do something he doesn’t want to do. Drinking is one of the few, and certainly the most conspicuous, exceptions that I am aware of. During Runa collective drinking, you are forced to drink. Forcing someone to do something becomes the general rule. In fact, not forcing someone to drink, as I sometimes did, moved by self-shame or pity for my guests, is taken as an example of someone who is not knowledgeable. Only young children who lack social skills do not force guests to drink. I have always been amazed at how hosts, no matter how drunk the guest is, still give him beer. As long as a guest stays conscious, any complaint, any protest or objection to drink more is always and swiftly ignored.



Drinking and Weakness of Will

79

In the act of drinking, those who get drunk surrender their own will to that of others. In becoming passive receivers, they allow others to affect them. The hosts who serve manioc beer manifest an evident pleasure in making people drunk. ‘I made him drunk!’ women exclaim joyfully as the guests leave the house. Stories about how one managed to make someone else drunk continue to circulate well after the drinking party is over. Grotti (2009) argues that among the Trio, nurturing others amounts to exerting a kind of ‘regressive control’ by which those who are fed turn into dependent children (see also Costa 2017). She draws an interesting parallel between the ‘chewing’ of manioc beer and the first chewed foods an infant receives, pointing to how Trio people temporarily become people who are fed, like newborns. Among the Runa, too, manioc beer has certain maternal connotations (see also Uzendoski 2004, 897): not only is beer the first food that babies consume, but it is also sometimes referred to as ‘liche’ (milk) due its white colour and texture. In letting themselves be transformed into passive dependents, the Runa demonstrate a great degree of trust: as elsewhere in Amazonia (Oliveira 2019; Sztutman 2008), manioc beer is not only a nourishing substance but also the primary vehicle through which enemies can poison you. Drinking is thus a realm where one can submit one’s will to that of others and where one can also exert a power over others. This is true of all drinking, not just the kind of collective drinking that I have described. Drinking relationships between non-kin people are characterized by a higher degree of coercion and forcefulness. However, this power dynamics never entirely disappears from any act of drinking, even within the domestic sphere. This is because drinking among the Runa is not conceived as an action that has its origin in the drinker. Getting drunk (machashca) or drinking generally (upina), is never understood as an action initiated by the drinker. Significantly, the Runa hardly ever say ‘I drank’ (upirani), but say instead, ‘I am made to drink’ (upichishcani). This is true in both practical and symbolic terms: manioc beer is something that you are always given to drink. For instance, a man does not drink manioc b ­ eer – h ­ e does not take a bowl, fills it with beer and then drink i­ t – ­but rather he always receives it from his wife. Similarly, while a woman beer maker might sip her own beer, someone else in the house, most often a daughter, will serve her the beer. Drinking beer always puts someone in the position of the receiver, in a passive position, and someone else in the position of the giver. Compare now this with drinking as it happens in a European context, for instance, in a pub or a bar. You engage in social drinking: you go to drink a beer together with your friends. You buy a beer, you buy your friends a beer, they reciprocate. In this context, there seems little doubt that it is you who decides to drink and get drunk. On some occasions, friends might attempt to get you drunk by buying you more and more drinks, on others they might even pour beer down your throat. Yet, in most social gatherings, getting drunk is an event over which one believes to exert some control. In the context of pub drinking, you would consider yourself as the person who decides to get drunk

80

Francesca Mezzenzana

and who therefore has the responsibility of having drunk too much. This is not at all the case for the Runa, for whom, getting drunk always has its origin in the actions of someone else. It is others who make you ­drunk – ­although you freely choose to submit yourself to their will. While people are free to choose whether to visit or not the house of the hosts, from the moment they start drinking, they entrust themselves to the will of others. It will be t­hey – t­he ­hosts – w ­ ho will literally decide what to do with you. From a Runa perspective then, the cause of drinking lays outside the self.1 This is one of the reasons that drinking among the Runa is not conceptualized, nor does it manifest itself, as a form of addiction. People do not start drinking on their own; they drink because someone else wishes them to do so. In fact, lone drinking is virtually absent in Runa villages. The only time in which I witnessed someone drinking alone was the case of Javier, an unmarried man in his thirties. Javier was unanimously identified as an alcoholic (alcoholico). He spent his days laying in a hammock in the house of his older brother. Nobody in the village invited him to join work parties since his family members were worried about his health. Javier was seen as a problematic person because he did not work, did not participate in family life and drank on his own. Javier’s case was indeed perceived as an instance of alcoholism just as it is commonly u ­ nderstood – h ­ is constant state of drunkenness had led him to the point of not being able to engage in any productive activity and he has been slowly excluded from meaningful social relationships. Javier’s case illustrates that collective Runa drinking and alcoholism, as we conceive it in addictive terms, are incompatible. Notice, in fact, the irony: ­Javier – t­ he only recognized alcoholic in the ­village – ­was also the only person not allowed to participate in collective drinking sessions. Javier, with his inability to work was excluded from collective drinking because in order to participate in such events one needs to be a fully productive adult, a real person. Runa communal drinking can only take place between people who are equals. Earlier on, I suggested that whereas the everyday ethos among the Runa focuses on independence and autonomy, occasions such as collective work parties make manifest the other aspect that is essential for a good life (sumac causana): interdependence. Drinking, with its ritualized etiquette, display of coercion and reluctance of the drinkers, makes obvious that ­people – e­ ven people so autonomous like the ­Runa – d ­ epend on others. And nevertheless, this display of interdependence lies on an essential caveat: interdependence can only take place between equals, between people who are sufficiently autonomous and responsible. Forcing Javier to drink would have been inappropriate (or even grotesque) since the play between submission and force would have not taken place amongst equal partners, amongst ‘real persons’. If drinking among the Runa is thus about celebrating simultaneously people’s dependence on each other and their equality, what to make about people’s claims that they should quit drinking (and the fact that they never do so)? From what I have described so far, to quit drinking for a Runa would not just



Drinking and Weakness of Will

81

mean giving up a habit but would rather entail a radical moral transformation. One could suggest that perhaps the Runa do not give up drinking because, despite its negative aspects, this constitutes a fundamental expression of the moral tension between independence and dependence, a tension which intimately shapes Runa subjectivity. The act of drinking together between kin and unrelated people in the village resolution was not a good resolution after all if he wanted to live as a Runa amongst his people. What I now wish to bring attention to is another interesting ethnographic fact: the apparent lack of concern the Runa display towards what we interpret as instances of weakness of will. When we detect discrepancies between what one’s proclaimed intentions and the actual actions that follow, we readily identify them with the negative denomination of weaknesses of will, as if the discrepancy between words and actions were a failure, a defect in the normal functioning of the self. The Runa, while also noticing the inconsistency, seem instead relatively unconcerned about them. When the Runa promise not to indulge in excessive drinking and routinely fail to act accordingly, no social shame nor stigma is attached to such ‘failures’. In the next section, I am going to provide some ethnographic examples of this general disinterest and reflect on some of its implications for thinking about weakness of will.

To Drink or Not to Drink? It is a humid morning, and I am sitting in the large oval house that hosts community events. Parents, teachers, and other members of the village are there for a secondary school teacher–parent meeting. The teachers, as often is the case in rural villages, are mestizos, mixed race people of Spanish descent who come from the city. Sitting in a circle with parents, village and school authorities, I prepare for what looks it is going to be, in every respect, a rather dull meeting. When a local indigenous officer – el teniente politico – suddenly enters the house, I wake up from my dormant state. The officer declares he found a bottle of aguardiente in the teachers’ sleeping quarters and identifies it as belonging to Alonso, a twenty-five-year-old mestizo teacher. Aguardiente is a strong alcoholic beverage made from sugarcane. In contrast to the locally produced beverage, manioc beer, aguardiente has a much higher alcoholic content. Aguardiente is often described as being the cause of many social ills. Fights, domestic violence, accidents and ­hepatitis – ­the latter being a major cause of death among Runa ­people – ­are often explicitly linked to the consumption of aguardiente. The local police officer begins to explain with a solemn voice that aguardiente is the reason why husbands beat their wives, the reason why men no longer spend time hunting and fishing but rather sit around to drink, leaving their families in hunger. He speaks for a long time, holding the incriminating bottle in his hand. He then addresses the teacher to whom the bottle belongs,

82

Francesca Mezzenzana

and says that the community did not expect him, someone who was there to give a ‘good example’ to youths, to bring a bottle of aguardiente to the village, and in so doing, contribute to the ‘loss of Runa culture’ (la perdida de la cultura Runa). A few men and women sitting in the circle nod in assent, many others are giggling at the teacher, whose face has suddenly turned purple and who sits with increasing unease on his chair. The amusement is palpable: after the words of the police officer, many others lift their hands to talk. Women have the harshest words for the teacher’s behaviour, claiming that aguardiente is destroying their families. A mestizo woman who has lived many years in the community reiterates the words of the police officer. She claims that mestizos who bring alcohol to Runa villages, put in danger the social fabric of the community. After many of such accusations, the teacher clumsily asks for permission to speak. He is visibly agitated: he defends himself by saying that he had brought the bottle only for personal consumption and not to ‘corrupt’ students. Alonso also adds he never comes to classes drunk and that he never offers alcohol to anyone. After the teacher’s words, a few other people raise their hands to speak. The first of them is a man in his mid-thirties, a parent of two children who attend the local primary school. He says that, while everyone agrees that aguardiente is bad, Alonso can do whatever he likes his spare time. ‘Nobody can tell him what to do,’ he concludes. After his speech, many other people stand up and talk in a similar manner. All of them condemn aguardiente consumption but defend the teacher’s freedom to drink it if he wishes. More people stand up to reinstate that no p ­ erson – l­et alone a m ­ estizo – s­ hould be allowed to bring aguardiente to the village. It becomes clear that the issue with the teacher is not that he might have drunk aguardiente during school hours but rather that, by bringing alcohol to the village, he could have corrupted the social fabric of the village. As people stand up to talk, others giggle, chat among themselves, other make humorous remarks. All of a sudden, a man, who has sat quietly throughout the meeting, stands up and starts talking vigorously in defence of the teacher. He is soon invited to sit and calm down: he, himself, is very drunk. People laugh loudly. At the end of the meeting, people agree that the bottle should not be returned to the teacher. Men begin joking with the police officer, trying to persuade him to give it to them for they know how to take good care of it. At the end, the bottle is emptied on the ground, in front of everyone, amidst laughter and amused shrieks. No action is taken against Alonso, who looks rather upset by the accident. The day after the meeting, it is a Saturday and a large party takes place in the schoolyard. Manioc beer and aguardiente are served in abundance and people, men and women, among them parents, village representatives and teachers who were present at the meeting the day before, begin drinking early in the morning and continue throughout the day. The party effectively ends the following Monday. Over the course of the weekend, practically everyone in the ­village – e­ xcept young ­children – ­is completely drunk.



Drinking and Weakness of Will

83

Aguardiente is often consumed by Runa people in occasion of collective work parties in a manner similar to manioc beer. However, aguardiente is unanimously associated with many social ills. Most Runa friends claim that aguardiente is the cause of marital fights and disease and for years, leaders have discussed the possibility of banning the consumption of aguardiente in the village (and one did so, albeit unsuccessfully as I described earlier). They have not succeeded, mainly because despite all the good resolutions, leaders themselves are the first to buy it in the city and bring it to the village to drink together with family and friends. What struck me about the teacher’s episode was that on the day following the incident, most people found it completely acceptable to get drunk with aguardiente. In fact, amongst the people at the meeting, there was n ­ one – i­ncluding ­women – ­who had never brought aguardiente to the village or consumed some quantities of it in company of family and neighbours. And yet, this did not stop people from making strong statements about the negative effects of bringing aguardiente to the village. During the meeting, when someone who was known to be an avid consumer of aguardiente stood up, people giggled and made jokes amongst themselves. I later told a friend how this had angered me. ‘This person had such a double face! How could he condemn the teacher if he was the first to drink it in large quantities?’ My friend did notice the irony: but rather than finding the man’s behaviour upsetting, he simply found it very funny. The idea that one might say something and then behave in a totally different way did not cause him any indignation. In fact, only the incriminated teacher seemed to find the situation upsetting. When I spoke to the teacher a couple of weeks after the meeting, he complained that he felt unjustly accused. ‘They are the first ones to bring aguardiente here! Not that we (referring to mestizos) don’t enjoy drinking! But we know when it is time to work and when it is time to drink! They don’t care!’ he said, adding that it was only since he had moved to the village that he had begun to drink heavily: ‘No matter how much I said no,’ he told me. ‘They laugh and just force me to drink.’ What I find interesting in the teacher’s episode is the relaxed attitude that Runa people display towards behaviours that appear to be inconsistent with their moral stances. The day following the meeting, the very same people who had spoken harshly against the teacher were happily drinking aguardiente together. When I jokingly made a reference to the scandal about the teacher’s behaviour, one of the men I was talking to, holding a cup filled with aguardiente, laughed heartily and said: ‘We shouldn’t be drinking too much, that is bad, that is true … drinking brings too many problems, illness, trouble in the family … but that is our life, that’s how we Runa live.’ Amidst cheerfulness and laughter, people talked about the school meeting and the teacher. ‘Poor man,’ laughed my friend Inacio who was present at the meeting the previous day: ‘he was so sad when we poured his aguardiente on the floor! But it is no good to bring aguardiente! He should have known better!’ Nobody seemed to care about the discrepancy between their actual behaviour and the claims they had

84

Francesca Mezzenzana

made the day before. What struck me as inconsistent behaviour, was something that people hardly noticed or cared about. Similarly, many cases that we would readily interpret as examples of a weakness of will seem to go relatively unnoticed by the Runa. For instance, during my last stay in the village of Wituk, my host Diego was diagnosed with stomach ulcers. He was forbidden from consuming alcohol. For a few months, Diego abstained from drinking manioc beer and other alcoholic beverages. Then, in occasion of a collective party he was offered manioc beer and he did not refuse. By the evening he was very drunk. In the days following the party, Diego laid in his hammock, looking miserable. He complained of intense stomach aches. I checked on him repeatedly to make sure he was fine. I was in the kitchen with his wife when we heard him moaning from his hammock. ‘That Diego’ commented his wife to me as she handed me some plantains to peel, ‘He complains that he is sick, he doesn’t drink anything for weeks, and then of all a sudden he gets drunk! There he is now, sobbing like a baby!’ She laughed loudly. When we went to see him, she handed him a plate full of food and mocked him: ‘I am not going to drink again! I am not going to drink again,’ she said in a falsetto voice. Diego took the plate and laughed slyly. His sons also joined the chorus and made some jokes about their father. Diego laid in the hammock, laughed feebly too, and then drifted into sleep. The matter was not discussed further. He repeatedly swore he would give up alcohol. Diego kept alternating periods of abstinence to periods when he drank heavily. Sometimes he was so much in pain he had to travel to the city to see a doctor. Diego, who failed and keeps failing to keep up with the promise he will not drink again, seems to be, from our analytical perspective, a classic example of a person who is weak-willed. His weakness would probably puzzle us: why does he not give up alcohol, we might try to persuade him to stop. After hearing his resolutions over and over, we might even start looking at him with contempt, knowing that he will never do what he promises. Among Diego’s close kin, however, nobody seemed to be concerned about his inconsistent behaviour. People do notice ­it – ­as his wife’s remarks ­demonstrate – ­but nobody reproached him for his lack of consistency. His family members listened silently to his promises and never spent a word noticing how many times he had failed in his attempts. The jokes also end as quickly as they begin. Diego’s behaviour did not provoke any strong reaction from his family members, except humour. His wife merely mocked him: she did not shame him, nor advise him, nor tell him what he should do for his own wellbeing. As I mentioned earlier on, the absence of advice can be explained by the Runa ethos of independence and autonomy, whereby decisions over one’s life are taken without the interference of others. One knows what is best for ­himself – ­and others hardly intervene to give advice or suggestions about how one should live. There is also a sense in which the decisions of o ­ thers – ­from a Runa ­perspective – ­are not to be discussed because they are only known to the agent. Whenever I asked elucidation for someone’s behaviour, a common



Drinking and Weakness of Will

85

answer I received was: pay yachan! ‘He/she must know!’ Alongside the value given to psychic privacy however, there is undoubtedly a relative indifference to the inconsistencies between Diego’s expressions of intentions and in actions. Such inconsistencies, what we would call weakness of will, did not matter2. What does this lack of interest tell us about the concept of weakness of will? In my concluding remarks I turn to explore how my Runa ethnographic materials interrogate some of the assumptions upon which the debate on weakness of will rest.

Conclusions I started this paper by asking an ethnographic question: why do Runa people say that drinking is bad and nevertheless keep drinking? I tried to address this question by bringing to light the role that collective drinking plays in the constitution of Runa moral selfhood. In particular, I suggested that the emphasis on interdependence that characterizes collective drinking provides an important counterpart to the everyday ethos of autonomy that shapes Runa life. I suggested that to refuse to participate in collective, extended drinking sessions would be much more than simply quitting a bad habit: it would amount to a real moral transformation. I believe that it is because of this that, despite what Runa people might claim, to quit drinking really look unfeasible if one is to live as a ‘proper’ person in a Runa community. From this perspective, people’s claims that they should stop drinking could be interpreted as empty words, not representing any real intention: words which are pronounced without believing in them too much since translating the intentions into actions would imply a too radical moral transformation (and surely not the most rational thing to do). However, I fear that this explanation, by suggesting that drinking might not be a real example of weakness of will (and thus implicitly suggesting that there might be other, clearer instances of weakness of will in the Runa context) really diverts the attention from a central fact that emerges from my ethnographic materials: Runa people’s relative indifference to what seemed to me inconsistent behaviour. Diego’s or Renan’s behaviour, much to my dismay, did not elicit any reaction from their kin or neighbours. They were not considered as untrustworthy by other villagers, nor did they have the stigma that is usually attached to those who are weak-willed (Horton 1999). People seemed to blissfully ignore their inconsistent behaviour, just like they did with their own. When I started writing this paper, I wondered about this indifference. Why do Runa people not recognize weakness of will as a problem? I have not found an answer to this question. This is perhaps because the question itself was ill-conceived. Posed as such, the question required me to think of the Runa social world and behaviour in the form of an absence, of a lack, rather than on its own terms.

86

Francesca Mezzenzana

And yet, the fact that is so difficult to think about weakness of will in a Runa context, in addition to the lack of social attention given to the issue enlightens the fact that rather than being a natural cognitive condition, weakness of will is a matter of ethical and social relevance only under specific social and cultural conditions. The concept of weakness of will arises because we tend to think that intentions should translate into actions, and when this does not happen, we are faced with an inconsistency. We perceive such inconsistency even if, as McKearney observes in this same volume, it is obvious that we author actions in many different ways during our lives, and that such a consistent will that takes into account all the possible options and then act on it does not correspond to how we behave in everyday life. And yet, despite this knowledge, every time that I observed inconsistent behaviour in the field, I was puzzled. My repeated efforts to ‘explain’ Runa absence of a weakness of will in this chapter, as well as a depiction of my surprise when people failed to acknowledge what to me seemed obvious and problematic inconsistencies of behaviour are revealing of some of the assumptions which underpin the debate on weakness of will and broader discussions around rationality. One such assumption is that inconsistencies of behaviour are necessarily problematic. When faced with inconsistent cognitive behaviour, anthropologists, like philosophers, experience a desire to make sense of it, to explain it. As anthropologist Charles Nuckolls (1993) has noticed, this desire for consistency derives from a particular history of Western scientific objectivist knowledge for which cognitive consistency is understood as being a basic, natural and universal capacity. It seems natural to us (where ‘us’ stands for Western academics) that inconsistent behaviour should be explained and made consistent ‘at some level’ in the analysis (Nuckolls 1993, 7). This is because our work as social scientists is built upon a specific cultural construction of explanation, one which heavily relies on the need for facts and actions to be consistent. My interpretation of the social value of drinking could be read only as such an attempt: given that drinking mediates experiences of dependence and independence, which are key to Runa social life, to quit drinking would actually not be the most reasonable thing to do at all. Under this perspective, Runa inconsistencies of behaviour are really not inconsistent and weakness of will is not really a weakness of will. I mentioned earlier how such interpretation, however reasonable, leaves me unsatisfied. This is because, by transforming inconsistent behaviour into rational behaviour my analysis circumvents the possibility, more difficult to conceive, that inconsistency is real and that in fact it does not seem to bother Runa people. What I think the Runa ethnography presented here clearly shows is that the desire to explain inconsistent behaviour such as weakness of will is itself a culturally specific desire. To ask the question ‘why are the Runa unconcerned about inconsistent behaviour?’ is already making the suggestion that consistency is a desirable, natural state. Instead, the question should be turned



Drinking and Weakness of Will

87

upside down and transformed: ‘why is it socially desirable for intentions and actions to be consistent?’ This does not amount to saying that Runa people never experience inconsistencies or never find them problematic. Instead, as Nuckolls (1993, 2) suggests, inconsistency might be domain-specific: while certain inconsistencies in the realm of formal explanation and cognition, such a weakness of will, might strike us, the Runa might puzzle around inconsistencies in other realms. The Runa people’s lack of interest toward questions of consistency of behaviour amplifies our own anxiety about finding a consistency between intentions and words and by doing so, it raises questions about the cultural and social circumstances that enable weakness of will to be considered as a problem worthy of attention and response in our own society. Francesca Mezzenzana is an anthropologist working with the Runa of the Ecuadorian Amazon. She has a PhD in anthropology from the London School of Economics and is currently the Principal Investigator of a Volkswagen Foundation Freigeist Grant, based at the Rachel Carson Centre in Munich, that explores children’s understanding of the natural world in different cultural settings. She is particularly interested in learning, child moral socialization and human–nonhuman relationships. Her research was awarded funding by the UK Economic and Social Research Council, the European Research Council, the Musée du Quai Branly and the National Geographic Society.

Notes   1. For a recent and insightful reflection on Amazonian theories of agents and causes see Kelly and Matos 2019.   2. While the Runa seem to display a general tolerance or indifference for inconsistent behaviour, not all inconsistent behaviour has the same value, and not all agents will be judged equally for it. For instance, suppose that someone took responsibility for hosting a drinking party and did not provide enough manioc beer. Given the importance of drinking for the Runa moral world, the failure to make others drink would certainly cause uproar among family and neighbours and would almost certainly lead to gossip and open criticisms. While people generally tolerate the unfulfilment of most commitments, some commitments, such as the organization of drinking ­parties – ­matter more than others. It also has to be emphasized that a gender asymmetry deeply shapes the evaluation of inconsistent behaviour. Certain inconsistencies of behaviour will not be lightly judged if undertaken by women and, importantly, if they concern the provision of food and drink. For example, if a woman fails to provide food or manioc beer (as her husband or kin would expect from her), then she would face serious consequences. The opposite case, where it is a man who fails to provide for his wife and children, is judged much more leniently by the wider community. While these are not exactly examples of inconsistent behaviours but more of ‘failed’ commitments toward others, they reveal not only how inconsistent behaviour might be evaluated differently across cultures but also that it is shaped by gendered power dynamics. Not everyone, so to speak, has the same freedom to engage in specific forms of inconsistent behaviour.

88

Francesca Mezzenzana

References Costa, L. 2017. The Owners of Kinship: Asymmetrical Relations in Indigenous Amazonia. Chicago: Hau Books. Dodd, D. 2009. ‘Weakness of Will as Intention-Violation’. European Journal of Philosophy 17: 45–59. Grotti, V. 2009. ‘Un Corps en Mouvement: Parenté, «Diffusion de l’Influence” et Transformations Corporelles dans les Fêtes de Bière Trio, Amazonie du Nord-est’. Journal de la Société des Américanistes de Paris 91(1): 73–96. . 2012. ‘Happy with the Enemy: Kinship, Pacification and Corporeal Transformation in Trio Beer Feasts, Northeastern Amazonia’. Anthropology and Humanism 37(2): 191–200. Harris, O. 2007. ‘What Makes People Work?’ In Questions of Anthropology, edited by R. Astuti, J. Parry and C. Stafford, pp. 137–167. Oxford: Berg. Heather, N. 2016. ‘Addiction as a Form of Akrasia’. In Addiction and Choice: Rethinking the Relationship, edited by N. Heather and G. Segal, pp. 133–148. Oxford: Oxford University Press. Holton, R. 1999. ‘Intention and Weakness of Will’. Journal of Philosophy 96: 241–62. Kelly, K. and Matos, M. 2019. ‘Política da consideração: ação e influência nas terras baixas da América do Sul’. Mana 25(2): 391–426. Nuckolls, C. 1993. ‘The Anthropology of Explanation’. Anthropological Quarterly 66(1): 1–21. Oliveira, A. 2019. ‘Manioc-Stem Transects: Vital Flows, Technical Processes and Transformations’. Vibrant: Virtual Brazilian Anthropology 16: 16552. Pierre, F. 1983 [1889]. Viaje de Exploración al Oriente Ecuatoriano. Quito: Abya Yala. Porras, P. 1979. ‘The Discovery in Rome of an Anonymous Document on the Quijo Indians of the Upper Napo, Eastern Ecuador’. In Peasants, Primitives, and Proletarians: The Struggle for Identity in South America, edited by D. Browman and R.H. Schwartz, pp. 13–47. Berlin: De Gruyter Mouton. Sztutman, R. 2008. ‘Cauim, substância e efeito: sobre consumo de bebidas fermentadas entre os ameríndios’. In Drogas e cultura: novas perspectivas edited by B. Labate, S. Goulart, M. Fiore, E. MacRae and H. Carneiro, pp. 219–50. Salvador: Edufba. Uzendoski, M. 2004. ‘Manioc Beer and Meat: Value, Reproduction and Cosmic Substance among the Napo Runa of the Ecuadorian Amazon’. Journal of the Royal Anthropological Institute 10(4): 883–902. Viveiros de Castro, E. 1992. From the Enemy’s Point of View. Chicago: Chicago University Press. Viteri, C. 2003. ‘Sumak Kawsai: Una Respuesta Viable al Desarrollo’. MA thesis. Universidad Salesiana.

4 Prayer, Demons and Akratic Sublation Jon Bialecki

This is an essay about how, and in what way, a particular set of anglophone American Christian believers at once are, and are not, akratic. This discussion is based on the intuition that there is something slippery about akrasia as a condition; or, to be more exact, there is something slippery, at least as far as akrasia is thought through as an anthropological problem when using the relatively narrow lens of North American religion. Because of what might look, to a secular vantage point at least, like a certain kind of internal contradiction in how akrasia operates in the domain, our discussion will be more of a tour of the terrain than an argument, though I hope to point out some of the more interesting features of the landscape. Our discussion will touch on ‘scaling’ in how and by who akrasia is expressed: we will see that it works at the level of individuals, but also that we can see akrasia (or something akin to akrasia) in larger groups as well. Though, to be fair, speaking of the ‘individual’ when considering akrasia is to miss the mark. As in so many other phenomena attended to by the discipline of anthropology, akrasia will be shown to be ­relational – ­or at least revealed to involve a blurring of boundaries between actors. But the relations involved here are not limited merely to the human; what we might code as ‘supernatural’ agents are also an important element in the network of social ties that brings akrasia to the fore. And these supernatural actors are central in both necessitating akrasia and in escaping akrasia, or what we will be calling akrasia for our purposes here. For as we will see, akrasia is not quite the right term, even if it not altogether wrong, either. We will be dealing with something more properly situated on the borderlands of akrasia, where we are dealing with a phenomenon that perhaps is better said to rhyme with akrasia. And when akrasia is present, it will be present only in its purely virtual form, as a potentiality or imaginary object that is never fully actualized. This is not because akrasia is absent in the set of scriptures considered to be canonical by most anglophone American Christian believers. The Koine

90

Jon Bialecki

Greek version of the word (ἀκρασία) is found twice in the New Testament. The first instance is in Matthew 23:25, where it appears when Jesus denounces the hypocrisy of the Pharisees; as part of a long litany of complaints directed against members of that religious school, he lashes out at them with the admonition, ‘Woe to you, teachers of the law and Pharisees, you hypocrites! You clean the outside of the cup and dish, but inside they are full of greed and selfindulgence [ἀκρασίας].’1 In 1 Corinthians 7:5, the word is invoked by Paul to suggest that dereliction of certain ‘marital duties’ might result in a loss of selfcontrol for the deprived spouse. ‘Do not deprive each other except perhaps by mutual consent and for a time, so that you may devote yourselves to prayer. Then come together again so that Satan will not tempt you because of your lack of self-control [ἀκρασίαν].’ Paul’s use of the term in 1st Corinthians is important, more because of what it suggests about his not using that term in seemingly applicable passages that have had more of an effect on ‘Christian life’ (with that noun phrase understood broadly, of course).2 In a meditation on the relation between ‘law’ (usually understood to refer to Deuteronomic taboos) and ‘spirit’, in Romans 7, Paul states that, We know that the law is spiritual; but I am unspiritual, sold as a slave to sin. I do not understand what I do. For what I want to do I do not do, but what I hate I do. And if I do what I do not want to do, I agree that the law is good. As it is, it is no longer I myself who do it, but it is sin living in me. For I know that good itself does not dwell in me, that is, in my sinful nature. For I have the desire to do what is good, but I cannot carry it out. For I do not do the good I want to do, but the evil I do not want to ­do – t­ his I keep on doing. Now if I do what I do not want to do, it is no longer I who do it, but it is sin living in me that does it. So, I find this law at work: Although I want to do good, evil is right there with me. For in my inner being I delight in God’s law; but I see another law at work in me, waging war against the law of my mind and making me a prisoner of the law of sin at work within me. What a wretched man I am! Who will rescue me from this body that is subject to death? Thanks be to God, who delivers me through Jesus Christ our Lord!

The importance of this passage in the historical development of Christian theology would be hard to exaggerate; this is a piece of writing that has an interpretive trail running from Augustine through to Aquinas, Luther, Calvin and Barth, and was the object of a significant amount of critical attention during the fraught century that gave birth to the protestant reformation.3 Theologians and other interpreters argue over whose voice Paul was speaking ­in – ­was he presenting the vantage of the non-Christian, the pre-conversation Christian believer, the practicing Christian, or perhaps he was writing in a confessional or autobiographical vein? But regardless of who Paul is understood as ventriloquizing, it does appear that even though the koine Greek lexeme is not used, what Paul is presenting is something close to akrasia as it has been articulated in contemporary philosophical discussions.4



Prayer, Demons and Akratic Sublation

91

Contemporary American Charismatic Christianity is, in at least one way, much like the Paul of Romans: it is a form of religiosity that may not use the word ‘akrasia’, but seems to be ruminating on the akratic possibility none the less. But just as in Paul, akrasia is eventually undone, even if it is not erased. As we will see, something very like akrasia is a common concern for these believers, even if the word is not used as such. We will see it also varies considerably in scope: akrasia in the sort of faith that we will be discussing can best be framed in a sort of ‘Weather v. Climate’ sort of way; this is to say that while akrasia is not necessarily present in any particular moment, the ubiquity of akrasia has effects. As commonly understood, any specific instance of behaviour may not be characterized as akrasia, but every person (believing or not) is presumed to engage in akrasia-like behaviours and acts on ethical issues at some point, and is thus still a sinner. Further, seemingly ‘correct’ ­behaviour – ­where, when viewed from the outside, one seems to both know the good and be able to successfully ­will – i­s no escape from this trap. For there is the danger that one is doing what is correct, but that action is vitiated because it is being done for the wrong reasons. One may be acting in a way motivated by pride, for instance, and not for love of God. And it should be noted that these instances of ‘akrasia without seeming akrasia’, is far from the last wrinkle or irony we will see in this discussion.

I Up until this point, I have discussed American Protestant Christianity as an abstract and idealized system of doctrines or as a decontextualized protocol for evaluation; it tells us little about what wrestling with akrasia is like on the ground. I want to show what this looks like in motion, in a particular collectivity. I’ll be drawing on fieldwork conducted with the Vineyard, a Charismatic Christian movement that originated in Southern California, but which is now worldwide.5 In my capacity as an ethnographer I have attended Vineyard Churches and spoken to Vineyard believers from literally all over the world, but it was in the wilds of Southern California where I spent the most time with them. Vineyard believers considered themselves to be the ‘radical middle,’ meaning that they have a roughly evangelical theological sensibility and social aesthetics, including an achingly intimate commitment to a psychologized sincerity. But they also practiced the Pentecostal-style pneumatological activities not commonly associated with their Protestant co-believers (Pentecostalism and Evangelicalism being two religious sensibilities which, until quite recently, were often considered to be antithetical to one another in the American Christian world). By pneumatological activities, I mean supernatural ‘special-effects’, such as speaking in tongues, prophecy, divine healing and [important for later on] casting out demons. While the constitution of Vineyard Churches varies a great deal depending on location and history, the

92

Jon Bialecki

set of people I worked with in Southern California were mostly white, mostly middle-class believers; they worked in professions that ranged from lawyers to clinical psychologists to physicians to research academics. Though I never heard the specific word ‘akrasia’ spoken of in the field, day-to-day Vineyard language was replete with what could be taken as synonyms for that term. Believers were very much convinced of what they would call their own fallenness, a term which, as they used it, could be glossed as a chronic, and indeed inescapable, akratic condition. They also spoke frequently about their ‘brokenness’, a term for the intellectual consciousness and affective weight of this chronic akrasia, as well as for a way of speaking of its particular individual specificities. People would, at times, use the possessive when speaking of their brokenness (‘my brokenness’), and while everyone had an acknowledged or unacknowledged brokenness, different people’s brokenness was assumed to not be the same as others. One person’s talk of brokenness could be centred on, say, compulsive sexual urges or an obsession with pornography, while someone else’s might be centred around pride, or an inability to engage in emotional relationships (though it would be rare to talk about it with that degree of openness and clarity outside of something like a public confessional speech genre or a Christian counselling scenario). Fallenness was also an important concept in justifying acts of charity and forgiveness. The logic here was that since humans are by nature characterized by something akin to what we are referring to as akrasia, punitive action towards anyone for any particular failure of the will would be to hold them to an unfair, and often to a degree impossible, standard. The anger of a coworker, say, or the inattentiveness of a spouse, is to be expected in this world, and while these sorts of behaviours are not to be encouraged, it thus would be wrong to punish these people merely for being human. This also did not mean that people could not take characterological faults into consideration in future ­encounters – ­fallenness and brokenness was not a get out of jail free card for future offenses, or reasons to engage in unreflective trust of proven bad a­ ctors – ­but this was different from holding a grudge. Also, not all brokenness was the same. Some forgiving took more affective labour, and was more of a continuing process that sometimes also asked for emotional work from the person to be forgiven, rather than the award of an instant grant of clemency. We should parenthetically note that, ironically enough, due to akrasia as the universal human condition, it was also expected sometimes that people would act in an akrasia-informed manner in forgiving people; ‘I know as a Christian that I should forgive my co-worker, but …’ The other aspect that is important to understand is that the inevitability of akrasia does nothing in regard to the self ’s own responsibility. Just because everyone was by nature akratic, it did not mean that the individual did not have responsibility for their own akrasia-informed history. One can forgive others for akrasia, one can even learn to forgive oneself, but forgiveness still presupposes a primary blameworthiness (though, as we will see, even here



Prayer, Demons and Akratic Sublation

93

this can only be said as a qualified truth). Even as escaping akrasia remains an impossible ethical standard, the negative moral charge of akrasia-fuelled failures hangs over believers.

