Philosophy of Suffering: Metaphysics, Value, and Normativity 0815361785, 9780815361787

Suffering is a central component of our lives. We suffer pain. We fall ill. We fail and are failed. Our loved ones die.

168 53 7MB

English Pages 331 Year 2019

Report DMCA / Copyright


Recommend Papers

Philosophy of Suffering: Metaphysics, Value, and Normativity
 0815361785, 9780815361787

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Philosophy of Suffering

Suffering is a central component of our lives. We suffer pain. We fall ill. We fail and are failed. Our loved ones die. It is a commonplace to think that suffering is, always and everywhere, bad. But might suffering also be good? If so, in what ways might suffering have positive, as well as negative, value? This important volume examines these questions and is the first comprehensive examination of suffering from a philosophical perspective. An outstanding roster of international contributors explore the nature of suffering, pain, and valence, as well as the value of suffering and the relationships between suffering, morality, and rationality. Philosophy of Suffering: Metaphysics, Value, and Normativity is essential reading for students and researchers in philosophy of mind, philosophy of psychology, cognitive and behavioural psychology, as well as those in health and medicine researching conceptual issues regarding suffering and pain. David Bain is Reader in Philosophy at the University of Glasgow, UK. Michael Brady is Professor of Philosophy at the University of Glasgow, UK. Jennifer Corns is Lecturer in Philosophy at the University of Glasgow, UK.

Philosophy of Suffering Metaphysics, Value, and Normativity

Edited by David Bain, Michael Brady and Jennifer Corns

First published 2020 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2020 selection and editorial matter, David Bain, Michael Brady, and Jennifer Corns; individual chapters, the contributors The right of David Bain, Michael Brady, and Jennifer Corns to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record has been requested for this book ISBN: 978-0-8153-6178-7 (hbk) ISBN: 978-1-351-11546-9 (ebk) Typeset in Times New Roman by Taylor & Francis Books


List of illustrations List of contributors Acknowledgements Introduction

vii viii xii 1



The nature of suffering 1 The world according to suffering

17 19


2 The disruption model of suffering



3 Painfulness, suffering, and consciousness



4 Suffering pains




Pain and valence 5 Valence, bodily (dis)pleasure, and emotion

101 103


6 Pain and mere tastes: Toward an attitudinal-representational theory of valenced perceptual experiences HILLA JACOBSON




7 Pain: An attitude with two heads




The value of suffering 8 Suffering as transformative experience

163 165


9 After motivational hedonism: Feeling bad can be good | feeling good can be bad



10 From suffering to satisfaction: Why we need pain to feel pleasure



11 ‘My horses and hogs and even everybody seemed changed’: Appreciating beauty in depression recovery




The normativity of suffering


12 Hedonic rationality



13 The agony of reason: The unsteady bond between suffering and human rationality



14 Some paradoxes of pain for rational agency



15 Suffering as a virtue






Figure 4.1 Pain and suffering: three subsumptive views


Table 4.1 Four differences between pain and suffering



Marilyn McCord Adams was Professor of Philosophy and Theology and an Episcopal priest at the time of her death in 2017. She taught at numerous universities, including UCLA, Yale, and Oxford, where she was the first woman and the first American to be Regius Professor of Divinity. At Oxford she was also Canon of Christ Church Cathedral. Her academic specialities were medieval and philosophical theology, and she was especially well known for her groundbreaking work on the problem of evil. She published five monographs and over 100 journal articles and encyclopaedia entries. David Bain is Reader in Philosophy at the University of Glasgow. His research centres on pain and affect, and he is increasingly interested in suffering and the emotions. Jointly with Professor Michael Brady, he was Principal Investigator of the Pain Project (2012–2013) and the Value of Suffering project (2013–2016). He has published on pain, affect, colour, and Wittgenstein’s private language argument ( Brock Bastian is an Associate Professor in the School of Psychological Sciences at the University of Melbourne. He is trained as a social psychologist and his research broadly focuses on the topics of ethics and well-being. Brock completed in his PhD in 2007 and since then has published more than 100 peer-reviewed journal articles and book chapters, including a popular science book, The Other Side of Happiness. Hagit Benbaji is Associate Professor of Philosophy at the Ben-Gurion University of the Negev. She has published on pain, emotion, action and causation, the mind–body problem, colour and colour perception, epistemology, the philosophy of Thomas Reid, and forgiveness and toleration. Michael Brady is Professor of Philosophy at the University of Glasgow, Scotland. His research centres on the philosophy of emotion, and its links with moral philosophy and epistemology. In 2013 his book Emotional Insight was published by Oxford University Press. He was Co-Principal Investigator, with Dr David Bain, on the Templeton-funded Value of

List of contributors


Suffering project, and his resulting book, Suffering and Virtue, was published by Oxford University Press in 2018. His introductory textbook Emotion: The Basics was also published. Outside of academia, he is philosopher-in-residence for the Manchester-based theatre company Quarantine. Havi Carel is Professor of Philosophy at the University of Bristol. She is a Wellcome Trust Senior Investigator, leading the Life of Breath project (2014–2020;, which won the Health Humanities’ Inspiration Award 2018. She is the author of Phenomenology of Illness (2016), Illness (2008, now in its third edition and shortlisted for the Wellcome Trust Book Prize), and of Life and Death in Freud and Heidegger (2006). She was voted by students as a ‘Best of Bristol’ lecturer in 2016. She previously held grants from the AHRC, British Academy and Leverhulme Trust ( Jonathan Cohen is Professor of Philosophy and faculty member of the Interdisciplinary Cognitive Science Program at the University of California, San Diego. Before coming to UC San Diego, he was a Killam Postdoctoral Fellow in philosophy at the University of British Columbia. He earned his PhD in philosophy at Rutgers University. His has published extensively in philosophy of perception (particularly on colour and on interactions within and between perceptual modalities) and philosophy of language (especially on issues about context sensitivity and the semantics/pragmatics interface). He is the author of The Red and the Real: An Essay on Color Ontology (OUP, 2009), and co-editor of Color Ontology and Color Science (with Mohan Matthen, MIT Press, 2010) and Contemporary Debates in the Philosophy of Mind (with Brian McLaughlin, Wiley-Blackwell, 2007). Sam Coleman is Reader in Philosophy at the University of Hertfordshire, and author of various articles mainly in the philosophy of mind. His current work explores the view that the mind is of a nature which is revealed to us in consciousness, but where mental operations nonetheless go on for the most part outside of consciousness. This theory is developed in detail in his forthcoming book with Oxford University Press, Dark Mind. Jennifer Corns is Lecturer in Philosophy at the University of Glasgow. Her research focuses on pain, suffering, affect, and death. Her forthcoming monograph with Routledge, The Complex Reality of Pain, employs contemporary philosophy, scientific research, and clinical reports to argue that pain, though real, is not an appropriate object of scientific generalisations or an appropriate target for medical intervention. She aims to use philosophical tools and evaluate empirical research to make progress on topics that matter within and beyond the academy. Matthew Fulkerson is Associate Professor of Philosophy at the University of California, San Diego in the United States. His research has focused on the


List of contributors sense of touch and haptic exploration. He is especially interested in the relationship between bodily awareness and our experience of the world, and in understanding how touch connects and interacts with the other senses. More recently, he has been exploring the relationship between emotions and perception, as well as writing papers on motivating sensory states and sensory pleasure and pain. He is the author of The First Sense: A Philosophical Study of Human Touch (MIT Press, 2014).

Hilla Jacobson is Associate Professor of Philosophy and Cognitive Science at the Hebrew University of Jerusalem, Israel. Her research focuses on consciousness, perception, desire, pain and affect. Much of her work stands at the point where philosophical concerns overlap with and draw upon empirical research in psychology and cognitive neuroscience. She is currently developing a general theory of perceptual valence. Antti Kauppinen is Professor of Practical Philosophy at the University of Helsinki. His work has focused on well-being, meaningfulness, moral emotions, practical rationality, and the nature of normativity. He is the coeditor (with Jussi Suikkanen) of Methodology and Moral Philosophy (Routledge, 2019) and Principal Investigator of the Academy of Finland research project Responsible Beliefs: Why Ethics and Epistemology Need Each Other (2019–2023). Ian James Kidd is assistant professor at the Department of Philosophy, University of Nottingham. His research interests include the philosophy of illness, phenomenology, and feminist philosophy of healthcare. He was co-editor (with José Medina and Gaile Pohlhaus, Jr.) The Routledge Handbook to Epistemic Injustice (2017) and is editing (with Heather Battaly and Quassim Cassam) Vice Epistemology (2020). His website is www.ia Colin Wayne Leach is a social and personality psychologist who studies status and morality in identity, emotion, and motivation. At Columbia University, he is Professor of Psychology & Africana Studies at Barnard College as well as Senior Research Scientist and Graduate Faculty in the Department of Psychology, School of Arts & Sciences. In addition to authoring nearly 100 journal articles and book chapters, Prof. Leach has co-edited the volumes Psychology as Politics (Political Psychology, 2001), Immigrant Life in the US (Routledge, 2003), The Social Life of Emotions (Cambridge, 2004), and Societal Change (Journal of Social & Political Psychology, 2013). Olivier Massin is professor of philosophy at the Institut of Philosophie of the University of Neuchâtel. He received two PhDs in philosophy, one from the University of Aix-Marseille and the other from the University of Geneva. He has previously been research scientist at the Institut Jean Nicod (Paris) and Professor at the Philosophy Department of the

List of contributors


University of Zürich. He has worked on questions such as: what are forces? What are mixtures? What are economic exchanges? What is ownership? What is continuity? What is pleasure? What is suffering? What is pain? What are desires? What are determinables? What are tryings? What are efforts? What is touch? Tom McClelland is a Research Fellow at the University of Warwick where he holds a Leverhulme Early Career Fellowship. His project Mental Action and Cognitive Phenomenology explores how we select what mental actions to perform, and what the experience of mental action is like. He has worked on a number of topics in philosophy of mind including the problem of consciousness, introspection, perceptual content and the self. He also dabbles in aesthetics and has published on the philosophy of film. Tasia Scrutton is an Associate Professor in Philosophy and Religion at the University of Leeds with interests in philosophy of religion, philosophy of psychiatry, and philosophy of emotion. Much of her current research looks at religious, spiritual, medical and other interpretations of (what in psychiatry are called) mental illnesses, examining how interpretations shape people’s experiences. Fabrice Teroni is Associate Professor in philosophy at the University of Geneva and project leader at CISA, the Swiss Centre for Affective Sciences. He works in the philosophy of mind and epistemology. His background is in the philosophy of memory, of perception and of affective states. He has published several articles and monographs on the general theory of emotions (The Emotions: A Philosophical Introduction, Routledge), on shame (In Defense of Shame: The Faces of an Emotion, Oxford University Press) and on memory.


This volume arose from The Value of Suffering Project, an interdisciplinary and international research project funded by the John Templeton Foundation. Its core team – Principal Investigators David Bain and Michael Brady; postdoctoral fellow (now lecturer), Jennifer Corns; and doctoral student, Abraham Sapien-Cordoba – were all philosophers at the University of Glasgow. The broader team (located in Australia, France, Norway, the US, and the UK) had expertise in philosophy of mind and cognitive science, theology, neuroscience, psychology, and clinical practice. From September 2013 to June 2016, the Project ran six workshops and three conferences, from which many of the contributions to this volume derive. We hope the present volume demonstrates the richness of suffering as a topic for research, and hope it stimulates further work in this area.

Introduction David Bain, Michael Brady and Jennifer Corns

This is a volume about the nature and value of suffering. Suffering is a central part of the lives of all human beings. Those of us who are lucky enough to escape the ravages of war, famine, poverty, and oppression still have lives in which suffering of various kinds plays a large part, for we all, to a greater or lesser extent, experience different forms of physical and emotional suffering: pain, hunger, fatigue, grief, guilt, and many others. Given this, it is surprising that suffering has received little in the way of serious consideration from researchers in core disciplines like philosophy, psychology, and neuroscience. Insofar as researchers in these subjects have considered the issue, they have often neglected questions about suffering as such. There has been considerable research on pain (somewhat to the exclusion of other kinds of physical suffering) and also on negative emotions, such as guilt, shame, and grief, but very little has been done to address the most general questions that arise. What do all these negative states have in common? How is their negativity related to their affective or hedonic dimensions – that is, to their being unpleasant? When, and in virtue of what, are these various states involved in, or themselves kinds of, suffering? And what is the role, purpose, or value of suffering? These questions have all been neglected. Or, at least, they have been until recently. For in the past few years, researchers from various disciplines have started to broaden their focus and ask systematic questions about negative affect and suffering. Some of this interest comes from psychology and neuroscience – with researchers interested in the relations between relatively well-known phenomena like pain and important but more technical notions such as valence, liking, and motivation. Some of this interest comes from philosophy. Those working in the philosophy of pain have started to investigate what we can learn about pain by looking at other forms of negative affective experience and at how theoretical developments in pain research might illuminate the nature of emotional suffering. Those working in philosophy of emotion and theology are increasingly considering questions about the value of suffering, and its purpose or function. The positive psychology movement has blossomed in recent years, along with the idea that suffering has educational value. At


David Bain, Michael Brady, Jennifer Corns

the same time, there has been increased focus in popular science on traits that are important to deal with suffering, such as resilience and grit. The present volume is part of this intellectual turn towards suffering. It aims to address four core issues about suffering in four parts: (I) the nature of suffering; (II) the nature of pain and valence; (III) the value of suffering; and (IV) suffering and other normative concepts. In what follows we’ll give a brief theoretical overview of the central issues under each heading, and then a brief summary of the respective chapters. In this way we hope to delineate the central questions and theoretical options that characterise this vitally important aspect of our lives.

Part I: The nature of suffering What is suffering? Theories of suffering aim to answer this question. Constructing a theory of suffering is easier said than done, however. For one thing, a great variety of things can be involved in suffering: sensations, perceptions, thoughts, evaluations, desires, bodily changes, and – some will say most important – negative affect. How are these involved when a person suffers? How do they relate to one another? Can we identify one – or perhaps a select number – as constituting what suffering really is? For another thing, suffering seems to come in different varieties. We can, for instance, distinguish physical and emotional suffering. But, even within each category, there is considerable diversity. And how do the diverse states and experiences involved relate to suffering? One idea would be that pain and grief, for instance, are simply specific kinds of suffering – species of a genus. But is this right? Even if pain, for instance, can be suffered, or involved in suffering, is it itself a kind of suffering? These are the sorts of questions that theories of suffering aim to answer. Given the complex nature of suffering, we’ll see that there are a wide variety of theoretical options here, with the chapters in this section promising to be a stimulus for future theoretical advances. The first chapter in Part I is by Antti Kauppinen. In ‘The world according to suffering’ (Chapter 1), he develops a novel theoretical account of the nature of suffering, which highlights the role suffering plays in transforming our world. Kaupinnen begins by noting that experiences of suffering – from the loss of a loved one, say, or from intense physical pain – are very different things. This raises the following question: what makes them both experiences of suffering? Kauppinen argues that it’s not just the unpleasantness of the experience; but nor, contra Brady’s recent proposal, is it the fact that we desire not to have the unpleasant experience. (Such a view has, Kauppinen thinks, too much of an inward focus.) Instead, he wants to argue that what we suffer from negatively transforms the way our situation as a whole appears to us. To cash this out, he introduces the notion of negative affective construal, which involves practically perceiving our situation as calling for change, registering this perception with a felt desire for change, and believing that the change is not within our power. We (attitudinally) suffer when negative affective



construal is pervasive, either because it colours a large swath of possibilities, as in the case of anxiety, or because it narrows our attention to what hurts, as in the case of grief. On this view, sensory or bodily suffering is a special case of attitudinal suffering: the unpleasantness of pain causes pervasive negative affective construal. Pain that doesn’t negatively transform our world doesn’t make for suffering. Central to Kauppinen’s account is the notion of felt aversion. On his view, felt aversion registers how a subject practically construes their situation, and is a part of what he calls affective construal. Such a construal involves a perception of invitations and affordances to act: one affectively conceives one’s situation in practical terms, but also and at the same time feels a motivation to do something. Kauppinen claims that attitudinal suffering is a matter of a subject’s pervasively affectively construing her situation in a negative way. He then goes on to argue that sensory suffering is just a special case of attitudinal suffering, one that is caused by the painfulness of a sensory input. Together, these claims constitute suffering as negative affective construal (SNAC): to suffer is to affectively construe one’s situation as negative in a pervasive enough fashion, and thus to experience felt aversion or attitudinal displeasure towards it. In Chapter 2, ‘The disruption model of suffering’, Tom McClelland articulates a different account, according to which your suffering a mental state – such as pain, grief, or guilt – consists in that state’s disrupting your wider mental life. Your intense pain might, for instance, induce distress, anxiety about the pain, and rumination about both the pain and the prospect of relief, thus preventing you from going about your ordinary life. And your grief might generate such emotions as anger and fear, require the revision of deepseated beliefs and desires, and pose questions that leave you feeling incomprehension. This, for McClelland, is the stuff of suffering. Neither pain nor its intrinsic unpleasantness (if such there be) suffice, therefore, and two people with equally unpleasant pains might suffer to different degrees. Tweaks are needed, McClelland concedes, to ensure this Disruption Model is not undermined by the fact that you need not suffer when, for instance, feelings of love disrupt your mental economy. (Hence McClelland specifies that the disruption constituting suffering must be ‘holistically unpleasant’.) Suitably tweaked, he argues, his view has a number of virtues, explaining why we cannot undergo a mere instant of suffering (contrast pain), since disruption takes time to unfold; explaining why we cannot be unbothered by our suffering (again contrast pain), since our suffering consists in being bothered; and explaining how it is that diverse cases of suffering might be phenomenally quite different while sharing a certain experiential commonality, namely the experience of unpleasant ‘mental impact’. If McClelland is right, there is also a surprising and important connection between suffering and agency. Since some of a disruptive state’s mental effects (e.g. rumination about our pain) are things we do rather than things that happen to us, suffering is partly agential, McClelland argues. Does this mean,


David Bain, Michael Brady, Jennifer Corns

implausibly, that we can simply decide not to suffer? No, the idea goes, because mental actions involved in suffering are responses to powerful urges, hence not entirely voluntary. (We can try to resist the urges, McClelland acknowledges, but that struggle is itself disruptive.) What do the urges urge us to do? To perform mental actions (e.g. attending, assessing, reflecting) that will assimilate ‘difficult’ mental states, restoring mental homeostasis or equilibrium. This, McClelland argues, is the role of suffering, and he thinks that some of us are more skilled at achieving that equilibrium than others. Some of us are better sufferers than others. Chapter 3, ‘Painfulness, suffering, and consciousness’, by Sam Coleman, is a wide-ranging exploration of three things: physical pain, painfulness (understood as pain’s distinctive unpleasantness), and suffering. These are indeed, Coleman claims, three distinct phenomena. After all, there are physical pains (‘sensory’ or ‘s-pains’) that are not painful. And, unlike pain and painfulness, suffering does not ‘present as having a specific bodily location’ (see also Chapter 4 below). But what makes a state a pain, or painful, or a state of suffering? In each case, Coleman invokes phenomenology. What makes a state a pain, he claims, is its ‘painy’ qualitative character or qualia. That character, he argues, explains how we are able introspectively to tell whether we are in pain and which kind of pain we are in. A parallel argument from introspection, Coleman continues, establishes the same conclusion about painfulness. What makes a state painful is also its qualitative character. At this point, a version of the ‘heterogeneity problem’ arises: introspection arguably reveals no ‘one feeling’ that is present whenever we are in painful pain. But, Coleman replies, while introspection reveals no one determinate quale, it might well reveal a single determinable quale. In sum, pain and painfulness are qualitative matters. But what is their relation to one another? Coleman’s answer is that painfulness is intrinsic to painful pain ‘as one constituent of a bi-partite self-representing state’. Turning to suffering, Coleman insists that this too is grounded in phenomenology. Desire-based accounts of suffering fail, he thinks, since we want suffered states to stop because they involve suffering, not conversely. And he thinks McClelland’s idea (above) that suffering consists in the ‘holistically unpleasant’ disruption of one’s mental economy fails by its own lights. For McClelland must concede, Coleman suggests, that were you experiencing ‘holistic unpleasantness’ as if such disruption were occurring, you would be suffering whether or not the disruption was real. Yet Coleman resists simply identifying suffering and ‘holistic unpleasantness’, instead aiming to capture the intimate relation between the unpleasantness of the suffered state and the suffering itself by claiming that suffering is the ‘higher-order’ property of ‘instantiating some or other unpleasant first-order property’. He concludes with a striking claim: while pain, painfulness, and suffering are indeed qualitative matters, qualitative character – absent ‘subjective awareness’ – does not entail consciousness, hence there might be unconscious



pains, unconscious painful pains, and even unconscious suffering. Analgesic hypnosis, which stops the subject reporting pain but leaves her blood pressure and heart rate raised, might – he thinks – provide a case in point. The final chapter in Part I is by Olivier Massin. ‘Suffering pains’ (Chapter 4) aims to clarify a number of distinctions and relations between pain and suffering, and clarify the conceptual space in which these things are located. In so doing, Massin develops his own original theoretical take on what suffering is, and looks at the implications of this for theorising about pain. Massin begins by noting that pain and suffering are typically assumed to be very similar phenomena – which might explain why philosophy of mind tends to focus much more on the former. But Massin argues that pain and suffering are utterly distinct. In order to make this case, Massin defends three distinct theses: (1) pain and suffering are not identical; (2) pain is not a species of suffering, nor is suffering a species of pain, nor are pain and suffering of a common (proximate) genus; and (3) suffering cannot be defined as the perception of a pain’s badness, nor can pain be defined as a suffered bodily sensation. This takes up the first four sections of Massin’s chapter. For the remainder of the chapter, Massin argues for the following three positive claims. (4) Pain and suffering are categorically distinct, with pain being a localised bodily episode, and suffering a non-localised affective attitude. (5) Suffering can be expressed, pains cannot. As a consequence, we can have compassion for the suffering of others, not for their pains. (6) The relation between pain and suffering is akin to the relation between danger and fear, injustice and indignation, wrongdoing and guilt: suffering is the correct reaction to pain. Massin then argues that his discussion undermines two influential views in the philosophy of pain. The first is that the experience of pain is incorrigible: the idea that one’s seeming to have a pain entails one’s having a pain. This view has been extremely influential, but Massin argues that an upshot of his chapter is that it is false. The second is that our ordinary conception of pain is ‘paradoxical’, on the grounds that pain is both located in the body and hence objective, whilst also being mental and hence not located in the body. Massin argues that the paradox disappears once we realise that pain is a state of our body, which one experiences and reacts to by suffering.

Part II: Pain and valence One way to enhance our understanding of suffering is to focus on suffering itself: what its components are, what instances of suffering have in common, what the folk and other disciplines think about suffering, and how the concept has come down to us through history. Another way to approach the subject is to try to understand closely related concepts, and see what we can learn about suffering from them. The final chapter in Part I took this kind of approach; and the chapters in Part II carry on the project by proposing and defending novel accounts of key phenomena that are closely related to


David Bain, Michael Brady, Jennifer Corns

suffering. One of these, as we have seen, is pain, and further theoretical advances here do indeed promise to shed light on what suffering itself is. But another is valence: the categorisation of a mental state or experience as negative or positive. Valence is extremely important to our understanding of suffering, given that perhaps the most general thing one can say about suffering is that it is (for all of its benefits) negative in some sense. That seems true, if not by itself very informative. What we need for a better understanding is some account of what valence is. The picture here, however, is also complex, given competing accounts of the notion. Some understand valence in terms of affective or hedonic dimensions such as pleasure and unpleasantness, others in terms of motivation, for instance the distinction between approach and avoidance behaviour; still others think that it is a sui generis category and cannot be reduced to these other elements. What seems true is that getting a better understanding of what pain and valence are will assist us in understanding our target concept of suffering. In the first chapter of Part II, ‘Valence, bodily (dis)pleasures, and emotions’ (Chapter 5), Fabrice Teroni investigates the nature of valence and its relation to other affective elements. Valence is a property of many states, including hedonic states, moods, emotions, and sentiments. All of these can be described as being either positive or negative. Teroni thinks that valence is distinctive of affective states, and differentiates them from states like beliefs and memories. This raises a number of issues about the nature of valence. First, what is valence? And what can valence tell us about the structure of the ‘affective domain’? Teroni examines a natural and popular answer to this question, the ‘core affect’ approach, and shows how it generates and endorses a particular view about the relation between valence, bodily (dis)pleasures, and emotions. This view is captured by two theses, which he terms explanatory priority and containment. The central task Teroni takes on in his chapter is to assess the prospects for these claims. According to the explanatory priority thesis, the valence of bodily (dis)pleasures is explanatorily prior vis-à-vis the valence of emotions. In short, we can explain what makes an emotion positive or negative by appealing to the positivity or negativity of hedonic states such as pleasures and displeasures. According to the containment thesis, emotions contain bodily (dis)pleasures. The containment thesis therefore provides a neat explanation of why the valence of (un)pleasant feelings has explanatory priority. After examining the core affect approach, Teroni then looks in detail at the internal structure of bodily (dis)pleasures and emotions. He maintains that bodily (dis)pleasures are intentional states whose valence is to be understood in terms of evaluative experiences, which take place automatically and for the most part at the subpersonal level. It turns out that on this view, the valence of bodily (dis)pleasures and emotions is the same; both are a matter of perceptually experiencing the value of some object. Given this account of the intentional structure of (dis)pleasure and of emotions, Teroni gives reasons to conclude that we should adopt neither explanatory priority nor containment.



The other two chapters in Part II focus on valence in the following sense: the pleasantness or unpleasantness of experiences. In ‘Pain and mere tastes: toward an attitudinal-representational theory of valenced perceptual experiences’ (Chapter 6), Hilla Jacobson addresses pain’s unpleasantness, but indirectly, via an argument concerning the valence of other perceptual experiences. Jacobson argues that a satisfactory account of pain’s unpleasantness must extend to the valence of other perceptual experiences. And to do this, she argues, it must accommodate the extensive inter-subjective and intra-subjective variability of those other experiences’ valence, for instance the possibility that for Dinna the taste of eggplant is pleasant whereas for Dan it is unpleasant, or that for Dinna it is only now pleasant, having been unpleasant for her in other contexts. Which theories of pain’s unpleasantness can be extended to accommodate such ‘valence variability’? Jacobson considers two: evaluativism and her own attitudinal-representational theory (ART). These agree that your pain experience is a perceptual representation of a certain bodily condition, but they differ about the valence, evaluativism claiming that your pain’s unpleasantness consists in its representing the bodily condition as bad for you, ART claiming that it consists in the pain involving a negative conative or desire-like attitude directed toward the represented condition. (A model: your pain represents a condition you want not to obtain.) For present purposes, the key difference is this: if evaluativism is right, a given pain’s unpleasantness will be either veridical or falsidical; if ART is right, it will be neither veridical nor falsidical (assuming, as Jacobson does, that conative attitudes don’t admit of veridicality and falsidicality). This difference underlies Jacobson’s argument in favour of ART. How might evaluativism and ART be extended to gustatory experiences? Evaluativism, Jacobson suggests, will say that pleasant and unpleasant experiences represent given features of whatever is tasted as, respectively, good or bad. ART will say that they involve, respectively, positive or negative desire-like attitudes towards the given features. Hence, crucially, advocates of this extended evaluativism must answer a challenging question that ART can side-step. Whose eggplant experience is accurate: Dinna’s or Dan’s? If they think eggplant cannot be both good and bad at the same time, evaluativists cannot reply ‘Both’. Nor, Jacobson argues, can they credibly claim that eggplant entirely lacks the relevant range of value properties and on this basis reply ‘Neither’, since this would commit them to implausible levels of inaccuracy in our gustatory experience. Might they instead reply ‘Just Dinna’s’ or ‘Just Dan’s’? Well, Jacobson argues, if they choose one of these answers, they need grounds for their choice, and she thinks it quite unclear what grounds there could be. At this point, evaluativists might return to the start and claim that eggplant can be both good and bad in the relevant sense: good-for-Dinna-in-her-circumstances, perhaps, and bad-for-Dan-in-his. But such relativising moves also lead to difficulties, Jacobson argues. So, she concludes, we should prefer ART in the gustatory case, hence in the pain case too.


David Bain, Michael Brady, Jennifer Corns

Whereas Jacobson contrasts evaluativism and ART, Chapter 7, ‘Pain: an attitude with two heads’, by Hagit Benbaji, argues that these two positions share a fatal problem. Suppose you have an unpleasant pain (hereafter ‘a pain’) in your hand. Both ART and evaluativism claim that your pain is both a reason for you to change something in your hand and a reason for you to eliminate the pain itself, for instance by taking a painkiller. Evaluativism, after all, says that your pain represents a state of your hand as being bad for you; and, on one version of the view (she cites David Bain’s), it thereby gives you a reason both to change the state of your hand and – being unpleasant, hence (by dint of the same evaluative content) itself bad for you – a reason for you to eliminate the pain itself. Similarly, according to ART, your pain involves a subjectively frustrated desire-like attitude for your hand not to be in the condition it seems to be in; and that attitude gives you reason to change the state of your hand and also – the desire-like state being ‘subjectively frustrated’ – a reason to eliminate the pain itself. But, Benbaji argues, this shared picture is incoherent. For no single state, she argues, whether evaluative experience or desire-like state or whatever, could give you a reason to change something in the extramental world and at the same time give you another reason to eliminate the former reason. No single state, as she puts it, ‘could live up to both ideals’. For if you act on the self-directed reason, you eliminate the world-directed reason Benbaji consider numerous replies. Evaluativists, for instance, might protest that their picture makes sense provided they distinguish different kinds of reasons and insist that, whereas your pain is a motivating reason to change your hand’s state, it is a justifying reason to eliminate the pain itself. For their part, ART theorists might try to distinguish what gives you reason to change the world (the desire-like attitude) and what gives you reason to eliminate the pain itself (that attitude’s subjective frustration). But neither reply, Benbaji argues, fully meets the deep worry behind her ‘conflicting-reasons’ objection. Hence she concludes that, in fact, your pain gives you a reason to eliminate itself and only to that end a reason to change the extramental world – which, she suggests, opens the door to traditional views, on which pains lack external objects altogether.

Part III: The value of suffering The first two parts of the book address questions about the nature of suffering: which theoretical approach is the most plausible, how related concepts might help us to understand better our core concept. The last two parts mark a turn towards normative questions. In this section we look at questions of the value of suffering – about the benefits that suffering sometimes bring, and the conditions in which it does so. That suffering can have significant value is compelling but often ignored – both in popular thinking and in academic research and reflection. But arguably suffering can indeed be valuable. Those who are incapable of pain typically die young, precisely because physical pain



has significant practical value in alerting us to bodily damage and threat and in motivating appropriate behavioural responses. Grief is a fitting, hence (arguably) valuable, response to loss; disappointment to violated expectation; hunger to lack of food; shame to social disgrace; and so on. Arguably, indeed, there would be something seriously amiss and inappropriate if we didn’t respond with the relevant forms of suffering to objects and events like these. Failure to feel guilt or shame might be good evidence of psychopathy; failure to grieve suggests lack of deep and loving feelings; and so on. This is not all. For, despite its negativity, forms of suffering are often essential to, or often enhance, things like pleasure and enjoyment. Think, for instance, of how much more pleasurable a cold glass of beer is when one is parched with thirst, or a warm house after being in the freezing cold. So reflection on our own experience, and on the experiences of others, strongly suggests that suffering has significant value. The chapters in this section examine this general line, and consider the value of suffering along a number of dimensions. The first chapter in Part III, by Havi Carel and Ian James Kidd, is ‘Suffering as transformative experience’ (Chapter 8). Carel and Kidd propose that many experiences of suffering can be illuminated as forms of transformative experience – to use the term coined by L.A. Paul. On Paul’s account, certain experiences are distinctive in virtue of their transformative capacity. Such experiences can be epistemically transformative, insofar as they provide forms of knowledge that were previously unavailable and inaccessible. Equally, such experiences are personally transformative insofar as they fundamentally change one’s values and preferences, and hence one’s identity, in substantive ways. So transformation doesn’t just involve getting more information or knowledge; it changes one’s values and interests, in such a way that one’s character and identity can change. Carel and Kidd argue that many kinds of suffering experiences share these features that Paul emphasises. In particular, they focus on suffering experiences that arise from the vulnerability, dependence, and affliction intrinsic to the human condition. These features can create a variety of positively, negatively, and ambivalently valenced forms of epistemically and personally transformative experiences. The kinds of experiences of suffering that Carel and Kidd focus on are not the type mostly discussed in the transformative experience literature. While Paul and others typically focus on positive experiences – those that people choose or elect to have – Carel and Kidd focus on two categories of unelected transformative experiences, which they label non-voluntary and involuntary. They argue that suffering experiences of these types promise to be potentially more transformative than others. In order to show this, Carel and Kidd develop a taxonomy of negatively valenced transformative experiences. They suggest three features that make such experiences ones of suffering, following Michael Brady’s definition: intensity, novelty, and attentional focus. Finally, they suggest that one possible explanation for the edifying capacity of suffering comes from it requiring more transformation than positive experiences.


David Bain, Michael Brady, Jennifer Corns

In Chapter 9, ‘After motivational hedonism: feeling bad can be good | feeling good can be bad’, Colin Wayne Leach argues against an orthodoxy that he claims blinds many psychologists to ways in which suffering can be valuable and pleasurable states harmful. The orthodoxy has a number of strands that weave through the chapter: the function of pleasant states is both to alert us to ‘advantageous states of affairs’ and to motivate us to prolong them; the function of unpleasant states is both to alert us to ‘disadvantageous states of affairs’ and to motivate us to avoid them; humans want to ‘approach pleasure and avoid suffering’ (this is the ‘motivational hedonism’ of Leach’s title); and pleasurable states manifest well-being and proper functioning whereas unpleasant states are bad for us and maladaptive. This orthodoxy, Leach argues, distorts our conception of the nature, importance, and motivational profiles of pleasant and unpleasant states. Among unpleasant states, Leach focuses on shame and anger. Psychologists, he claims, tend to think of shame as bad for us, either ‘internalised’ in self-destructive behaviour or ‘externalised’ as hostility to scapegoats. And they tend to think that shame motivates aversive behaviour – avoidance of whatever ‘disadvantageous’ circumstance it alerts us to – rather than positive or ‘approach’ behaviour. But, Leach argues, this picture, arising from the orthodoxy, is inaccurate. For, in a subject who believes that the personal failure occasioning the shame is reparable, Leach claims, shame can in fact motivate not avoidance (‘running away’, ‘hiding’, ‘wishing to disappear’) but positive action, and moreover constructive action, aimed at selfimprovement, not self-destruction. So not only can shame be good for us, Leach concludes, but its motivational profile is more positive (less about ‘aversion’). constructive, and context-dependent than the orthodoxy would have us believe. Leach makes similar points about anger. What it motivates us to do depends on our beliefs, he argues; and, in those who believe that the injustice occasioning the anger can be addressed, anger can motivate constructive action, aimed at – among other things – restitution. In ‘From suffering to satisfaction: why we need pain to feel pleasure’ (Chapter 10), Brock Bastian too draws on psychological research to argue that suffering is a powerful contributor to and even necessary condition on our experiences of pleasure, satisfaction, and meaning. He begins with two observations. We tend to draw sharp distinctions among states of suffering, broadly construed – between mild unpleasantness and intense suffering, for instance, and between physical pain and ‘social’ suffering – where, in terms of underlying psychological and biological mechanisms, there is (he claims) continuity. Relatedly, we distinguish negative and positive experiences sharply, and focus on the harms of the former, regarding those who report enjoying them as aberrant. Consequently, Bastian argues, we tend to miss that even intense suffering is routinely a ‘component’ of our experiences of meaning, community, well-being, and pleasure.



Bastian turns to detail the various ways in which suffering can enhance pleasure and our lives more broadly. He argues, for instance, that our adaptability means that the pleasure of a hot spa fades unless punctuated with uncomfortable plunges into cold water. Unpleasant experiences also enable relief, he adds, which – thanks to opponent processing – consists not merely in the removal of the unpleasant experience, but in an ‘opioid overshoot’ and pleasure, hence runners’ highs. Moreover, he argues, we are more likely to remember and rate as meaningful highly pleasurable and highly painful events, which the preceding suggests are interdependent. And he presents evidence to suggest that, when we numb pain with painkillers, we thereby numb pleasures too. Turning from pleasure to other ways in which suffering can enhance our lives, Bastian presents evidence that pain can reduce feelings of guilt, and also reduce other unpleasant emotions – especially approach-related emotions, like anger – which is arguably one of the many reasons people self-harm. Moreover, Bastian argues, shared suffering – from Americans’ experience of the 11 September 2001 terrorist attacks (which produced a spike in volunteering) to group consumption of hot chilies – can increase group solidarity, trust, cooperation, support, and creativity. Bastian also presents evidence to suggest that suffering contributes to resilience – the capacity to regard adverse circumstances as challenges rather than threats – hence, in one study, rats conditioned to painful stimuli exhibited reduced ‘learned helplessness’ and, in another, humans with moderate (as against low) life-time adversity reported greater well-being and less pain when their hands were immersed in ice water. Bastian does not recommend increasing suffering. And he is careful to note the importance of individual differences in how inclined we are to think of our resources and capacities as equal to given demands, hence to experience those demands as challenges (causing adrenaline release) rather than threats (causing cortisol release). He also registers that there are more and less harmful ways to produce suffering, suggesting that self-harmers might profit from seeking pain in intense exercise instead. But his key point is this: even were it possible, eliminating human suffering entirely might undermine rather than promote our well-being. Tasia Scrutton’s focus in ‘“My horses and hogs and even everybody seemed changed”: appreciating beauty in depression recovery’ (Chapter 11) is on how suffering can have significant aesthetic value, and enhance our experience of beauty. It has long been thought that suffering can have positive benefits for the sufferer, at least once they have lived through and recovered from it. Scrutton points out that most of the ‘meaning-making’ literature and accounts of mental suffering such as depression and melancholy point to the development of moral characteristics such as compassion, increased trust and love, creativity, and an appreciation of what is really valuable, following mental suffering (see Chapter 8 of this volume for the ways in which suffering can be personally transformative and edifying). Those who have experienced and lived through suffering become morally better – more compassionate,


David Bain, Michael Brady, Jennifer Corns

trusting, hopeful, less egotistical, as a result. However, Scrutton notes that another, less-theorised theme in accounts of mental suffering shifts focus away from moral development, and towards aesthetic. Scrutton argues that those who experience mental suffering often have a heightened appreciation of beauty, in particular of the beauty of the natural word, both during their experience of suffering, and afterwards. In order to examine this idea, Scrutton focuses on accounts of recovery from depression and melancholy from a number of sources. Some of these are from published autobiographies, while others are gleaned from accounts discussed by William James. A third important source is the collection of reports from the archives of the Alister Hardy Centre for Religious Experience. Although many of these accounts are religious, and although there might be links between spirituality and natural beauty, Scrutton focuses on the more general issue of how suffering in depression and melancholy might enhance appreciation of beauty, and how, as a result, suffering can be of value to the sufferer.

Part IV: The normativity of suffering The chapters in Part III focused, in different ways, on the relation between suffering and value. But suffering is also related to other normative domains – in particular, to morality and rationality. To take the former: it is a staple of certain religious traditions – Christianity, Islam, Buddhism – that suffering can be central to moral agency and moral virtue. Think, for instance, of the ‘virtue solution’ to the problem of evil, wherein suffering is putatively justified on the grounds that it is necessary for meaningful choice, and the moral and spiritual development that is grounded in this. Or think of ideas prominent in the Koran and the Bible of the ways in which suffering is just punishment and so facilitates social virtue, or is a test of faith and so strengthens trust and commitment to the divine. Moreover, certain forms of suffering – such as grief and shame – are central to moral development and having a good moral character. To take the latter: forms of suffering can arguably enhance, as well as hinder, our capacities for rational reflection and problem-solving, our ability to make sense and meaning of our lives. Forms of suffering can promote and secure long-term prudential goods, and so be strategically rational – as when remorse motivates us to apologise and make reparations, and thereby facilitates our re-entry back into the social group. Often this takes place in opposition to our conscious and rational reflection on what it’s best for us to do. The chapters in this section all focus, in various ways, on the connections between suffering, moral agency, and rationality. In Chapter 12, ‘Hedonic rationality’, Jennifer Corns argues that the pleasantness and unpleasantness of our mental states is ‘reason-responsive’. We can have reasons not only to believe and to act and even to undergo certain emotions, she argues, but also to ‘feel pleasantly or unpleasantly’. In support of her thesis, she describes a case in which a subject, Glenda, grieves. It



would, Corns claims, be quite natural to say that the unpleasantness of Glenda’s grieving states (perceptions, feelings, thoughts, and imaginings) is rational. And in other cases, she argues, it would be quite natural to say that the unpleasantness of a subject’s states is irrational or even unintelligible. But Corns’s argument does not turn only on what it is natural to say. She argues that key ‘indicators’ of reason-responsiveness apply to the ‘hedonics’ (the pleasantness and unpleasantness) of mental states. For instance, she argues, we can ask after and sometimes find reasons that speak in favour of one’s mental states being unpleasant, or ways in which the unpleasantness coheres with one’s other states. Moreover, she argues, the most threatening objections to her conclusion can be countered. Some, for instance, will object that reason-responsive states must be voluntary, or content-bearing, or have an aim, or be non-perceptual. But in each case Corns argues either that the requirement is in fact met by pleasantness and unpleasantness, or that it should be abandoned. For instance, while on one conception of voluntariness, unpleasantness is involuntary, so too are beliefs – the very paradigms of reason-responsiveness – yet on another conception, which explains voluntariness in terms of rationality, beliefs count as voluntary after all, but so too (on Corns’s argument) does unpleasantness. Corns’s concludes with her view’s potential implications. Her thesis might, she says, suggest that the unpleasant dimensions of certain mental states are more susceptible to ‘rational interventions’ than we tend to think. It might also suggest, she adds, that the very distinction between what is and what is not reason-responsive should be abandoned. The second chapter of Part IV also addresses rationality, but from quite a different angle. In ‘The agony of reason’ (Chapter 13), Matthew Fulkerson and Jonathan Cohen argue that ‘states of suffering’ occupy a distinctive place in our rational lives. On the one hand, they rationally motivate. On the other, they are not reason-responsive, being distinctively resistant ‘to our broader rational concerns’. Consequently, the authors claim, while suffering states play a central role in our practical deliberations, those states’ integration in our mental lives is peculiarly limited. They can seem like (even though they are not) ‘a non-rational, outside influence’. To illustrate, Fulkerson and Cohen contrast two cases, Beach and Vaccination. These involve competing desires: in Beach, to relax and to meet a work deadline; in Vaccination, to be healthy and to avoid a painful injection. In both cases, you deliberate and decide to act on the former desire, hence going to the beach and having the vaccination. Yet, Fulkerson and Cohen argue, there is an important difference. Although in both cases the rejected desire continues to be a pro tanto reason to perform the rejected action, and in both it might continue to exert some motivational ‘pull’, that pull is at least weakened by your decision in Beach, whereas there is no such weakening in Vaccination. When the needle touches your arm, your anti-pain desire ‘continues to exert control … even after rational deliberation has done its work’. You might have the vaccination anyway, but only after ‘fighting through’ the pain.


David Bain, Michael Brady, Jennifer Corns

This ‘stickiness’ of suffering cannot, the authors argue, be reduced to more familiar phenomena such as akrasia or non-intentional action. In Vaccination, they argue, you do not ‘give in’ to the anti-pain desire against your better judgement, but ‘fight against’ it. And even had you acted on the antipain desire, your pulling away from the vaccination would not have been a ‘pain-induced reflex’, since what suffering ‘forces’ is not movements, but decisions to act. So the stickiness of suffering is real, Fulkerson and Cohen conclude. Moreover, it is explanatory. For it contributes, they suggest, to the depression and other ‘secondary suffering’ that chronic pain can induce. Those with chronic pain, on this picture, know their pain serves no purpose and should not motivate them, yet it does, thereby intruding into their thoughts and impeding their actions. (There are connections here to Chapter 2 of this volume by McClelland.) In Chapter 14, ‘Some paradoxes of pain for rational agency’, Marilyn McCord Adams examines the tension between the idea that pain is necessary for moral agency and meaning-making, but at the same disrupts both of these. On her view, pain systems are at the biological level ambivalent. Pain has important cognitive, motivational, and educational roles; and first-hand experience of pain is central to empathic engagement with others and the moral consideration and agency that flows from this. Moral competence requires us to be sensitive to and empathise with the pain and suffering of other people. At the same time, the worst types of pain and suffering are overwhelming and undermine our capacity to function as moral and indeed rational agents altogether. By the same token, pain and suffering are important for meaning-making: they are ‘plot complicators’ that lead to creative problem-solving and bring out the best in people. We find those who have experienced and overcome pain and suffering more interesting, more admirable. At the same time, pain and suffering can disrupt attempts to make sense of or give meaning to one’s life: they are socially alienating, sources of trauma, things that can destroy positive meaning. So pain and suffering are central to, yet threaten to disrupt, meaning-making. Adams then examines these paradoxes from the standpoint of traditional Christianity. Can God appreciate what it is like for us to suffer, and how bad this is? Can God experience pains and suffering that are overwhelming? She introduces Linda Zagzebski’s attempt to address this puzzle by appeal to God’s ‘omnisubjectivity’ – that God knows what it is like for each creature to have the experiences that they do, while being aware that God is not any of these creatures. Adams thinks that this puts God in a position to resolve the paradox of pain, but she raises worries when it comes to the problem of evil. These might be addressed if we recall that God has the capacity to compensate us for the pain and suffering we experience. Similarly, the notion of God creating us to be capable of making positive sense of our pain and suffering, whilst allowing that pain and suffering can be destructive of our capacity to make meaning, seems problematic. Here again, though, God might be viewed as answering pain-generated despair with solidarity in suffering, through the



person of God the Son. This can function to give positive meaning to the suffering of humans, for those very humans. As a result, the God of traditional Christianity helps us to escape, at least to an extent, the paradoxes of pain. The final chapter in the volume focuses on the relation between suffering and moral character. In ‘Suffering as a virtue’ (Chapter 15), Michael Brady proposes that suffering can itself be intrinsically valuable. In particular, he maintains that experiences of suffering can constitute virtuous motives, and that dispositions to suffer can constitute virtues of our physical and emotional systems. The idea that forms of suffering can be virtuous shouldn’t strike us as too bizarre, he argues, given that things like pain and remorse can be appropriate or fitting responses to disvalue, such as bodily damage and moral wrongdoing, and can enable us to deal with them effectively. Brady borrows from Linda Zagzebski’s account of virtue – according to which virtues involve affective motivational elements which are reliably successful in bringing about some end – to flesh out his account of suffering as virtuous. For the negative feelings at the heart of things like pain and remorse have clear motivational value; and a disposition to feel these things in the right circumstances to the right degree can enable the subject to be reliably successful in protecting the body, and making appropriate reparations. Brady then defends his view against two classes of objection: that the relevant feelings cannot be virtuous because they are neither acquired nor part of our characters; and that forms of suffering cannot be virtuous motives because they are intrinsically bad, and virtuous motives are intrinsically good. In addressing these objections, Brady develops the idea that forms of physical suffering can constitute faculty (rather than trait) virtues, and that they can count as intrinsically valuable according to Thomas Hurka’s ‘recursive’ account of value, insofar as they are forms of hating, or being against, what is bad.