II Now, what is of note is that Vineyard believers have a technique to handle akrasia, to manage it, which simultaneously overcomes yet maintains this ethical condition in something not unlike a dialectical sublation, which works as both an erasure and a ratification. The name of this technology is prayer.6 But we should note that managing akrasia is different from controlling it. As we will see, because this technology is predicated on the weakness of the human will and the susceptibility to inhuman or greater-than-human actors, this is more akin to shaping a landscape than it is to containing or negating a difficulty. The sort of prayer being discussed here is something a hair more intense than the classical image of confessions and requests said while one is kneeling bedside at the end of the day. Solitary, ritualistic prayer does h ­ appen – ­and believers are also encouraged by pastors, family and all sorts of American evangelical self-help books to regularly spend some ‘time alone with the Lord’ every day. In my ethnographic experience, for most, this is more of an early morning activity, though, of course, the timing varies from person to person. It is also usually engaged in someplace set aside for the activity; more than one person I have known has a literal prayer closet. This prayer being discussed here is rather intended to be more like the ‘garden of Gethsemane’ than like some rote practice with a flat affect, though those who regularly practice it often admit that such heightened levels of solitary intensity are much rarer than they would prefer. But prayer can also be a social and collective activity, and here it is easy to ethnographically identify the contours of both akrasia and what we will later see as akrasia’s supplemental cure or erasure. One of the common social forums for collective yet individuate prayer among Vineyard charismatics occurs during regularly scheduled (often weekly) and relatively small gatherings of fellow believers, often, but not always, drawn from the same church. These groups can have all sorts of names: Bible study, prayer group, cell group, fellowship group, etc., but one thing that is common to most of them (at least among the Charismatic Christians I spent time with) is that they always involve a particular form of collective prayer.7 The form typically goes like this: a set of people would sit in a circle in someone’s living room, and one by one, each believer present would talk about the struggles and successes of the week, and also about the emotional state that accompanied them. This is done using a casual, almost conversational vocabulary for the most part, though shards of biblical language can also

94

Jon Bialecki

occasionally be mixed in the linguistic composite. This might sound like some of the ‘checking in’ practices that characterize a great number of self-help groups, and to a degree, it is in the way that this practice builds familiarity and cohesion within the group. What sets this practice apart from other similar checking-in exercises is that people can also use this check-in time to request prayer. This prayer can occur either when everyone has gone around the circle and had a chance to speak, or alternately take place right there and then when the request is made (this choice depends on the practice that the group has settled o ­ n – t­hese different structures aren’t thought of as carrying any theological or ethical import). An example taken from my fieldwork would be a ­person – J­acob – who is interested in purchasing a downtown condominium but is troubled by the fact that the unit is being built in a ‘redeveloping’ part of the city. The condo would actually be located in the same place where, as an employee of a para-Church organization dedicated to numerous forms of outreach, he does ministry work with the homeless, a population who would be grievously negatively impacted by the inevitable closure of resources that gentrification would bring. He is tempted because this would be a surprisingly reasonably priced unit, there is not a lot of money in this kind of ministry, and typically real estate in Southern California is rather expensive. In short, it is a contest between what two desires: one informed by a market logic of protective selfinterest, and a Christian logic of sacrifice and services. Prayer for Jacob would then begin. When the prayer starts, those immediately around the individual requesting the prayer (or, put alternately, the ‘patient,’ with the term used here in the grammatical sense) will put their hands on his or her back. Others further away in the prayer circle will lift up one of their hands, with the palm facing the person, as if one would be touching the back of the person, were it not for the intervening distance. Someone (usually, but not always, a person with a great deal of seniority) will then ‘open’ the prayer by restating the patient’s concerns, while addressing God in the second person, and referring to the patient in the third person. This person would take the originally expressed concern, and speak as if interceding on the patient’s behalf: ‘help Jacob have a sense for what your will is, give him the strength to do what it is you desire, help calm and focus him in this time of trouble,’ and so forth. Often times, other people in the circle would take pauses in the petition to present their own takes: let Jacob ‘keep in mind your love of the poor,’ someone might add, or they might use their turn in praying to rearticulate the central problem using a different core metaphor, in this case framing the problem as not about knowledge of the effects of this possible decision, but about avoiding the demonic lure of greed and money. During these moments, alternative formulations of the problem may bring out aspects heretofore not articulated in the prayer, and like good jazz improvisers, contributions offered later on in the prayer will often build on prior statements. At a certain point someone, usually the individual who ‘opened’ the prayer, will sense the energy



Prayer, Demons and Akratic Sublation

95

in the room shift and ‘close’ the prayer by restating the original petition in a pithy yet slightly different way, perhaps while also incorporating some of the other aspects of the situation that came out as others took turns improvising. Then, the prayer will be officially closed, often by something along the lines of saying that this is asked in Jesus’ name. The hope is that either during the prayer, or more likely sometime subsequent to it, Jacob will know God’s will, which in this case was not to purchase the desired condominium. This is, in a sense, a double operation, an identification of what God’s will is, but also a request for the human will to be supplemented by God’s aide, since the presumption is that these prayer-driving questions are more than just diagnostic but also require the capacity to perform what is needed. Assuming something like this clarity or determination does ­occur – ­and in Jacob’s case, it ­did – ­what will appear to have happened when Jacob demurs from buying the condominium is that he escaped falling into akrasia. As had been stated, it was a contest between worldly desire and a divine injunction. But while this may be the analysis of Jacob’s ultimate decision, a certain vantage point, that is not exactly how things will be framed following the internal logic of the system, as understood by these Vineyard believers. First, though, an aside. It must be acknowledged that the Charismatic Christians I spent time with did at times talk as if any clarity that occurred following prayer was the result of an autonomous decision. Embedded in a larger secular shared milieu, these believers are good at what linguists call code-switching, which is to say jumping from one mode of speech to another, depending on the circumstances and one’s interlocutors.8 This was particularly the case with my middle-class informants, who had to use the language of naturalistic causal forces and autonomous individualism regularly thanks to their professional entanglements. It would be easy for someone in Jacob’s position to say that he had come to a decision or threw himself into what he ‘knew’ was the right thing. But that was only the way that he would talk in front of certain audiences, or only when particular evaluations were being made. When discussing religious phenomena with like-minded believers, he was not obliged to resort to either the language of autonomous individualism or materialist causation. Rather, it followed the logic of a divine communicative event that was indifferent to both the logic and the experienced subjectivity of bounded, discrete persons. This difference in how it would be presented to fellow believers also indicates why sometimes prayer can be so striking. The response to prayer stood out because the divine voice is alien – though that is not quite the correct term, since it often has an intimacy to it. But as we will see, it certainly is strange. That is because in much of charismatic practice, one of the markers of a divine message is that it breaks in some way from what is understood as the quotidian. If this world is a fallen world, as the implicit logic goes, then an act of God would be something that breaks with the natural fallen order as they understand it.

96

Jon Bialecki

There are two chief modes through which this can occur. In the first mode, a divine message can occur when there is an event that is intuitively sensed to be improbable to the point of being considered statistically impossible. An example is the coincidence of suddenly thinking of someone that you have not had contact with for years, and having them phone you, or your running into them, the very next day; this tells you that this was not a chance meeting, but the working of the hand of God. Another classic example of divine approval is needing money for some event, and either finding or being given the exact amount of money necessary, which is taken as unqualified divine approval for the project. These are mere cases in point, but the underlying principle should be clear: any seemingly unlikely coincidence that could be understood as fraught to meaning is a candidate for being a divine message. This turns the uncanniness of the (subjectively measured) statistically impossible into something more: a divine sign that has some communicative value. Such an experience can be understandably arresting. But the answer to a prayer can be signalled in an even more striking manner. It is equally common for a divine message or act to take the form of some thought or affective state that falls upon the person; and this thought or state is so unlike the normal thoughts of the individual that they seem to have a source beyond that individual’s control. This thought or affective state is thus seen as not one that the individual has brought about through her or his own efforts, either because it is so out of sorts with that which the person enjoys, presumes a knowledge the individual does not have, or has a kind of vividness to it that makes it unlike normal cognition. These messages are considered to be divine thoughts introjected into the person. The will that carries our morally troubled individual past the abyss of akrasia is therefore not his or her own. And this is acknowledged in how people speak about these events, or rather, how they credit them. The Lord comforts, guides, carries, leads; the only real exercise of will demanded of the subject is his or her submitting to the divine. Thus, as far as consequences go, the person escapes akrasia. But as far as attribution goes, this is not because of that individual’s strong will, but despite the will’s weakness.9

III Prayer does not always address ethical conundrums, of course. As we will see shortly, a person can pray for the health or welfare of themselves or others, or for knowledge about any given state of affairs. But when prayer does touch on the ethical, it does so in a way that rhymes with akrasia, while not exactly mapping on directly to akrasia’s contours. Prayer mimics the akratic formula in that it posits a state where there is a sense of an ethical imperative, but also a wrestling desire to do otherwise. But it also collapses judgment and will, in so much that a suspicion that one is locked into an akratic situation is either



Prayer, Demons and Akratic Sublation

97

undone through a divine message confirming desire over suspected judgment, or alternately through a simultaneous confirmation that one is correct in intuiting that judgment regarding the good and one’s will to do the good are not in alignment, even as the arresting significance of the divine sign in effect serves as a supplement or prothesis to the human will, a jolt that allows previously wanting willpower, allowing them to thereby escape their prior conflict. But more follows. The request to bring willpower and judgment into alignment does not always have to happen at the level of the individual. This phenomenon scales upwards, to the level of say, a church, a denomination-like movement or even (for those of a prophetic bent) to the level of the nation itself. There are moments where a larger entity (such as a church) can feel that, by circumstances or by divine obligation communicated directly to leadership, they have to take on as a project, even though they suspect that the collective desire to engage in this work is not present. This can be cured through prayer. A church, for instance, that wishes it has more of a heart for the poor, for the stranger or (in more politically conservative churches) for the unborn, for the Israeli state or for the ‘homosexual’, can collectively implore God for the resolve that it feels it lacks. In short, the church can desire to have God change their collective desire, or to have God deepen their currently manqué commitment. It is important to note that even though the person voicing the appeal may use language that suggests a collectivity (‘let us be a people’, for instance, was a repeated term used in prayer for church-level aspirational prayer), the church, denomination or nation is, in essence, a single unit, with its constituent members sharing the same flaw with the group and each other in a fractallike manner. At one level, such requests (which usually, but not always, come from Church leadership), may seem like a rhetorical ploy to mobilize desire towards what is thought to be a worthy collective goal, and away from what is often described as a cold or powerless Christianity. And that is no doubt at least sometimes the perlocutionary effect of such language. But should whatever collectivity make such a turn, this will be discussed as a move of God, the Holy Spirit or some other similar, beyond-human agency. This is not, however, the only way that this pattern can express itself. The language and practices used in divine healing, for instance, follow a similar pattern to prayers for guidance or resolve, only here God is not supplementing a weak human will, but rather confronting the recalcitrant will of the disease, which is personified with various levels of explicitness depending on whether the group of individuals engaged in prayer lean towards or away from a naturalistic description of disease in their petition. The alternative to naturalistic accounts of disease is where disease or injury is discussed as if it was a malevolent outside agent who acts with intent. One can pray for, say, pain, illness, or madness, to go away, or alternately for the ‘spirit of ’ pain, illness, or madness to go away.

98

Jon Bialecki

If this term, ‘spirit of,’ sounds familiar, it is because it is also the language used during deliverance from demons.10 Believers cannot be possessed by ­demons – ­this sense of some kind of ultimately unbreakable, personal autonomy that marks the difference between Pentecostal/Charismatic deliverances and Catholic ­exorcisms – ­but they can be demonized. There are multiple ways that demonization can ‘present’ (to use an overly medical vocabulary); a classic form is a sudden and violent acting out during a church function. However, demons more commonly express themselves not through dramatic ‘acting out,’ but through some characterological issue, such as drug abuse, alcoholism or even engagement in particular sets of sexual activities that believers consider unhealthy or sinful. The demonized is not necessarily completely i­ nnocent – d ­ emons can attach to an individual thanks to a pre-existing character flaw. But demons can also attach due to some psychic injury or wound, such as familial or sexual abuse. And the cause can even be for ill-doings before the individual was even born, as in cases where the aetiology of the demon is rooted in parental experimentation with the occult. Demons are attracted to individuals injured in ways that thematically resonate with the demons’ nature, hence the ‘spirit of ’ designation, where one is at once speaking of the physical, psychological or spiritual symptom either caused or exacerbated by the demon and referring to the demon by title as ­well – ­but the person does not need to be the author of their own injury, which is to say that through an unfortunate history, an individual can be demonized, and thereby act in an akratic manner, not due to some internal weakness, but due to external attack or the (literal) sins of their fathers (or mothers). Sexual abuse can open up the abused to demonic attack, even though the victim is understood to be blameless; a weakness for paganism or astrology can open up a couple’s unborn children to demonic oppression, even if they were ignorant of their parent’s fault. Assuming that the demonic attack isn’t too florid, there is no way of knowing whether this is a case of an individual’s weak will or a demon’s strong will. Sometimes people have their will warped by demons, but sometimes people are just weak-willed in the first place. Indeed, this is so undecidable that most of the time, there is no real investigation into the ‘origin’ of demonization, and in softer cases, prayers may be purposefully couched in ways that are ambivalent about whether they are prayer for character reformation or for demonic deliverance. However, there are exceptions: behaviour is definitely considered to be demonic if it is both extreme in the intensity of affect displayed, or if the behaviour can be read as, on the surface, openly hostile to religion. But given that there is a continuum between disease and demonization, and also that negative psychological or characterological traits can be tied to demonic assault, absent concerted prayer, there is no way for most believers to know for certain if a flaw is human fallenness or spiritual oppression. Which brings us to the following point: the divinization and demonization of the ecology of the will here means that akrasia is at once present when it



Prayer, Demons and Akratic Sublation

99

should not be, and absent when it should. In a strange inversion of the akratic formula, a person can remain too weak to do what he or she knows is best and still do it, or alternately, a believer can know what is right, have the necessary willpower to carry it out under normal circumstances, and yet still be incapable of the act despite it being physically possible thanks to the insertion of an alien entity that works to cloud judgment and subvert the person’s will. And what is more, these conditions can be enchained: a demonization can trigger a casting out which culminates in the demonized individual calling out to Jesus or God the Father, and hence using divine will to achieve the repudiation of the evil spirits that the individual could not reject prior to this. Likewise, prayerful individuals engaged in the practice of submitting to divine will are seen as objects of demonic affliction by evil spirits who wish to hinder God’s work. I recall one person reporting how, on the eve of a weekly prayer meeting, he and his wife were often beset by various infelicities such as misbehaving kitchen appliances, but also uncharacteristic bouts of annoyance with and petty behaviour towards one another; he read these misfortunes and minor misbehaviour as the result of demonic attacks inspired by the upcoming meeting. There is no doubt that this strange, inside-out akrasia does social work: it encourages a certain humbleness of self-presentation in seemingly devout people, and if they should disgrace themselves, it also explains their fall. And charting the eddies and rip currents of the will in this manner no doubt results in heightened levels of self-examination as one looks for indices of divine and demonic action in one’s own psyche. And over time, repeated practice does change one’s habitual sensorium.11 But to view this particular case teleologically, as a practice solely meant to catalyse the ethical self-work, would be a mistake. For most of the time, this behaviour is not a prelude to or means of self-work.12 Nor is this primarily a set of techniques for engaging in some kind of pneumatic everyday ‘ordinary’ ethics, tactically massaging the will so as to get oneself or others to act in desired ways in specific moments as one fumbles along.13 That is because all this behaviour is, to a degree, hostage to forces that are expressed in gradations of surprise: unexpected divine signs in the form of a thought, a strange buckling of the will as one falls for a temptation or suffers a disease or symptom that would not normally fell that person. Surprise is a function of that which is unexpected, even if the levels of surprise necessary and the heuristic for identifying the unexpected are capable of being shifted through extended participation in the kind of religion. So, this is more along the lines of usually quite small events that precipitate a certain kind of subject, rather than which shape a subject or are implemented by a subject. Prayer as a technique for simultaneously allowing for akrasia and removing akrasia’s sting is, in the end, more a technology for viewing the world in a certain ­way – ­or perhaps considering that these shifts in the will are, in essence, events that befall subjects, for making the world be a certain ­way – ­than prayer is a technology for acting in the world in a certain manner.

100 Jon Bialecki

This is to say that a discussion of akrasia (at least in this instance) is as much a discussion of ontology as it is ethics, because the controlling dynamics are a function of what kind of forces one’s informants assume to be operative in the world, and not solely determined by either values or virtues.14 The pseudoakrasia of prayer, deliverance and demonization may suggest that the way will, at times, buckles or overcomes, may have effects that may be usually situated in the register of ethics, but have effects far outside that domain. Jon Bialecki is Lecturer in Anthropology at the University of California, San Diego. He is the author of A Diagram for Fire: Miracles and Variation in an American Charismatic Movement, which won the 2017 Sharon Stephens Prize of the American Ethnological Society and was named a finalist for the 2018 Clifford Geertz Prize by the Society for the Anthropology of Religion. He also wrote Machines for Making Gods: Mormonism, Transhumanism, and Worlds Without End (published by Fordham Press in 2022) and is the coeditor of the book series ‘New Directions in the Anthropology of Christianity’.

Notes   1. New International Version Translation.   2. Parenthetically, it seems that we can refer, with some degree of comfort, to both 1st Corinthians and Romans as being written by Paul, and not merely being pseudepigrapha; both the consensus of historical-critical interpreters, as well as contemporary dataattribution analysis, identify the two epistles as having most likely been written by the same author. See Savoy 2019.   3. See van Den Beld 1985; Elliot 2008.   4. On the question of whether Romans 7 is properly considered an instance of akrasia, see van Den Beld 1985 and Wasserman 2007.   5. See also Luhrmann 2012 for a different anthropological depiction of Vineyard religious practice and learning.  6. Cf. Blanton 2016, who more narrowly ties the concept of prayer as technology to forms of mechanical broadcast mediation.   7. For more on group prayer, see Bialecki 2016.   8. On code-switching, see Woolard 2004. For a similar account of cognitive doubling found in Charismatic Evangelicals operating in a secular setting, see Lurhmann 2012.   9. On the role of surprise and the transformation of the will resulting from the divine sign, see Bialecki 2017. 10. Bialecki 2017, 146–165. See Csordas 1997, 165–199 for a similar discussion involving American Charismatic Catholic beliefs about demonization. 11. Luhrmann 2012a. 12. Such as that discussed in Faubion 2011 and Laidlaw 2013. 13. See, e.g. Lambek 2010. 14. Similarly, see Keane 2013, though in that work the argument is made generally, and not specific to a particular case, as is the claim being made here.



Prayer, Demons and Akratic Sublation 101

References Bialecki, Jon. 2016. ‘Diagramming the Will: Ethics and Prayer, Text, and Politics’. Ethnos, 81(4): 712–734. . 2017. A Diagram for Fire: Miracles and Variation in an American Charismatic Movement. University of California Press. Blanton, Anderson. 2016. ‘The Apparatus of Belief: Prayer, Technology, and Ritual Gesture’. Religions 7(6): 69. Csordas, Thomas. 1997. The Sacred Self: A Cultural Phenomenology of Charismatic Healing. Berkeley, CA: University of California Press. van Den Beld, A. 1985. ‘Romans 7: 14–25 and the Problem of Akrasia’. Religious Studies 21(4): 495–515. Elliot, Mark. 2008. ‘Romans 7 in the Reformation Century’. In Reformation Readings of Romans, edited by Kathy Ehrensperger and R. Ward Holder, pp. 171–88. London: T&T Clark. Faubion, James D. 2011. An Anthropology of Ethics. Cambridge: Cambridge University Press. Keane, Webb. 2013. ‘Ontologies, Anthropologists, and Ethical Life’. Hau 3(1): 186–91. Laidlaw, James. 2013. The Subject of Virtue: An Anthropology of Ethics and Freedom. Cambridge: Cambridge University Press. Lambek, Michael. 2010. ‘Towards an Ethics of the Act’, in Ordinary Ethics, edited by Michael Lambek, pp. 39–63. New York: Fordham Press. Luhrmann, T.M. 2012a. When God Talks Back: Understanding the American Evangelical Relationship with God. New York: Knopf. . 2012. ‘A Hyperreal God and Modern Belief: Toward an Anthropological Theory of Mind’. Current Anthropology 53(4): 371–395. Wasserman, Emma. 2007. ‘The Death of the Soul in Romans 7: Revisiting Paul’s Anthropology in Light of Hellenistic Moral Psychology’. Journal of Biblical Literature 126(4): 79–3816. Woolard, Kathryn. 2004. ‘Codeswitching’. In A Companion to Linguistic Anthropology, edited by Alessandro Duranti, pp. 73–94. Malden, MA: Blackwell.

5 Troubleshooting Humans Modelling the Pathways to Inertia, Backsliding and Moral Transgression on Indonesia’s Hypnotherapy Circuit Nicholas J. Long

Hypnotherapy is becoming big business in Indonesia. At a time when the ‘traditional’ medical services provided by sorcerer-healers (dukun) are increasingly derided as ‘backward’ and Islamically heterodox and hospital medicine remains unreliable and prohibitively expensive, ever more Indonesians are turning to hypnotherapists for help with their physical, psychiatric and personal problems, and ever more are training to become hypnotherapists themselves. As of December 2022, over 37,500 Indonesians were registered with the Indonesian Board of Hypnotherapy (IBH), the nation’s largest and oldest professional hypnosis association, founded in 2002. Many more have become hypnotherapists after training with other organizations, apprenticing themselves to a practising therapist, or learning about hypnosis online. These therapists hail from a wide variety of regional, ethnic and socioeconomic backgrounds, although they are most frequently male, Muslim, in their twenties or thirties, and with a prior history of work or study in the healing professions, stage magic or fields related to maths and computing. But whatever their background, they usually see themselves as vanguards: not just in promoting a secular, scientific and cosmopolitan therapeutic form that has been imported from overseas (most initially train in a model inspired by the theory and practice of the US hypnotherapist Milton Erickson), but also in the innovative ways in which they have adapted hypnotherapy to the Indonesian context. For example, Indonesia has seen an unprecedented incorporation of hypnotherapeutic procedures into ‘non-clinical’ relationships, such as those between teacher and student via hypnoteaching, or within the family home via ­hypnoparenting – p ­ ractices that, whilst not unknown elsewhere, are widely



Troubleshooting Humans 103

denounced as ‘wacky’ (Mukherjee 2016), ‘weird’ (Li 2016), or even outright ‘creepy’ (Julianelle 2016). Many therapists are also active in the field of hipnometafisika, which draws on twenty-first century understandings of psychology, healing, and Islam to understand, replicate and augment long-standing Indonesian cultural practices that were once considered ‘supernatural’. To be active on Indonesia’s hypnotherapy circuit is thus to be at the forefront of a movement grappling with complex questions regarding the nature of reality and of human experience. Foremost amongst these are questions regarding motivation, the relationship between action and judgment, and the nature and origins of practices that might be deemed ‘akratic’ – which is to say, following the definition given by McKearney and Evans (this volume), that they involve a subject willingly, consciously and freely choosing to do something that goes against their own judgment about what is, all things considered, the best thing for them to do. Many Indonesians seek out hypnotherapy because they are struggling to act in accordance with their better judgment, even though they appear free to do so, and therapists regularly encounter cases of procrastination, phobias, difficulty in resisting temptations and ‘low motivation’.1 But understanding how and why akratic (or ‘akrasia-like’ action) occurs is not only important for treating individual clients; hypnotherapists also see it as providing a vital service to the nation. There is a widespread perception in Indonesia that the country suffers from an endemic culture of corruption, opportunism and indolence. Some view this as a vestige of the nation’s autochthonous fishing cultures, others consider it a toxic legacy of repeated exposure to authoritarian and feudal rule. Either way, there is a consensus that such a ‘mindset’ (the English term is used) should be overturned. To the extent that hypnotherapy can do just that, hypnotists see themselves as at the forefront of what President Joko Widodo (2014) has termed Indonesia’s ‘Mental Revolution’: a ‘national movement’ seeking to ‘transform Indonesia’s destiny’ mind by mind. However, even as they sought to use hypnosis to overcome their clients’ akratic action, hypnotherapists were confronted with a parallel problem: that trance induction could itself precipitate behaviours that went against their clients’ better j­udgment – s­ omething they had to account for, and mitigate against. Drawing on fieldwork conducted in Java and the Riau Islands, this chapter examines the models of self that Indonesian hypnotherapists used to respond to these questions. These were models they had first encountered on the hypnosis circuit and then taught to their own students, apprentices and clients, sometimes incorporating their own elaborations or modifications.2 Very often they involved metaphors and imagery drawn from the world of cybernetics and computing. As Turkle (1984, 299) suggests, such metaphors may have an intuitive appeal, because they help us explore aspects of ourselves that we do not feel in control of. In Indonesia, their popularity was further heightened by the strong influence of neuro-linguistic programming (NLP) on the nation’s hypnosis and self-improvement circuits, a school of thought which actively

104 Nicholas J. Long

teaches therapists that the human mind is like a computer that can be ‘programmed’.3 However, the metaphors that my interlocutors worked with went well beyond the computer technology of the 1 ­ 970s – ­the decade when NLP was pioneered. Theirs were state-of-the-art visions of human as laptop, tablet or smartphone, tapping into Wi-Fi, cellular data and The Cloud in ways that not only felt contemporary but could also accommodate the Islamic metaphysical principles to which the vast majority of them subscribed. With this argument, I do not mean to paint a homogenizing portrait of Indonesia’s hypnosis scene. Hypnotherapists vary considerably in their understanding of how hypnosis works, with some (the Ericksonian symbolists) believing the ‘power of suggestion’ to derive from the way that the hypnotic subject interprets and responds to the symbolic suggestions proffered by the therapist, and others (magnetists and mesmerists) drawing inspiration from traditional Javanese metaphysics, New Age philosophy, and Franz Anton Mesmer’s theory of ‘animal magnetism’ to suggest that hypnosis is essentially physical in nature. In this latter view, an efficacious hypnotherapist is not someone who judiciously chooses the symbols that will be most effective for reshaping the client’s internal ‘mythic world’ (Dow 1986, 58–60). Instead, they use their own private reserves of ‘energy’– variously understood as an electromagnetic force field, a ‘vibrational energy’ inspired by quantum physics, or reserves of ascetically (or supernaturally) augmented spiritual p ­ otency – ­to force a suggestion into the mind of the client via speech or gesture (for further discussion, see Long 2018). However, the importance of these differences should not be overstated. The official line of the IBH is that hypnosis has both a psychological and an energetic, or ‘physiological’, component, and that therapists should balance these respective elements of their practice in accordance with their own preferences and talents. Moreover, the two schools hold relatively similar views of the selves and minds to which they are providing therapeutic assistance. This chapter outlines these Indonesian models of the self, showing how they differ subtly but importantly from those that predominate in scholarly discussions of akratic action, both in terms of the mechanisms by which they saw such actions as arising, and in terms of the normative and evaluative implications of akratic action taking place in the world. Juxtaposing the two types of models proves instructive in several regards. Firstly, my interlocutors’ self-theories call attention to possibilities and dynamics that either go unmarked within or stand at odds with mainstream philosophical traditions of thinking about akrasia. For example, the conventional notion that one’s ‘better judgment’ is straightforwardly discernible is complicated by Islamic ontological precepts which posit that experiential access to one’s very best judgment can become blocked following worldly activity. The implication is that the ‘subjective akrasia’ long debated by philosophers is only one modality of acting against one’s better judgment, and that additional varieties of akrasia (such as what I term ‘Islamic akrasia’) should be considered in cross-cultural enquiry.



Troubleshooting Humans 105

Secondly, the distinct ethical affordances of the self-theories my interlocutors use to account for a­ krasia – ­which emphasize the formative influence of social ­relations – ­can help explain why there is such an appetite for incorporating hypnotherapy into such intimate and formative spaces as the family home and the ­classroom – s­ paces in which hypnosis would rarely be used in the EuroAmerican contexts from which the hypnotherapeutic tradition originated. This point is important because it shows how different cultural responses to akrasia can have world-building force, inviting us to critically reassess the worlds that arise from the way Euro-American philosophical traditions conceptualize and respond to akratic action. Finally, the juxtaposition reveals how difficult it is to generate definitive understandings of why human beings act as they ­do – ­inviting, I suggest, an analytic outlook that does not insist upon a singular, transcendent theory of action but rather provincializes conceptions of ‘agency’ in the way that previous generations of scholars reconceptualized ‘sex’ and ‘gender’.

The Akratic Self in Philosophy, Critical Theory and Anthropology Akrasia has historically presented philosophers with two key puzzles: whether it really occurs, and how real or apparent instances of akratic action can be explained. As Charlton (1988, 3) notes, philosophers’ intuitions on these questions vary, reflecting their differing foundational conceptions of the human being. Some would hold that one ‘will always choose the course of action which will lead him in his opinion to the greatest sum of good’ (Ramsey 1931, 174) or assert that it simply ‘does not happen that a man sets himself to consider, in its totality and all its parts, an argument forbidding him to do a certain act, while at the same time he yields to a strong desire and does as the desire wishes’ (Robinson 2010, 157). In such views, akrasia is essentially impossible. Others see the phenomena of self-deception and akrasia as so self-evident that they deem them a ‘pseudo-problem’, holding that any philosopher who finds them puzzling or denies them has already made a philosophical ‘mistake’ (Lemmon 1962, 144–145; Mele 2012, 115), and must suffer, ironically, from ‘an astonishing degree of philosophic self-deception’ (Rorty 1986, 115). At stake in these debates is the integrity of a particular conception of rational agency: one associated with the folk theories of personhood that carried normative force in twentieth and twenty-first century Euro-America, which views the individual as ‘a rational, motivated agent with a protected interior core of beliefs, desires and memories’ to whom agency is ‘parcelled out’, as if it were property (Melley 2000, 10–14). Rorty (1986, 115) characterizes this vision of the self as like a modernist city, in which decisions emanate from the rational government buildings located in a grand central square and travel out to the suburbs along broad radial avenues, binding the whole city together in a systemic integrated plan. Such a self ought not to be capable of

106 Nicholas J. Long

anything as irrational as akrasia. Some philosophers thus characterize akrasia as a significant vice, and others declare it a technical impossibility. For many anthropologists and critical theorists, however, this vision of rational selfhood is not only specific to a particular place and time (see e.g. Geertz 1974; Stäheli 2013), but also descriptively inadequate even in contemporary Euro-American settings. Tackling what Brickhouse and Smith (2010, 227–228) have flagged as a paradigmatic example of akratic a­ ction – ­the subject who knows they ought to eat healthily, but actually eats ­badly – ­Berlant (2011, 116) advances the notion of ‘lateral agency’ as a counterpoint to what we might call the ‘vertical agency’ implicit in discourses of consumer-citizens as autonomous individuals with ‘sovereign’ control over their lives. In the contemporary United States, and especially amongst the poorest, who face gruelling labour conditions at work and the stress of careful budget management at home, ‘working life exhausts practical sovereignty, the exercise of the will as one faces the scene of the contingencies of survival’. Eating, and particularly the unhealthy pleasures of salt, fat and sugar, are thus ‘a form of ballast against wearing out, but also as a counter-dissipation … an activity releasing the subject into self-suspension … a fitting response to a stressful environment’ (2011, 115–116). With this argument, Berlant does not seek ‘to replace a notion of cognitive will with a notion of involuntary or unconscious activity’ but to theorize life as both a project and a ‘site of episodic intermissions from personality’. The burden of self-reproduction, the obligation to be reliable, proves such a drag that much of what we do ‘involves not being purposive but inhabiting agency differently in small vacations from the will itself ’; a process Berlant terms ‘self-interruption’ but that could also be understood as a form of akrasia. The implication of Berlant’s argument is that such akratic moments are inevitable and that visions of complete practical sovereignty and subjective super organization are mere fantasies, embraced for the affective sense of control, efficacy and security they evoke but lacking descriptive accuracy (2011, 97). Other scholars have turned to evolutionary anthropology in order to develop an account of human nature that encompasses and naturalizes akratic action, freeing it of its more pejorative associations. Writing shortly after her completion of a Master’s degree in anthropology, the philosopher Amélie Rorty (1986, 118) argues that ‘an agent capable of responding appropriately to the potentially conflicting demands of a rich and rapidly changing environment is an agent with potentially conflicting traits’. In order to co-operate with others, humans require strong empathic capacities. However, these risk internally dividing a person and so must be counterposed with strong, relatively conservative habits of behaviour and thought and critical capacities of unbounded enquiry. Consequently, a subject will frequently be in situations where ‘his preferred c­ ourse – e­ ven when it is rationally formed, and linked to some rational h ­ abits – ­will fail to draw the rest of his motivational structures with it’ (1986, 120). Far from being an aberration, akrasia is an inadvertent



Troubleshooting Humans 107

by-product of ‘the psychological mechanisms … that operate in standard rational habits’ (1986, 122). Indeed, in the anthropologist Albert Piette’s (2016) recent version of this argument, these mechanisms might even have been the key evolutionary advantage that Homo sapiens held over Neanderthals. Rejecting the image of self-as-modernist-city, Rorty (1986, 116) proposes that the self is better understood as like a medieval city (a loose confederation of relatively autonomous neighbourhoods) onto which a system of boulevards, radial avenues and central government has subsequently been imposed. From afar, that latter system might appear to be the ordering system of the city, but anyone on the ground knows that it is the interconnections between neighbourhoods that determine how the city is run and how it can most easily be navigated. Akratic behaviours may subvert expectations of rationality, but this does not render them inexplicable or impossible. They simply ‘reveal the various strata and layers of the self, its complex archaeology’. Models of the kind proffered by Rorty and Berlant are prima facie appealing to anthropologists. By recognizing the centrality of social relations, political economy, and situational context to human subjectivity and behaviour, they resonate w ­ ith – a­ nd indeed, are directly informed b ­ y – t­ enets that have long been central to anthropology as a discipline. Against these accounts, the vision of a fundamentally rational self appears to be very much an ideological product of Euro-American late modernity. But here we must exercise caution. These seemingly superior models of selfhood are no less products of history than those they seek to displace, and while they are good to think with, and can generate illuminating interpretations of ethnography, embracing them too wholeheartedly as theoretical tools risks leading us into an analytic echo chamber that distances us from the actual stakes of the action that we are using them to discuss, as experienced by the people undertaking and affected by it. With this in mind, I now turn to the theories of the akratic self that were most prominent amongst the Indonesian hypnotherapists with whom I worked, examining how and to what effect they differed from the theories I have just outlined, and what implications those differences might have.

The Spinning Beach Ball of Death, or, Subjective Akrasia Core to all Indonesian hypnotherapy, whether Ericksonian or mesmeric, is a distinction between the realms of conscious and subconscious thought. Trainee hypnotherapists usually first encounter this idea through the metaphor of an iceberg, learning that only 12 percent of mental capacity floats ‘above the surface’ and is experientially accessible. This represents ‘conscious thought’ (pikiran sadar). The remaining 88 percent is ‘subconscious thought’ (pikiran bawah sadar) and is experientially inaccessible, despite being integral to action and experience. In a computing-related metaphor used by some Indonesian hypnotherapists, conscious thought is likened to the desktop

108 Nicholas J. Long

in which one operates, while subconscious thought represents all the background code and operational processes necessary for that desktop to function as intended. Indeed, in his best-selling book Quantum Ikhlas, Indonesian self-help guru Erbe Sentanu (2015, 28) presents the subconscious as akin to a computer’s hard disk: [In a mechanical computer,] the bit of hardware that contains all software and programs is the hard disk. In a human self, the hard disk is our subconscious data store (gudang informasi); the place where information from every cell in the body is recorded.

What both metaphors capture, but which the iceberg visualizes especially effectively, is that the subconscious is the ultimate seat of human action and judgment. No matter how effortful one’s conscious strivings, these can only ever represent 12 percent of human capacity and will not translate into action unless one’s subconscious is in alignment. Here, then, we see the origins of what can be called ‘subjective akrasia’– the classic case debated by philosophers, in which one subjectively experiences oneself as not acting in the way one consciously judges to be best. Quite simply, one’s subconscious is oriented towards a different goal. However, reorienting the subconscious is not straightforward because of the existence of a critical factor (English used) which prevents thoughts and suggestions that are not compatible with the subconscious’s extant programming from lodging within it.4 There are, nevertheless, certain ‘states of consciousness’ in which the critical factor is able to be bypassed. These can arise when the subject is on the threshold between sleeping and waking, enjoying the drowsy satisfactions of having eaten a full meal, on the brink of orgasm, or intently focused on one particular object, perhaps because they have been captivated by a book, news article, or television show, or perhaps because they are in thrall to an authority figure or compelling narrator. At these times, the subconscious can be directly accessed, explored, and edited. The paradox, however, is that these are all times at which the conscious exercise of volition is also disabled. Under normal circumstances, subjects thus have limited, if any, direct control over their own subconscious mind, which becomes filled with ambient suggestions drawn from the surrounding environment. This leaves them vulnerable to various problems, including ‘programming errors’, ‘computer viruses’ and ‘software incompatibility issues’. As Sentanu (2015, 95) explains, if two incompatible programs are installed on the same computer then it is likely to crash or ‘hang’ such that either or both of the programs becomes unusable. My interlocutors in the Riau Islands, for instance, claimed that Indonesians growing up in one of the nearby Malay fishing villages might easily acquire the conviction that there is no need to inconvenience oneself by working hard to grow food for the future when one can just catch fish and enjoy the present ­moment – ­a program that would undermine any business endeavours



Troubleshooting Humans 109

that relied on forward planning. If they then tried their hands at business, they would ‘hang’, and their businesses would fail. A more critical perspective, articulated by some Southeast Asian intellectuals and politicians, might argue that, if anything was subverting Riau Islanders’ entrepreneurship, it was internalized stereotypes of Malays as lazy and ill-suited to capitalism. Either way, though, failure reflected a problem with would-be entrepreneurs’ internal programming. For Indonesian hypnotherapists, then, subjective akrasia has social origins, and the nature of consciousness means that it requires a social solution. One cannot straightforwardly amend one’s own s­ ubconsciousness – e­ specially without extensive prior training in hypnosis.5 Thus, in the face of subjective akrasia, those who are ‘motivated to have a motivation’ (Zagzebski 1996, 119) require help from an external agent. The mere exercise of willpower will not be enough. Moreover, many hypnotherapists subscribe to the New Age doctrine of ‘quantum mysticism’, which holds that all matter is energy vibrating at particular frequencies and that a universal ‘Law of Attraction’ draws entities vibrating at the same frequency toward each other (see Long 2019, 336– 337). For these hypnotherapists, even if conscious willpower could overcome subconscious dispositions, one’s strivings would not translate into concrete results. The vibrational energy of one’s subconscious ­programming – ­more powerful than conscious vibrations, by a ratio of 88 to 1 ­ 2 – ­would still have an overriding influence on the life outcomes delivered by cosmic energy flows. The only way one could improve one’s lot was thus with outside assistance. This is where hypnotherapy can help. The practice revolves around inducing a ‘trance state’ in which the subject’s critical factor can be circumnavigated and new suggestions introduced directly into the subconscious by the therapist, typically in accordance with the conscious wishes expressed by the client during a preceding consultation. They can also implant an ‘anchor’ (jangkar), linking certain suggestions to a bodily gesture that the subject can make while conscious, endowing them with more control over their subconscious influences than they had previously. Hypnotherapy thereby offers important opportunities for resolving the problem of akratic action, allowing its practitioners to contribute to the Mental Revolution in ways that they believe to be more fundamental and long-lasting than the injunctions to conscious self-cultivation and spiritual reform associated with various other political and piety movements (e.g. those documented by Hoesterey 2016; Rudnyckyj 2010). But hypnotherapy also presents risks. By relaxing their critical factors in the presence of a hypnotherapist, clients face the prospect of their therapy deliberately or inadvertently implanting subconscious suggestions that create new possibilities for akratic action. Indeed, the hypnosis circuit is rife with tales of clients who, upon entering or awakening from trance, behaved in ways that they might never have wished to in their ordinary waking state: marauding like tigers, screaming abuse at their therapists or embarking upon sexual seductions. Prospective clients, meanwhile, are often wary of hypnosis,

110 Nicholas J. Long

fearing therapists will use their skills to rob them, sexually assault them or extract secrets for use in blackmail. It has thus become imperative for hypnotherapists to account for and guard against these unwanted outcomes.

Games of the Subconscious Euro-American philosophical traditions have often struggled to grasp the full experiential complexities of action conducted ‘under hypnosis’. Philosophers of akrasia such as Mele (2006, 127) and Sultana (2006, 230) have, for example, historically excluded action under hypnosis from their discussions, claiming it is neither ‘free’ nor ‘voluntary’: ‘the agent is just as unfree as the marionette in the hands of the puppeteer’ (Tappolet 2017, 567). Similar exclusions are found in Euro-American legal traditions. The American Law Institute’s Model Penal Code (1985, 19) exempts ‘conduct during hypnosis or resulting from hypnotic suggestion’ from the category of ‘voluntary acts’ for which one can be criminally liable. Legal commentators even question whether conduct under, or following hypnosis constitutes ‘an act’ or is better described as ‘a happening’ since, in their view, the hypnotized subject does not act but is acted on (Katz 1987, 128–135). Such argumentation, however, seems more informed by the fictional archetype of the nefarious hypnotist as ­puppeteer – ­first popularized in George du Maurier’s Trilby (1998 [1894]), and still a staple of popular culture (see e.g. Jordan Peele’s 2017 film Get Out) – than by the experiences of actual hypnotic subjects. Euro-American hypnotic practitioners and research psychologists have, for their part, countered such claims with the proposition that ‘all hypnosis is self-hypnosis’: The hypnotist can recite an induction procedure and make suggestions for various experiences, but it is the subject who must actively participate in the process; without that active participation, nothing happens … The main reason that the hypnotist has so little impact on what goes on in hypnosis is, simply, that virtually all of the action is in the subject. (Kihlstrom 2008, 24)

This stance certainly proves effective at deflecting lawsuits and reassuring anxious subjects that hypnosis is safe and will not render them excessively ­vulnerable – a­ concern that led to the mantra that ‘all hypnosis is self-hypnosis’ sometimes being voiced on the Indonesian circuit too. However, like the legal and philosophical arguments it counters, it stands in thrall to a principle of originationism (the moral intuition that a person is only responsible for acts that they ­originated – ­see Kaye 2007, 370–371), leading situations full of crosscutting influences to be inappropriately purified into attributions of unilateral responsibility. Testimonies from researchers who have undergone hypnosis, by contrast, emphasize ‘the double-edged feeling most subjects of hypnosis experience’ in which ‘one part of you knows you need not go along with the



Troubleshooting Humans 111

hypnotist’s suggestions, and another part of you simply can’t be bothered to resist’, in which suggestions may prompt unexpected and out-of-character behaviours, but in which there is an enduring sense that you ‘can snap out of it if you want to’ (Epps 2010, 276; Waterfield 2002, 230–231). An appreciation of these nuances could still be used to support an originationist position. Waterfield (2002, 232–234), for instance, ultimately characterizes hypnosis as a state of lowered inhibitions, such that one’s response to suggestions is reflective of one’s ‘usual character and tendencies’. However, most of my interlocutors, following in the Ericksonian tradition (see Beels 2001, 213) insisted that hypnotherapy was fundamentally a collaborative endeavour in which hypnotist and client co-created figured worlds and new patterns of subconscious thought through their respective provision and receipt of suggestions. Establishing a successful collaboration with a client’s subconscious was a complex endeavour. The subconscious was so vast, so full of software which the client themselves may not be aware of, that even experienced hypnotherapists could not be certain how a therapeutic suggestion would interact with programs already installed. It might be co-opted by an existing program or activate programming that had previously been dormant, leading to behaviours opposite to those the therapist i­ntended – a­ ll processes that Doni,6 a senior trainer from Jakarta, referred to as ‘the games of the subconscious’ (permainan bawah sadar). For this reason, most trainers recommended that novices stick to simple techniques that reinforced the motivational programs they believed humans all had installed inside of them. Typical therapies involved enabling clients to visualize their desired goals or reconnect with past periods of ‘peak motivation’ – techniques that were less risky than trying to actively outwit a program that was holding a client back. Yet even such ‘solution-focused’ therapy could sometimes backfire, as when Zakki, a mildmannered hypnotherapist from West Java, found himself having to fend off the sexual advances of a housewife named Salma after he roused her from trance. Salma’s behaviour could easily be seen as a subconscious response to the semiology inherent in the hypnotic encounter itself: the feel of Zakki’s reassuring hand on her shoulder as she fell backwards into trance, the enactment of total submission to a fresh-faced young man plying her with a­ ffirmations – ­if these elements of the therapy had interacted with pre-existing subconscious memories, suggestions, or desires, then a strong emotional or psychosexual response seems wholly explicable. What was shocking about her actions was that they exceeded all bounds of propriety. She was married. So was Zakki. In the intensity of the therapeutic moment, Salma appeared to have forgotten herself, and to have abandoned her moral compass. Such outcomes could not be explained in terms of the ‘games of the subconscious’ alone. Instead, my interlocutors drew on religious metaphysics to make sense of them, thereby outlining a form of action that I call ‘Islamic akrasia’.

112 Nicholas J. Long

From Default Factory Settings to Porn Sites Everywhere, or, Islamic Akrasia Just as every laptop, tablet and smartphone leaves its factory in perfect condition, so my respondents insisted that the default factory setting of human beings is to be morally good. They have committed no sins and can discern (and thus avoid) sinful courses of action. This is principally due to the operations of a spiritual organ known as the ‘hati nurani’, which literally means the ‘illuminated heart’ but is commonly translated as the ‘heart of hearts’ or ‘conscience’. Sentanu (2015, 28) deems it the ‘human computer’s … operating system’. Widya, a businesswoman from the Riau Islands, described the hati nurani as a ‘large organ’ that existed within ‘the spirit’ (jiwa) and that served as conduit by which God could send divine revelation (wahyu) and inspiration (ilham) to human subjects, sounding an ‘alarm’ (alaram) if ever they contemplated a course of action that was either intrinsically impious or that would catapult them on a pathway towards inevitable sin (Long 2017, 717). From an Islamic perspective, the judgment of the hati nurani truly was one’s ‘best judgment’. However, it did not necessarily translate into the phenomenological experience of rational ‘best judgment’ that pervades philosophical discussions of akrasia. As humans grow older and are socialized into worldly habits and cultural patterns, suggestions lodge in the subconscious, directing them towards particular courses of action. These impulses, and the psychosocial affective urges associated with them, could directly occlude the ‘alarm’ of the hati nurani. Moreover, many of my interlocutors claimed that every person’s soul was ‘held’ (dipegang) by four angels and four devils, who would each give ‘whispers’ (bisikan), encouraging or discouraging particular courses of action suggested by subconscious programs: a cacophony of whispers that made the voice of the hati nurani difficult to discern (see Long 2017, 716–717). Some also believed that sorcerers and spirits had the capacity to insert their own whispers into other people’s streams of consciousness, occluding the hati nurani even further. There was thus no guarantee that one’s phenomenologically experienced ‘best judgment’ was ‘best judgment’ in a morally objective sense. Muslims thus faced the prospect of what I term ‘Islamic akrasia’ – freely acting against what their hati nurani would have told them was their best judgment, had they only been able to discern it. These challenges were compounded by the inherent fragility of the hati nurani. A Minangkabau teacher from the Riau Islands described it as being like a physical ­heart – ­though it began working perfectly, ignoring its needs by eating too much fatty food could cause it to become blocked and malfunction. Similarly, ignoring the calls of the hati nurani and living a life of sin could render it defective. As Asad (2003, 90) writes, in many Islamic traditions, one’s punishment for living impiously is ‘to become the kind of person that one is’; incapable of the forms of moral perception that might lead one onto a redemptive path, one is left unable to distinguish true speech from false, divine speech



Troubleshooting Humans 113

from human (or devilish), and thus unable to live the virtuous life that God requires. Salma’s case appears to exemplify this principle. When her hypnosis session aroused impious t­houghts – ­whether at Zakki’s direct behest, or inadvertently, through the games of the ­subconscious – ­then her hati nurani ought to have kicked in, sounding its alarm so that she refused to comply, perhaps even exiting trance altogether. That it had failed to do so suggests it was defective. This is not quite the same argument as Waterfield might make regarding Salma’s ‘usual character and tendencies’. There is no presumption in this Islamically inflected account that Salma had any underlying history of adultery, or a character disposed towards sexual forwardness. Rather, the intimation is that her previous sinful acts – whatever they may have ­been – ­had left her, and her hypnotherapists, unable to rely upon her hati nurani to stop the games of her subconscious from precipitating impious action while she was in trance. Islamic akrasia was a matter of considerable anxiety for many of the people I met during my fieldwork. The thought that Indonesia might be populated by people like Salma, who no longer knew wrong from right, raised bleak prospects for their nation’s future. And if they themselves became so disconnected from their hati nurani that their lives became a litany of sinfulness, this could have unenviable consequences in and beyond the grave. Many of my interlocutors were thus highly circumspect towards their everyday moral judgments, openly recognizing that they might be fallible. At the same time, they were heavily invested in finding strategies to avoid Islamically akratic action and ensure they were living on a relatively pious path. One such strategy was deference to the pronouncements of Islamic authorities, on the grounds that if such edicts resulted in sinful action, the fault would accrue to the person who had issued the pronouncement, rather than any Muslim who had followed it in good faith (for further discussion, see Long 2017). Complementary strategies were focused on reshaping individual consciousness so that the cries of the hati nurani could be more straightforwardly discerned. It was to these that many of the hypnotherapists I knew believed they could make an important contribution. For many hypnotherapists, the key to avoiding Islamic akrasia was to heighten a subject’s attunement to their hati nurani through intensive visualizations, in what they saw as a spiritual equivalent to ‘requesting a new password’ for an operating system from which one had been locked out (Sentanu 2015, 39). Similar ideas pervade popular ‘spiritual training’ programs, such as the Emotional and Spiritual Quotient (ESQ) interventions documented by Rudnyckyj (2010), which invite participants to reflect on their misdeeds and imagine the punishments they will receive in the grave as a means of opening their hati nurani (referred to in the programs via the Anglicized language of the ‘God Spot’) and kick-starting a reformed life of atonement and piety. However, ESQ clients frequently complained that the impact of these sessions was not long-lasting. My hypnotherapist friends were not surprised: for

114 Nicholas J. Long

an already-impious subject, the reconnection with their hati nurani would quickly be submerged by impious worldly suggestions and their associated whispers. They believed hypnosis could secure longer-lasting outcomes by installing an ‘anchor’ in the subconscious to guard against such backsliding. By using emotive imagery, typically involving the sickness or death of a mother, they invited their clients to connect with the feelings of altruism, care and unconditional love that stemmed from their primal and inaugural relationships in the world, effectively regressing the subjects to their ‘default factory setting’ in which their hati nurani reigned supreme. They then provided suggestions that encouraged their clients to link these earliest images and suggestions to mundane gestures that could be deployed during waking consciousness, allowing the wakeful subject to reconnect with the most pious reaches of the subconscious via something as simple as the squeeze of their hand. Such procedures aimed to generate ‘anchored subjects’ who could use their hati nurani to navigate all aspects of their lives with motivation and piety, and thus have more certainty that they were acting according to their very best judgment. Ansar, a therapist from Central Java’s north coast, had developed a pioneering new therapy that he described, in a typically computer-based metaphor, as a ‘total reboot’. This therapy, named zero magnetisme, combined New Age, energy-based understandings of hypnosis-as-mesmerism with Islamic metaphysics. Ansar believed that he induced trance by manipulating his own electro-magnetic field so as to make people responsive to his suggestions. Zero magnetisme involved manipulating these fields of energy to such an extent that they completely overwhelmed both therapist and client, causing each to ‘go zero’, i.e. to fall into a deep trance in which their conscious thought was rendered completely empty. Upon waking, it would not be long before subconscious programs began once again to repopulate their consciousness. However, there would be a brief period, just as they awoke, when these programs would not yet be fully operational, and when the only impulse able to inform consciousness was that of the hati nurani. Concerned about his own moral fallibility, Ansar would now ‘go zero’ whenever he had a difficult decision to make. He could thereby avoid the risk of his (or his clients’) actions being ‘akratic’ from either an Islamic or a subjective standpoint, using his magnetism to temporarily saturate his consciousness with ‘best judgment’. A final strategy was preventative and centred on installing what my interlocutors termed ‘anti-virus software’. This was described most passionately by Zukhruf, a hypnotherapist and motivator living in West Java, whose background was in nursing but who also held a fascination for hipnometafisika. Zukhruf ’s particular interest was in the ‘calling of shades’, or ngamat, a technique used in the martial art of pencak silat to access the jurus (short sequences of movements) of deceased grand masters by calling their spirits into the bodies of the living (Farrer 2009). As he came to learn more about hypnosis, and particularly the NLP practices of ‘matching and mirroring’ (in which one



Troubleshooting Humans 115

tries to synchronize thoughts and feelings with others by copying their physicality as precisely as possible) and ‘modelling excellence’ (in which one tries to achieve one’s goals by living exactly like an idolized exemplar), he began to wonder whether NLP and ngamat were different versions of the same thing. Drawing further inspiration from what his explorations into psychotherapy had taught him about the Jungian notion of a collective supra-consciousness, Zukhruf had provisionally concluded that the skills and knowledge of others were all programs that existed in something akin to the Android ­Store – ­a cloud-based repository of software that could be downloaded and installed in the subconscious at will, provided one were in a state of consciousness that allowed the critical factor to be circumnavigated. He had spent many years trying to perfect the practice of ‘downloading’ other people, which he did by focusing intently on the person whose programming he wanted to access while in a state of trance. His endeavours were driven by a deep curiosity regarding human potential, the economic allure of developing a unique product for Indonesia’s self-improvement market, and a genuine excitement for the ways in which downloading could enhance the lives of Indonesians. He himself used it not only to develop his personal skillset, but also to satisfy his curiosity regarding current affairs (by downloading Islamic terrorists, perhaps he could understand what had led them to take such destructive, impious action), to better understand his clients’ problems and to provide intelligence about criminals to the police. It was still a work in progress; adopting the metaphor of a smartphone, he explained how sometimes there was a poor signal, or the GPS was inaccurate. However, if he tried downloading someone overnight, this usually involved a prolonged enough period for him to achieve a complete download. Shortly after he began to explore these matters, Zukhruf decided to download one of Indonesia’s most celebrated hypnotists. He put himself into trance, accessed the hypnotist and fell asleep. The next morning, however, things were not as he had hoped. He struggled to put the new techniques to any effect; he felt grogi – a term derived from the English ‘groggy’, but with a semantic range that has ‘totally shifted from its [original] meaning’ (Lowenberg 1991, 134), which indexes the kind of dazed, dizzy, uncomfortable awkwardness that one might experience when suffering stage fright in front of a large crowd. Upon asking around, he discovered the hypnotist he had downloaded had principles and habits that were fundamentally incompatible with his own. Most especially, although the hypnotist was Muslim, he did not pray five times a day. Indeed, he had scolded his apprentices for breaking off their studies in order to pray. Zukhruf hadn’t been aware of these details at the time of the download, but was now convinced that they had been integral to his own precipitous decline in efficacy, lucidity and wellbeing. When he had downloaded the famous hypnotist, he had downloaded everything in that hypnotist’s subconscious, including his impious leanings. These had come to pollute a consciousness that Zukhruf retrospectively realized had been kept relatively clear and efficacious by his own pious conduct. ‘It’s like when we download a computer

116 Nicholas J. Long

program,’ he explained, ‘If there’s a virus, we might end up being shown lots of porn sites when we try to run it. It’s the same when we download a person’. David Worotikan, a hypnotherapist from Banten, elaborates further: ‘even one virus in our software can make it slower-running and prone to hang. [The virus] will try to sabotage every program, so that none of them run optimally’ (2012). Zukhruf ’s infection had created a situation in which his efficacy as an actor, and his mental ­clarity – ­including his acuity and ability to accurately apprehend and adjudicate situations, both morally and ­practically – h ­ ad become compromised. This was a condition especially conducive to action that was Islamically akratic, and indeed to action that was subjectively akratic as well. Two possible aetiologies may have underpinned it. The very fact of having installed the computer viruses could be seen as fundamentally corrupting Zukhruf ’s spiritual operating system, injuring his hati nurani and leaving him with exactly the kind of consciousness that any sinner would have, even though he himself had not committed any sins between commencing the download and waking up the next morning. Alternatively, his hati nurani could still be intact, but with his consciousness so crowded by devilish whispers associated with the new impious suggestions in his subconscious that it had become much harder for him to discern his hati nurani, or the whispers of his angels, leaving him in a compromised condition. Either way, he was now in a state where he was more likely to commit morally transgressive actions and thus damage his hati nurani himself. I have dwelt on Zukhruf ’s story at some length because it illuminates vividly how the relationship between judgment and action can become problematic in a way that is not readily captured by conventional notions of akrasia as ‘weakness of will’ or ‘moral incontinence’ (cf., e.g. Thero 2006). Zukhruf was, admittedly, in a situation where he found his ‘will’ – that is to say, following Throop (2010), his capacity to successfully undertake goal-directed effortful ­action – w ­ eakened, but only as a result of having undertaken an effortful goal-directed action the night before. Nor does it seem quite right to think of Zukhruf ’s problems in terms of his ‘character’ given the radical transformation occasioned by importing the contents of someone else’s subconscious into his own. Indeed, we might reasonably ask who or what ‘Zukhruf ’ was, given that the discourse of downloading presents him as an assemblage of programs, that, in theory, anyone could access and install. What made him distinguishably ‘Zukhruf’, then, was the combination of programs that he had installed, coupled with the unique history of how these had been cybernetically managed. In this regard, the source of moral transgression and of akratic action could be seen not as a failure of ‘will’ but a deficiency in his systems management. Zukhruf told me how his experiences had prompted several epiphanies. Firstly, he realized how important it was to keep his ‘anti-virus software’ up to date (see also Sentanu 2015, 30; Worotikan 2012). He now uses an anti-virus self-hypnosis script every night to, in his words, ‘filter’ all data he encounters:



Troubleshooting Humans 117

Whenever and wherever, with my conscious mind, body and spirit, I only permit information or data to enter into myself if it is, exclusively, information or data that is compatible with myself, that will empower myself, and that will keep myself healthy. (Emphasis in original)

He took additional measures when planning to download a terrorist or criminal. ‘I know that I’ll be exposing myself to malicious programs,’ he explained, ‘and it’s not nice afterwards. I can feel them operating in my consciousness. So, I have to make sure that my anti-virus is very strong, and afterwards I keep giving myself suggestions, reprogramming myself.’ It should be emphasized that this ‘reprogramming’, and indeed the installation of the anti-virus software, all needed to operate at the level of the subconscious to be effective. Having spent many years on the hypnosis circuit, Zukhruf was confident that he could do this unaided, via self-hypnosis. Many others, however, were less ­skilled – t­ hey would need assistance with these basic tasks of self-maintenance. Furthermore, Zukhruf ’s experiences of downloading the malicious programmes of impious trainers, terrorists and murderers had allowed him to see, through the intense contrasts precipitated by an immediate download, the slow degradation that any human being socialized in an impious environment would experience over time. In order to function in a social world, humans had no choice but to install and upgrade programs all the time, putting themselves as continual risk of viruses. Anti-virus software was not just imperative for those experimenting with downloading; it was imperative for everybody. These realizations shaped Zukhruf ’s relationship to hypnotherapy in profound ways. Having initially taken an interest in hypnosis because of its implications for pain management in his hospital, he no longer wanted to restrict his hypnotherapy to medical settings. He devoted himself to raising awareness of language, communication science and consciousness amongst the public so they could be more self-reflexive in their everyday interactions and protect themselves and future generations from the viruses in common circulation. He gave talks at local kindergartens, explaining to parents how impressionable their children were, and warning them against the dangers of using language that might inadvertently instil negative suggestions. He set up a motivational clinic for local kids in which he sought to profile each child’s unique abilities (divined via fingerprint analysis and psychological tests) and advise their parents on how to foster a linguistic environment that would work with, rather than against, those strengths and talents. And, like many advocates of hypnoparenting (e.g. Worotikan 2012), he extolled the virtue of acting on these insights regularly within the home to fortify motivations and, most crucially, to keep everyone’s anti-virus systems at peak performance. This did not necessarily need to involve deliberate trance induction, although it could. It could also involve parents using the natural rhythms of consciousness to make sure their children were furnished with enabling suggestions at times when their critical factors were most relaxed. In this way, parents, and

118 Nicholas J. Long

the hypnotherapists who advised them, would be crafting a generation better equipped than their own to overcome the perils of both subjective and Islamic akrasia. From the perspective of the self-theories that circulated amongst Indonesian hypnotherapists, there was nothing ‘weird’ or ‘creepy’ about hypnotherapy’s incursion into the intimate relations of schooling or family l­ife – ­a reaction that makes most sense in contexts where hypnosis is imagined as a complete takeover of one’s will, and where such decentring of individual control elicits a sense of ‘agency panic’ (Marlovits 2013, 143–145; Melley 2000, 12) Instead, hypnoteaching and hypnoparenting were simply more deliberate and directed versions of the social influences that shaped subjectivity on an everyday basis. Their risks could be easily mitigated provided the subjects still had a strong relationship with their hati nurani and the therapy was delivered in such a way as to circumvent potentially counterproductive ‘games of the subconscious’. Under these circumstances, hypnotherapy was not an act of control, turning a child into a mere ‘puppet’, but an act of care, reaching into subconscious recesses that children could not access themselves, instilling suggestions that could protect them from the corrosive effects of socialization and its viruses, whilst still affording them the capacity to self-determine a constructive and moral future on their own terms. It thereby exemplified the principles set out by President Joko Widodo (2014), as he ruminated on how to launch the Mental Revolution that Indonesia so urgently needed and decided that the answer was to begin ‘from each of us ourselves, starting with our family, household and workplace, and then spreading out across our cities and across the nation’.

The Akratic Self, Revisited The models of self that predominate on Indonesia’s hypnosis circuit are just that: models. Yet, if taken as philosophical interventions in their own right, they speak back in important ways to models that currently predominate in philosophical and critical theory discussions of akrasia. Most obviously, they indicate the poverty of what Rorty (1986, 121) has described as ‘the standard model of action as an outcome of beliefs and desires … that can be overridden by rational will’, instead revealing the complexity of motivation and action. This is an important point, though hardly a novel one. Instead, the more interesting contributions may come from the challenges they pose to ideas that have already been advanced as alternatives to this ‘standard model of action’. Berlant’s writings on ‘lateral agency’ have important implications for the ways in which anthropologists theorize action, drawing our attention to the importance of self-interruption within the constitution of subjectivity. Moreover, Berlant’s emphasis on energetics highlights how broader social structures such as labour relations are complicit in subverting moral



Troubleshooting Humans 119

intentions, prompting stressed, exhausted subjects to indulge in ‘spreading out’ activities, even when these go against their better judgment. As a political or ethical framework, however, this model becomes vulnerable when confronted with individual variation in the ability to withstand similar pressures. If some people can exhibit fortitude in adversity, self-indulgence, however understandable, reappears as a sign of weakness or lack. Berlant’s insights may thus be less effective than hoped in challenging the ‘heroic morality’ of ‘performing willpower’ characteristic of many contemporary models of personhood (Gooldin 2008, 289). My interlocutors would certainly not deny Berlant’s basic claim that there are energetic limits on any human’s capacity for intentional action. As Sentau (2015, 28) writes, ‘both mechanical and living computers … overheat if they are used too much’. However, hypnotherapists’ nuanced attention to the manifold circumstances in which human programming can ‘crash’, ‘hang’ or experience ‘software incompatibility issues’ would lead them to see lateral agency as not merely energetic but also psychosocially produced. They would, moreover, recognize important differences between the various lateral agencies that can prompt self-interruption. Disruptions of practical sovereignty could as easily result from a pious alarm being sounded by the hati nurani as from a psychosocial suggestion or a ­virus – ­but these would be instances of subjective akrasia with a profoundly different moral resonance. Similarly, they would highlight the importance of distinguishing between two different forms of ‘best judgment’ – that which is subjectively experienced, and that which exists in the hati nurani but which cannot necessarily be straightforwardly discerned. Their models thus highlight the value of approaching the question of action with reference to different varieties of akrasia, and invite an understanding of the self as a complex assemblage of whispers, impulses and suggestions. Such notions of self as ‘confederation’ are, of course, close to Rorty’s theory of the self as a city with a complex archaeology. But while Rorty offers a persuasive descriptive analysis of why self-deception and akrasia are not just possible but pervasive, she encounters difficulties when considering how we should normatively respond. She refuses to ‘justify or accept them’, maintaining that they are ‘potentially dangerous’, but also notes that they are ‘not by any means the worst among the psychological and intellectual vices’ and preferable to the ‘consistent unconflicted vice’ that could arise from a righteous determination to stamp them out (1986, 129). Her conclusion is that we should eschew sweeping proprieties in matters of akrasia in favour of investigating the relative advantages of specific courses of action. This seems reasonable until one realizes that such an enterprise itself demands the exercise of rational deliberation. If done prospectively, then rationality will have liberated the self from the akrasia that Rorty herself considered inevitable, and her model of the self is fundamentally undermined. If done retrospectively, we are left with the bleak prospect of life haunted by an unavoidable problem of moral luck,

120 Nicholas J. Long

in which we face the unhappy prospect of accountability for choices that we could not reasonably have made otherwise. Incorporating the hati nurani into this model might complicate Rorty’s secular intimation that subjective akrasia is always a vice, but would do nothing to challenge her portrait of action as intrinsically morally risky. Where my interlocutors might say that Rorty has gone wrong is in her decision to present the city as a historical accomplishment rather than the dynamic hub of ‘creative destruction’ and ‘accumulation by dispossession’ (Harvey 2007) that their own encounters with Indonesian planning authorities might lead them to envision. Rorty emphasizes, like my interlocutors, the constitutive nature of relationality. Indeed, it is our capacity to ‘acquire the beliefs and desires of others’ that simultaneously enables us to be social agents and sets the conditions of akrasia (1986, 124). The agent self is thus a ‘loose configuration of habits, habits of thought and perception and motivation and action, acquired at different stages, in the service of different ends’ (1986, 130). But by emphasizing habit acquisition as a historical process, in which sociality sediments into the stratigraphy of the self, Rorty underplays the self ’s capacity to be dynamically reformulated through the collaborative management of different states of consciousness and of the surrounding suggestive environment. Yet it is this possibility, encapsulated in the notion of the self as a networked piece of computer hardware that animates my interlocutors’ lives, filling them with hope for a more prosperous future in this life and the next.

Implications: Cultures of Akrasia and a Comparative Anthropology of Action Indonesian hypnotherapists’ models of the self, then, draw our attention to issues and dynamics that have been underplayed in existing discussions of akrasia. As well as showing how the relationship between action and judgment is problematized differently in Indonesia than in Euro-American philosophy, they also hold certain advantages over Euro-American models as transcendent theories of human nature. But should we as anthropologists embrace them as such? When Giovanni da Col and David Graeber advocated a ‘return of ethnographic theory’ in the inaugural issue of the since-beleaguered journal HAU, they declared that they had no intention that the publication ‘limi[t] itself to the highlands of Papua New Guinea or the forests of Amazonia’; ethnographic contexts everywhere were potential homes to exotic, alien and treacherous ‘stranger-concepts’ that needed to be ‘invited in, hosted as honoured guests’ and subsequently ‘recognized as affines, and eventually, even ancestors’ (2011, vii). Nevertheless, many anthropologists might understandably baulk at the prospect of adopting a theoretical framework inspired by traditions of selfimprovement and self-optimization that they are more used to critiquing for



Troubleshooting Humans 121

complicity with neoliberal capitalism and may even suspect to be pseudoscience. While all of the hypnotherapists I have cited in this paper seemed driven by a genuine intellectual curiosity regarding the human condition, and appeared to be sincerely advocating the models outlined above as their best understanding to date of how people inhered and acted in the world, their conclusions may nevertheless reflect their susceptibility to confirmation bias and interpretive drift (see Luhrmann 1989) – especially given the strong commercial incentives they had for believing in the dominant hypnotherapeutic model. On the other hand, there is little doubt that their intense attention to suggestion, influence, and what Blackman (2008, 23–24), following William James, highlights as ‘the problem of personality … the riddle of what makes us both “one and many”’, draws attention to aspects of personhood and action that have been underplayed in much recent social theory. Historians of the social sciences have written at length about how and why the notion of suggestibility waned in popularity over the twentieth century, one compelling argument being that it was less compatible with the operations of financial capitalism than more liberal, individualized notions of self and mind (see Stäheli 2013). In short: if Indonesian theories of hypnosis and suggestibility may partly have been shaped and maintained by economic and commercial interests, then so too might the models of personhood and mind to which they stand askance. Rather than committing ourselves to an ‘Indonesian’ – or any o ­ ther – ­theory of action, then, perhaps the conclusion that anthropologists are best advised to draw is that human action, that most mundane corollary of being in the world, remains remarkably difficult to explain. Or at least, it is impossible to explain definitively, in a way that is neutral and free from the influences and interests populating the explanatory milieu. Just as our existence within bodies that have diverse genitalia is an existential fact that demands explanation, but can be accounted for in manifold ways, so we are born ‘biologically unfinished’ (Errington 1990, 11) when it comes to our capacity to act. We have been biologically endowed with the capacity to make decisions and originate action, but also to be influenced by others and to act in ways that go against our better judgment and conscious desires. How we interpret these different aspects of personhood and make sense of the relations between them can vary considerably across different contexts, giving rise to multiple persuasive models of agency and action. Throop (2010, 34–36) has made valuable strides in this direction by highlighting how different traditions may hyper- or hypo-cognize ‘own-ness’, ‘goal directedness’ and ‘effortful-ness’ as three experiential correlates of ‘willing’. When the field of ‘action’ is expanded from that which is wilful to that which can also be akratic, an even more varied palette of experiential correlates comes into view. One’s akrasia may feel more or less one’s own, more or less conflicted, and it may reflect a compulsive engagement with the world or a confused or apathetic detachment from it. There is scope here for a fascinating

122 Nicholas J. Long

piece of ‘comparative phenomenology’, investigating whether different cultural traditions of conceptualizing action are simply different interpretations of a common set of experiences, or lead acting subjects to experience themselves in fundamentally different ways (cf. Luhrmann 2020). There is also important work to be done on the historical question of how and why human societies’ interpretations of action have come to be differentiated in the particular ways they are. In the meantime, simply documenting different cultural responses to the ambiguities of action remains a crucial anthropological task, allowing anthropologists and their readers to better understand the stakes of action in particular societies, whilst denaturalizing locally hegemonic theories of action and asking what might be gained (or, indeed, lost) if human action were to be conceptualized in alternative ways. In this regard, it is important to examine how particular theories of personhood and action offer distinct ethical affordances for akratic people as they confront their failures to act according to their best judgment, and the distinct modes of sociality to which such ‘cultures of akrasia’ can give rise. Traditions that have historically seen akrasia as a problem of the individual ultimately saddle akratic people with the burden of moral failure or, at best, moral ­luck – ­affordances that invite judgmental, atomized modes of relating, and anguished experiences of self. By contrast, Indonesian hypnotherapists used the problem as a spur to new forms of collective action, collaboration, care and mutual support. They did this, I argue, because of their conviction that the actions of the self, including those that were akratic, were fundamentally constituted (though not coerced) by relations with others. By working with the metaphors of computer viruses, hard disks and downloads, hypnotherapists and their clients were able to find hope in the idea that we are all part of the same cybernetic network. And so it is that the popularity of hypnotherapy continues to grow.

Acknowledgements Research was supported by the ESRC (ES/T013796/1) and the LSE Research Infrastructure and Investment Fund. Dr Nicholas J. Long is an Associate Professor of Anthropology at the London School of Economics and Political Science (LSE), where he works on the anthropology of Indonesia and on responses to COVID-19 in the UK and Aotearoa New Zealand. He won the 2019 Stirling Prize for Best Published Work in Psychological Anthropology for his article ‘Suggestions of Power: Searching for Efficacy in Indonesia’s Hypnosis Boom’.

Troubleshooting Humans 123



Notes   1. Clients may also seek out hypnotherapy for help with issues that are not directly linked to the relationship between action and ­judgement – ­e.g. the management of chronic pain, anaesthesia during minor surgical procedures, and the recovery of lost memories.  2. As such, the models charted in this chapter should be understood as distinctly Indonesian articulations of hypnotherapy. While there will undoubtedly be parallels with how other hypnotherapeutic traditions understand akratic action, a full analysis of such similarities and differences is beyond the scope of the present discussion.   3. NLP is a psychotherapeutic tradition combining the cybernetic theory of self advanced by anthropologist Gregory Bateson with insights from the Ericksonian tradition of ‘conversational hypnosis’ and Chomsky’s theory of ‘transformational grammar’ to provide a practical toolkit for helping people overcome roadblocks in their thought.   4. The term critical factor derives from mid-twentieth century Euro–American hypnosis discourse, which noted that the ‘critical factor of the human mind’ (Turner and Williams 1959, 40) would lead subjects to dismiss therapeutic suggestions if they were exposed to them in their normal waking state. The critical factor here referred to an aspect or dimension of thought. Some authors describe it as a Cerberus-like ‘guardian at the gate’ (e.g. Stewart and Thomas 1995, 779). However, many contemporary overviews of hypnosis, both in Indonesia and beyond, represent the critical factor as a physical boundary separating the conscious and subconscious minds.   5. Some of my most experienced interlocutors practiced self-hypnosis techniques upon themselves, but noted that this was an advanced and very challenging technique.   6. All names are pseudonyms.