Part I

The nature of suffering


The world according to suffering Antti Kauppinen

The experience of suffering can take many forms. Consider the following fragment of Andrew Solomon’s description of his major depression: I felt that my mind was immured, that it couldn’t expand in any direction. I knew that the sun was rising and setting, but little of its light reached me. I felt myself sagging under what was much stronger than I … In depression, all that is happening in the present is the anticipation of pain in the future, and the present qua present no longer exists at all. (Solomon 2001, 18, 29) Or consider, for contrast, my experience when a wisdom tooth got infected a few years ago. At first, there was just a soreness in the gums. When it started getting painful, I took a painkiller, thinking that the feeling would soon pass, as had happened before. But it didn’t, and I booked a dentist for a few days later, taking another pill. Alas, it had no effect. Soon I could think of nothing else. I couldn’t sit, or stand, or lie on the floor. The kids’ everyday requests – help with this, give that – irritated me no end. I tried to lock myself in my room and listen to music, but I couldn’t concentrate. Only one topic fit in my mind. It was getting late, and in panic, I went online again to find a dentist who would be on call. After a few failures, I got through to one who was at home with his kids, but agreed to meet me at his clinic in half an hour. I drove there in a hurry, and he gave me a shot of anaesthetic and booked me for an operation in the morning. I could have fallen on my knees to thank that beautiful man.1 Given that suffering comes in many forms, from Solomon’s depression and my mercifully short-lived agony to grief and loneliness and hunger, it’s a good question to ask what unifies them – what makes them all instances of suffering, and as such prima facie bad for the sufferer, with further motivational and normative consequences down the line. As Michael Brady (2018) has recently persuasively argued, it won’t do to appeal to simply to the unpleasantness of the experiences, since we need not suffer even if we undergo an unpleasant experience, even if it is an intense one. But Brady’s own proposal, on which suffering is roughly a matter of having unpleasant experiences we


Antti Kauppinen

occurrently desire not to have, has a problematic inward focus, or so I’ll argue. I claim that what is essential to suffering is instead that what we suffer from negatively transforms the way our situation as a whole appears to us. To spell this out, I introduce the notion of negative affective construal. It involves three key components: practically perceiving or conceiving of our situation as calling for change, registering this perception with a felt desire for change, and believing (or perceiving) that the change is not within our power. It is thus simultaneously a matter of how the world appears to us and how we are poised to act with respect to it. As many have recently pointed out, it can be an intrinsically unpleasant experience to desire that not p and to believe that p; it’s worse yet to desperately wish that not p and see no way to bring it about that not p, especially when this involves construing oneself as faulty in some way. In experiences of suffering, negative affective construal is pervasive, either because it colours a large swath of possibilities or because our attention is narrowed to what we’re averse to. Forms of what I’ll call attitudinal suffering, such as depression or grief, are themselves constituted by specific kinds of negative affective construal. In contrast, sensory suffering is pervasive negative affective construal caused by experiences like toothache or hunger. Sensory suffering is pain that has a meaning for the subject in this sense. In effect, I’m going to claim that sensory suffering is a special case of attitudinal suffering, a negative transformation of the experienced self and world caused by unpleasant bodily experience. Pain that doesn’t cause such a transformation doesn’t make for suffering, however intense it is.

Approaching suffering When I talk about suffering in this chapter, I mean it in the experiential sense, in which some experiences constitute suffering. As Brady (2018) notes, this is important to emphasize, since the term is also used more broadly for any kind of harm that might occur to something, for example when we say that a car suffered damage from a collision. As a kind of experience, suffering is always psychological (or mental). It is strictly speaking a misnomer to talk about ‘physical suffering’, although some suffering obviously has bodily causes. (I will instead speak of sensory suffering below.) So what makes a mental experience one of suffering? Here are some platitudes that can serve as tentative fixed points:     

Suffering is unpleasant. Suffering is intrinsically bad for the sufferer, though not necessarily allthings-considered bad for her. Anyone has a pro tanto reason to relieve anyone’s suffering, if they can. Animals and children can suffer, not only adult humans. It is possible for a person to desire that she herself suffer, for example because she thinks that it’s a fit punishment.

The world according to suffering 


We can suffer from many kinds of things, including pain, hunger, exhaustion, the loss of a loved one or a job, lack of promising future prospects, lack of friends, injustice, or lack of meaning.

I think all these platitudes are correct. First, suffering has a hedonic dimension. Second, it is bad for you to suffer, as far as it goes, even if without suffering, you would miss out on something of great value, so that putting everything together, the state of affairs in which you suffer is better than its alternatives. Third, and related, everyone has some reason to relieve your suffering, though the strength of the reason varies considerably depending on their relationship to you and their possibilities, among other things, and it may be outweighed by other reasons for action. Fourth, in coming up with an account, we should not overintellectualize, since the range of suffering subjects is broad. Fifth, even though suffering is intrinsically bad and you have reason to avoid it, it is nevertheless something you might intelligibly seek in special circumstances – indeed, on some first-order views, it might be merited. The last item on the list calls for some clarification because of the slippery grammar of ‘suffering from’. I’m using it here to indicate the source of the suffering, the thing that makes us suffer and that we’d rather not be the case. When you suffer from a significant loss, you might be grieving, and we could also say that you suffer from grief. When you suffer from lack of promising future prospects, you might be depressed, and we might then say you suffer from depression. But in the latter use, the ‘from’ indicates something different. Grief and depression aren’t sources of suffering, but ways of suffering. They may constitute it. This distinction will be important in my argument. Considering these platitudes makes it very plausible to maintain that suffering is not identical with pain. The first reason is that as is now widely accepted, not all experiences pain are unpleasant or bad. It is common in pain science to distinguish between sensory-discriminative and affectivemotivational aspects of pain (e.g. Grahek 2007). Cases of pain asymbolia, in which a person reports feeling pain in response to bodily damage while not finding it painful, are often seen as evidence of this. Second, as Brady (2018) points out, some forms of suffering, such as exhaustion or anxiety, are not aptly described as painful (although they are unpleasant). Third, bodily pain is a bad model for experiences like depression or ennui, which lack the kind of localizable focus that bodily pain has, and don’t strictly speaking hurt, though they are certainly unpleasant. Finally, not all pains that do hurt amount to suffering – if I step on a Lego block the kids left on the floor, it sure hurts, but I couldn’t claim I’m suffering. As Brady (2018, 23) observes, some of these considerations generalize to the broader category of unpleasant experiences. You can be tired or hungry or lonely, for example, without suffering. Brady successfully dismisses suggestions that suffering is determined by the intensity of unpleasantness or the importance of the object of negative affect for the person.


Antti Kauppinen

For example, boredom can amount to suffering in spite of not being intensely unpleasant. Instead, he develops the thought that we suffer when we have unpleasant feelings and mind having them. Here is a more precise definition: Suffering as undesired unpleasant experience Suffering […] involves two essential elements: (i) an unpleasant feeling or experience of negative affect, which is a central part of our experiences of pain, grief, loneliness, hunger, and the like; and (ii) an occurrent desire that this unpleasant feeling or negative affective experience not be occurring. (Brady 2018, 29) In the light of what follows, it’s important to take note that Brady defines occurrent desire in functional rather than phenomenal terms, by reference to effects on attention, motivation, and deliberation (ibid.) Does this account meet all the desiderata for a theory of suffering? While it certainly makes progress, I don’t think it does. The core problem is that it is inwardly focused. In particular, the key elements are unpleasantness of a feeling and a desire directed at one’s inner state. This model arguably better fits experiences like bodily pain. (I’ll return to this below.) But it is problematic when it comes to negative emotions, which are world-directed. Consider here the case of grief. Brady himself observes that someone who grieves a loved one can suffer experientially. But must such a person have an occurrent desire that they don’t have this unpleasant experience in order to suffer? Surely not. It would be curious if a mother who lost her child in a tragic accident wanted her unpleasant state to end. She may well not want that – insofar as she focuses on her grief, she finds it the right response to her situation.2 (Indeed, one’s grieving is interrupted when one reflects on one’s own state rather than focusing on one’s loss.) She does, to be sure, have an occurrent desire, or rather a wish, that her child miraculously came back unharmed, or something of the sort. Quite possibly, everything around her may seem to cry out for the child to be climbing and dancing and playing again, and nothing she casts her eye on affords fun or even work. What she wants is for the world to be different, to go back to what it was, not that her own experience of the world is different. When the grief pervades the way she experiences the world, I believe this suffices perfectly well for suffering. What we need, then, is an account that captures both sensory and worlddirected suffering. I think that John Hick’s proposal is a good starting point. He says: ‘I would suggest that by suffering we mean that state of mind in which we wish violently or obsessively that our situation were otherwise’ (Hick 1966, 354). I think this nearly gets things right when interpreted suitably. As I will myself put it, to suffer is, in rough terms, to affectively construe one’s whole situation negatively, which is registered by a pervasive felt aversion towards it.

The world according to suffering


Suffering and felt aversion To begin unpacking the suggestion that suffering should be understood in terms of affective attitudes towards one’s situation, let’s consider what is and what isn’t part of one’s ‘situation’ in the relevant sense. The key to a principled distinction is the difference between my own take or perspective on things and what I take a stand on. Facts about the world independent of me, such as the fact that a loved one died, are clearly part of my situation. (Perhaps what makes them part of my situation is that they bear on the satisfaction of my concerns, or the things I care about.) But so is my pain. The pain I feel when I bang my knee doesn’t constitute my take on banging my knee.3 It’s just an unfortunate fact about my situation. This is important, because it allows for a unified account of sensory and thought-dependent suffering along Hick’s lines. In each case, there’s an aspect of our situation that we fervently wish was different, whether it’s the searing pain resulting from an infected wisdom tooth or the fact that we seem to lack any promising prospects. The pain asymbolic lacks such a wish, so she doesn’t suffer even if she has pain. The proposal also handles suffering from loneliness, since one need not have a localizable hurt to violently wish to be close to someone. And it makes good sense of grief, since the grieving mother really, really wishes her child was still alive, although she doesn’t want her unpleasant feeling to go away, given that her wish cannot be granted. We must, however, say more, since it is obviously possible to wish that one’s situation were otherwise without yet suffering. That’s no doubt why Hick talks about violent or obsessive wishing. But what do these qualifications amount to? It can’t be just the motivational strength of desire that is at issue – again, it is possible to be highly motivated to change one’s situation without suffering, and in any case wishes need not be motivating at all in one’s actual circumstances, since we may wish for what we know to be impossible, as Aristotle already observed (Nicomachean Ethics 1111b22). We must instead draw on a different dimension of intensity. To get at the relevant sense, we should begin by observing that we seem to speak of desires in two different senses. As Chris Heathwood notes, this is supported by the intelligibility of the following response a person might give to their partner while discussing common plans: ‘I don’t want us to do what I want us to do; I want us to do what you want to do’ (Heathwood 2017, 9; cf. Parfit 2011, 43). There’s no contradiction here, because we can talk of wanting in two different ways. In the first or functional sense, for me to desire to F is just for me to be disposed to (try to) F to some degree, or to be in a state with a world-to-mind direction of fit (Smith 1987). Whenever I voluntarily F, a desire to F can be attributed to me (Nagel 1970). This is the sense in which I can want to do something I don’t want to do. In the second, phenomenal or felt sense, if I desire to F (or want p to be the case), I have a genuine attraction to F-ing, or p being the case (see e.g. Oddie 2005, Aydede and Fulkerson 2018). I view it with some enthusiasm, and take pleasure in the prospect. In the phenomenal sense, I may well not want to do something I


Antti Kauppinen

nevertheless voluntarily do, for example for the sake of our friendship. In this sense I can also want something I know to be impossible (in which case we often talk about wishing). While my desires to pay my bills or to catch the 9.15 bus are typically flavourless forms of goal-directedness, we all know that some desires involve longing or aching or yearning for their object. A Springsteen fan wanting to get a ticket to the last show of his Broadway run is in a distinctive kind of phenomenological state. The same goes for aversion, mutatis mutandis. Waiting for results from a sample, I really, really don’t want the lump in my throat to turn out to be cancerous. There’s a recognizable phenomenology here, too. And if it turns out that it is cancer, my wish for it to be otherwise is not merely flavourless goal-directedness. In my view, it is such experiences of felt aversion to one’s seeming situation that play a key role in explaining the unpleasantness of suffering.4 Part of what it is to suffer is to want something that seems to be the case not to be the case and to take it that at least for the time being, there’s nothing one can do to change this aspect of things, even if it’s only because of being so damn weakwilled or cowardly. Suffering seems to always involve a sense of lack of control over the unwanted situation (and possibly one’s own action). In such a scenario, one has a desire that is subjectively frustrated, as we might say – after all, it could be that unbeknownst to us, things actually are as we want. (Maybe the loved one survived the accident after all, and is just about to come back.) But even so, we may suffer, as long as we think otherwise. I emphasize that the claim isn’t that it suffices for suffering that one’s desires are subjectively frustrated, or that what we call feeling frustrated is special. In such experiences, our attention is focused in part on what we want. If we’re feeling lonely, what we want is human connection, not getting what we want. Nevertheless, when we desire for things to be other than we think they are, and can’t for the present do anything to change them, we experience what I call a felt aversion to our situation. A felt aversion to one’s experienced situation is an intrinsically unpleasant state. It’s not just a matter of having a frustrated desire plus some displeasure, but an experience that has the phenomenal content it does in virtue of its intentional contents. This kind of unpleasantness is not external to one’s subjective take on the situation, but an aspect of it. It is at least similar to what Fred Feldman calls attitudinal displeasure, a hedonic attitude of being displeased that something is the case. Though he says that such attitudes need not have any feel (Feldman 2004, 57), he apparently only means that they lack a sensory feel (according to Lin 2018, n20). If such attitudes are not reducible to felt desire for things to be otherwise, they are another candidate for constituting felt aversion. In these terms, when you suffer, you’re attitudinally displeased with the way you take things to be.

Affective construal and attitudinal suffering When we suffer, we wish things to be otherwise but can’t seem to be able to do anything about it at least right now, and this is unpleasant. But it is also

The world according to suffering


characteristic of suffering that it transforms our world in the sense of the horizon of our experience. I’ll argue next that in attitudinal suffering, the felt aversion registers how we practically construe our situation – indeed, it is an inseparable part of what I call affective construal.5 To explain what I mean by affective construal, I’ll begin with the broader notion of practical construal. Theorists of perception have long observed that we don’t simply perceive our environment in terms of neutral qualities like shapes and surfaces. We also perceive what J. J. Gibson termed affordances, or potentials for action. He argued that affordances are neither objective nor subjective, but rather relational. Here is one of his examples: If a surface is horizontal, flat, extended, rigid, and knee-high relative to a perceiver, it can in fact be sat upon. If it can be discriminated as having just these properties, it should look sit-on-able. If it does, the affordance is perceived visually. (Gibson 1979, 120) For Gibson, affordances are the primary object of ordinary perception. Such perception of possibilities for action, which may also be a part of imagining non-actual scenarios, is a key part of what I call practical construal of a situation. (Here ‘perception’ is used in a rather broad sense, which may not entail causal impingement on the subject.) But there is also another important aspect of practical construal: the perception or conception of invitations to act, not just possibilities for acting. Like perception of affordances, this is an integral part of how we ordinarily relate to the world. To modernize an example by Kurt Koffka (1936), suppose your phone makes the incoming message sound. Most likely, you can’t help hearing this sound as inviting you to check the message, even if you’ve resolved not to check your messages until after you’ve finished your task. The phone appears to you as to-be-checked – it has a ‘demand character’, in Kurt Lewin’s (1935, 77) terms. For the gestalt psychologists, the ‘silent organization’ of the perceptual field in terms of demand character is correlated with the subject’s needs or concerns, even if not necessarily occurrent motivation (e.g. Koffka 1936, 360ff.). Similarly, Susanna Siegel (2014) has recently argued that perception of what she calls soliciting affordances can come apart from being motivated – she holds that we might, for example, hear a song as inviting us to dance without having the least inclination to do so. (Note that the modality of the call to action can vary, as the contrast between the terms ‘invitation’ and ‘demand’ suggests; I’ll mostly use ‘invitation’ as a general term here.) So, I’ll take it that practical construal comprises both the perception (or conception) of opportunities and invitations to act, of affordances and solicitations. On the view of emotions that I favour, their intentionality can be understood in terms of motivationally registered practical construal, which I’ll label affective construal. While I can’t make a proper case for this here, I will


Antti Kauppinen

give some tentative arguments. The first part of the case appeals to phenomenological contrast. Simply put, the claim is that the same scene will appear to a subject differently depending on his emotional state. Suppose I see Peter slipping on an icy footpath and falling in a comical fashion. First, I regard the scene with amusement, but then I realize he can’t get up. Suppose I respond to this by feeling pity. Here is how Sartre describes the effect: ‘I feel pity for Peter and I come to his aid. For my consciousness, one thing alone exists at that moment: Peter-having-to-be-aided. This quality of “having-tobe-aided” is to be found in Peter’ (Sartre 1936/2004, 18). When I pity Peter, I see him as inviting assistance in virtue of his hurt and helplessness. The onset of pity involves changing my practical perception of the situation. Indeed, my attitude toward Peter wouldn’t be pity without some such change in either perceiving or conceiving of him. Similarly, consider looking at a snarling dog, which I first believe to be in chains, but then realize is actually free. Part of fearing the dog is a transformation of the situation as it appears to me. The dog now appears at to-be-fled-from, and simultaneously my environment appears not to afford flight (otherwise I wouldn’t be afraid, but just leave). So I think that Sartre and others in the phenomenological tradition are right in saying that emotion is a way of apprehending the world (Sartre 1939/ 1948, 52). Each emotion is a mode of consciousness of its target, or possibly of the subject’s situation in general.6 However, what makes a construal affective is not just conceiving one’s situation in practical terms, but also occurrent, felt motivation to act accordingly. In the case of basic emotions, this is plausibly related to bodily action readiness, as Rebekka Hufendiek has recently argued. For her, emotions are ‘embodied action-oriented representations of affordances’ (Hufendiek 2015, 161). On her naturalistic account, evolution has provided us with bodily reactions that prepare us to act in functional ways in response to invitations and affordances.7 Interestingly, something similar can be found in Sartre, according to whom ‘in emotion it is the body which, directed by consciousness, changes its relations with the world in order that the world may change its qualities’ (Sartre 1939/1948, 61). However, while these suggestions are very plausible when it comes to basic emotions, not all emotions seem to involve bodily action readiness – think of envy, for example. But I think that occurrent emotions do necessarily contain something related, namely felt desires or aversions. It’s not contingent that in envying someone, we both construe something she has as to-be-made-mine or at least to-be-taken-from-her (and ourselves has having a lack-to-be-filled), and have a phenomenal desire or wish to get what the other has, or for her not to have it.8 (What makes envy unpleasant is that we simultaneously construe the situation as not affording an answer to these invitations.) In the language I’ve been using, the phenomenal desire registers the summons.9 In emphasizing the judgment-independent intentionality of emotion, this type of view is evidently related to perceptualist accounts, which take emotions to consist in perceptual experiences of value. But while for perceptualists, fear, for example, is a perceptual experience of its target as fearsome or

The world according to suffering


dangerous, according to the present proposal, fear presents its target as to-befled-from and as not-affording-safe-continuation (for example) and the situation as not-affording-flight. These are lower-level properties that are more directly linked to potential action. As such, they can be grasped by creatures without evaluative capacities. For example, based on their behaviour, small babies and frogs seem to be guided by their perception of affordances rather objective physical qualities (Gibson and Walk 1960; Ingle and Cook 1977). Of course, as we acquire concepts, it becomes possible for us to construe situations in more abstract terms and respond in more subtle ways. As far as we know, not many creatures can construe an act as something to-be-apologized-for, and be moved to apologize in response, for example. Nevertheless, there are excellent evolutionary reasons to think that even if the facts to which our emotions respond to are evaluative, the primary function of emotions isn’t to inform us about them, which is roughly what perception does, but rather to prepare us to act in a way that is appropriate in the light of our concerns, whether hard-wired, inculcated, or chosen. This is not to deny that they can be attuned to what is valuable, and consequently serve an informational role, if the background concerns are themselves evaluatively fitting. For example, someone who has a moral concern for fairness may rightly construe a union-busting law as to-be-resisted without being able to put a finger on just why, and consequently judge it to be unjust.10 Although emotions don’t represent their targets as good or bad, if I’m right, they are nevertheless valenced, positive or negative. In the light of the above considerations, valence can be understood in two ways. First, valence can be concern-based. As Roberts (2003), among others, has emphasized, underlying emotions are background concerns: we wouldn’t fear for someone if we didn’t care about them, nor do we take joy in a rival’s victory. Such concerns need not be obvious to us – indeed, they might be revealed by the very fact that we find ourselves feeling in a certain way because of an event. Along these lines, Lazarus (1991) proposed that emotions are negative when they result from (conscious or unconscious) appraisals of events as goalincongruent.11 Adapted for the practical construal model, this can be cashed out in terms of construing the salient situation as not satisfying or affording the satisfaction of one’s concerns.12 I care about certain things, where caring entails background desires, among others, to protect, promote, or serve the target, depending on its nature. When we practically construe the situation as frustrating such background desires and register this with a (new) felt desire for it to be different, our emotion will be a negative one. The second, less obvious kind of valence might be labelled immanent. It is illustrated by experiences like fear, which involve practically construing one’s situation as calling for action that it doesn’t afford. This mismatch, when registered by felt desire for the situation to be otherwise, gives rise to a negative emotion. Similarly, if the smoothness of the road, the power of your engine and the brightness of the day summon you to step on the gas, but the presence of overly cautious driver in front of you blocks you from responding


Antti Kauppinen

accordingly, it is no wonder if you respond with irritation, and with anger if you take the driver to be responsible for the non-affordance. We can capture these phenomenological and functional considerations as follows: Emotions as valenced practical construals Emotions are (at least) valenced practical construals registered by felt desires or aversions, often associated with bodily preparedness to act. I don’t pretend that this is anything worth calling a theory of emotion – perhaps it’s better thought of as a desideratum: any good theory of emotion should explain the phenomenological and motivational features it sums up. It generalizes beyond emotions strictly speaking to moods and broader affective states. After all, what do stress, anxiety, depression, and loneliness, say, have in common with standard negative emotions? My answer is not that they are unpleasant (though that is also true), but that they also involve construing one’s situation as not satisfying or affording the satisfaction of one’s concerns, or a mismatch between invitations and affordances. For the depressed, very little in the present or future is inviting, and nothing seems to afford anything worthwhile. (Consider the quote from Solomon in the introduction.) As Matthew Ratcliffe (2014) emphasizes, there is a felt loss of certain kinds of possibility, and consequently a sense that one doesn’t belong in one’s world.13 For the anxious, in turn, everything appears as to-be-watched-out for, which is manifest in bodily tension and jumpiness. When does negative affective construal amount to suffering, then? Not all negative emotions and conditions involve suffering. You can be afraid, feel lonely and, perhaps, be depressed without suffering (though that would be a borderline case of depression). Nor is the answer to be found in the intensity of negative construal. I could, after all, be very angry and violently wish someone to be punished without suffering. But if we suppose that negative affective construal is pervasive in either of two ways, it does seem to amount to suffering. First, it can have a broad scope, like depression or anxiety. Wherever I look, I see options that appear pointless or hopeless, leaving room for little motivation other than a wish for change, or threatening and hard to evade, resulting in felt motivation to hide and proceed with caution. Any positive affective construals are crowded out or short-lived. Vice versa, if you’re lonely without suffering, you do wish you could share your world without someone, but there are nevertheless many other things in your practical horizon summoning your attention and interest. Second, negative affective construal can be pervasive in the sense that my thoughts revolve around its object. It becomes the center of gravity of my attention. This is the way we characteristically suffer in the case of grief, jealousy, or grievance. Everything reminds us of the loved one lost – the song on the radio, the smell in the kitchen, the creak of the stairs – and receives a melancholy cast (in more technical terms, they appear to call for the presence

The world according to suffering


of the loved one, which the situation doesn’t afford). This is evidently a matter of degree – even the most jealous person is occasionally distracted, but just how often and for how long will vary. It is worth emphasizing that pervasive negative construal of our environment and opportunities is also reflected in how the subject sees herself. After all, there is a correspondence between what my environment affords and what I can do. If we attend to ourselves, we show up as unable to transform our situation in the desired way, and quite probably as inadequate or faulty. Generally speaking, then, pervasive negative affective construal of the situation is reflected in negative self-construal. Here is my claim in thesis form: Attitudinal suffering For S to attitudinally suffer from X is for S to pervasively affectively construe her situation (and typically herself) negatively in virtue of X. The source of suffering, then, is whatever is the source of pervasive negative affective construal. To relieve suffering is to bring about a change in affective construal – among other things, by changing the subject’s situation, changing the subject’s beliefs about it, changing background concerns (this is the Stoic way – ceasing to care about temporal things), reducing the pervasiveness of the construal (by, say, broadening the scope of one’s attention to include good things), or manipulating the motivational states that register practical construal (which is how some medications work, I suspect).

Sensory suffering What about sensory suffering, then? Recently, it has become popular to think that pains are representational states – for example, my toothache represents the tooth as damaged. One common argument appeals to naturalist tracking theories of intentionality, according to which, roughly, mental states in general represent the items that trigger them under optimal conditions (Cutter and Tye 2011), or the items that they have been ‘set up to be set off by’ (Prinz 2004) in the course of evolutionary history. Since my toothache is just about invariably caused by tooth damage, and has no doubt evolved as a signal of just such damage, it appears to fit the picture. And of course, my toothache motivates me to do to the dentist (to get the tooth fixed), and not just to take a painkiller (get rid of the unpleasantness). Yet there are also obvious differences between pains and standard representational states like the belief that my tooth is damaged or fear that my tooth is damaged. We don’t say that pains are correct or incorrect (not to mention true or false), or rational or irrational, or reasonable or unreasonable. In my view, this is not just the result of some quaint prejudice. One way to capture the difference is to note that pains are not judgment-sensitive attitudes, like beliefs and emotions. They don’t involve a commitment to


Antti Kauppinen

things being so-and-so, or construe things as being in a certain way. They don’t from part of my perspective on the world, but are part of the world that I have to deal with. If we accept a tracking theory of representation, perhaps we must distinguish between representing p and having p as a correctness condition. Let’s grant, in any case, that pains somehow signal bodily damage, even if they don’t constitute our perspective towards it, and thus form part of our situation.14 In the standard case of pain, then, we can distinguish three related items: the pain sensation itself, the damage that causes the pain, and finally the unpleasantness of the pain. If we allow that the pain sensation itself need not be experienced as painful, it seems to be ruled out as either constituent or source of sensory suffering. The same goes for bodily damage itself, though of course it can be a source of attitudinal suffering, whether or not it is painful. If my fingers were cut off or immobilized, I could no longer play the guitar, which would likely make any situation that highlighted the impossibility very unpleasant – for example, going to band practice and watching others do their thing would probably result in pervasive negative affective construal of my situation, and thus in attitudinal suffering. It seems that distinctively sensory suffering must have to do with unpleasant sensory experiences, from painful pains to persistent hunger or the muted soundscape of partial hearing loss. What makes them unpleasant could be a specific hedonic quality or tone present in each experience (Bramble 2013), or a phenomenal anti-damage desire together with seeming damage (Aydede and Fulkerson 2018; Jacobson 2018), or the experience of intrinsically (and perhaps phenomenally) desiring not to have a sensation while having it (Brady 2015; cf. Lin 2018), or perhaps an evaluation of the bodily damage as bad (Bain 2013). As Brady (2018) observed, it’s implausible that what makes these experiences of suffering is the intensity of the unpleasantness. Take the aural experience that results from partial hearing loss due to fluid in the middle ear. All sounds are muffled, inaudible, or booming, as if one were underwater. Conversations are hard to follow and participate in, and music sounds unlistenable. While the experience isn’t painful or intense, when it persists and impacts significantly on how we experience possibilities, it can make for suffering. Brady’s suggestion was that we suffer when we have unpleasant experiences we mind. I criticized the view for its inward focus, but here the thought does seem apt. But I don’t think minding in the sense that makes for suffering is just a matter of desiring not to have the unpleasant experience. Rather, in sensory suffering, we are attitudinally displeased by the attitude-independent fact that we’re having an unpleasant experience, and this has pervasive effects on how we construe our situation. On this account of sensory suffering, we suffer from pain or tiredness much as we suffer from the perceived loss of a loved one. It becomes the feature of our situation that explains why negative construal pervades our experience of the world. There is, then, a unity to all suffering. Often, as in the case of the toothache I described at the beginning,

The world according to suffering


the pervasiveness results from thoughts revolving around the pain or other unpleasant experience, which appears as something to-be-ended and correlatively generates felt desire for change. Piercing pain is nothing if not an attention magnet. But the effects of unpleasant experience are not restricted to attention, but also involve change in how we practically construe the objects of our attention. First, as Fredrik Svenaeus observes, pain and other forms of unpleasant bodily experience turn our body from a transparent medium with which we explore the world into an object in that world – one is no longer ‘at home with one’s body’ (Svenaeus 2014, 413). It’s no longer just what we use to overcome obstacles, but an obstacle to be got around of. Second, these experiences change how we relate to the objects and opportunities around us. Consider here Havi Carel’s phenomenological description of the effects of her chronic breathlessness on her practical perception (as I would call it): The world shrinks and becomes hostile. The sense of possibility that accompanies objects disappears. A bicycle is not an invitation for an afternoon of fresh air and freedom. It is a relic of days bygone. Hiking boots now sit leaden in a cupboard. They are no longer ‘something to be worn when going for a hike’; they have long been too heavy and hiking too hard. The inviting smell of mud and hills has faded from their soles […] (Carel 2016, 111) Here, to be sure, it’s not clear whether it is the effects of bodily damage (the respiratory disease) on possible actions or the unpleasantness of anticipated sensory experiences that negatively transforms the sufferer’s world. No doubt both often play a role. When sensory suffering results from long-lasting disease, torture, imprisonment in harsh conditions, or some other factor that the subject cannot control, it can also result in negative affective construal of one’s whole future and call one’s identity into question – which is to say, it will be a partial cause of deep depression, anxiety, or disorientation.15 So I’m arguing that sensory suffering is really just a special case of attitudinal suffering – attitudinal suffering caused by painfulness or unpleasantness of sensory input. In this sense, sensory suffering is pain that has a meaning for us – an identical, equally painful pain sensation that didn’t for some reason change how we affectively construe the world wouldn’t constitute suffering. (It may be worth distinguishing a broader category of bodily suffering which encompasses sensory suffering, but also attitudinal suffering caused by bodily disease or damage directly.) It worth noting, finally, that in the experiences of suffering I’ve discussed in this section, there are two sources of unpleasantness – first the sensory displeasure, which is painful in itself, and then the pervasive attitudinal displeasure brought about by changes in relation to our body and the world caused by the sensory displeasure (or bodily damage). In that sense, suffering redoubles the


Antti Kauppinen

unpleasantness of the pain. Perhaps this amplification of displeasure explains why bodily suffering is so frequently thought of as the paradigm of suffering.

Evaluating the proposal The account I’ve been developing can be summed up as follows: Suffering as negative affective construal (SNAC) To suffer is to affectively construe one’s situation as negative in a pervasive enough fashion, and thus to experience felt aversion or attitudinal displeasure towards it. This seems to meet the desiderata for an account of suffering that I discussed in the first section. Insofar as we accept the claim that an experience of feeling aversion to what seems to be the case beyond one’s control is unpleasant, SNAC guarantees that suffering is unpleasant. (Additionally, the source of suffering may itself be unpleasant in the case of pain.) One worry about the role of felt aversion might be that motivational anhedonia, or the lack of any desire, is a form of suffering in spite of the absence of felt aversion to one’s situation (Tully 2017). (In my terms, this could be pervasive negative practical rather than affective construal.) My response to this concern is to deny the existence of such cases. To be sure, major depression can at least in principle extinguish all desires to act. But it is highly implausible that the depressive who suffers doesn’t think something along the lines of ‘Please God, not this!’. The absence of any such desire would amount to resignation or acceptance of the pointlessness of one’s existence, which is no longer a state of suffering. (In effect, I’m defending the Buddhist idea that desire is at the root of suffering.) Alternatively, we might also think that a total absence of positive affective construal suffices for suffering of a sort. This might be plausible in cases of extreme boredom – though there, too, it’s natural to assume a frustrated wish for having something interesting to do. Second, pervasive negative affective construal is plausibly intrinsically bad for you. Its unpleasantness may suffice in itself. But we can in fact say something more. When your overall affective construal of your situation is negative, you are also guaranteed to be unhappy, according to most persuasive views of happiness. I lack the space here to defend it, but I take it that Dan Haybron’s (2008) emotional condition view of happiness is by far the best existing contender for meeting the desiderata for such theories.16 Pervasive negative affective construal is evidently inconsistent with Haybron’s ‘psychic affirmation’ of one’s life. Few would deny that unhappiness is intrinsically bad for the unhappy, even if it’s not the only intrinsically bad thing. Third, given plausible views of the link between value and reasons, this suffices to explain anyone who has a reason to care about us has a reason to relieve our suffering, if they can. And fourth, given the cognitive undemandingness of affective construal, any creature that is capable of practical perception and experiencing phenomenal desires can suffer, according to SNAC.

The world according to suffering


Finally, the account allows for the intelligibility of desiring oneself to suffer. This requires some spelling out, since Brady argues that views that involve aversion to one’s situation, like Hick’s view (or mine), can’t explain this: ‘Hick’s account seems to rule out the possibility of someone’s being glad that they are suffering, or the possibility of someone’s willing acceptance of their suffering’ (Brady 2018, 27). A criminal, he notes, might accept her suffering as just in the light of her crimes, and an Opus Dei member might scourge his flesh as a mark of devotion. While they suffer, ‘both, plausibly, do not wish that their situation – a situation in which they willingly accept pain as a mark of devotion to Christ, or as merited punishment for their crimes – were otherwise’ (ibid.). To see why this doesn’t work, let’s be clear on what is and isn’t part of their situation. Suppose the repentant criminal is forced to do taxing physical labour. She suffers when her aversion to this pervades her affective construal of her overall situation. It follows that she wants her situation (of being forced to do taxing labour) to be otherwise. Nevertheless, she can accept as just that she is in a situation she wants to change, in which case she doesn’t want to change this higher-order fact. If she had read this chapter (most criminals haven’t), she might say ‘I deserve to affectively construe the world negatively’. She could even be glad she affectively construes her overall situation negatively – a little bit of gladness is consistent with suffering. Of course, such positive attitudes are in tension with suffering, which is why the whole thought of being glad that you’re suffering is borderline paradoxical to begin with. Indeed, it’s a plus for a theory of suffering like SNAC that it makes the tension clear.

Conclusion My guiding thread in this chapter has been the thought that suffering transforms the sufferer’s world, whether by dulling its colours or sharpening its edges. This transformation is registered by motivational states that have phenomenal content, so it involves a felt wish for things to be otherwise, and is unpleasant to experience. In emphasizing the world-directed intentionality of suffering, the view differs steeply from what may well be the currently leading account in analytic philosophy, Michael Brady’s higher-order desire theory. It also sharply distinguishes between suffering and bodily pain – though in some sad cases, we suffer because of pain and its effects. But though all suffering is fundamentally attitudinal, it is often also manifest in the body, given the embodied nature of affect. When our intentional horizon clears, or the light steals in through a crack, the effect really is much like a weight falling off our shoulders.

Notes 1 If you’re ever have tooth trouble while in Cork City, the number of Canty Dental is 021 4344111.


Antti Kauppinen

2 Brady could hold, to be sure, that the mother has both a desire for her unpleasant experience to end and a higher-order desire for it to continue. While this is a possible move, the insistence on the presence of the reflexive first-order desire in such cases would seem to be theoretically motivated, and at odds with the phenomenology. 3 According to evaluativists about pain, such as Bain (2013), pain represents its cause as bad. But even if we granted this, it’s a further and less plausible claim that this evaluation is part of my perspective, since it is not a judgment-sensitive attitude (see Part IV). 4 Hilla Jacobson (2018) and Brady (2015, 2018) have made related claims about the unpleasantness of pain. Jacobson (2018, 22), for example, says that ‘the phenomenal character of painfulness is constituted by a subjectively frustrated desire that the bodily condition, which is represented as obtaining, does not obtain’. 5 I borrow the term ‘construal’ from Roberts (2003), but I don’t mean to endorse all aspects of his characterization of what construals amount to. 6 On this view, we can have unconscious emotions only in the sense that we don’t realize that the way in which we conceive of the object is constitutive of an emotion – for example, I may conceive of Peter as calling for aid due to his weakness and be motivated to help him without realizing that I pity him. 7 On Hufendiek’s enactivist account, there isn’t a gap between the embodied representation and action readiness, but the one and the same bodily response both represents and prepares us to act. My worry about this sort of view is that representation seems to take place on a sub-personal level, leaving it unclear how the formal objects of emotion are represented in my conscious experience and not just by the bodily responses. (For this kind of worry, see Deonna and Teroni 2012, 73–74.) 8 For an astute account of envy, see Protasi (2016). 9 What about aesthetic emotions like aesthetic admiration, which are often presented as counterexamples to linking emotion with motivation? I don’t think these are counterexamples. If I admire a painting, it appears to me as to-be-examinedcarefully and to-be-recommended-to-others, among other things, and insofar as admiration is occurrent, I also have a felt desire or wish to do these things (see also Kauppinen forthcoming). 10 I’ve argued that as proto-evaluative construals, emotions can constitute moral intuitions (Kauppinen 2013a). 11 Some deny that this is true for all emotions. Prinz (2004, 168), for example, claims that ‘Some emotion episodes seem to be wholly independent of our goals and plans’. But his examples, like fear felt when someone suddenly throws a stone through the window, are not counterexamples to concern-based accounts – concerns are not concrete, consciously adopted aims, but caring attitudes that underlie such selections. 12 Perception of affordances also covers Lazarus’s ‘secondary appraisal’ involving assessment of control and coping potential (Lazarus 1991, 133). 13 Unfortunately, I came across Ratcliffe’s congenial work only in the final stages of preparing this chapter, so I cannot properly discuss it here. Where I’m inspired by Sartre, he draws on Husserl and Heidegger. 14 On this view, there’s no such thing as ‘social pain’, though there are of course unpleasant experiences resulting from social relationships, as Jennifer Corns (2015) persuasively argues. 15 It is far too strong to claim, as Eric Cassell (1991) does, that suffering by definition involves the ‘destruction’ or ‘disintegration’ of the person, or threat thereof. But there’s no reason to deny that it is a possible consequence. 16 However, for some reservations, see Kauppinen (2013b).

The world according to suffering


Acknowledgements I’m grateful to Michael Brady and Lilian O’Brien for generous comments on an earlier draft of the paper, as well as audiences in Turku and Jyväskylä, especially Miira Tuominen.

References Aydede, M. and Fulkerson, M. (2018). Reasons and theories of sensory affect. In D. Bain, M. Brady and J. Corns (eds), Philosophy of Pain (pp. 37–69). Abingdon: Routledge. Bain, D. (2013). What makes pains unpleasant? Philosophical Studies, 166(1), 69–89. Brady, M. S. (2015). Feeling bad and seeing bad. Dialectica, 69(3), 403–416. Brady, M. S. (2018). Suffering and Virtue. Oxford: Oxford University Press. Bramble, B. (2013). The distinctive feeling theory of pleasure. Philosophical Studies, 162(2), 201–217. Carel, H. (2016). Phenomenology of Illness. Oxford: Oxford University Press. Cassell, E. (1991). The Nature of Suffering and the Goals of Medicine. Oxford: Oxford University Press. Corns, J. (2015). The social pain posit. Australasian Journal of Philosophy, 93(3), 561–582. Cutter, B. and Tye, M. (2011). Tracking representationalism and the painfulness of pain. Philosophical Issues, 21(1), 90–109. Deonna, J. and Teroni, F. (2012). The Emotions: A Philosophical Introduction. Abingdon: Routledge. Feldman, F. (2004). Pleasure and the Good Life: Concerning the Nature, Varieties, and Plausibility of Hedonism. Oxford: Oxford University Press. Gibson, J. J. (1979). The Ecological Approach to Visual Perception. New York: Houghton Mifflin. Gibson, E. J. and Walk, R. D. (1960). The ‘visual cliff’. Scientific American, 202, 67–71. Grahek, N. (2007). Feeling Pain and Being in Pain (2nd edn). Cambridge, MA: MIT Press. Haybron, D. M. (2008). The Pursuit of Unhappiness: The Elusive Psychology of Well-being. Oxford: Oxford University Press. Heathwood, C. (2017). Which desires are relevant to well-being?. Noûs, online ahead of print. Hick, J. (1966). Evil and the God of Love. London: Macmillan. Hufendiek, Rebekka (2015). Embodied Emotions: A Naturalist Approach to a Normative Phenomenon. Abingdon: Routledge. Ingle, D. and Cook, J. (1977). The effects of viewing distance upon size preference of frogs for prey. Vision Research, 17, 1009–1019. Jacobson, H. (2018). Not only a messenger: towards an attitudinal-representational theory of pain. Philosophy and Phenomenological Research, online ahead of print. Kauppinen, A. (2013a). A Humean theory of moral intuition. Canadian Journal of Philosophy 43(3), 360–381. Kauppinen, A. (2013b). Meaning and happiness. Philosophical Topics, 41(1), 161–185.


Antti Kauppinen

Kauppinen, A. (2019). Ideals and Idols: On the Nature and Appropriateness of Agential Admiration. In A. Archer and A. Grahlé (eds), The Moral Psychology of Admiration (pp. 29–44). Lanham, MD: Rowman and Littlefield. Koffka, K. (1936). Principles of Gestalt psychology. Philosophical Review, 45(4), 412–415. Lazarus, R. S. (1991). Emotion and Adaptation. Oxford: Oxford University Press. Lewin, K. (1935). A Dynamic Theory of Personality: Selected Papers (trans. D. K. Adams & K. E. Zener). New York: McGraw-Hill. Lin, E. (2018). Attitudinal and phenomenological theories of pleasure. Philosophy and Phenomenological Research. Nagel, T. (1970). The Possibility of Altruism. Oxford: Clarendon Press. Oddie, G. (2005). Value, Reality, and Desire. Oxford: Oxford University Press. Parfit, D. (2011). On What Matters (vol. 1). Oxford: Oxford University Press. Prinz, J. J. (2004). Gut Reactions: A Perceptual Theory of Emotion. Oxford: Oxford University Press. Protasi, S. (2016). Varieties of envy. Philosophical Psychology, 29(4), 535–549. Ratcliffe, M. (2014). Experiences of Depression: A Study in Phenomenology. Oxford: Oxford University Press. Roberts, R. C. (2003). Emotion: An Essay in Aid of Moral Psychology. Cambridge: Cambridge University Press. Sartre, J.-P. (1936/2004). The Transcendence of the Ego: A Sketch for a Phenomenological Description. Abingdon: Routledge. Sartre, J.-P. (1939/1948). The Emotions: Outline of a Theory. New York: The Philosophical Library. Siegel, S. (2014). Affordances and the contents of perception. In B. Brogaard (ed.), Does Perception Have Content? (pp. 39–76). Oxford: Oxford University Press. Smith, M. (1987). The Humean theory of motivation. Mind, 96(381), 36–61. Solomon, A. (2001). The Noonday Demon: An Atlas of Depression. New York: Simon and Schuster. Svenaeus, F. (2014). The phenomenology of suffering in medicine and bioethics. Theoretical Medicine and Bioethics, 35(6), 407–420. Tully, I. (2017). Depression and the problem of absent desires. Journal of Ethics and Social Philosophy, 11(2).


The disruption model of suffering Tom McClelland

Introduction Suffering takes many forms. We might suffer from an acute feeling of pain, or from a deep-seated feeling of guilt or from a persistent feeling of grief. But there are a number of reasons for thinking that suffering involves more than having such feelings. One reason is that the magnitude of one’s suffering needn’t match the severity of the suffered state. As Klein observes, ‘pain and suffering can vary independently in their intensity’ (Klein 2015a, p. 51). Consider a severe headache that persists at the same level over three days. On the first day it causes you great distress: you feel anxious about the pain and scared that it won’t abate, you’re unable to go about your normal life, you attend to nothing but the pain and ruminate endlessly on its possible source and the prospect of relief. On the second day you are more stoical: you’re still bothered by the pain but your emotions are more in check, you’re better able to function, with effort you can concentrate on other things and your thoughts aren’t always concerned with your pain. On the third day you have a strong dose of morphine: the pain is just as severe as before but now you aren’t bothered by it at all.1 Here it is natural to say that although the intensity of your pain has remained constant over the three days, the extent of your suffering has diminished. But if the pain itself hasn’t got any better, what exactly has improved?2 I suggest that what’s improved is how much the pain disrupts your wider mental life. Pain is an individual mental state that has an impact on your wider mental economy (i.e. the total web of mental states that constitute your mental life). On the first day you experience negative emotions and the pain compromises your ability to pursue your normal priorities, to control the focus of your attention and to govern the contents of your thoughts. Here the pain is at its most disruptive. On the second day the pain continues to disrupt your mental life but to a lesser degree. And on the third day you enter a chemically induced state of mind in which the pain no longer interferes with your mental life (although the drugs doubtless do). Perhaps, then, suffering just is the unpleasant disruption to your mental life that the suffered state causes. To understand how bad it is for you to be in a given mental state we cannot just


Tom McClelland

look at the intrinsic unpleasantness or ‘badness’ of that mental state (if such there is). Instead we must consider how that mental state impacts your wider mental life. I argue that this disruption model of suffering can address a number of important puzzles and help us to make better sense of what the experience of suffering is like. The disruption model of suffering has interesting implications for the role of agency in suffering. A disruptive mental state, such as a pain, impacts upon many different facets of our mental life. Some of the results of the unpleasant state are involuntary: perhaps the anxiety you feel with the onset of a severe pain is outside your control. But some of the other results are agential: ruminating on the possible source of your pain is something you do not something that happens to you. So if we identify suffering with mental disruption, then we must identify suffering with something that is at least partly agential. I suggest that this surprising conclusion should be embraced: suffering is, in part, the result of an active struggle with that which is suffered, and accounts on which suffering is passive fail to capture this struggle. Of course, the suggestion that our suffering is partly under our control does raise some questions. If suffering is partly constituted by things that we mentally do, why don’t we just not do them? If part of suffering from a pain is continually thinking about that pain then why don’t we alleviate our suffering by simply not thinking about it? After all, a lot of the time we know that those mental actions aren’t doing us any good anyway! I propose that we perform such mental acts because suffered states, such as pains, trigger urges to act in certain ways. We feel pressure to perform certain mental actions. Acting on that urge disrupts our mental life, but resisting that urge is itself effortful and disruptive. So although some aspects of suffering are agential there’s no simple way of voluntarily lowering one’s suffering. These mental urges are associated with any number of suffered states including feelings of guilt, shame, loss and disappointment. Our guilt, for example, is associated with a compelling urge to ruminate upon that guilt, and this urge will disrupt our mental life whether we elect to submit to it or resist. Suffering is an active struggle, and these urges underwrite that struggle. In the foregoing sketch of my proposal I have made three main claims: i That to make sense of the puzzles surrounding suffering we should adopt a disruption model on which suffering is constituted by the unpleasant impact of a suffered state on our wider mental lives. ii That to make sense of the disruption constitutive of suffering we should acknowledge that suffering is, to some extent, under our agential control. iii That to make sense of the agential status of suffering we should recognise the presence of disruptive mental urges triggered by suffered mental states. In the first section I make a case for the first claim. In the second section I make a case for the other two claims. In the third section, I extrapolate some

The disruption model of suffering


lessons about the value of suffering, how best to navigate the experience of suffering, and the influence of cognitive sophistication on suffering.