References American Law Institute. 1985. Model Penal Code: Official Draft and Explanatory Notes. Philadelphia: The American Law Institute. Asad, Talal. 2003. Formations of the Secular: Christianity, Islam, Modernity. Stanford: Stanford University Press. Beels, C. Christian. 2001. A Different Story: The Rise of Narrative in Psychotherapy. Phoenix: Zeig, Tucker, and Theisen. Berlant, Lauren. 2011. Cruel Optimism. Durham: Duke University Press. Blackman, Lisa. 2008. ‘Affect, Relationality and the “Problem of Personality”’. Theory, Culture & Society 25(1): 27–51. Brickhouse, Thomas C., and Nicholas D. Smith. 2010. Socratic Moral Psychology. Cambridge: Cambridge University Press. Charlton, William. 1988. Weakness of Will. Oxford: Blackwell. da Col, Giovanni, and David Graeber. 2011. ‘Foreword: The Return of Ethnographic Theory’. Hau: Journal of Ethnographic Theory 1(1): vi–xxxv. Dow, James. 1986. ‘Universal Aspects of Symbolic Healing: A Theoretical Synthesis’. American Anthropologist 88(1): 56–69. Epps, Garrett. 2010. ‘“When You Awake You Will Feel No Remorse”: Stage Hypnotism and the Law’ In Law and Magic: A Collection of Essays, edited by Christine A Corcos, pp. 263–77. Durham: Carolina Academic Press. Errington, Shelly. 1990. ‘Recasting Sex, Gender, and Power: A Theoretical and Regional Overview’ In Power and Difference: Gender in Island Southeast Asia, edited by Jane Monnig Atkinson and Shelly Errington, pp. 1–55. Stanford: Stanford University Press.

124 Nicholas J. Long

Farrer, D.S. 2009. Shadows of the Prophet: Martial Arts and Sufi Mysticism. Dordrecht: Springer. Geertz, Clifford. 1974. ‘From the Native’s Point of View: On the Nature of Anthropological Understanding’. Bulletin of the American Academy of Arts and Sciences 28(1): 26–45. Gooldin, Sigal. 2008. ‘Being Anorexic: Hunger, Subjectivity, and Embodied Morality’. Medical Anthropology Quarterly 22(3): 274–96. Harvey, David. 2007. ‘Neoliberalism as Creative Destruction’. The Annals of the American Academy of Political and Social Science 610: 22–44. Hoesterey, James Bourk. 2016. Rebranding Islam: Piety, Prosperity, and a Self-Help Guru. Stanford: Stanford University Press. Julianelle, Mike. 2016. ‘Hypnotizing Your Kids into Submission Is Not a “Parenting Style”’. Accessed 7 August 2019. https://www.scarymommy.com/hypnoparenting/. Katz, Leo. 1987. Bad Acts and Guilty Minds: Conundrums of the Criminal Law. Chicago: University of Chicago Press. Kaye, Anders. 2007. ‘The Secret Politics of the Compatibilist Criminal Law’. Kansas Law Review 55(2): 365–427. Kihlstrom, John F. 2008. ‘The Domain of Hypnosis, Revisited’. In The Oxford Handbook of Hypnosis: Theory, Research, and Practice, edited by Michael R. Nash and Amanda J. Barnier, pp. 21–52. Oxford: Oxford University Press. Lemmon, E.J. 1962. ‘Moral Dilemmas’. Philosophical Review 71(2): 139–58. Li, David K. 2016. ‘Hypnotism is the Latest Weird Parenting Trend’. New York Post, https://nypost.com/2016/07/21/hypnotism-is-the-latest-weird-parenting-trend/ (accessed 7 August 2019). Long, Nicholas J. 2017. ‘On the Islamic Authority of the Indonesian State: Responsibility, Suspicion, and Acts of Compliance’. Journal of the Royal Anthropological Institute (N.S.) 23(4): 709–26. . 2018. ‘Suggestions of Power: Searching for Efficacy in Indonesia’s Hypnosis Boom’. Ethos 46(1): 70–94. . 2019. ‘“Accept and Utilize”: Alternative Medicine, Minimality, and Ethics in an Indonesian Healing Collective’. Medical Anthropology Quarterly 33(3): 327–44. Lowenberg, Peter H. 1991. ‘English as an Additional Language in Indonesia’. World Englishes 10(2): 127–38. Luhrmann, T.M. 1989. Persuasions of the Witch’s Craft: Ritual Magic in Contemporary England. Cambridge: Harvard University Press. . 2020. ‘Mind and Spirit: A Comparative Theory about Representation of Mind and The Experience of Spirit’. Journal of the Royal Anthropological Institute (N.S.) 26(S1): 9–27. Marlovits, John. 2013. ‘Give Me Slack: Depression, Alertness, and Laziness in Seattle’. Anthropology of Consciousness 24(2): 137–57. Mele, Alfred R. 2006. Free Will and Luck. Oxford: Oxford University Press. . 2012. Backsliding: Understanding weakness of will. Oxford: Oxford University Press. Melley, Timothy. 2000. Empire of Conspiracy: The Culture of Paranoia in Postwar America. Ithaca: Cornell University Press. Mukherjee, Meghna. 2016. ‘Wacky Trend: Hypno-Parenting’. Times of India, https://​ timesofindia.indiatimes.com/life-style/relationships/parenting/Wacky-trend-Hyp​ no-parenting/articleshow/53539861.cms (accessed 7 August 2019). Peele, Jordan (dir). 2017. Get out. Universal Pictures.



Troubleshooting Humans 125

Piette, Albert. 2016. ‘The Minimality, Contradictions, and Origin of Indifference’. HAU: Journal of Ethnographic Theory 6(1): 18–22. Ramsey, Frank Plumpton. 1931. The Foundations of Mathematics and Other Logical Essays. London: Routledge. Robinson, Richard. 2010. ‘Aristotle on Akrasia (book vii.1–11)’, in Aristotle’s ‘Nicomachean Ethics’, edited by Otfried Höffe, pp. 149–66. Leiden: Brill. Rorty, Amélie Oksenberg. 1986. ‘Self-Deception, Akrasia and Irrationality’. In The Multiple Self, edited by Jon Elster, pp. 115–31. Cambridge: Cambridge University Press. Rudnyckyj, Daromir. 2010. Spiritual Economies: Islam, Globalization, and the Afterlife of Development. Ithaca, NY: Cornell University Press. Sentanu, Erbe. 2015. Quantum ikhlas: Teknologi aktivasi kekuatan hati. Updated edition. Jakarta: PT Elex Media Komputindo. Stäheli, Urs. 2013. Spectacular Speculation: Thrills, the Economy and Popular Discourse. Eric Savoth, trans. Stanford: Stanford University Press. Stewart, A.C., and S.E. Thomas. 1995. ‘Hypnotherapy as a Treatment for Atopic Dermatitis in Adults and Children’. British Journal of Dermatology 132(5): 778–83. Sultana, Mark. 2006. Self-Deception and Akrasia: A Comparative Conceptual Analysis. Rome: Editrice Pontificia Università Gregoriana. Tappolet, Christine. 2017. ‘Self-Control and Akrasia’. In The Routledge Companion to Free Will, edited by Kevin Timpe, Meghan Griffith and Neil Levy, pp. 565–76. New York: Routledge. Thero, Daniel P. 2006. Understanding Moral Weakness. Leiden: Brill. Throop, C. Jason. 2010. ‘In the Midst of Action’. In Toward an Anthropology of the Will, edited by Keith M. Murphy and C. Jason Throop, pp. 28–49. Stanford: Stanford University Press. Turkle, Sherry. 1984. The Second Self: Computers and the Human Spirit. London: Granada. Turner, Dudley W., and Alexander S. Williams, III. 1959. ‘Medical Hypnosis’. Journal of the National Medical Association 51(1): 40–2. Waterfield, Robin. 2002. Hidden Depths: The Story of Hypnosis. Basingstoke: Pan Books. Widodo, Joko. 2014. Revolusi mental. Kompas, Available online at http://nasional.kom​ pas.com/read/2014/05/10/1603015/Revolusi.Mental (accessed 23 February 2015). Worotikan, David. 2012. ‘Mendidik anak dengan hypnoparenting’. http://bantenhip​ notiscenter.blogspot.com/2012/04/mendidik-anak-dengan-hypnoparenting.html (accessed 30 June 2019). Zagzebski, Linda Trinkaus. 1996. Virtues of the Mind: An Inquiry into the Nature of Virtue and the Ethical Foundations of Knowledge. Cambridge: Cambridge University Press.

6 The ‘Replication’ of Caste as a Form of Collective Akrasia Ivan Deschenaux

Replication Dalits are among South Asia’s most ostracized people. The low position ascribed to them in the traditional caste hierarchy continues to afflict them in numerous ways.1 Macroeconomic indicators, for example, show that they face severe socioeconomic disadvantages throughout the subcontinent. This holds true not only in India, but also Nepal, the country which I focus on here (CBS 2005, 2011; WB–DFID 2006; Bennett, Sijapati and Thapa 2013).2 As my own and other ethnographers’ observations indicate, Dalits also continue to be treated as ‘inferior’ in everyday life and interpersonal relations. In Sammagaon,3 the village in the Himalayan foothills of Eastern Nepal where I conducted 18 months of ethnographic fieldwork from 2013 to 2015, this happens in at least two ways. The first is a set of stereotyping practices which portray Dalits as ‘dirty’, ‘stupid’, ‘inept’, ‘rude’, ‘dishonest’, ‘drunk’, ‘irresponsible’, ‘boisterous’, etc. The second is a regimenting of space and interpersonal relations that falls short of full-blown segregation but nevertheless clearly denotes the lower status of Dalits and harks back to their ‘impure’, ‘polluting’ or ‘defiling’ character in traditional caste ideology. This regimenting of space and relations, which I will refer to as ‘untouchability’ or ‘purity practices’ from here on, comprises several elements. Among these, three are particularly prominent. The first is an avoidance of commensality: non-Dalits will not eat food cooked by a Dalit, as this food is thought to carry the Dalit’s impurity and have the potential to pollute the non-Dalit.4 The second is a regulation of space: a Dalit never enters a non-Dalit house, lest he or she defile that space.5 The third is a very strict avoidance of intermarriage: Dalits and non-Dalits do not intermarry. Violations of this rule, which



The ‘Replication’ of Caste as a Form of Collective Akrasia 127

are extremely rare, may lead to the non-Dalit’s family disowning him or her entirely.6 There is, therefore, in the hills of Eastern Nepal a pronounced divide between Dalits and non-Dalits, ­marked – ­albeit less visibly than in the ­past – b ­ y ongoing stereotyping and discriminatory practices. Nathaniel Roberts, in his ethnography of a slum in Chennai, in the South East of India, describes the divide between Dalits and others as ‘qualitatively distinct’ from the more mundane ways in which non-Dalit castes are distinguished from each other (2016, 3). This holds true in Nepal. What is singular about the separation between Dalits and others is a sense of fundamental, unbridgeable inequality. Dalits are treated as inherently different, lowly or impure, rather than as a banally distinct group. But the category ‘Dalit’ itself is not homogeneous; it contains further divisions which run, arguably, as deep as that which separates Dalits and nonDalits. In Sammagaon, the specific people among whom I lived and conducted ethnography are members of the ‘Bishwakarma’ caste. They are Dalits and, as such, find themselves on the receiving end of stereotyping and untouchability practices daily. Where things become more complicated is in the relations which the Bishwakarma maintain with other Dalits. In Sammagaon, the issue is most evident in their interactions with the Pariyar, another Dalit caste. Despite denouncing caste-based discrimination as a nefarious phenomenon, the Bishwakarma treat the Pariyar in the very same way as non-Dalits treat Dalits in general. Non-Dalits will avoid food cooked by all Dalits (including the Bishwakarma), prevent Dalits from entering their houses and strictly avoid marrying them. In the exact same way, the Bishwakarma will avoid food cooked by the Pariyar, prevent Pariyar people from entering Bishwakarma houses and strictly avoid intermarrying with the Pariyar. The Bishwakarma, thus, are simultaneously on the ‘receiving’ and ‘giving’ ends of untouchability. Similar forms of intra-Dalit discrimination occur throughout South Asia. Classic anthropological literature calls the phenomenon ‘replication’ – the replication among Dalits, that is, of the macroscopic division that separates all Dalits from others. My goal in this chapter is analytical. I want to see what insights there are to gain by bringing together the anthropological idea of replication and a body of philosophical literature discussing weakness of the will and akrasia, or ‘acting against one’s better judgement’. I do not argue that the Bishwakarma are simply ‘weak-willed’ for reproducing the kind of caste-based discrimination which they verbally denounce, but rather show why such a simplistic interpretation is untenable. Making this point has value on its own, because it is not uncommon in Nepal to hear non-Dalits accuse the Bishwakarma of inconsistency. Such accusations vary in sophistication, but they revolve around the idea that the Bishwakarma do not deserve a treatment that they are incapable of extending to others. To counter this accusation, I show that there is more to the Bishwakarma’s behaviour than a simple, straightforward failure to adhere to ethical principles they demand others obey.

128 Ivan Deschenaux

But I have two more analytical aims in bringing together replication and akrasia. The first is to see whether we can gain new insights into replication itself. Anthropology has produced several explanations for this phenomenon, which I will briefly review in the next section. These explanations have various shortcomings. I want to see if we can understand the phenomenon in new ways by thinking about it against the background of literature on weakness of the will and akrasia. The second aim is to explore whether, through such a reflection, there is anything to learn about akrasia and weakness of the will. Thus, I am interested both in how the philosophy can inform the ethnography, and how the ethnography can inform the philosophy.

Existing Explanations of Replication Existing literature has primarily debated whether replication indicates that Dalits live in consensus with the social system that causes their oppression. Earlier scholarship on caste, to a much greater extent than current anthropological literature on the topic, viewed caste as an overarching, integrated system of social organization (Gupta 2005; Das 2015). Louis Dumont, in a highly influential monograph ([1966] 1980), argued that Indian society is fundamentally hierarchical and organized around core notions of ritual purity and pollution. He further argued that Indians agree with this form of social organization regardless of their position in the social hierarchy. Drawing on this view, Marston Michael Moffatt (1979) wrote the first detailed analysis of replication. Although it is based on research conducted several decades ago in South India, his book is a necessary starting point because it set the terms for the debate that followed. Later anthropologists working with Dalits in various parts of South Asia, including Nepal, have engaged with his claims. While many have rejected them, the debate cannot be understood without briefly attending to Moffatt’s views. Observing not only that some Dalits treated other Dalits as untouchable, but also that their social, ritual and religious practices closely resembled those of Brahmins (so-called ‘upper-castes’), Moffatt concluded that Dalits did not have their own, distinct culture.7 In line with the Dumontian perspective, he claimed that replication was an indication of Dalits’ ‘cultural consensus’ with Brahminical views and with the moral and social order of caste more generally: The cultural system of Indian Untouchables [Dalits] does not distinctively question or revalue the dominant social order. Rather, it continuously recreates among Untouchables a microcosm of the larger system. (Moffatt 1979, 3, emphasis added)

In this chapter, I am not concerned with the same breadth of practices as Moffatt and the authors who wrote about replication after him. I do not look, for example, at Dalits’ ritual practices, and I am not concerned with the extent



The ‘Replication’ of Caste as a Form of Collective Akrasia 129

to which these resemble Brahminical ones. The convergence of Dalit and Brahminical cultural markers is a topic which has complexities of its own in Nepal and is best treated separately.8 I examine ‘replication’ in my own ethnography more narrowly, in relation only to the way in which the Bishwakarma treat other Dalits as untouchable.9 Using this narrower definition, we could try to transpose Moffatt’s argument to present-day Nepal. It would run as follows. Nepali Dalits are treated as lowly and polluting by non-Dalits. Nepali Dalits are divided into Bishwakarma and Pariyar. The Bishwakarma treat the Pariyar as lowly and polluting, in the same way as non-Dalits treat Dalits as a whole. From this, we could infer (following Moffatt’s reasoning) that the Bishwakarma are in consensus with notions of hierarchy and purity. They view society as hierarchically organized, as evident in the fact that they situate the Pariyar ‘below’ themselves. Despite their verbal protestations against caste ideology, they believe that people of different castes are more or less ‘pure’, as evident in the fact that they treat the Pariyar as untouchable.10 This line of argumentation leads to a contentious conclusion. It implies that the Bishwakarma do not oppose the fundamental ­tenets – ­hierarchy and ­purity – ­of a social system which profoundly disadvantages them. In Moffatt’s own words, replication ‘indicates in a most forceful manner the degree to which [Dalits] are in consensus with a system that defines them as fundamentally low’ (Moffatt cited in Deliège 1992, 157). It is true that not all Dalits systematically oppose everything related to caste. This much is unsurprising, given the cultural and historical context which they live in. In Nepal, caste was a dominant social category for much of the country’s history. It used to be powerfully sanctioned by the State (Höfer [1979] 2004; Gellner 2007), and while this is no longer officially the case, caste continues to be important in many spheres of life. We should not be surprised, then, to see some Dalits manifesting some signs of agreement with a castebased social organization, in some contexts. But the proposition that Dalits live in consensus with caste is empirically untenable. By and large Dalits do oppose caste and ascriptions of inferiority. Countless ethnographies have shown the many ways in which this is the case, and I observed the same in Nepal. My Bishwakarma interlocutors in Sammagaon often explained to me that caste-based discrimination is morally wrong. They defended instead what I have come to think of as an ‘ethic of castelessness’, a normative view according to which caste and caste-based discrimination should not exist. Primarily, they defended this by denouncing and decrying any instance of caste-based discrimination which they heard about, to me and among themselves. In addition, they commonly used three idioms when explaining to me why a social organisation based on caste is wrong.11 First, they often stated that all humans have the same blood, and the everybody’s blood is the same colour, red. This was a way of appealing to a unique, shared humanity across caste, a way of countering any notion that there is something essentially different about

130 Ivan Deschenaux

people of different castes. Second, they told me that caste was something that had been made by humans rather than the gods. They explained that humans had created caste to exercise power over other humans. I interpret this as a statement that caste is ontologically ‘shallow’: it is a politically motivated phenomenon that could cease to exist, not an ontologically deep category of divine origin.12 Third, they often used the expression, ‘there are only two kinds of people, men and women’. The emphasis was always on the ‘only two kinds’. In context, this was not so much a way of emphasizing a separation between the ­sexes – s­ omething which my interlocutors took for g­ ranted – ­as a way of attracting attention to the lack of further division. The point was that caste is not like sex, and that while sex is a real division, caste is not, or at least should not be. Moffatt’s assertion that Dalits consent to notions of hierarchy and purity is, therefore, untenable in contemporary Nepal. This leaves us with a problem because replication remains an empirically observable reality. We need an alternative explanation. Given that the Bishwakarma are vocally opposed to caste-based discrimination and the idea of caste itself, how can we explain the fact that they treat the Pariyar as untouchable? One answer requires a radical re-interpretation of replication. This is the approach followed by Karanth, who suggests that ‘replication can also be seen as a challenge to the dominant social order and as an effort on the part of the Untouchable castes to pursue that from which they are excluded’ (Karanth 2004, 155). In this perspective, when Dalits behave similarly to Brahmins, this is not a passive act of consent. Instead, it is an active challenge to the social order, a claim that they too can, legitimately, engage in Brahminical practices. This argument may hold for the aspects of replication which are not the focus of this chapter, i.e. the adoption by Dalits of Brahminical ritual practices. It does not, however, help us square the tension which is my central focus: between the Bishwakarma’s explicit ethic of castelessness and their behaviour towards the Pariyar. Another explanation for replication is that Dalits’ rejection of caste is limited or incomplete. They endorse an ethic of castelessness while still partly accepting ideas of impurity and hierarchy. Deliège adopts this line in an article re-evaluating Moffatt’s work: There is indeed a degree of agreement between high castes and [Dalits] about the foundations of the system, but … it by no means follows that the agreement is as all-pervasive or as general as the idea of ‘consensus’ seems to suggest. (Deliège 1992, 171)

What this helps us recognize is that, contra Moffatt, treating others in one way does not necessarily imply consenting to being treated in the same way. The fact, for example, that my Bishwakarma interlocutors refuse to eat food cooked by a Pariyar does not mean that they like it when non-Dalits avoid food which they, the Bishwakarma, cook. They may well comply, but this does



The ‘Replication’ of Caste as a Form of Collective Akrasia 131

not mean they consent. For example, when they organize a wedding and invite non-Dalit guests, the Bishwakarma in Sammagaon will ensure that there are two kitchens. The cooks in one are Bishwakarma; they prepare food for guests of the same caste. The cooks in the other are hired non-Dalits who cook for the non-Dalit guests. When I asked my interlocutors why they do this, they explained that it was expected of them, that non-Dalit guests might leave or cause a scandal if things were done differently. But they resented the fact that this was expected of them in the first place and wished they could have only one kitchen. Deliège also helps us recognize a degree of contextual variation in the Bishwakarma’s attitudes. The same Bishwakarma individuals who deplored having to organize multiple kitchens at their weddings would often balk at the idea of eating food cooked by a Pariyar. This kind of variation is unsurprising, according to Deliège, because we should not presume that ­Dalits – ­or anybody else, for that m ­ atter – h ­ ave a single, unified worldview. We should not expect them to be consistent. It is worth quoting in full the concluding passage of his article as it gives a sense of how fundamentally incoherent Deliège thinks humans are: It is wrong to assume that humans are always rational, coherent and unambiguous beings who subscribe totally and exclusively to a single ideology. The world is rather made up of people who are always complex, sometimes contradictory. One can be at the same time Christian and communist, democrat and elitist, pacifist and violent. The only communist militant of Alangkulam is also the pusari (priest) of the Muni temple and a member of a high, oppressive caste. These elements are theoretically incompatible but this does not bother Kandaswamy, who assumes his own contradictions as everyone of us does. This is also true of a social group like the untouchables who are pulled between two different, if not opposed, ideologies: they are ‘those who are forced to be down’ or even militant dalits, but at the same time they also reproduce the system to maintain those lower to them in a degrading position. (Deliège 1992, 171–172, emphasis added)

Deliège is successful in debunking Moffatt’s ascription of consensus to Dalits. And he may well be right that humans are ‘always complex’ and ‘sometimes contradictory’ (for a more recent contribution emphasizing this point, see Berliner et al. 2016 and other articles in the same special issue). But he ultimately leaves us with the impression that Dalits do not even see a tension between their explicitly endorsed anti-caste principles and their casteist behaviour towards other Dalits; or that, if they do perceive it, they are, like Kandaswamy in the quotation above, ‘not bothered’ about these merely ‘theoretical incompatibilities’. This is not true for many of the Bishwakarma in Sammagaon. As they told me directly, they are aware that their behaviour towards the Pariyar is at odds with the anti-caste views they espouse. And there are signs, also, that this tension sits uneasy with them.

132 Ivan Deschenaux

The Bishwakarma display a sense of unease or compunction when treating the Pariyar as untouchable. Often, for example, a Bishwakarma individual will try to dissimulate that caste is driving their behaviour towards a Pariyar. This is especially true when the Pariyar individual has gained a respectable status through, for example, class mobility. Thus, when turning down food, the Bishwakarma individual might say in an apologetic tone that they have just eaten or are not hungry, that they have a stomach ache, etc. Similarly, a Bishwakarma wishing to prevent a Pariyar from entering his or her house might say something like, ‘why don’t we sit here in the sun, where it is warm?’ Such excuses are used to prevent the embarrassment that would result from saying out loud what everybody involved ­knows – ­that the real reason for avoiding the food or sitting outside comes from purity practices.13 These are subtle clues that the Bishwakarma sense that something is amiss in their treatment of the Pariyar. Similarly, whenever my host and adoptive ‘father’ in Nepal, a Bishwakarma man called Ganga, observed or participated in an instance of replication, he would come to me and explain that this kind of thing continued to happen out of habit (bānī), not because the Bishwakarma actually think the Pariyar are untouchable. The implication, which he sometimes spelled out explicitly, was that although replication was an undesirable habit, it was hard to get rid of. For instance, Ganga had long been involved in local politics and various forms of Dalit activism, and his ongoing participation in public life meant that he regularly interacted with members of other castes and ethnic groups, including one of the senior Pariyar people of Sammagaon. On occasion, the latter would come to Ganga’s house. When this happened, the Pariyar man would not be invited indoors, unlike many other guests. After this occurred, Ganga would come to me and explain, in a somewhat embarrassed tone, that while yes, the reason for not asking this man indoors was his caste, doing so was ‘only’ an ‘old habit’ (purānō bānī) and he did not actually think the Pariyar man could defile or pollute him. Not all the Bishwakarma in Sammagaon were as candid as Ganga, but in conversation with me, many expressed, in one way or another, unease about their behaviour towards the Pariyar. Of course, one possibility is that they were not being honest. My interlocutors may have adjusted their statements according to what they thought I would find agreeable. This is a risk inherent in the ethnographic method, but I can offer two points of reassurance. The first is that statements of this kind emerged with minimal prompting on my part. The second is that, as I spent more time in Sammagaon and as I became more familiar with my hosts, I encountered them more and more regularly. Together, these facts lead me to believe that there is a widespread sense among the Bishwakarma that their behaviour does not accord with their endorsed ethical principles and that this is something which ought, in some sense, to change. In sum, none of the explanations of replication which I have reviewed here adequately capture my ethnographic observations. The proposition that the



The ‘Replication’ of Caste as a Form of Collective Akrasia 133

Bishwakarma are in ‘consensus’ with caste ideology (Moffatt) is empirically untenable. The notion that they replicate to dissent with caste (Karanth) does not explain their behaviour towards the Pariyar. And while it is true that they are incoherent (Deliège), there is an awareness of, and embarrassment at, this incoherence, which needs to be accounted for.

Is Replication Akratic? Can replication be understood as a case of weakness of the will or akrasia? Although akrasia and weakness of the will are sometimes used synonymously, contemporary philosophical literature draws a distinction between them. Following Holton (1999), weakness of will is understood primarily as a discrepancy between a subject’s resolutions and behaviour. I am weak-willed when, for example, I resolve to finish writing an academic paper by a certain date, but then proceed to do something more enjoyable and fail to complete the paper on time. In this sense, weakness of the will is a propensity to violate previously formed intentions. In contrast, following Davidson ([1970] 2001), akrasia is usually understood as a discrepancy between judgement and action. It occurs when a subject acts against what they judge, in the moment, to be the best course of action. It would be akratic of me not to finish writing an academic paper if, when I am sitting down at my desk, I judge that finishing the paper would be the best course of action.14 Given that Dalits engage in replication while espousing anti-casteist principles, could we understand their behaviour as either an instance of weakness of the will or akrasia? The Bishwakarma say that they should act differently but manifestly find it hard to do so. It is regularly the case, in Sammagaon, that a Bishwakarma individual intentionally treats a Pariyar as untouchable despite experiencing unease as they do so, and despite recognizing, upon reflection, that they ought not to, according to the ethic of castelessness, which they explicitly subscribe to. So, are the Bishwakarma weak-willed? Are they acting against their own better judgement?

Is Replication Non-Akratic Weakness of the Will? As a first analytical step, it should appear clear that replication cannot be described as an instance of non-akratic weakness of the will. As mentioned above, on the standard account (Holton 1999), weakness of the will is a propensity to violate previously formed intentions. More precisely, it occurs when a person has resolved to act, in the future, in a way that they might be inclined against, but then, when the time comes, the person does not act according to their intention. Instead, they give in to a more immediate inclination. Weakness of the will thus conceived is one kind of irresoluteness. It is

134 Ivan Deschenaux

a diachronic phenomenon involving a conflict between a person’s long-term and short-term judgements (see also Henden 2016, 119 for a comparison of akratic and non-akratic weakness of the will). The reason replication does not fit this description is, quite simply, that none of the Bishwakarma individuals I know in Sammagaon appear to form intentions to stop treating the Pariyar as untouchable. Certainly, no one ever told me of such intentions, and I did not observe any of them trying to treat the Pariyar differently. I cannot exclude that some of the Bishwakarma in Sammagaon formed such intentions without changing their behaviour in any observable way. Still, having spent eighteen months in the field, if such intentions had at all been common, I suspect I would have observed some manifestation of them, verbal or behavioural. Why not? If many Bishwakarma judge that they ought not to treat the Pariyar as untouchable, why don’t they form intentions to stop doing so? One possibility is that the ‘habit’ of treating the Pariyar as untouchable, described by Ganga, is simply too forceful. But there is a simpler explanation. There are factors which would make such intentions excessively hard to live up to, as will become clear shortly.

Is Replication Akrasia? Having rejected the idea that replication is a case of weakness of will, I now want to examine whether replication can be understood as a case of akrasia, a discrepancy between judgement and action. Davidson’s classic definition of akratic action is that: In doing b an agent acts incontinently [akratically] if and only if: (a) the agent does b intentionally; (b) the agent believes there is an alternative action a open to him; and (c) the agent judges that, all things considered, it would be better to do a than to do b. (Davidson 1970, cited in Stroud 2014)

At first glance, it may appear that the instances of replication described earlier fit this characterization. But what the Bishwakarma themselves say shows that this is not the case. When asked about the way in which they treat the Pariyar, besides invoking habit, the Bishwakarma offer a second explanation. They state that their behaviour is due to the judgement of others. This explanation is even more common than the previous one and, as a matter of fact, a version of it is also used by non-Dalits to explain why they treat Dalits as untouchable (but I will not discuss this more general case here). Thus, a Bishwakarma individual will commonly say that the reason they treat the Pariyar as untouchable is not that they themselves believe that the Pariyar are impure or defiling. Rather, it is because of what other Bishwakarma people would think if they behaved any differently. Thus, the real risk one takes by letting a Pariyar into one’s house or eating food cooked by a Pariyar is not exposing oneself to pollution. The real danger is that one might become the subject of gossiping and



The ‘Replication’ of Caste as a Form of Collective Akrasia 135

backbiting, and in the extreme, that one might be shunned or ostracized by one’s own kinsfolk and community. The first time I encountered this type of explanation, it reminded me of a sentence I heard all too often while growing up in a remote Swiss village: ‘I’m not racist, but …’. A racist comment would inevitably follow. What the two cases had in common, I surmised, was a denial of one’s individual responsibility in maintaining a social phenomenon widely perceived as undesirable. Later in my fieldwork, as I discovered how people who enter intercaste marriages are treated by their families, I started to re-evaluate. Marrying somebody who is traditionally seem as ‘untouchable’ to one’s own caste carries a very real risk of being disowned. To be certain, this is more extreme than letting the ‘wrong’ person into one’s house or eating the ‘wrong’ food, but even these comparatively mundane actions can still lead to shaming by one’s relatives and community. My initial assumptions were therefore misguided. It was not simply that my interlocutors were deflecting and denying their role in perpetuating castebased discrimination. The shaming structure described in the previous paragraph is, arguably, at the core of the caste system. Taking the risk of shaming into account, it becomes easy to see why a Bishwakarma individual who would genuinely prefer not to treat the Pariyar as untouchable might do so anyway: the alternative course of action would ruin their standing and reputation within their own community. This, in turn, is reason enough to reject the suggestion that replication is a matter of individual akrasia, as defined by Davidson. Take the example of a Bishwakarma person preventing a Pariyar from entering their house. In many cases, the two first conditions in Davidson’s definition will be met: (a) the Bishwakarma individual will intentionally prevent the Pariyar from entering and (b) the Bishwakarma will believe that there is an alternative course of action, viz. allowing the Pariyar to enter, which is available to them.15 Condition (c), however, will often not be met. While the Bishwakarma person in this example might judge that it would be better, in a moral or ethical sense, to let the Pariyar in, they will not believe that this is the best course of action, all things considered. To put things differently, in light of the ethic of castelessness only, the Bishwakarma person should indeed choose to let the Pariyar in, and failing to do so may seem akratic. In light of the further knowledge that letting the Pariyar in comes at a strong cost to their own reputation and with the risk of being ostracized by their kin, it appears unreasonable to do so. Thus, it is not i­rrational – l­et alone a­ kratic – f­ or any individual Bishwakarma to treat a Pariyar as untouchable, even if the Bishwakarma individual subscribes to the ethic of castelessness. Furthermore, to return to a point made in the previous section, it would be very strange for a Bishwakarma individual who fears shaming to form intentions to let the Pariyar in.

136 Ivan Deschenaux

Is Replication a Case of ‘Collective’ Akrasia? So how do we explain the fact that people who are subjected to highly discriminatory practices and explicitly oppose these nevertheless subject others to the very same? In particular, if the Bishwakarma’s behaviour is neither weak-willed nor akratic in the usual sense, how else can we capture the ambivalence inherent in their behaviour, their unease when they treat the Pariyar as untouchable, their realization that doing so goes against ethical norms which they otherwise endorse? If we are to do justice to this ethnographic observation, our analysis will have to account for the fact that (at least some) Bishwakarma individuals feel that they should behave differently as a people, even when they have good reason not to do so individually. Might developing a concept of collective akrasia be helpful, here? Most contributions describe akrasia exclusively as an individual phenomenon, but there are some exceptions. Pettit (2003), for example, argues that there are conditions in which a group may be capable of akrasia.16 His strategy, by and large, is to identify what kind of group is sufficiently like an ­individual – ­that is, what kind of group sufficiently resembles the model of a rationally unified ­agent – ­for the concept of akrasia to carry over without issue. In particular, he is concerned with the way in which such a group would make and carry out decisions. He starts by eliminating groups which are insufficiently unified in their decision-making process, declaring the following incapable of akrasia: • mere collections of individuals that happen to display similar intentions at a given point in time; • cooperatives that reach collective decisions by voting but have no mechanism in place to avoid rationally inconsistent results from one decision to the next; and • unified cooperatives that reach collective decisions by voting and happen to produce rationally consistent results from one decision to the next, but do so without any deliberate mechanism or process ensuring consistency.17 Only what he calls self-unifying cooperatives fit the bill. These are groups that reach collective decisions by voting and explicitly look out for rational inconsistency from one decision to the next. To decide whether this can help us understand what is happening in the case of the Bishwakarma, we must dive deeper into Pettit’s argument. To show how a self-unifying cooperative can display akrasia, Pettit considers the example of a three-member editorial board of an academic journal (Pettit 2003, 82ff.). Say the members of such a board vote on several issues over a given period. They initially decide, through majority voting, to freeze the price of the journal for subscribers. Subsequently, they decide, through majority voting again, that when determining whether to publish a given piece, they will be bound by the



The ‘Replication’ of Caste as a Form of Collective Akrasia 137

recommendations of external reviewers. Subsequently still, the question arises as to whether the journal should publish highly technical papers which are costly to produce. Agreeing to this would be inconsistent with the two prior decisions, Pettit states, because it would be impossible to guarantee a price freeze while simultaneously agreeing to publish expensive technical papers that external reviewers are likely endorse without any regard for cost. In this situation, Pettit shows that a ‘discursive dilemma’ might arise: the group may, through majority voting, agree to all three propositions without any single member voting yes to all three.18 If the group is ‘self-unifying’ – that is, if it checks for consistency from one decision to the n ­ ext – i­t should recognize that these three decisions are incompatible. Yet, recognizing this issue might not, on its own, be sufficient for any of the group members to change their (individually rational) voting pattern. The group as a whole, then, may display collective akrasia even when no individual member is akratic or irrational. The Bishwakarma, as an entire people, do not even begin to resemble a self-unifying cooperative. They are a large and dispersed group comprising some 1.2 million individuals, living in all parts of Nepal. But what of the Bishwakarma of Sammagaon, the village in which I conducted my fieldwork? In this case, we are talking about a small number of households that live close to each other, belong to the same patriline and, while not strictly a corporate group, nevertheless interact on a regular basis. This is certainly closer to the kind of group Pettit has in mind, but still does not fit the bill, because the Bishwakarma of Sammagaon use no formal process to make decisions. They do not cast votes on how to behave and they do not critically examine the consistency of their common decisions. Therefore, even this small group does not satisfy Pettit’s requirements for a self-unifying cooperative. Does this mean there is no insight to gain here? Perhaps we ought not to give up so quickly. If we follow a different strategy than Pettit’s to define collective akrasia, we will see that the concept may meaningfully apply to the Bishwakarma. Thus, I propose we define collective akrasia by transforming Davidson’s definition so that it no longer pertains to individual action and judgement, but to their collective counterparts instead:19 In doing b, a group displays collective akrasia if and only if: (a) members of the group do b intentionally; (b) members of the group believe there is an alternative action a open to them; and (c) members of the group judge that, all things considered, it would be better for all members of the group to do a than to do b.

Replication fits this definition. For the Bishwakarma, b might be any instance of treating the Pariyar as untouchable (not letting a Pariyar into their house, avoiding food cooked by a Pariyar, etc), whereas a would be not treating them as such (letting them in, eating the food, etc.) For the sake of clarity, I have omitted in the definition above some qualifiers that would, strictly speaking, be necessary. Say, for instance, that a single member of a group does not believe that course of action a is better. If the

138 Ivan Deschenaux

group is large enough, we would not want an isolated instance of dissent to compromise our attribution of collective akrasia to the group. We should therefore reformulate condition (c) as ‘a sufficient number of members of the group judge that, all things considered, it would be better to do a than b’. Similar remarks can be made for points (a) and (b). In a more formal treatment, we would have to identify what counts as a ‘sufficient number’. It is not clear what strategy we should follow to do so, but in any case, a fully formalized treatment is not my aim here. Instead, my goal is more simply to determine whether there is value to approaching collective akrasia from this angle. Less formally, collective akrasia in this perspective is the phenomenon that arises when enough group members believe that collectively acting in a particular way would, all things considered, be better than not doing so, but these same group members do not act accordingly. In moving from individual akrasia to collective akrasia, very little has changed at the definitional level. We have only replaced individual action with collective action. An important consequence of this change, however, is that collective akrasia can occur without group members being individually akratic (this is comparable, in some respects, to Pettit’s ‘discursive dilemma’). Indeed, judging that it would be best for all members of a group to act in a particular way does not necessarily mean that one ought to act accordingly, even when one belongs to the group. It may be that such a judgement only entails that one should act accordingly on the condition that other members of the group will do the same. This is a more relaxed definition of collective akrasia than the one proposed by Pettit because it does not specify at the outset what kind of group is capable of akrasia. Pettit’s strategy is to declare that only a rationally unified agent could display akrasia, and then ask under what conditions groups can be rationally unified. As I have shown above, this is a non-starter for the Bishwakarma. Pettit’s strict constraints seem to be guided by his desire to ultimately discover something about individual akrasia by studying the collective case. In particular, he is keen to apply a ‘constitutional model’ of the self in his analysis, according to which a subject is composed of multiple, potentially discordant ‘voices’. He suggests that individual akrasia might arise analogously to what happens in the collective case when the voices that compose a self are discordant and generate a ‘discursive dilemma’. While I find this specific point convincing, Pettit’s constraints are too strict. They forgo any possibility of collective akrasia that differs, in structure, from the individual case. Collective akrasia as I define it is, in essence, a coordination problem. Coordination problems have been explored systematically in the literature on collective action, using game theoretic concepts. A detailed presentation is beyond the scope of this chapter (see List 2014), but a brief description is straightforward. There are broadly two possible courses of action for the Bishwakarma: ‘treating the Pariyar as untouchable’ and ‘not treating the Pariyar as untouchable’. Both courses of action are stable and self-reinforcing. That is, as long as an overwhelming majority of the Bishwakarma follow



The ‘Replication’ of Caste as a Form of Collective Akrasia 139

one of these courses of action, it is very risky and difficult for any individual Bishwakarma to diverge from it.20 Thus, in a hypothetical world in which no Bishwakarma treat the Pariyar as untouchable, there would be no incentive for any given individual Bishwakarma to start doing so. In fact, switching to treating the Pariyar as untouchable would be costly because it would mean acting against one’s own anti-casteist ethical norms. What keeps this hypothetical world from coming about, however, is the fact that in a situation where all, or more accurately, almost all the Bishwakarma do treat the Pariyar as untouchable, not doing so comes at a high reputational cost for any given individual, and this reputational ‘cost’ outweighs the ‘benefits’ which one would accrue from acting in line with one’s anti-casteist ethical norms. Using this kind of analysis yields important insights into why certain practices endure. It gives a formal description of what it means for a practice to be self-reinforcing. Additionally, it shows why it is so difficult for a group or population to switch from one course of action to another. Any trailblazing, individual Bishwakarma who chooses to stop treating the Pariyar as untouchable must do so at a considerable personal, reputational cost. It is only if many Bishwakarma collectively decide to simultaneously stop treating the Pariyar as untouchable that this reputational issue can be avoided (see Bicchieri 2016, 106ff. on coordination and norm change). Existing literature has used a similar analytical framework to explain why certain harmful social practices endure. Mackie (1996), for example, argues that the sudden end of foot binding in China is best understood as a rapid shift from one equilibrium to another, in a coordination problem. He argues that, as long as foot binding was necessary to ensure a daughter’s good marital prospects, it would have been too costly for any individual family not to bind a daughters’ feet. The practice ended with the creation of alternative marriage markets, in which the parents of boys promised not to marry their son to a woman with bounded feet and the parents of girls promised not to bind their daughter’s feet. Such markets created a space in which binding one’s daughter’s feet brought no tangible advantage and was therefore disincentivized. Key to Mackie’s analysis is the fact that it was only through the coordinated or collective creation of these alternative markets that the practice could be eliminated. The overall lesson from such an analysis is that social practices, including ones perceived as nefarious by the very people who engage in them, can be self-sustaining: the simple fact that others do something can be sufficient to make it rational and reasonable for individuals to follow along. An analysis of ­replication – ­or indeed any social ­phenomenon – ­couched in these terms is necessarily a simplification. Such an explanation cannot exhaustively explain replication. Social reality is complex, and there are many reasons a Bishwakarma might treat a Pariyar as untouchable. I am developing a deliberately simplified and abstracted model of the phenomenon, in which I purposely ignore much of the complexity at hand. I am doing this in the hope of better capturing one specific aspect of replication, namely, the extent to

140 Ivan Deschenaux

which it might be a collective, self-reinforcing process. In other words, I do not think that replication is merely a coordination problem, nor do I think it can be reduced to a matter of ‘social convention’, a term sometimes applied in the literature on coordination problems. I do, however, think that the extent to which replication is self-reinforcing has gone broadly unnoticed thus far and helps us explain the tension better than previous explanations.21 What distinguishes collective akrasia from a run-of-the-mill coordination problem? The answer lies in the observation reported earlier, namely, that the Bishwakarma are cognisant of what is going on. The Bishwakarma are aware that their treatment of the Pariyar is, in no small part, driven by the gaze of other Bishwakarma. They talk about the way in which others would perceive them, were they to act differently. In the view of collective akrasia that I am developing, this conscious realization is key. A run-of-the-mill coordination problem can occur without anybody involved realizing what is going on. In contrast, collective akrasia occurs when individuals consciously realize that their behaviour is not in line with ideals which they subscribe to, but they also realize that they cannot afford to change their individual behaviour without the support of others. This view of collective akrasia raises several questions. Among these, three are particularly intriguing. First, does this conception of collective akrasia teach us something about individual akrasia? The question is similar to Pettit’s in the piece cited earlier, but the answer may be quite different. A description of agents as rationally unified seems to underlie the literature on individual akrasia. I wonder if the view of collective akrasia I am pushing for here has the potential to undermine this assumption. Take, for example, Amelie Rorty’s (1980) taxonomy of akratic ‘breaks’, which lists the different levels at which akrasia might occur within an agent. A ‘break’ might occur between a subject’s ethical evaluations and intentions, between ethical evaluations and the interpretation of a situation, between intentions and actions, etc. The unspoken implication behind this taxonomy is that, under normal functioning conditions, there should be no break. A normally functioning agent should act in a rationally unified way: they should adequately apply their ethical principles when evaluating a given situation, derive intentions from such evaluations, and their actions should follow seamlessly. What reflecting on collective akrasia suggests is that ‘breaks’ might occur even when an agent is acting in a unified way, and in such cases, these breaks may not be akratic as traditionally understood. Such breaks might occur simply because of discrepancies between the best course of action for an individual and for the collective that individual is part of. Second, what is the phenomenology of collective akrasia? What is it like to be in a situation where one consciously violates one’s ethical principles because respecting them could upset the collective one is part of? In the case of the Bishwakarma, what is it like to choose to preserve one’s reputation over applying one’s commitment to castelessness? The reason a phenomenological



The ‘Replication’ of Caste as a Form of Collective Akrasia 141

study will be useful is because it might reveal the thinness of an approach that focuses too narrowly on intentions and actions. We saw earlier in this chapter why it is rational for any individual Bishwakarma to continue treating the Pariyar as untouchable, despite judging this immoral, and despite knowing full well what it is like to be on the receiving end of such a treatment. Though I cannot speak of my interlocutors’ inner experiences, I suspect that, in general, the rationality of a course of action does little to reduce the ambivalence and inner conflict which one might experience in such a situation, particularly in social contexts where rationality is not given inherent value. Judging that a course of action is best for the group may not entail that one should individually follow this course of action, but might it not still exercise a ‘pull’ of some kind? One way a phenomenological approach might help us understand this pull is by showing whether collective akrasia involves collective intentionality. In this chapter, I have used descriptions such as ‘believing that the group ought to do’ something. It may turn out that such beliefs are better captured as instances of ‘we-intentionality’ (Sellars 1974, Tuomela and Miller 1988, Tuomela 2005). The difference, from a first-person perspective, is between ‘I believe it would be best if the group I am part of did a’ and ‘I believe we ought to do a’. The direct inclusion of the self in the second formulation might help explain the inner conflict that, if my guess is correct, occurs in collective akrasia. A third question raised by this approach to collective akrasia concerns social norm change. This is a field which I am not familiar with, but in which an approach based on coordination problems has proven fruitful (I have already mentioned Mackie (1996) and Bicchieri (2016)). My description of collective akrasia has drawn on these approaches and focused on the question of whether a subject is aware of the coordination issue. There is scope, then, to look at the impact of this awareness on norm change over time.

Conclusions Replication, as Deliège showed, should not be read as an indication that Dalits consent with caste ideology. Neither should it be seen, however, as just one more manifestation of humans’ fundamental lack of coherence, as Deliège would have it. Instead, we should acknowledge that Dalits may often be aware of the incoherence inherent in replication, and explain why this awareness does not suffice to change their behaviour. The explanation is found in the disparate demands of anti-casteist norms on the one hand, and the preservation of one’s individual reputation in a casteist collective on the other. When we consider these disparate demands under the analytical lens of akrasia, it becomes clear that no Bishwakarma individual can adequately be said to display individual weakness of the will, akratic or otherwise, when they treat the Pariyar as untouchable. Non-akratic weakness of the will is

142 Ivan Deschenaux

excluded because the Bishwakarma do not form intentions to stop behaving the way they do in the first instance, and akratic weakness of the will is excluded because it is rational for the Bishwakarma to preserve their reputation at the cost of violating an ethical norm which they otherwise endorse. These considerations are more than just analytical: as mentioned earlier, it is not uncommon in Nepal to hear non-Dalits blame the Bishwakarma for their apparent inconsistency and suggest that their behaviour towards the Pariyar discredits their demands for equal treatment. Showing why the Bishwakarma’s behaviour is not weak-willed in the traditional sense should help us dismantle this kind of accusation. Arguing that their behaviour fits the formal structure of a coordination problem should help us show why replication, like other instances of caste-based discrimination, is a social and collective behaviour through and through. It is structural, and my hope is that describing the phenomenon as a case of collective akrasia says something about its structure. On the philosophical side, I have developed a view of collective akrasia which, unlike that proposed by Pettit, is more than an extension of the individual case to very specific, limited collectives. I have shown why collective akrasia occurs when members of a group judge, all things considered, that it would be better for all members of the group to follow a particular course of action a, but the members of the group instead follow a course of action b. This definition, which deliberately mirrors Davidson’s classic definition of individual akrasia, suggests that collective akrasia has the same formal structure as a coordination problem in which actors are ‘stuck’ in the worse of two equilibria. What distinguishes collective akrasia, in this view, is the conscious realization on part of the agents involved that they are in such a situation. This view raises several further questions. Most importantly, it demands that we re-evaluate the assumptions of rational unity omnipresent in the literature on individual akrasia, it suggests that there would be much to gain from a phenomenological approach, and it demands closer focus to the dynamics of collective intentionality. Ivan Deschenaux is an LSE Fellow in Qualitative Methods in the Department of Methodology. He has conducted extensive ethnographic fieldwork in Nepal, focusing on caste and caste-based discrimination, as well as some online experimental work in India. His research lies at the interface between sociocultural anthropology and social cognition. He is particularly interested in what makes some social norms endure even in contexts where people oppose them. He is also interested in the integration of ethnographic methods with other experimental and quantitative approaches.



The ‘Replication’ of Caste as a Form of Collective Akrasia 143

Notes   1. The term ‘Dalit’, which translates as ‘broken’, is used by activists and often considered inherently political. In India, a more neutral appellation, favoured by the State, is ‘Scheduled Caste’ (SC). The latter is also often used in academic literature. In Nepal, ‘Dalit’ is in use both by activists and the State, regardless of political alignment. Accordingly, I use it here to designate all people who belong to castes that were once considered ‘untouchable’ (achhūt, pāni nachalne), not only those who are politically organized.   2. There is a lack of recent data disaggregating poverty by caste in Nepal, hence the somewhat older sources cited here. The soon-to-be-finalized Nepal Living Standards Survey (NLSS-IV) should provide more recent data.   3. Following standard anthropological practice, places and people have been given pseudonyms throughout this chapter.   4. There are more precise rules dictating which foods, cooked and raw, a non-Dalit may and may not accept from a Dalit, but there is no need to enter further detail here.   5. In this region in the past, and in other parts of Nepal today still, a number of public spaces were off-limits for Dalits, including tea shops, restaurants, temples and others. This is generally no longer the case in the part of the country where I conducted fieldwork.   6. In extreme cases, intermarriage can lead to the non-Dalit family committing violent crimes against the Dalit spouse and their family. I observed no such case during my fieldwork, but their existence is common knowledge.   7. In Moffatt’s words, ‘[Untouchables] recreate among themselves the entire set of institutions and of ranked relations from which they have been excluded by the higher castes by reason of their extreme lowness’ (1979, 5).  8. An adequate presentation of this topic requires a discussion of the historical ‘Hinduization’ of Nepal and recent ethnic politics. This is beyond the scope of this chapter.   9. Some may argue that, with such a restrained definition, I am discussing an altogether different phenomenon. Although I do not think this is the case, it would not be a problem. What I am primarily interested in here is the way in which some Dalits treat others as untouchable. This would remain an empirically observable fact even if we ultimately choose not to call it ‘replication’. 10. For brevity’s sake, I continue to use a simplified model here in which I consider only the Bishwakarma and the Pariyar. The reality is more complex, and there are multiple divisions among Nepali Dalits. One could easily expand the argument to include these additional complexities. 11. It is worth noting that these idioms are also used by progressive non-Dalits who oppose caste. Whether the idioms have the same valence when used by non-Dalits is not clear to me. They may take on a different character, becoming less of a normative claim that caste should not exist, and more of an instance of ‘caste blindness’, i.e. a false notion that society has already moved past it. 12. This claim takes on further significance if contrasted to the Manusmriti. In this text, which many in Sammagaon are aware of, different caste groups (varna) are said to have emerged from different body parts of the creator of the universe. 13. The same is in fact true of non-Dalits who oppose caste-based discrimination. NonDalits people use the same euphemistic sentences when turning down food cooked by the Bishwakarma, for example. In this case, however, nobody would suggest that they are replicating a system which is detrimental to them, so the rest of the argument does not apply. 14. Note that the two are not the same. Say I resolve to finish writing my paper tomorrow but, when the time comes, I discover that I feel quite tired and judge that it would be best

144 Ivan Deschenaux

15. 16.

17. 18. 19. 20. 21.

for me to take a nap instead of writing. In this case, I may be weak-willed (too readily violating a previous intention), but not akratic (because I judge that taking a nap is preferable). This is to say, they are aware that this is in the realm of possible actions, not that it is a likely alternative. Pettit offers his own definition of akrasia, which is analogous to the Davidsonian variety: ‘… an agent is akratic when the following conditions are fulfilled. The agent holds by intentional states in the light of which he or she sees that a certain response is required; the states involved may be beliefs or desires, judgments or intentions, and the required response will typically be an action. The agent functions in conditions that are intuitively favourable, and within limits that are intuitively feasible, for acting as required; there is nothing abnormal about how things transpire within his or her constitution or ­circumstances – n ­ o malfunction, for example, or perturbation. But nevertheless the agent fails to act in the required manner’ (2003, 68). Pettit himself is somewhat unsure whether such cooperatives may occur in reality, but he lists the possibility to be exhaustive. Details in Fig. 3.2 in Pettit (2003, 83). This does not require, as should be clear from the definition itself, any concept of group mind or super-individual agent. We might view these courses of action as Nash equilibria. I also think that a version of this argument might carry over to caste-based discrimination more generally, but this is a topic for elsewhere.

References Bennett, L., B. Sijapati, and D. Thapa. 2013. Gender and Social Exclusion in Nepal: Update. Kathmandu: Himal Books. Berliner, David, Michael Lambek, Richard Shweder, Richard Irvine, and Albert Piette. 2016. ‘Anthropology and the Study of Contradictions’. HAU: Journal of Ethnographic Theory 6(1): 1–27. https://doi.org/10.14318/hau6.1.002. Bicchieri, Cristina. 2016. Norms in the Wild: How to Diagnose, Measure, and Change Social Norms. Oxford: Oxford University Press. Bloch, Maurice. 2013. Anthropology and the Cognitive Challenge. Cambridge: Cambridge University Press. CBS. 2005. ‘Poverty Trends in Nepal 1995/96 and 2003/04’. Central Bureau of Statistics. . 2011. ‘Poverty in Nepal’. Central Bureau of Statistics. Das, Veena. 2015. ‘Caste’. In International Encyclopedia of the Social & Behavioral Sciences, edited by James D. Wright, Second Edition, pp. 223–27. Amsterdam: Elsevier. Davidson, Donald. 1970. ‘How Is Weakness of the Will Possible?’. In Essays on Actions and Events, pp. 21–42. Oxford: Oxford University Press. https://doi.org/10.1093​ /0199246270.003.0002. Deliège, Robert. 1992. ‘Replication and Consensus: Untouchability, Caste and Ideology in India’, Man: 155–73. Dumont, Louis. (1966) 1980. Homo Hierarchicus: The Caste System and Its Implications. Complete rev. English ed. Chicago: University of Chicago Press. Gellner, David. 2007. ‘Caste, Ethnicity and Inequality in Nepal’. Economic and Political Weekly 42(20): 1823–28.



The ‘Replication’ of Caste as a Form of Collective Akrasia 145

Gupta, Dipankar. 2005. ‘Caste and Politics: Identity over System’. Annual Review of Anthropology 34: 409–27. Henden, Edmund. 2016. ‘Addiction, Compulsion, and Weakness of the Will: A DualProcess Perspective’. In Addiction and Choice: Rethinking the Relationship, edited by Nick Heather and Gabriel Segal, pp. 116–31. Oxford: Oxford University Press. Höfer, András. (1979) 2004. The Caste Hierarchy and the State in Nepal: A Study of the Muluki Ain of 1854. Kathmandu, Nepal: Himal Books Classics. Holton, Richard. 1999. ‘Intention and Weakness of Will’. The Journal of Philosophy 96(5): 241. https://doi.org/10.2307/2564667. Karanth, G.K. 2004. ‘Replication or Dissent? Culture and Institutions Among ’Untouchable’ Scheduled Castes in Karnataka’. Contributions to Indian Sociology 38(1–2): 137–63. https://doi.org/10.1177/006996670403800106. List, Christian. 2014. ‘Three Kinds of Collective Attitudes’. Erkenntnis 79(S9): 1601– 22. https://doi.org/10.1007/s10670-014-9631-z. Mackie, Gerry. 1996. ‘Ending Footbinding and Infibulation: A Convention Account’. American Sociological Review 61(6): 999. https://doi.org/10.2307/2096305. Moffatt, Michael. 1979. An Untouchable Community in South India: Structure and Consensus. Pinceton, NJ: Princeton University Press. Pettit, Philip. 2003. ‘Akrasia, Collective and Individual’. In Weakness of the Will and Practical Irrationality, edited by Sarah Stroud and C. Tappolet, pp. 68–96. Oxford: Oxford University Press. Roberts, Nathaniel. 2016. To Be Cared for: The Power of Conversion and Foreignness of Belonging in an Indian Slum. Pdf. Oakland: University of California Press. Rorty, Amelie Oksenberg. 1980. ‘Where Does the Akratic Break Take Place?’. Australasian Journal of Philosophy 58(4): 333–46. https://doi.org/10.1080/000484​ 08012341341. Sellars, Wilfrid. 1974. Essays in Philosophy and Its History. Philosophical Studies Series. Dordrecht: Springer Dordrecht. https://doi.org/10.1007/978-94-0102291​-0. Stroud, Sarah. 2014. ‘Weakness of Will’. In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Spring 2014. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/spr2014/entries/weakness-will/. Tuomela, Raimo and Kaarlo Miller. 1988. ‘We-Intentions’. Philosophical Studies 53(3): 367–389. Tuomela, Raimo. 2005. ‘We-Intentions Revisted’. Philosophical Studies 125: 327–369. WB–DFID. 2006. ‘Unequal Citizens: Gender, Caste and Ethnic Exclusion in Nepal’. Kathmandu: World Bank and Department for International Development (DFID).

7 Is Grit Irrational for Akratic Agents? Lubomira Radoilska

Introduction Contemporary analytic philosophers tend to see akrasia, or acting against one’s better judgment, as a problem of motivation. According to this standard view, akratic actions are paradoxical since akratic agents know that they have a better alternative but nevertheless take up the worse, akratic option. In other words, akratic agents know what they are doing. They do not make any epistemic mistakes ­but – ­inexplicably – engage in behaviours that they correctly identify as wrong. The thought that akratic agents are not flawed as inquirers and knowers but only as agents plays a key role in turning akrasia into a textbook example of motivational only, or practical irrationality. This paper will aim to revise the standard view by emphasizing the epistemic dimensions of the phenomenon; that is, the ways in which akrasia affects both how agents understand their own involvement and how they handle evidence about their prospects of success. The ambition is to show that akratic agents typically rationalize their akrasia. They do not recognize it as paradoxical or irrational. Instead, they reinterpret it as separate, goal-directed actions undertaken under conditions that are not ideal for them. This rationalization of akrasia is closely related to another epistemically deficient habit: akratic agents pay too much heed to evidence that they are unlikely to succeed. In so doing, they display too little of what philosophers have described as ‘epistemic resilience’ or ‘grit’. This result is significant for a number of reasons. First, it helps shed light on the relationship between the motivational and the epistemic sides of akrasia. Second, it offers a fuller understanding of the phenomenon as a multi-faced process that unfolds over time rather than a sequence of paradoxical actions. Third, it avoids issuing conflicting normative requirements toward agents who, like the akratic, already find themselves in an irrational state.



Is Grit Irrational for Akratic Agents? 147

This philosophical project hopes to make three related contributions to the ethnography of akrasia as conceived in this volume. First, it develops a more nuanced conceptual apparatus that would enable us to capture more fully the richness and complexity of the phenomenon under consideration and to acknowledge the potentially conflicting and irreducible standpoints from which it can be experienced and assessed. In this respect, it expands on the constitutive thought that akrasia is not well understood in terms of mere inconsistency or inner conflict since it only becomes an issue when a certain set of normative expectations of what it is to be an agent are at work (McKearney and Evans, in this volume). For, as evidenced by Francesca Mezzenzana’s contribution to this collection, inconsistency between one’s judgment and action would not amount to akrasia in the absence of such expectations. By articulating these normative expectations in greater detail, the present chapter could help elucidate contexts within which meeting them is apt, in contrast to others where this would be misguided. A second contribution is to shed critical light on the nature and scope of the normative implications of identifying or misidentifying an action as akratic. As illustrated by Rebecca Lester and Patrick McKearney in their respective chapters, the stakes can be extremely high, especially when the persons whose actions are interpreted as akratic belong to vulnerable groups. In the case study discussed by Lester, the decision to fund or not the muchneeded treatment of an eating disorder hangs on just such an interpretation by a medical insurance company representative. In a similar vein, McKearney’s comparative study of care settings for people with learning disabilities across three cultures reveals the significance of identifying their putative choices, akratic appearances notwithstanding, where these choices are seen as directives for the carers to follow. The three personal stories explored in the present chapter broaden this line of inquiry to ascertain how and why close attention to past evidence of one’s own vulnerabilities might affect the self-attribution of akrasia in problematic ways. Third and final, the proposed philosophical analysis puts forward a pluralist conception of rationality, whose normativity is both context- and agent-­sensitive. Operating with such a conception helps avoid introducing unexamined assumptions about what rationality requires when approaching instances of apparent akrasia, whether in familiar or unfamiliar contexts. Darin Weinberg’s contribution to this volume indicates how and why this could offer a welcome corrective to much theorising about excessive substance use in terms of akrasia. For, as Weinberg compellingly argues, on many occasions, when equating rationality with across-the-board abstinence or restraint, we miss out on the life-enhancing, and hence, rational role that substance use plays. The discussion proceeds as follows. First, I consider the so-called Evidential Threshold Account of grit, an epistemic attitude that allows agents to bracket out some unsettling evidence about their own chances of success. On this account, grit is rationally permitted only when agents operate in favourable or

148 Lubomira Radoilska

neutral circumstances. I then demonstrate how this account applies to akratic agents by outlining three personal stories of addiction treatment. In light of these stories, akrasia appears to be a particularly unfavourable circumstance, to the point of making grit no longer rational, if the Evidential Threshold Account is correct. Next, I introduce my own philosophical account of akrasia as necessarily less-than-successful conduct, a mode of action that avoids straight failure but is never fully successful. I then expand on this idea to show how the epistemically deficient habit of rationalizing one’s own akrasia leads to a motivational perseverance of the wrong kind. This allows me to contrast and compare my view of akrasia as a long-term process where the akratic conflict remains out of focus for the affected agents with the standard view, which insists instead on their full awareness of conflict at the time of action. I argue that the alternative I propose can offer a clearer picture of why akrasia is a case of irrationality, both practical and epistemic, rather than a concealed form of rationality or a sub-personal kind of agency, beyond the scope of rational assessment. In conclusion, I return to the question of how unsettling evidence is handled in the context of akrasia. I show that a requirement to refrain from grit would lead akratic agents into further irrationality rather than help them get out of their irrational state.

The Evidential Threshold Account of Grit According to a recent account put forward by Morton and Paul (2019), grit is an epistemic attitude of resilience toward evidence that we are not likely to succeed in projects of ours we have already invested in considerably. The authors highlight the motivational significance of remaining confident in one’s own success, especially when faced with difficulties. In other words, it would not be practically rational for agents who have worked toward a challenging long-term objective to give up at the first sign of difficulty. Grit provides such committed agents with an epistemic ­buffer – ­they are rationally permitted to set aside unsettling or contrary evidence about their own prospects of success. This permission, however, is limited. Once faced with strong evidence that success is unlikely, even committed agents are no longer rationally permitted to resist or downplay it. So, the room for rational grit lies between two epistemic extremes. On the one extreme, an agent places too much weight on contrary evidence, thus compromising her own chances of success by giving up too readily on more demanding, longer-term projects. On the other extreme, however, she pays too little heed to contrary evidence and perseveres in tasks long after her efforts have proven fruitless, thus also undermining her chances of succeeding at some alternative undertaking. Rational grit therefore excludes instances of excessive epistemic resilience where ‘we must rely on tactics like avoiding or ignoring the evidence or nurturing positive illusions about our abilities or the extent of our control’ (Morton and Paul 2019, 183).



Is Grit Irrational for Akratic Agents? 149

To distinguish between the two, Morton and Paul develop their Evidential Threshold Account. It tells us that there is a range of rationally permissible ways, or policies, to treat contrary evidence once an agent commits to a project. This range varies from the policies available to either an independent observer faced with the same evidence or that same agent before her committing. In either of these two cases, the evidential threshold required will be lower than that for a committed agent. That is to say, both an observer and an agent who contemplates a particular project should draw negative conclusions about the prospects of this project on grounds that might not be sufficient for a committed agent. Such an agent would be permitted to suspend judgment as long as her rationale does not involve any of the tactics listed earlier and, more generally, steers away from what Morton and Paul (2019, 194) term ‘delusional optimism’. Importantly, the higher evidential threshold to which committed agents are entitled is ‘ecologically constrained’: to remain epistemically rational, these agents should consider not only the content of the contrary evidence available to them. They should also factor in the overall context in which they operate. This context is agent-specific: what would look like neutral, if not favourable, circumstances for some agents would amount to stark adversity for others. Morton and Paul (2019, 200) refer to the vast array of precarious circumstances that would impact the evidential threshold for committed agents as ‘resource scarcity’. This may include poverty or other forms of socioeconomic disadvantage but also traumatic personal experiences or lack of emotional support. Since the evidential threshold of committed agents should not only be responsive to data about their chances of success but also to the context in which these data are to be interpreted, commitment under extreme resource scarcity might not provide greater leeway with respect to contrary evidence. This is the crux of Morton and Paul’s argument: once we recognize that evidence is not just content, but content in context, rational grit turns out to be out of reach for agents who apparently need it most. The point is made as follows: … it may be that in situations of extreme scarcity agents should have reasoning habits that lead them to remain maximally sensitive to evidence of potential failure even after adopting a difficult goal. Put simply, perseverance may not serve such agents well. One reason is that such environments tend to be more unpredictable, and so the agent’s initial assessment of the likelihood of success might be less robustly justified. Another is that for an agent with scarce resources, events that would constitute small setbacks for someone else might be devastating … Given the high stakes of failure, retaining low evidential threshold may be more rational than the alternative … For agents who regularly operate in unsupportive or even discriminatory contexts … grit can lead to the investment of more effort than is effective or healthy. Consequently, it may be that agents in contexts of severe material and emotional scarcity ought not to have an evidential policy that enables grit at the expense of caution or self-protectiveness … they should remain highly responsive to evidence that pure effort will not be enough. (Morton and Paul 2019, 201–2)

150 Lubomira Radoilska

The authors admit that this result is ‘distressing’. The reasoning habits that would amount to rational grit under more favourable conditions do not support the achievement of agents operating under resource scarcity but lead to their ‘over-efforting’ instead. As a consequence, the limited resources they have at their disposal are further depleted, in vain. Morton and Paul (2019) illustrate the link between misplaced grit and over-efforting with patterns of underachievement that affect students from underrepresented groups at elite universities. Unlike their peers from more privileged backgrounds, when faced with good predictive evidence that they are unlikely to complete a competitive course, many of these students do not opt for an alternative that would allow them to satisfy equivalent academic requirements but keep struggling with the original course. As a result, they end up failing their degree. Following Steele (2010) where this example comes from, Morton and Paul explain this instance of over-efforting as a counterproductive response to ­‘stereotype threat’. The term was first introduced by Steele and Aronson (1995) and refers to feeling at risk of confirming, as a self-characteristic, a negative stereotype about one’s social group or identity. For instance, in the social psychology experiments conducted and analysed by Steele and Aronson, Black undergraduate students performed worse than their White peers on standardized tests but only when race was made salient in advance of the task. When race was not made salient in this way, Black students performed better than or just as well as White students. The hypothesis put forward by Steele and Aronson is that performance in academic contexts can be impeded by awareness of how one’s conduct might be perceived through the lens of identity stereotypes; hence, the term ‘stereotype threat’. Returning to the example discussed in Morton and Paul (2019), students from underprivileged backgrounds display excessive grit out of fear that by taking the easy way out, they would effectively confirm the negative stereotype that people like them are not academically strong. The present discussion will focus on a different set of cases altogether. Agents of interest to us would satisfy the conditions of extreme resource scarcity as defined by Morton and Paul. In addition to lack of emotional support and sometimes material deprivation, this scarcity would, however, also ­derive – ­partly but ­importantly – ­from their own records as agents. The feature that these agents have in common is that some exercises of their agency have been affected by a distinct form of akrasia, of which addiction is an example. Drawing on the Evidential Threshold Account we just explored, we should expect that grit cannot be rational for such agents. For even without notable scarcity in terms of external resources, akratic agents are subject to extreme scarcity in terms of internal resources: poor past performances linked to akrasia cast a shadow over any future commitments of theirs. Yet, there are strong reasons to doubt that grit may not be rational for them. To see what these reasons might be, let us first lay out three personal stories where agents with addiction try



Is Grit Irrational for Akratic Agents? 151

to pursue but ultimately fail at difficult, long-term projects in spite of sincere commitment. I shall then articulate what makes these stories illuminating for akratic agency overall. Expanding on this analysis, I shall return to the q ­ uestion of rational grit to argue pace Morton and Paul (2019) that greater resilience to contrary evidence is not only rationally permitted but also required in instances of extreme resource scarcity, such as akrasia.

Three Stories of Akrasia The following personal stories come from Radoilska and Fletcher (2016). Although fictional, they are informed by Fletcher’s four-decade-long clinical practice as an NHS consultant psychiatrist.

Mr Miller Mr Robert Miller is a 65-year-old retired chief executive. His mother died at the age of 82 from ‘old age’. His father died at the age of 58 from carcinoma of the oesophagus having been a heavy drinker throughout his adult life. Mr Miller was an only child and described a happy and stable childhood despite his father’s drinking. He excelled at school, enjoyed good peer relationships and obtained a first-class honours degree at university. He married in his late 20s, had two children in his 30s, and in his mid-40s became the chief executive of a national company. He is described by his family as a good husband and father, with a reputation for honesty, integrity and fairness. Throughout his working life, he drank alcohol most days, attributing this to the stress of his job and frequent socializing. In his early 60s, he developed a tremor in his hands in the morning, which he suspected was due to anxiety. His wife and children became increasingly concerned about his drinking, especially as he was known on occasions to drink and drive. Under considerable family pressure, he saw his GP and was referred for CBT to treat anxiety, stress and depression. He attended these sessions regularly but did not find them helpful and his drinking pattern did not change. Following a blood test to check thyroid function, his GP detected markedly deranged liver function tests and referred him to a consultant psychiatrist who diagnosed moderate alcohol dependence. Mr Miller declined the offer of medication, believing that he was strong-willed enough to reduce his drinking on his own, but he did accept two counselling sessions with a substance misuse liaison worker. When he was 64 years old, he arrived home one evening after drinking and fell out of his car in a very intoxicated state. An ambulance was called and Mr Miller was taken to the A&E department. He was ‘terrified’ that he would be reported to the police for driving under the influence of alcohol, but this did not occur. The shock and embarrassment of this episode led him to accept

152 Lubomira Radoilska

treatment advice from his consultant psychiatrist, who arranged for a home detoxification followed by treatment with acamprosate 666 mgs t.d.s., and disulfiram 200 mgs daily, which his wife promised to supervise ‘religiously’. For six months prior to his retirement, Mr Miller complied with treatment. His wife, however, gave up supervising disulfiram after three months as she had started to ‘trust’ her husband again. His mood was buoyant, his work performance strong and he looked physically fit, having lost weight. Against the advice of his consultant psychiatrist, Mr Miller stopped taking medication one month prior to retirement so that he could ‘enjoy’ his farewell party. This is because, when mixed with alcohol, disulfiram produces strong uncomfortable feelings similar to hangover almost immediately after consumption. Mr Miller was convinced that there would be no problems with alcohol after retirement in view of his clinical progress and the future stress-free lifestyle he anticipated. He drank at his retirement party, relapsed back into uncontrolled heavy drinking and spent his early retirement days feeling depressed, deeply ashamed and bored. His very caring family were desperate for him to stop drinking and asked his psychiatrist if he could be ‘sectioned’. After some persuasion, Mr Miller had another home detoxification and restarted treatment with acamprosate and supervised disulfiram. He drank on top of his medication and started to talk about ‘checking out’, by which he meant committing suicide.