The disruption model of suffering Suffering as mental disruption The proposal on the table is that suffering is the disruption to one’s mental life caused by a suffered mental state. To unpack this view it might be helpful to step away from the example of pain and to consider the experience of grief. The experience of grief can be a particularly awful example of suffering, and the dramatic way in which grief can disrupt your mental life makes it a vivid illustration of the disruption model. After the death of a loved one you have a new mental state that needs to be mentally digested – a mental state representing the deceased as gone. The ensuing mental processes will encompass all facets of your mental life. On the level of emotion, you might experience feelings of longing for the absent person, sadness that you won’t see them again, fear for your future without them, anger at the injustice of their absence and guilt about how you treated them when they were alive. On the doxastic level, any belief premised on the presence of the deceased will have to be revised. This could include deepseated beliefs about your future with the deceased, and perhaps even about your identity as a person. There will be any number of questions raised by the death that you have to think through: How could this have happened?, What did I do to deserve this?, What am I going to do with my life now?. These questions may be associated with meta-cognitive feelings of incomprehension. On the level of desires, any desire premised on the presence of the deceased will be frustrated and must slowly be updated. You will likely also experience an overwhelming desire for the deceased to come back – a desire that can never be satisfied. And any number of desires unrelated to the deceased could also be impacted: desires you have for certain kinds of food, drink or leisure activity might disappear after your loss, rendering you incapable of enjoying the things you normally enjoy. As these mental processes unfold your attention will be dragged toward your bereavement and you may find it hard to concentrate on anything else. These different mental processes interact in complex ways. Your emotional experience may unfold in phases. On the classic Kübler-Ross model of grief, for example, the bereaved go through a phase of anger before entering a period of depression (Kübler-Ross 1969). And new realisations can trigger new emotions. The realisation that a loved one won’t be present for some future life event might lead to an experience of dread that itself has to be dealt with. Here we find a kind of domino effect, where confronting one mental state triggers a new bad mental state that is itself disruptive. The process of grief encompasses this whole network in all its rich causal-temporal structure.


Tom McClelland

Although grief is an extreme example of suffering, it is representative of how a suffered mental state can impact our experience. States of pain, shame, guilt, doubt etc. all disrupt our mental equilibrium, though they each do so in very different ways. It is this disruption that I claim is constitutive of suffering. The suffered mental state is like a rock dropped in a pond and suffering is the disturbance of the water. This is the driving thought behind the disruption model, but some refinements are needed. I suggest that with three refinements we can capture a plausible account of suffering. The first refinement is motivated by the following complication: not all cases of one’s mental life being disrupted are cases of suffering. Feelings of love, for example, can radically disrupt one’s mental life but here the impact on your wider mental economy is welcome.3 So if suffering is constituted by mental disruption, it had better be a particular kind of disruption. A natural proposal is that the kind of mental disruption constitutive of suffering is unpleasant mental disruption. Whereas love generally disrupts our mental life in a way that is pleasant for us, pain generally disrupts it in a way that is unpleasant for us. At the risk of stretching the rock-in-the-pond analogy, love tends to send out good mental ripples but pain tends to send out bad mental ripples. On this account, experiencing unpleasantness is a necessary condition of suffering. This is a commitment with which I am comfortable. To my mind, it would be unintelligible to say ‘I was suffering intensely but there was nothing unpleasant about it’. However, I am not simply equating suffering and unpleasantness here. On the disruption model, experiencing an unpleasant mental state does not suffice for suffering. To say that the disruption to one’s mental life is unpleasant is to say something about your overall experience at a given time. When you’re suffering from a headache, an unpleasant pain might be part of an overall experience that it is unpleasant for you to be in. But if you score a winning goal with a flying header, an unpleasant head pain might be part of an overall experience that is far from unpleasant. The winning goal case lacks what we might call holistic unpleasantness. Accordingly, the disruption model would describe only the former example as a case of suffering.4 The idea of a pain causing unpleasant experiences distinct from the pain itself can be found elsewhere in the literature. Price, for instance, introduces the concept of ‘secondary pain affect’ and Fields has a concept of ‘secondary unpleasantness’ (see Aydede and Guzeldere 2002). In both cases, a type of unpleasantness is posited that can vary independently of the intensity of the pain and that is in some way sensitive to one’s ‘higher functions’ such as one’s beliefs about the pain. However, neither thinker is targeting the unpleasantness of one’s overall conscious state. They are positing an extra unpleasant state that may or may not accompany a pain, but such individual unpleasant states can still be distinguished from the overall unpleasantness of one’s experience. On my account, if we want to know how much someone is suffering from a pain we need to ask how unpleasant it is to be them right now. By looking at

The disruption model of suffering


the unpleasantness of a pain state and a ‘secondary’ state we risk missing the forest for the trees. Another welcome consequence of this proposal is that a suffered mental state needn’t be intrinsically unpleasant (i.e. the suffered mental state needn’t be such that it is necessarily unpleasant for the subject to be in that state). Although suffering is essentially unpleasant, one’s overall conscious experience can be unpleasant without the object of one’s suffering being unpleasant. When we experience grief, for example, the state that disrupts our mental economy is plausibly the belief that a loved one is a dead. This belief is neither intrinsically pleasant nor unpleasant to hold. What’s unpleasant is the impact that this belief has on your mental life. On some views, even pain states are not intrinsically unpleasant. If this were the case, the disruption model would accommodate it in much the same way: being in pain would be unpleasant for a subject in virtue of its unpleasant impact on the subject’s overall mental life rather than in virtue of some intrinsic feature of the pain itself. The second refinement concerns the relationship between suffering and consciousness. Our mind is a network of conscious and unconscious states. If suffering is disruption to this network, perhaps suffering could be purely unconscious. Your unconscious mental life could be negatively disrupted by a pain while your conscious life is unaffected. In such a case, would the subject be suffering? I would argue not. As discussed, suffering is unpleasant mental disruption and this unpleasant disruption must be conscious. I would be tempted to argue that unpleasantness is essentially conscious, and that unpleasant mental disruption is therefore essentially unconscious. However, in this volume Sam Coleman has made a serious case for the existence of unconscious unpleasantness. Nevertheless, I would maintain that unconscious unpleasantness could never constitute suffering. To say that you are suffering is to say something about what it’s like to be you at the moment, and unconscious states do not constitute what it’s like to be you. So the mental disruption constitutive of suffering must be conscious unpleasant disruption. One final refinement is in order. Not all cases of holistically unpleasant mental disruption qualify as suffering. Imagine you have a very mild headache. This mild headache might disrupt your mental life a little, and this disruption might be mildly unpleasant. But you would not ordinarily say that you are suffering because of your mild headache. Instead, you might say that you’re a little bothered by it.5 We can make sense of this by saying there’s a continuum of unpleasant mental disruption running from the mildly unpleasant to the extremely unpleasant. At the bottom end of the continuum, we have cases where we’re just bothered by a mental state. At the top end of the continuum, we have cases of suffering. And in the middle of the continuum we have fuzzy cases such that it’s hard to say whether we’re mildly suffering or whether we’re just bothered in a way that falls short of suffering. With the foregoing refinements we can now finalise the disruption model as follows:


Tom McClelland Disruption model Suffering is disruption of one’s conscious mental life by a suffered mental state such that (a) the disruption is holistically unpleasant and (b) its unpleasantness meets a suitable threshold of severity.

The benefits of the disruption model Why adopt the disruption model? Although the model is meant to be prima facie plausible, I suggest that it gains a lot of support from its ability to offer solutions to a range of puzzles that surround the concept of suffering. For example, we can make sense of an instant of pain caused by a pinprick, but cannot make sense of an instant of suffering (Edwards 2003, p. 65). Why might this be? Well, the disruption model has a simple answer. The disruption constitutive of suffering takes time to unfold. Where the pain is so brief that it has no wider impact, we don’t suffer. And if we do suffer, this is something that unfolds after the cessation of the pain and lasts more than an instant. Another question: why does it make sense to say you aren’t bothered by your pain but not make sense to say you aren’t bothered by your suffering? Again, the disruption model has a simple answer: your suffering is your being seriously bothered – the significant negative impact on your mental life – so it is impossible for you to suffer without being bothered by your suffering (Edwards 2003, p. 65). Other accounts of suffering can, of course, confront such puzzles in different ways. But a virtue of the disruption model is that it can address so many puzzles. There are four other puzzles that the disruption model promises to solve that I will unpack a little more slowly. The puzzle of partial dependence Any account of suffering must be able to capture the subtle relationship between suffering and the state that is suffered. On the one hand, we cannot equate suffering with the possession of the suffered state. Having a pain state is one thing but suffering from it is another: the degree of one’s suffering can vary independently of the pain and one might even be in a pain state without suffering at all. On the other hand, we cannot describe suffering as a discrete state that may or may not accompany pains and the like. For if it were a discrete state we should be able to suffer without having any state that is suffered, but this is impossible. You can’t just suffer – there’s no suffering quale that can characterise our phenomenology independently of our other mental states. So whatever suffering is, it is something that requires the presence of the suffered state. When we suffer from a pain, our suffering thus has a puzzling partial dependence on the pain: it cannot be reduced to some aspect of the pain but nor does it exist independently of the pain. The disruption model addresses this puzzle straightforwardly. Suffering is the unpleasant mental disruption resulting from the pain. The impact of a

The disruption model of suffering


pain is not reducible to the pain itself any more than the ripples in a pond are reducible to the stone that causes them. But the suffering is still dependent on the pain: there is no disruption without some state that does the disrupting. The disruption model thus predicts that suffering is partially dependent on the state that is suffered.6 The puzzle of partial phenomenal commonality The phenomenology of different cases of suffering has something in common. What it’s like to suffer from a pain, what it’s like to suffer from remorse and what it’s like to suffer from fear differ significantly. The phenomenal qualities distinctive to pain, remorse and fear are totally different. And yet there’s something that each of the experiences share. If you’ve never felt remorse you can’t know exactly what it’s like for someone who is feeling remorse, but if you’ve ever suffered from any other state then you know something about what it’s like for them to be suffering from their remorse. There’s a certain commonality between experiences of suffering. And yet there’s no such thing as an isolable suffering quale: there’s not one quality that all and only suffered states have – a quality that can accompany pain, remorse, disorientation or any other experience. Suffering thus has a puzzling partial phenomenal unity: there’s no one suffering quale and yet different experiences of suffering resemble one another phenomenally. The disruption model can help us make sense of this puzzle. Mental states that are different in their intrinsic phenomenology may disrupt our mental lives in similar ways. On the disruption model, the experience of suffering just is the experience of this unpleasant mental impact, so qua impact the phenomenology of suffering a pain can resemble the phenomenology of suffering remorse. I am not suggesting here that the impact is just the same in these different cases. A pain will not have exactly the same effect on our emotions, beliefs, desires and attention as remorse does. In fact, no two cases of pain will ever have exactly the same effect. The commonality between different cases of suffering is more abstract than this. There’s something it’s like to have one’s mental life unpleasantly disrupted by a mental state. Exactly what it’s like will differ from case to case, depending on both the suffered state and the way it disrupts our mental life, but we can see the commonality between these different cases. This doesn’t mean there’s an unpleasant disruption quale accompanying all cases of suffering. Instead we should think of the phenomenology of mental disruption on analogy with something like the phenomenology of perceiving: there is something it’s like to perceive and all experiences of perceiving share this broad phenomenology even though there is no ‘perceiving quale’ shared by each experience. Instead different perceptual experiences are all different realisations of the general phenomenology of perceiving. Similarly, different experiences of suffering are different realisations of the general phenomenology of unpleasant mental disruption.


Tom McClelland

The puzzle of partial cognitive penetrability Here once more suffering has a puzzling status that falls somewhere between two poles. First, it is tempting to say that suffering is cognitively impenetrable. Klein (2015a, p. 184) says, ‘Whatever suffering is, it appears to be fairly strongly encapsulated. It’s hard to talk yourself out of suffering from pain.’ Similarly, Cohen and Fulkerson (2017, p.235) say, ‘one of the things that make suffering so awful is its peculiar one-sided relation to rationality: suffering provides reasons, but … it is oddly resistant to influence by our other reasons’. And yet there are long-standing traditions that encourage us to alleviate our suffering precisely by changing our beliefs. In Stoicism the right beliefs about the real value of things can release you from suffering. Similar claims are integral to the Buddhist tradition. There the emphasis is more on changing your desires than your beliefs, but the required alteration to desire is again driven by coming to see things as they truly are, which is best characterised doxastically. In a contemporary context Corns has suggested that ‘it may be that suffering is more susceptible to reason-giving than we realized. It may be that, on more occasions than we think, the most efficient way to ease our pain is not to take a pill, but to have a think’ (Corns 2015). How can suffering encourage such divergent claims? Again, it appears that suffering has a subtle intermediate status, this time between the cognitively penetrable and the cognitively impenetrable. On the one hand, we can’t normally use what we believe about a situation to reason ourselves out of suffering, and yet the extent of our suffering can depend upon our background beliefs and on changes to those beliefs. The disruption model can help us to make sense of this situation. If suffering is the unpleasant disruption of our mental economy by a suffered state, then the extent of our suffering will clearly be influenced by what’s already present in that mental economy, including our beliefs. If we change our beliefs, the suffered state might be less disruptive to our mental life, thus alleviating our suffering. This means there is an extent to which suffering is cognitively penetrable. This complements the Stoic Epictetus’s suggestion that ‘It is not what happens but how you react to it that matters’ (Epictetus 1995). The occurrence of a pain may be out of our control, but suffering is a matter of how we react to that pain, and with the right beliefs and attitudes our reaction to the pain needn’t be so bad. That said, beliefs aren’t the only things in our mental economy. Consider the distress caused by the pain: this distress can be an important aspect of your suffering, yet it is not immediately responsive to reasons. Suffering is thus partially cognitively penetrable because the disruption constitutive of suffering is partially cognitive. Some aspects of suffering are thus responsive to reasons while others are not. The puzzle of absent suffering There is a range of puzzling cases in which subjects have mental states of a kind that would normally be associated with suffering and yet they do not

The disruption model of suffering


appear to suffer. Pain asymbolia appears to provide examples of pain without suffering. Klein introduces the condition as follows: ‘Pain asymbolia is a rare condition caused by lesions to the posterior insula. Asymbolics say that they feel pain, but they are strikingly indifferent to it’ (Klein 2015b, p. 493). How can subjects undergo severe pains without suffering? The disruption model offers a framework that can answer this question. Suffering is the unpleasant disruption of a subject’s mental life so, in principle, one might have a pain without it having any negative impact on one’s mental life. You can be in pain without necessarily worrying about it, having one’s attention dragged toward it, or ruminating on its origin and its prospects of relief. In normal circumstances a severe pain would cause these things, but in abnormal circumstances it might not. Pain asymbolia is just such an abnormal circumstance: the condition makes one indifferent to the pain in such a way that it doesn’t disrupt one’s mental life. The details of why this is so would be needed to complete the story – Klein (2015b) has an interesting account on which subjects no longer care about the physical integrity of their body – but a broad framework in which pains don’t have the mental impact constitutive of suffering certainly looks like the right way to go. I have mentioned already that the disruption model can dodge the question of whether some mental states are intrinsically unpleasant. This is particularly important in the context of asymbolic pain. It is very difficult to judge whether the asymbolic’s pain is still unpleasant or not, but on the disruption model this isn’t really the important question. The important question is whether the pain asymbolic’s overall experience is holistically unpleasant, and it is fairly clear that the answer is ‘no’. Regardless of whether their pain state is intrinsically unpleasant, the absence of holistically unpleasant disruption means that the pain asymbolic is not suffering. Moving beyond pain, here’s another puzzling case: There I was, walking from the library down the silent, lonely hallway to the bathroom, when I was struck. That’s the only word that really works. Struck as though someone had hit me. Mom was dead. Two years after we’d buried her next to Dad, I suddenly ‘got it’. (Bursack 2014) Here the subject seems to undergo a terrible loss without suffering. Instead, her suffering is delayed: she only starts suffering two years after her loss. The disruption model can make sense of this. The subject was not deluded about the loss of her mother – she correctly believed that she was dead. But her belief had somehow failed to have any impact on her wider mental life. The belief only starts to have a serious impact on her mental economy when she ‘gets it’. Put another way, she only started to mentally digest what her loss meant for her after the delay, so it was only then that the suffering really started. In this section I have presented the disruption model and shown how it promises to address several challenging puzzles that surround the concept of


Tom McClelland

suffering. The next step is to extrapolate some of the implications of this account and to flesh out some more of its details along the way.

Suffering, agency, and mental urges Some of what the mind does is not agential: perception, emotions and unconscious processes are outside our direct control, or at least only available for control in a very limited way. Other mental processes are agential: there are acts of attention, acts of evaluation, acts of contemplation and so on. The extent to which our mental life is agential is hotly contested (see Soteriou and O’Brien 2009), but it is fair to say that at least some mental events are indeed mental actions. The disruption model states that the process constitutive of suffering encompasses all facets of our conscious mental life. This includes not just the non-agential mental events but the agential ones, which implies that suffering is to some extent agential. Consider again the process of grief. Perhaps things like feelings of anger or depression are involuntary consequences of your loss. But many of the things that happen in your mind because of your loss are agential. Directing your attention to our loved one’s death is an action. So too are the actions of thinking through their death, including reflecting on how this injustice could have occurred, who you are now they are gone, whether you will ever be able to move on with your life, what the future will hold for you in their absence etc. These myriad mental processes are constitutive of the unpleasant disruption to your mental life caused by the suffered state, and this disruption is constitutive of your suffering. The result is that your suffering is partly constituted by your own mental actions. Although this result may seem counter-intuitive, it does fit with some of the ways in which we talk about suffering. Suffering is often described in terms of struggling – a struggle that is mental rather than bodily. We talk, for example, of struggling with pain (McCracken 1998; Toye et al. 2017), struggling with grief (Neimeyer et al. 2010), struggling with shame (Osherson and Krugman 1990), struggling with guilt (O’Connor et al. 1997) and even struggling to find meaning in life (du Plessis et al. 2013).7 In each of these cases there’s a mental state that we’re struggling to fit into our mental life, and it is natural to see the struggle as an activity that is in some way constitutive of your suffering. Interestingly, one thing we don’t talk about struggling with is our suffering. Perhaps this is because our suffering, in at least these cases, is partly constituted by our mental struggle rather than being an object of struggle. Crucially, struggling is agential. The exact form our struggle takes will be different from case to case but in each of these examples there’s an unpleasant mental state with which we have to mentally struggle in order to bring ourselves closer to a mental equilibrium. Reflection on the place of mental struggle in suffering may make us more comfortable with the conclusion that suffering is partly agential. Nevertheless, a residual worry remains. Actions are voluntary so, to the extent that suffering is constituted by mental actions, suffering is voluntary. And if suffering is in

The disruption model of suffering


part something we do voluntarily then suffering is, to that extent, something we are free not to do. We don’t have to attend to our pain and we don’t have to think about our loss. In fact, there are cases where we know that there’s no point in doing these mental actions. There’s no point, for instance, in thinking about dental pain or phantom limb pain as they don’t involve real bodily damage. We could ameliorate the disruption to our mental life by not performing these futile mental actions. To the extent that the impact of a suffered state on our mental lives is voluntary, we are free to curtail our suffering by simply not performing the relevant action. Something has certainly gone wrong here. Suffering isn’t something we can simply choose to alleviate by not performing particular mental actions, and the experience of suffering is not an experience of freely choosing to attend to your pain, say, rather than attending to something more useful. To overcome this worry I think we should hold on to the claim that suffering is partly agential but resist characterising these mental actions as wholly voluntary. To achieve this, I suggest we appeal to mental urges. Urges are mental states that direct us to act in certain ways. A terrible itch, for example, is characterised (or perhaps constituted) by an urge to scratch. Urges are a little different to desires. Desires tend to be directed toward some target state – such as getting a job – and guide our actions toward the achievement of that state e.g. by motivating us to complete the application form. Urges are more directly associated with action. Urges to scratch, to cough or to urinate do not represent some target state then leave it to you what to do to reach that state (Hall 2008). Instead, urges motivate our actions directly. Moreover, they do so in a way that is unresponsive to reasons: you can’t think yourself out of the urge to scratch. I suggest that as well as urges to perform bodily acts, we experience urges to perform mental acts. Mental urges are important to the dynamics of suffering. When we experience a pain in our ankle, for example, we might experience an urge to take our weight off our foot. I suggest we also experience a mental urge to direct our attention to the pain, to think about its source and its prospects of alleviation etc. Following these mental urges unpleasantly disrupts our conscious mental life, and such disruption is partially constitutive of our suffering. When we follow these urges our action is agential. That said, urges put us under considerable pressure to perform the relevant action. Resisting urges is difficult, so even if following an urge would be unpleasant for us we can’t straightforwardly elect not to do it. Moreover, even if we do resist an urge, our doing so is itself unpleasantly disruptive. Resisting an urge is effortful and distracting and requires our continued attention and energy. Having to make this constant effort is itself disruptive to our mental life. So the mental urges associated with your pain leave you in a ‘damned if you do, damned if you don’t’ situation: following the urge unpleasantly disrupts your mental life, but resisting the urge unpleasantly disrupts your mental life too, albeit in a different way. Of course, pains aren’t the only state from which we can suffer. I propose that suffered states are often associated with mental urges that underwrite the


Tom McClelland

suffering they cause. The specific mental action we feel the urge to perform will vary from case to case: acts of attention, evaluation, explanation, contemplation, imagination, recollection etc. Where the relevant act is one that would be unpleasantly disruptive to our mental economy, and where the force of the urge is such that resisting it would itself be unpleasantly disruptive, such urges can contribute significantly to our suffering. So even though performing the relevant mental action is agential, it’s not the case that we can ameliorate our suffering by simply not performing that action. Some of what I have said resonates with ‘imperativist’ theories of pain. Klein, for instance, offers a theory of pain on which pain states are states with imperatival content commanding us to protect our body (Klein 2015a). On this view, an ankle pain is not just associated with an urge to relieve the pressure on one’s foot. Rather, it’s in some sense constituted by a state with the content ‘relieve the pressure on your foot!’. If we accept this theory, we might want to go further and posit imperatival content that commands us to perform certain mental actions. Indeed, Watzl (2017) offers an excellent imperatival theory of attention on which salient objects demand you to redistribute your attention in a certain way, and I’ve toyed with the idea of imperatives to perform mental action elsewhere (McClelland 2017). However, I have two reasons for not going down this route here. First, it’s not clear that we’re justified in positing states with imperatival content. The familiar category of urges seems to be able to do the job without having to posit more sophisticated states that command us to act in certain ways. Second, even if we do have imperatival states, it’s not clear that they could constitute states like pain. It’s one thing to say that pains command us to act in certain ways, but quite another to say that pain is nothing more than being in a state with such content (Bain 2011). Overall then, I think we can embrace the disruption model’s implication that suffering is constituted in part by agential mental processes. Because suffered mental states are typically associated with urges to respond mentally to them in certain ways, we can understand those mental acts as things we are pressured to do rather than things we can elect to do or not do on a whim. Positing such mental urges helps capture how our mental lives get unpleasantly disrupted and to make sense of the mental struggle that characterises so many experiences of suffering.

Three lessons Our understanding of suffering has a range of important ramifications. The disruption model of suffering sheds light on some of the important questions raised by suffering. The function of suffering What is the purpose of suffering? If the disruption model is true, we can rephrase the question as follows: what is the purpose of having one’s mental

The disruption model of suffering


life unpleasantly disrupted by the suffered mental state? I suggest that the purpose of suffering is the assimilation of a difficult mental state. We suffer in order to mentally digest something that has happened to us – to learn from it, adjust to it and grow from it. Unpleasant disruption is how we make the transition from having a mental state that is at odds with our wider mental economy to adjusting our mind to accommodate that mental state. We can again use grief as an illuminating example. Earlier on I described the different stages of grief, each of which is unpleasantly disruptive to one’s mental life. On the Kübler-Ross model of grief the final stage is acceptance. Acceptance is a kind of mental homeostasis – a sustainable psychological dynamic. This doesn’t mean returning to the same kind of overall mental state you were in before your bereavement: the whole point of the process is to dramatically revise your overall mental state in light of your loss. Instead you find a way of being able to go on despite your loss. This acceptance is achieved because of one’s suffering. You can’t learn of a loss then leap directly to a totally new mental economy that is no longer premised on the presence of your loved one. There’s a process you have to go through – a process that is unpleasantly disruptive – in order to get yourself to this new mental situation. Your suffering only abates once this mental process has run its course. We can generalise this to capture the value of other kinds of suffering. The function of suffering is to adjust our mental economy to deal with some new mental state that is at odds with our psychological status quo. Feelings of pain have to be assimilated into our mental lives, as do frustrations of desire, pangs of guilt, or any of the other of the varied states that we suffer. We have to go through the unpleasant mental disruption constitutive of suffering in order to adjust to these states. Only then can we achieve a kind of mental equilibrium. This ties in with the place of urges in suffering. Urges function to motivate actions that return us to bodily homeostasis (Klein 2015a, pp. 13–25, says the same thing about imperatives). But if bodily urges to scratch, eat or urinate function to bring us back to bodily homeostasis, perhaps mental urges to attend, assess and reflect function to bring us back to mental homeostasis. Suffered mental states knock our mind out of balance, and suffering is the process of readjusting to find our mental balance again. This vindicates Corns’s characterisation of the value of suffering: ‘sometimes instead of doing everything we can to rid ourselves of suffering, we should let it hurt. Sometimes, perhaps, it is good for us to feel grief, guilt, anger, or pain’ (Corns 2015). Suffering is the disruptive process of adapting our mental economy to a problematic mental state, and such adaptation is crucial to our growth. Does this mean we should always welcome suffering? Not quite. For one thing, there are cases where the process of suffering stalls and mental homeostasis is never achieved. In cases of persistent grief subjects are no better adjusted to their loss after years of sorrow. In post-traumatic stress disorder a traumatic event continues to disrupt subjects’ mental life years after it has happened. The disruption model can make sense of this kind of case as a


Tom McClelland

failure to assimilate. Here the mind’s attempts to mentally digest a loss or a trauma fail. We suffer because we have to mentally digest things, but some things are just indigestible. Like a stuck record, we continually go through the same unpleasant mental processes without getting any closer to a mental equilibrium. Successful therapy for this kind of persistent suffering is therapy that helps subjects make sense of what has happened to them – to adjust their wider mental life in a way that successfully assimilates their loss or their trauma. And this is exactly what therapists dealing with grief (e.g. Neimeyer et al. 2010) and post-traumatic stress disorder (e.g. Osatuke and Stiles 2010) have found. But in the absence of such outside help our suffering isn’t achieving what it’s supposed to and we’d be better off without it. Nevertheless, we can still get a grip on how suffering generally has value despite these troubling cases. Of course, recognising this value doesn’t make the unpleasant disruption of our mental lives any less unpleasant. But at least we can see here why such unpleasantness is required. The navigation of suffering Suffering isn’t easy and, notwithstanding the value suffering has, it would be good if we could suffer less. In the foregoing I’ve emphasised the agential aspects of the process of suffering. Most actions we perform can be performed in better or worse ways. This opens up the possibility that we can be better or worse at suffering. Perhaps one way to ameliorate suffering is to become a more skilled sufferer. A number of accounts have been given of how skilled agents learn to navigate a landscape of possible actions (see Dreyfus 2002). A skilled tennis player, for example, will experience a landscape of possible actions, and must navigate this landscape skilfully to achieve her optimal performance. Applying this to suffering, the barrage of urges triggered by a suffered mental state constitutes a challenging mental landscape for us to navigate. If you lack the requisite skills, you might try to resist all these urges. Having to effortfully resist these urges is difficult, and if you resist them you will fail to assimilate those states into your wider mental life. An unskilled sufferer might instead follow every urge, allowing their mind to be thrown into a chaotic storm of activity that is terrible to endure and that is ill-suited to bringing one toward mental equilibrium. The skilled sufferer follows the right urges, in the right way, at the right time. She navigates the landscape of urges in a way that brings her toward mental equilibrium as efficiently as possible. Put another way, she is better at coping with her disruptive situation than her unskilled counterpart. This involves a certain necessary amount of suffering but, crucially, involves no more suffering than is required for the assimilation of the suffered state. The disruption model encourages us to develop an account of the differences between skilled and unskilled suffering, and to devise ways of teaching ourselves and others to be better at suffering.

The disruption model of suffering


Suffering in other minds The disruption model identifies suffering with the impact of an unpleasant state on one’s wider mental economy. Whether and how one suffers thus depends on what kind of mental economy one has. As discussed, a Stoic might have a particularly resilient mental economy that is relatively undisturbed by an unpleasant pain, say. Other people might have a much more fragile mental economy that is dramatically disrupted by the most minor of unpleasant feelings. This is a contrast between different kinds of mature human mind. But what if we move to different kinds of mind? The mental economy of an infant, or of a non-human animal, is totally different to that of most mature humans. How do we make sense of suffering in such minds? The disruption model might suggest that a subject with a richer mental life has a greater capacity for suffering: the bigger and more complex the mental economy, the greater the potential impact of an unpleasant state. The suffering we experience in grief is so bad because it encompasses so many different facets of our mental lives: a rich range of emotions, a vast set of beliefs and expectations, a complex identity, a host of short and long-term desires and aspirations. If our mind weren’t so complex, our suffering wouldn’t be so bad. On the other hand, the disruption model might suggest that those with a more sparse mental life are more vulnerable to suffering: the smaller and more simple the mental economy, the more easily disrupted it is by an unpleasant mental state. Where we suffer a severe pain we can generally see beyond it to the ‘bigger picture’: many of our most deeply-seated desires are entirely unimpeded by the pain so our overall assessment of our situation needn’t be so bad. A stone dropped in a lake only disturbs a little of the water. But what if one’s mind is less like a lake and more like a pond? The stone would send ripples through the whole surface. So in this sense the pain suffered by an infant or by a dog could be worse. Its impact is more comprehensive and there is no ‘bigger picture’ available to the subject that might offer consolation. What should we make of these two lines of thought? Clearly the disruption model doesn’t entail that infants and animals don’t suffer, nor does it straightforwardly entail that their suffering is any better or worse than our own. What it does entail is that if we want to understand what a subject’s suffering is like, we have to understand what their mental life is like. Perhaps the phenomenology of the sensation of pain can be exactly the same for the adult, the infant and the dog. But the phenomenology of the suffering caused by that pain will be as different in character as their minds are different in nature.

Concluding remarks In this paper I have offered a model of suffering that promises to overcome a range of puzzles regarding the partial dependence of suffering on the suffered


Tom McClelland

state, the phenomenal commonality among disparate experiences of suffering, the partial cognitive penetrability of suffering and the possibility of a subject not suffering despite their possession of a mental state that would typically cause considerable suffering. Some of the most important implications of this model concern the role of agential mental processes in the experience of suffering. Reflection on the place of agency in suffering has led me to propose that suffered mental states are typically associated with mental urges pushing us to perform a range of mental actions through which we mentally digest and adapt to the suffered state, assimilating it into our wider mental economy. Following these urges inevitably disrupts our mental life, but effortfully resisting them is similarly disruptive. This has the result that even though we have some agency over our suffering, this agency is under considerable pressure. The disruption model helps us make sense of how suffering relates to unpleasantness whilst dodging certain vexed questions regarding intrinsic unpleasantness. It also helps us to make sense of the value of suffering, the possibility of skilled suffering and the nature of suffering in other minds.

Notes 1 Here I’m assuming that the best account of morphine pain is one that accommodates patients’ reports that the severity of one’s pain is unaffected by the drug (Klein 2015b). 2 Another question that arises here is whether the unpleasantness of the pain changes over these three days. Has the pain remained as intense but become less unpleasant, or has its unpleasantness remained constant but with some further factor changing? Unpleasantness raises some issues that I address in due course. 3 A similar point is made by Bain (2019) regarding ‘intrusiveness’, noting that there are plenty of states that are intrusive but not unpleasant. 4 Here one might think of the overall unpleasantness of one’s experience in a summative way: you take the unpleasantness of all your conscious states and their sum total gives you the unpleasantness of your overall conscious state. I’m tempted to think more holistically here and to deny that we can just ‘read off’ the unpleasantness of one’s overall experience from the unpleasantness of one’s token states. In fact, following Bayne (2010) I’d suggest that phenomenal qualities belong to one’s overall conscious state at a time, and that token conscious states are abstractions from this overall phenomenology. 5 Thanks to Emily Caddick-Bourne for suggesting this. 6 In correspondence, Sam Coleman has raised the interesting possibility that one might suffer as a result of a state that one takes oneself to be in but which does not exist. If it seems to you that you are in pain when there is no such pain state, this merely intentional pain state might disrupt your mental life. And if such mental disruption constitutes suffering, we can suffer without there being any real state that is suffered. Here I think it is crucial that there is at least a notional state that is suffered. I can maintain that suffering requires there to be a state that is suffered but add the qualification that this state might be merely intentional. 7 It may seem unnatural to say that we struggle with mental states. In grief, for example, we’re struggling with the loss of a loved one, not with our belief in the loss of a loved one. Nevertheless, there’s still a clear sense in which it’s the presence of that belief that’s disrupting our mental life. After all, if the loved one died without

The disruption model of suffering


our knowing about then we wouldn’t suffer. Perhaps we can reconcile these thoughts by saying that we sometimes struggle with a suffered mental state transparently. That is, the struggle is directed at the content of the state rather than at the presence of the state itself.

References Aydede, M. and Guzeldere, G. (2002) Some foundational problems in the scientific study of pain. Philosophy of Science Supplement, 69(3), 265–283. Bain, D. (2011) The imperative view of pain. Journal of Consciousness Studies, 18 (9–10), 164–185. Bain, D. (2019) Why take painkillers? Nous, 53(2), 462–490. Bayne, T. (2010) The Unity of Consciousness. Oxford: Oxford University Press. Bursack, C. B. (2014) Delayed grief: why caregivers experience grief months or years after a death. Retrieved from upport/Delayed-Grief-Why-a-Caregiver-Will-Often-Get-Hit-With-Grief-Months-orYears-After-a-Death.htm (accessed January 2015). Cohen, J. and Fulkerson, M. (2017) Pain and rationality. In J. Corns (ed.), The Routledge Handbook of Philosophy of Pain (pp. 235–244). New York: Routledge. Corns, J. (2015) Narrating our suffering. Retrieved from (accessed February 2016). Dreyfus, H. L. (2002) Intelligence without representation – Merleau-Ponty’s critique of mental representation the relevance of phenomenology to scientific explanation. Phenomenology and the Cognitive Sciences, 1(4), 367–383. du Plessis, A. L. and Breed, G. (2013) A possible solution for corruption in South Africa with the church as initiator: a practical theological approach. Hts Theological Studies, 69(2), 1–10. Edwards, S. D. (2003) Three concepts of suffering. Medicine, Health Care and Philosophy, 6, 59–66 Epictetus (1995) The Discourses of Epictetus, trans. R. Hard. London: Everyman. Hall, R. J. (2008) If it itches, scratch! Australasian Journal of Philosophy 86(4), 525–535. Klein, C. (2015a) What the Body Commands. Cambridge MA: MIT Press. Klein, C. (2015b) What pain asymbolia really shows. Mind, 124(494), 493–516. Kübler-Ross, E. (1969) On Death and Dying. New York: Macmillan. McClelland, T. (2017) AI and affordances for mental action. Proceedings of the 10th AISB Symposium on Computing and Philosophy, 2017, 372–379. McCracken, L. M. (1998) Learning to live with the pain: acceptance of pain predicts adjustment in persons with chronic pain. Pain, 74(1), 21–27. Neimeyer, R. A., Burke, L. A., Mackay, M. M. and van Dyke Stringer, J. G. (2010) Grief therapy and the reconstruction of meaning: from principles to practice. Journal of Contemporary Psychotherapy, 40, 73–83. O’Connor, L. E., Berry, J. W., Weiss, J., Bush, M. and Sampson, H. (1997) Interpersonal guilt: the development of a new measure. Journal of Clinical Psychology, 53(1), 73–89. Osatuke, K. and Stiles, W. B. (2010) Change in postraumatic stress disorder: a disruption model account. In G. Dimaggio and P. H. Lysaker (eds), Metacognition and Severe Adult Mental Disorders: From Basic Research to Treatment (pp. 285–300). London: Routledge.


Tom McClelland

Osherson, S. and Krugman, S. (1990) Men, shame, and psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 27(3), 327–339. Soteriou, M. and O’Brien, L. (eds.) (2009) Mental Actions. Oxford: Oxford University Press. Toye, F., Seers, K., Hannink, E. and Barker, K. (2017) A mega-ethnography of eleven qualitative evidence syntheses exploring the experience of living with chronic non-malignant pain. BMC Medical Research Methodology, 17(1), 116. Watzl, S. (2017) Structuring Mind: The Nature of Attention and How it Shapes Consciousness. Oxford: Oxford University Press.


Painfulness, suffering, and consciousness Sam Coleman

Introduction Having just escaped prison, Edmond, the future Count of Monte Cristo, is shot while sailing with a crew of smugglers. But he was almost happy at this skirmish and his wound. These hard tutors had taught him how he viewed danger and bore suffering. He had laughed at danger and, as the shot pierced him, said like a Greek philosopher ‘Pain, you are not an evil.’1 Such examples show that there are two broad kinds of suffering—what we could call physical and mental suffering, respectively. Edmond suffers physically—from the pain—but not mentally. Mental suffering plausibly involves mental perturbation,2 but physical suffering can occur with or without mental suffering. There may be creatures psychologically so simple that they are incapable of mental suffering—there is little to nothing to perturb. But still it seems they could suffer physically; on account of severe pain, say. My main topic is physical suffering, especially that occasioned by physical pain. Later I will return to mental suffering. Physical pain need not be unpleasant. The pain of a yoga stretch can feel nice, and likewise those of certain other kinds of bodily release. That is not to mention the pains of plucking a hair, picking a spot, or of a lover’s pinch. In fact some pains can be extremely pleasant, hence sought after. Presumably that is partly behind the appeal of a burning-hot curry.3 One can suffer even on account of a pleasant experience, in an indirect sense, because of what it signifies to one. For example, one might feel pleasure at something inappropriate, and then experience considerable worry from having reacted that way. Such indirect suffering is clearly mental in character. When it comes to unpleasant physical pain there is another sense of suffering, whereby one suffers simply because of how the pain feels, regardless of whatever else it signifies—call this direct suffering. This is the sense in which Edmond suffers from his wound. When one suffers directly from pain, that is because how it


Sam Coleman

feels is unpleasant: you cannot suffer from a pain purely in respect of how it feels if it feels perfectly pleasant.4 Pain is not the only unpleasant sensation that can be directly suffered—consider a case of nausea. But the particular form of unpleasantness distinctive to pain is painfulness. So I will talk interchangeably of pain’s unpleasantness and painfulness in what follows. There is, therefore, a close connection between pain’s unpleasantness, or painfulness, and the direct suffering pain can yield. My first aim in this paper is to clarify the nature of, and relationships among, pain, painfulness, and direct suffering—which I will simply term suffering, henceforth. My second aim concerns the attitude of pain researchers—comprising philosophers and scientists—to consciousness, i.e. phenomenal consciousness. In one way researchers are hostages to consciousness: no criteria exist to detect pain and suffering independent of subjective reports, which answer ultimately to the subject’s consciousness of her pain state.5 Relatedly, even if the posit of unconscious pain is accepted in certain quarters, the notion that suffering could literally go on unconsciously is almost universally rejected. When it comes to pain and suffering, then, consciousness is widely taken to be the be-all and end-all. An important example illustrates this point: Pain asymbolics report being in pain, but do not report it being unpleasant. Researchers infer that asymbolics feel pain, but do not feel it as unpleasant. From this they further infer that the asymbolic instantiates a pain that lacks affect. Thence they conclude that there are two separable components to pain, ‘sensory’ pain (‘s-pain’) and affect. Finally, because the asymbolic is understood to be in pain that lacks unpleasant affect, researchers infer that she is not suffering from that pain. The complexity of this chain of inferences is not usually laid bare, and once it is there are evidently plenty of steps and accordingly considerable room for error. To take the first two steps: asymbolics might be lying about their sensations. Or they may be sincere, but unable to cognitively access, hence report, their pain’s unpleasantness. Bracketing these worries, the remaining steps crucially rely on the assumption that pain’s unpleasantness, and the kind of suffering this occasions, are exclusively conscious phenomena. Without this assumption there is no legitimate inference from the absence of conscious unpleasantness to its absence full stop, and no inference to the lack of suffering. The popular view on which unpleasant pain consists of two dissociable components, and on which there may be pains that wholly lack affect, is thus the product of a theoretical deference to consciousness. The same is true of the thesis that suffering is exclusively a conscious phenomenon. Pain researchers defer to consciousness, but in my view they do not properly heed its lesson regarding pain, painfulness, and suffering. I will argue that consciousness actually gives us a double-edged message about these phenomena. Introspection reveals pain and painfulness to be essentially kinds of qualia, or qualitative character—a thesis I defend from the ‘heterogeneity problem’. But introspection also prompts a conception of pain and painfulness on which these are capable of unconscious existence. This implies, in turn, that suffering

Painfulness, suffering, and consciousness


may well occur unconsciously, something I argue for in part by criticising rival models of suffering. Taking consciousness seriously as an epistemic source for the natures of pain, painfulness, and suffering, thus has the surprising result that consciousness is removed from the metaphysics of pain, painfulness, and suffering. My second aim is to motivate this more nuanced attitude to consciousness in the study of pain.

Painfulness is a quale There is excellent reason to hold that painfulness consists in qualitative character, such that defusing objections to this view would be tantamount to establishing it. The view has strong champions,6 and even dissenters acknowledge it as the default.7 But before I address the most prominent of those objections, it is worth making the excellent reason explicit as an argument. When pain is conscious one normally knows one is in pain, and this knowledge comes from ‘introspection’, a form of attention to one’s experiential state as such, providing non-inferential knowledge. That is why the default view is that s-pain and painfulness are qualia: for what we feel and what we attend to in introspection are kinds and configurations of qualitative character. By ‘qualia’, or ‘qualitative character’, I mean those properties that underwrite the resemblances and differences among conscious experiences, and which supply ‘what it is like’ to have such experiences. When asked what an experience is like we advert to the qualia of which we are aware. For instance, tasting black tea on its own is a type-different experience to tasting it mixed with bergamot. Feeling anger is a type-different experience to feeling desire. Arguably, consciously thinking Mick Jagger is still cool is a type-different experience to thinking that the Eiffel tower stands in Paris (see Pitt’s argument below). In each case, the qualia are different. The last two cases, of emotions and thoughts, are important: it is often taken, and this seems especially so in the debate around pain, that qualia are ‘blank’ feels, contentless in a certain sense. But in the case of emotions it seems quite obvious that: (i) emotional states have content, e.g. they represent the subject or world as thus-and-so (empty and washed out, in depression); and (ii) emotional states carry this content by means of their qualitative feel. It is with this enriched—I would say realistic—notion of qualia that I operate. It is not very controversial that s-pains are qualia. Clark accepts pains ‘typically have some sort of sensory character’, even while denying ‘that what makes these episodes painful is a quale’.8 I will make explicit reasons for holding that s-pain is a quale—what I will call ‘paininess’.9 I will then show that these reasons apply equally to painfulness, pain’s negative affective aspect. Therefore it will not be possible to hold the line as Clark intends: if s-pain involves qualia so does painfulness. Let us begin with a related case. Pitt argues that thoughts have a distinctive phenomenology such that: (i) the conscious experience of thinking is different


Sam Coleman

from any other kind of conscious state; and (ii) a given conscious thought content is experienced as different from the experience of thinking any typedistinct content. Thoughts thus have proprietary and distinctive qualia.10 Pitt’s argument can be summarised as follows: TQ TQ1. One can tell by introspection that one thinks, and what type of conscious thought one is having. TQ2. This would not be possible unless conscious thoughts had proprietary and distinctive qualitative character. TQC. Conscious thoughts have proprietary and distinctive qualitative character. I will not investigate TQ’s soundness. Premise one, particularly, seems open to objection. Nonetheless, an argument of this sort seems readily applicable to pains. Note that the success of the arguments below does not hang on TQ’s soundness. Indeed, Pitt mentions pain as a parallel case in a way that suggests he thinks the reasoning and conclusion more secure for pain than for thought; hence the dependence, if anything, runs conversely. Consider, then, an adaptation of TQ to s-pain: PQ PQ1. One can tell by introspection that one is in conscious s-pain, and what type of s-pain one is in. PQ2. This would not be possible unless conscious s-pains had proprietary and distinctive qualitative character. PQC. Conscious s-pains have proprietary and distinctive qualitative character. PQ is very plausible. Take PQ1: One can tell directly on the basis of attending to experience, by introspection, whether one feels a pain, a tickle, a caress, or an itch…etc. One need not infer that one is in pain. Even if one does arrive at knowledge that one is in conscious pain by inference (one suddenly notices one’s limping, say), if the pain is conscious one can know, or continue to know, about it by introspection. The directness of this form of knowledge does not imply that one cannot wrongly judge about pain. But it is surely the case that even where one judges wrongly (e.g. confuses cold for pain, as in ‘dental fear’), further attention to the experience can correct the verdict. So, one can tell introspectively that one is in s-pain. Moreover, as part and parcel of this direct knowledge, one can tell what kind of s-pain one is in: nagging, sharp, burning, aching, tearing, and so on. The pain literature is replete with examples of the many kinds of pain subjects can directly distinguish between. And it does not seem that one could tell introspectively that one was in pain without having a sense of what kind of pain it was. That is, one apprehends the determinable of conscious s-pain by way of apprehending a determinate form of s-pain.