Amy Parker Amy Parker is a 21-year-old mother of one child. She never knew her biological father. Her mother had multiple boyfriends who often brought alcohol and drugs into the home. As a young girl, she was given alcohol and was sexually abused by a number of her mother’s temporary partners. Her educational performance was poor, and she socialized with a group of students on the fringe of school life. At the age of 11, she started smoking cigarettes, and as a 13-yearold, she self-harmed by scratching the inside of her thighs with scissors, but this behaviour never came to the attention of her teachers or GP. By the age of 15, she had used a wide range of ‘party’ drugs. Social Services were temporarily involved when Amy was found living on the streets having stopped going to school. At the age of 17, she smoked heroin, and within three months she was injecting into her arms and hands. Amy also used street diazepam, cheap alcohol and occasionally shared a pipe of crack cocaine. When she was 18 years old, she developed a left-sided deep vein thrombosis after injecting into her groin and was found to be hepatitis C positive. She became pregnant at the age of 19 and this led to a remarkable change in her behaviour. Amy began to attend a Community Substance Misuse Team (CSMT) where she was started on a methadone prescription. Her medication was supervised on a daily basis at a local supermarket pharmacy and the dose was gradually increased to



Is Grit Irrational for Akratic Agents? 153

120 mls methadone mixture 1mg/ml. This, together with the support of a substance misuse worker, appeared to help her stop using heroin and diazepam. A number of consecutive urine and swab tests were negative for illicit drugs. In view of being hepatitis C-positive, Amy was offered appointments at her local hospital’s antenatal department, which she attended regularly. Towards the end of the second trimester, she returned to live with her mother. Amy said that she was determined to give her baby the ‘best possible chance’ and was ‘desperate’ to be a good mother and to care for her child well. Throughout her pregnancy, Amy received close support from a community midwife, the social services and the CSMT. By the third trimester she was considered to have made excellent progress. In view of this, and continuing regular negative tests for illicit drugs, the pick-up regime of methadone was reduced to twice weekly. A small-for-dates baby boy was born in good health (apart from a squint) at 38 weeks’ gestation by spontaneous vaginal delivery. Amy experienced a short period of baby blues and did not take to breast feeding. Even with close support, she found the routine of caring for her baby demanding and exasperating. Within two months of the birth, Amy was no longer picking up her methadone on a regular basis and began to make excuses for failing to attend her key worker appointments at the CSMT. When she did attend, she said she was exhausted. A drug screen taken at 12 weeks’ post-delivery tested positive for heroin, cocaine and diazepam. Conflict with her mother accelerated when Amy started going out in the evenings leaving the baby in her mother’s care. Her mother told the CSMT that Amy was ‘seeing’ drug users and dealers she had relationships with in the past. Despite strenuous efforts and serious warnings from the CSMT, a Health Visitor and Social Workers from the Child Protection Team, Amy returned to her old pattern of injecting drug use and unstable relationships. Despite Amy’s promises of improvement and pleas for clemency, her son was eventually removed from her care and put up for adoption.

Peter Phillips Peter Phillips is a 27-year-old, ex-Army Corporal with no family history of psychiatric disorder. He was an average student, sporty, popular and outgoing. After leaving school, he joined the British Army and excelled during basic training. He loved Army life, enjoying the hard work, discipline and camaraderie. At weekends, he would drink heavily with his friends, but this did not seem to impact on his work performance. His military unit was closely knit, especially after their first tour of duty in Afghanistan. Whilst leading a night patrol during a second tour in Afghanistan, the soldier behind him stepped on a landmine. Peter was spattered with blood and shrapnel fragments but was able to continue. The patrol came under heavy fire and the men ran for cover. Peter found himself in an irrigation channel with two friends. Whilst they

154 Lubomira Radoilska

attempted to provide covering fire, Peter showed great bravery (later formally recognized), running back to the wounded soldier and dragging him 20 metres into the ditch. Attempts were made to provide first aid, with tourniquets being applied to both leg stumps, but despite their best efforts the soldier died. Following this, Peter said that his nerves were ‘shredded’. He felt constantly in danger, irritable, aggressive and guilty. After the tour in Afghanistan was over, the unit was sent to Cyprus for R&R. Peter got drunk every day, was argumentative and started getting into fights. Back in the UK, he lost interest in Army life and continued to drink heavily. He made the decision to apply for premature voluntary retirement. His unit medical officer referred him to a CPN at the military Department of Community Mental Health. The CPN thought that Peter had Post Traumatic Stress Disorder (PTSD), so he provided an abbreviated form of trauma-focused CBT and suggested to the unit medical officer that a prescription of mirtazapine, 30 mg at night, might help. The treatment proved beneficial. Peter subsequently left the Army but found it difficult to obtain work. He continued to suffer intermittent nightmares of the incident in Afghanistan and drank half a bottle of vodka most nights as he was ‘frightened to go to sleep’. He was unable to maintain a stable relationship with a girlfriend and due to continuing unemployment, he came under financial pressure. His previous symptoms of PTSD returned ‘with a vengeance’. His drinking spiralled out of control, he wet the bed regularly and suffered a bad bout of pancreatitis after which his GP told him to ‘completely and permanently abstain from alcohol’. However, Peter considered that using alcohol was the only way he could get to sleep and suppress the vivid memories, sense of danger, jumpiness and anxiety he experienced. Peter was arrested after attacking a stranger in a pub who criticized the Army and he ended up on a probation order. His probation officer arranged for referral to psychological services, but after waiting four months for an assessment, Peter was told that nothing could be done for him until he stopped drinking. Following referral to a community substance misuse team, he received an in-patient detoxification, during which he was re-referred to psychological services. Peter continued to have nightmares of Afghanistan, feelings of anger and aggression, and panic attacks. He kept away from all reminders of military life and avoided watching TV news programmes. Within two weeks of leaving the detoxification unit, he started to drink a bottle of vodka a day. He was again turned away from psychological services because of his alcohol consumption. Peter has managed to get a job as a car park attendant but is still drinking very heavily and suffering from PTSD. He says he ‘hates the taste’ of alcohol and wants to stop drinking, but he fears he might kill himself as he cannot cope with his nightmares, loneliness and sense of guilt.



Is Grit Irrational for Akratic Agents? 155

Akrasia as Necessarily Less-Than-Successful Conduct As shown by the three stories, none of the agents succeeds in their respective endeavours. For instance, Mr Miller ignores his ‘drink problem’, then tries to solve it by h ­ imself – ­unsuccessfully – looks for professional help, gets on the way to recovery, relapses, and ultimately, gives up trying. The pattern of good initial progress followed by relapse and despair is also replicated in the stories of Amy Parker and Peter Phillips. In the following, I will aim to shed light on lack of success as a distinctive feature of the exercises of agency illustrated in all three stories, exercises that we may qualify as akratic. To do so, I will expand on the account of akrasia as necessarily less-than-successful conduct that could manifest itself in patterns of either weakness of will or addiction I proposed in Radoilska (2013). In essence, the central feature we encounter across these closely related phenomena is understood as follows. First, being necessarily less-than-successful qualifies a conduct or a particular strand of a person’s agency; for instance, excessive use of alcohol and related behaviours like drinking in secret from family rather than this person’s agency overall. Second, being necessarily less-than-successful points to a distinctive kind of agency, the manifestations of which cannot be fully successful to the extent that they arise at all. They avoid straight failure; yet they do not qualify as achievements. In this sense, akrasia can be aptly defined as perseverance of the wrong kind. What is missing here is the openness to both success and failure that comes with engaging in a new project. So even if akratic conduct cannot be successful, it still provides a degree of certainty in getting at least something that is not readily available to agents.1 This certainty is akin to the sense of familiarity or feeling at home afforded by some habitual behaviours (cf. Carlisle 2014). Having said that, it is important to note that if akrasia is a habit, it is neither unthinking, nor ‘out of control’, but a distinctive mode of perceiving, interpreting, feeling, reasoning and doing that results in necessarily less-than-successful conduct.2 The reason for this becomes apparent when we take seriously ‘achievement’ as unifying idea that underpins all credible manifestations of agency (cf. Bradford 2015). Any action in this broad sense is meant to be a kind of achievement, at least on its own terms. Agential success here is defined across two complementary axes: not only production (bringing about an effect), but also assertion (an agent’s articulating a particular commitment of theirs). Only when these two axes are well-aligned is an action successful on its own terms. Such an action would amount to actualization on the part of the agent: her effecting a change in the world articulates a commitment of hers. By contrast, when production and assertion are misaligned in a distinctive and sustained way, rather than just coming apart, the ensuing actions are necessarily less-than-successful. The misalignment defining akrasia takes the following form: the relevant conduct is successful as production to the extent that it is unsuccessful as assertion. Following on from this account, addiction looks like a recalcitrant and particularly frustrating

156 Lubomira Radoilska

form of secondary akrasia that derives from a considerable difficulty, if not subjective inability to address an initial conflict between valuing (the assertive dimension of agential success) and intending (the productive dimension of agential success). Returning to the personal stories for illustration, when Mr Miller goes off his medication to enjoy his retirement party, what he does is successful in terms of production (he drinks alcohol at the party) to the extent that it is also unsuccessful in terms of assertion (it undermines his sincere commitment to a life free from alcohol). Importantly, the underlying conflict between assertion and production, valuing and intending is not immediately apparent to an akratic agent. Although neither complete, nor present in every instance of addictive conduct, this lack of self-awareness is significant. It brings together two prominent features in the phenomenology of secondary akrasia which are also present in our three stories: frustration at one’s own actions coupled with a sense of hopelessness for the future. These features point to an underlying conflict between assertion and production that stays out of focus and, instead of being addressed, keeps coming back to eventually solidify into necessarily less-than-successful conduct. The ensuing sense of under-achievement often serves as inducement to further necessarily less-than-successful exercises of one’s agency. As pointed out earlier, albeit it frustrating, this conduct comes with a reliable outcome. This might not be what akratic agents appreciate or even want. It is nevertheless something that they can cling on to. Recall Peter Phillips’s story: he hates the taste of alcohol and wants to stop drinking, even though he never spends a day of abstinence outside of the clinic.

Akratic Self-Awareness and Rationalization At first blush, it might seem paradoxical that frustration at one’s own actions does not offer a platform for the realignment of actual plans to sincere commitments. On closer look, however, it becomes clear that experiences such as these feed back into the lack of self-awareness that typically accompanies necessarily less-than-successful conduct, for they help mask the underlying conflict between its assertive and productive dimensions. The latter are seemingly broken down into separate one-dimensional actions whose outcomes would not strike us as mutually exclusive. This kind of re-description defines the retirement party planning by Mr Miller: the party is treated as a one-off occasion, which, in one sense, it clearly is. This, however, does not apply to drinking alcohol, which in Mr Miller’s case has been a near daily routine for decades. By failing to appreciate that drinking alcohol at his retirement party is an instance of the same necessarily less-than-successful conduct that he is trying to overcome, Mr Miller makes further instances such as this more likely to occur. And as we learn from his story, the relapse indeed leads to a sense of



Is Grit Irrational for Akratic Agents? 157

shame and hopelessness that in turn underwrite further necessarily less-thansuccessful actions of his. In fact, we can already see at the start of the story how akrasia gets reinterpreted as separate actions that belie the underlying process. Mr Miller accounts for his regular alcohol consumption as a goal-directed response to specific events in his professional life, such as frequent socializing with business partners. Following this line of thought, he does not anticipate drinking alcohol after his retirement. This partly explains his unpreparedness to address the recurring episodes of alcohol consumption during the first months of his retirement. For, on the goal-directed, action-by-action re-description Mr Miller was giving to his drinking, these episodes would not happen once the rationale for consuming ­alcohol – ­keeping up with professional ­obligations – ­no longer applied to him. What transpires at this point is that necessarily lessthan-successful conduct is undergirded by a stable, self-fulfilling mechanism that makes alternative approaches to agential achievement, to which success and failure are equally open, far riskier and less attractive as a result. Recent work on rationalization shows that relevantly similar mechanisms might be at work more widely, maintaining less-than-successful exercises of individual agency that would not qualify as akratic in the strict sense. As Schwitzgebel and Ellis (2017, 170) point out: Rationalization occurs when a person favours a particular conclusion as a result of some factor (such as self-interest) that is of little justificatory epistemic relevance. The thinker then seeks an adequate justification for that conclusion but the very factor responsible for her favouring it now biases how the research for justification unfolds. As a result of an epistemically illegitimate investigation, the person identifies and endorses a justification that makes no mention of the distorting factor that has helped guide her search.

There is a potential dissonance here between three categories of reasons: 1) actual motivating reasons that the process of rationalization keeps out of focus; 2) explanatory reasons that make this process intelligible on reflection; and 3) justificatory reasons that the agent appropriates post factum (cf. Alvarez 2013). On this picture of reasons for belief and action, rationalization would be especially pernicious when it papers over not just some dissonance but straight contradiction between different categories of reasons. The upshot is to make detection and eventual resolution extremely unlikely. As Schwitzgebel and Ellis (2017, 171) observe, rationalization not only ‘obstructs the critical evaluation of one’s own reasoning’; in addition, ‘it impedes the productive exchange of reasons and ideas among well-meaning interlocutors’. Being necessarily less-than-successful, akratic conduct rests on this kind of more pernicious rationalization that obscures akratic agents’ self-understanding as conflicted and precludes their effective communication with meaningful others. As illustrated by Mr Miller’s story, these two trends go hand in hand. By rationalizing his daily alcohol consumption as an efficient and

158 Lubomira Radoilska

context-dependent means to secure professional goals, he blocks the realization that this routine clearly jeopardizes important commitments of his. This rationalization succeeds in reducing the visibility of the conflict between assertion and production, valuing and intending from the first-person perspective of the akratic agent himself. This ambivalent success, however, is achieved at the expense of a related failure: Mr Miller must remain at odds with his family and keep ignoring, for as long as possible, that he has a problem with alcohol. The proposed analysis explains why the action-by-action re-description of one’s own akratic conduct paints akratic actions as either fully successful (e.g. Mr Miller’s take on his drinking before the A&E episode) or clear failures (e.g. Mr Miller’s view on his continuing to drink after retirement). For re-descriptions such as these require that success in action is interpreted as one-dimensional and dependent on a single question, on this occasion, that of whether the intended outcome is achieved or not. Importantly, even if the implicit notion of agential success at work here is scalar to allow for degrees of achievement rather than binary, this additional nuance would not be enough to transcend the limitations of the one-dimensional model on which success in action is assessed. As a consequence, akratic incentives are not identified as a subclass of a distinctive category of reasons, explanatory ones. Explanatory reasons, also known as ‘reasons why’ (cf. Dancy 2000) lead to action without necessarily motivating it in a way that is transparent to the agent. For instance, I might stay at home instead of joining a party because I forgot about it or because I find social interactions too demanding. In each of these cases, the reason that explains my staying at home is not a reason, in the light of which I act. This is fairly obvious in the first scenario, where I simply forget about the party. By contrast, in the second scenario, I might be tempted to rationalize my staying at home by pointing to a good (normative) reason I could have had for doing so, such as needing to rest or to complete a piece of work with a pressing deadline. The problem with this rationalization of mine is that it misdescribes what I do by misrepresenting my motives. This reason-swapping also undermines my self-understanding as an agent: I can keep engaging in less-than-successful conduct, such as avoiding parties without realizing that this is what I am effectively doing. In a similar vein, the failure to recognize akratic incentives as fundamentally explanatory rather than either motivating or normative reasons is part of the underlying mechanism that obfuscates the necessary correlation between successful production and unsuccessful assertion in akrasia. One-dimensional models of action, whether attuned to a binary or a scalar notion of success, do not offer a corrective. Instead, they inadvertently reinforce the pernicious rationalization that fuels akrasia.3



Is Grit Irrational for Akratic Agents? 159

Acting Against One’s Better Judgment as a Puzzle of Practical Irrationality Before returning to grit and its putative irrationality for akratic agents, there are two related questions we need to consider. First, how does the account of akrasia employed in the preceding analysis relate to the standard account, according to which akrasia consists in acting against one’s better judgment at the time of action? And second, why rely on this alternative account rather than the standard one when addressing the issue of whether grit is irrational for akratic agents? The standard account of akrasia introduced in Davidson (1970, 2001), and further developed in Mele (1987), focuses on discrete, self-contained actions that take place in a short space of time, such as the following:4

One Drink Too Many Alex has an important presentation to make next day early in the morning. Before heading back home after work, her colleagues invite her to join them for a drink. She decides to go out but have no more than two drinks as she knows that drinking any more will affect the quality of her performance next morning. As the evening progresses, she is offered a third drink, which, after a brief moment of hesitation, she takes. Against her better judgment, she ends up having a fourth drink as well. The akratic conflict here is meant to be stark. The action goes against the agent’s own better judgment. It is also free and uncompelled. And, since it all happens at the same time, it bears the hallmarks of an enacted head-on contradiction that singles out akrasia as paradigm case of practical irrationality. This is a neat, theoretically attractive picture. Yet, part of its appeal derives from leaving the phenomena under consideration fundamentally underdescribed. It is beyond the scope of the present discussion to offer a sustained argument in support of this claim.5 Instead, I shall briefly outline the advantages of filling in the standard account in the way proposed earlier. One of these advantages derives from conceiving akrasia as a process that unfolds over time rather than a series of one-off events. This is not to say that acting against one’s better judgment never takes the form suggested by scenarios, such as ‘one drink too many’. The interest of a diachronic interpretation is instead to indicate that on many occasions the akratic conflict is not immediately apparent and can be seen for what it is only if a longer-term view on exercises of agency is adopted. As shown by the stories we considered, first-personal rationalizations of the akratic process contribute to an action-by-action redescription, where the underling conflict is diffused: the akratic agent seems to be doing first one thing for a good reason and then, at a later time, another, very different thing, for a good reason also. All of these instances of akrasia

160 Lubomira Radoilska

are likely to slip under the radar on the synchronic understanding implied by the standard account. For cases, such as ‘one drink too many’ are only the tip of the iceberg. By focusing on them as central, we are likely to lose sight of how different facets of akratic conduct fit together and why this amounts to practical irrationality. Addressing this issue is another, related advantage of conceiving akrasia in terms of necessarily less-than-successful conduct. It helps solve a fundamental puzzle put forward by Davidson (2001): practical irrationality is difficult to pin down. If we explain its instances too readily, we turn it into a concealed type of rationality. If, by contrast, we treat these instances as ultimately unintelligible, we reallocate them to the domain of non-rationality or sub-personal agency. We observed a relevantly similar tension in Mr Miller’s story: his early rationalization of his akratic conduct presents it as fully rational, albeit unorthodox, goal-directed behaviour. By contrast, his family’s request to have him sectioned as well as his own later despair at failing to stop drinking relegate a cluster of his agency to something regrettable that is merely happening to him. In either case, a central feature of akrasia is missed out: the conduct to which it gives rise is criticizable in virtue of being irrational by the akratic agent’s own lights. Scenarios like ‘one drink too many’ are good at highlighting the puzzle; however, this is achieved at the expense of simplifying it. The synchronic approach to akrasia is only part of the problem. For the notion of akratic conflict as head-on contradiction, on which this approach rests, also suggests that akrasia is a fundamentally motivational issue: the agent sees the better course of action and yet engages in an opposite and inferior one. As a consequence, the epistemic dimension of akrasia is left unattended. Yet, looking at the preceding personal stories, akratic agents’ knowledge of their own actions is compromised in a distinctive way. Moreover, this is a constitutive feature of the phenomenon rather than a by-product of it. Being able to account for the interactions between the motivational and the epistemic aspects of akrasia is crucial for understanding it as practical irrationality without over-emphasizing its proximity to either rationality or non-rationality. Rethinking akrasia as necessarily less-than-successful conduct rather than merely an individual action undertaken against one’s own better judgment offers a promising ground for such an account.6

Grit and the Akratic Handling of Evidence Having explored a possible account of akrasia as necessarily less-than-successful conduct, in light of three personal stories, we can now appreciate fully the challenge presented by the Evidential Threshold Account of grit outlined at the start of this paper. To recap, according to this account, committed agents are rationally permitted to handle adverse evidence in a different way than impartial observers or even the same agents before committing. The difference



Is Grit Irrational for Akratic Agents? 161

consists in a higher threshold for treating such evidence as decisive in the case of committed agents. For, once they assess a long-term project of theirs as very unlikely to succeed in the light of newly acquired evidence, they have to give up, in spite of the time and effort they have already expended on it. By contrast, when faced with evidence that speaks against a particular undertaking, both observers and not yet committed agents only need to redirect their attention without incurring any loss. The distinction between committed and noncommitted agents comes with an important caveat. The higher threshold provision concerning adverse evidence does not apply uniformly to all committed agents. Those operating under conditions of extreme resource scarcity are excluded from it. What rationality requires from them is to continue handling contrary evidence as noncommitted agents or mere observers would. This modified requirement should also apply to akratic agents. As the preceding discussion shows, they fully satisfy the conditions of extreme resource scarcity set out by the Evidential Threshold Account. In many cases, as in the stories of Amy Parker and Peter Phillips, this would include reduced external resources, such as unsupportive environment and financial strain. However, as Mr Miller’s story illustrates, even in cases where akratic agents do not suffer from extreme external scarcity, they would still be subject to extreme internal scarcity generated by their own record of necessarily less-than-successful conduct. And as we pointed out earlier, this conduct demonstrates perseverance of the wrong kind. So, it would seem natural to conclude that epistemic resilience would only hinder akratic agents who better stay alert to contrary evidence as rational, noncommitted agents would. In spite of its initial appeal, this possible conclusion is open to a challenge. Assuming a normative understanding of rationality, there should be a point to being rational. Things should go better for agents who heed what rationality requires, not worse. The underlying expectation covers not only what we might term steadily rational exercises of agency. It is even more to the point when agents have strayed away and are now trying to get back on track. As Broome (2007, 365) observes: ‘Just because you are in an irrational state, that does not mean rationality can be expected to impose conflicting requirements on you. You should expect rationality to require you to get out of your irrational state, not to get in deeper, into … further irrationality’. Turning to akratic agents, this normative expectation looks like the following: just because they have a record of necessarily less-than-successful conduct, thus operating under (at least internal but probably also external) resource scarcity, that does not mean rationality can be expected to impose conflicting requirements upon them. They should expect rationality to require them to get out of their irrational state of akrasia, not to get in deeper, into further irrationality. The requirement to forgo grit, however, does seem to require akratic agents to get in deeper into further irrationality. As highlighted earlier, Morton and Paul (2019, 202) admit that grit’s unsuitability to the pursuit of difficult goals

162 Lubomira Radoilska

under extreme resource scarcity is a distressing finding. In light of the present discussion, this admission points to a normative conflict that the Evidential Threshold Account ultimately imposes on committed agents that meet with a lot of adversity. For it asks such agents to keep low evidential threshold when deciding whether to give up on demanding goals that their context has made even harder to achieve. Looking at instances of akrasia, this implicit contradiction cannot be adequately explained as a tension that results from the irrational state in which the affected agents find themselves, independently of this Account’s requirement on how they should handle contrary evidence. Instead, the contradiction is introduced by the requirement itself: as your evidence points out, you are in a desperate situation, so do not try to protect yourself from despair! To see why grit is, after all, rational for akratic agents, let us look back at the overall shape relevant exercises of agency take across the three personal stories. In each story, we witness the undertaking of a difficult longterm o ­ bjective – o ­ vercoming one’s necessarily less-than-successful conduct that has become a clear threat either to specific core commitments (in the stories of Mr Miller and Amy Parker) or to getting on with one’s life overall (in the story of Peter Phillips). This undertaking, however, ends up uniformly in failure, in spite of some initial success. How is this shape of agency to be explained? Is it plausible to submit that things go wrong for the akratic agents because of excessive, viz. irrational grit? Such an explanation might seem supported by the impact rationalization has on akratic agents’ self-awareness as conflicted agents. As argued earlier, such agents tend to re-describe their necessarily less-than-successful conduct in terms of self-standing actions aimed at particular outcomes. These redescriptions make the akratic conflict less conspicuous and more manageable from the agents’ own perspective. At the same time, however, they entrench akratic conduct further. For, the inverse correlation between unsuccessful assertion and successful production that underpins it does not come to the agents’ attention. Clearly, the akratic handling of evidence here is wanting. What is less clear is that the shortcoming results from inappropriate epistemic resilience. For there are no traces of delusional optimism in the personal stories. I take this observation to be self-evident regarding Amy Parker and Peter Phillips as neither overestimates their chances of success at any point. A strong case can also be made regarding Mr Miller, even though he takes himself to be strong-willed enough to cut down on alcohol without medical help. With hindsight, his self-assessment is proven wrong; however, there is no reason to attribute this error to irrational grit: contrary evidence, such as frequent relapses, has not yet become available. What is more, as a high-achiever, Mr Miller has a lot of evidence to support his self-belief. If undue epistemic resilience is not the source of trouble for these akratic agents, how are we then to account for their failures to achieve the goals to which they have committed? Here is a possible alternative: akratic agents tend



Is Grit Irrational for Akratic Agents? 163

to fail because of their handling of evidence; in particular, they keep applying low threshold to contrary evidence after commitment, as the Evidential Threshold Account requires from them. As the following discussion will aim to demonstrate, this approach is epistemically irrational since it takes evidential content out of its original context; it is also practically irrational since it paints one’s own project as ineffectual trying and, as a result, promotes selffulfilling despair. The agents in all three personal stories have a feature in common: they keep a close eye on their own less-than-successful performances, both past and present. These are treated not just as indicative evidence of how they are likely to perform in the future. In fact, this evidence is accorded high predictive value and treated as conclusive. This shared pattern of contrary evidence handling is apparent in the overwhelming effect that first significant setbacks have in the stories of Mr Miller and Amy Parker. Mr Miller reverts to uncontrolled heavy drinking once it becomes clear that his life after retirement will not be plain sailing. Amy resumes her use of illicit drugs soon after she discovers that looking after a baby is much harder than she anticipated. Peter Phillips’s story is even more to the point. He never attempts to stop drinking outside of the clinic: as bad as it is, drinking has so far helped him resist the urge of committing suicide. Peter’s evidence handling is fully consistent with what the Evidential Threshold Account recommends for agents who, like him, operate under conditions of extreme resource scarcity. Yet, it amounts to sinking deeper into further irrationality rather than addressing the underlying akratic conflict. Following this line of thought, we can now clearly see why akratic rationalizations derive from giving too much credence to contrary evidence rather than ignoring or underestimating it. Peter re-describes his routine alcohol intake as a means to an end, a straightforward action as production, not a conflicted exercise of agency. Nevertheless, he acknowledges it as a means of last resort. But seeing oneself as someone with no alternatives can only support perseverance of the wrong kind, that of necessarily less-than-successful conduct. It might be objected that, albeit distressing, Peter’s conclusion is not epistemically irrational: he is indeed in a bad way. His conclusion, however, goes further than this admittedly accurate assessment: there is no other way to keep going than drinking oneself unconscious. This over-pessimistic account is the upshot of keeping one’s threshold for contrary evidence low when operating under conditions of extreme resource scarcity. For such a policy contributes to a pair of harmful (and irrational) reasoning habits: the first is to focus on evidence that is partial, inconclusive and context-sensitive in the wrong way; the second is to ignore one’s position as agent when assessing evidence relative to one’s own actions. Let us explore each habit in turn. Keeping a close eye on evidence that one is unlikely to succeed has an obvious flipside: being less sensitive to evidence in one’s favour. For this low evidential threshold is not meant to be applied uniformly but to capture putative

164 Lubomira Radoilska

contrary evidence. As a consequence, supportive evidence is acknowledged as such at a higher threshold. Moreover, in all cases, the agents are looking at probable future states of affairs, not facts. So, no matter how strong adverse evidence they encounter, this evidence is, by definition, inconclusive. This point is unlikely to register with them as long as they stick to a low threshold for contrary evidence which, of course, abounds under conditions of extreme scarcity. The underlying reasoning habit also affects how evidence that registers is assessed. It is not calibrated to the task that akratic agents are effectively engaged in. To appreciate what the stakes are, let us briefly consider an example from the literature on skill. In an illuminating paper, Hawley (2003) sketches the following scenario: to escape from ruthless assailants, a person has to manage to drive away on a frozen lake while under heavy fire. The question as to whether she can drive in these action-movie circumstances is categorically different from that whether she can drive tout court. It would be misleading to think of driving as one and the same task performed in different circumstances. For such an understanding is likely to lead us to underestimate the gap between the standard of success in the extraordinary case as opposed to that in the ordinary one. For instance, failure to perform well in the former is no evidence against this person’s likelihood to succeed in the latter. Returning to the issue of contrary evidence handling in the context of akrasia, there is a valuable lesson we can learn from the preceding thought experiment. Akratic agents treat the projects they have committed to as difficult only in the sense of their own poor prospects to succeed. In so doing, they end up assessing their performances as though failing at another, much easier (and irrelevant) task, such as refraining from alcohol when things go well, in the absence of alcohol dependence. In contrast, if what they are trying to achieve were to be considered in its original context, missteps would be judged less harshly and their predictive value more readily questioned. Let us now move to the second and related reasoning habit: ignoring one’s position when handling evidence of the odds of being successful as an agent. There is something distinctly disempowering in thinking about one’s own commitments primarily as a matter to settle by weighing up evidence, like an impartial observer would. For unlike observers, agents have the power to influence the relevant course of events: they make things happen in virtue of their acting. So, when considering how projects of theirs might unfold, agents are not confined to the available evidence that speaks in favour of or against a happy end.7 To require them to inhabit such a detached perspective, on pain of irrationality, amounts to undoing their standing as agents. At the very least, it generates the conditions for split-up, if not conflicted, exercises of agency. If I am trying to get something done while looking for signs that I would better give up, my trying cannot be wholehearted. And when the project that I am engaged in is difficult anyway, my keeping watch for contrary evidence is likely to turn what I am doing into just trying. But seeing myself trying and never

Is Grit Irrational for Akratic Agents? 165



succeeding is likely to push me toward more predictable undertakings.8 In this context, necessarily less-than-successful conduct could offer such an agent much-needed certainty.

Concluding Remarks In this paper, I argued that akrasia is best understood as necessarily lessthan-successful conduct. Akratic actions are successful as productions to the extent that they are unsuccessful as assertions. By treating these actions as parts of a longer-term process of rationalization instead of paradoxical one-off events, it becomes possible to investigate the epistemic side of the phenomenon along with its better researched motivational side. Reflecting on three personal stories, I tested the initially plausible claim that grit or epistemic resilience might not be rational for akratic agents ­since – ­as a result of their ­akrasia – t­ hey should remain alert to contrary evidence that difficult projects of theirs are unlikely to succeed. In the light of a normative understanding of rationality according to which being rational should benefit agents, it transpired, however, that refraining from grit would only make akratic agents worse off. This is because the rationalizations that underpin their necessarily less-than-successful conduct already overexpose them to contrary evidence. A requirement to stay focused on such evidence when pursuing difficult goals can only amplify the disempowering effects of rationalization.

Acknowledgements I would like to thank the editors of this collection, Patrick McKearney and Nicholas Evans for their helpful comments on earlier drafts. I would also like to acknowledge the constructive feedback I received from audiences at LSE and the Universities of Kent and Johannesburg. The research on this chapter was supported by a British Academy Newton Award, Ref.: NAFR1180082. Lubomira Radoilska is a Professor of Philosophy at the University of Kent, UK, and Editor-in-Chief of Ethical Theory and Moral Practice. She is the author of Addiction and Weakness of Will (OUP, 2013), L’Actualité d’Aristote en morale (PUF, 2007) and multiple papers on topics including the norms of action and belief, reasons, rationality and responsibility.

Notes   1. In this respect, there is a superficial resemblance between akratic conduct and skilled agency, viz. secure competence. See, for instance, Raz (2012, p. 2012): ‘We acquire and are aware of having a sphere of secure competence, consisting of a range of actions that,

166 Lubomira Radoilska

in normal circumstances, we reliably expect that we shall successfully perform if we set out to perform them, barring competence-defeating events (which are very rare)’. For, the reliability that comes with akratic conduct is not that of success but that of not ending up with nothing.   2. See Pollard (2010) for an account of habit as a complex disposition across various manifestations of agency. Douskos (2017) adds on an insightful contrast between such an understanding of habit and existing notions of skill.   3. I speak of one-dimensional models in the plural as a pernicious rationalization of the kind described here may also be supported by an underlying understanding of action as assertion only rather than production as the case is in Mr Miller’s story. To see how this might work, let us briefly consider Amy Parker’s story: Amy’s lack of self-awareness as a conflicted agent is grounded in her sense of whole-hearted commitment to be a good mother and to give her baby the best chance in life (normative-cum-motivating reason). Akratic incentives (explanatory reasons) are then dismissed as (normatively) insignificant, which in turn makes their enduring efficacy even more perplexing from the point of the akratic agent. And since Amy continues to operate with a one-dimensional model of action as assertion, she assumes that the issue lies with the very object of her commitment: looking after her baby is experienced as an exhausting and exasperating activity.   4. The standard account of akrasia is sometimes presented as a theoretical offshoot from Ancient Greek treatments of the topic, in particular that by Aristotle in Nicomachean Ethics, Book 7 (2001). Yet, the relationship is far more complex, with several points of contrast; for instance, unlike Ancient Greek discussions of akrasia, contemporary counterparts tend to see the issue as one of rationality rather than morality. Charles (1984) offers an illuminating analysis of Aristotle’s theory of action in comparison to that of Donald Davidson, which until recently dominated the contemporary debate. See also Hoffmann (2008), where major views on akrasia throughout the history of Western philosophy are critically explored in their original intellectual contexts.   5. See, however, Radoilska (2013).   6. See Holton (2009) for an alternative diachronic conception of the phenomenon. Its irrationality is interpreted in terms of irresoluteness instead of perseverance of the wrong kind: akratic agents change their minds too readily. This conception helps solve some of the difficulties faced by the standard synchronic account. However, it still prioritizes the motivational dimension of the phenomenon at the expense of its epistemic significance.   7. See Radoilska (2017) for an argument in favour of treating beliefs about one’s own actions as a special case where the truth norm can be satisfied in two ways: not only by following the available evidence that something is or is not the case but also by making that something is or is not the case through the exercise of one’s own agency. For instance, I typically do not need to consider any evidence to conclude (correctly) that I will fulfil a promise of mine. My commitment to do so suffices as long as what I have promised to do resides within the sphere of my secure competence (see note ii above).   8. When just trying, one can get lucky but not really succeed. See Yaffe (2010, Ch.9) on the distinction between inadequate and impossible attempts. Applying this distinction, we could say that the attitude of ‘just trying’ turns relevant actions into impossible rather than merely inadequate attempts.



Is Grit Irrational for Akratic Agents? 167

References Alvarez, Maria. 2013. Kinds of Reasons: An Essay in the Philosophy of Action. Oxford: Oxford University Press. Aristotle. 2001. Nicomachean Ethics, edited and translated by Roger Crisp. Cambridge: Cambridge University Press. Bradford, Gwen. 2015. Achievement. Oxford: Oxford University Press. Broome, John. 2007. ‘Wide or Narrow Scope Commitment?’, Mind 116(462): 359–70. Charles, David. 1984. Aristotle’s Philosophy of Action. London: Duckworth. Dancy, Jonathan. 2000. Practical Reality. Clarendon Press: Oxford. Davidson, Donald. 1970/2001. ‘How Is Weakness of the Will Possible?’. In Essays on Actions and Events, edited by Donald Davidson, pp. 21–42. Oxford: Clarendon Press. Douskos, Christos. 2017. ‘Pollard on Habits of Action’, International Journal of Philosophical Studies 25(4): 504–524. Carlisle, Clare. 2014. On Habit. London and New York: Routledge. Hawley, Katherine. 2003. ‘Success and Knowledge-How’. American Philosophical Quarterly 40(1): 19–31. Hoffmann, Tobias (ed.). 2008. Weakness of Will from Plato to the Present. Washington, DC: The Catholic University of America Press. Holton, Richard. 2009. Willing, Wanting, Waiting. Oxford: Oxford University Press. Lester, Rebecca J. 2023. ‘Trigger Warnings: Danger, Desire, and Declensions of the Will in Eating Disorders Treatment’. In Against Better Judgment: Akrasia in Anthropological Perspective, edited by Patrick McKearney and Nicholas H.A. Evans. Oxford: Berghahn Books. McKearney, Patrick. 2023. ‘Relational Akrasia: Care and the Distribution of Action’. In Against Better Judgment: Akrasia in Anthropological Perspective, edited by Patrick McKearney and Nicholas H.A. Evans. Oxford: Berghahn Books. McKearney, Patrick and Nicholas H.A. Evans. 2023. ‘Against Better Judgment: An Anthropology of Akrasia’. In Against Better Judgment: Akrasia in Anthropological Perspective, edited by Patrick McKearney and Nicholas H.A. Evans. Oxford: Berghahn Books. Mele, Alfred R. 1987. Irrationality: An Essay on Akrasia, Self-Deception and SelfControl. New York: Oxford University Press. Mezzenzana, Francesca. 2023. ‘To Live Like ‘People’: Drinking and Weakness of Will Among the Runa of the Ecuadorian Amazon’. In Against Better Judgment: Akrasia in Anthropological Perspective, edited by Patrick McKearney and Nicholas H.A. Evans. Oxford: Berghahn Books. Morton, Jennifer M., and Sarah K. Paul. 2019. ‘Grit’. Ethics 129: 175–203. Pollard, Bill. 2010. ‘Habitual Actions’. In A Companion to the Philosophy of Action, edited by Timothy O’Connor and Constantine Sandis, pp. 74–81. Oxford: WilleyBlackwell. Radoilska, Lubomira. 2013. Addiction and Weakness of Will. Oxford: Oxford University Press. . 2017. ‘Aiming at the Truth and Aiming at Success’, Philosophical Explorations 20(sup1): 111–26. Radoilska, Lubomira, and Keron Fletcher. 2016. ‘Lessons from Akrasia in Substance Misuse: A Clinicophilosophical Discussion’, BJPsych Advances 22(4): 234–41. Raz, Joseph. 2012. ‘Agency and Luck’. In Luck, Value and Commitment: Themes from

168 Lubomira Radoilska

the Ethics of Bernard Williams, edited by Ulrike Heuer and Gerald Lang, pp. 133– 61. Oxford: Oxford University Press. Schwitzgebel, Eric, and Jonathan Ellis, J. 2017. ‘Rationalization in Moral and Philosophical Thought’. In Moral Inferences, edited by Jean-François Bonnefon and Bastien Trémolière, pp. 170–90. Routledge, New York. Steele, Claude, M. and Joshua Aronson. 1995. ‘Stereotype Threat and the Intellectual Test Performance of African-Americans’. Journal of Personality and Social Psychology 69: 797–811. Steele, Claude M. 2010. Whistling Vivaldi: How Stereotypes Affect Us and What We Can Do. New York: W.W. Norton. Weinberg, Darin. 2023. ‘Three Problems with the Addiction as Akrasia Thesis that Ethnography Can Solve’. In Against Better Judgment: Akrasia in Anthropological Perspective, edited by Patrick McKearney and Nicholas H.A. Evans. Oxford: Berghahn Books. Yaffe, Gedeon. 2010. Attempts in the Philosophy of Action and the Criminal Law. New York: Oxford University Press.

8 Relational Akrasia Care and the Distribution of Action Patrick McKearney

It was a clear autumnal morning one Saturday in October, and I had been supporting Bob to get up according to his weekend routine. A few months prior, Bob’s principal carer, Peter, had taught me how best to help Bob in the mornings: put the peanut butter on his toast like this, run his bath like that, talk to him like so. And I had followed his advice to the letter. But Bob still did not seem as comfortable with me as he did with Peter, by a long way. While Peter sailed through the mornings with Bob happily chatting, Bob was often sullen while I supported him, not wanting to engage in conversation, and displeased with many things I did. Bob is fifty-five years old and has cognitive disabilities significant enough to place him in need of care throughout his life. To be more specific, Bob has been diagnosed as having a high degree of autism. He fares well in the security of daily rhythms, familiar ways of talking, and material objects such as his daily newspaper that he carries at all times. Bob’s doctor and social worker also think he has what in the UK is termed a ‘learning disability’: a lifelong cognitive impairment significant enough to affect his capacity to do daily tasks and so to get by on his own.1 It is for that reason that Bob’s local authority pays an NGO to provide his accommodation, upkeep and care in a group home for five people with similar disabilities in a large British town. I had arrived to do research on this organization a few months before this day in October, by taking up a job as a full-time carer in Bob’s home, a position that I would hold for a year. I worked alongside a team of ten carers who rotated on shifts to provide support for Bob and his housemates during mornings, evenings and at weekends. About half of that team had, like me, started the job that calendar year. We were trained by more experienced carers and managers in both formal

170 Patrick McKearney

and informal ways. As part of my induction into the daily rhythms of the job, one of Peter’s key bits of advice to me was not to talk to Bob too much in the mornings, and especially not to ask him questions. This was easy for Peter as naturally quite a calm and self-assured person. I, by contrast, found the feeling that Bob simply did not like me so uncomfortable that I was repeatedly overcome by the pathological urge to chat to Bob to try to win him over. I had been doing well to resist that urge that cold Saturday morning, and perhaps that was also why Bob seemed a little better disposed towards me. But this discipline of silence went against not only my nervous disposition, but also some of my obligations as a carer. Changes to legislation governing professional caring relationships in Britain now prohibit paternalism by making it mandatory for care givers to provide support in line with the care recipient’s wishes (for a summary, see Welshman and Walmsley 2006). The training I received in a mandatory government course online, and in formal sessions run by the organisation, informed me that I had to offer people with learning disabilities choices for their care. In this context, asking someone what they want is a very good way to care for them. Indeed, it is essential in situations where the carer does not know what their wishes are, but must act on their behalf nevertheless. One cannot simply decide for the care-recipient, one must offer them the opportunity to choose for themselves. As Bob sat down in front of the TV that morning, this felt like one of those occasions. Bob cannot put DVDs on himself, but he normally loves watching the sitcom Friends on a Saturday morning. Not asking Bob what he wanted in this situation risked going against Bob’s wishes. I might put the DVD on when he would rather sit quietly; or I might not put it on when he actually wanted to watch it. So, I simply asked him: ‘Bob do you want to watch Frie—’ Before I had even managed to finish the sentence, Bob responded with a loud and affirmative ‘Aaalright’, and then proceeded to talk excitedly about the show in his rhythmic, lyrical tones: ‘Right, I’m pretty sure it’s the first episode now. Either the last episode, or the first episode of Friends, or both. Friends on a Saturday. My gosh.’ Success. Bob and I were talking. He had what he wanted. I had a clear choice from him to guide my caring actions. At this point, however, I realized a fatal flaw in my plan. I had forgotten to get Bob to clean his room and change his ­sheets – ­something he is meant to do every Saturday morning. This is, again, in line with current care policy. Carers are there to assist people to live their lives, and to help them increase their capacities for everyday tasks. Carers are not supposed to do everything for their charges. Bob can push a hoover around his room. He will begrudgingly do this at the right point in his routine; that is, just after he brushes teeth, before the carer takes over to finish the job. The problem was that it was, by now, way past that point in the routine. And so, when I told him he needed to clean his room, he ran angrily upstairs before I had a chance to follow him. He returned quickly and very unhappy indeed. As he sat there silently, I asked



Relational Akrasia 171

him once again if I should put the DVD on for him. He replied, ‘No. I hate Friends. Get it out of here. Friends is on fire, seriously. I’ve had enough now. This is a joke.’ What was going on? Perhaps Bob’s mood had simply changed, and so he didn’t want to watch Friends anymore. Or perhaps his decision to watch Friends had not changed and he was simply talking in this way to express his anger; maybe he was not confusedly cursing Friends, but very clear-mindedly cursing me. Or maybe he was not communicating what his judgment was, but ‘only talking’ (Aristotle in Davidson 2001, 28). Or could something more complex have happened to him as a decision-making subject? Maybe he now had two intentions, desires and even judgments; one that watching Friends would be an excellent thing that he would love to do, the other that it would be an unpleasant, scary or poor choice. Perhaps this was even a clear-cut example of akrasia in which Bob’s judgment to watch Friends had somehow failed to translate into the action of actually instructing me to put it on, and perhaps he had intentionally instructed me to do something else entirely. Or does Bob lack the cognitive ability to do the kind of evaluations of his different desires necessary to make such a judgment in the first place? I suspect that you are already beginning ‘to play the amateur psychologist’ (Davidson 2001, 28) and form an opinion, in one way or another, about Bob’s mental state. But academics are not the only ones to have such opinions. Carers also engage in these deliberations about the relationship between Bob’s judgments, will and actions in order to decide what practically to do (see also McKearney 2020). In this paper, I step back from giving my own answer as to what is going on inside Bob in order to attend to how questions around akrasia become practically significant for carers. In particular, I ask an ethnographic question: why does arriving at the right understanding of Bob’s psychological state matter so much to them? I demonstrate that Bob’s care is premised upon two competing ethical worries: that he cannot always direct his own actions, and that he ought to be able to do so (McKearney forthcoming; see also Redley and Weinberg 2007; Redley 2018). I show that this particular way of responding to dependence aims to create a direct link between Bob’s choices and his carer’s actions in order to extend the reach of Bob’s selfhood into the world. But I also demonstrate that this link, extended as it is across two people, is always fragile. It breaks down, most notably, when Bob’s intentions are not clear. It is hardest for carers to know how to act on Bob’s behalf when his different desires seem inconsistent and incompatible. When carers offer psychological interpretations of Bob’s inner state, it is in in an effort to repair that connection, to find a way to reconcile his desires so that they can act in line with his intentions, and to continue to enlarge his capacity for action. In the course of proposing this answer to the question as to why Bob’s carers are so focused on his mental state, I explore whether akrasia and weakness of will can exist between people. An aside: the distinction between the

172 Patrick McKearney

two concepts is not especially important for my purposes as I draw freely on both Donald Davidson’s and Richard Holton’s influential treatments of the concepts. Indeed, I use diverse sources in the akrasia debate as an analytical resource with which to analyse my ethnographic m ­ aterial – a­ nd thus cite from these authors as a way of developing these kind of connections. To continue: both akrasia and weakness of will are concerned with the relationship between diverse parts of the self, and between that self and its expression in the world. They share a concern with the fragility of the self and its attempts to extend its influence in action. I contend that Bob’s caring relationship responds to a similar predicament: of how Bob’s different parts are put together such that his intentions can translate efficaciously into reality. But in this care, the concern is distributed across two human subjects rather than contained within one. Akrasia exists when there is meant to be a relationship between judgment and action that does not, in fact, become realized. Similarly, I maintain that this kind of care is meant to translate Bob’s judgments into actions but regularly fails to do so. Does that mean that there is such a thing as ‘relational akrasia’? I take this kind of care as a striking example of a more general possibility: that social relationships can distribute a judgment and an action across subjects, fragmenting the relationship of identity, presupposed within Socrates’ thought, between the judge and the actor. This chapter thus offers a way of seeing social relationships as routinely playing with precisely the features of subjectivity that are at stake in the akrasia debate. To develop these analogies between the akrasia debate and social relationships more widely, I compare British care to two different ideal types that represent alternative ways in which such relationships of dependence can be constructed. I use philosophical work on akrasia and weakness of will to analyse these diverse ways of constructing relationships between people, and between their intentions, judgments and actions. Doing so helps me to show how other ways of distributing responsibility for dependence might close down the possibility of relational akrasia, and how the British one very much opens it up. This chapter thus raises a further question as to whether it is only in certain types of social relations that akrasia becomes a psychological possibility, a moral problem and an analytical puzzle in the first place. Is individual akrasia possible only in certain kinds of social relations?

Going out Bob lives with Martha, a thirty-five-year-old woman who also has a learning disability and normally sits with Bob while he watches TV. In the idiosyncratic monosyllables that are her main form of communication, Martha pleads with carers throughout the week to get them to take her to a pu (pub, restaurant or café) to get kee: cake and a cup of tea. But when it comes to actually leaving the house on a Saturday morning, Martha frequently does not do so.



Relational Akrasia 173

Instead, when a carer tells her that it is time to go, she often sits on the sofa in the sitting room with her head in her hands. Sometimes she is more defiant, saying ‘No!’ and slapping the air in front of her. At others she seems more upset, crying into her hands and looking up at her carer imploringly. On some occasions, Martha finds a way to overcome this situation and leave the house. But even when she does so, and is some way into the journey, she might start to say hom (home) and tug on her carer’s arm to drag them back to the house. And sometimes in these moments she hits them, square in the chest. Speculation abounds about what is going on in these moments among those involved in Martha’s life (McKearney 2021). Her mother, for instance, says that what Martha needs from those responsible for her in these moments is confidence and reassurance. The sense that Martha wants to stay inside or go home is likely to dissipate when you do not give in to her but instead make it clear that you are in charge and will keep her safe. But she also thinks that Martha is ambivalent about situations in which she is under another’s authority. For instance, carers try to persuade Martha to bathe each morning, but she often refuses to do so, or hits them when she has got in the ­bath – ­in a move that her mother interprets as Martha trying to assert power in a situation where Martha feels she lacks it. Many of Martha’s carers also put forward the idea that these conflicts stem from what they call an ‘attachment’ to certain people. When Martha’s current favourite carer is supporting her, she will typically bathe or leave the house with them with little fuss. I was, for some months, that person to Martha and not only left the house with her with ease, but also took her on long trips around the city and even on a weekend to Cambridge. All, for the most part, without incident. I was quite happy to interpret this as evidence that, unlike Bob, who was clearly a poor judge of character, Martha had apprehended my hidden caring abilities and considerable personal charm. Others carers did not. They told me that this kind of intense relationship always comes with a distrust of other carers. On that same trip to Cambridge, Martha tried to pull out some of my partner’s hair, and during this whole period, she very rarely left the house with other carers with anything approaching the ease that she managed with me. Carers did not interpret Martha’s attachment to me as a decision, but rather as emotionally overwhelming her capacity to make choices. With them, she still wanted to leave the h ­ ouse – ­but she was so afraid of being supported by someone else that she became unable actually to act or decide upon this desire. This should not be given into, but challenged, so as to enable Martha to return to her senses and to a greater sense of freedom. Her sister, though, thought that Martha’s preference for some carers over others was evidence that she had a healthy respect for her own body and who got close to it. She maintained that Martha was most definitely making a choice for some people over others, and one that should be respected. Martha was, then, forming desires and preferences in a perfectly adequate way (after

174 Patrick McKearney

all, what kind of deliberation does one need to do in this regard other than have a preference for who one likes and who one does not?). There was no conflict within Martha to be solved through some kind of emotional pedagogy; there was only a conflict between her carers’ obligations to respect her wellformed decisions and their responsibility to protect her from harm. When Bob is at the pub eating a burger, or at his sister’s house with rice and chicken in front of him, he typically seems very pleased indeed and will continue to talk excitedly about these experiences throughout the week. Sometimes if you ask Bob on a Wednesday whether he wants to do either of these activities at the weekend, he will say ‘Yes’. But when it comes to Saturday and Bob is already watching Friends, if you ask him again, he will likely say ‘No!’ in a distressed tone of voice. And he will also run off to the cloakroom, throw some coats on the floor, and stand by the front door bouncing up and down, and often cursing. He will normally leave the house with you, but he will often be very angry as he does so. Bob has, for a very long time, had a complicated relationship with leaving the house. When he moved into the care home, he had a stronger reaction to going out: typically ripping down all the wall-hangings in the corridor whenever anyone tried to make him leave. This was in the heyday of de-institutionalization, when his carers were desperate to give him the kind of life people with disabilities had been denied by society until that point. And so, they did not take Bob’s resistance as a ‘decision’ that should constrain them, but rather as a sign of how confined he had been. They thus tried to get him out of the house as much as possible. An experienced carer in an organization called Marla told me that they eventually came to think they should listen to Bob’s desires more. They stopped trying to force him out of the house quite so much. Having previously refused, they bought a television to enable him to remain inside more happily. Marla continued to interpret Bob’s ambivalence about going out in line with this change. She felt that Bob’s displays of anger were an expression of a clear decision to stay in, one that needed to be taken seriously. She did not settle on a possible, simpler, interpretation that watching TV was his only desire: and that, therefore, it was clear that carers should let him stay inside. She thought, instead, that Bob had two desires and that he could not work out how to put them together. But Bob’s principle carer, Peter, had a different approach. He felt that Bob’s aversion to going out did not indicate the presence of a decision, but the absence of the capacity to take one. He saw Bob as deeply afraid of going out. He described Bob’s desire to watch the TV as another kind of ‘attachment’. Peter saw it as a comfort blanket that renders him ‘psychologically unable’ to choose, rather than a freely made choice in and of itself (Davidson 2001, 27; see also Williams 1993; Laidlaw 2013). He felt that Bob’s anxiety had the capacity to ‘distort judgment’ and so prevent Bob from landing upon a ‘fully formed’ desire, intention and decision (Davidson 2001, 28). Bob’s anxiety ‘fetters the

Relational Akrasia 175



reason’ (Aquinas in Davidson 2001, 33), curtailing Bob’s ability even to ‘opt for pleasure’ (Davidson 2001, 30). Bob, Peter believed, could only manage to ‘opt’ in this way under pressure from a carer. So, Peter endeavoured to change this situation by forcing Bob out the house to do activities he wanted, in the hope that this would eventually make Bob less anxious and less attached to the ­T V – ­and therefore able to choose more freely.

Care Why do carers in this organization reflect so much on the seemingly technical, philosophical question of precisely what was going on in Martha’s and Bob’s minds at these moments? One answer to that question would draw attention to Bob’s and Martha’s cognitive impairments, which might seem in and of themselves to draw attention to the presence or absence of the capacity to choose. Bob and Martha’s lives invite us to think about what kinds of mental processes are necessary to arrive at a judgment of the best thing to do. But they also raise questions about how important such judgments a­ re – ­or, more specifically, what kinds of social relations render such judgments important and which do not. I argue that the close psychological scrutiny and ethical concern that Bob and Martha receive in the UK is not universal by considering two alternative ways in which care can be conducted in relation to this question. These examples are not, by virtue of the relatively minor quantity of ethnographic information I have been able to collate about them, fully developed anthropological descriptions. They are, instead, impressionistic ethnographic portraits of ideal types. I use them to suggest alternative possibilities, rather than to focus on the (undoubtedly present) complexities of what they involve in ethnographic practice. The evidence I have to hand, however, suggests them to be real possibilities rather than merely thought experiments. And it is for that reason I have chosen them, and for which I shall be very glad when we have more data on these examples and other comparable ones. I conducted eight months of fieldwork on adults with intellectual disabilities in the Indian state of Kerala. Most of my work was in the context of families, and in particular, on the parents who typically provide care to adults with intellectual disabilities in this context. Compared to carers in the UK, the parental caregivers I worked with were typically far less interested in whether or not their children’s decisions were the result of attachment, fear or careful deliberation. And that seems to be, in part, because they take it as their responsibility to take decisions on behalf of their offspring, even when the latter are upwards of 30 years old. They manage and direct aspects of their conduct that many British parents would not dream of intervening in: whether or not they should be married, what they eat and drink, and even when they stand and sit. They literally pull and push their children to get them to be

176 Patrick McKearney

seated or on their feet. In this context, what a person with disabilities intends has very little effect on what their carers do to them. This is, I suspect, why precisely what is going on in the mind of such a person is not nearly so important as in the UK: because it has far fewer practical implications than in the British context where such a person’s will is meant to govern the caring actions of others (see also Mody 2020b, 2020a). It simply does not matter whether that person’s will is, in any particular moment, in one state or another, because that will not change who is actually taking the d ­ ecisions – a­ nd it will very rarely be the person with disabilities themselves. The second example comes from Francesca Mezzenzana’s work on the Runa of the Ecuadorian Amazon (Mezzenzana 2020; see also Mezzenzana this volume and McKearney 2021). Mezzenzana describes how the Runa regard everyone as having individual wills (munay) that are not to be interfered with. They do distinguish between the degrees to which individuals have cultivated the capacity to think (yuyay) in a way that characterizes ­adulthood – ­and thus do not think that children, addicts or those with mental illnesses possess this to the same extent as others. But a judgment that another’s yuyay is not fully developed or present does not legitimize interference in their life. The Runa thus do not offer to such people the kind of care that Euro-Americans tend to think such situations require. They prioritize respecting the person’s autonomy instead, because everyone has munay regardless of the degree to which they possess yuyay. In this setting, another person’s psychological state has no consequences for one’s own actions. One continues to respect another’s freedom regardless: principally, by not interfering, whether that be through punishment, advice or care. In this context, a person manages their own l­ife – ­no one else will intervene to persuade them otherwise, to help them to do actions they are incapable of alone, or to prevent them from managing their own affairs as they decide. This ethical ‘commitment rules out as impossible’ any intervention in a person’s life contrary to their own ‘evaluative judgment’ about it (Stroud 2014, 10). The care that Bob and Martha receive in Britain is an uneven combination of these two approaches (Redley 2018). As in Kerala, Martha and Bob’s inability to take certain decisions are taken as cause to provide them with relationships of dependence. But those inabilities are not, as they are in Kerala, taken to lessen the importance of their intentions. In the UK, Bob and Martha are still meant to be in control of their own actions, just as they would be among the Runa. Put differently, Bob and Martha’s decisions are meant to direct, in a way foreign to the Keralan parents, a caring relationship whose very existence is foreign to the Runa. In this relationship, Bob and Martha’s decisions have a much more complicated connection to action. Others perform actions on their behalf, but that mediation is meant to be directed by their decisions. This helps us understand the extent of the scrutiny of their intentions and the questions about how rational they are being in any given moment. Such issues are essential to care in Britain in a way that they simply are not in



Relational Akrasia 177

the two other examples. In both of the other cases, the person with disabilities’ intentions are relatively unimportant for others’ actions: in South India, because those intentions need not be taken into account in caring for them, and among the Runa, because there is no scope for intervention in the first place. In Britain, by contrast, Bob and Martha’s decisions are taken to be ineffective, and so others are meant to act on their behalf to achieve those objectives. But the actions performed by others are meant to have a clear and direct relationship to those ­decisions – ­however difficult Martha and Bob find it to put them into action on their own. But these two ­imperatives – ­to act in place of another, but to act in line with that persons’ ­intentions – ­are famously difficult to reconcile (Redley and Weinberg 2007; Banks 2017; Levinson 2010; Davis 2012). In reality, the British system contains both the Runa and South Indian options. As among the Runa, when people are psychologically capable, one should still leave them to their own devices. When they are entirely incapable, one should intervene as in South India. But the British system also creates the possibility that there are people like Bob and Martha who combine these possibilities. They do not have the kind of mental capacities that would put others in a position to leave them alone. But they nevertheless have preferences and intentions that define the ethical status of others’ interventions into their life. This creates a particularly fraught relationship between the will and action in their ­care – ­in which the will is meant to connect up to action, even as it is meant also to be mediated by others. It is constantly unclear whether and how carers should intervene to make that possible (McKearney 2019, 2020). A carer’s intention to protect can result in coercion, and their nonintervention can easily turn into neglect. The only thing that can ensure their caring actions are immune to this kind of moral luck (Williams 1981) is understanding the exact status of Bob’s and Martha’s intentions in any particular moment, which is why there is so much emphasis on getting this right. The problem is that the caring system does nothing to help carers know that. In fact, it simply multiplies the ­possibilities – ­making it possible to draw all sorts of different and subtle boundaries between mental states that each have significant consequences for how one should act. The complex and varied psychological theories that carers develop are responses to this predicament of openness: attempts to work out ethical models that would tell them how to act appropriately. If, at any particular moment, Bob really knows what he wants in a straightforward way, one absolutely should act in line with it. But if it is really the case that Bob is akratic, caught between two judgments about what it is best to do, or overwhelmed by his emotions then there may be little clarity about what his will is, or whether it should be followed.

178 Patrick McKearney

Demanding Decisions Richard Holton argues that we ought to think about weakness of will in terms of the relation between an intention made with the aim of directing the actions of our future self and a departure from that intention at a later stage when we act in a contrary manner (Holton 1999, 241). These ‘Intentions’ (Holton capitalizes them to make clear they are a distinct kind of intention) are made in a ‘propitious moment to reason’ and are done with the aim of shoring up a certain course of action so that we can protect it against uncertainty, indecision and future temptation (Holton 1999, 243). In particular, they are often made with the explicit aim of countering a future psychic state in which we know we shall not want to undertake a course of action we currently judge best. Such intentions thus create a connection to our future self that offers a way for us to extend temporally certain parts of ourselves into upcoming situations where we are not sure they will e­ ndure – ­to make sure that our decisions become efficacious in the world. Indeed, they aim to ‘control’ our future self and to grant our self some stability over time (Bratman in Holton 1999, 242). They constitute, to put it differently, a kind of ‘intrapersonal’ relationship with ourselves (Holton 1999, 243). That is, in no small part, why cases of weakness of will are so interesting philosophically: because, in them, a posited unity between decision and action within the human subject breaks down. Martha and Bob are in care because they are judged to be incapable of acting on their own. Their care is meant to respond to the possibility that they cannot put their judgments, decisions and desires into action. The specific idea at work here is not just that they are dependent, nor even that they are physically incapable of performing a particular action. Rather, it is that they are not able to think about how to put their decisions into action, in such a way that their decisions would result in anything, and that they are not able to understand what decisions they need to take (McKearney forthcoming). Their care is based on a judgment that they do not know how they are dependent and thus do not choose in a way that reliably leads to them surviving and thriving. Such care thus takes on responsibility for arranging, and sometimes even performing, actions that other adults do for themselves. While Bob and Martha both have clear and obvious desires to eat, and to eat certain kinds of food, they do not understand well how to earn money, nor how to spend it on food, nor how to cook that ­food – ­nor even that all these actions are necessary to having food on their plate. So, carers both think about that necessity and perform those actions in order to make sure they eat. In the South Indian ideal type, Bob and Martha would eat and ­bathe  – ­and the fact that they might be judged to have difficulty understanding or performing those tasks would not affect this. But in such a system, Bob and Martha’s intentions and decisions are not meant to have any effect on those outcomes. In the British system, by contrast, Bob and Martha’s desires about  when  and  how  to eat are indeed supposed to exert influence over



Relational Akrasia 179

those actions performed on them and on their behalf. But in order for this to happen, Bob and Martha have to form something like one of Holton’s intentions. They have to do something that neither I have to do when I cook for myself, nor a recipient of South Indian care has to do in order to eat, and that is to make a decision and communicate i­ t – t­ hat is, to command someone else to do it. In Holton’s example, our intentions are authoritative commands to our future self. Bob and Martha need to command someone else. In each case, the current individual self does not straightforwardly perform the action; instead, it tries to extend itself through an order towards another (that is, a future self, or a separate person). Models of akrasia are premised upon a subject with certain kinds of mental capacities and are centred upon the subject’s relations with itself. Focusing on people who receive care for mental dependencies takes us into a murky territory of intersubjectivity in which wills and actions do not so obviously adhere to individuals, but can be distributed in more complex ways. We are so inclined to see Bob and Martha as simply lacking ‘normal’ ability that we are likely to pass over how specific the set-up of their care is. The philosophical tendency to assume that ‘all things considered’ judgments and intentions are a perfectly normal and healthy way for subjectivity to function does not help us here ­either – ­for it leads us to imagine that this is what nondisabled people are doing all the time. I am far from convinced, however, that the rest of us do need, most of the time, to make the kind of decisions, judgments and intentions that Bob and Martha have to in care. Think, for instance, of the difference between making something for lunch and deciding what to order in a restaurant. Making something for lunch does not require anything like a will or a judgment. One can have a reason to cook something, and end up eating that thing, without ever having sat down to consider all the possible options at one’s disposal. At a restaurant, however, one is required to have a definite ­choice – ­precisely because one must author this clearly in order for the waiter or waitress to bring you anything at all. I can cook steak while still being undecided if pasta would be better. But if I tell the waiter that I am undecided between having steak and having pasta, they cannot act on my behalf. The situation requires us to take an unusual kind of more global judgment in relation to ­food – ­that is, not simply to intend to eat, but rather to come up with an intention that another will then carry out for us. And that is precisely because we author the decision but not the action. We thus distribute to another something that we might do for ourselves. In a similar way, Bob and Martha must go an extra step that many of us do not have to perform. Wanting to have a cup of tea or an afternoon snack becomes a decision they need to land on and communicate to others, rather than an action they perform for themselves. If they are to eat when and what they want, they must communicate this decision to others. Indeed, it is a striking irony that the very care given to Bob and Martha because they are

180 Patrick McKearney

judged to have a difficulty with making judgments simultaneously asks them continuously to arrive at a particularly developed kind of judgment. It is as if they are always living in a restaurant, deciding what course of action to take. Their everyday life relies on a level of decision-making that most of us have to engage in only when we go to places like restaurants. Another way to put this is that it is as if they are always having to form and communicate intentions that are reliable and enduring enough expressions of their will that a carer can faithfully perform such actions on their behalf. They are like a person who is constantly concerned that their future self will not act on their behalf and so feels like they must form strong ­intentions – ­rather than just act as they see fit in each moment. This is why the caring relationship has so many parallels to cases of weakness of will: because a potential tight knit between decision and action is distributed across a less stable terrain, but nevertheless with the aim of maintaining the connection. When akrasia arises, ‘judgment and action … have diverged’ (Stroud 2014, 1) and there is a troubling indeterminacy between a person’s decisions and their own actions. People with mental disabilities constantly live with such a troubling indeterminacy and divergence precisely because they do not, within care, have that degree of control over many of their most intimate actions to begin with. The relationship between their intentions and their realization in the world is, as in weakness of will, always fragile. What does weakness of the will mean if your will does little to direct many of your actions in the first place?

Weakness of Will between People I have just described the restaurant and the care home as unusual situations that go against the grain of a normal unity between judgment, decision and action. But that may not be accurate. Anthropology has long challenged liberal  ideals of autonomy by demonstrating more complex relationships between decisions and actions; that is, distributions of these facets of subjectivity between individuals (Ferguson 2013; Mahmood 2012; Robbins 2007; Dumont 1980). It demonstrates that we routinely extend ourselves and our capacities for action through relationships not only with ourselves but also with others (see Holton 1999, 242–43; see also Strathern 1990; Lamb 2000; Busby 1997). Hierarchical commands are the most obvious example of this, in which one person’s intentions radiate into the bodily actions of others (Ferguson 2013). The social parallel of self-­control – ­the determined exercise of the will over one’s ­action – ­is the direction of some people’s actions by another’s will (Holton 1999, 249; see also Foucault 1992; Faubion 2011). Work on ritual has long shown how shaping social dynamics in certain ways can render participants in a ritual as actors while simultaneously taking away their status



Relational Akrasia 181

as decision-makers or intention-formers (Bloch 1974; Laidlaw and Humphrey 2006; Humphrey and Laidlaw 1994; Staal 1979). The actions one performs in ritual are often defined not by one’s own intentions, or even as the product of one’s own decisions (see also Aulino 2016, 2019). The hierarchical aspect of this reflects, to some extent, how care works in Kerala in that a parent’s decisions control a great deal of the actions of their child. In British care, by contrast, such control cannot manifest as c­ ommand – ­but often instead emerges in the form of carers attempting to persuade their charges (McKearney 2021). Instead, although we are not accustomed to seeing it this way, it is the care-recipients who are technically giving the commands: who are extending their wills through the actions of another (Kittay 2007). Bob’s and Martha’s capacities are taken to be impaired enough that they cannot extend their intentions into action by themselves, and thus require another to act for them. But Bob’s and Martha’s intentions are meant to govern the actions of their carers. Care in South India does nothing to change this fragility. Nor does the absence of care among the Runa. Neither way of organizing dependence makes a disabled person’s intentions more likely to translate into reality. But a relationship of care in Britain attempts to extend the reach and stability of their ­decisions – n ­ ot only temporally but also relationally. The carer aims to use their own body and mind to ensure that a person’s intentions are realized in a situation where it is uncertain. This kind of care is thus like one of Holton’s intentions: it is a way to amplify the stability and force of the decision-making self in the world. It aims to repair that ­connection – ­to give the intention the kind of force it should have (Holton 1999, 247). Here, the aim of carers is not to control others or the self, but for one’s actions to be controlled, governed, guided by another’s choices. For carers, a person with disabilities’ decisions have a distinctive character. They are not merely one possible course of action among others, just as my own individual ‘best judgment’ is not just one thought among many (see also Stasch 2008). Rather, they have a ‘special connection’ to the actions carers must ­perform – t­ hey constitute a ‘command or imperative’ that is ‘intended to guide conduct’ (Stroud 2014, 2). The answer to the carer’s question ‘What shall I do?’ lies in the judgment of the person with disabilities (Hare in Stroud 2014, 2). Just as the problem of weakness of will arises because intentions can be broken, so this kind of care is meant to be directed by the will of the person with disabilities but in reality is extremely vulnerable to going awry. In Holton’s example of intentions that aim to guide one’s own future self, decision and action are spread across two moments in time within the same subject. In the case of British care, decision and action are spread across two subjects. The relationship between decision and action breaks down, in Holton’s example, when the intention made at one point does not guide the future self. The formerly made intention does not speak loudly, persuasively or clearly enough. The current self does not hear or heed its call. The relationship between

182 Patrick McKearney

decision and action in Bob or Martha’s care can break down when a carer decides to ignore what they know Bob or Martha want. But it can also break down simply due to the difficulty of knowing what Bob or Martha intend, as if in a ritual, when one were unclear how the rite should go on. In the South Indian case, we find a disarticulation between a person with disabilities’ decisions and the caring actions performed for them. There is no normative connection, in this context, between what a person wills and what happens to t­ hem – ­no particular ‘ties between evaluative judgment and action’ (Mele in Stroud 2014, 5). Whether a person with disabilities ends up leaving the house, for instance, cannot be explained by what they themselves judge best. In the Runa case, by contrast, whether a person ends up leaving the house can only be explained by what they themselves ­do – ­no one will intervene. Here, people’s own actions are so independent of relationships of support that they cannot be motivated by another’s will. The South Indian system prioritizes the provision of caring actions over the decisions of the care-recipient, while the Runa system does the reverse. In each, the reach of a person with disabilities’ decisions are unchanged. But the British system attempts to combine these priorities by exploring the possibility that the care-recipient’s decisions might be translated into caring actions. Indeed, it makes this connection normative in the way that intentions do: it raises the decision to the position of what should happen to the point that an explanation will be needed if it does not indeed come to pass (Holton 1999, 247). But carers have not, as yet, found a reliable way to maintain that connection between decision and action with people who do not always author and communicate clear and consistent decisions. So, in the British case, there is a relationship between decision and action, but it is not wholly ‘stable’ (Bratman in Holton 1999, 242). The care recipient’s intentions are supposed to direct the carer’s actions but do not always do so. These different relational responses to dependency have analogies to the three major philosophical positions on the possibility of akrasia (Stroud 2014, 4–5). The South Indian case parallels the externalist position which holds that none of our actions can be explained by our decisions. When we look at what actually happens to a person with disabilities in this setting, it is only ‘in some cases’ that their ‘desire and belief caused an action, while in other cases’ their decisions ‘merely led’ to their ‘judgment that a course of action was desirable’ with no effect on what their carers actually do to them (Davidson 2001, 32). What a person with disabilities reasons ‘is best to do’ does not have the direct effect of ‘settling’ the carer’s ‘question of what to do’ (Bratman in Stroud 2014, 5). A person with disabilities’ reasons are ‘motivationally inert’ for the carer (Stroud 2014, 6 on Davidson). The Runa case, by contrast, parallels the internalist position in which what we end up doing is entirely identical with what we decide, judge best and want (Davidson 2001, 26; Stroud 2014, 3). To adapt Davidson’s paraphrasing of this position to the case of the Runa: ‘there is no distinguishing’ the independent ‘conditions under which an

Relational Akrasia 183



agent’ decides to act from the independent ‘conditions under which he acts’ (Davidson 2001, 32). The case of British care, meanwhile, directly parallels articulations of akrasia and weakness of will. In this form of care, unlike in the externalist position, a person’s decisions do have an explanatory and causal relationship to action. But, unlike in the internalist position, this relationship does not take the form of a direct, unmediated, and reliable connection.