Painfulness, suffering, and consciousness


PQ2 assures us that if we can tell these things about conscious s-pains that is thanks to their qualitative character. Specifically, to be able to tell pains from non-pains introspectively requires that there be s-pain-specific qualitative character—paininess. It does not require that all s-pains share some single determinate qualitative unit—a distinctive feeling. Perhaps s-pains share nothing that can be isolated independent of their classification as pains (see the discussion of painfulness in the next section). But still, s-pains might all have a qualitative character such that we can tell directly from experience whether we are in pain. Here is an analogy that will be useful throughout: Some colour experiences arguably have something literally in common—experiences of red and orange, for instance. But many, perhaps most, do not—experiences of blue and red, for instance. So there is no determinate qualitative unit that all colour experiences have in common. Nonetheless we know, directly on the basis of experience, whether we are having a colour experience, and which one we are having. That is because colour experiences have proprietary and distinctive qualitative character. PQ2 tells us the same is true of s-pains. An alternative account might be that we have some reliable, but brute, identification mechanism that delivers conscious beliefs about what kind of s-pain we are in. Such a mechanism would not justify attributing qualitative characters to s-pains. But this suggestion evidently does not do justice to how we actually know we are in conscious pain. Identifying what kind of s-pain one is in, in the sense relevant to PQ, is not like finding that one can answer a quiz question though unaware of quite where the answer came from. In fact in such cases there is a question mark over the justification of the belief, even if it is correct and reliably produced. But there is no corresponding question mark over one’s justification for introspective beliefs about the kind of s-pain one is in. When one ascertains what kind of conscious s-pain one is in one does this by attending to one’s present experience—and this is clearly to pay attention to the kinds of qualitative character within one’s introspective purview. So PQ2 is true. PQ’s conclusion may seem obvious and PQ otiose. Participants in the pain literature, and beyond, routinely describe s-pain as involving qualitative character. But PQ is useful since exactly the same is true of painfulness, with exactly the same upshot. Here is the argument: PFQ PFQ1. One can tell by introspection that one’s conscious pain is painful, and the manner in which it is painful. PFQ2. This would not be possible unless conscious painfulness had proprietary and distinctive qualitative character. PFQC. Hence, conscious painfulness has proprietary and distinctive qualitative character. Consider just one dimension of painfulness: intensity. PFQ1 is clearly supported: we can tell directly, by attending to experience, just how intensely


Sam Coleman

painful a pain we are undergoing. This is not to deny that we are ever in painful states whose intensity and other characteristics we fail to notice—this can happen when one is distracted from pain, e.g. in sport. Indeed, I will argue later that s-pain and painfulness (even suffering) might occur without consciousness altogether. The point of PFQ1 is to highlight an ability we have, and the case of painful intensity speaks strongly in support. Consider again pain asymbolics: they say they are in pain that is not painful. How can they tell? They are introspecting. In fact we take their introspective reports about their s-pains and the painfulness—or not—of these exceedingly seriously, building a whole problematic for pain research thereupon. This also supports PFQ1. The argument for PFQ2 is as in previous iterations of the argument: what this kind of direct, non-inferential, introspective knowledge latches onto are qualitative characters. A particular objection is worth considering here, however. A desire theorist might allege that pain’s unpleasantness consists in its being the target of a non-instrumental desire that it stop. And it might seem plausible that one can introspectively identify the relevant desire, hence painfulness, in a given case. Yet desires, the objection continues, lack qualia. Hence PFQ2 is false. In reply, introspectibility seems a very strong reason to attribute qualitative character to a mental state. If a state lacks all qualitative character, all ‘what-it-is-like-ness’, so that there is nothing it is like for the subject to be in the state, then just what is it she latches onto when introspecting? What marks the supposedly quality-free state out against other introspectible states, notably those that do possess qualia? The desire would rather seem to be an unidentified flying mental object, if it truly lacked all qualitative character. If there is nothing it is like to have the state, then, it seems, it may as well be unconscious.11 Thus, it appears, either desires are introspectible and have qualitative character, perhaps of a special ‘conative’ sort,12 or they lack qualia and are not introspectible after all. We would be back to the brute identification mechanism dismissed above in connection with PQ2. So PFQ2 stands. This is where the point that qualia are not mere blanks, but can carry content—in fact cannot help but carry it (consider the way perceptual qualia set the scene for one13)—is important. Hence, my objection to desire theories of painfulness is not that desires lack qualia; we may well have some desirish-phenomenology when in unpleasant pain, perhaps even unavoidably. The objection is the simple wrongness of saying that the stop-desire determines pain’s unpleasantness rather than vice versa. If the pain does not hurt already why should I want it to stop? I conclude that painfulness is a quale. PQ shows that s-pain has the hallmarks of being a quale. Its conclusion is widely accepted. PFQ points out that painfulness has these same hallmarks. Hence, given the plausibility of PQ, it is hard to avoid the conclusion that painfulnesss, too, consists in qualia. Hence we reach the qualia view (QV): QV S-pain and painfulness consist in qualitative character.

Painfulness, suffering, and consciousness


To illustrate QV, consider a thought experiment. As Meditation I closes, doubting Descartes has whittled his being down to a core of pure conscious experience—what he finds undoubtable. He points out that though he may not see, for he may lack eyeballs, he surely seems to see—he has visual experience. The same gap between experience and its object does not exist for pain: even if one can be in pain without feeling it, one cannot feel pain without being in pain. Now, is it conceivable that a disembodied Cartesian spirit could exist in a state of excruciating pain? Unfortunately it is. Such a spirit would be a pure consciousness, and the contents of such a consciousness would be practically exhausted by forms of qualitative character,14 among them the sensory, affective, and cognitive. This confirms that s-pain and painfulness are qualia.

The heterogeneity problem QV’s stiffest challenge is the ‘heterogeneity problem’ (HP)—other objections pale in comparison.15 Such is HP’s prominence, and such QV’s initial plausibility,16 that refuting HP is tantamount to establishing QV.17 HP says painfulness cannot be a quale, because painful pains vary so much in qualitative character that there is nothing they could have in common to constitute their unpleasantness. Sidgwick is usually quoted at this point. Of pleasure he famously says: [W]hen I reflect on the notion of pleasure … the only common quality that I can find in the feelings so designated seems to be that expressed by the general term ‘good’ or ‘desirable’ … Hence, while I cannot define Pleasure—at least when we are considering its ‘strict value’ for purposes of quantitative comparison—as the kind of feeling which we actually desire and aim at, I still recognize as its essential quality some relation to desire or volition. I propose therefore to define it as feeling which, when experienced by intelligent beings, is at least implicitly apprehended as desirable.18 Philosophers take these remarks as applicable to painfulness, mutatis mutandis. A closer look, however, reveals Sidgwick to be less clear-cut. As has been noted, Sidgwick speaks of pleasure as a feeling. In fact the whole passage seems compatible with the view that pleasure is a quale disposed by nature to arouse desire—and conversely, presumably, for painfulness. The key interpretational issue would then be what makes a painful pain unpleasant. If the unpleasantness lies in the sensory quality of the painful pain, then it can ground a judgement of undesirability: this would be a version of QV. But if the feeling is not unpleasant except in so far as judged(/judgeable) as undesirable, then QV is false. Sidgwick, at best, vacillates between these alternatives. The point to glean is that an essential relation to (un)desirability does not itself render QV false.


Sam Coleman

But HP can be formulated more simply. Here is Heathwood: ‘there doesn’t seem to be any one feeling…common to all occasions on which we experience [pain’s unpleasantness]’.19 Let us grant the introspective claim. If QV entailed that every painful state featured a single determinate quale common to all such episodes, then the phenomenological observation behind HP would surely falsify it. But a more flexible, hence plausible, view is that the painfulness quale is a determinable, and each painful state involves one of its determinates. This emendation suffices to evade most formulations of HP. However, some have rejected even the emendation. Heathwood further denies that pleasures(/pains) share ‘hedonic tone’, which is meant to exclude determinable theories. But there is an ambiguity hereabouts that yields two versions of the determinable theory, only one of which is vulnerable to Heathwood’s criticism. One view would be that painfulness is a determinable each determinate of which bears a recognisable internal resemblance to the others, considered pairwise, in the way determinates of redness clearly bear such a resemblance relation to one another; a relation internal to the determinable, and which unifies it: nothing that does not appear reddish can be a determinate of red.20 Plausibly there is no such determinable of painfulness, given the very wide range of ways feelings can be painful. But here is a second view: While painfulness is a determinable quale, and while some of its determinates recognisably resemble (e.g. the painfulness of a pinprick and that of a sharp nail-pinch), others seem not to resemble at all (e.g. the painfulness of a burn and that of a stomach ache). What then licenses talk of painfulness as a determinable quale? Is there any model we can appeal to of a determinable quale where no internal resemblance relation unifies the class of all determinates that fall under it? Colour qualia seem to offer a perfect model. Many colour qualia determinates do resemble—e.g. redness and orange.21 But many do not: e.g. blue and red (or determinates of these). Thus there is no internal resemblance relation between all determinates of the determinable colour quale. There is nothing that unifies them independently of their being colourful, i.e. belonging to the determinable. If one sought a qualitative factor stable in all colour qualia, or even an internal resemblance dimension along which all determinate colour qualia stood, one would fail.22 Heathwood asks: ‘What is it about each of these things in virtue of which it is correctly classified as a [painful] pain?’23 If we asked that question about colour qualia, we could give no answer independent of belonging to the determinable. One might then infer that colour qualia are united only by our disposition to classify them as such. But this would be a clear mistake—of course colour quale is a genuine determinable. It follows that HP is not a well-founded objection to QV.24 Someone might reply that all colour qualia have as a component a hue value, and that is what grounds the internal unity of the determinable—the common factor. Yet the point just observed about colour qualia even more palpably applies to hues: they have no internal resemblance relation taken

Painfulness, suffering, and consciousness


as a class. Certain hues simply do not resemble, considered pairwise, except in the sense of being members of the determinable hue. Yet few would say that the metaphysics of hue consists in our disposition to judge a property a hue. HP fails. Perhaps we catalogue the painful states by invoking their dispositions regarding desire, say. But that may nonetheless be a procedural necessity rather than an indicator of their metaphysics. Similarly, what can we say about colour qualia except that we find them colourful? But we find them colourful because they are colourful, not conversely. QV says the same about painfulness.25

S-pain, painfulness, and pain How are s-pain (i.e. paininess) and painfulness related? The points below are largely independent of s-pain being paininess qualia, so I will mostly talk simply of s-pain. First, we can be confident that s-pain and painfulness are distinct items. Pain asymbolia seemingly proves this, since asymbolics report having pain that does not hurt. As per PQ, they recognise their pain by introspecting. As per PFQ, they recognise that their pain does not appear painful (or only minimally so) likewise. If s-pain can exist without painfulness—or affect of any sort—then these are different things. There is a threat to this reading on which asymbolia pain differs from standard painful pain,26 hence to this argument that s-pain and painfulness are distinct. According to Klein, though asymbolics actually have standard painful pain, what they lack, which explains their indifference to pain and bodily damage, is an attitude of bodily care.27 If that lack explains the phenomena there is no need to say that painfulness is missing, which blocks the inference to the distinctness of s-pain and painfulness. Now it may be that we should attribute asymbolics a loss of this attitude, to explain why, say, one asymbolic walking in the road cheerfully ignores an approaching truck. But there remains good reason to think asymbolia pains are non-standard. For were it only the attitude of bodily care lacking, we really should expect asymbolic reports of dreadful pain which they nonetheless did not mind. But, strikingly, we do not get such reports.28 To repeat: we take introspective reports seriously when it comes to asymbolia. It is because asymbolics report being in pain that we ascribe s-pain to them. Therefore, since they omit talking about, even deny, painfulness, we ought to infer that painfulness is not typically detected. It follows that in asymbolia either painfulness is absent altogether, or it is unconscious. I will investigate this latter possibility later. For now, we can reason as follows: since s-pain can be present when painfulness is either unconscious or absent, s-pain and painfulness are different things. They are different since one can be conscious where the other is unconscious, or is even absent—hence something is true of one that is not true


Sam Coleman

of the other.29 It is possible that they are properties of a single state of the subject, or aspects of one property (if we allow aspects into our ontology). In that case a state of the subject is, perhaps, partially felt and partially not. But still it follows that they are distinct properties, or aspects, of said state. Given my earlier argument, plus the fact that subjects (e.g. asymbolics) can discriminate s-pain from painfulness introspectively, it additionally follows that paininess (i.e. the s-pain quale) and painfulness are type-distinct qualia. It is important to be clear about what asymbolia does not show. It has been taken to show that s-pain can occur without affect. But, as noted, that inference relies on assuming that painfulness is exclusively conscious. If painfulness can be unconscious, then even asymbolia pain may have affect— unconscious painfulness. But then the rationale for thinking s-pain could exist without affect would be undermined: for until asymbolia, and related phenomena such as morphine pain, were discovered, the notion that s-pain might lack affect altogether would have struck us as highly counterintuitive. Obviously, the thesis that painfulness might exist unconsciously is also counterintuitive to many—especially on my model, which construes painfulness as qualitative character, a property widely held to require consciousness. We would need good reason to believe in unconscious painfulness qualia. I will canvas some reasons later on. The point for now is that until and unless unconscious painfulness is ruled out, asymbolia fails to demonstrate that pain can lack affect. All it shows is that s-pain and painfulness are distinct, not separable. How are s-pain and painfulness related in standard painful pain? On one hand, it strongly appears from introspection that in such cases s-pain itself is painful—pain’s unpleasantness is somehow bound up with the s-pain in an intimate fashion. Rachels is right that ‘When you twist your ankle or jam your finger, the experience itself seems to hurt; the unpleasantness seems to be right there in it.’30 It thus appears introspectively that painfulness attaches to s-pain, such that painfulness is intrinsic to a painful pain state. I think this introspective datum counts against views on which what (constitutively) makes an s-pain unpleasant is completely disjoint from it, such that what makes pain unpleasant is not intrinsic to the pain state (e.g. a desire theory). It categorically cannot be said that a desire is literally part of a pain state. Rather, pain’s unpleasantness on desire theory consists in an s-pain’s bearing a relation to something outside the pain state itself, the relevant stop-desire.31 On the other hand, as asymbolia showed, it does seem that s-pain and painfulness are different things. So here are two theses we want to capture: A: Painfulness attaches to s-pain, such that painfulness is intrinsic to a painful pain state. B: S-pain and painfulness are different things.

Painfulness, suffering, and consciousness


It would be easy to assert A were s-pain and painfulness identical in instances of painful pain. But B rules this out. On the other hand it would be easy to endorse B if the phenomenology were not such that painfulness seems to accrue tightly to s-pain, such that painfulness is intrinsic to the pain state. We seem to want a relationship closer than that of two discrete states, but looser than identity. The obvious relation to appeal to is constitution. Here, then, is a model capable of honouring A and B. It springs, additionally, from a remark of Crisp’s, who notes that enjoyment, though a quale, is intentional: it is directed upon another mental state, viz. the experiential property one enjoys.32 This is represented as pleasant. I believe the same holds for painfulness with respect to s-pain: painfulness intentionally qualifies s-pain. This is an observation about the phenomenology. So we have thesis C to honour as well: C: S-pain and paininess, more than merely properties of the same state, stand in an intentional relationship that underpins a phenomenological intimacy. In a standard case of painful pain we have two items, a paininess and a painfulness. I suggest that they constitute a bi-partite state—this is pain properly so-called. Following Crisp, when painfulness occurs it is clearly directed on the paininess it qualifies—it is not an isolated quale associated with paininess only by compresence. Imagine you experience multiple different painful pains simultaneously—you shut your thumb in the door while stepping on a pin, all against a background migraine. The paininesses felt are qualitatively distinct, and so are the relevant painfulnesses. Moreover, each painfulness feels to belong to the relevant paininess. That, somehow, is part of the experience of painfulness. There is no question as to which painfulness attends which paininess in such cases, and that is because the qualia settle the matter directly. This directedness speaks of intentionality—the painfulness quale presents, represents, or qualifies, the paininess quale—sufficing for it to be felt as painful. Adapting a model of Kriegel’s for the analysis of conscious states, I suggest that the painfulness element represents the paininess as painful. But because they are constituents of one conjoint mental state, in directly representing the paininess the painfulness, additionally, indirectly represents the whole of which it is a constituent. Kriegel posits indirect representation as a species of non-causal representation that plays a key role in consciousness. Here I strip it of that role—I do not suggest that a painfulness and a paininess standing in this relationship must comprise a conscious state, even if I am currently discussing conscious states. Roughly, X indirectly represents whole Y when X directly represents a large or significant portion of Y—e.g. a painting directly representing the front of a house indirectly represents the whole house.33 Thus the entire conjoint pain state is represented as painful, by means of painfulness being ascribed in the first place to the paininess. What makes the state


Sam Coleman

conjoint? Again following Kriegel, I appeal to phenomenological unity—the paininess is felt as painful as much as the painfulness is felt as qualifying the paininess. This means the intentional-cum-phenomenological relationship they bear is essential to the state, and that makes it a complex, or conjoint, state. Now we can say, what seems right, that painfulness is intrinsic to painful pain. It is intrinsic to pain as one constituent of a bi-partite self-representing state. Moreover the intimate attachment of painfulness to paininess is captured. But we also allow that paininess and painfulness are metaphysically distinct. This model thus accommodates theses A and B, plus C, Crisp’s observation about intentionality. Pain, then, is a single complex state held together by an internal representational relation that standardly manifests, phenomenologically, as sensory-pain-as-qualified-by-painfulness.34 On this model, overall, s-pain and painfulness have to do with the intrinsic properties of the pain state, with its qualitative character, not with ways this qualitative character is further represented, conceived of, or ‘attitudinalised’. It follows that if the kinds of qualia we find in consciousness are capable of existing without consciousness, there could exist unconscious painful pain. I return to this below.

Painfulness and suffering pain What is the relation between painful pain and the (direct) suffering it yields? On one hand, when painful pain is suffered this seems to be something tightly related to the sensation. It might require undergoing a painful experience to remind yourself of this, but there is a sense, pondering the phenomenology, that this is all there is to my suffering the pain. On the other hand, there is good reason to think one’s suffering is not simply identical to the instance of painful pain. Massin (Chapter 4, this volume) canvasses some of these reasons, but perhaps the most powerful is this: whereas s-pain and painfulness present as having a specific bodily location, suffering does not. We do not say: I am suffering just there—indicating a painful bodily zone.35 Again we are pulled in two directions: suffering pain is tightly bound to the pain, so the phenomenology tells us, yet it must also be something distinct from the pain. Most theories certainly veer too far in the direction of separating pain and suffering. Corresponding to the theory that painfulness consists in desiring that s-pain cease, there is a desire theory of suffering on which to suffer pain is to desire that it not exist. But this faces its analogue of the major problem for the desire theory of painfulness: just as we should say that we want pain to stop because it hurts, not conversely, here we should say that we desire painful pain to stop because we are suffering on its account, not conversely. Relatedly, suffering is itself an unpleasant state to be in, and the phenomenology tells us that suffering’s unpleasantness derives from the unpleasantness of the pain suffered—somehow the suffering inherits this. A good theory

Painfulness, suffering, and consciousness


must explain this unpleasantness, and the ‘inheritance relation’, satisfactorily. This is a key constraint: I call it the demanding desideratum (DD). The desire theory cannot meet it, since it must say suffering’s unpleasantness lies in our desire that it stop, when we evidently desire that it stop due to its antecedent unpleasantness. Hence this theory does not account for suffering’s unpleasantness—it confuses condition with symptom. McClelland (Chapter 2, this volume) suggests suffering consists in unpleasant disruption of one’s ‘economy’ of mental states. So in the case of suffering from pain, we have two loci of unpleasantness: the pain itself is unpleasant, and the disturbance it causes to one’s mental states is also experienced as unpleasant—it might provoke anxiety, say, or distract one from nicer thoughts. McClelland labels the latter ‘holistic unpleasantness’, and it is where he thinks suffering comes in. This theory permits pleasant disruptions to mental life, as in an all-consuming love. That would not be a case of suffering. Presumably one could suffer from love, even though love felt pleasant, if it provoked unpleasant mental disruption. This is all plausible. McClelland is concerned to avoid a phenomenology-based theory of the sort I favour. Hence his emphasis on mental disruption in addition to holistic unpleasantness phenomenology. But I doubt that he manages to sail safely past the qualia rocks. He allows that one might suffer a notional state, if one wrongly took oneself to be in physical pain, say, and was unpleasantly mentally disrupted as a result. But what if the disruption was only notional? If one experiences holistic unpleasantness, and takes oneself, wrongly, to be mentally disrupted, does one suffer in McClelland’s sense? Imagine that we point out to the subject that they are cognitively unimpaired, perform everyday tasks normally, and so on—will this make them feel any less that they suffer? It might only increase their experience of distress. In fact at that point they may well become mentally disrupted. Far be it for us to tell them that they are not suffering when they have every experience of suffering (it might feel just like the last time they suffered, when they genuinely were mentally disrupted). It would appear strange to overrule them on the basis that a brain super-scanner reveals only an illusion of mental disruption. Hence holistic unpleasantness phenomenology suffices for McClelland-style suffering, and his turns out to be a phenomenology-based account. So much the better. Still, I think McClelland fails to capture the intimacy of suffering and pain. In a case of suffering pain McClelland posits two instances of unpleasantness, painfulness and holistic unpleasantness. The latter is caused by the former, to be sure, but it seems rather distantly related—unconnected. McClelland, recall, allows that I might suffer even if my pain is only notional—when I take myself to feel painful pain though I do not. Relatedly, the theory does not really explain how the unpleasantness of suffering derives from that of the suffered state—rather these are discrete, merely causally-related, hence mutually permutable, items. So McClelland’s theory does not meet DD. I think it is best suited to modelling mental suffering. But even then it seems that what one directly mentally suffers is the holistic unpleasantness


Sam Coleman

phenomenology, and mental disruption is only a normal cause of this, not a constitutive factor in suffering. McClelland’s model of suffering from physical pain is really a model of indirect suffering: one suffers the pain indirectly, not directly, in virtue of its disruptive mental tendency. But nonetheless there is a state one directly suffers: the holistic unpleasantness phenomenology. Thus indirect suffering, whether mental or physical, bottoms out in direct mental suffering. Massin’s theory in Chapter 4 is that suffering is the correct psychological reaction to painful pain. It is correct in part because the pain is unpleasant, and it is appropriate to suffer on that account. But this does not tell us what suffering is, nor why it is unpleasant. In fact Massin shirks the question of saying substantively what suffering is, averring that pain and suffering are primitive phenomena. Why would reacting correctly to an unpleasant state be itself an unpleasant state? Massin’s account faces the same difficulty he levels against the view that suffering is perceiving pain’s badness, namely DD. Generally, perceiving something unpleasant need not be an unpleasant state—it can even be a positive state if it helps one overcome the unpleasant thing. Similar remarks apply to reacting correctly to an unpleasant thing. So Massin’s account fails DD. It neither explains suffering’s unpleasantness nor how this derives from pain’s unpleasantness.36 These accounts suffer from making suffering too detached from what is suffered. The upshot is that they struggle to capture suffering’s unpleasantness—the unpleasantness intrinsic to suffering. We noted that suffering’s unpleasantness seems to flow from the unpleasantness of painful pain. But we cannot identify the two, either (recall Massin’s point about bodily location). So here is another suggestion. I propose that suffering pain is simply the property of instantiating pain’s painfulness. More generally, (directly) suffering an unpleasant state, mental or physical, is the subject’s property of instantiating that state’s unpleasantness. This makes suffering a ‘higher-order’ property—it is the property of instantiating some or other unpleasant firstorder property. Suffering pain is thus a property distinct from pain, but closely related to it. This seems to get the balance right. In particular, we can explain why suffering pain is unpleasant in a way that derives from the unpleasantness of pain. For if pain is unpleasant—feels painful, on my account—then it will also be unpleasant to have the property of instantiating pain’s unpleasantness. For who wants to have the property of instantiating an unpleasant property? The higher-order property clearly inherits the unpleasantness in a suitable way, without being identical to pain’s unpleasantness. Call this the instantiation theory of suffering (ITS). ITS captures mental and physical direct suffering under the same umbrella: ‘suffering’, the idea goes, is a general term we use to denote a subject’s property of instantiating one or other qualitatively unpleasant first-order property. This generic character may help to explain the widely noted fact that it is easier to empathise with someone’s suffering than with the specific form of state from which they suffer—for we have all suffered.

Painfulness, suffering, and consciousness


A worry about ITS is that higher-order properties plausibly lack causal efficacy—this objection looms over role-functionalist analyses of mental properties, for instance. But the threat is less severe regarding suffering. It does not seem wrong to say that suffering owes its difference-making powers to the suffered state—the ‘role’ player. But neither is it a negligible matter to have the property of instantiating a painful pain, say—that is indeed a property of which one may reasonably want, and seek, to be relieved. But note, when suffering pain one wishes to be relieved by removal of the painfulness, not mere removal of the suffering (were that possible): this fact supports my account. Another objection is that I cannot say, with Massin, that suffering is a reaction to pain. If suffering is a reaction then it is plausibly caused by pain, but the property of instantiating pain’s unpleasantness is not caused by pain or its unpleasantness. But I think—against McClelland, Massin, Klein, and others—that it is wrong to say that pain causes direct suffering (it certainly can, indeed must, cause indirect suffering—as in McClelland’s account). The relation is closer than that. If pain really could cause direct suffering then suffering the pain could conceivably continue once the pain had ceased—like ripples in a pond where the stone has long since sunk. But it makes no sense to directly suffer a property one no longer feels nor instantiates. On the other side, causation permits a time-lag between cause and effect: a period when which the first exists without the second. But this implies one could instantiate an agonising pain without suffering it for some time. That sounds wrong. Surely being in such a state suffices for suffering it. And, though we must strictly shun ‘reaction’ talk, that all supports ITS. We can still say pain yields, or occasions, suffering, understood non-causally. Suffering pain on ITS is not a quale. But it is, ultimately, the property of instantiating the bi-partite quale of a painful-paininess. Thus ITS does justice to the observation that one can tell one is suffering by introspection. ITS, additionally, has the virtue of explaining why suffering pain lacks a presented bodily location despite being closely related to pain, which has one. For the property of instantiating a property that presents a bodily location does not itself present a bodily location. It is, rather, a ‘global’ property of the organism. ITS thus seems to get the parameters of pain and suffering correct.37

Suffering and (un)consciousness It may seem strange to follow up a phenomenology-based account of pain, painfulness, and suffering by claiming that these can occur unconsciously. But introspection, though it can tell you for sure that a mental state is present, cannot rule that it is not present—just ask Oedipus. That is significant. Some researchers allow that pain can exist unconsciously, but vastly more are loath to countenance the existence of unconscious painfulness, let alone suffering. The prevailing view ties suffering tightly to consciousness.38 The same


Sam Coleman

view dominates, more confusedly, in the public sphere, as evidenced during the Terri Schiavo case. Though Schiavo was legally ruled to be incapable of suffering precisely due to her unconscious vegetative state, clinicians and journalists nonetheless made strenuous assurances that the removal of her feeding tube would not cause her discomfort or suffering.39 We must be clearheaded, considerate, and fair when evaluating the possibility of unconscious suffering, since the consequences, if it exists and we ignore it, are terrible to contemplate. On my account the question of whether suffering can occur unconsciously aligns with the question of whether a subject could instantiate, unconsciously, such properties as painful pain qualia. Many hear ‘qualia’ as including consciousness—if so think ‘qualitative character’ where I say qualia, properties qualitatively of the same kind only with subjective awareness subtracted. That very notion might sound oxymoronic to the same many (or more)—but now at least we are into substantive issues. Despite common philosophical wisdom, which often cites pain as the paradigmatic essentially conscious quality, there is decent evidence for unconscious pain qualia, from such cases as pains suffered during sport that are felt intermittently (if at all) until afterwards, and migraines that last for days—seemingly day and night—and which can wake the sufferer from dreamless sleep. If pain really wakes her, as opposed to a pain-free neural substrate, the subject was instantiating unconscious pain. These examples show at the least the coherence of the notion of unconscious pain qualia.40 What about painfulness? Earlier I said it was not ruled out that asymbolics instantiate unconscious painfulness. Assuming what Block famously challenged, namely a tight link between phenomenal consciousness and the capacity to cognitively access, including report, one’s mental states,41 unconscious painfulness may be unreportable. But the lack of any apparent pained physiological response in asymbolics is quite strong evidence that they are free of painfulness. For unconscious painful pain would still be…painful. The conventional interpretation of asymbolia is thus likely, if not certain, to be correct. There are, however, better cases. Block provides two such: pain under hypnosis, and under general anaesthetic.42 Hypnosis is used analgesically, and often effectively, in that the hypnotised subject does not report pain during her procedure. Still, there is evidence of pain from heightened blood pressure and heart rate—just the signs that closely accompany conscious painful pain. Moreover, the ‘hidden observer’ technique seems to make contact with a part of the person who does feel—hence can report—excruciating pain matching the period of the procedure. The second case is of pain undergone during dental work. Patients were given general anaesthetic, but later air travel in unpressurised cabins ‘reactivated’ the pain in the relevant areas. It was then found that using local anaesthetic on one side of the mouth prevented this effect for that side—so one way or another local really kills pain! Block holds that sensory qualities are necessarily felt, hence he casts these as cases of cognitively inaccessible conscious pain. But given the

Painfulness, suffering, and consciousness


coherence of the posit of unconscious qualia, including painful pain, it is far more plausible to preserve the link between consciousness and cognitive access, and say these pains are unconscious, but no less painful for all that— as the physical symptoms show. In a similar vein, Hardcastle stresses the complexity of unconscious pain processing, and the negative long-term consequences of such processing in chronic patients regardless of indications of consciousness.43 She concludes that ‘pain processing’ is bad whether conscious or not. But I am inclined to view the processing as evidence of unconscious painful pain, i.e. qualia, especially given the upshots for patients. Given ITS, it follows that the relevant subjects suffered unconsciously. This result will serve as reductio for some. But our intuitions here are actually nuanced. Very arguably, we would not weigh unconscious pain and suffering as heavily as conscious pain and suffering: I would surely take the option, if available, of having my pain made unconscious, if not eliminated, during a painful procedure. But that does not mean we are blasé about unconscious pain and suffering. No-one would consent to being unconsciously tortured throughout their sleeping hours by an ingenious electro-shock treatment which, while causing a good deal of unconscious pain, left no physical or psychological mark in waking life—not even if money was at stake! It is easy to confuse the relative lack of central cognitive and subjective organisation in unconsciousness for the absence of a subject who has anything at stake, or can suffer. But unless we believe, with Descartes, that we are only subjects in so far as we are conscious, then indeed the person has unconscious aspects, including sensory ones, and can suffer in respect of those aspects. Given what hangs on it, this is a possibility about which we should be serious and circumspect.

Notes 1 Dumas (1844/2003: 217). 2 See Chapter 2 of this volume, as well as my further discussion below. 3 It is implausible that these cases feature unpleasant (or non-pleasant) pain that causes pleasure. Then we would see the pain state as something to be borne to get to the pleasure. But we seek, or relish, such pains themselves. 4 N.B. Not every unpleasant pain is suffered. A mildly unpleasant pain may only bother us, a state at the low end of the spectrum on which suffering is at the high end (McClelland, Chapter 2, this volume). 5 E.g. Loeser (2000: S2–S6). I will assume reported pain is phenomenally, as well as access, conscious (contra Block 1995)—see also my further discussion below. 6 E.g. Kripke (1980: 151). 7 E.g. Clark (2005). 8 Clark (2005: 180). 9 Thanks to David Bain for this term. 10 Pitt (2004). 11 This may be unfair to unconscious states, since I later suggest that there is often something it is like to have them—they can have qualia. 12 Kriegel (2015). 13 Horgan and Tienson (2002).


Sam Coleman

14 Two non-qualitative introspectible elements of consciousness are its structure, and the alleged ‘awareness of awareness’ (Kriegel 2009)—neither are candidates for the natures of s-pain/painfulness. 15 1. Introspection has been used as a source of objection to QV. But Aydede (2014) notes that qualia theorists and opponents alike invoke it, meaning introspection cannot be a clear source of objection. Moreover, QV’s opponents plausibly invoke introspection in order to advance HP: their argument is that introspection reveals no common painful unit, hence painfulness is not a quale. Thus the objection from introspection is no stronger than HP. 2. The argument that QV struggles to explain pain’s motivational character wrongly assumes qualia must lack intentional content. 3. Attitudinal theories may seem suited to a reduction of pain/painfulness— but this is only an argument against QV if one finds reductionism appealing. HP, hence, is by far the strongest independent objection to QV. 16 Heathwood (2007: 26). 17 Crisp (2006: 629). 18 Sidgwick (1893: 128). 19 Heathwood (2007: 26). He speaks of pleasure, but says the point applies to pain—I assume he includes painfulness in this. 20 Likewise, plausibly, for the determinable qualities of sweetness, warmth, and coffee taste, sometimes invoked in discussion of HP. Cf. Kagan’s (1992) analogy of volume. 21 Qualitative resemblance is not sufficient for a determinable (orange is not a determinate redness) but some evidently think it necessary—those who advance HP against determinable QV. 22 Colour qualia are on a spectrum, in that one can get from any determinate to any other by means of tiny, almost imperceptible, differences. That might be deemed a condition on a determinable property. But it is not implausible that the same is true of painfulness’s determinates (likewise paininess), albeit in a determination space larger than that of the colours. Compare the space for smells (Young et al. 2014). 23 Ibid. 24 Crisp (2006) offers a ‘determinable’ QV in response to HP, but I find his formulations unclear as between the two versions I distinguished. 25 It is inessential that the presentation is in terms of colour (/hue) qualia, as opposed to colours—i.e. construed as external properties. In fact the point might apply even more strongly if colours are identified with physical properties, given the lack of plausible resemblance relations between the multifarious physical realisation bases of colours. 26 E.g. the reading of Grahek (2007). 27 Klein (2015). Later he embraced the view that asymbolic pain consequently lacks unpleasantness. 28 Cf. Bain (2014). 29 See also Ploner, Freund, and Schnitzle (1999) case of a man who reported feeling pain’s unpleasantness without s-pain. 30 Rachels (2000: 196). Cf. Carruthers (2004)’s bee-sting case. 31 It is not that a stop-desire causes a non-unpleasant pain to change its intrinsic character. Otherwise we could focus on the constitutive unpleasantness factor intrinsic to the pain state. 32 Crisp (2006: 628). 33 Kriegel (2009: 116). 34 For cases of pleasant pain, we can imagine a pleasantness quale entangled with a paininess quale in much the same manner. 35 Note this is not to say s-pain or painfulness have bodily locations (nor that suffering lacks one): this is a point about their intentional content. 36 In fairness Massin denies that suffering derives its unpleasantness from the pain suffered. But this seems at odds with the claim that we suffer pain (see his ‘small

Painfulness, suffering, and consciousness

37 38 39 40 41 42 43


stone’ case), and with the closeness of pain and suffering. Nor does this relieve Massin of the obligation to say what makes suffering unpleasant. So his account is unsatisfying. Similarly, suffering grief, say, will be the property of instantiating grief ’s unpleasantness; and one does not suffer from grief just there, exactly where the grief is located in one’s mental space. See e.g. Edwards (2003), Kahane and Savulescu (2009), and DeGrazia (2014). See e.g. The majority of European doctors believe an unconscious patient undergoes no pain (Dermertzi et al. 2009)—therefore a significant minority demur. Cf. Rosenthal (1991). Block (1995). Ibid., p. 244—see his paper for further references. Hardcastle (2018).

Acknowledgements Thanks to Tom McClelland, David Bain, and Jennifer Corns for very helpful discussion and suggestions.

References Aydede, M. (2014) How to Unify Theories of Sensory Pleasure: An Adverbialist Proposal. Review of Philosophy and Psychology 5: 119–133. Bain, D. (2014) Pains that Don’t Hurt. Australasian Journal of Philosophy 92(2): 1–16. Block, N. (1995) On a Confusion about a Function of Consciousness. Behavioral and Brain Sciences 18: 227–247. Carruthers, P. (2004) Suffering without Subjectivity. Philosophical Studies 68(2): 316–336. Clark, A. (2005) Painfulness is not a Quale. In M. Aydede (ed.), Pain: New Essays on Its Nature and the Methodology of Its Study (pp. 177–197). Cambridge, MA: MIT Press. Crisp, R. (2006) Hedonism Reconsidered. Philosophy and Phenomenological Research 73(3): 619–645. Degrazia, D. (2014) What is Suffering and What Sorts of Beings Can Suffer? In R. M. Green and N. J. Palpant (eds), Suffering and Bioethics. New York: Oxford University Press. Demertzi, A., Schnakers, C., Ledoux, D., Chatelle, C., Burno, M.-A., Vandaudenhuyse, A., Boly, M., Moonen, G. and Laureys, S. (2009) Different Beliefs about Pain Perception in the Vegetative and Minimally Conscious States: A European Survey of Medical and Paramedical Professionals. Progress in Brain Research 177: 329–339. Dumas, A. (1844/2003) The Count of Monte Cristo. London: Penguin. Edwards, S. D. (2003) Three Concepts of Suffering. Medicine, Healthcare and Philosophy 6(1): 59–66. Grahek, N. (2007) Feeling Pain and Being in Pain (2nd edn). Cambridge, MA: MIT Press. Hardcastle, V. (2018) When is a Pain not a Pain? In D. Bain, M. Brady and J. Corns (eds), Philosophy of Pain: Unpleasantness, Emotion, and Deviance. London: Routledge.


Sam Coleman

Heathwood, C. (2007) The Reduction of Sensory Pleasure to Desire. Philosophical Studies 133: 24–44. Horgan, T. and Tienson, J. (2002) The Intentionality of Phenomenology and the Phenomenology of Intentionality. In D. J. Chalmers (ed.), Philosophy of Mind: Classical and Contemporary Readings (pp. 520–533). New York: Oxford University Press. Kagan, S. (1992) The Limits of Well-Being. Social Philosophy & Policy 9(2): 169–189. Kahane, G. and Savulescu, J. (2009) Brain Damage and the Moral Significance of Consciousness. Journal of Medicine and Philosophy 34: 6–26. Klein, C. (2015) What Pain Asymbolia Really Shows. Mind 124(494): 493–516. Kriegel, U. (2015) The Varieties of Consciousness. New York: Oxford University Press. Kriegel, U. (2009) Subjective Consciousness: A Self-Representational Theory. New York: Oxford University Press. Kripke, S. (1980) Naming and Necessity. Cambridge, MA: Harvard University Press. Loeser, J. D. (2000) Pain and Suffering. The Clinical Journal of Pain 16(2): S2–S6. Pitt, D. (2004) The Phenomenology of Cognition, Or, What is it Like to Think that P? Philosophy and Phenomenological Research 69(1): 1–36. Ploner, M., Freund, H. K. and Schnitzler, A. (1999) Pain Affect Without Pain Sensation in a Patient with a Postcentral Lesion. Pain, 81: 211–214. Rachels, S. (2000) Is Unpleasantness Intrinsic to Unpleasant Experiences? Philosophical Studies 99(2): 187–210. Rosenthal, D. M. (1991) The independence of Consciousness and Sensory Quality. Philosophical Issues 1: 15–36. Shaver, R. (2016) Sidgwick on Pleasure. Ethics 126: 901–928. Sidgwick, H. (1893) The Methods of Ethics. London: Macmillan. Young, B. D., Keller, A. and Rosenthal, D. (2014) Quality-Space Theory in Olfaction. Frontiers in Psychology 5: 1–15.


Suffering pains Olivier Massin

Within contemporary philosophy of mind, pain gets a lot of attention, suffering very little.1 This stands in contrast with other research areas: 


Theories of retributive justice and punishment put the concept of suffering at their core (Zaibert 2017, 2013; Fingarette 1977; Gray 2010; Hart 2008; Walen 2015). Wrongdoers, some claim, deserve to suffer. Punishment should therefore consist in making wrongdoers suffer in return: inflicting pain is rarely ever mentioned. Torture is standardly defined in terms of suffering (Roberts 2016). Torturing essentially involves making one’s victim suffer, but does not essentially involve inflicting pain on them. Waterboarding, sensory deprivation, disorientation cause intense suffering but no pain. Asymbolics can be tortured, although not pain can be inflicted on them. Compassion is typically characterised in terms of suffering (e.g. suffering for someone else’s sake; Nilsson 2011). Likewise, Schadenfreude is understood as a delight in the suffering—not in the pain—of others. Discussions about the moral status of animals typically hinges on their capacity to suffer (DeGrazia and Rowan 1991). Within Christian theology, Christ is said to have suffered for humans and sinners are promised eternal suffering (Roberts 2003; Talbott 2017). Clinicians assume that is the suffering of their patient that needs to ultimately be dealt with, not—or not just—their pains. “It is suffering, not pain, that brings patients into doctors’ offices in hopes of finding relief”, writes Loeser (2000; see also Cassell 1982, 1995). Although utilitarians sometimes speak of our duty to diminish the amount of pain, utilitarianism is arguably better spelled out in terms of a duty to relieve suffering (Mayerfeld 1996, 1999).

How is it, then, that philosophy of mind focusses mostly, if not exclusively, on pain? One reason, I surmise, is that pain and suffering are assumed to be very closely related phenomena. That is, one tends to assume either that pain and suffering are one and the same phenomena under different names, or that one is a species of the other, or that they both belong to a same genus, or that


Olivier Massin

one can straightforwardly be defined in terms of the other. If pain and suffering are such close cognates, then it is quite natural to assume that studying pain is not overlooking suffering, but rather laying the basis for a proper understanding of suffering. The view defended here, by contrast, is that pain and suffering are not only distinct, but utterly distinct: they neither belong to a common kind, nor does the one figure in the essence of the other. Suffering is an attitude, pain is a sensation located in our body. Their relation is not subsumptive or essential but rather psychological and normative: suffering is the usual and correct reaction to pain. The first section below rebuts an argument to the effect that the pain/suffering distinction is a distinction without a difference. I then point out four key differences between pain and suffering in the following section. Next I object to the view that pains are just suffered bodily sensations. I then reject the view that suffering is ever perceiving pain’s badness. I propose a positive account of the pain/suffering relation, dubbed the “reactive account”, before answering three objections to the reactive account. Finally, I argue that the reactive account undermines both the influential view that pain is incorrigible and the influential view that the ordinary conception of pain is paradoxical. A methodological comment is in order before we start. The following is an essay in descriptive psychology: it intends to describe the distinctions and relations between pain and suffering as we ordinarily speak about, think about and experience them. This means, first, that any questions about the relation between pain (or suffering) and brain states will be bracketed. This means, second, that the ordinary understandings of pain, suffering, and their relation is assumed to prevail by default: we need to give reasons if we are to abandon or revise them.

A distinction without a difference? Ordinary language distinguishes between having pain, feeling pain, suffering pain, and still other attitudes or psychological episodes one may entertain towards pain (enduring pain, enjoying pain, being indifferent to pain, etc.). But do these linguistic distinctions capture genuine psychological differences? Perhaps the grammatical complexity of “having a pain”, “feeling a pain” or “suffering a pain” does not reflect any metaphysical complexity. One influential consideration to this effect pertains to so-called cognate accusatives, such as “feeling feelings”, “dreaming dreams” or “thinking thoughts”.2 In such expressions, it is claimed, the verb and its direct object are so closely related that their grammatical distinction corresponds to no metaphysical difference.3 The argument against the pain/suffering distinction then goes as follows: P1 In “suffering a pain”, “a pain” is the cognate accusative of “suffering”. P2 The distinction between verbs and their cognate accusatives does not correspond to a metaphysical distinction.

Suffering pains


C There is no metaphysical distinction between pain and suffering. I believe that both premisses are false. “Pain” is not a cognate accusative of “suffering”. Things other than pains might be suffered from: one can suffer from a loss, an injustice, a disease, a discomfort … Second and more importantly, even if “pain” were a cognate accusative of “suffering”, it would not follow that pain and suffering are identical. Contrary to the prevailing assumption, the presence of a cognate accusative does not entail the lack of an underlying difference. Consider: “loving a beloved”, “tasting a taste”, “smelling a smell”. Trivial as they are, such expressions certainly do not point to an identity of the loving and the beloved, of the tasting and taste, of the smelling and the smell. One may in fact construe cognate accusatives for all intentional states (e.g. seeing visibilia, hearing audibilia, etc.). Such expressions are trivial not because of the absence of an act/object distinction, but rather because of the lack of any substantive information about the object: all we are told about the object of such attitudes is … that they are the objects of such attitudes. Thus, the triviality at stake here is akin to that found in “I see what I see”, “I know what I know”: it arises from the violation of the Gricean maxim of quantity, not from there being a distinction without a difference. Cognitive accusatives therefore provide no reason to reject the view that, in “suffering a pain”, the pain is distinct from the suffering.4

A distinction with four differences If cognate accusatives provide no argument against the pain/suffering distinction, can we find arguments in favour of the distinction? And then how should we account for the relations between them? A first proposal is that pain is a kind of suffering, namely physical suffering.5 A second proposal is that suffering is a kind of pain, namely, mental pain. A third proposal is that pain and suffering are both of the same general kind (e.g. negative affect; unpleasant sensation—see Figure 4.1). What these three proposals have in common is the idea that pain and suffering stand in some sort of subsumptive relation (genus/species, determinable/determinate…) to each other. In this sense, pain and suffering are held to be of the same kind.

Figure 4.1 Pain and suffering: three subsumptive views


Olivier Massin

I believe, on the contrary, that pain and suffering do not stand in any subsumptive relation: one is not subsumed under the other, and they are not both subsumed under the same genus. This is because suffering is an attitude while pain is a non-intentional episode. The first essential difference between pain and suffering is then this: Intentionality Pains are non-intentional episodes, whereas suffering is intentional: it is an attitude directed towards something distinct from itself. Which kind of attitude is suffering? Suffering has a negative hedonic valence, which suggests that suffering is an emotion. Although this way of construing the pain/suffering distinction is rather unusual within analytic philosophy, Feldman (1997, 2004) has worked out a cognate distinction between two kinds of pleasures—attitudinal pleasures and sensory pleasure—that has been influential. The chief thing that distinguishes sensory from attitudinal pleasures, Feldman claims, is that sensory pleasures are non-intentional sensations or feelings (the feelings we get from a massage, from an orgasm, from a fresh beer), while attitudinal pleasures (“taking pleasure in” or “enjoying”) are intentional: they are attitudes directed towards something distinct from themselves. Some quibbles aside,6 I suggest that we model the pain/suffering distinction on Feldman’s distinction between attitudinal and sensory pleasures—although I shall disagree with Feldman’s account of the relation between attitudinal and sensory affects. Importantly, Feldman’s distinction between attitudinal and sensory pleasures is not a distinction between two species of the same kind. Hence, the fact that we use the word “pleasure” to refer to both kinds of phenomena is somewhat misleading. There is indeed a link between attitudinal and sensory pleasures but the link is not that of having something in common. What Feldman’s calls the “heterogeneity problem”—the problem of spelling out what pleasures have in common in virtue of which they are pleasures—is restricted to sensory pleasures (and, he might have added, to attitudinal pleasures). But sensory and attitudinal pleasures have nothing in common. (More cautiously, since they may fall under the genera temporal beings, phenomena, affects … they do not fall under a same proximate genus.) The question that arises with respect to these two kinds of pleasures is not what they have in common, but how they are related—what Feldman calls the “linkage problem” (to which I shall come back to later). The same applies, I submit, to pain and suffering. They belong to different categories, and the question is not what they have in common (Feldman’s “heterogeneity problem”) but how they are related (Feldman’s “linkage problem”). Many philosophers will object to Intentionality that pains are widely agreed to be experiences or mental states, and that these are intentional. Representationalists, for instance, equate pains with the perception of bodily

Suffering pains


damages; evaluativists equate pains with the perception of some bodily disturbance as bad. I do not wish deny that there is an extended sense of the word “pain” in which it is used to denote mental states. Nor do I wish to deny that pains are systematically experienced, in a way colours or sounds aren’t (the strength of the necessity at stake here is a tricky issue). However I do believe that the view that word “pain” primarily refer to a mental states is a philosopher’s invention. A first argument to this effect is that one finds no intentional constructions with the word “pain” in ordinary language. One cannot “pain” something. While “taking pleasure in something” arguably expresses attitudinal pleasure, it jars to say “taking pain in something”.7 I can think of four possible counterexamples to this claim, none of which are conclusive. First, one may reply that we speak of the “experience of pain”. However the “of” here is not specificatory (as in “a book of economics”) but intentional (as in “a perception of a tree”). An experience of pain, is not an experience of the pain sort, but an experience directed at pain. Second, one may point out that when we say of a person that she is in pain, we mean that she in some specific mental state. I agree that being in pain is mentalistic. But this is not because “pain” here refers to a mental state. This is because “being in” stands for “experiencing”, “suffering”, “being absorbed in”, “enduring” or some other mental attitude one may entertain towards pain. Something similar may hold for “having a pain” which is often used to mean “experiencing a pain” (an extended use which is understandable, given that having a pain typically goes along with experiencing that pain). In this respect, it is quite telling that many representationalist and evaluativist theories of pain take as their explanandum not “pains” but “pain experiences” or “having a pain”. By doing so, I suspect, they change the explanandum from pains to experiences of pain—and should therefore consistently equate pains with the bodily damages, or bad bodily damages, such experiences are allegedly about. Third, one may stress that although “taking pain in something” is indeed incorrect, “taking great pains to do something” is correct. And this, one may then press, is clearly an intentional pain-construction. But note first that in many translations of the construction “taking pain to”, the word “pain” disappears (e.g. “se donner du mal/de la peine” in French, “sich bemühen/sich Mühe geben” in German). Second and more importantly, “taking pains to do something” does not mean that pains are mental states directed at the action, but that the action is accomplished with great care. It is not that pain are directed at the action, it is rather the action is effortful. Finally, one may press that common constructions such as “her marriage pained me”, or “the pain of losing a child” are clearly mentalistic. But these are presumably extensions of the primary meaning of “pain”, such as when we say that we are have an itch to do something, that we are burning with curiosity, that an idea tickles us, that we are touched by a compliment, that our conscience pricks us, that we feel a pang of guilt, or that remorse stings.


Olivier Massin

Brentano (1995: 84) claims that the primary sense of “pain” is mental, and that it is only by equivocation that we come to call “pain” the bodily state corresponding to it—an equivocation akin to one at play when we say that our nose is our shame, or that the loss of a friend is a great sorrow. I maintain that exactly the reverse is true. Brentano is right that in some cases (shame, sorrow), primarily mental locutions come to apply, by extension, to physical objects. But he fails to notice that in other cases, we come to apply to mental episodes terms that originally apply to bodily states. This is precisely often the case with bodily sensations, as just suggested: “tickles”, “itches”, “prangs”, “burns”, “stings”, “pricks” … and “pains,” primarily refer to bodily states but can, by extension, be used in a mentalistic fashion. Ordinary language therefore clashes with the view that pains are intentional mental states. The second argument in favour of the view that pains are not mental states directly stems from the second essential difference between pain and suffering, which, I submit, is this: Location Pains have a bodily location, whereas suffering has no location. We experience, think and speak about pains as being located in our body, even if sometimes diffusely. Pains are not the only negative sensations than have bodily location: itches, hunger pangs, nauseas, general bodily feeling such as fatigue or being cold are also felt as located in our body. But suffering is not one of these. We do not experience suffering, nor do we think and speak about suffering as being located. Like enjoying and other attitudes, suffering is not given as located.8 While common sense, ordinary language and experience unhesitatingly ascribe bodily location to pains, philosophers have been reluctant to take the idea that pains have a bodily location at face value. Two main worries have been raised. The first directly follows from the philosopher’s view that pains are mental: P1 Pains are mental phenomena. P2 Mental phenomena lack bodily location (from the standpoint of descriptive psychology). C Pains lack bodily location. I have already suggested that P1 is false on linguistic grounds. Assuming the truth of P2, I consider this argument to be in effect a reductio of P1. If a philosopher’s view entails that pains lack bodily location, then that view is probably false. Faced with a choice between pains being mental or pains being located, we should unhesitatingly embrace the later. The second objection commonly raised against the location of pains stems from a failure of transitivity: from the fact that I have a pain in my finger, and that my finger is in my mouth, it does not follow that I have a pain in my

Suffering pains


mouth. Tye draws the conclusion that ordinary talk about pain as bodily located, taken literally, is false: In saying that the pain is in a leg, we speak as if the pain itself is inside the leg, when in reality it is a representation that represents something else inside (or on the surface of) the leg. (Tye 2017: 481) While this sounds like a dismissal of our ordinary way of speaking, Tye suggests that it is not, for talk about pains as located is not meant to be taken literally: Such talk is common. A drowning feeling is not itself drowning. Rather it is a feeling that represents to its subject that he or she is drowning. […] Correspondingly, a pain in a leg is a feeling that is in a leg only in the sense that it represents something (to wit, a disturbance) in a leg. (Tye 2017: 481) The expression “the pain in my foot”, Tye suggests, is a mere façon de parler that is not intended to be taken literally. I disagree. While there is no doubt that “a drowning feeling” is a façon de parler, when we speak of the pain in our foot, we do mean it literally: we mean that the pain is really there in our foot—and that it is really a pain. This corresponds to our experience: we also feel the pain as being located in our foot, and we can feel a pain move up our thigh. From both the standpoint of ordinary language and descriptive psychology, pains are really located in our bodies. On the other hand, considerations about what pains represent—if pains represent anything—are utterly absent from our naïve picture. Thus Tye’s proposal that pain is not bodily located is nolens volens strongly revisionary.9 What then should we say about the intransitivity of pain’s location? That is, we need to establish how we should handle the following inconsistent triad: i Pain is located in body parts. ii Being located in is transitive. iii If pain is located in a body part, and if that body part is located in some container, the pain is not necessarily located in the container. The correct answer is not to reject (i), pace Tye, but (ii): being located in is not always transitive (Noordhof 2001, 2005). True, from the fact that Paul has a pain in his finger and that his finger is in his mouth, it does not follow that he has a pain in his mouth. But from the fact that Paul has a tumour in his finger and that his finger is in his mouth, it does not follow either that he has a tumour in his mouth. Yet his tumour clearly has a bodily location.10 How is it that being located in is not transitive in such cases? The sense in which pains, tumours, but also holes, pressures, vibrations etc. are located is


Olivier Massin

peculiar in two respects. First, these phenomena are located not in empty space but in other bodies. Second, for these phenomena, being located in another body is not just being located at the place were a part of this body is located. To be located, for pain, holes or vibrations, is also to modify or affect the body in which they are located. Following Hyman (2003) and Noordhof (2005), pains belong in the category of modes or states: they are not merely located at the place occupied by body parts; they modify those parts. Pain location, I submit, must then satisfy two conditions: Pain location A pain p is located in a body part if: (i) p occupies a place within that body part; (ii) p modifies that body part at that place (i.e. the body part is hurting as a result). The reason why pain location is non-transitive lies in (ii): the location of pain, holes, tumours, vibrations, pressures or flaws (unlike the location of bodies, persons or cities) is a kind of “modifying location”. There appears to be no good reason to give up the common sense view that pains are located in body parts. A more exotic way of challenging the location criterion of distinction between pain and suffering, is to grant that pain is located, but to argue that suffering also is. One may think of two arguments to that effect. First, materialists may retort that suffering is located in our brain. But recall that the view defended here is meant to be descriptive: in our natural understanding, enjoying, believing, seeing, remembering, suffering … are mental episodes that have temporal location, but lack spatial location. How far is your seeing of a dog from your pride at having won a game? In which direction? Such questions make no sense. Besides, even if it were granted that suffering is located in our brain, it would remain true that pains, by contrast to suffering, can be located in other parts of our body. Finally, one may try to defend the location of suffering by relying on expressions such as “suffering in the flesh” and “suffers in her body”. I rejected Tye’s suggestion that “I have a pain in my foot” was a mere façon de parler. Yet it seems clear that “suffering in the flesh” and the like are mere façons de parler. The “in” here does not indicate the place of the suffering but rather the place of what we suffer from. The third distinction between pain and suffering pertains to expressibility. The proposal is that while pains, being sensory, cannot be expressed, suffering can. Expressing one’s state is not the same thing as reporting it. One can indeed report that one has a pain in the toe. But one cannot express that pain. Aren’t the grimace on our face or the exclamation “ouch!” expressions of that pain? No: these express our suffering from that pain. To see the plausibility of this proposal, consider a tinnitus. One can report having tinnitus, but one cannot express one’s tinnitus. One can however express one’s annoyance at it. Or consider a slight sensation of pressure on one’s neck. One can report

Suffering pains


having such a sensation, but one cannot express pressure sensations. One can however express one’s delighting in it. Likewise for pain: pain being a nonintentional bodily episode, like tinnitus and bodily pressures, one cannot express one’s pain, one can only express one’s suffering it. Expression We cannot express our pain, but we can express our suffering. This third difference between pain and suffering helps explain the fourth and last difference: one cannot have compassion or sympathy for the pain of others—that is, one cannot have pain with them—but one can have compassion for their suffering—that is, one can suffer with them. When someone has an intense pain in the foot, one cannot have a pain with him, but one can suffer with him: it is with the suffering arising from his pain that we sympathise, not with his pain. Likewise, Julie cannot sympathise with Paul’s tinnitus, but she can sympathise with the annoyance that the tinnitus causes to Paul. This is not to say that we can have no representation of the pain of others. Rather, the best one can do here is it to imagine oneself having a similar pain, to put oneself is somebody else’s shoes. Scheler calls “feeling-after” (Nachfühlen) the imagination of the pain of others, where we affectively reproduce a similar pain in ourselves, and contrast it with the “feeling-with” (Mitgefühl) such as compassion, which presupposes that we access the very mental state of the other person.11 The fourth distinction between pain and suffering is then this: Compassion One can have compassion for the suffering of others, but not for the pain of others. The idea traces back to Adam Smith, who endorses a weaker version of the view to the effect that sympathy for pains must be of a lesser degree than sympathy for suffering: […] this is the case of all the passions which take their origin from the body: they excite either no sympathy at all, or such a degree of it, as is altogether disproportioned to the violence of what is felt by the sufferer. It is quite otherwise with those passions which take their origin from the imagination. The frame of my body can be but little affected by the alterations which are brought about upon that of my companion: but my imagination is more ductile, and more readily assumes, if I may say so, the shape and configuration of the imaginations of those with whom I am familiar. A disappointment in love, or ambition, will, upon this account, call forth more sympathy than the greatest bodily evil. Those passions arise altogether from the imagination. The person who has lost his whole fortune, if he is in health, feels nothing in his body. What he suffers is from the imagination only, which represents to him the loss of his dignity,


Olivier Massin neglect from his friends, contempt from his enemies, dependance, want, and misery, coming fast upon him; and we sympathize with him more strongly upon this account, because our imaginations can more readily mould themselves upon his imagination, than our bodies can mould themselves upon his body. (Smith 2002: 25)

Scheler endorses the stronger view that compassion for the pain of others is impossible: Two parents stand beside the dead body of a beloved child. They feel in common the “same” sorrow, the “same” anguish. It is not that A feels this sorrow and B feels it also, and moreover that they both know they are feeling it. No, it is a feeling-with-each-other [Mit-ein-ander-fühlen]. […] It will be evident that we can only feel mental suffering [Leiden] in this fashion, not physical pain [Schmerz] or sensory feelings. There is no such thing as a “pain-with” [Mitschmerz]. Sensory types of feelings (feeling-sensations [Gefühlsempfindungen] as Stumpf calls them), are by nature not susceptible of this highest form of feelings-with [Mitgefühls]. They must somehow become “objectual”. They arise only compassion [Mitleid] “with” and “upon” the suffering of pain by the other person. Likewise, there is certainly such a thing as rejoicing-with [Mitfreude] and upon another’s sensory pleasure, but never sensory-pleasure-with [Mitlust]. (Scheler 2008: 12–13)12 Why is it, as Smith puts it, that “our imaginations can more readily mould themselves upon his imagination, than our bodies can mould themselves upon his body”? How is it that we can have compassion for the suffering of others, but not for their pain? One natural explanation is this. Since pain cannot be expressed, one cannot perceive the pain of others—no more than one can hear their tinnitus. By contrast, since suffering might be expressed, typically through grimacing, one can perceive their suffering. Now to sympathise with someone’s emotions, one must arguably have direct awareness, typically perceptual, of these emotions. If this is true, only expressible states can become objects of sympathy. Under this plausible assumption, the third criterion of distinction between pain and suffering, Expression, entails the fourth one, Compassion. The four distinctions between pain and suffering are recapped in Table 4.1.13 Table 4.1 Four differences between pain and suffering

Pain Suffering





✓ ✕

✕ ✓

✕ ✓

✕ ✓

Suffering pains


Pains are not suffered bodily sensations I have rejected two views about the relation between pain and suffering: (i) that they are identical; (ii) that one is a species of the other, or that the two are species of a common kind. Those tempted to downplay the pain/suffering distinction might still argue, however, that although pain and suffering indeed belong to distinct kinds, one can be defined via or reduced to the other. I now turn to rejecting proposals of this sort. Pain and suffering are not only different in kind: neither figure in the nature of the other. The first account I want to rebut is this: Suffering account of pain Pains are suffered bodily episodes. This account of pain belongs to the same family as accounts that equate pain with an averted or disliked sensation. One chief proponent of a suffering account of pain is Feldman (1997, 2004). Feldman argues that sensory pleasures are alike in virtue of being the objects of attitudinal pleasures, and that sensory pains are alike in virtue of being the object of attitudinal pains (which I prefer to call suffering). This allows him to solve both the heterogeneity and linkage problems: what unites the heterogeneous sensory pains is their relation to attitudinal pain. For x to be a sensory pain is (simplifying a bit) for x to be the object of some attitudinal pain. Here are six problems for the suffering account of pain. 1.


Hunger. Bodily sensations other than pains can be suffered. One can suffer from an itch, a discomfort, the sensations of cold associated with the flu, hunger, etc. As noted above, not all unpleasant bodily sensations are pains, although many if not all may give rise to suffering. One way out for the suffering account of pain would be to isolate which kinds of bodily sensation needs to be suffered in order to count as a pain. But then why not call “pains” those very sensations, prior to their being suffered (which would also avoid the other problems below)? And why endorse the suffering account in the first place if we are able to solve the heterogeneity problem—the question of what all pain sensations have in common—before having introduced suffering? Asymbolia. Some subjects, though they claim to have pains, also claim not to be bothered by them. Taking these subject’s reports at face value, one is led to the conclusion that pains do not essentially displease, are not essentially disliked or suffered. Faced with such cases, upholders of the suffering account of pain have to dismiss the reports of the subjects who claim that they feel pain but do not suffer it.14 It would however be better not to dismiss as false such reports. A more common example comes from the enjoyment of spicy food. Up to a threshold above which pains becomes unbearable, chili likers tend to enjoy the burning pain caused by chili (Rozin and Schiller, 1980).


Olivier Massin


Euthyphro. Do we suffer pains because they feel painful/bad, or do pains feel painful/bad because we suffer them? The suffering account is arguably committed to the latter. Indeed, it would be quite odd to claim that suffering a sensation turns it into a pain and to deny that suffering a sensation makes it seem painful/bad. However, the view that pains seem painful or bad because we suffer them puts the cart before the horse. The natural order of explanation is that we suffer pains because they seem painful/bad.15 Misidentification. The suffering account of pain misidentifies what we suffer (from). When we walk on a small stone, what we suffer from is the ensuing pain and not the non-pain pressure sensation which, once suffered, becomes a pain. Pain is already there when we start suffering it. We do not introduce pain into the world by suffering initially neutral sensations. Justification. Like other emotions, suffering can, on the face of it, be justified: one may wonder why—for what reason—a pain is (or should be) something we suffered from. Intuitively, the answer to that question, will have to be either (i) because pain is painful or (ii) because pain is bad. But neither of these answers are open to the suffering account of pain, since the account is committed to pain being painful or bad because we suffer it. On the suffering account of pain, suffering alights somewhat randomly on bodily sensations, without reason. Location. The suffering account of pain is not compatible with the view that pains are entirely located in the body. On this view, it is true indeed that the bodily sensation constitutive of pain is wholly located in our body. However that sensation, qua suffered, is not entirely located there. The reason is that the suffering element essential to pain is not itself given as located in the body. So an essential part of any pain lacks bodily location. What is strictly speaking located in the body is just the neutral sensation: not the pain, but its indolent substrate.




The suffering account of pain should therefore be rejected: pains cannot be defined through suffering, or at least not in this way.

Suffering is not perceiving pain’s badness Instead of defining pain in terms of suffering, one might attempt to define suffering in terms of pain. Suffering, we have seen, is an emotion which, like other emotions, is intentional, can be expressed, with which one can sympathise, and which is liable of justification. One influential account of emotions equates them with perceptions of value. Let us call the “perceptual approach to suffering” the following view: Perceptual account of suffering To suffer is to perceive/feel the badness of a pain. Here are six problems for this account:

Suffering pains 1.







Loss. That we can suffer from other things that pains (hunger, but also loss, disease, poverty…) shows, as we saw, that the suffering account of pain fails to give a sufficient condition for having pains. It also shows that the perceptualist account of suffering fails to give necessary conditions for suffering: it is not necessary to perceive the badness of a pain to suffer, one might as well perceive the badness of a loss. Valence. Suffering, like most emotions, has a hedonic valence: suffering is essentially unpleasant. But feeling or perceiving x are not essentially valenced episodes. The perceptualist’s idea, however, is that to perceive x as bad must be unpleasant. That is, the presentation of disvalue necessarily turns the presentation itself into something negative. But this idea is far from being obvious. Perceiving a ladybird as red does not turn the perceiving into a red state; thinking of an event as past does not turn the thinking into a past episode. Why should perceiving a bodily episode as bad turn the perceiving into a bad or unpleasant episode? In other words, the perceptual account of suffering faces the problem of explaining how the attitude inherits or gets coloured by the disvalue represented in its content. Absent such an account, we should stick to the view that the perception of the badness of a pain is, like other perceptions, neither positive nor negative. Messenger shooting. If suffering just informs us about the badness of our pain, are we irrational each time we want to be relieved of our suffering? Shouldn’t we want, on the contrary, to face the truth and continue to be informed about the bad episodes occurring in our body? Why shoot the messenger?16 Torturing. Here is another way to press this point. Suppose that to torture a person is to intentionally make her suffer. Why would the torturer want to inform his victim about the badness of her pain? And if torturing is just informing or making apparent to the victim the badness of her bodily episodes, why is it morally problematic? Is it impossible to torture an omniscient being? Justification. Another problem for the perceptual account of suffering stems again from the fact that suffering is capable of justification. We may wonder for what reason one suffers a sensation or external episode, to which the answer will typically be: because it is or feels bad in some way. But while “Why—for what reason—do you suffer?” makes sense, the question “Why— for what reason—do you experience your pain as bad?” does not. Perceptions, experiences, feelings, presentations are not the kinds of states which are liable of justification, by contrast to suffering, emotions or beliefs.17 Reaction. We say both that we suffer pain, and that we suffer from pain. The former is intentional and raises no problem for the perceptual account of suffering. The later however suggests that suffering is a reaction to pain. But perceiving or feeling are not reactions: we can suffer from pain, but we cannot perceive or feel from pain. To suffer is not to perceive pain’s badness.


Olivier Massin

Suffering is the correct reaction to pain I have argued that pain and suffering are distinct, that neither is subsumed under the other or under a common kind, that a pain is not a suffered bodily sensation, and that suffering is not experiencing the badness of pains. What then are pain and suffering and how are they related? My answer to the first question is dull: pain is pain, suffering is suffering. From the standpoint of descriptive psychology, pain and suffering cannot be reduced, analysed of defined in terms of each other, nor, I suspect, in terms of other phenomena. They have to be accepted as primitive. This does not mean that nothing can be said about them. First, we can specify the category to which they belong (attitudes, emotions, for suffering, bodily episodes, for pains) and some distinctive features they have in virtue of their primitive nature (e.g. suffering is expressible). Second, we might describe their relations. What then are the relations between pain and suffering—identity, subsumption, and essential participation having been excluded? There are three main relations between pain and suffering, I submit. The first is normative: suffering a pain is the correct emotion to entertain towards it: Correctness Suffering is the correct emotional reaction to pain.18 Thus the relation between suffering and pain is akin to the relation between fear and danger, indignation and injustice, guilt and misdeeds, etc.: the former are the fitting/correct/appropriate affective answer to the latter. For the sake of this paper, I will take Correctness to be equivalent to: (i) suffering is the appropriate reaction to pain; (ii) suffering is the fitting attitude towards pain; (iii) pain is worthy of being suffered; (iv) if one is to affectively react to pain, then pain ought to be suffered; (v) there are normative reasons to suffer pain. Correctness is a normative relation which I won’t attempt to specify further, apart from the two following points. First, correctness is a non-moral relation (Ewing 1947; d’Arms and Jacobson 2000; Mulligan 1998; Rabinowicz and Rønnow-Rasmussen 2004). To say that suffering is the correct, appropriate or fitting reaction to pain is not to say that suffering pain is morally good, praiseworthy or that we have some ethical duty to suffer pain. We can imagine baroque cases where our ethical duty would be to adopt an incorrect reaction to pain, such as to enjoy it. Second, to say that suffering is the correct emotional reaction to pain, is not to say it is required to suffer in reaction to pain. Another permitted option is to fail to react affectively to pain. Here is a short consideration in favour of Correctness. Pain is widely thought to be bad in some way (typically to be bad finally, personally and pro tanto; see Massin 2017). It is also typically agreed that bad things call for negative affective reactions. One is then naturally led to wonder what the correct reaction to pain is. It is hard to think of a better candidate than

Suffering pains



suffering. Of course, pain may also, and appropriately so, annoy us, make us despair, angry, distressed, fearful, depressed, etc. But such reactions are arguably not just reactions to the pain, but to specific aspects of the pain or to elements associated with it. We might despair at the chronicity of a pain, or be annoyed by its recurrence. A pain can make us angry because it was avoidable, etc. By analogy, that dangers may appropriately depress us, annoy us or tire us, constitutes no objection to the view that fear is the fitting affective reaction to danger. The second relation between pain and suffering is psychological: we normally suffer pain because it feels bad (and not the reverse, as mentioned above against the suffering account of pain). That is, our suffering from a pain is an answer to its seeming bad:20 Reaction Suffering from a pain is an emotion that arises in reaction to the felt badness of that pain. This echoes the recurring claim that emotions are answers, responses or reactions to (the objects of) evaluations, a claim which entails that emotions are not themselves evaluations (pace the perceptualist account of emotions). What is it to affectively react to pain being presented as bad? One possibility is that “x is a reaction to y” simply means “x is caused by y”: the experience of pain’s badness would cause us to suffer. While I think it is true that suffering is caused by the feeling of pain’s badness,21 it is certainly not the case that any emotions caused by such a feeling should count as a reaction to it. There is more to reacting than simply being caused. What? To react to the felt badness of pain is also to adopt the ensuing attitude for this very reason. When Marie suffers her pain in the toe, she does so for the reason that her pain is bad, as she feels it. The felt badness is both a cause of and a motivating reason for the suffering reaction. One question that naturally arises at this point is this: is pain bad because it is worthy of being suffered, or worthy of being suffered because it is bad (or neither of these). Let us adopt the following labels: Buckpassing Pain is bad because it is appropriate to suffer from pain. Naive realism Suffering from pain is appropriate because pain is bad. Correctness is compatible with both Buckpassing and Naive realism. However, Reaction strongly suggest Naive realism. For if the felt badness of pains is the reason for which we normally suffer pains (Reaction), and if the reason for which one should suffer pain never involves pain’s badness (by Buckpassing), we are forced to conclude that we are always wrong when we suffer pains on account of their felt badness. Unless we want to divorce motivating reasons to


Olivier Massin

suffer from normative reasons to suffer, we should endorse Naive realism. This is, I submit, the third relation between pain and suffering: Naive realism Suffering from pains is correct because pains are bad. The badness of a pain is not only the reason for which one normally suffers it when one feels it—a motivating reason–, it is also the reason for which one should suffer the pain—a normative reason. Let us call the conjunction of Correctness, Reaction and Naive realism the Reactive account of the pain–suffering relation. Reactive account of the pain–suffering relation (i) Correctness: suffering is the correct emotional reaction to pain. (ii) Reaction: suffering from a pain is an emotion that arises in reaction to the felt badness of that pain. (iii) Naive realism: suffering from pains is correct because pains are bad. The reactive account avoids all the problems raised against the suffering account of pain: 1.

2. 3.

4. 5.


Hunger. That bodily sensations other than pains can be suffered is not a problem for the reactive account, since it neither claims nor entails that all suffered bodily sensations are pains. Asymbolia. Since on the reactive account it is possible to have pain without suffering it, asymbolics’ reports can be taken at face value. Euthyphro: On the reactive account, pain is not unpleasant or bad in virtue of being suffered, but pain is typically suffered because it is unpleasant or bad, in line with the natural order of explanation. Misidentification: On the reactive account, it is not the neutral bodily substrate of pain which is suffered, but pain itself. Justification: While the suffering account of pain cannot say that we suffer from pain for the reason that it is bad or painful, this is explicitly what the reactive account claims. Location: In the reactive account, suffering is not part of the nature of pains, hence pains can be wholly located in the body.

The reactive account also avoids the pitfalls of the perceptual account of suffering: 1.


Loss. Since the reactive account does not define suffering in terms of the perception of pain’s badness, things other than pain can be suffered from—a loss, a disease… Valence. The reactive account is not committed to the controversial view that the perception of something negative is itself negative. The negativity of suffering is not derived from the negativity of the pain suffered.

Suffering pains 3.





Messenger Shooting. On the reactive account, suffering is not the perception the pain’s badness but arises in reaction to the perception of pain’s badness. Hence wanting to be relieved from our suffering is not refusing to face the facts. Torturing. For the same reason, the reactive account does not equate torturers with informers. (One may indeed inform somebody about something bad in order to make him suffer, but informing is here a means to cause suffering, it is not itself the suffering.) Justification. Since on the reactive account, suffering is an emotion and is not reduced to a perception, the question “Why—for what reason—do you suffer?” is legitimate (contrary to the question “Why—for what reason—do you experience your pain as bad?”), and finds a natural answer: because pain is painful/bad. Reaction. The reactive account straightforwardly accounts for the fact that we both suffer pain, since suffering is intentional, and that we suffer from pain, since suffering is equated with an intentional reaction.

Objections answered In spite of these advantages over the suffering account of pain and the perceptual account of suffering, the reactive account faces three objections that I now want to address. Objection 1: It is not appropriate to suffer slight pains An objection to Correctness is that is it intuitively unfitting to suffer from little pains. One natural answer is that suffering from slight pain is fitting as long as the suffering is also slight. But the gist of the objection may be precisely that there is no such thing as slight suffering. In the same way that nothing can be mildly sublime, fantastic or radiant, or that one cannot be slightly exhausted or thrilled, one cannot slightly suffer. Granting that “suffering” forbids “slight suffering”, one may however argue that there is a perfectly intelligible broader concept of emotional reaction to pain, which encompass suffering for intense pains, but also being bothered or being annoyed, for less intense ones. Ordinary language’s lexicon may be too coarse-grained, nonetheless the distinction between pain and the affective reactions to it is clearly marked in its syntax and semantics. Objection 2: We suffer from pains because they are painful, not because they are bad One may object to Reaction that it is not the badness of pain that one reacts to when one suffers it, but to its unpleasantness or painfulness.


Olivier Massin

My answer is that the distinction between pain’s badness and pain’s unpleasantness or painfulness is a spurious one. Painfulness is a thick axiological property. I can’t elaborate on this view here (see Massin 2017 for a full defence), but here is a short argument. Pain feels painful; pain also feels bad. But pain does not feel bad on top of feeling painful. Once we experience the painfulness of a pain, we do not need to experience a further property to be presented with its badness. Unless the experience of pain is systematically delusive, painfulness and pain’s badness are one and the same property. Objection 3: There is no difference between feeling a pain as bad and suffering it A second objection to Reaction is that the distinction between feeling x as bad and suffering x is without a difference. What is it to feel a pain’s badness, if not to suffer it? The claim that value feelings are distinct from affective reaction to them is unorthodox, but not unprecedented.22 Here are two considerations in its favour, one general, one more specific to pain. First, it is quite difficult to offer reasons for emotions without appealing, explicitly or implicitly, to the values of the objects of these emotions. We are amused by a joke for the reason that it was funny, angry about somebody because his remark was offensive, enjoy a wine because it is balanced, fear a lion because it is dangerous, or like a vase because it is delicate. These properties are admittedly all thick values. Now for these to count as motivating reasons for our emotions, they must (i) be accessed by us,23 and (ii) be distinct from the emotion they motivate. Hence there must be some mental episodes prior to emotions that present us with the value of the object of our emotions. Furthermore, (iii) recalcitrant emotions suggest that such episodes are often non-doxastic. Hence there are some non-doxastic mental episodes with evaluative content that are distinct from and prior to emotions and that provide the motivating reasons for these emotions. These are the episodes which I stipulatively call “feelings of value”. Some sceptics about feelings of value have tried to satisfy these three requisites by denying that motivating reasons or justifiers for emotions are evaluative, and by claiming instead that emotions are justified by the natural properties grounding values (Deonna and Teroni 2012a: 94–97; 2012b). Paul’s reason to fear the lion would not involve its dangerousness (an evaluative property) but only its having sharp teeth (a non-evaluative property). However, this proposal heavily relies on a conversational implicature between having sharp teeth and being dangerous. 24 Once the implicature is cancelled, the proposal loses any intuitive appeal. Consider: JULES:

For what reason do you fear that lion?

PAUL: Because it has very sharp teeth! JULES: And these make him dangerous? PAUL:

No, not in the least. In fact, the lion seems totally harmless to me.

Suffering pains


Second, there seem to be cases where we feel the value of something without affectively reacting to it. One can thus feel the funniness of a joke without being amused by it (Mulligan 2008). After several hearings, the pleasure we take in listening to a piece of music decreases, although the aesthetic value we hear in it does not disappear. Sceptics about value feelings typically reply this way: in such cases one believes or judges that the joke is funny, while failing to really feel or perceive its funniness (which, on perceptualist accounts, is an emotion). More generally, where defenders of value feelings claim that one can grasp the value of a thing while remaining unmoved—that one can be value-blasé—their enemies reply that one then simply fails to see its value—that one is value-blind. This is where pain asymbolia provides a neglected argument in favour of value feelings. Asymbolics report feeling pain after a pinprick, but also report that this does not bother them in the least—and behave accordingly. The standard take on pain asymbolia is that pain is a sensation with two components, one sensory, the other affective, and that asymbolic pains lack the affective component. There are two problems with the standard account. First, it amounts to dismissing asymbolics’ reports: when asymbolics claim to have a pain in the finger, what they should in effect say, on the standard account, is that they have a sensation which is like pain in only one respect. Second, the claim that asymbolics’ pains lack any affective component—that they are, as is commonly put, “pains that don’t hurt”—is at odds with the way asymbolics describe their experience. Looking at the (very few) reports available, asymbolics do report not only feeling pain, but also describe these pains as hurting, painful and bad: “Thank you, this was very good, it hurts a lot” (“Danke schön, das war ganz gut, das hat mir so weh getan”; Schilder and Stengel 1928); “I feel it indeed; it hurts a bit, but it doesn’t bother me” (Pötzl and Stengel 1937: 180; see also Grahek 2007: 45); “it was bad, please something else” (“das war schlecht, bitte schön, noch etwas”; Schilder and Stengel 1928). It is hard to make sense of such reports if one maintains that asymbolics’ pains lack the “affective dimension” of pain altogether, failing to distinguish the painfulness/badness of pains from their being disliked/suffered. Once the distinction between the felt badness of pain and its being suffered is recognised however, such reports make plain sense: asymbolics experience their pains as painful/bad, but fail to react to them by suffering them. They are not unable to fully feel pain, but to correctly react to it. Asymbolics’ pains are not pains that don’t hurt. They are full-blown pains, that hurt, are painful and feel bad. They are just pains that asymbolics don’t suffer from.25

Against incorrigibility The reactive account of the relation between pain and suffering sheds new light on pain’s alleged incorrigibility. I take Incorrigibility to be the following view:


Olivier Massin Incorrigibility Seeming to have a pain entails having a pain.

Incorrigibility has been very influential. It is a key assumption of Kripke’s argument against identity-theories of mind. Since pain is exactly as it seems, Kripke maintains, the identity theorist cannot explain away the intuition that the relation between pain and brain states is contingent by pointing out that it is in fact the relation between the sensation of pain and brain states which is contingent (as he can do with heat): Someone can be in the same epistemic situation as he would be if there were heat, even in the absence of heat, simply by feeling the sensation of heat […] No such possibility exists in the case of pain and other mental phenomena. To be in the same epistemic situation that would obtain if one had a pain is to have a pain. (Kripke 1980: 152) Incorrigibility is also a key premiss of the so-called “paradox of pain” (Hill, 2004, 2005). Our ordinary conception of pain would be paradoxical on the following grounds. On the one hand, we naturally think of pains as being located in body parts, and in this respect we seem to be committed to the view that pains are objective bodily conditions. On the other hand, we conceive of pains as being mental, notably because they seem incorrigible. If no mental episodes are located in body parts, the ordinary conception of pain is paradoxical. I believe that Incorrigibility is false and that the pain/suffering distinction explains both its attraction and why it is false. The reason why Incorrigibility is false is relatively straightforward. Referred pains—where a pain is felt in a location than other than the one in which it is actually located—shows that the felt location of pain can be illusory. “But still, there is a pain,” the incorrigibilist replies, “it is just elsewhere.” No, there is possibly no pain at all. Phantom limb pains—where a pain is felt in an amputated limb—show that experiences of pain can be hallucinatory. One cannot have a pain in a limb that does not exist.26 And unlike referred pains, phantom limb pains are not actually located elsewhere in the body. “But still, there is a pain,” the incorrigibilist replies, “it just has no location.” No, pains essentially have locations, this is one of the central differences between pain and suffering, as we saw. Suppose there were pains without location, would there be any natural kind to which located and unlocated pains belong? “How can you claim that a person complaining about an intense pain in her limb in fact has no pain? Doesn’t she know?” When Julie insists, sincerely, that she has an intense pain in her amputated limb, what she says is false—she has no pain in her limb, because she has no limb. But what she expresses—her suffering—is not doubt real. Julie is genuinely suffering from a hallucinatory

Suffering pains


pain. The plausibility of Incorrigibility relies on a conflation between pain and suffering. One may suffer a pain that one does not have, in the same way that one may fear a danger that one does not face. Suppose Julie hallucinates that a tarantula is above her and insists that she is in real danger. She is not infallible about dangers for all that, quite the contrary. But she really is frightened by an hallucinated danger. Likewise for her phantom limb “pains”: she has no pain, but her pain hallucinations prompts genuine suffering. When we say, with an air of paradox, that Julie is in pain although she has no pain, what we mean is that Julie is genuinely suffering in reaction to a hallucinated pain. So Incorrigibility is false. I won’t elaborate on how this affects Kripke’s objection to materialism, but I want to say a few words in conclusion about how this counts in favour of the common sense conception of pain. Such a conception has been claimed to be paradoxical on the grounds that it both maintains that pains are mental and that pains are located in our body. I believe it is simply not the case that the common sense conception of pain equates pains with mental episodes. Pains are neither experiences nor emotions—but there are experiences of pain, and emotional reaction to experienced pains. For common sense, pain is a state of our body, which one experiences and to which one usually and correctly reacts by suffering. In other words, the reactive account is (part of) the common sense account of pain and suffering. This ordinary account is not only perfectly consistent, it is also preferable to several alternative accounts of pains. First, many alternative accounts of pain diagnose an ambiguity in the ordinary term “pain” which they then disambiguate by introducing various technical terms (“pain-experience” versus “pain-quality”; “paine” versus “paino”; “central state” versus “peripheral pain”; see e.g. Tye 2006a, 2006b; Aydede 2009; Hill forthcoming). The present proposal dispenses with any technical jargon by relying instead on the ordinary language distinction between pains, experiences of pain, and sufferings from pain. Second, the reactive and ordinary approach to pain and suffering proves more fine-grained than several revisionary accounts. Thus, while the standard scientific account of pain relies on a twofold distinction between the sensory and the affective components of pains, the ordinary approach relies on a threefold distinction between having, experiencing and suffering from pain. Second, while the standard scientific approach rests content with componential and causal relations between the two components of pain, the ordinary approach introduces normative relations into the picture. Third, most alternative accounts of pain are led to equate pains with experiences or feelings (of bodily damages, of evaluative properties), and as a consequence struggle to account for the location of pains. The ordinary conception of pain, by contrast, does not compromise on pain location.

Notes 1 See however Klein (2015a) and Brady (2018a).


Olivier Massin

2 The terminology varies: Ryle (1990) speaks of “cognate accusatives”; Kenny (1966: 133) speaks of “cognate objects”, “nominalization accusatives” or “internal accusatives”; and Twardowski (1999) speaks of “internal complements”. 3 See Dummett (2014: 103), Hall (1956), Kenny (1966), and Ryle (1990). 4 Essentially the same point is made by Twardowski (1999) with respect to agentive expressions such as “racing a race” or “jumping a jump”, which, he claims, reflect two phenomena: the action and its product. 5 See Carruthers (2004) and Brady (2018a). 6 See Massin (2013) for discussion. 7 While Feldman (2002, 2004: 84) uses the expression “attitudinal pains” to refer to the polar opposite of attitudinal pleasures or enjoyment, I think “suffering” is the right term. Unlike “attitudinal pain” (and “disenjoyment”, which Feldman also uses), “suffering” has the advantage of not being a term of art. Besides, “attitudinal pain” is unfortunate, for pain primarily refers to a kind of bodily sensation (unlike “pleasures”). The view adopted here—“suffering” is the opposite of “enjoying”—is endorsed by Scheler (1973: 27 n.23) and Mulligan (2008). 8 See also Scheler (1973: 333, 413). 9 For a full defence of the view that pains are genuinely located in our body, see Hyman (2003). 10 One might reply that there is a sense in which we do have a tumour in our mouth when we have a finger with a tumour in our mouth. But in that sense, I submit, we also have a pain in our mouth when our painful finger is in it. 11 Compare De Vignemont and Jacob (2012). 12 Translation modified; see also Scheler (1973: 335, 340–341). 13 These do not exhaust the differences between pain and suffering. Scheler (1973: 334 sqq.) notices further differences, which I only mention here en passant: (i) Pain cannot be emotively remembered (re-felt), whereas suffering can. (ii) There is make-believe or imaginative suffering, but there is no make-believe pain. (iii) Attending to suffering typically diminishes it, but attending to pain does not modify pain. (iv) Pain is transient, suffering is endurant (in a quite specific sense; ibid.: 90–92). Scheler also seems to think that (v) Pain and bodily pleasure are not essentially incompatible (a pain and a bodily pleasure can occupy the same bodily place), but suffering and enjoyment essentially are (one cannot enjoy and suffer exactly the same thing, under exactly the same respect, in exactly the same way). 14 Here is Feldman (2004): “what happens in such cases [in which people experience pain but do not mind] is that the individual experiences a sensation that any normal person would find painful—perhaps the individual himself formerly found similar experiences painful, or would find them painful were it not for the drugs or hypnosis or whatever is affecting him. Thus, there is some justification for calling the sensations ‘pains’. […] But because of the unusual circumstances, the person does not take intrinsic attitudinal pain in his feelings. Thus the feelings (as felt by him on that occasion) are not strictly pains. […] we call them ‘pains’, but on the proposed analysis they are not.” 15 A similar problem afflicts desired-based account of pain, see Bain (2012) and Aydede (2014). Brady (2018b) advances a new version of the desire-based account of pain which avoids this Euthyphro worry. On this new version, it is not the averted pain sensations that is painful, but the experience of having a pain sensation that one wants to cease. If I am right that pains are pain sensations—and that pain experiences are experiences about pains—Brady’s new desire view entails pains are not painful (only pain experiences are). 16 This argument was initially raised by Bain (2012) as a potential problem for the evaluativist theory of pain, according to which to have a pain is to have an experience of a bodily disturbance as bad. Jacobson (2013) generalises the worry to any view that analyses pain in terms of a cognitive attitude.

Suffering pains


17 This problem is more generally a problem for any perceptual theory of emotions (see Deonna and Teroni, 2012; Müller, 2017). See however Tappolet (2016: 40) for a possible answer, and Maguire (2018) for doubts about the view that there are reasons for emotions. I cannot address these worries here. 18 Perhaps Correctness should be conditional on pain being felt. Unfelt pains, if they exist, may not call for suffering. I shall here ignore this complication. 19 “Enduring pain” could be an alternative. But what is to endure a pain? My hunch is that enduring a pain is suffering it plus some mental or behavioural element: not to be distracted by it, persisting in one’s effort or activity, not complaining, etc. If this is true, then enduring is not really an alternative to suffering. 20 I am here assuming following Brogaard (2010, 2012) that the logical form of “x feels F” (in which “feel” is neither comparative, nor epistemic) is “Feels[x is F]”. Badness is a property that pains are presented as having. 21 Klein (2015a: 48sqq) also points out that pain normally causes suffering. 22 The distinction has been defended by Scheler (1973), Reinach (1989: 104, 279–312), von Hildebrandt (1969), and Mulligan (2008, 2009, 2010a, 2010b), Müller (2017). See Vendrell Ferran (2008) and Salice (2015) for presentation of the role of distinction among early phenomenologists. 23 I am here assuming with Dancy (2000) that motivating reasons are reasons in the light of which one acts (in the present case: react). 24 See Müller (2017) for a detailed defence of that point. 25 See Klein (2015b) for a partly converging account of pain asymbolia and Bain (forthcoming) for discussion. 26 As compellingly argued by Bain (2007).

Acknowledgements I am very grateful to Kevin Mulligan, Uriah Kriegel, Walter Horn, Michael Brady and Daniel Schulthess for extended discussions and suggestions on this paper. I am also indebted to Fabrice Teroni, Arnaud Dewalque, Robin McKenna, John DeMouy, Larry Tapper and audiences at the conferences Suffering and Phenomenal Consciousness, Glasgow, 2016; Phenomenology of Emotion, Liège, 2017 for their useful comments. Thanks to Riccardo Braglia, CEO and managing director of Helsinn Holding SA and the Fondazione Reginaldus (Lugano) for financial support of the work published here.

References Aydede, M. 2009. Is Feeling Pain the Perception of Something? Journal of Philosophy 56(10): 531–567. Aydede, M. 2014. How to Unify Theories of Sensory Pleasure: An Adverbialist Proposal. Review of Philosophy and Psychology, 5, 119–133. Bain, D. 2007. The Location of Pains. Philosophical Papers 36(2): 171–172. Bain, D. 2012. What Makes Pains Unpleasant? Philosophical Studies, 166(1): 69–89. Bain, D. 2019. Why Take Painkillers? Nous 53(2): 462–490. Brady, M. S. 2018a. Suffering and Virtue. Oxford: Oxford University Press. Brady, M. S. 2018b. Painfulness, Desire, and the Euthyphro Dilemma. American Philosophical Quarterly 55(3): 239–250. Brentano, F. 1995. Psychology from an Empirical Standpoint. London: Routledge.


Olivier Massin

Brogaard, B. 2010. Do “Looks” Reports Reflect the Contents of Perception? Manuscript. Brogaard, B. 2012. What do We Say When We Say How or What We Feel? Philosophers’ Imprint 12. Carruthers, P. 2004. Suffering without Subjectivity. Philosophical Studies 121: 99–125. Cassell, E. J. 1982. The Nature of Suffering and the Goals of Medicine. New England Journal of Medicine 306: 639–645. Cassell, E. J. 1995. Pain and Suffering. In W. Reich (ed.), Encyclopedia of Bioethics, 2nd edn. New York: Macmillan. Dancy, J. 2000. Practical Reality. Oxford: Oxford University Press. d’Arms, J. and Jacobson, D. 2000. The Moralistic Fallacy: On the ‘Appropriateness’ of Emotions. Philosophical and Phenomenological Research 61(1): 65–90. DeGrazia, D. and Rowan, A. 1991. Pain, Suffering, and Anxiety in Animals and Humans. Theoretical Medicine and Bioethics 12(3). Deonna, J. 2012b, From Justified Emotions to Justified Evaluative Judgements. Dialogue 51: 55–77. Deonna, J. and Teroni, F. 2012. The Emotions: A Philosophical Introduction. London: Routledge. De Vignemont, F., and P. Jacob. 2012. What Is It like to Feel Another’s Pain? Philosophy of Science 79: 295–316. Dummett, M. 2014. Origins of Analytical Philosophy. London: A. & C. Black. Ewing, A. 1947. The Definition of Good. London: Macmillan. Feldman, F. 1997. Utilitarianism, Hedonism, and Desert: Essays in Moral Philosophy. Cambridge: Cambridge University Press. Feldman, F. 2002. The Good Life: A Defense of Attitudinal Hedonism. Philosophy and Phenomenological Research 65: 604–628. Feldman, F. 2004. Pleasure and the Good Life: Concerning the Nature, Varieties, and Plausibility of Hedonism. Oxford: Oxford University Press. Fingarette, H. 1977. Punishment and Suffering. Proceedings and Addresses of the American Philosophical Association 50(6): 499–525. Geiger, M. 1913. Phänomenologie des aesthetischen Genusses. Jahrbücher für Phänomenologie, Halle 1913: 567–684. Grahek, N. 2007. Feeling Pain and Being in Pain, 2nd edition. Cambridge, MA: MIT Press. Gray, D. C. 2010. Punishment as Suffering. Vanderbilt Law Review 63: 1619. Hall, E. W. 1956. Ghosts and Categorial Mistakes. Philosophical Studies 7: 1–6. Hall, R. J. 1989. Are Pains Necessarily Unpleasant? Philosophy and Phenomenological Research 49: 643. Hart, H. L. A. 2008. Punishment and Responsibility: Essays in the Philosophy of Law. Oxford: Oxford University Press. Hill, C. S. 2004. Ouch! An Essay on Pain. Advances in Consciousness Research 56: 339–362. Hill, C. S. Forthcoming. Ow! The Paradox of Pain. In M. Aydede (ed.), Pain: New Essays on its Nature and the Methodology of its Study. Cambridge, MA: MIT Press. Husserl, E. 1970. Logical Investigations. London: Routledge. Hyman, J. 2003. Pains and Places. Philosophy 78: 5–24. Jacobson, H. 2013. Killing the Messenger: Representationalism and the Painfulness of Pain. The Philosophical Quarterly 63: 509–519. Johansson, I. 2001. Species and Dimensions of Pleasure. Metaphysica 2: 39–71.