Conclusion It is no coincidence that it is in the British form of care that the person with disabilities’ minds are subject to the most scrutiny. It is necessary for carers to understand precisely what Bob’s and Martha’s intentions are in order to know how to act. And it is no coincidence that it is also in this way of managing dependency that questions about the relationship between those intentions are a significant practical concern. If Bob appears to want to watch TV and to want to go out, a carer has no way to defer to his intentions and act on his behalf. It is as if there were two priests in a church service, telling you different things. As if you tell the waiter you want two dishes, but only one of them. One intellectual option at this point is to hold up our hands and say with Davidson, ‘Remember the enormous variety of ways a man can believe or hold something’ (Davidson 2001, 28). Maybe Bob just wants many things. Earlier, I was trying to draw our attention to how normal this is when I argued that a restaurant exerts an unusual demand on us to funnel our various judgments, desires and actions related to food into a single overall one about what to eat for lunch. We can act as if we knew something, and yet profoundly doubt it; we can act at the limit of our capacity and at the same time stand off like an observer and say to ourselves, ‘What an odd thing to do’. (Davidson 2001, 28)

As Nussbaum and Davidson both argue, the idea that all ethical reasoning about how to act will yield only one true result is deeply suspect (Davidson 2001, 34; Nussbaum and Hursthouse 1984; Nussbaum 1986, 2000). It may not, in other words, be any flaw in Bob’s capacity to reason that leads him to the position where he has conflicting ­desires – ­reasons to perform actions that are logically incompatible. While the end result may not resemble a Kantian kind of autonomy and self-consistency, there is not necessarily anything ‘badly done’ about this kind of diversity, fragmentation and complexity within Bob’s selfhood (Laidlaw 2013). Reasons ‘for acting’ are ‘irreducibly multiple’ and Bob’s reasoning does not display any defect for being multiple in precisely this way (Davidson 2001, 34). But the carer, just like the waiter, needs to know what to do. If carers hold up their philosophical hands, to fail to move their caring hands is not an

184 Patrick McKearney

option legally available to them. They must continue to care. And to do this they need to know, just like the waiter, what Bob’s order is. ‘It is not enough to know the reasons on each side’: the carer ‘must know how they add up’ (Davidson 2001, 36). Bob’s (perfectly normal) plurality is not enough. When he does not communicate a clear result, a carer must author one for him. They must decide for Bob which of those various ways he has of believing and holding something ought to move them to action. And that will be to resort to a paternalistic form of care in which another’s decisions control what happens to him, rather than his own. That option loses sight of the aim and hope of this care: that there can be a connection between Bob’s intentions and their actions. This is why carers attempt to interpret things in a way that will ‘add up’ to a single order from Bob that they can execute. Many of these include the same kind of reasoning that Holton himself engages in, when he distinguishes between intentions ‘made in circumstances that allow clear thought’ from wishes that arise ‘in circumstances that prevent clear thought’ (Holton 1999, 248). The notion of ‘attachment’, for instance, does similar work of distinguishing between when Bob and Martha are considering and communicating their true preferences and those in which they are overcome by something that prevents them from accessing them. This model delineates a relationship between these individuals’ intentions that gives one intention more authority than a­ nother – a­ nd thus enables carers to know what to do. This is characteristic of the akrasia debate far more ­generally – i­ t offers models of the psyche that delineate which forms of thought, desire and willing are properly ‘reasonable’ and which are distortions (e.g. Holton 1999, 248–49, 254; see also Davidson 2001 on the various models in the debate). Akrasia and weakness of will are concerned with the relationship between the self and its expression in the world. They arise in the space between the extreme externalist and extreme internalist position; in which we suppose that there should be a connection between judgment and action but also make room for the possibility that it may not always hold in practice. The caring relationship responds to a similar predicament. In it the relationship between decision and action is distributed across two human subjects rather than contained within one. The normative connection between decision and action is distributed between them, as care-recipients with minds judged disabled, and their carers as those who must act under their orders. Akrasia and care share a concern with the fragility of attempts to extend the self, and thus the relationship between parts of the self, as well as between the self and the world, are central to both. The intellectual problem of akrasia arises when we puzzle over why a decision or judgment fails to translate, as we imagine it should, into an action. Because we posit a normative connection between the two, there is something to explain when it does not happen. And this is, in no small part, because the debate imagines that humans have such a thing as a ‘best’ or ‘comprehensive’



Relational Akrasia 185

judgment that has a special status as an imperative or command. That inevitably leads to questions when that judgment fails to move us on some occasion and not others. The way that caring relationships are constructed in Britain requires of people with intellectual disabilities a similarly totalizing kind of judgment. Carers, if they are successfully to act on such a person’s behalf, need more than just an array of different desires. They need the care-recipient to author one over others, and they need them to author it not just as a preference but as a command or an imperative. This requires of people with disabilities precisely the kind of ‘comprehensive’ or ‘all things considered’ judgment that the existence of akrasia relies on (Davidson 2001; Stroud 2014). It requires, in other words, them to turn their different considerations between different options into a definite stance. They are required not simply to have desires and judgments, but to have a will. The puzzle, for carers, begins for carers when they fail to display this, and thus offer no imperative or command that will guide the carer’s action. Why has this person failed to author the kind of binding intention that is required? The akrasia debate, in general, takes it as read that forming intentions, resolutions, and higher order judgments of this nature is necessarily more reasonable and ­rational – ­precisely because the more comparing and reconciling of our different reasons we can do, the better. I have two things to say to this. The first is that this idea, even as it is used to oppose the Socratic stance on akrasia, echoes the Platonic demand of rationality that we render all things comparable to one another: whether as calculable in terms of their potential cost and benefits, or as logically relatable propositions or beliefs. It is far from obvious that this is how life is, and that rationality demands we view the world as such. It may, as Nussbaum argues, be more rational and clear-sighted to value and believe things in ways that do not add up in this way (Nussbaum 1990). The second is that it ignores the ways in which those same philosophers will often consider it unreasonable to press certain kinds of questions that others pursue with great fervour. The point is made by the debate surrounding the famous anthropological example of the Azande, who ask upon every death why it happened to this person now (Evans-Pritchard 1976). They accept the naturalistic explanation of the cause of death but also look for a social cause of that naturalistic process. To simply accept the proposition ‘this is just how it happened’ is for them to leave an unacceptable amount unexplained (Wilson 1974; Tambiah 1990). Most philosophers will consider this to be asking for too much; rationality, in this case, may seem, instead, to demand that we cease trying to find order and reconciliation. Just how much order and coherency we take to be ‘rational’ is not self-evident. Carers so frequently wonder whether Bob and Martha are akratic not because their action conspicuously lacks the hallmarks of rationality. The reason, instead, is that Bob and Martha’s care requires them to author actions

186 Patrick McKearney

in a distinctive way; that is, in precisely the way the akrasia debate requires of us too: with a clear and consistent will. But if we attend to the diversity of ways that we actually author decisions in our everyday lives, it is far from clear that we always make comprehensive judgments and have such a will. It is also far from clear to me that it is a defect in rationality that we do not ‘will’ in this way. It is only under certain social and ethical frameworks that such a will is required, and its absence becomes something problematic to explain. Could it be that it is only under these same conditions that akrasia becomes an analytical puzzle, a moral problem, and even a psychological possibility in the first place?

Acknowledgments I would like to thank the participants of the workshop on akrasia that Nicholas Evans and I organized at the LSE for their insightful thoughts and reflections on this paper and akrasia in general. My deepest thanks as ever go to the participants in my research for their openness to supporting my attempts to understand them and their concerns. Patrick McKearney is an Assistant Professor at the University of Amsterdam conducting research in the UK, India and Italy. His recent articles on disability, care, ethics, and religion include publications in Social Analysis, Ethnos and JRAI. He has also edited two special issues on cognitive disability in The Cambridge Journal of Anthropology and Medical Anthropology.

Note   1. Learning disability is the most widely used term in the UK to refer to those who acquire, before the age of 18, a cognitive impairment that affects daily functioning. Intellectual disability is the more widely used term in international academic literature.

References Aulino, Felicity. 2016. ‘Rituals of Care for the Elderly in Northern Thailand: Merit, Morality, and the Everyday of Long-Term Care’. American Ethnologist 43(1): 91–102. . 2019. Rituals of Care: Karmic Politics in an Aging Thailand. Ithaca: Cornell University Press. Banks, Carys. 2017. ‘Are People with Learning Disabilities Really Being Empowered?: An Ethnography Exploring Experiences of Empowerment Policies in UK Social Care Support’. Unpublished thesis, University of Bath. https://researchportal.bath​ .ac.uk/en/publications/are-people-with-learning-disabilities-really-being-empo​ wered-an-e.



Relational Akrasia 187

Bloch, Maurice. 1974. ‘Symbols, Song, Dance and Features of Articulation: Is Religion an Extreme Form of Traditional Authority?’ European Journal of Sociology / Archives Européennes de Sociologie 15(01): 54–81. Busby, Cecilia. 1997. ‘Permeable and Partible Persons: A Comparative Analysis of Gender and Body in South India and Melanesia’. The Journal of the Royal Anthropological Institute 3(2): 261. https://doi.org/10.2307/3035019. Davidson, Donald. 2001. Essays on Actions and Events Donald Davidson. 2nd ed. Oxford: Clarendon. Davis, Elizabeth Anne. 2012. Bad Souls: Madness and Responsibility in Modern Greece. Durham, NC: Duke University Press. Dumont, Louis. 1980. Homo Hierarchicus: The Caste System and Its Implications. Chicago: University of Chicago Press. Evans-Pritchard, Edward Evan. 1976. Witchcraft, Oracles and Magic among the Azande. Oxford: Oxford University Press. Faubion, James D. 2011. An Anthropology of Ethics. Cambridge: Cambridge University Press. Ferguson, James. 2013. ‘Declarations of Dependence: Labour, Personhood, and Welfare in Southern Africa’. Journal of the Royal Anthropological Institute 19(2): 223–42. Foucault, Michel. 1992. The History of Sexuality, Vol. 2: The Use of Pleasure. London: Penguin Books. Holton, Richard. 1999. ‘Intention and Weakness of Will’. The Journal of Philosophy 96(5): 241–62. https://doi.org/10.2307/2564667. Humphrey, Caroline, and James Laidlaw. 1994. The Archetypal Actions of Ritual: A Theory of Ritual Illustrated by the Jain Rite of Worship. Oxford: Clarendon Press. Kittay, Eva Feder. 2007. ‘Beyond Autonomy and Paternalism: The Caring Transparent Self ’. In Autonomy & Paternalism: Reflections on the Theory and Practice of Health Care, edited by Thomas Nys, Yvonne Denier, and Toon Vandevelde, pp. 23–70. Leuven: Peeters. Laidlaw, James. 2013. The Subject of Virtue: An Anthropology of Ethics and Freedom. Cambridge: Cambridge University Press. Laidlaw, James, and Caroline Humphrey. 2006. ‘Action’. In Theorizing Rituals: Classical Topics, Theoretical Approaches, Analytical Concepts (Numen Book Series) (Studies in the History of Religions), edited by Jens Kreinath, Joannes Augustinus Maria Snoek and Michael Stausberg, pp. 264–85. Leiden: Brill Academic Pub. Lamb, Sarah. 2000. White Saris and Sweet Mangoes: Aging, Gender, and Body in North India. Berkeley: University of California Press. Levinson, Jack. 2010. Making Life Work: Freedom and Disability in a Community Group Home. Minneapolis: University of Minnesota Press. Mahmood, Saba. 2012. Politics of Piety: The Islamic Revival and the Feminist Subject. Princeton: Princeton University Press. McKearney, Patrick. 2019. ‘The Weight of Living: Autonomy, Care, and Responsibility for the Self ’. Journal of Disability and Religion. https://doi.org/10.1080/23312521​ .2018.1483219. . 2020. ‘Challenging Care: Professionally Not Knowing What Good Care Is’. Anthropology and Humanism 45(2): 223–32. https://doi.org/10.1111/anhu.12302. . 2021. ‘What Escapes Persuasion: Why Intellectual Disability Troubles “Dependence” in Liberal Societies’. Medical Anthropology 40(2): 155–68. https://​ doi.org/10.1080/01459740.2020.1805741.

188 Patrick McKearney

. forthcoming. ‘At the Margins of Liberal Care: Ethics Between Dependence and Freedom’. Current Anthropology. Mezzenzana, Francesca. 2020. ‘Between Will and Thought: Individualism and Social Responsiveness in Amazonian Child-Rearing’. American Anthropologist 122(3): 540–53. Mody, Perveez. 2020a. ‘Care and Resistance’. Anthropology and Humanism 45(2): 194–201. https://doi.org/10.1111/anhu.12301. . 2020b. ‘Kinship Care’. In Spaces of Care, edited by Loraine Gelsthorpe, Perveez Mody, and Brian Sloan, pp. 183–200. Oxford: Hart Publishing. Nussbaum, Martha Craven. 1986. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Cambridge: Cambridge University Press. . 1990. Love’s Knowledge: Essays on Philosophy and Literature. Oxford: Oxford University Press. . 2000. ‘The Costs of Tragedy: Some Moral Limits of Cost-Benefit Analysis’. The Journal of Legal Studies 29(S2): 1005–36. Nussbaum, Martha Craven, and Rosalind Hursthouse. 1984. ‘Plato on Commensurability and Desire’. Proceedings of the Aristotelian Society, Supplementary Volumes 58: 55–96. Redley, Marcus. 2018. ‘Full and Equal Equality’. Tizard Learning Disability Review 23(2): 72–77. Redley, Marcus, and Darin Weinberg. 2007. ‘Learning Disability and the Limits of Liberal Citizenship: Interactional Impediments to Political Empowerment’. Sociology of Health & Illness 29(5): 767–86. Robbins, Joel. 2007. ‘Between Reproduction and Freedom: Morality, Value, and Radical Cultural Change’. Ethnos 72(3): 293–314. Staal, Frits. 1979. ‘The Meaninglessness of Ritual’. Numen 26(1): 2–22. https://doi.org​ /10.2307/3269623. Stasch, Rupert. 2008. ‘Knowing Minds Is a Matter of Authority: Political Dimensions of Opacity Statements in Korowai Moral Psychology’. Anthropological Quarterly 81(2): 443–53. Strathern, Marilyn. 1990. The Gender of the Gift: Problems with Women and Problems with Society in Melanesia. Berkeley: University of California Press. Stroud, Sarah. 2014. ‘Weakness of Will’. In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Spring 2014. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/spr2014/entries/weakness-will/. Tambiah, Stanley Jeyaraja. 1990. Magic, Science, Religion, and the Scope of Rationality. Cambridge: Cambridge University Press. Welshman, John, and Jan Walmsley (eds). 2006. Community Care in Perspective. Basingstoke: Palgrave Macmillan. Williams, Bernard. 1981. Moral Luck: Philosophical Papers 1973–1980. Cambridge: Cambridge University Press. . 1993. Shame and Necessity. Berkeley: University of California Press. Wilson, Bryan R. 1974. Rationality. Oxford: Blackwell.

Afterword Akrasia in Its Social Context Richard Holton

It is always an ­education – ­and a ­joy – ­to discuss topics from one’s own discipline with people from another. And so it has been here. In its discussion of akrasia, analytic philosophy has tended to focus on simplified examples in order to control the variables: cases are typically made up, with their features stipulated or they are taken, in caricatured form, from literature. I am as guilty as anyone. Rarely do the cases come from the real world. Moreover, analytic philosophy tends to be resolutely individualistic. Akrasia is typically seen as a very personal characteristic; social context rarely gets a look in. So it is refreshing to turn to real-world examples, with the uncertainty and confusion they bring with them; and to examples that focus on the social context as much as on the individuals, and hence on what the editors call ‘endemic akrasia’. It is this latter idea that will be my focus. A couple of clarifications before starting. The first is terminological. I will stick with a distinction that I have used before.1 By ‘akrasia’, I mean an agent’s free intentional action that is contrary to their judgment of what it is, all things considered, best for them to do. By ‘weakness of will’, I mean an agent’s free intentional action that is contrary to a resolution that they have previously formed. The distinction is somewhat stipulative, although I think that the characterization of weakness of will is broadly in line with normal usage.2 Typically, akrasia and weakness of will go together since one normally resolves to do what one judges best, and so a failure to do one is a failure to do the other. But sometimes they come apart: one might resolve to do something that one doesn’t judge best; or one’s judgments may change over time; or one may judge something best but never resolve to do it. The second clarification is conceptual, and it concerns akrasia, not weakness of will. The main philosophical debate, stretching from Socrates, does not question the rationality of akrasia; almost everyone concedes that it is in some sense irrational. Rather, it denies its possibility. On a certain conception of human beings, and of their attitudes, a judgment of what is best is inescapably

190 Richard Holton

linked to desire, and desire to action. So, the person who genuinely judges that a certain course of action is best is bound, whether or not they are rational, to act on it. There are various possible responses here. One is to accept the Socratic position that akrasia is impossible, but to soften the blow. What had been taken as akrasia is mere weakness of will; no one thinks that is impossible. The apparently akratic person has changed their mind; they do not act on their resolution, but that is because, perhaps in thrall to the power of temptation, they no longer judge that doing so would be best. Suppose though that we want to deny the Socratic position. Two strategies are available. One is to deny the connections between judgment, desire and action that Socrates posited: to say that one can judge something best without desiring to do it, or that one can desire to do something, and yet fail to act on that desire. The second is to accept those Socratic contentions, but to point out that they only hold within an agent, and that a given human being may contain more than one agent: the conscious and the unconscious, for instance, as in psychoanalysis, or in the Indonesian hypnotherapy, so vividly described by Nicholas Long, which distinguishes the conscious 12% of the mental iceberg from the 88% that is hidden. Or, in a slightly different register, the two opposed ‘wilful subjects’, the true client and ‘Ed’, the eating disorder, as posited in the Cedar Grove rhetoric that Rebecca Lester discusses. I present these t­ wo – d ­ enying the connections, or dividing the s­ ubject – ­as stark alternatives, but accounts often take a position somewhere between them. If judgments don’t affect actions, the actions must be somewhat insulated from them. How great does the internal insulation have to be before you get a different agent? There may be no clear answer. Plato’s famous metaphor of the chariot pulled by two horses, one wayward, one obedient, can be read, on the first approach, as presenting the human being as a single agent, the charioteer, contending with different internal forces; or it can be read, on the second, as presenting the human being as three distinct agents: the two horses and the charioteer.3 On Plato’s picture though, and in almost all of the philosophy that has followed it, these divisions, whether between kinds of desire or between discrete agents, have been understood as firmly individual and internal. The agent is internally divided: it is the psyche that is somehow fractured. Anthropology’s stress on the society raises the idea that we might look for the primary fracture elsewhere. To see why this doesn’t inevitably just collapse back into the ­internal – t­ o the view that what matters is not the society but the individual’s beliefs about the s­ ociety – i­t is worth looking at externalist accounts of the mind more broadly. There has been a highly influential strain of thought since Kripke and Putnam that you cannot characterize a mental state just by looking inwards.4 A thought about a tiger, say, or about arthritis, or about the moon, needs to be characterized in terms of the thing it is about: the external type or property



Afterword 191

or object. More recently has come the thought that not just the content of thought but the process of thinking needs to be understood more externally: philosophers have increasingly started to think about embodied cognition, which takes us out of the brain, or extended cognition, which takes us right out of the body. For example, consider the much-discussed issue of how we position ourselves to catch a ball. Imagine a fielder in a game like cricket or softball, running to the spot where the ball will land, and adjusting that run as the ball is affected by the wind or other obstacles. Even novices are not bad at this (the bigger difficulty is in getting there fast enough, or in catching the ball when we do). But, as a bit of physics, the calculation is quite difficult, and it requires estimating a lot of different forces. How then do we do it? A celebrated suggestion by Seville Chapman is that the person moves their body, i.e. runs, in such a way as to keep a certain feature constant.5 That feature, it turns out, is not exactly trivial: even when the ball is approaching head on, it is the rate of change of the tangent of the elevation angle of the ball. Clearly that is not something that the agent explicitly calculates. But the suggestion is that w ­ e– ­and many other a­ nimals – ­might have evolved the ability to recognize when it holds: to recognize when our movement keeps that rate of change constant, and hence to adjust our movement to keep it so (the elevation angle of the ball is equivalent to the viewing angle, so it is plausible that we can effectively calculate this by knowing the angle at which the head is tilted). And then the problem is solved without the need to estimate all of the various forces involved, to perform computational corrections when the ball is moved by a gust of wind and so on. As it is sometimes put, the agent knows how to reach the point where the ball will land, even if they do not ­know – ­cannot s­ pecify – ­where the ball will land. How should we describe how the agent achieves the calculation? We might try to focus just on what is going on in the brain: the calculation of the rate of change of the tangent given the information of the tilt of the head. But clearly that is inadequate. The calculation is achieved in the context of a feedback system involving the agent’s body and the field and the ball: the agent is moving across the field, tracking the ball with their eyes in order to provide the information for the calculation, and then that calculation in turn determines the rate at which they move. We need to describe the whole ­system – b ­ ody, field, ball, ­brain – ­in order to describe what is going on. This explanation doesn’t involve other persons or social expectations; but once the general point is ­made – ­that the best explanation of what we at first think of as an internal cognitive process might involve factors that extend beyond the a­ gent – ­the way is left open for introducing it. The idea of the divided person that is common across philosophy, even in much of the most insightful (such as Amélie Rorty’s lovely (1986) image of the old city with highways overlain) has remained highly internal and hence individual. It is the inner self that is divided up: the bit that wants the cake warring with the

192 Richard Holton

bit that doesn’t. Other people rarely get a look in. Yet from the anthropological perspective, these other people are central: witness, in this volume, Jon Bialecki’s discussion of relational akrasia in Charismatic Christianity, or Ivan Deschenaux’s discussion of how a member of one caste understands the significance of their actions in the eyes of others, especially those of different castes. Even the cases that look more like the classic philosophical examples turn out to have crucial social dimensions. Darin Weinberg’s discussion of addiction moves to the cases of Del and Neal, whose apparently irrational actions turn out to be properly responsive to how they are socially perceived. And the focus in the Cedar clinic on personifying eating disorders as the internal Ed frequently turns out to be a cover for conflicts between the clients and their parents and clinicians. How does this change things? If desires are thought of as intrinsic, as properties of some inner self, then the clash of desires that is characteristic of akrasia pushes us, as we have seen, towards the idea of numerous selves, or at least numerous distinct foci of desire, within the agent. But once desire is seen as more external, as the reflection of action within a network of social relations and expectations which channel and reinforce it, then it becomes less mysterious and more open to explanation. Given that those networks will be varied, and actors will move between them, it is not surprising that the desires will be conflicting, and not surprising that they will often come adrift from the judgments that the agent makes: that they will be akratic. Does this make for different social selves, or a single self with distinct social relations? I doubt that it much matters quite what we say. But we should be careful about where we ascribe akrasia, since, as many of the authors in this collection point out (most centrally Lubomira Radoilska) the judgments are also likely to shift. In other places I have stressed, within a cognitive dissonance framework, how this can happen in a very internal way, as agents try to make sense of themselves (Holton 2009, Ch. 5). Equally important is a more social shift, as agents use the testimony of different social groups to provide content to their judgments.6 In such cases their actions and their judgments will align. So any akrasia will be diachronic, or merely potential: the agent had previously formed judgments that were contrary to their current actions, or will form them again, or will be such that they could be brought to form them in a different social environment. I have focused so far on akrasia, the divergence between judgment and action. What of weakness of will, the divergence between resolution and action? Much of what I said about akrasia transfers straight over. Just as a new social environment can lead to actions that conflict with earlier judgments, so it can lead to actions that conflict with earlier resolutions. But since the forming of a resolution is often a public a­ ct – ­a speech ­act – ­weakness of will raises a number of new issues in the social context. One concerns the involvement of further agents in the execution of an action, as discussed by Patrick McKearney. Should an earlier decision to go out to a café, made by someone



Afterword 193

in sheltered accommodation, overrule a clear later desire not to? If it were just up to the individual, clearly the later desire would win out. But once a carer is involved, one, moreover, who has been instructed to respect and promote the autonomy of the individual, the earlier decision can look authoritative. In reality here, I suspect that the carer’s decision will be influenced by their experience of the outcomes of the different policies. If overruling later objections and bundling the person out of the door typically leads to a good day, one remembered and savoured through the week, that looks like a good reason to stick with the resolution. If, on the other hand, it typically leads to a day of misery and subsequent recrimination and regret, then the force of the resolution looks largely evidential; there is still space for a notion of autonomy, but the resolution works as evidence of an ongoing desire that should be respected rather than as the deciding autonomous act. But as McKearney so forcefully brings out, no policy is going to be straightforward. In other cases, the role of publicly announced resolutions may be different again. In my earlier work, I had only seen them as harbingers of future action. But since they are speech acts in their own right, they may have a very different point. Philosophers have typically followed Hume in thinking that announcing a resolution, like publicly making a promise, increases the pressure on the actor to go through with the action. But Peter Gollwitzer (2009) has found that often announcing a resolution decreases the likelihood of further action. The actor is under pressure to do something; publicly announcing a resolution to act becomes a substitute for the action itself (one is reminded of the modus operandi of certain governments).7 Perhaps these thoughts can shed light on the perplexing attitude to alcohol, especially to the imported spirits, aguardiente, in the Runa community that Francesca Mezzenzana describes. Resolutions to stop drinking are constantly made and then broken. There is nothing surprising in that, and nothing surprising that people come up with stories that give themselves a defence for so doing them. What is surprising is the lack of sanction that comes from other people to such weakness of will. Mezzenzana suggests that the core issue is simply that the members of the community are unconcerned about consistency. Perhaps that is right, but as she says, this isn’t generally the ­case – ­only in certain contexts. If so, what is special about the contexts? More broadly, if resolutions are so unlikely to be acted on, and if they bring so little public sanction when broken, then why bother to make them in the first place? One possibility is again that the resolutions are playing some different role. Perhaps they are a way of voicing regret at the bad consequences of drinking too much aguardiente: the fights, the domestic violence, the hepatitis. We often think of ­regret – ­this is almost axiomatic in ­philosophy – a­ s the reflection of a wish that we had done otherwise, and hence as involving a commitment to actually do otherwise in the future. A good choice, it is often assumed, is one that can be looked back on without regret. But that cannot be generally right. A nice example from Lloyd Humberstone (1980): someone bets £20 on

194 Richard Holton

a horse. If the horse loses, they regret not saving their money; if it wins, they regret not putting £40 on it. What do they do next time they are in a similar situation? They bet £20. Regret can reflect simply the fact that things could have been better. The Runa people might very much regret the bad effects of the aguardiente without thinking that they provide a reason to stop drinking it. And perhaps, especially if they are a community in which the public expression of resolution has traditionally had little role, their resolutions serve primarily to express that regret. This last point does highlight a more general consideration, one that is all too familiar to anthropologists. Philosophers are prone to look for the essence of things. That may be a natural human tendency (Gelman 2003), and it works well for much of the natural world. But when the subject matter is a social phenomenon like akrasia, we are much more likely to see a range of things: phenomena closely enough related to deserve the name and to be illuminated by cross-cultural comparison, but distinguished by the divergent cultural settings in which they are to be found. The essays in this volume make that very clear. Richard Holton is Professor of Philosophy at the University of Cambridge, and a fellow of Peterhouse. He works on a wide array of issues around moral psychology, and does his best to read more than just the philosophical literature.

Notes   1. For a full discussion, see (Holton 2009, Ch. 4)   2. There is some debate here, and the usage of the expression ‘weakness of will’ turns out to be complex. See (Mele 2010) and in response (May and Holton 2012). I take it that ‘akrasia’ is a technical term in English, and so can be defined by stipulation.   3. Consider also the passage from Davidson (1982) about ‘overlapping territories’ quoted above by Weinberg p. 125.   4. For the classic texts see (Kripke 1980, Putnam 1975, Burge 1979).   5. (Chapman 1968). Although highly influential, the account is not uncontested. See, for instance (Postma et al. 2017; Höfer et al. 2018).   6. Chalmers and Clark (1998) famously argue that the contents of one’s phone can provide an extension of one’s mind. But what if the phone provides access, as it will, to very different sets of opinions?   7. I describe this as a consequence of the social presentation of the speech act of resolving, but there is a first-person analogue involving some very recognizable rationalization:

Had I been less firmly resolved upon settling down definitively to work, I should perhaps have made an effort to begin at once. But since my resolution was explicit, since within twenty-four hours, in the empty frame of the following day where everything was so well arranged because I was not yet in it, my good intentions would be realized without difficulty, it was better not to start on an evening when I felt ill-prepared. The following days were not, alas, to prove more propitious. (Proust 1919, 623–4)

Afterword 195



References Burge, Tyler. 1979. ‘Individualism and the Mental’. Midwest Studies in Philosophy 4: 73–121. Clark, Andy, and David Chalmers. 1998. ‘The Extended Mind’. Analysis 58: 7–19 Chapman Seville. 1968. ‘Catching a Baseball’. American Journal of Physics 36: 868–70. Davidson, Donald. 1982. ‘Paradoxes of Irrationality’. In Philosophical Essays on Freud, edited by Richard Wollheim and James Hopkins, pp. 289–305. Cambridge: Cambridge University Press. Gelman, Susan. 2003. The Essential Child: Origins of Essentialism in Everyday Thought. New York: Oxford University Press. Gollwitzer, Peter, et al. 2009. ‘When Intentions Go Public: Does Social Reality Widen the Intention-Behavior Gap?’ Psychological Science 20: 612–18. Höfer, S., et al. 2018. ‘No Free Lunch in Ball Catching: A Comparison of Cartesian and Angular Representations for Control’. PLoS ONE 13(6): e0197803. Holton, Richard. 2009. Willing, Wanting Waiting. Oxford: Oxford University Press. Humberstone, I. L. 1980. ‘You’ll Regret It’. Analysis 40: 175–76. Kripke, Saul. 1980. Naming and Necessity. Oxford: Blackwell. May, Joshua, and Richard Holton. 2012. ‘What in the World Is Weakness of Will?’ Philosophical Studies 157: 341–360. Mele, Al. 2010. ‘Weakness of Will and Akrasia’. Philosophical Studies 150: 391–404. Postma, Dees, et al. 2017. ‘When a Fly Ball Is Out of Reach: Catchability Judgments Are Not Based on Optical Acceleration Cancelation’. Frontiers in Psychology 8: 535. Proust, Marcel. 1919. A l’ombre des jeunes filles en fleurs. Trans. C.K. Scott Moncreiff and Terence Kilmartin. London: Chatto and Windus, 1981. (Page references to the translation.) Putnam, Hilary. 1975. ‘The Meaning of “Meaning”’. In Philosophical Papers, Vol. II, pp. 215–71. Cambridge: Cambridge University Press. Rorty, Amélie. 1986. ‘Self-Deception, Akrasia and Irrationality’. In The Multiple Self, edited by Jon Elster, pp. 115–31. Cambridge: Cambridge University Press.

Index

addiction, 15, 50–65, 150–51, 155 as akrasia, 51 and freedom, 53–54 willing vs. unwilling, 43–44 alcohol, 72–74, 82, 84, 151–57 alcoholic, 71–73, 75–76, 80–81, 84 alcoholism, 55, 58, 59, 71, 73, 76, 80 beer 76–79 manioc beer, 71–72, 74–84, 87n2 agency, 40, 105–6, 118, 121, 155–57, 160, 162–65 agency panic, 118 lateral, 106, 118 akrasia in the bible, 89–91 collective, 22–23, 136n and eating disorders, 41–42 endemic, 22–23 externalist approach to, 18, 24–26, 182 internalist approach to, 13–14, 18, 24–26, 182 Islamic, 104, 111–18 possibility of, 2, 12–13, 17–21, 189–90 relational, 169–186 subjective, 104, 107–10 See also collective Amazonia, 75, 78–79, 87n1, 176 anchoring, 109, 114 anorexia, 41, 42–43 anti-virus, 114–18 Aristotle, 17–18, 23 autonomy, 58–59, 78, 80, 84–85, 95, 183 Berlant, Lauren, 47, 106, 118–19 bulimia, 44–45

care, 45–47, 65, 74–75, 82–3, 147, 153, 169–86 in Britain, 170, 176–77, 185 managed care, 36 Cedar Grove, 35 character, 1, 17, 92, 98, 181 children, 71–72, 75–76, 78–79, 82, 87, 176 Christianity, 24, 89–100 city, the as metaphor for the self, 105, 107, 119–20 collective, 97 colonialism, 6, 78 community, 71–85, 87n2 computer virus, 108–22 consistency, 12–17, 43 see also inconsistency coordination problem, 137 Costa, Luiz, 79 critical factor, 108–17, 123n.4 Davidson, Donald, 18–19, 50, 52, 61–65, 133–35, 137, 160, 171–75, 182–85 demons, 98 desire, 11, 13, 17, 24–25, 43–44, 94–97, 111, 171–74, 178, 183–85 Douglas, Mary, 10–11 drinking, 75–87, 87n, 151–56, 159–60 collective, 78–80, 85 eating disorder, 34–49 Ed (eating disorder), 39–40 ethnographic theory, 3, 26–27, 120 Evans-Pritchard, E.E., 6–7, 9–11, 185 evidence, 148–50, 161–64

Index 197



Frankfurt, Harry, 43, 59–60 freedom, 14, 16, 23, 41, 73, 82, 87n2 grit, 148–49, 159, 162 hati Nurani (conscience), 112–19 Heather, Nick, 51–54, 76 healthcare, 36 Holton, Richard, 19–20, 24–25, 72, 133, 178–82, 184 hypnotherapy, 102–22 hypnosis, Ericksonian, 102, 104, 111 hypnoparenting, 102, 117–18 inconsistency, 1–2, 4–12, 23–26, 45–46, 86, 87n2, 131, 147, 171, 183, 190 see also consistency independence, 74–75, 78, 80, 84, 86 and dependence 80–1, 86, 172, 178, 182–83 India, 175–79, 181–82 Indonesia, 102–22 intention, 1–4, 16–17, 19–20, 24–25, 41–44, 46, 52–54, 133–34, 140–41, 171–72, 176–85 collective intentionality, 141 See also resolutions intermarriage, 126–27 irrationality, see rationality Islam, 24, 102, 103, 104, 111–18 judgment, 1–2, 4–17, 41–44, 52–53, 62, 96–97, 103–4, 159, 171–72, 175–86 all things considered, 56, 179, 185 best judgment, 12–14, 19, 42, 112, 119, 112, 175, 181, 184–85 capacity to make, 178 kin, 75–6, 78–79, 84 kinship, 8 Leach, Edmund, 8–9 Lévy-Bruhl, Lucien, 9, 11 magnetism, 104, 114 Marxist anthropology, 7–8, 14–15 mental revolution, 103, 109, 118 mesmerism, 104, 107, 114 moral, 81, 87, 87n2 luck, 119, 122, 177

selfhood, 85 stances, 83 moral incontinence, 1, 52, 116 Nepal, 126–43 neuro-linguistic programming (NLP), 103–4, 114–15, 123n3 Nuckolls, Charles, 86–87 Nussbaum, Martha, 183, 185 originationism, 110–11 parties, 71–73, 76–78, 80, 83, 87n2 pencak silat (martial art), 114 persuasion, 78 Pettit, Philip, 22, 136–38, 142s philosophy and anthropology, 3, 19, 26–27, 105–7, 127–28, 142, 147, 185–86, 189–94 of the mind, 25 moral, 5 utilitarian, 5 Plato, 2, 13, 17, 185, 190 prayer, 89–100 as technique, 93–96 quantum mysticism, 109 rationality, 1–16, 22–23, 41–42, 61–62, 86, 105–7, 112, 118, 119, 141, 185–86, 189 debate in anthropology, 6 irrationality, 19, 146, 159–65 rationalization, 57, 156–58, 165 resolutions, 20, 73, 83–84, 133, 189, 193 See also intentions replication of caste, 127 self-sustaining practices, 139–40 responsibility, 35, 39, 47, 92, 172, 175 Individual, 75 Rorty, Amélie, 18, 20, 22, 105–7, 118–20, 140, 191 self-control, 17, 56–64, 90, 178, 180–81 self-cultivation, 18, 24–25, 99 self-hypnosis, 110, 116–17, 123n5 Socrates, 2, 12–13, 16–20, 172, 185, 190 suggestion, 104, 108–21

198 Index

shame, 38, 81 stigma, 73, 81, 85 Singer, Peter, 4–5, 8–9 strength of will, 23 of demons, 98 Sztutman, Renato, 75, 79

untouchability, 126–27 Uzendoski, Michael, 75, 79

thought conscious, 103, 107–9, 114, 117, 121 subconscious, 107–18 trance, 103, 109, 111–17 triggers, 37–39, 46 trigger warnings, 34–35

weakness of the will, 1, 18–20, 40, 52, 71–73, 81, 84–87, 180–81, 189, 192–93 non-akratic weakness of will, 133–34 see also strength of will will, the, 35, 39–41, 45–46, 176–77

Vinyard, 91–92 Viteri, Carlos, 71 Viveiros de Castro, Eduardo, 76