Suffering pains


Kenny, A. 1966. Action, Emotion and Will. London: Routledge and Kegan Paul. Klein, C. 2015a. What the Body Commands: The Imperative Theory of Pain. Cambridge, MA: MIT Press. Klein, C. 2015b. What Pain Asymbolia Really Shows. Mind 124(494): 493–516. Kripke, S. 1980. Naming and Necessity, rev. ed. Cambridge, MA: Harvard University Press. Loeser, J. D. 2000. Pain and Suffering. The Clinical Journal of Pain 16: 2–6. Maguire, B. 2018. There Are No Reasons for Affective Attitudes. Mind 127(507): 779–805. Massin, O. 2013. The Intentionality of Pleasures and Others Feelings, A Brentanian Approach. In D. Fisette and G. Fréchette (eds), Themes from Brentano. Amsterdam: Rodopi, pp. 307–337. Massin, O. 2017. Bad by Nature: An Axiological Theory of Pain. In J. Corns (ed.), Routledge Handbook of Philosophy of Pain. London: Routledge, pp. 321–333. Mayerfeld, J. 1996. The Moral Asymmetry of Happiness and Suffering. Southern Journal of Philosophy 34(3): 317–338. Mayerfeld, J. 1999. Suffering and Moral Responsibility. Oxford: Oxford University Press. Müller, J. M. 2017. How (Not) to Think of Emotions as Evaluative Attitudes: How (Not) to Think of Emotions as Evaluative Attitudes. Dialectica 71(2): 281–308. Mulligan, K. 1998. From Appropriate Emotions to Values. The Monist 81: 161–188. Mulligan, K. 2008. On Being Blinded by Value, Feeling Feelings and their Valence. Manuscript. Mulligan, K. 2009. On Being Struck by Value. In B. Merkel (ed.), Leben mit Gefühlen. Emotionen, Werte und ihre Kritik. Paderborn: Mentis-Verlag, pp. 141–161. Mulligan, K. 2010a. Husserls Herz. In M. Frank and N. Weidtmann (eds), Husserl und die Philosophie des Geistes. Frankfurt: Suhrkamp, pp. 209–238. Mulligan, K. 2010b. Emotions and Values. In P. Goldie (ed.), Oxford Companion to the Philosophy of Emotions. Oxford: Oxford University Press, pp. 475–500. Nilsson, P. 2011. On the Suffering of Compassion. Philosophia 39: 125–144. Noordhof, P. 2001. In Pain. Analysis 61(270): 95–97. Noordhof, P. 2005. In a State of Pain. In M. Aydede (ed.), Pain: New Essays on its Nature and the Methodology of its Study. Cambridge, MA: MIT Press. Rabinowicz, W. and Rønnow-Rasmussen, T. 2004. The Strike of the Demon: On Fitting Pro-attitudes and Value. Ethics 114: 391–423. Reinach, A. 1989. Sämtliche Werke, ed. K. Schuhmann and B. Smith. Munich: Philosophia Verlag. Roberts, R. 2003. Emotions: An Essay in Aid of Moral Psychology. Cambridge: Cambridge University Press. Roberts, R. 2016. Emotions in the Christian Tradition. The Stanford Encyclopedia of Philosophy (Winter 2016 edition), ed. E. N. Zalta. Retrieved from https://plato.sta Rozin, P. and Schiller, D. 1980. The Nature and Acquisition of a Preference for Chili Pepper by Humans. Motivation and Emotion 4(1): 77–101. Ryle, G. 1990. The Concept of Mind. London: Penguin. Salice, A. 2015. Actions, Values, and States of Affairs in Hildebrand and Reinach. Studia Phaenomenologica 15: 259–280. Scheler, M. 1973. Formalism in Ethics and Non-Formal Ethics of Value. Evanston, IL: Northwestern University Press.


Olivier Massin

Scheler, M. 2008, The Nature of Sympathy, trans. P. Heath. New Brunswick, NJ: Transaction Publishers. Schilder, P. and Stengel, E. 1928. Schmerzasymbolie. Zeitschrift für die gesamte Neurologie und Psychiatrie 113(1): 143–158. Smith, A. 2002. The Theory of Moral Sentiments. Cambridge: Cambridge University Press. Talbott, T. 2017. Heaven and Hell in Christian Thought. The Stanford Encyclopedia of Philosophy (Fall 2017 edition), ed. E. N. Zalta. Retrieved from https://plato.sta Tappolet, C., 2016. Emotions, Value, and Agency. Oxford: Oxford University Press. Twardowski, K. 1999. On Actions, Products and Other Topics in Philosophy. Amsterdam: Rodopi. Tye, M. 2006a. Another Look at Representationalism about Pain. In M. Aydede (ed.), Pain: New Essays on Its Nature and the Methodology of Its Study. Cambridge, MA: MIT Press, pp. 99–120. Tye, M. 2006b. In Defense of Representationalism: Reply to Commentaries. In M. Aydede (ed.), Pain: New Essays on Its Nature and the Methodology of Its Study. Cambridge, MA: MIT Press, pp. 163–176. Tye, M. 2008. The Experience of Emotion: an Intentionalist Theory. Revue internationale de Philosophie 62: 25–50. Tye, M. 2017. Are Pains Feelings? The Monist 100: 478–484. Vendrell Ferran, I. 2008. Die Emotionen, Gefühle in der realistischen Phänomenologie, Berlin: Akademie Verlag. von Hildebrand, D. 1969[1922]. Sittlichkeit und ethische Werterkenntnis. In K. Mertens (ed.), Die Idee von der sittlichen Handlung. Sittlichkeit und ethische Werterkenntnis. Darmstadt: Wissenschaftliche Buchgesellschaft, pp. 127–268. Walen, A. 2015. Retributive Justice. The Stanford Encyclopedia of Philosophy (Winter 2016 edition), ed. E. N. Zalta. Retrieved from win2016/entries/justice-retributive. Zaibert, L. 2013. The Instruments of Abolition, or Why Retributivism is the Only Real Justification of Punishment. Law and Philosophy 32: 33–58. Zaibert, L. 2017. On the Matter of Suffering: Derek Parfit and the Possibility of Deserved Punishment. Criminal Law and Philosophy 11(1): 1–18.

Part II

Pain and valence


Valence, bodily (dis)pleasure, and emotion Fabrice Teroni

A fundamental property of bodily (dis)pleasures, emotions, moods and sentiments (love, hate, etc.) is their valence or polarity: they are readily described as being either positive or negative. Let us call all valenced states “affective”. This is not meant to be stipulative – we pre-theoretically think that these states form a family, and this is surely due to the fact that they are valenced. Three presuppositions about valence underlie my discussion. First, that valence is distinctive of affective states and sets them apart from other psychological states such as beliefs, imaginative experiences and memories. We can surely describe particular beliefs, imaginative experiences and memories as positive or negative. The first presupposition is that we do so in virtue of their relations to affective states. As opposed to this, the valence of affective states is independent of the relations they have to other psychological states.1 The second presupposition is that valence is among the phenomenal or experiential properties of affective states. This is to take for granted that, when we pre-theoretically group together bodily (dis)pleasures, emotions, moods and sentiments, we do so on account of the way they manifest themselves in consciousness. The third and final presupposition is that affective states have their valence essentially: they could not be what they are if they had a different valence, or no valence at all. For instance, a bodily displeasure could not be the psychological state it is if it was pleasant or neutral. In a nutshell, I shall presuppose that affective states form a family of psychological states essentially characterized by the positive or negative way they occupy consciousness.2 The fact that psychological states are affective on account of their valence means that issues surrounding valence are key to the understanding of the affective domain. It is useful to divide these issues into two groups.3 On the one hand, we may wonder whether valence is a monadic or relational property, what (if relational) its relata are and whether they include evaluative properties. This first group of issues regards the nature of valence. On the other hand, we may wonder as to how affective states relate to one another, and more specifically whether some of them are more basic than others. This second group of issues regards the structure of the affective domain. Among these structural issues, I want to concentrate on two issues concerning the


Fabrice Teroni

relations between bodily (dis)pleasures and emotions. First, is the valence of bodily (dis)pleasures explanatorily prior vis-à-vis the valence of emotions? Saying that it is endorsing what I shall call the explanatory priority thesis. Second, do emotions contain bodily (dis)pleasures? Saying that they do is endorsing what I shall call the containment thesis. My aim in this chapter is to assess the prospects of these two structural theses in light of what we can reasonably think about the nature of valence. The discussion proceeds as follows. First, I introduce the popular core affect approach to affective experience and explain why it is committed to the explanatory priority and containment theses. Given that this approach gives pride of place to bodily (dis)pleasures, the second section below explores the nature of their valence. I argue that bodily (dis)pleasures are intentional states whose valence is to be understood in terms of evaluative experience, and identify similarities and dissimilarities between their intentional structure and that of emotions. On this backdrop, I then assess the prospects of the explanatory priority and containment theses. I offer reasons to conclude that we should adopt neither of the two theses.

The core affect approach To say that a psychological state is valenced is to characterize its phenomenal or experiential properties. A good starting point for our discussion is therefore to examine how valence falls within a more comprehensive characterization of affective experience. A natural way of approaching affective experience is to emphasize two dimensions of it. First, and unsurprisingly, affective experience is more or less pleasant or unpleasant. Bodily pleasures, as well as episodes of happiness, amusement, admiration and pride are pleasant. Bodily displeasures, as well as episodes of anxiety, shame, despair and fear are unpleasant.4 Second, affective experience is bodily. In bodily (dis)pleasure, obviously, but also in moods and emotions, the body somehow intrudes into consciousness: affective states implicate a variety of bodily changes, and we feel some of these changes.5 Given how natural it is to capture affective experience in terms of (dis) pleasure and bodily feelings, it is not surprising that a popular approach is built around these two dimensions. According to the core affect approach championed by psychologists such as James Russell (2003) and Lisa Feldman Barrett (2006), each affective state occupies a specific area of a two-dimensional valence/arousal space (see also Russell and Barrett 1999). The term “core affect” denotes this two-dimensional nature of affective experience. The first dimension of core affect, valence, corresponds to how pleasant or unpleasant the experience is. The second dimension, arousal, constitutes a specific understanding of the bodily dimension of affective experience: the claim is that it is constituted by how activated or sleepy the body feels.6 As I see it, the first advantage of the core affect approach is that it rests on these two salient dimensions of affective experience.

Valence, bodily (dis)pleasure, and emotion


In order to appreciate fully how this approach pictures affective experience, we should bear in mind that core affect is supposed to have two important properties. The first property of core affect concerns its qualitative nature. One may worry that the approach under discussion pictures affective experience in an unconvincingly fragmented way. When we are anxious at the prospect of a medical exam, proud of a past deed or bored at a family reunion, we have a hard time factoring out how pleasantly or unpleasantly we feel from how excited or sleepy we feel. There is no cause for concern here. According to the advocates of the core affect approach, the distinction between valence and arousal should not encourage the idea that affective experience is an aggregate of one feeling corresponding to arousal and another feeling corresponding to valence. The idea is rather that any point on the two-dimensional space corresponds to a feeling that is “simple at the subjective level”, affective experience being a feeling state whose phenomenal properties are an “integral blend of hedonic (pleasure–displeasure) and arousal (sleepy–activated) values” (Russell 2003: 147). The core affect approach has it, then, that affective experience is a unitary experience blending amounts of arousal and valence.7 This attempt at unifying two salient dimensions of affective experience is another of its assets. The second property of core affect concerns whether it is intentional or not. Core affect is at any moment constituted by an experience that we can plot without remainder on the two-dimensional qualitative space. This has suggested to many that core affect is constituted by “raw feels” (i.e. purely qualitative, non-intentional states). The core affect approach raises many worries with which I shall not be concerned here.8 Given my aim, I shall only focus on how it understands the relation between valence and bodily phenomenology. The reader may have noticed something odd in the idea just presented that valence and bodily arousal blend in a unitary raw feel. After all, one would expect a blend of valence and bodily arousal to result in pleasant or unpleasant bodily feelings. Bodily feelings may lack well-defined spatial boundaries and be more or less diffuse, yet they are about the body. There are three reasons that support the idea that this is the most charitable interpretation of core affect. The first, mainly exegetical reason is that advocates of the approach perceive themselves as heirs of James – and one would fail to see what is Jamesian in an account in which bodily feelings play no role.9 The second reason is that we readily understand “sleepy”, “activated” and “more or less energetic”, when they modify feelings, as referring to diffuse ways the body feels. This is further illustrated by many familiar descriptions: when we say that we feel light, heavy, tight, etc., we refer to characteristic bodily feelings. The third reason is that, if we do not understand arousal in terms of feelings that are about the body, then it is difficult to see which other object these feelings could have. As a result, we would saddle the core affect approach with the claim that there are non-intentional feelings. This claim is controversial, since the existence of non-intentional psychological states is a hotly debated issue. For these reasons, the most convincing understanding of core affect


Fabrice Teroni

is that it corresponds to (un)pleasant bodily feelings that can be plotted on the sleepy-activated axis.10 We are in a position to realize that the core affect approach, a natural and popular way of understanding affective experience, adopts the two structural theses introduced above. First, it claims that the (un)pleasant bodily feelings that constitute core affect explain the valence of all affective states, and of emotions in particular. This is to endorse the explanatory priority thesis. Now, emotions surely have properties besides those of core affect. For instance, they are about distal objects or events – one is afraid of an approaching lion or happy that a friend won a prize. Whatever additional properties affective states may possess, the key claim of the core affect approach is that what makes them affective are (un)pleasant bodily feelings.11 Second, the core affect approach also endorses the containment thesis: emotions contain (un)pleasant bodily feelings – given that emotions are valenced on account of these (un)pleasant bodily feelings, emotions in fact contain these feelings essentially. These are the two theses that I want to discuss in what follows. As will emerge, they are difficult to reconcile with what we know about the nature of bodily (dis)pleasures and emotions. The purpose of the next section is to familiarize ourselves with the relevant ideas.

The intentional structure of bodily (dis)pleasures and emotions We have ended the previous section by observing that the core affect approach endorses the explanatory priority and the containment theses: the valence of (un)pleasant bodily feelings explains the valence of emotions, since emotions are valenced in virtue of containing (un)pleasant bodily feelings. The next question is of course: what are (un)pleasant bodily feelings? The aim of this section is to look into these feelings and their (un)pleasantness. I present some considerations suggesting that (un)pleasant bodily feelings have a complex intentional structure, and that this structure is shared by emotions. On this backdrop, I will assess, in the next section, the prospects of the explanatory priority and the containment theses. My starting point is the fact that bodily (dis)pleasures are more complex than we may pre-theoretically think. In particular, it is important to distinguish the purely proprioceptive from the valenced (i.e. (un)pleasant) aspect of bodily (dis)pleasures. Whenever we are in a state of bodily pleasure or displeasure, we represent proprioceptively a more or less precise and extended area of our body. This spatial representation is of course insufficient to individuate most bodily sensations or feelings. In addition to spatial information, bodily feelings often carry qualitative information: an area of the body may feel swollen, cold, burning, wet, pulsating, etc. For our purposes, an important type of information concerns functional integrity – many philosophers (e.g. Tye 1995; Bain 2019) have defended the claim that pain feelings differ from other sensations in virtue of representing a body part as damaged.

Valence, bodily (dis)pleasure, and emotion 12


Suppose that this is the case. Still, the representation of damage does not account for the unpleasantness of pain. To explain why, let me press into service pain asymbolia, a dissociative condition that is well documented by clinical studies of pain (e.g. Grahek 2007). Asymbolic patients have the capacity to refer to and recognize their pains, yet the latter do not move them at all. When an experimenter pierces their skin with a needle, they tend to smile and say something along the lines of “That’s painful, but I don’t care”. In my opinion, we should understand pain asymbolia as follows: what asymbolic patients refer to are distinctive but not valenced (i.e. neutral) bodily feelings (Grahek 2007, Teroni 2018). Given that asymbolics describe their feelings as painful, one may initially think that this understanding of the condition is not compatible with the evidence. Doesn’t this mean that their feelings hurt – the archetypical valenced state? No. We have just introduced the idea that pain feelings differ from other bodily feelings in virtue of representing damage. This opens the possibility that asymbolic patients use “painful” to describe bodily feelings that represent damage.13 The peculiarity of their condition is that their pains (i.e. feelings that represent parts of their body as damaged) are not negatively valenced or unpleasant. To drive the point home, consider the following analogy: one may represent one’s property (one’s car, say) as being damaged without this feeling unpleasant. One may have had a wonderful day or be too exhausted for this to happen. What is disconcerting about asymbolic patients is that their proprioceptive representations of bodily damage (i.e. their pains) always “leave them cold” in the way representation of damage to our property sometimes leaves us cold. Of course, this is not to suggest that the representation of a body part as damaged is immaterial to an account of pain. The suggestion is rather that the role of this representation is not to explain valence. Its role is to distinguish pain from many other types of sensations that are unpleasant but not pain(ful). Tickles and itches may be unpleasant, but they are not painful for this reason. Among bodily feelings, pain(ful) feelings are those that contain or are constituted by the representation of a body part as damaged. These feelings are typically unpleasant – but if we understand pain asymbolia in the way suggested here, this is not the case for anyone at any time. This means that we still need an explanation of the chief aspect of (un) pleasant bodily feelings, i.e. their valence. The explanation has to choose between two main options: we may try to capture the (un)pleasantness of these feelings by appealing either to motivation or to evaluative representation. Adopting the first option, we end up claiming that what makes a feeling (un) pleasant is the subject’s desire to have it continue (cease) (e.g. Sidgwick 1874; Heathwood 2007). This is not very appealing, since it goes against the intuitively compelling thought that we desire some feelings to continue and others to cease on account of their (un)pleasantness.14 It is thus preferable to adopt the second option, whose distinctive claim is that feelings are (un)pleasant when they combine with evaluative representations (Nelkin 1994; Bain 2019;


Fabrice Teroni

Carruthers 2018).15 A feeling becomes (un)pleasant when the bodily condition it represents is evaluated positively or negatively.16 The idea is intuitively appealing, since we readily refer to evaluation when we describe the relevant psychological states, for instance when we say that an area around our wrist feels good or bad. In addition, appealing to evaluation leaves room for explanatory relations between (un)pleasant feelings and desires: we typically desire a feeling to continue or to cease because of its (un)pleasantness, i.e. because of the positive or negative evaluation of our bodily condition with which the feeling combines. How should we understand these evaluations? It would of course be illadvised to understand them as intellectually demanding deployments of evaluative concepts (e.g. Nelkin 1994) – what makes a sensation (un)pleasant is surely not the fact that we deploy a concept to carve out distinctions within the sensational domain. The idea is rather that the relevant evaluations take place automatically and mostly at the subpersonal level in such a way that, at the personal level, the subject is simply struck by how pleasant or unpleasant the sensation is.17 The phenomenology of (un)pleasant bodily feelings is a matter of this (un)pleasantness somehow imposing itself on us. Having said that, how exactly should we understand this phenomenology of (un)pleasantness? One natural idea is to use the model of perception. Perceptual experience – be it visual, auditory, etc. – possesses three features that make it an attractive model of valence. First, perceptual experience is intentional. The visual experience of a violet chrysanthemum is about this flower, its colour and shape; the auditory experience of a B flat is about a sound of this tonality. Second, there are good reasons to think that the intentionality of perceptual experience does not depend on the deployment of concepts. One can see a violet chrysanthemum without possessing the concepts of violet or chrysanthemum, as one can hear a B flat without possessing the concepts of this pitch and tonality (e.g. Crane 1992). Third and relatedly, the intentionality of perceptual experience is hardly detachable from its phenomenology: to have a visual experience that is about a violet chrysanthemum is closely related to the way this experience fills one’s stream of consciousness.18 The hope is to apply the model of perception to valence, i.e. to claim that what makes (un)pleasant bodily feelings (un)pleasant imposes itself on us in the cognitively undemanding and phenomenologically salient way an object’s colour or a sound’s pitch imposes itself on us when we see or hear it. Of course, this requires that we be explicit about the property that is perceived in (un)pleasant bodily feelings. We know that this property is evaluative and that it is exemplified by the body.19 At this juncture, I suggest we borrow a key idea from David Bain’s account of pain’s unpleasantness (Bain 2019). According to Bain, pain feelings are proprioceptive representations of bodily damage. A pain feeling is unpleasant (i.e. negatively valenced) when the bodily damage is the object of a perceptual experience that represents it as bad for us. Suppose a pain of yours represents a damage in an area around

Valence, bodily (dis)pleasure, and emotion


your wrist. On Bain’s account, your pain is unpleasant if you seem to perceive this damage as bad for you.20 Two aspects of this account are worth emphasizing. The first is that what explains unpleasantness is the perceptual experience of a specific evaluative property: a final, personal and negative value. The perceptual experience of an instrumental, impersonal or positive value would not explain the unpleasantness of pain. This is intuitively convincing, since we think of pain as disvaluable for its own sake (final value), given that it diminishes the quality of the subject’s life for her (personal value). The second aspect of the account is that it is built around two evaluative properties. First, there is the final, personal and negative value we have just met. Given that veridical perceptual experiences attribute this value to bodily damages, its bearers need not be, and in the case at hand are not, psychological states. Second, there is unpleasantness – a value that we can also describe as being final, personal and negative. According to the perceptual account of pain’s unpleasantness under discussion, this second value is closely related to the first: it consists in the perceptual experience of a bodily damage as being of final, personal and negative value.21 Given that unpleasantness is constituted by such an experience, its bearer must be a psychological state – and, I take it, more specifically a conscious psychological state.22 Let me now draw attention to the intentional structure that is, on this account, characteristic of unpleasant pain. Call “cognitive base” the proprioceptive experience that allows the subject to access the relevant part of the body and its being damaged. This cognitive base may exist alone, although it is typically accompanied by a perceptual experience of disvalue that makes the episode unpleasant. The cognitive base grounds this evaluative experience, which elaborates on it. In unpleasant pain, the evaluative experience represents the final, personal and negative value of a bodily damage. This presupposes a representation of damage in the guise of a proprioceptive experience. The proprioceptive experience grounds the evaluative experience, which gives access to an additional, evaluative property of the bodily damage. This intentional structure is hardly exclusive to unpleasant pain, and one would expect this account to apply to other bodily feelings as well.23 Bodily feelings are (spatial-cum-qualitative) proprioceptive experiences that may exist alone. However, some of them typically function as the cognitive bases of evaluative experiences, which elaborate on them. Proprioceptive experiences become unpleasant when they combine with a perceptual experience of the final, personal badness of what happens in the body. Remember that we have adopted the idea that damage representation is distinctive of pain feelings. Other bodily feelings may represent other properties at various places in the body. We all know that a wide range of bodily feelings may constitute the cognitive bases of displeasure. But what makes all of them unpleasant is something they have in common: the fact that they ground an evaluative experience giving access to the final, personal badness of a bodily condition. Correspondingly, what turns proprioceptive experiences into


Fabrice Teroni

pleasant feelings is the fact that they function as the cognitive bases of evaluative experiences giving access to the final, personal goodness of the relevant bodily condition. The fact that the perceptual account of valence applies smoothly to the (un)pleasantness of other bodily feelings is reassuring. We should strive for a unified (or at least not too fragmented) account of valence, and the fact that it offers such a unified account is a point in favour of the core affect approach. It would be disquieting to conclude that the valence of pain feelings is a completely different property from the valence of other bodily feelings. It would be equally disquieting if the perceptual account could not be extended to other affective states, and to emotions in particular. Fortunately, it turns out that the intentional structure of (un)pleasant bodily feelings that we have just disclosed is shared by emotions. Suppose that you are afraid of a lion coming towards you, or that you admire Titian’s Assumption of the Virgin. In order to fear the lion or admire the painting, you must access these objects and some of their properties. For instance, you must see the lion as approaching you, or see the specific way colours are distributed on the panel. Given that these visual experiences play the same role in these emotions as proprioceptive experiences play in (un)pleasant bodily feelings, let us extend the courtesy and describe them as the cognitive bases of these emotions. There is of course a key difference between (un)pleasant bodily feelings and emotions: the cognitive bases of emotions typically are not proprioceptive experiences.24 This is true, but irrelevant given our interest in the intentional structure that is shared by (un)pleasant bodily feelings and emotions. Cognitive bases no more account for the valence of emotions than for the valence of (un)pleasant bodily feelings. In the same way as bodily feelings may leave us cold, we may remain emotionally unaffected by the visual experience of an approaching lion or of Titian’s painting. We still need an explanation of emotional valence, and we face a similar choice between an account in terms of motivation or in terms of evaluative representation. Again, and for parallel reasons, the best move is to understand valence in terms of evaluative representation (Teroni 2018). The similarities between (un)pleasant bodily feelings and emotions are in fact numerous. In particular, the phenomenology of emotions is also a matter of something of positive or negative import imposing itself on us. No more than in the case of (un)pleasant bodily feelings can we understand valence in terms of the deployment of evaluative concepts. We should again insist that the evaluation takes places automatically and mostly at the subpersonal level – what happens at the personal level is that the subject is simply struck by the positive or negative import of what is presented to her by the cognitive base. According to an influential approach, more or less automatic and non-conceptual appraisal processes underscore emotions, which make themselves manifest at the personal level as specific evaluative takes on the environment.25 In the same way as for (un)pleasant bodily feelings, perceptual experience of value accounts for emotional valence.26

Valence, bodily (dis)pleasure, and emotion


Perceptual experience of which value? The answer is a function of the emotion we consider. A recurring theme in emotion research is that different thick values 27 are the formal objects of different emotion types.28 We can understand the idea as follows. Thick values are divided into two main groups: there are negative (fearsome, disgusting, shameful, ugly, etc.) and positive (admirable, beautiful, funny, lovely, virtuous, etc.) thick values. An emotion has negative valence because it contains a perceptual experience of a negative thick value. Fear of the lion has negative valence because it contains a perceptual experience of the lion’s fearsomeness. Other emotions get their negative valence because they represent different negative thick values (the disgusting, the shameful, the appalling, etc.). On the bright side of things, an emotion has positive valence because it contains a perceptual experience of a positive thick value. Admiration of Titian’s painting has positive valence because it contains a perceptual experience of the painting’s admirable character, while other positive emotions get their valence in virtue of representing different positive thick values (the successful, the funny, etc.). It is difficult to tell whether the thick values that account for emotional valence belong to a specific formal family. Are they all personal values? If they are, they need not be personal values for the subject undergoing the emotion – one can be afraid for someone, for instance.29 And the thick values at stake in admiration or guilt may not be personal.30 You may remember that I have emphasized two aspects of the perceptual account of unpleasant pain. It is important to realize that these two aspects carry over to the generalized account of valence under discussion. First, we have seen that the unpleasantness of pain is explained by the perceptual experience of a final, personal and negative value. According to the generalized account, the (un)pleasantness or valence of emotions is explained by the perceptual experience of the relevant thick value. Second, we have observed that the account of pain’s unpleasantness is built around two evaluative properties. The same is true of the generalized account. The first evaluative property is the thick value that the perceptual experience represents – the lion’s fearsomeness or the painting’s admirable character. The bearer of this first evaluative property sometimes is a psychological state (one may be afraid of a recurring dream, say); this is comparatively rare and the bearer often is a distal object or event. The second evaluative property is the emotion’s valence or (un)pleasantness. According to the generalized account, this valence consists in the perceptual experience of the object’s thick value. As in unpleasant pain, the fact that (un)pleasantness is constituted by such an experience means that its bearer must be a conscious psychological state. According to the generalized account of valence, negative valence or unpleasantness is a matter of perceptual experience of disvalue. How does unpleasantness relate to suffering? This is a complex issue, and I shall rest content with some preliminary thoughts. Is unpleasantness necessary for suffering? It seems to me that it is: absent unpleasantness (i.e. negative affect), one cannot suffer. Is unpleasantness sufficient for suffering? It seems to me


Fabrice Teroni

that it is not: there is suffering only when unpleasantness reaches a certain threshold and/or consumes attention. Mild disappointment at a missed opportunity is not a way of suffering. Deep shame at a misdeed is. Intense unpleasant affects are ways of suffering.31 The upshot of this section is this: (un)pleasant bodily feelings and emotions have the same intentional structure. In both cases, a cognitive base grants access to an object. The object is then assessed in a cognitively undemanding way. This assessment makes itself consciously manifest as the experience of a value, which accounts for the valence of the psychological state. The fact that (un)pleasant bodily feelings and emotions have the same intentional structure leaves room for significant differences. The cognitive bases of emotions are not, as opposed to those of (un)pleasant bodily feelings, exclusively proprioceptive. And the thick values at stake in emotions are specific to them – understanding the relation between these values and those at stake in (un)pleasant bodily feelings will turn out to be a key issue in the next section.

Assessing the explanatory and the containment theses Our next task is to bring together the results of the preceding sections and see how they relate to each other. Are the two structural theses about the affective domain – the explanatory priority thesis (the valence of (un)pleasant bodily feelings explains the valence of emotions) and the containment thesis (emotions contain (un)pleasant bodily feelings) – that emerged as key commitments of the core affect approach supported by, or even compatible with the perceptual account of valence that we have adopted? According to this account, the valence of bodily (dis)pleasures and emotions is a matter of experiencing a value of the object represented by a cognitive base. To assess how the perceptual account of valence links with the explanatory priority and containment theses, we have to address the following question: what is the relation between the valence of (un)pleasant bodily feelings and the valence of emotions? In light of the account, this turns out to be a question about the relation between the values that we experience in (un)pleasant bodily feelings – final, personal values – and the thick values that we experience in emotions. To address this question, I proceed as follows. First, I examine whether we can make sense of this relation in a way that is compatible with the explanatory priority thesis. I argue that this is not possible. Second, I examine whether we should accept the containment thesis even though bodily (dis)pleasure do not explain emotional valence. I give some reasons to think that we should not. The first issue is whether we can understand the relation between the values that we experience in (un)pleasant bodily feelings and the values that we experience in emotions in a way that is compatible with the explanatory priority thesis. Let me summarize the little that has surfaced so far regarding this relation. When I presented the perceptual account of pain’s unpleasantness, I followed Bain in characterizing the value experienced simply as being

Valence, bodily (dis)pleasure, and emotion


final, personal and negative. This suggests that the value at stake in unpleasant pain (and, probably, in other (un)pleasant bodily feelings as well) is a thin value – the damage is experienced as being finally and personally bad tout court.32 As opposed to this, the values experienced in negative (the fearsome, the disgusting, the shameful, etc.) and positive (the admirable, the successful, the funny, etc.) emotions are thick values (i.e. more determinate and descriptively richer ways of being bad and good). Let us suppose then that the (un)pleasantness of bodily feelings is a matter of representing a positive or negative thin value. How do these thin values relate to the thick values represented in emotions? The most natural way of understanding this relation is as that of two determinable properties (thin goodness and thin badness) to their respective determinates (the fearsome, the admirable, etc.). What does this mean? One nice illustration of the determinable/determinate relation is the relation between being coloured and, say, a determinate shade of vermilion. It is often claimed that being coloured is a determinable with different determinates in order to draw attention to two aspects of their relation (Johnson 1921). First, that an object is always coloured in virtue of exemplifying a determinate colour, such as a determinate shade of vermillion. Second, that we cannot define the determinate property in terms of the determinable together with a determinate property: the nuance of vermilion cannot be defined in terms of the determinable of being coloured together with a determinate property (the only suitable determinate property is the nuance itself, and this does not constitute a definition). For present purposes, we can concentrate on the first aspect. If (un)pleasant bodily feelings are (un)pleasant in virtue of containing the experience of a positive or negative thin value, it is tempting to understand the relation of these thin values to the thick values experienced in emotions as being that of two determinables to their determinates. There is a lot to be said in favour of the idea that something cannot be of final, personal value tout court – it is always so in virtue of being of a determinate final, personal value (being disgusting, shameful, successful, etc.). Understanding the relation between the values at stake in (un)pleasant bodily feelings and emotions as that of two determinables to their determinates is not very convincing, however. First, this seats uneasily with the explanatory priority thesis. According to this thesis, the valence of (un)pleasant bodily feelings explains the valence of emotions. We are exploring the possibility of understanding the (un)pleasantness of bodily feelings in terms of the experience of thin values, and the (un)pleasantness of emotions in terms of the experience of thick values. But if something is the bearer of a thin value only in virtue of exemplifying a thick value, one would expect this ontological priority to generate a representational priority as well. To represent something as having a thin value would seem to be a derivative of thick value representation. As the case may be, it would come down to represent it either as having a given thick value (to represent the lion as bad is to represent it as fearsome, say), or as having one does not know which thick value


Fabrice Teroni

(to represent the lion as bad is to represent it as either fearsome or disgusting or sad, etc.).We end up with an order of explanation that is the reverse of that advanced by the explanatory priority thesis.33 I am not fully convinced by these considerations, however. A second, more potent reason to refuse to understand the relation at issue on the determinable–determinate model emerges once we remind ourselves that we are interested in a specific kind of representation, i.e. perceptual experience. The considerations just advanced indeed apply especially well to perceptual experience: seeing, hearing or touching a determinable property always comes down to experiencing a determinate property. We always see determinate colours and shapes, for instance. This is why “Anna sees the sofa’s colour” only makes sense in a distributive way: to see an object’s colour is to see either this or that determinate colour it has.34 Not only does this go against the explanatory priority thesis, it demonstrates that advocates of the perceptual account of valence cannot understand the relation between the values experienced in (un)pleasant bodily feelings and emotions as that of a determinable to its determinates. The perceptual account of valence turns out to require that the values represented in (un)pleasant bodily feelings are as thick as the values represented in emotions. We should say that the idea that these feelings represent thin values is an unwanted side effect of our purely formal characterization of the relevant values as final and personal. What we actually experience when we undergo (un)pleasant bodily feelings are thick values exemplified by the body. Thick values, we have seen, are determinate and descriptively rich ways of being good or bad. Now, it may prove difficult to be informative as to the particular way of being finally and personally (dis)valuable which is distinctive of the thick values experienced in (un)pleasant bodily feelings. This shouldn’t cause too much trouble: it is difficult to characterize most values informatively, and there may be primitive thick values.35 Let us thus agree that the values we experience in (un)pleasant bodily feelings are thick values. Where does that leave us regarding the explanatory priority thesis? We are exploring an alternative account of the relation between the values at stake in (un)pleasant bodily feelings and in emotions: it is not anymore the relation of a determinable to its determinates, but the relation of one family of thick values to another. Is this relation compatible with the explanatory priority thesis? Here are some considerations for concluding that it is not. The basic insight of the perceptual account of valence is that value experience has the potential to explain the (un)pleasantness of psychological states. On the face of it, this potential accrues to the experience of any thick value: perception of a thick value accounts for the unpleasantness of pain, perception of fearsomeness accounts for the unpleasantness of fear, perception of funniness accounts for the pleasantness of amusement, etc. There is a lot to be said for this understanding of (un)pleasantness in terms of the experience of different thick values. For instance, it offers an attractive explanation of

Valence, bodily (dis)pleasure, and emotion


both the similarities and the dissimilarities in the (un)pleasantness characteristic of different affective states. While the similarities stem from the fact that all affective states represent values, the dissimilarities stem from the fact that different affective states represent different thick values. That being said, the prospects of the explanatory priority thesis now turn on whether a feature of the thick values we experience in (un)pleasant bodily feelings supports a claim to explanatory priority. It is hard to figure out what this feature could be. We obviously cannot set our sights on thickness as such. Nor can we appeal to the fact that the values experienced in bodily (dis) pleasures are personal values – the values experienced in many emotions are also personal, and these emotions would therefore come out as explanatorily on a par with bodily (dis)pleasure. For the same reason, we cannot refer to the fact that the values experienced in bodily (dis)pleasures are final values. There may be a case for the claim that fearsomeness is an instrumental value, but this looks like an exception rather than the rule. Acknowledging that the values experienced in bodily (dis)pleasures and in emotions are different thick values leaves one without the resources to maintain the explanatory priority thesis.36 This puts an end to my examination of the first issue. We have explored two plausible ways of making sense of the relation between the values experienced in bodily (dis)pleasure and in emotions – as a determinable/determinate relation or as a relation between different thick values. None of the two is compatible with the explanatory priority thesis, which we should therefore abandon. The second and final issue is whether this leaves us in a position to accept the containment thesis, i.e. to claim that emotions contain (un)pleasant bodily feelings. Now that we have rejected the explanatory priority thesis, we have to see if there are reasons other than the attempt to account for emotional valence to claim that emotions contain (un)pleasant bodily feelings. We should acknowledge without hesitation that some emotions contain bodily (dis)pleasures. Unpleasant bodily feelings accompany some episodes of fear or shame, and pleasant bodily feelings accompany some episodes of joy or amusement. That being said, our discussion puts us in a position to appreciate two things. First, that emotions contain (un)pleasant bodily feelings contingently and, second, that when emotions contain such feelings, a subtle balance should be maintained for both to exist. Let me conclude by giving some substance to these thoughts. In light of the perceptual account of valence, the relation between emotions and bodily (dis)pleasures comes out as contingent. On this account, valence is constituted by value experience: the sense there is to say that an emotion is (un)pleasant is exhausted by the experience of the relevant value(s). To determine the modal status of the relation between emotions and bodily (dis) pleasures, we should thus wonder as to what is (un)pleasant when we undergo an emotion. We know that the answer must refer to the experience of a thick value – what is pleasant in an episode of amusement is the experience of the


Fabrice Teroni

joke’s funniness, what is unpleasant in an episode of fear is the experience of the lion’s fearsomeness, etc. Are these answers complete, or should we always add “and the body feels good or bad”? When we undergo emotions, there often seems to be nothing good or bad in the way the body feels – there is nothing pleasant in how the body feels in many episodes of amusement, for instance, nor is there anything unpleasant in how the body feels in many episodes of fear. More controversially, one may contend that some emotional episodes – in particular, intense ones – contain bodily feelings whose valence is the opposite of the emotion’s. Two examples come to mind: the bodily feelings in intense fear, which many find (intrinsically?) pleasant when they know that there is nothing to fear, and those in intense amusement, which may be rather unpleasant. This may constitute an additional reason to claim that the relation between emotions and bodily (dis)pleasures is contingent – why think that intense fear and amusement could not be what they are if they did not contain these “conflicting” bodily feelings? Emotional intensity also helps put some flesh on the idea that, when emotions contain (un)pleasant bodily feelings, a subtle balance has to be maintained for both to exist. We have seen that the intentional structure of emotions and bodily (dis)pleasures is a matter of value experience. In order to realize that their relation is not only contingent but also fragile, we should ask how value experience relates to attention. Many scholars insist that affective states have an essential relation to attention: they tend to focus our attention to the object whose value we experience.37 Our attention is drawn to the fearsome lion, the admirable painting or the painful wrist. Paying no attention at all to the object of an occurrent affective state seems to be impossible. We can of course resist this attentional pull, but this takes effort and the effort it takes is a function of emotional intensity: the more intense the emotion, the more irresistible the attentional pull. If this is along the right track, an immediate implication is that the relation between emotions and bodily (dis) pleasures is fragile. The more intense an emotion, the more it focuses the subject’s attention to its (typically distal) object – this makes it more and more incompatible with bodily (dis)pleasure. Conversely, the more intense a bodily (dis)pleasure, the more it focuses the subject’s attention to her body – this makes it more and more incompatible with emotional consciousness. This is something we all know: intense bodily (dis)pleasures interfere with emotions by focusing attention inwards, intense emotions interfere with bodily (dis)pleasures by focussing attention outwards. This is another truth that the advocates of the core affect approach ignore at their own risk.

Conclusion Our discussion has focused on the relations between bodily (dis)pleasures and emotions. I have argued that the core affect approach to affective experience is committed to two theses regarding the structure of the affective domain. According to this approach, the valence of bodily (dis)pleasures is

Valence, bodily (dis)pleasure, and emotion


explanatory prior vis-à-vis the valence of emotions and emotions are valenced because they contain bodily (dis)pleasures. These two theses are in my opinion irreconcilable with what we can reasonably claim to know about valence. The valence of bodily (dis)pleasures and emotions is a matter of value experience. I have argued that the nature of value experience as well as the values that we experience in bodily (dis)pleasures and emotions should lead us to reject these theses.

Notes 1 Some philosophers and psychologists mention surprise as a neutral or not valenced emotion. I tend to think that the phenomenon described as neutral surprise is not an affective state. There are surely positive and negative episodes of surprise, but surprise is – in the same way as beliefs, imaginative experiences and memories – valenced in virtue of its relations to affective states. 2 Some philosophers, most notably Solomon (2001), have emphasized that there are several reasons for which we may assess emotions positively or negatively. Nothing in what follows goes against this form of pluralism. I shall be interested in one important ground of these assessments, namely valence understood in view of the three presuppositions just presented. 3 Distinguishing these two groups of issues does not mean that there are no interesting relations between the two. For instance, some claims about the nature of valence have consequences for the structure of the affective domain. Some of these relations will emerge in what follows. 4 None of this implies that there are no ambivalent affective states (nostalgia being a case in point). The presuppositions laid out in the introduction bear on the account of affective ambivalence, but they leave room for this phenomenon. For a discussion, see Massin (2018). 5 Many debates surround the place and role of bodily feelings in emotion. I shall presuppose here that the phenomenal properties of affective states are (at least partly) a function of how the body feels. For discussion, see Deonna and Teroni (2017, forthcoming). 6 While this bodily aspect of affective experience may not be distinctive of affective states as types, it may help differentiate emotional episodes of the same type. Compare, for example, the joy felt after a good discussion with a friend, which is rather low in excitation, with the joy felt after having won a prestigious prize, which is typically quite high in excitation. 7 An analogy may help. Auditory experience is unitary, yet it can be plotted on several axes (pitch, timbre, etc.). These axes correspond to dimensions along which auditory experience can change, which does not imply that it is fragmented into separable experiential chunks. 8 For a discussion of some of these worries, see Deonna and Teroni (2017, forthcoming) and Teroni (2017). 9 That being said, the core affect approach may be in tension with James’s view, which does not recruit bodily pleasures and displeasures exclusively. 10 I advertised this interpretation as being the most charitable one. Doesn’t it clash with the idea, repeatedly put forward by Russell, that core affect is “objectless”? In my opinion, it does not: in insisting that core affect is objectless, it seems to me that Russell only wants to deny that core affect is about distal objects. 11 Given my purposes here, I shall leave aside the way advocates of the core affect approach try to capture these additional properties of emotions. The approach is often coupled with so-called dual component theories of emotions originating in






16 17 18 19 20


Fabrice Teroni

the work of Schachter and Singer (1962), according to which emotions are constituted of core affect and interpretations of its distal causal origin. For discussion of dual component theories, see Deonna and Teroni (2017). This raises two important issues I shall not discuss in what follows. First, one may wonder as to what it means exactly to represent damage in the way pain does. Second, one may harbour doubts about the intentionalist idea that often accompanies the claim that pain represents damage, i.e. the idea that the phenomenal properties of pain can be fully explained by damage representation (Tye 1995). The essays collected in Aydede (2005) are a good guide to the debates surrounding these issues. The question regarding what unifies the category of pain feelings is debated. Is it – as the intentionalist maintains – damage representation, or rather a feeling tone or a family of feelings tones (e.g. Crisp 2006)? And is it a merely contingent linguistic fact that there is no parallel subcategory of positive sensations? For an interesting discussion, see Trigg (1970). This is to cut a long story very short. For a detailed criticism of several variants of the motivational view of pain’s unpleasantness, see Bain (2019). In Teroni (2018), I offer some reasons to reject two variants of this view of emotional valence. Brady (2018) puts forward a motivational view of unpleasantness that may avoid this worry. According to this view, an unpleasant experience is a compound constituted by a sensation and the subject’s desire that this sensation does not occur. This is obviously not the place to discuss Brady’s view in any detail. Still, let me observe that a potential liability of the view is that the desire that the sensation does not occur – which is a proper part of the unpleasant experience – cannot be rationalized at the first-person level in terms of a desire that an unpleasant sensation ceases. One central aim of the evaluative view is precisely to preserve these kinds of explanations. One may think that in representing damage, pain feelings are already evaluative. According to the approaches under discussion, this is not the case: damage is a functional property that may be evaluated positively, negatively, or not at all. This is the lesson of the analogy with damage to one’s property that we have just discussed. A consequence of the evaluative view, which is shared by the motivational view, is that the valence of proprioceptive sensations is not essential to them. An influential model of these evaluations and how they unfold is to be found in appraisal theory. For variants of this theory, see Lazarus (1991) and Scherer (2001). The much-discussed property of transparency (e.g., Martin 2002) is closely related to this last observation. This is at least so in the veridical cases – and a perceptual approach typically leaves room for illusions, i.e. cases where the relevant object does not have the property it perceptually seems to have. In adopting Bain’s account of unpleasantness in terms of perceptual experience of disvalue and, as we shall see later, a similar account of valence in general, I leave aside my misgivings about the very idea of a perceptual experience of value and its application to emotions (Deonna and Teroni 2012; Teroni 2018). I leave these misgivings aside, as the issues I am interested in remain basically the same however one understands the nature of evaluative experience – provided, that is, that one applies the same account to the valence of (un)pleasant bodily feelings and emotions. In claiming that unpleasantness is constituted by a perceptual experience of value, the account aims at identifying a type of representation that accounts for the phenomenology of unpleasantness. The claim is that perceptual representation – as opposed, say, to conceptual representation in the guise of a value judgement –

Valence, bodily (dis)pleasure, and emotion


23 24

25 26 27

28 29 30


32 33


explains this phenomenology. One of the aims of the perceptual account of unpleasantness is to avoid the famous “messenger-shooting objection” (Jacobson 2013 and, for discussion, Bain 2019). In a nutshell, the idea is that perceptual representation generates a phenomenology which can explain and rationalize distinctive types of attitudes towards our experiences, attitudes that cannot be rationalized by other types of representation. The perceptual account of pain’s unpleasantness explains valence in terms of value representation. Does it imply that valence is a relation, whose relata are a psychological state and a value? This depends on one’s account of perceptual experience. As I observed in footnote 19, most advocates of the account agree that pain feelings remain unpleasant even if the perceptual experience is illusory, i.e. even if there is nothing of final, personal and negative value at the relevant place in the body. If so, then valence is at best a pseudo relation. I am not sure whether Bain, whose account of pain’s unpleasantness I borrow, would be happy to generalize in this way. Any kind of psychological states or combination thereof can function as the cognitive base of an emotion – we may regret a deed we remember, hope that an event we foresee will occur, be sad that a theory we are thinking about is wrong, etc. On the nature and roles of the cognitive bases of emotions, see Deonna and Teroni (2012). The appraisal theories of Lazarus (1991) and Scherer (2001), which I mentioned in footnote 17, have been primarily applied to emotions. Variants of the perceptual approach to emotions are defended in Prinz (2004), Roberts (2003) and Tappolet (2016). Williams (1985) introduced the distinction between thin and thick values in the contemporary literature. Thick values are specific ways of being good or bad – being sublime is a way of being (aesthetically) good, being unjust a way of being (morally) bad. Recent discussions of thick values and of how they relate to thin values can be found in Kirchin (2013, 2017). We shall have the opportunity to come back to the relation between thick and thin values in the next section. On the idea that thick values are the formal objects of emotions, see Kenny (1963) and Teroni (2007). An interesting issue is whether there are parallel cases related to unpleasant pain. One candidate is the sort of empathic response we may have when we see someone injured. To say that Mary admires what is admirable to her is to say that she admires what is, according to her, admirable. This does not mean that what is admirable is personally as opposed to impersonally valuable, i.e. good in virtue of contributing to the quality of the subject’s life for her (on this and, more generally, on the relations between emotions and the self, see Teroni 2016). Admiration may have personal value – this is a different issue to which we will soon turn. A further issue is whether some emotions represent instrumental as opposed to final values. I briefly come back to this issue in the next section. Brady (2018) argues that boredom creates a problem for this approach: isn’t boredom a way of suffering that is precisely characterized by flat rather than intense affect? I do not think that boredom is a way of suffering – one can suffer from being bored, but this is a matter of one’s boredom eliciting negative affect. We shall see below that there are reasons to resist this claim. Let me observe in passing that this reverse order of explanation hardly makes sense at the psychological level. Acknowledging this reverse order of explanation, the perceptual account would have to claim that bodily feelings are (un)pleasant in virtue of the experience of a thick value, where this experience takes the shape of an emotion. Bodily feelings surely elicit emotions, as when one is afraid of a sudden change in heartbeat. Yet, (un)pleasant bodily feelings are not primarily unpleasant because they elicit such emotions.


Fabrice Teroni

34 These observations relate to the fine-grainedness of perception, which is often adduced as a reason to claim that the content of perception is non-conceptual (e.g. Crane 1992). It would be ironic if, after having insisted that affective experience cannot be explained in terms of conceptual representation, one did not pay sufficient attention to the nature of perceptual representation. If I am on the right track, the fine-grainedness of perception is precisely what creates difficulties for the explanatory priority thesis coupled with a perceptual account of valence. 35 Why not say that the values perceived in (un)pleasant bodily feelings are hedonic values, so as to distinguish them from aesthetic or moral values? The reason is that this turns out to be less informative than one may think. The values at issue are those that we perceive in the body. As I have emphasized, we should not confuse these perceived values with the unpleasantness that consists in perceiving them. This means that we cannot take “hedonic” to be informative in the way we would most naturally take it to be, i.e. as specifying goodness in terms of a way of occupying consciousness. More generally, a good strategy for characterizing thick values is in terms of the restrictions they put on their potential bearers (e.g. Kenny 1963). For instance, a remark can be offensive because a speech act may manifest the intention to demean. This is not the case of a pebble or a logical rule. Yet, it is not clear which sort of illuminating restriction is connected to the body, i.e. to the only serious candidate for being the exclusive bearer of the thick values we are after. 36 As we have seen in the previous footnote, claiming that the values we experience in bodily (dis)pleasures are hedonic values is not a viable option. The same is true of the idea, which we have abandoned along the way, that pain is unpleasant in virtue of representing damage. If damage is a thick value, it is a value that can hardly claim explanatory priority over fearsomeness, disgustingness and the other thick values accounting for emotional valence. 37 I do not have the space to dwell on the nature of attention – and it seems to me that the observations in the text are sufficiently basic to constrain any account of attention. I shall also remain silent on the function of attention in emotion. For two different approaches, see Brady (2013) and De Sousa (1987).

Acknowledgements I am grateful to Michael Brady for his helpful comments on a previous version of this chapter.

References Aydede, M. (ed.) (2005). Pain. New Essays on its Nature and the Methodology of its Study. Cambridge, MA: MIT Press. Bain, D. (2019). “Why Take Painkillers?”. Nous 53. 2, 462–490. Barrett, L. F. (2006). “Solving the Emotion Paradox: Categorization and the Experience of Emotion”. Personality and Social Psychology Review 10, 20–46. Brady, M. (2013). Emotional Insight. The Epistemic Role of Emotional Experience. New York: Oxford University Press. Brady, M. (2018). Suffering and Virtue. New York: Oxford University Press. Carruthers, P. (2018). “Valence and Value”. Philosophy and Phenomenological Research 97. 3, 658–680. Crane, T. (1992). “The Nonconceptual Content of Experience”. In T. Crane (ed.), The Contents of Experience (pp. 136–157). Cambridge: Cambridge University Press.

Valence, bodily (dis)pleasure, and emotion


Crisp, R. (2006). Reasons and the Good. Oxford: Oxford University Press. Deonna, J. and Teroni, F. (2012). The Emotions. A Philosophical Introduction. New York: Routledge. Deonna, J. and Teroni, F. (2017). “Getting Bodily Feelings Into Emotional Experience in the Right Way”. Emotion Review 9. 1, 55–63. Deonna, J. and Teroni, F. (forthcoming). “Emotional Experience: Affective Consciousness and its Role in Emotion Theory”. In U. Kriegel (ed.), Oxford Handbook of Philosophy of Consciousness. New York: Oxford University Press. De Sousa, R. (1987). The Rationality of Emotion. Cambridge, MA: MIT Press. Grahek, N. (2007). Feeling Pain and Being in Pain. Cambridge, MA: MIT Press. Heathwood, C. (2007). “The Reduction of Sensory Pleasure to Desire”. Philosophical Studies 133, 23–44. Jacobson, H. (2013). “Killing the Messenger: Representationalism and the Painfulness of Pain”. Philosophical Quarterly 63. 252, 509–519. Johnson, W. (1921). Logic (Part 1). Cambridge: Cambridge University Press. Kenny, A. (1963). Action, Emotion and Will. London: Routledge and Kegan Paul. Kirchin, S. (ed.) (2013). Thick Concepts. New York: Oxford University Press. Kirchin, S. (2017). Thick Evaluation. New York: Oxford University Press. Lazarus, R. (1991). Emotion and Adaptation. New York: Oxford University Press. Martin, M. (2002). “The Transparency of Experience”. Mind and Language 17. 4, 376–425. Massin, O. (2018). “Bitter Joys and Sweet Sorrows”. In C. Tappolet, F. Teroni and A. Konzelmann Ziv (eds), Shadows of the Soul: Philosophical Perspectives on Negative Emotions (pp. 50–59). New York: Routledge. Nelkin, N. (1994). “Reconsidering Pain”. Philosophical Psychology 7. 3, 325–343. Prinz, J. (2004). Gut Reactions: A Perceptual Theory of Emotions. New York: Oxford University Press. Roberts, R. (2003). Emotions: An Essay in Aid of Moral Psychology. Cambridge: Cambridge University Press. Russell, J. A. (2003). “Core Affect and the Psychological Construction of Emotion”. Psychological Review 110. 1, 145–172. Russell, J. A. and Barrett, L. F. (1999). “Core Affect, Prototypical Emotional Episodes, and Other Things Called Emotion: Dissecting the Elephant”. Journal of Personality and Social Psychology 76. 5, 805–819. Schachter, S. and Singer, J. (1962). “Cognitive, Social and Physiological Determinants of Emotional State”. Psychological Review 69. 5, 379–399. Scherer, K. (2001). “Appraisal Considered as a Process of Multilevel Sequential Checking”. In K. Scherer, A. Schorr and T. Johnstone (eds), Appraisal Processes in Emotion: Theory, Methods, Research (pp. 92–120). New York: Oxford University Press. Sidgwick, H. (1874/1981). The Methods of Ethics. Indianapolis: Hackett. Solomon, R. C. (2001). “Against Valence (‘Positive’ and ‘Negative’ Emotions)”. In R. C. Solomon, Not Passion’s Slave: Emotions and Choice (pp. 162–177). New York: Oxford University Press. Tappolet, C. (2016). Emotions, Value, and Agency. New York: Oxford University Press. Teroni, F. (2007). “Emotions and Formal Objects”. Dialectica 61. 3, 395–415. Teroni, F. (2016). “Emotions, Me, Myself and I”. International Journal of Philosophical Studies 24. 4, 433–451.


Fabrice Teroni

Teroni, F. (2017). “In Pursuit of Emotional Modes: The Philosophy of Emotion After James”. In A. Cohen and B. Stern (eds), Thinking About the Emotions: A Philosophical History (pp. 291–313). Oxford: Oxford University Press. Teroni, F. (2018). “Emotionally Charged: The Puzzle of Affective Valence”. In C. Tappolet, F. Teroni and A. Konzelmann Ziv (eds), Shadows of the Soul: Philosophical Perspectives on Negative Emotions (pp. 10–19). New York: Routledge. Trigg, R. (1970). Pain and Emotion. Oxford: Oxford University Press. Tye, M. (1995). “A Representational Theory of Pains and Their Phenomenal Character”. Philosophical Perspectives 9, 223–239. Williams, B. (1985). Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press.


Pain and mere tastes Toward an attitudinal-representational theory of valenced perceptual experiences Hilla Jacobson

Recently, I put forth a new account of the unpleasant, painful, aspect of the phenomenal character of pain. According to this account, which I term the “attitudinal-representational theory” (hereafter, “ART”), painfulness is constituted by a subjectively frustrated first-order negative conative (desire-like) attitude that is directed toward a bodily condition the pain represents as obtaining.1 In what follows, I elaborate on ART, and locate it among alternative accounts. I put a special emphasis on the contrast between ART and painevaluativism, which I take to be its most formidable rival. I then suggest that theories of painfulness should be put to a scope challenge – namely, examining the plausibility of their applicability to other valenced perceptual experiences. To assess their plausibility with respect to this challenge, I present extended, general versions of both ART and evaluativism. In the rest of the chapter, I propose a new kind of consideration – one that pertains to the far-reaching variability of valenced gustatory experiences – that may prove fruitful in adjudicating between ART and evaluativism. I argue that ART better accommodates the phenomenon of valence variance. As far as pains are concerned, then, the argument will proceed from pains to “mere tastes” and back again. The chapter as a whole has two (interrelated) purposes: first, to lend further support to ART as a theory of unpleasant pain, and, second, to make some headway toward vindicating ART as a general theory of valenced perceptual experiences.

Introducing ART As already mentioned, ART is an account of a particular aspect of the phenomenal character of pain – namely, its painfulness or negatively valenced phenomenal aspect. It is thus perfectly consistent with ART that there are pains that are not unpleasant (notably, those had by pain asymbolics). As to non-unpleasant pains or to the neutral aspects of unpleasant pains, ART endorses the idea that they are representational – they (perceptually) represent the obtaining of a certain bodily condition. Yet, according to ART, what turns the experiences that represent the obtaining of bodily conditions into unpleasant experiences is their involving, in addition, a negative conative (desire-like) attitude toward the represented conditions.2


Hilla Jacobson

The postulated desire-like attitude is a “negative desire” in that it is directed against a particular condition or state of affairs that is represented as obtaining. Furthermore, whether this bodily condition in fact obtains is irrelevant for the subject’s being in unpleasant pain. Rather, what is required for one’s being in an unpleasant pain is that the bodily condition be (perceptually) represented as obtaining, and that this representation be accompanied by a negative conative attitude toward this condition. That is, the conative attitude is directed against a condition that seems to obtain. In this sense, unpleasant pain is constituted by the mental fact of having a subjectively frustrated desire. Why do I refer to the attitudes postulated as constitutive of painfulness as desire-like attitudes rather than simply as desires? I take the desire-like elements constitutive of pains to be similar only in certain central respects to “ordinary” desires construed as conceptually structured full-blown propositional attitudes. Though key similarities obtain between a desire-like attitude towards p (hereafter, “DEp”, for a desire of an experiential sort) and the corresponding desire (more on this below), there are also notable differences between them. This is analogous to claims that most theorists (including evaluativists) accept (and which I, too, endorse) about the representational elements constitutive of pains and of “regular perceptual experiences”: the representational element constitutive of experiencing that p (hereafter, “REp”) shares key characteristics with the belief that p (e.g., both are truth or veridicality apt), yet it nonetheless differs from this belief in important respects. REp’s and DEp’s are experiential, more rudimentary, counterparts of beliefs and desires.3 Among other things, REp’s and DEp’s differ from beliefs and desires in their possessing non-conceptual contents and in their being poised to play somewhat different functional roles.4 It is especially significant to note that REp’s and DEp’s are (at least largely) cognitively impenetrable, and in particular mostly unsusceptible to direct influences by beliefs (see, e.g., Jacobson 2017). This explains why even a subject who believes there is nothing wrong with her body may still have a subjectively frustrated DEp, which constitutes her being in an unpleasant pain. It is worth emphasizing that ART is a phenomenological view of painfulness – namely, one that takes painfulness to constitute part of what it is like for you to undergo your pain. Moreover, by “phenomenology” I do not just mean the sort of what-it-is-like-to-be-in-a-mental-state that many think characterizes all occurrent (personal-level) mental states, including propositional attitudes such as beliefs. Rather, I take pain to be phenomenal in a stricter sense – specifically, in a way that is characteristic of perceptual states. ART, that is, takes painfulness to be part of the phenomenal character of unpleasant pain, and its goal is to account for that phenomenal character. Locating ART among its rivals, we may begin by noting that ART is a first-order view of pain. That is, the desire-like attitude that it postulates as constitutive of unpleasantness is a first-order attitude – it is directed at the bodily condition, which is an item within the extramental world; it is not directed at another mental state. Thus, ART substantially differs from the

Pain and mere tastes


traditional desire theory of pain. The latter is a second-order theory, according to which unpleasantness is constituted by a desire directed against the experience of pain.5 Next, Bain (2017, p. 40) refers to views according to which “pains have, partly in virtue of their unpleasantness, truth conditions” as “cognitivist views”. According to this characterization, ART is not a cognitivist view: ART maintains that, in contrast to its neutral aspect, the unpleasant aspect of pain, which is constituted by a rudimentary desire-like attitude, is not truth or veridicality apt. This is already one respect in which ART differs from evaluativism, namely, the view that unpleasantness is constituted by the fact that the bodily condition (represented by the neutral experience of pain) is also represented as bad for the subject. According to evaluativism, an unpleasant pain is veridical just in case the bodily condition represented by the pain in fact obtains and this condition is indeed bad for the subject.6 This difference between ART and evaluativism regarding their evaluability in terms of veridicality will figure prominently in what follows. Furthermore, not only is ART a “non-cognitivist view”, it is also a nonintentionalist view (it is not a “content-view”). According to intentionalism, the unpleasant phenomenal character of pain consists in its having the right intentional content (see, e.g., Bain 2017, p. 41 and Cutter and Tye 2014, p. 427) – at minimum, painfulness supervenes on pain’s intentional content. In contrast, according to what may be called “attitudinal views”, the phenomenal character of unpleasant pain does not supervene on its intentional content. Rather, painfulness is determined, in addition, by the psychological attitude (or mode) adopted toward the content, which, in the case of ART is a desire-like attitude. I take the distinction between attitudinal and intentionalist views to be significant (hence, my preference for calling my view the Attitudinal-Representational Theory.) This distinction marks an important difference between ART and evaluativism, the central tenet of which is that the phenomenal character of pain supervenes on its (rich, evaluative) representational content. But, further, it distinguishes ART from imperativism. According to imperativism, the phenomenal character of pain is explained in terms of the receipt of a bodyissued, experiential, command to do or avoid a certain action (such as “see to it that this bodily state does not exist!”), and this, in turn, is understood intentionally – i.e., as the idea that the phenomenal character of pain supervenes on its imperative content. 7 Imperative content, just like representational or indicative content, is a sort of intentional content. Both evaluativism (which is a version of representationalism – see below) and imperativism are intentionalist views.

Highlighting the attitudinal aspect of ART Let me elaborate on the distinction between attitudinal and intentionalist (especially, evaluativist) views a little further. A few points may seem to blur


Hilla Jacobson

this distinction. First, the difference may seem to be blurred by the fact that both intentionalist views and attitudinal views may be versions of functionalism. On both sorts of views, that is, painfulness may be explained in terms of pain’s occupying the right sort of causal role, or (invoking psychofunctionalist terminology) by its being characterized by the right mode of processing. But, whereas on intentionalist views, pain’s occupying the right sort of functional role constitutes its possession of a specific, evaluative (indicative or imperative) content, according to attitudinal views, the right sort of causal role determines the pain’s involving a particular (desire-like) psychological attitude. Second, as evaluativists now emphasize, according to evaluativism it is not that any representation of a bodily condition of yours as bad for you constitutes an unpleasant pain. Clearly, having a belief with that content need not be painful for you in any sense, and a fortiori it is not painful in what might be called “the bodily sense”. Rather, it is your experientially or perceptually representing a bodily condition of yours as bad for you that constitutes painfulness.8 In order to form a state with negative phenomenology, the content should be entertained “experientially”. How, then, are we to understand the notion of an “experiential representation”? Plausibly, this notion is to be understood attitudinally: experientially representing that your body is in a condition that is bad for you involves adopting an experiential or perceptual psychological attitude toward your body being in a condition that is bad for you.9 Rather than involving the mode of believing that p, the relevant representations involve the (non-doxastic) mode of your experiencing p, or of its seeming to you that p. This may seem unsurprising. Pain evaluativism is an instance of representationalism – i.e., the general view that phenomenal character is explained by representational content – and, like most versions of representationalism, it is not a form of pure representationalism: it does not explain the phenomenology of experiences solely in terms of their representational contents. Yet, once again, the distinction between attitudinal and intentionalist views may seem to be blurred, because on both views the psychological attitude seems to play a constitutive role in the phenomenality of pains. Yet, in this context too, the distinction between attitudinal and intentionalist views can be made clear. The most important clarification appeals to a familiar distinction between two questions that any complete theory of phenomenality must address. First, there is the question of what constitutes the fact that a mental state is phenomenally conscious – in virtue of what does a mental representation have any phenomenal character? Second, there is the question of what constitutes the fact that a phenomenally conscious state has the specific phenomenology that it has – in virtue of what does it have its particular phenomenal character? It is with the second question that both intentionalist (and specifically, evaluativist) views and ART are directly concerned – they are theories of the determination of particular phenomenal characters. Thus, intentionalists are perfectly entitled to appeal to attitudes

Pain and mere tastes


adopted toward contents in reply to the first question. They may say that involving a particular (specifically, experiential) attitude is a necessary condition for a representation being phenomenally conscious. But their most basic commitment – namely, that phenomenal character supervenes on, and is to be explained by, intentional content – entails that they cannot appeal to the kinds of attitudes the experiences involve in reply to the second question. ART, in contrast, appeals to attitudes precisely in the context of the second question. According to ART, two states with the same contents may differ in their phenomenal characters, because phenomenal character supervenes on, and is to be explained by, both content and attitude. Another role that some intentionalists assign to modes in determining phenomenology is that of distinguishing between the phenomenal characters of experiences belonging to different perceptual modalities. As Byrne (2001, p. 5) writes: “Intermodal intentionalists hold, while intramodal intentionalists deny, that the phenomenal differences between perceptual modalities – between visual and auditory experiences, for example – is determined by a difference in content.” Tye is an intermodal intentionalist, and there are some reasons for thinking that so is Bain.10 In any event, even if an evaluativist endorses only intramodel intentionalism, the crucial distinction between his view and ART still obtains. According to the former view, within a given perceptual modality, all differences in phenomenal character are due to differences in representational content, and according to the latter view, such differences may be due to the attitude adopted toward the representational content. For example, whereas according to evaluativism the difference between neutral and unpleasant pains consists in their respective neutral vs. evaluative contents, according to ART the contents of these states are identical, yet they differ in the attitudes involved, in that unpleasant pain involves an additional negative conative attitude. Lastly, there is yet another factor that may seem to blur the distinction between ART and standard versions of evaluativism – namely, that the particular attitude invoked by evaluativists may seem to bear striking resemblance to that of desiring. Consider some of the explanatory work that is supposed to be done by taking the evaluative representations constitutive of pains to be experiential rather than doxastic. First, it is to explain why pains motivate us to get rid of the bodily conditions they represent (see, e.g., Bain 2013). Clearly, being motivational is standardly taken to be a characteristic of desires. Second, note also that the reason-giving force of those representations is at least partly derived from their psychological attitude or mode. This is because states with the same contents lack this reason-giving force (ibid). Yet again this feature is taken to be a characteristic of desires, which are reasons per se – i.e., ones whose reason-giving force is (at least partly) rooted in their very existence rather than in the represented objects. Third, the representations being experiential is to explain why being in pain gives us a reason not only to get rid of the represented bodily condition, but also, and independently of whether this bodily condition obtains, gives us a reason to get rid of the pain


Hilla Jacobson

itself. That is, it is to explain why, whereas there is nothing intrinsically bad about, e.g., believing that something is bad for you, being in pain is intrinsically bad.11 I have argued at length that there are reasons for thinking – quite independently of the pain debate – that having a subjectively frustrated desire is intrinsically bad (Jacobson 2018). Yet, there are no reasons for thinking – independently of the pain debate – that experiential (in contrast to doxastic) representations of badness are intrinsically bad.12 Lastly, the relevant representations being experiential can be taken to explain the necessary connection (supported, e.g., by cases of pain asymbolia – see, e.g., Bain 2014) that is thought to obtain between unpleasant pains and caring about (or nonindifference toward) one’s body. Needless to say, quite independently of the pain debate, caring about the intentional object of an attitude is largely taken to be the mark of desire. At this stage, my own tendency is to conclude that if the attitude postulated to be constitutive of painfulness has so many of the characteristics of a desire of some sort, then, well, it is a desire of some sort. In order to deliver the explanatory goods, the postulated attitude must have certain characteristics that are paradigmatic of desires. Thus, rather than postulating a sui generis attitude that shares so many of the characteristics of desiring, it is better to take the attitude to be a desire. But I strongly suspect that my rivals will not be convinced. Indeed, so far, they have not been. It is also worth mentioning that one plausible reaction on their part is to concede that the attitude they postulate is a “desire” under a very specific – cognitivist – conception of desire: there is notable proximity between these postulated attitudes and desires under their cognitivist or perceptualist construal.13 In fact, evaluativists actually would be invoking desire, just under a different name, if a cognitivist construal of desire, were correct. On this construal, desires in the ordinary sense are best conceptualized as perceptions of goodness – i.e., as states in which the desired object appears good to the desiring subject. This proximity may seem to bring ART and the relevant versions of evaluativism closer together again, and, moreover, to make it difficult to see how to adjudicate between them. Yet, the debate between ART and evaluativism is very real. At this point in the dialectic, we should return to the distinction between cognitive and non-cognitive attitudes understood as the distinction between those attitudes that are truth (or veridicality) apt and those that are not. Whereas it is crucial for the cognitivist conception of desires that desires are truth (or veridicality) evaluable, according to the traditional conception of desires, they are not. Similarly, whereas it is a defining feature of evaluativism that experiential representations are veridicality-apt, the desirelike attitudes to which ART appeals are not veridicality-apt. Thus, there is a straightforward way of adjudicating between ART and evaluativism – namely, by examining whether the unpleasant aspect of pain is veridicality-apt. The argument to be presented in the remainder of this chapter will bear directly on this question.

Pain and mere tastes


Beyond pain: the uniformity assumption, and the extended versions of ART and evaluativism Recall that on both evaluativism and ART, unpleasant pain is, or involves, a valenced perceptual experience. Now, so far, both evaluativism and ART have been presented first and foremost as theories of pain, and pain, its importance notwithstanding, is but one instance of a valenced perceptual experience. There are also, for example, delicious and distasteful gustatory experiences, delightful and disgusting olfactory experiences, and, as recent scientific data suggests, even intrinsically pleasurable and unpleasurable visual experiences. Indeed, valence may be ubiquitous in the visual phenomenal realm (see., e.g., Lebrecht et al. 2012). When repeatedly sipping an exquisite wine, moving away from the intolerable squeak of a fire alarm, or moving closer to the edge of a cliff to get a better view of a magnificent scene, it seems that the experiences, similarly to pain experiences, are valenced, and have a motivational, reason-giving force. The sensible world, with its colors, odors, tastes and shapes, is not given to us in an evaluatively neutral, inert manner; rather, our perceptual interactions with it appear to be infused with valence. Indeed, it seems that this is what makes it so glorious (or, at times, bleak). Due to the fact that the valenced aspect of pain is often so pronounced, and given the significance of pain in our life, it is quite natural, when attempting to account for valenced perceptual experiences, to begin with the apparently paradigmatic case of pain. Indeed, philosophers have now been focusing on pains for more than a decade. However, I suggest that even for those interested primarily in pain, the time is ripe to turn our attention elsewhere. This is because of the initial plausibility of what can be termed the uniformity assumption – namely, that in order for an account of the valenced aspect of pain to be plausible, it must be applicable, mutatis mutandis, to the valenced aspects of other perceptual experiences. If, indeed, valence is intrinsic both to the phenomenal character of pain and to the phenomenal character of other perceptual experiences, we would expect that these phenomenal characters are constituted by similar elements.14 Thus, theories of pains should meet what we may call a scope challenge. It is worth noting that evaluativists, in particular, may have additional reasons to accept the uniformity assumption. One of the main reasons for the attractiveness of evaluativism is that it is a species of representationalism – i.e., to repeat, the view that phenomenal character is explained by representational content – and representationalism is the most prominent theory of phenomenal character. The attractiveness of representationalism, in turn, is partly due to its promise to naturalize phenomenal consciousness. If experiences have their phenomenal characters in virtue of their specific representational contents, and if there is a plausible naturalistic account of that in virtue of which experiences have their representational contents (i.e., a plausible naturalistic psychosemantics), then phenomenality is thereby naturalistically explained. In fact, even independently of the promise for naturalization, it is


Hilla Jacobson

widely accepted that if representationalism is true at all, it applies to all sensory experiences – representationalists should opt for the view in its unrestricted form (see, e.g., Byrne 2001). In particular, representationalism must be extended to pains, which were traditionally considered especially challenging for the view. Thus, the strengths of representationalism provide reasons for seriously considering evaluativism, and evaluativism, in turn, is crucial for the vindication of representationalism. If the uniformity assumption is challenged, evaluativism in particular and representationalism in general are undermined. Thus, both evaluativism (in and of itself as well as an extension of representationalism) and ART must extend beyond pains. Also, in the case of both theories the general outline of how the extension is to proceed is rather clear. Evaluativism would postulate that the valenced aspects of “positive” and “negative” perceptual experiences are due to those experiences’ representing various conditions as good or bad for the subject (respectively). ART would postulate that (subjectively) frustrated negative conative attitudes toward the conditions represented as obtaining constitute the valenced aspects of unpleasant experiences, and that (subjectively) satisfied positive conative attitudes toward the conditions represented as obtaining constitute the valenced aspects of pleasant experiences. In any event, the contrasting commitments of each view regarding the veridicality question must be maintained: evaluativism, but not ART, should maintain that the valenced aspects of pleasant and unpleasant perceptual experiences are veridicality-apt. In what follows I will put both evaluativism and ART to the test of the scope challenge. I will argue that there are reasons for thinking that only ART can meet this challenge. The phenomenon that will stand at the core of the following argument is that of valence-variance: an examination of perceptual experiences other than pains reveals that their valenced aspects exhibit a large degree of inter-subjective as well as intra-subjective variability. The argument will focus on the example of gustatory experiences. It will rely, inter alia, on intuitions to the effect that there’s no disputing about taste, as well as on empirical findings concerning the sort of factors that determine gustatory valenced aspects and the kinds of processes leading to their instantiation. The tentative conclusion will be that the valenced aspects of gustatory perceptual experiences are not veridicality-apt. If so, given the uniformity assumption, we have reasons for preferring ART over evaluativism.

The argument from valence-variance against evaluativism Consider a few mundane, familiar cases of variance in gustatory valenced experiences (hereafter, “g-valenced experiences”). Case A – Inter-subjective mere preferences Dinna’s gustatory eggplant experiences are extremely pleasant, while Dan’s gustatory eggplant experiences are deeply unpleasant.

Pain and mere tastes


Case B – Intra-subjective mere preferences Dinna used to dislike the taste of eggplants (her gustatory eggplant experiences used to be unpleasant), but now she enjoys it (those experiences have become pleasant). Case C – Sporadically valenced sensory experiences Dan likes both vanilla ice cream and chocolate ice cream. On some occasions, he finds vanilla ice cream more pleasant, and on other occasions, chocolate ice cream strikes him as tastier. Similarly, on certain occasions his taste for ice cream (vanilla or chocolate) slightly decreases. Case D – Sudden dramatic valence shift Dinna loves cocoa. One morning, when being served cocoa for breakfast, she finds it unpalatable. The challenge to evaluativism is as follows. Our starting point is that in these cases: 1.

The two experiences of the (same) substance have different valences, and thus differ in their phenomenal character.

Evaluativism is committed to the supervenience of phenomenal character on representational content, hence this commitment in conjunction with (1) implies that: 2.

According to evaluativism, the two experiences of the substance differ in their representational contents (specifically, in the evaluative aspect of their contents).

Given that evaluativism (on its standard, Russellian versions) is committed to the supervenience of representational content on represented properties, it follows that: 3.

According to evaluativism, the two experiences represent the substance as possessing different g-values.15 And so:


Evaluativism is committed to holding one of the following two hypotheses: Either: Each substance has (on the given occasion) at most one of the g-values the two experiences respectively represent it as having, in which case at most one of those experiences is veridical. Or: Each substance may have (on the given occasion) both of the g-values the two experiences respectively represent it as having, in which case, both experiences can be veridical.16


Hilla Jacobson

Let us call the former hypothesis the value-uniqueness hypothesis (VUH), and the latter hypothesis the value-plurality hypothesis (VPH). I will argue that: 5.

Neither VUH nor VPH is plausible. Hence,


Evaluativism (as a theory of valenced gustatory experiences) is implausible.

In the next two sub-sections, I try to establish (5). I will assume that, for evaluativism (as a species of representationalism), both a commitment to g-valenced experiences’ being always (or even largely) falsidical and a commitment to their being always veridical (i.e., self-verifying) are problematic. Indeed, these assumptions ought to be shared by most representationalists. Widespread misrepresentation in a certain domain, especially among what appear to be normal perceivers, is widely taken to be implausible (see, e.g., Tye 2006). And, with respect to the idea that representations are self-verifying, many go so far as to take the possibility of error as necessary for the very notion of a representation (see, e.g., Dretske 1986; Bain 2009). Thus, if Cases A–D support the conclusion that the evaluativist cannot avoid these commitments, then his position, when applied to g-valenced experiences comes with a significant cost. Hence, if there is a viable alternative theory of g-valenced experiences that better accommodates valence-variance, these commitments give us reasons for preferring that theory. Let me add a final remark regarding the strategy of appealing to considerations from variance. Forceful arguments from variance have been presented especially in the context of color experiences. To mention but a few examples, Ned Block (2000, 2003) has argued that the variability of color experiences undermines representationalism, and Cohen (2004, 2009) has argued that it provides strong reasons for color-relationalism. Yet, there are many interesting differences between color experience variability and g-valenced experience variability. For now, suffice it to note two of them. First, (as is reflected by Cases A–D) the variability in g-valenced experiences is far greater than that of color experiences. Second (and likely relatedly), from the outset, we seem to lack clear-cut intuitions regarding the veridicality or nonveridicality of any g-valenced experiences. For these reasons, even those who are unconvinced by arguments from color experience variability may find the present argument more compelling. The value-uniqueness hypothesis As is reflected by Cases A–D, I will focus on human g-valenced experiences. Within the context of human experiences, theories of g-value that fall under

Pain and mere tastes


VUH take it to be a non-relational, inter-subjectively available property. G-value is the property of being (gustatorily) good/bad to a certain degree – a property that the same substance has for us all in all circumstances. According to one such view, an objectivist (i.e., mind-independent) view, g-value is a biological property of substances (e.g., being apt to harm, or being beneficial for, our bodies to a certain degree). And according to another, initially more plausible subjectivist (i.e., mind-dependent) view, g-value is a disposition to appear (i.e., taste) in characteristic positive or negative ways to normal perceivers under normal conditions.17,18 VUH faces difficulties already in light of the case of inter-subjective valence variability – namely, Case A. According to it, there is a real disagreement between Dinna and Dan regarding the “real” g-value of the eggplant, about which at most one of them is correct. This hypothesis actually encompasses two possibilities: either both Dan’s negatively valenced experiences and Dinna’s positively valenced experiences are falsidical, or one of them is veridical and the other is falsidical. The first possibility (i.e., that both experiences are falsidical) is quite unmotivated. From the outset, the idea that any of these experiences is falsidical is counterintuitive, all the more so that they both are. More importantly, given that the representationalist wishes to avoid the claim that g-valenced experiences are systematically falsidical, we may stipulate that one of the relevant experiences is veridical. However, before we do so, and thus move to the second possibility, it is important to note that one objection to views that fall under VUH is that, according to them, in the vast majority of cases, both experiences will in fact turn out to be falsidical. If, in contrast to what will be suggested below, there is a fact of the matter as to which of the experiences is veridical, then vast non-veridicality looms. Thus, those views indeed seem to be committed to the problematic assumption that g-valenced experiences are systematically, and moreover, quite drastically, illusory. As is illustrated by Case A, this is because both Dinna’s and Dan’s g-valenced experiences, as well as many other experiences in between the extremes of extremely pleasant and deeply unpleasant eggplant experiences, are all quite prevalent. According to the views in question, at most one of the relevant experience types along this wide spectrum is veridical, and so all the rest are falsidical. The second possibility – i.e., that one experience is veridical – faces an additional challenge, namely that any decision regarding which of the two experiences is veridical is highly arbitrary and stipulative. I will argue that any consideration to the effect that one experience reveals the way eggplants really taste, whereas the other experience is misleading, could equally be used to argue for the superiority of the other experience. In particular, given the sorts of factors that are empirically known to determine g-valenced phenomenal aspects, it is hard to see why the experiences of either Dan or Dinna are abnormal, and so what reasons there are to privilege one of the experiences as veridical. These factors include genetic predispositions, age, gender, exposure (or familiarity with specific foods) and acquired affective associations (see, e.g., Birch 1999).


Hilla Jacobson

At this stage, the argument runs parallel to Block’s argument against representationalism from the phenomenon of shifted spectra (Block, 2000, 2003). With respect to many hues, the color experiences of different apparently normal subjects (e.g., typical males and females) are phenomenally shifted with respect to each other, and there is no justification for arguing that perceivers of only one type see specific shades of colors as they truly are. First, like the subjects in Block’s argument who all pass standard tests for normal color vision, Dan and Dinna, we may safely assume, have intact discriminatory gustatory abilities. Relatedly, while some differences in g-valence can be explained by a difference in the sensory aspects of the relevant experiences – e.g., Dan could have been a supertaster to whom eggplants taste bitter – others cannot.19 Similarly, both the relevant visual experiences and Dan’s and Dinna’s gustatory experiences are representative of the experiences of many other subjects, as both liking and disliking eggplants are common; neither Dan nor Dinna have atypical experiences. Finally, and quite importantly, as Block emphasizes, the variability in color experiences is due to factors such as gender, race and age, thus stifling the reaction that one group should be regarded as normal and the other as defective (that reaction, he says, would amount to sexism, racism or ageism). As is attested by the empirical literature (see, e.g., Birch 1999), in a similar vein, we may realistically assume that the relevant factors that distinguish Dan and Dinna as gustatory perceivers are their being accustomed to different culinary traditions (Dinna is native to Middle Eastern cuisine, while Dan’s inclinations are due to an Eastern European culinary upbringing), their gender difference, or their age difference (Dinna is an adult, while Dan is a youngster).20 Thus, the challenge to the non-relational representationalist view is that there seems to be no principled, non-arbitrary way of singling out, or privileging, one of the relevant valenced experiences as the veridical one – i.e., as the one that reveals the (evaluative) way eggplants really taste. VUH, in short, seems to embody an unwarranted gustatory dogmatism. Perhaps, then, we should opt for a pluralistic approach, according to which both experiences are veridical? Perhaps there is no real dispute between Dinna and Dan, because neither is misled about the evaluative properties of the eggplant? Indeed, this option may seem to do justice to the intuition captured by the aphorism “There’s no disputing about taste”. This option is VPH. The value-plurality hypothesis The theories of g-value that fall under VPH take it to be a relational, intersubjectively unavailable property. The property represented by g-valenced experiences is not being (gustatorily) good/bad (for humans) to a certain degree. Rather, it is being (gustatorily) good/bad to a certain degree for a particular type of perceiver, Px, under a particular type of condition, Cy. For example, one natural version of g-value relationalism is a (subjectivist) dispositional view, according to which g-values are dispositions (of certain

Pain and mere tastes


substances) to cause g-experiences with specific valenced aspects in (normal) perceivers of type Px under (normal) conditions of type Cy.21,22 And according to yet another natural version of the view (hereafter, “the relationalist biological view”), g-values are objective (mind-independent), biological properties, of being beneficial or harmful (to a certain degree) for (the bodies of) perceivers of type Px under conditions of type Cy. In light of the difficulties inherent in g-value dogmatism, the present pluralist approach, which postulates a great plurality of g-values (the eggplant has different evaluative properties for different subjects in different conditions), may seem attractive. Yet, I shall now argue, pluralist, relationalist views are also implausible. Whereas the objectivist, relationalist biological view stumbles over the problem of massive, radical non-veridicality, the subjectivist (experience-dependent) relationalist view confronts the opposite problem. The problem with the experience-dependent relationalist view, I will argue, is that in the case of g-values (in contrast, perhaps, to that of color), once we go down the pluralistic road, it seems that there is no stable stopping point. It isn’t only that more g-valenced experiences turn out to be veridical, rather, it turns out that, plausibly, all such experiences are veridical. Let us begin with the latter (experience-dependent) relationalist view. The challenge to the representationalist is now the following: rather than motivating the idea that one of the experiences (in Cases B–D) is veridical (as in the argument against VUH), she should now show that (in at least one of these cases) at least one of the experiences is falsidical. Since B–D seem to be rather exhaustive as kinds of intra-subjective variability, if none of the relevant experiences is falsidical, it seems that the representationalist is indeed heading down the road of self-verification. In fact, even proponents of relationalism (specifically, color-relationalism) admit that they face a serious challenge concerning their ability to account for perceptual misrepresentation. Thus, as Cohen writes: the relationalist’s ease in reconciling apparently conflicting representations of a single stimulus comes with a concomitant cost: while relationalism makes it extremely easy for a representation to be veridical (hence the possibility of reconciling apparently conflicting variants), it can seem that the relationalist lacks resources for saying that a representation of x’s color is erroneous. Surely, though, a theory of color that makes errors of color perception impossible cannot be correct. (Cohen 2007, pp. 335–336) But even if this challenge can be met in the case of color – after all, even intuitively, there are cases that present themselves as clear cut examples of color misrepresentation – I will argue that it is hard to see how it can be rebutted in the case of g-values.23 Consider, first, Case B (in which Dinna has acquired a taste for eggplants). The same considerations that showed that there’s no way of privileging one experience


Hilla Jacobson

as veridical in Case A, show that there is no way of privileging one experience as falsidical in Case B. Not only is there no intuition to the effect that one of Dinna’s experiences is falsidical, the hedonic shift she undergoes is plausibly due to an age-difference, or to effects of exposures and/or affective-associations, none of which seem to warrant taking one of her experiences to be falsidical. Consider next Case C (which exhibits some fluctuations in the valence of Dan’s vanilla ice cream experiences). Yet again, I argue, there is no intuition to the effect that either of Dan’s experiences is falsidical or even that it is less accurate than his other experience. Moreover, the kind of factors that are plausibly responsible for the hedonic shift, cannot privilege one of his experiences vis a vis the other. For example, a decrease in tastiness may be due to Dan’s being hungry or thirsty (he may find ice cream tastier after a good meal, and less tasty if being served as an appetizer), or to his being cold (the ice cream may seem tastier on a hot day). But none of these conditions seem to warrant the verdict that he is misled by his experience. (It is quite unmotivated to suppose, that, e.g., one’s g-experiences on a hot day are more accurate relative to the ones one enjoys on a cold day.) Furthermore, the hedonic shift may simply be due to his concurrent psychological states, such as what he feels like eating when looking at the desert menu. In light of the complete symmetry between a craving for chocolate ice cream and a craving for vanilla ice cream, in this case, yet again, there seems to be nothing misleading about the way the particular flavor of ice cream – both prior and after the onset of the craving – appears to him. Finally, consider Case D, in which Danna’s cocoa experiences undergo a sudden dramatic valence shift. This case may seem to be the best candidate for misrepresentation, and thus the most promising for the representationalist. Thus, the representationalist may admit that, just as different subjects might vary in their tastes with no implication of error, so might one subject’s taste change over time without giving us reason to suspect non-veridicality (as in Case B). Similarly, minor intra-subjective fluctuations in the tastiness of a particular flavor are no cause for suspecting misrepresentation (as in Case C). Yet, our representationalist may insist, misrepresentation can arise against the backdrop of some intra-subjective consistency in the g-valences of experiences of the same substance, when there is a rapid hedonic shift. Illusory g-valenced experiences are to be found when there are sudden significant deviations from a stable intra-subjective pattern of valenced experiences of the same stimuli. In reply, first, even in this case, there is no clear intuition to the effect that the atypical experience is illusory. For what it’s worth, my intuition clashes with the claim that whereas the cocoa now appears to Dinna to have a negative value, actually, in her present condition, it does have – in a manner that is somehow hidden from her – a positive value for her. We need to appeal to more theoretical considerations to consider whether the atypicality is any indication of non-veridicality. In particular, I recommend that we consider the processes that lead to such atypical experiences, and correspondingly, to the conditions that prevail when

Pain and mere tastes


the subject has such experiences. Are the processes leading to rapid hedonic changes radically different from those responsible for acquiring tastes? I will argue that the answer is negative. The most radical examples of rapid hedonic shifts are exhibited by what is known as conditioned taste aversion (CTA), or the Garcia effect (see, e.g., Garcia et al., 1974). CTA occurs when a distaste for a particular substance develops due to a single event that leads to the association of negative symptoms (primarily, nausea) with that substance. As the name attests, “conditioned taste aversion” is considered a (somewhat special – see below) example of classical conditioning or associative learning. More specifically, CTA is considered to be an instance of evaluative conditioning (EC), which is characterized by the psychological literature as the associative transfer of valence – i.e., as the process of acquiring likings/dislikings through associative learning or conditioning (see, e.g., De Houwer 2001, Rozin and Zellner 1985). EC occurs when changes in the valence of experiences of a stimulus result from pairing that stimulus with other “positive” or “negative” stimuli (i.e., stimuli that evoke experiences with positive or negative valence, respectively). Throughout multiple events (or laboratory trials), a neutral stimulus (e.g., a sound or a “neutral” taste) is paired with an affective stimulus (e.g., a “positive” taste), resulting in the transfer of the valence from experiences of the affective stimulus to experiences of the “neutral” (or differently valenced) stimulus. EC is canonically considered the standard process of acquiring tastes.24 I now argue that in light of the close relation between CTA and EC, and given that, at least under subjectivist versions of VPH, EC – as the standard psychological process of acquiring tastes – cannot plausibly be considered as leading to value illusions, neither should CTA. Again, CTA is considered an instance – albeit, a somewhat special instance – of EC. The clearest apparent difference between CTA and EC concerns the number of pairings between the neutral stimulus and the affective stimulus that is required for the transfer of valence. Whereas, in general, EC occurs as a result of a number of neutralstimulus/affective-stimulus pairings and there is a tendency for the EC effect to increase with increasing number of pairings (see survey of studies reported in De Houwer 2001), CTA demonstrates successful strong conditioning that is achieved on the basis of a single pairing. However, the difference in the number of pairings does not seem to be sufficient grounds for singling out experiences resulting from (standard) EC as veridical, while insisting on the non-veridicality of the experiences that result from CTA. Insisting that, for example, whereas g-valence experiences resulting from four pairings are veridical, those resulting from a single pairing are illusory, seems stipulative and ad hoc. Nor, as far as I can see, is there any other reason for privileging “standard EC experiences” and denouncing CTA experiences. Thus, it is hard to see how the conjunction of evaluativism with respect to g-valenced experiences and the experience-dependent relationalist view of g-value can leave room for misrepresentation.


Hilla Jacobson

Yet, the evaluativist has one final option – she may appeal to the relationalist biological view of g-value. To recall, according to that view, g-value is a relational but experience-independent (“objective”) property that is to be understood in biological terms – i.e., that of being conducive or adverse to biological flourishing. Surely, in this “biological-flourishing” sense, the same substance – e.g., some orange juice – might be very good for me when dehydrated or when suffering from hypoglycemia, and less so when I am not suffering from either of these conditions. According to the present view, g-valenced experiences thus track the relevant relational, mind-independent values – being good/bad (to a certain extent) for the particular individual on the particular occasion. It might be argued that if this is indeed the case, there would be a fact of the matter that settles the question whether each particular g-valenced experience is veridical, and that that question thus turns into a scientific one. However, this view is implausible in light of the fact, which is greatly strengthened precisely by scientific considerations, that e.g., contingent associations – and, specifically, ones that do not track values in the biologicalflourishing sense – play such a prominent role in determining the valence of g-experiences. As we saw, evaluative conditioning is a most standard mechanism for determining the valence of experiences, and factors such as exposure, as well as cultural influences, are among the most salient determiners of valence. Such factors suffice for determining whether a subject loves or detests the taste of eggplants. Since these factors have little, if anything, to do with the dietary values of eggplants, and hence with their value in the relational biological-flourishing sense, the conjunction of evaluativism and the relationalist biological view implies systematic, gross misrepresentation of g-valenced experiences. In this sense, note, the move from a non-relationalist view to a relationalist view is not very helpful for the evaluativist: the sort of factors that regularly affect the determination of g-valence do not correspond to substances’ nutritious values. Here is another way of noticing the problem: the variability in g-valenced experiences of subjects who, as far as their biological, nutritious needs are concerned, are similar, is simply too great (and, of course, the same is true of the variability in the experiences of the same subject at different periods of time). For such subjects, that is, the biological values of the same substances are similar, yet their g-valenced experiences will quite standardly differ, and moreover, they will often differ to a great extent. Thus, the view implies that the non-veridicality of g-valenced experiences is not only widespread, it is also, quite standardly, radical – our valenced gustatory experiences are standardly bound to mislead us in significant ways. The extent to which g-valenced experiences can vary, independently of variations in relational biological values, is important to the argument. In light of the shifted-spectra phenomenon, some representationalists are willing to bite the bullet of prevalent misrepresentation in the case of color experiences. However, the degree of variability in color experiences is very limited in comparison to the radical variability in g-valenced experiences. The same shade is seen by one

Pain and mere tastes


subject as pure green, by another subject as bluish-green, and by yet another subject as yellowish green; but none of the normal subjects sees it as purple or red. Indeed, representationalist attempts to rebut the shifted spectra challenge appeal to the fact that whereas non-veridicality of fine-grained color experiences is prevalent, veridicality does prevail at the level of coarse-grained color experiences (see, e.g., Tye 2006; Byrne and Hilbert 1997). Alas, no such rebuttal is available in the case of g-valenced experiences, because their variability clearly extends to the coarse-grained level: the eggplant can taste either delicious or repugnant to the same subject (in the same nutritional circumstances). Thus, the phenomenon of “shifted g-valenced experiences” seriously threatens the conjunction of evaluativism and the relationalist biological view of g-value. To recap, subjectivist, experience-dependent views that fall under VUH fail because there seems to be no principled way of privileging one of the relevant experiences as veridical, and experience-dependent views that fall under VPH fail because there seems to be no principled way of singling out one of the relevant experiences as falsidical. As to objectivist, experience-independent views, such views – even those that fall under VPH – lead to radical, gross misrepresentation. In accordance with the argument, evaluativism faces serious difficulties accommodating the variability of g-valenced experiences.

The argument from valence variance for ART The gist of the argument from the variability of g-valenced experiences in favor of ART may now be clear. What underlies the difficulties evaluativism faces in light of the variability of g-valenced experiences is that, due to this variability, it faces serious problems concerning the assignment of veridicality/ non-veridicality values to such experiences. Views that take g-values to be subjective, experience-dependent properties stumble over the arbitrariness of taking g-valenced experiences to be either veridical or falsidical. As we have repeatedly seen, for such views, there appears to be no fact of the matter that would support assigning veridicality/non-veridicality values to the relevant experiences. Objectivist views may be thought to avoid the arbitrariness problem, yet they are undermined due to their commitment to the relevant experiences being systematically, and rather commonly, radically falsidical. Given that taking experiences of a certain domain to be either systematically falsidical or systematically veridical (in fact, self-verifying) is highly problematic, both objectivist and subjectivist views face difficulties accommodating the variability of g-valenced experiences. Thus, the obvious move to take is to conclude that the valenced aspects of g-experiences are neither veridical nor falsidical – i.e., to adopt the hypothesis that they are not veridicality-apt. This, however, is precisely what ART predicts. According to ART, the attitude constitutive of the valenced aspect of perceptual experiences is a conative, desire-like attitude (under the standard, non-cognitivist understanding of desire), and, as such, this attitude is not evaluable for truth or veridicality. The notion of truth or veridicality is just inapplicable to conative attitudes.


Hilla Jacobson

The valenced aspects of perceptual experiences are not veridicality-apt because these aspects are not representational. Dinna’s having a pleasurable eggplant g-experience – her enjoying the eggplant – does not consist in her representing the eggplant as good for her in some sense. Rather, her enjoying the eggplant consists in her positive desire-like attitude toward the eggplant. This, rather than her experience’s revealing the eggplant’s value, is the best explication of what we naturally describe as her (currently) “liking it”. The present hypothesis is vindicated, I argue, due to its ability to do justice to the phenomenon of valence variability. The fluctuations in the valence of Dan’s and Dinna’s eggplant g-experiences are neither due to their more often than not misrepresenting the eggplant’s evaluative properties (as per VUH), nor are they due to the transient nature of the eggplant’s value (as per VPH). Rather, they are due to the fact that, on different occasions, the tasters’ desirelike attitudes toward the eggplant vary. They are simply fluctuations in whether, and to what extent, the eggplant – which is represented by their experiences in a non-evaluative manner – is currently (gustatorily) attractive or appealing to them. Correlatively, there is a sense in which, on different occasions of having g-valenced eggplant experiences, the eggplant – and not only their experiences of it – genuinely (in contrast to veridically) is good or bad for the subjects to varying degrees, and they have reasons for or against eating it. This is the sense in which a (positive or negative) desire toward an object – in virtue of its very existence – renders its obtaining good or bad for the desirer, and provides her with reasons for maintaining or eliminating the object. That ART provides the best account of valence variability is further strengthened both by considerations pertaining to the sort of causal processes leading to the determination of the valenced aspects of experiences, and by considerations regarding practical reasoning. Briefly, as to those processes, for one thing, their dependence on associations – often transient, non-repeatable ones – may support the idea that their function is not that of tracking the external evaluative properties of the stimuli. Relatedly, it supports the idea that the causal etiology leading to the valenced aspects of experiences is best conceived of as just that – a merely causal progression of events that does not suffice to ground facts regarding the veridicality of the resulting states. As to practical reasoning, one need not hold an exclusively desire-based view of reasons for action in order to accept that some of our reasons depend just on what we want, or find appealing. Thus, many hold that whereas my reason for choosing one rather than another path for my daily stroll may derive from my responsiveness to some of the path’s (desire-independent) evaluative features, it may also be anchored in the simple fact that I am presently inclined toward it. I suggest that the reasons provided by pleasant and unpleasant perceptual experiences for pursuing or avoiding their objects are excellent candidates for being such attitude-dependent reasons. I have argued that the best account of the variability of the valenced aspects of gustatory experiences treats their constitutive evaluative elements as

Pain and mere tastes


attitudinal. If this argument is successful, then, by dint of the Uniformity Assumption, the same is true of the valenced aspect of pains. Yet the Uniformity Assumption does not merely license the move from the plausibility of ART as a theory of g-valenced experiences to its plausibility as a theory of unpleasant pains; it also licenses the move to the plausibility of ART as a general theory of valenced perceptual experiences. If so, the phenomenally conscious aspects of many of our experiences are not exhausted by our representing the relevant aspects of the world; rather, they are also constituted by our attitudes toward those aspects.25

Notes 1 For an elaboration of ART, see Jacobson (2014, 2018). For a psychofunctionalist theory whose main tenets bear resemblance to ART, see Aydede (2014) and Aydede and Fulkerson (2018). Bain (2017, 2019) refers to this theory as the firstorder desire view, but in order to highlight some of its distinctive features (as I explain below) I prefer to term it ART. 2 In the following, I will often use “pain” as shorthand for “unpleasant pain”. 3 I use below the expression “psychological attitude” to refer to the attitude component of DEps and REps, although some reserve this expression to attitudes taken toward full-blown conceptual propositions. I use this expression more loosely. 4 A full explication of ART requires investigating the functional profile of DEp attitudes, an investigation that would be partly empirical. It is important to note that I do not take DEp attitudes to differ from desires in essentially involving some phenomenology. Rather, it is only subjectively frustrated DEp attitudes (and, for that matter, subjectively frustrated desires) that have such phenomenology. 5 Proponents of the traditional desire view include Armstrong (1962), Pitcher (1970), Heathwood (2007) and Brady (2018a, 2018b). 6 Proponents of evaluativism include David Bain (2013, 2017, 2019), Bennett Helm (2002), and Brian Cutter and Michael Tye (2011, 2014). 7 For defenses of pain-imperativism, see, for example, Hall (2008), Klein (2007 and 2015), and Martínez (2011). In his (2015), Klein restricts his (first-order) imperativism to neutral pains – he does not apply it to unpleasantness. 8 As already mentioned, evaluativists also need not hold that the contents of the pain and the belief are precisely identical – e.g., they can hold that whereas the latter content is conceptual the former is non-conceptual. The present point is that the difference in content, even if there is such, does not suffice to explain the difference in their phenomenality. Also, I’d like to note that on my view the subjective frustration of any desire, and not only that of an “experiential desire” constitutes a state with negative phenomenology. The difference between sorts of subjectively frustrated conative attitudes corresponds to the difference between sorts of painfulness (e.g., bodily and mental). 9 Another way of understanding the notion that there may be different modes under which some psychological state of yours can represent that your body is in condition that is bad for you appeals to differences in Fregean modes of presentations. I am concerned here with Russellian rather than Fregean versions of representationalism. 10 In his (2003), Bain attempts to explain the phenomenal difference between visual experiences of bodily damage and pain experiences by adverting to differences in contents, not modes.


Hilla Jacobson

11 The challenge of explaining why representations that something is bad for you are themselves bad for you is known as the Normative Objection or as the Messenger-Shooting Objection (see e.g., Bain 2013, 2019; Jacobson 2014, 2018). 12 See, though, Bain (2017, 2019) for replies. 13 For the cognitivist conception of desire, see Stampe (1987) and Oddie (2005). 14 I will not defend here the assumption that valence is intrinsic to the phenomenal character of various perceptual experiences. For present purposes, suffice it to note that if (as evaluativists and attitudinalists agree) this assumption is granted with respect to pains, it is highly plausible that it should also be granted with respect to other apparently valenced perceptual experiences. This assumption is also accepted by the scientific literature regarding valenced gustatory-experiences to which I shall later appeal. 15 Hereafter, I will often use the term “gustatory values” (“g-values”, for short) to refer to the putative evaluative properties of objects that according to evaluativism are represented by valenced gustatory experiences. Note that “value” refers to either positive or negative value. 16 Strictly, there is a third option – namely, that the substance has no evaluative properties. This would render all valenced g-experiences falsidical. As I write below, this consequence is considered implausible. 17 Two clarifications are in order. First, if (as is plausible) the relevant property is restricted to humans – i.e., a substance has one g-value for humans and another for, e.g., canines – then, strictly speaking, even the views discussed in this section are relational ones. The argument will focus on human experiences, and the distinction that is relevant to the argument is that between intersubjective and intrasubjective properties (“subjects” and “perceivers” being understood as human subjects and perceivers). Second, note that even an intersubjective evaluative property can be described as “being good/bad for the subject”, but the expression “the subject” in this locution is to be understood as “that subject – whoever s/he may be – that consumes the substance”. In contrast, in the case of an intersubjectively unavailable property (or a “relational property”, as I use this term) the identity of the subject is relevant to the individuation of the property. 18 I believe that the view that g-value is a non-relational, objective physical property of substances (useful or hurtful for our bodies) – i.e., one that is understood in biological terms – is implausible from the outset. It is highly unlikely that there is a unique g-value that substances have for all subjects, regardless of those subjects’ age, state of nutrition, etc. The more plausible version of the “biological view” of g-value is a relational one – i.e., one on which the relevant properties are useful or hurtful for a particular Perceiver (Px) in a particular condition (Cy). I will argue against the latter (relational) version in the next section, and if the argument is successful, then, a fortiori, it undermines also the former (non-relational) version. In the rest of the present section (specifically, in the considerations regarding the arbitrariness involved in deciding which of two experiences is veridical) I will focus on the “non-biological”, experience-dependent view. 19 Studies show that the increased sensitivity of super-tasters is pronounced in the case of bitter tastes. 20 This is one of the contexts in which it is worth keeping in mind that whereas variations in color experiences pertain to fine-grained shades, in the gustatory case the variations are immense – for Dinna, eggplants are delicious, and for Dan, they are unpalatable. Thus, important lines of reply to the shifted spectra argument – ones that appeal to the claim that whereas normal perceivers misperceive fine-grained shades, they nonetheless veridically see coarse-grained colors – are inapplicable to the argument from the variability of tastes (see, e.g., Tye 2006; Byrne and Hilbert 1997).

Pain and mere tastes


21 The notion of a “normal perceiver” is relevant to subjectivist relational views in the form of a “normal state” of a perceiver (see Block 1999, p. 71). 22 To clarify the way I distinguish between relational and non-relational experiencedependent dispositionalist views, according to views of the relational sort, there are numerous types of (human) perceivers, and numerous types of sets of conditions, each of which are part of the specification of a different disposition instantiated by the same substance. According to views of the non-relational sort, in contrast, there is one (broad) type of (human) perceivers, and one (broad) type of conditions, which specifies a single disposition that is instantiated by each substance. (The relevant types are broad in that they are more inclusive relative to those that figure in the specifications of relationalist dispositionalist views: perceivers and conditions that are different according to relationalist views, fall under the same types according to non-relationalist views.) 23 I here remain agnostic about the plausibility of color-relationalism. 24 Interestingly, the scientific literature seems to use the term “valence” indiscriminately for both the pleasant and unpleasant aspects of experiences and for the putative positive and negative values of the stimuli. Thus, the expression “transfer of valence” that is frequently used in that literature can also be read as referring to transfer of value. This betrays an implicit endorsement of a relationalist view of g-values. 25 My research was supported by the Israel Science Foundation, Grant 1001/17. For helpful comments and conversations, I am very grateful to Ned Block, Arnon Cahen, David Chalmers, Jonathan Cohen, Susanna Siegel, and Preston Werner, all of whom have greatly improved this paper. I am especially indebted to David Bain and Hagit Benbaji, who have read numerous drafts of this paper.

References Armstrong, D. M. 1962. Bodily Sensations. London: Routledge and Kegan Paul. Aydede, M. 2014. How to Unify Theories of Sensory Pleasure: An Adverbialist Proposal. Review of Philosophy and Psychology 5: 119–133. Aydede, M. and Fulkerson, M. (2018) Reasons and Theories of Sensory Affect. In D. Bain, M. Brady, and J. Corns (eds.), The Philosophy of Pain: Unpleasantness, Emotion, and Deviance. Abingdon: Routledge. Bain, D. 2003. Intentionalism and Pain. Philosophical Quarterly 53: 502–523. Bain, D. 2009. McDowell and the Presentation of Pains. Philosophical Topics 37: 1–24. Bain, D. 2013. What Makes Pains Unpleasant? Philosophical Studies 166: S69–89. Bain, D. 2014. Pains that Don’t Hurt. Australasian Journal of Philosophy 92(2): 305–320. Bain, D. 2017. Evaluativist Accounts of Pain’s Unpleasantness. In J. Corns (ed.), The Routledge Handbook of Philosophy of Pain (pp. 40–50). Abingdon: Routledge. Bain, D. 2019. Why Take Painkillers? Noûs 53(2): 462–490. Birch, L. L. 1999. Development of Food Preferences. Annual Review of Nutrition 19: 41–62. Block, N. 2000. Sexism, Racism, Ageism and the Nature of Consciousness. In R. Moran, J. Whiting, and A. Sidelle (eds.), Philosophical Topics (pp. 71–88). Fayetteville, AR: University of Arkansas Press. Block, N. 2003. Mental Paint. In H. Martin and B. Ramberg (eds.), Reflections and Replies: A Book of Essays on Tyler Burge. Cambridge, MA: MIT Press. Brady, M. S. 2018a. Painfulness, Desire, and the Euthyphro Problem. American Philosophical Quarterly 55(3): 239–250. Brady, M. S. 2018b. Suffering and Virtue. Oxford: Oxford University Press.


Hilla Jacobson

Byrne, A. 2001. Intentionalism Defended. Philosophical Review 110: 199–239. Byrne, A. and Hilbert, D. R. 1997. Colors and Reflectances. In A. Byrne and D. R. Hilbert (eds.), Readings on Color, Vol. 1: The Philosophy of Color. Cambridge, MA: MIT Press. Cohen, J. 2004. Color Properties and Color Ascriptions: A Relationalist Manifesto. The Philosophical Review 113(4): 451–506. Cohen, J. 2007. A Relationalist’s Guide to Error About Color Perception. Noûs 41(2): 335–353. Cohen, J. 2009. The Red and the Real. Oxford: Oxford University Press. Cutter, B. and Tye, M. 2011. Tracking Representationalism and the Painfulness of Pain. Philosophical Issues 21: 90–109. Cutter, B. and Tye, M. 2014. Pains and Reasons: Why it Is Rational to Kill the Messenger. Philosophical Quarterly 64(256): 423–433. De Houwer, J., Thomas, S. and Baeyens, F. 2001. Associative Learning of Likes and Dislikes: A Review of 25 Years of Research on Human Evaluative Conditioning. Psychological Bulletin 127(6): 853–869. Dretske, F. 1986. Misrepresentation. In R. Bogdan (ed.), Belief: Form, Content, and Function. Oxford: Oxford University Press. Garcia, J., Hankins, W. G. and Rusiniak, K. W. 1974. Behavioral Regulation of the Milieu Interne in Man and Rat. Science 185(4154): 824–831. Hall, R. 2008. If It Itches, Scratch! Australasian Journal of Philosophy, 86(4): 525–535. Heathwood, C. 2007. The Reduction of Sensory Pleasure to Desire. Philosophical Studies 133: 23–44. Helm, B. 2002. Felt Evaluations: A Theory of Pleasure and Pain. American Philosophical Quarterly 39(1): 13–30. Jacobson, H. 2014. Phenomenal Consciousness: From an Evaluative Point of View. Riga: Scholars’ Press. Jacobson, H. 2017. Pain and Cognitive Penetrability. In J. Corns (ed.), The Routledge Handbook of Philosophy of Pain. Abingdon: Routledge. Jacobson, H. 2018. Not Only a Messenger: Towards an Attitudinal-Representational Theory of Pain. Philosophy and Phenomenological Research. doi:10.1111/phpr.12492. Klein, C. 2007. An Imperative Theory of Pain. The Journal of Philosophy 104(10): 517–532. Klein, C. 2015. What the Body Commands: The Imperative Theory of Pain. Cambridge, MA: MIT Press. Lebrecht, S., Bar, M., Feldman, B. and Tarr, M. J. 2012. Micro-Valences: Perceiving Affective Valence in Everyday Objects. Frontiers in Psychology 17: n.p. doi:10.3389/ fpsyg.2012.00107. Martínez, M. 2011. Imperative Content and the Painfulness of Pain. Phenomenology and the Cognitive Sciences 10(1): 67–90. Oddie, G. 2005. Value, Reality, and Desire. New York: Oxford University Press. Pitcher, G. 1970. The Awfulness of Pain. Journal of Philosophy 68: 481–492. Rozin, P. and Zellner, D. 1985. The Role of Pavlovian Conditioning in the Acquisition of Food Likes and Dislikes. Annals of the New York Academy of Sciences 443: 189–202. Stampe, D. 1987. The Authority of Desire. Philosophical Review 96: 335–381. Tye, M. 2006. The Puzzle of True Blue. Analysis 66: 173–178.


Pain An attitude with two heads Hagit Benbaji

What can be added to a mere representation of some bodily condition to make it painful? A recent provocative reply rules out anything that is not about the bodily condition. On this view, what is required might be some additional content of the same representation, or perhaps some additional attitude towards the same content. Either way, the view says, the result is a bad motivational attitude that constitutes a reason to change the bodily condition and a reason to seek its elimination. Such an attitude pushes us in opposing directions: it should motivate us outward to change the world, but also inward so as to eliminate the attitude itself. A bad motivational attitude is directed both outward and inward, as if (like Dr. Dolittle’s “pushmi-pullyu”) it had two heads at opposite ends of its body. We can say the following about bad motivational attitudes (or BMAs): they provide us with world-directed reason (that is what the “motivational” stands for1), and they provide us with self-eliminating reason (that is what the “bad” stands for). The BMA view of pain purports to explain the most conspicuous aspect of bodily pain, its painfulness, without invoking experience-directed states or unpleasant qualia, and without compromising the normative and motivational nature of this negative affective state. A non-compromised account of painful pain (henceforward, I will just use “pain”) respects the truism that it is rational to take painkillers to eliminate pain in virtue of nothing above and beyond how it feels. I shall call this claim Badness: The painfulness of pain is a (justifying and motivating) reason to eliminate it simply by virtue of how it feels. Pain gives us a reason to eliminate it neither in virtue of extrinsic states, such as a frustrating desire to get rid of the pain, nor only in virtue of its consequences, such as, its being enormously disrupting. In short, pain is intrinsically and non-instrumentally bad.2 I argue that an attitude that constitutes a world-directed reason—a reason to change the world—cannot be bad. So the claim that pain is a world-directed reason is in conflict with the claim that it is a self-directed reason. The argument from conflicting reasons, as I shall call it, shows that an attitude that


Hagit Benbaji

gives us reasons for its own elimination cannot be guided by normative standards motivating us to change the world. The idea that pain is a BMA is not without allure. I begin by explaining why the main theories of pain construe it as a BMA. I then present the general argument from conflicting reasons against the possibility of a BMA and render it less abstract by showing how it applies to the main theories of pain. I then discuss the suggestion that on the main theories of pain, pain is not a genuine BMA by first distinguishing two kinds of practical reasons and then by distinguishing a complex state from its component. Both suggestions, I will argue, result in compromising the normative and motivational force of pain. The argument from conflicting reasons should be distinguished from the more familiar messenger objection (Hall 1989; Bain 2013; Jacobson 2013; Brady 2015). The messenger objection argues that representational views of pain, that take pain to be a mere representation, cannot explain what it is that renders a representation of a bodily disorder bad in itself. The argument from conflicting reasons is wider and stronger than the messenger objection. It is wider since it applies to any intentional attitude that provides us with a reason to change the world. Any intentional attitude with a motivational force, and not only a representational one, makes the perfectly rational act of taking painkillers seem as irrational as shooting the messenger. It is stronger since the messenger objection can be dismissed, at some point, with the quietist response that explanations come to an end somewhere (Bain 2017: 483). The argument from conflicting reasons shows that we cannot explain how an intentional attitude that motivates us to change the world is bad, because a motivating intentional attitude cannot be bad.

Variety of BMAs Echoing Wittgenstein’s subtraction problem (“what is left over if I subtract the fact that my arm goes up from the fact that I raise my arm?”; Wittgenstein 1953: §621), the contemporary literature on pain has raised an addition problem: A perceptual representation of bodily disorder + x = painfulness. The “what can be added?” question assumes that any theory of pain is a composite of a perceptual state plus something. Pains are not merely akin to perceptual states, they are literally composed of a perceptual state somatosensorily representing some bodily condition as being thus-and-so, specifically, a kind of disturbance, bodily damage, or a condition threatening damage (Bain 2013: 72). The appeal to perceptual experience is motivated by intentionalism that purports to explain the phenomenal character of experience by its representational content (Byrne 2001; Tye 1995). In the case of pain, intentionalism claims that what it is like to have severe neck pain is to be in a perceptual state that somatosensorily represents a bodily disorder located in the neck (Tye 2006; Byrne and Tye 2006; Bain 2003).

Pain: an attitude with two heads


However, just as an action is not a mere bodily movement, pain cannot be a mere perceptual representation. Just as mere bodily movements are purposeless, perceptual experience is affectless, and motivationally inert. But pain is a negative affect, and its being painful endows it with motivational power. It drives us to fight it (Korsgaard 1996). Typically, pain provides us with both motivating and justifying reasons to perform what Bain calls “avoidance behavior” (e.g., lifting your foot away from scalding bathwater). Further, pain provides us with both motivating and justifying reasons for taking painkillers. Having motivational power by virtue of its affective phenomenology, pain cannot be a mere perceptual representation, and the “what can be added?” question is raised. Granting intentionalism, the correct substitution in the one-variable equation cannot be a non-intentional state, qualia or the like; it has to be an attitude with content. The content can be directed either inward or outward. On the first alternative, x = an intrinsic desire to get rid of a perceptual representation of bodily disorder (Armstrong 1962; Brady 2015; Heathwood 2007; Pitcher 1970). The second-order desire theory adds to the perception of bodily disorder an aversion from it. What makes the perceptual experience of bodily disturbance painful is that we do not want it. As is well known, the view is susceptible to a version of Plato’s Euthyphro dilemma: is pain bad because we do not want it, or rather, do we not want pain because it is bad? On the second alternative, the BMA view, x = a con-attitude directed at the bodily disorder. The originality of the BMA view of pain is that it looks for a motivational attitude that is directed at the world. The missing ingredient that makes a mere perceptual experience painful is pro- or con- attitudes: attitudes which include an evaluative dimension that expresses our aversion from the bodily disturbance, that is, the intentional object of the experience, and not the experience itself. Con-attitudes prescribe certain courses of action which account for the normative and motivational force of pain. There are three main versions of such attitudes: evaluative perception, desire, and imperative. Each suggests a different substitution for the variable: 

Evaluativism: x = an experience of a bodily disorder as bad. This view enriches the content of painful perception such that it does not merely represent a bad disturbance but represents it as bad; that is, a perception of badness3 (Helm 2002; Cutter and Tye 2011; Bain 2013, 2017). First-order desire (henceforward, FOD): x = a desire that the bodily disorder cease. This view enriches the attitude of painful perception such that it does not include a mere perceptual representation of bodily disturbance, but a desire that this bodily disturbance ceases; that is, a subjectively frustrated4 desire (Aydede and Fulkerson forthcoming; Jacobson 2018). Imperativism: x = an experience of an imperative that orders that the bodily disorder should cease. This view enriches the perceptual experience’s indicative (or truth-apt) content with imperative content,5 for


Hagit Benbaji instance—“See to it that this bodily state no longer exists!” (Hall 2008; Klein 2015; Martinez 2011).

On each view, the missing element concerns the bodily condition.6 It ascribes an additional property to it (evaluative), or directs a different attitude to it (desire or imperative). How does adding con-attitudes or evaluative properties to the perception of bodily disturbance give us painfulness? Each view is based on a different intuition about the matter. For evaluativism, the point is to see “how natural and intuitive is the idea that its seeming to you that things are bad for you in some way can itself be bad for you in another way” (Bain 2013: 86). FOD theories claim that “there is a strong intuition to the effect that subjectively frustrated desires are bad” (Aydede and Fulkerson forthcoming: 25). Jacobson expresses a similar intuition: “having a subjectively frustrated desire7 is, well, frustrating (or otherwise disturbing); and being frustrated is bad” (Jacobson 2018).8 Imperativists could (but don’t9) argue that having a subjectively frustrated order is, well, threatening, and being threatened is bad. Evaluativists, FOD theorists, and imperativists trying to capture the badness of pain might claim, respectively, that, in a nutshell:10   

Perceiving badness is bad. Desiring something without getting it is bad. Being ordered to do something and not obeying11 is bad.

Let me emphasize that these claims (roughly and freely) express the intuition of the final stage of each explanation, which I will consider more thoroughly after presenting the argument from conflicting reasons, and its application to theories of pain. The rationale behind these theories is that pain is a BMA. Pain is motivational since it provides us with reasons to change the world as a first order con-attitude directed at the world (an evaluative perception, a desire, or an imperative). Pain is bad since it provides us with a reason to eliminate it.

The argument from conflicting reasons Consider a BMA. Since it is a motivating attitude, it gives us a world-directed reason to change the world. Since it is a bad attitude, it gives us a reason to eliminate itself. These two reasons, I argue, impose contradictory rational requirements on the attitude: the attitude should change the world and it should disappear. But if the attitude disappears it cannot, trivially, be a reason to change the world. There is something very odd about a state that is a reason to change the world and a reason to eliminate itself, since acting on that second reason would eliminate the motivation to respond to the first reason. An attitude cannot live up to both ideals: it cannot motivate us to change the world if it is eliminated. Hence, a motivational attitude that gives

Pain: an attitude with two heads


us a reason to change the world cannot be bad on pain of losing its motivational rationale. A BMA is, then, a source of conflicting reasons. On the face of it, nothing is more familiar than the phenomenon of conflicting reasons. I have a desire to drink coffee and a desire to get rid of this desire. Nothing objectionable about that. Granted—there is nothing objectionable about conflicting reasons that are constituted by different attitudes or different normative facts (e.g., a desire to drink coffee, the fact that coffee is bad for me, a second order desire to be rid of the first one). However, a BMA is, by its very nature, the sole source of both reasons to change the world and to eliminate itself. Here’s a model that may be helpful. There’s nothing incoherent about my urging or persuading you to give money to the poor and someone else urging or persuading you to shut me up. Further, there is nothing incoherent about my doing both those things, as long as they are independent of each other. Thus, if what is most important to me is to get to you to give money to the poor, and to that end I urge or persuade you to do so, it would be rather selfdefeating for me at the same time to urge or persuade you to shut me up, but it would not be incoherent. Given that my reason to urge you to shut me up is independent of my reason to urge you to give money to the poor, I can do both things. What is incoherent is when my urging you to give money to the poor is so nagging, perhaps a torment to you that it gives you a reason (justifying, not merely motivating reason) to shut me up. In this case, the nagging to do something good constitutes a reason to eliminate itself, thereby undermines its motivational power to benefit the poor. It might be objected that a BMA does not constitute any tension at all, let alone incompatible requirements, for the reasons it provides are of different kinds, usually called “right and wrong kind of reasons”, or “object-given and attitude-given reasons”. For example, epistemic reasons for beliefs bear on the truth or justification of the content and are thus content-related reasons (Parfit 2001; Piller 2001; Hieronymi 2005).12 By contrast, pragmatic reasons for having beliefs do not bear on whether they are epistemically justified or true, but on the utility of being in the state of having this or that belief, making them attitude-related reasons.13 There is no tension between having an epistemic reason to believe that not p, and a pragmatic reason (getting money for believing that p) to bring about this belief. Yes, it is true at the same time that the belief should (and tend to) disappear given the evidence against it, and that it should be brought about given the consequences of holding it, but the first “should” has a different sense from the second. It should disappear14 since it does not fit the world; but it should be held since it is having good consequences. Wrong kind of reasons are practical reasons, reasons to do things (including forming, continuing with, or getting rid of certain attitudes), thus they are quite different from reasons to think things true, justified or fitting.15 Indeed, crucially, one might argue that the notion of a BMA invites precisely this very distinction. The reasons of badness—reasons to eliminate the


Hagit Benbaji

attitude—are anchored to the attitude, while the world-directed reasons are anchored to its content. There is no tension between the two kinds of reasons, as there is no tension between epistemic and pragmatic reasons for belief. The temptation to appeal to the distinction between two kinds of reasons must, however, be resisted. The distinction marks different senses of reasons for attitudes, while the reasons constituted by a BMA are practical reasons for action. An attitude has content, which leaves room for distinguishing content-related reasons that show that an attitude’s content is true, justified or appropriate, and attitude-related reasons to act so as to bring about (or eliminate) the attitude, for, say, its good consequences. But actions do not leave room for this distinction. The reason to eliminate the BMA and the reason to change the world motivate and justify actions (eliminating the attitude, changing the world), and actions are justified in the same manner, namely, by showing that there is something in favor of the action. In the next section, I will clarify the argument and make it less abstract by considering its application to the BMA view of pain.

Two applications of the argument from conflicting reasons Let me now show that both the central theories of pain (the FOD theory and evaluativism) fall prey to the argument from conflicting reasons. The attitudinal-representational theory (hereafter ART), as it is developed by Jacobson, takes pain to be a complex mental state comprising the perceptual experience of bodily disorder and the desire that this bodily disorder not obtain. This complex mental state has precisely the two properties of a BMA, namely, motivational and bad. First, it is motivational. Desiring that not-p and experiencing that p are the conditions under which we are motivated to change the world: to eliminate p and bring about not p. Thus, the experience of bodily disorder and the desire that the bodily disorder not obtain are the conditions under which we should act on our desire: they constitute a reason (justifying and motivating, I’ll drop the distinction till the next section) to make the bodily disorder cease. But these very conditions, namely, desiring that not-p and experiencing that p, are precisely the conditions under which the desire is guaranteed to be (subjectively) frustrated, i.e., bad. A frustrated desire is, then, a BMA. Turning to evaluativism, as it is developed by Bain, its core claim is that pain is not an ordinary but an evaluative perceptual experience. The rationale behind postulating an evaluative content is to endow painful experience with motivational power. For example, Bain argues that a perception (or an illusion) of badness, cannot leave us cold: “when the badness for you of a state of your own body is impressed on you, this—independently of further desires— defeasibly motivates you to do something about that bodily state” (Bain 2013). Thus, evaluative experience is distinguished from ordinary experience in being a motivating attitude. The evaluative experience of the bodily condition as bad is a reason to eliminate the bodily condition. But an experience of

Pain: an attitude with two heads


badness is, Bain contends, itself bad, and is thus, a reason to eliminate itself. Perceptual experience of a bad bodily disorder is, then, a BMA. Now, in ruling out these views on the basis that they construe pains as BMAs, I am not denying that we can eliminate the frustrated desire that the bodily disorder not obtain (ART), or the experience that the bodily disorder is bad (evaluativism), and, nevertheless, be motivated to change the bodily condition simply because we believe that the bodily condition obtains or that it is bad. Certainly. But in such cases it is not the pain that motivates us to change the bodily condition, but other mental attitudes (for example beliefs) that accompany it. The point of the argument is not that we lose any reason we may have, just that an attitude that constitutes a reason to eliminate itself cannot also be a reason to change the world. To appreciate the conflict involved in the notion of a BMA, it should be noticed that ART and evaluativism are views on which what explains the painfulness of pain (e.g. evaluative content, anti-damage desires) is what explains the badness of that painfulness. By contrast, there are views which invoke one thing to explain the painfulness and another to explain the badness of the painfulness. Such an “additionalist” view16 takes the painfulness of pain to be constituted by, e.g., an evaluative experience of a bad bodily disorder, to which it adds a desire not to have this experience of a bad bodily disorder. The additional desire is not directed at the bodily disorder, that is, it is not an anti-damage desire; rather, it is directed against the evaluative experience, the painful experience, that is, it is an anti-painfulness desire that the painful experience cease. The anti-painfulness desire explains the badness of a painful pain, but there may be painful pains that do not invoke such additional desires, that is, there may be painful pains that are not bad (Cutter and Tye 2014). Thus, an additionalist view compromises Badness, because a painful pain is not intrinsically bad. The badness of a painful pain is conditioned on the additional anti-painfulness desire. The unpleasant pain composed of the evaluative experience and the desire to eliminate this experience is not a BMA, since each component is the source of a different reason. The evaluative experience is the source of a world-directed motivated reason to eliminate the bad bodily condition. The desire that the evaluative experience not obtain is the source of the reasons of Badness (Bain 2017). To be sure, it is a complex one component of which eliminates the other. But such a complex is quite common, as it exists in cases of, e.g., alienated desire, a desire that is accompanied by a second order attitude against it. Further, such a complex does not seem to reflect the normative aspects of pain. One could find oneself having to decide whether to act on the reason provided by the painful pain or the reason provided by the desire, whereas painful pain, I argue, never generates such a choice. When we are in pain, we do not find ourselves deliberating between eliminating the bodily disorder or the pain. Pain, in itself, does not impose such a deliberation on us. Nevertheless, odd as it may be, the complex is coherent. On an additionalist view there


Hagit Benbaji

is no one attitude that is the source of incompatible rational requirements. A BMA, by contrast, does generate such a conflict of reasons. Might my opponent reply that the normative division of labor is internal to the attitude, by which I mean that different aspects of the attitude are the source of distinct reasons. After all, pain has both content and phenomenology. Naturally, the reason to eliminate the bodily disorder relates to the content of the attitude, while the reason to eliminate the pain relates to its phenomenology, namely, its painfulness. Pain constitutes a reason to eliminate the bodily disorder in virtue of its content, and a reason to eliminate itself in virtue of how it feels, that is, painful. This will not do. Apart from the anti-intentionalist implications of separating content and phenomenology, the thought behind the BMA views of pain is that the reasons involved in pain are both grounded in its painfulness,17 thus, both reasons have the same source. On ART, it is the desire that the bodily condition not obtain upon perceiving (or seeming to perceive) that it does obtain that motivates us to change it. And it is the desire that it not obtain upon perceiving (or seeming to perceive) that it does obtain that is painful, and thus constitutes the reason to eliminate the frustrated mental state. On evaluativism, what motivates us to change the bodily condition is perceiving (or seeming to perceive) the badness of it (an evaluative belief with the same content may be inert). It is the perceptual encounter with a worldly badness that constitutes a reason to fight it. And it is the perceptual encounter with a worldly badness that is itself bad, and thus, a reason to fight itself. The idea of the BMA views is to explains the painfulness of pain via nothing but the intentional attitude that is directed at the bodily condition. The painfulness consists in a motivational directedness towards the bodily condition. So, the reasons of painfulness stem from the same source that grounds the reasons to change the world. However, it is precisely this that I claim to be impossible. Now, at this point, proponents of BMA views may deny the conflict of reasons, in the manner of the general strategy mentioned above, of distinguishing different kinds of reasons, or dividing the labor between different components. One instance of that strategy was to distinguish content-related and prudential reasons; the current move available to evaluativism is to distinguish motivating and justifying reason. Another instance of the strategy was to distinguish components of the attitude; the current move available to ART is to separate the motivational component from the bad complex. In the final next sections, I will elaborate and respond to each objection respectively.

Evaluativism: justifying and motivating reasons Bain’s evaluativist version of BMA views distinguishes between motivating and justifying reasons. He claims that first, an evaluative experience

Pain: an attitude with two heads


constitutes merely a motivating reason to change the world, and second, a bad attitude constitutes merely a justifying reason to eliminate itself. The first claim denies that an evaluative experience gives us a justifying reason to change the world. Only normative facts can be justifying reasons. The justifying reason to change the bodily condition is the badness of the bodily condition, not the perception of it. The justifying reason to eliminate the pain is the badness of the pain, that is, the perception of the badness of the bodily condition. By eliminating the evaluative perception, I do not thereby lose the justifying reason to change the world, because that reason is objective: it is the badness of the bodily disorder. By eliminating the attitude, I only lose an access to the reason, i.e., a motivating reason to change the world, not a justifying reason to do so. Thus, pain is not a BMA, for although the evaluative perception is bad, thus constitutes a justifying reason to eliminate itself, it is not a justifying reason to change the world. In this respect, there is a crucial difference between evaluativism and ART. On ART, the content of the experience, namely, that there is bodily damage, is not bad in itself, rather, it lacks any normative significance; what makes the existence of a bodily damage bad is desiring that it not obtain. That is, it is the existence of a con-attitude or negative desire that the bodily damage will not obtain that makes its obtaining bad. Thus, the attitude of desiring not p upon perceiving (or seeming to perceive) that p is the source of conflicting reasons to both change the world and eliminate itself. By contrast, on evaluativism, the bad attitude is merely a motivating reason to change the world and not a justifying reason to do so. The world itself constitutes the reason to change it, not the bad attitude of perceiving the worldly badness. The second claim denies not only that an evaluative perception is a justifying reason to change the world, but also that it is a motivating reason to eliminate itself. Bain argues that the badness of the evaluative perception, namely, the painful pain, is a justifying reason to eliminate itself, but not a motivating reason to do so. Of course you are also motivated to eliminate your pain, but you are not motivated to do so by the pain’s unpleasantness per se, “only by desires (intrinsic and extrinsic) for that unpleasantness not to be occurring” (Bain 2017: 485). Given the distinction between motivating and justifying reasons, Bain can reject my claim that pain is a BMA attitude that provide reasons both to change the world and eliminate itself. Pains don’t do this on evaluativism. Or rather, they provide only motivating reasons to change the world, and only normative reasons to eliminate themselves. Pain justifies its elimination, but does not motivate it; and pain motivates the elimination of the bodily condition, but does not justify it. Pain is not, then, a BMA because the alleged incompatible reasons are of different kinds, motivating and justifying. Separating motivating and justifying reasons may allow evaluativism to evade the letter of the argument from conflicting reasons, but not its spirit. For pain constitutes a special sort of motivating reason, namely, a motivating reason by dint of being a presentation of, or an encounter with, an objective


Hagit Benbaji

justifying reason. Thus, an encounter with a justifying reason to change the world presents us with what the world requires us to do, thereby requiring us to eliminate itself, namely, our motivation to act in light of what the world requires of us. Although the evaluative perception is not itself the source of conflicting justifying reasons, it is submitted to incompatible requirements: it should motivate us to change the bodily condition, for this is what the world requires; and it should eliminate itself. It presents us with a normative requirement set by the world and requires us to eliminate the motivation to satisfy that requirement. By analogy, it is as if the perception of suffering would require us to turn our gaze from the poor. Perception of suffering is unpleasant, as Bain suggests. But if it were the case that its unpleasantness constitutes a reason to eliminate itself, we could not also perceive the requirement to relieve the suffering. Korsgaard quotes Hutcheson’s apt point on this issue: “If our sole intention, in compassion or pity, was the removal of our pain, we should run way, shut our eyes, divert our thoughts from the miserable object, to avoid the pain of compassion, which we seldom do” (Korsgaard 1996: 149). Either we shut our eyes, or we face the pain, namely, perceive it, and move to relieve it. We cannot do both. To be sure, you can eliminate the pain, and still do what the world gives you reason to do without the pain. I may be motivated to change the world by remembering that my body is in bad shape. This memory is not (physically) painful, and yet, it may be a reason to eliminate the bodily disorder. In this case, notice, the pain is no more than a messenger—it is not itself the reason (not even motivating reason) to change the world. The point of the argument from conflicting reasons is that in pain, I cannot succumb to the normative requirement to change the world at the same time as eliminating the pain. Although the pain does not constitute the normative requirement to change the world, this requirement inherently makes itself felt in pain in such a way that I am motivated to fulfill it. Thus, in pain, I am subject to incompatible normative requirements: in perceiving the badness of the bodily disorder, I am justified to eliminate the bodily disorder and at the same time, that very perception. But if I satisfied the latter requirement, I cannot live up to the demand to change the bodily disorder. In any case, downgrading an evaluative perception to the status of mere a motivating reason seems too ad hoc to dispel the conflict. A perception that p is the best epistemic access to the fact that p, that is, it constitutes knowledge that p. Thus, when one acts in light of perceiving that p, one acts in light of the fact. One cannot do more to improve one’s epistemic situation vis-à-vis p. All the more so when perceiving that p is cashed out in terms of an encountering with p’s badness, being opened to it, or badness of p being “impressed on you”. Bain’s perceptual strategy appeals to these notions of perception: what is bad for you about your unpleasant pain is not your merely representing damaged states of your body as bad for you, but rather your

Pain: an attitude with two heads


putatively encountering those states’ badness-for-you, or your having their badness-for-you putatively impressed on you. (Bain 2017: 484) The metaphors emphasize the idea that when one perceives that p (or that p is bad), the fact itself is presented, not a representation of it, so that when one acts to eliminate p, one acts in light of the fact that p is bad, not in light of one’s merely believing that p is bad. Thus, in perceiving that p is bad, we are justified in fighting p, not merely motivated to do so. This is what is needed to generate the conflict. One may object that in cases in which there is not really any damage, e.g. phantom or psychogenic cases, one cannot act in light of the fact that the bodily condition is bad, for it is not bad (there is no bodily damage). Indeed, in such cases, one does not perceive that the bodily condition is bad, but merely seems to perceive this. But in such a case, there is a sense in which you are and a sense in which you are not justified in trying to fight p, rather like the case in which you get on a train because of a false belief that it’s going where you need to go (Bain 2013). The fact that there is a sense in which you are not justified in trying to fight p when there is no bodily damage, emphasizes that this is so only because you don’t act in light of the fact—you do not perceive that p is bad. Thus, in the “good” cases (in which p is bad), you are justified to fight p, because you do act in light of the fact. So far, I have argued that evaluative perception is a special sort of motivating reason—a motivating reason to act in light of the fact—thus, it presents the subject with incompatible normative requirements. Given Bain’s second claim that an evaluative perception is not a motivating reason to eliminate the pain, it may not be a full-blown BMA. But even if an evaluative perception does not generate a conflict of motivating reasons, I’ve argued that it generates conflicting justifying reasons. Regarding Bain’s second claim that an evaluative perception does not constitute a motivating reason to eliminate itself, let me close the discussion on evaluativism by pointing to a potential tension it may raise. The denial that pain motivates its elimination presupposes the conceivability of a subject who does not mind unpleasant pains, or even more strangely, likes them (Bain calls them Coolman and Strangelove respectively, in his 2017: 479).18 The unpleasant pain of such subjects does not motivate them to get rid of it, because they lack the desire to get rid of it (Coolman), or even desire the unpleasant experience to continue (Strangelove). This means that if Coolness, say, remains indifferent to the phenomenal character of the unpleasant pain, there has to be a property of this phenomenal character, to which his indifference is directed. Importantly, it has to be a property of the experience, rather than of the bodily condition, as Coolness is not indifferent regarding the bodily disorder—he is feeling bad about it. However, given representationalism, when Coolness tries to focus upon the experience, “he ‘sees’ right through it, as it were, to the entities it represents” (Tye 2006: 110). Although


Hagit Benbaji

Coolness may be aware that he is having an unpleasant experience, he is not “aware of a sensation or an experience: what he is directly aware of is an entirely physical condition of his body (whether or not this awareness is veridical)” (ibid.). The transparency of the experience implies that the phenomenal character of the experience is constituted by the badness of the bodily disorder, so to remain indifferent to the phenomenal character of the experience seems to remain indifferent to the badness of the bodily disorder.19 But this is in tension with the assumption that Coolness’s experience feels bad, that is, that he is not indifferent to the badness of the bodily disorder.

ART: separating desire and frustrated desire Jacobson too may object that pain is not a BMA, for what is bad is not what provides us with the motivational reason to change the world. What is bad is the complex condition of a subjectively frustrated desire, that is, the perception that the bodily condition obtains plus the desire that it not obtain. What is motivational is the second component of the complex, namely, the desire that the bodily disorder not obtain. Thus, there is no one attitude that is motivational and bad. The frustrated desire is bad, but it is not motivational. The desire alone is motivational, but not bad. The complex is bad, and it motivates us to change the world in virtue of the desire that composes it. Indeed, on this construal, there seems to be no conflict of reasons. The argument I presented above concludes that if a BMA is a reason to eliminate itself, then it cannot, trivially, be a reason to bring about any change in the world. A subjectively frustrated desire can, however, be eliminated by eliminating the perception that the bodily condition obtains, instead of the desire that it not obtain. Thus, we can eliminate the frustrated desire without losing the world-directed reason, that is, the desire that the bodily disorder not obtain. However, eliminating the complex condition by eliminating the perception that the bodily condition obtains does not leave the desire intact. On the contrary, the elimination of the perception that the bodily condition obtains eliminates the desire that it not obtain as well. For if p is already the case, the desire that p should (and tends to) disappear; it may survive as a favoring attitude towards p, such as liking, but not as a motivational state towards p. This is part of the difference between beliefs and desires regarding their opposing direction of fit. Beliefs have a mind-to-world direction of fit, such that in cases of misfit, it is their “fault”—they should be discarded. Desires have a world-to-mind direction of fit, such that in cases of misfit the world is to blame—it should be changed to fit the desire. Conversely, in cases of fit, the belief should continue while the desire should disappear. The desire motivates us to change the world only when there is (or seems to be) a misfit between its content and the world. That is, only when it is or seems to be not yet satisfied but—to some extent—frustrated. Any desire is, then, subjectively frustrated (to some extent). This means that as soon as you want things, according to ART, you’re in a bad situation (to that extent).

Pain: an attitude with two heads


It might be argued that the generalized conclusion that any desire is subjectively frustrated involves a misconception of the notion of frustration. A subjectively frustrated desire is not just the conjunction of desiring that p and an awareness (or seeming awareness) that not-p. Being frustrated means realizing (or seeming to realize) that the desire cannot be satisfied, or that all efforts to fulfill it are in vain. It is only such realization that is painful, not just any thought that p is not the case while desiring that p. However, on this understanding of frustration, a frustrated desire does not give us a reason to change the world at all. If a frustrated desire is a desire conditioned on believing that it cannot be satisfied, than such desire can only give us reason to eliminate it, not to satisfy it (since we believe that it cannot be satisfied). If a frustrated desire that p is to give us a reason for bringing about p, it should be accompanied by a realization that not p, not by the pessimistic thought that it cannot be satisfied. But, again, frustration in this weaker sense is part and parcel of any desire. Since any desire that p, as a motivational state, is conditioned on its being the case (or seeming to the subject) that not p, a subjectively frustrated desire is a BMA. The complex of desiring that p and perceiving that not p is, both, motivational and bad. Although the desire is the source of the world-directed reason, its motivational power is conditioned on the other component: the perception that it is not satisfied. The elimination of the perception that the bodily disorder obtains undermines the motivation to eliminate the bodily disorder simply because we have lost the perception that there is something to be done. Eliminating either the desire or the perception that frustrates it undermines the attitude’s motivational force to change the world. Hence, pain, on ART, is a BMA. Jacobson may concede that pain is a BMA, but argue that a BMA is not the source of conflicting reasons. On her novel theory of desire, a subjectively frustrated desire does not give us a reason to satisfy it; rather, it gives us a reason to satisfy or eliminate it. Thus, although a frustrated desire is a BMA, it is special in that it does not imply conflicting reasons but a disjunctive reason. A frustrated desire does not motivate us to change the world and to eliminate itself; rather it motivates us to change the world or to eliminate itself. Rather than pushing and pulling in opposite directions it leaves one to deliberate between two alternatives. In reply, I contend that the disjunctive conception involves a revisionist concept of desire. Given that frustration (to some extent) is the predicament of any desire, whenever I desire that p (or not p) I must always, according to the disjunctive conception, deliberate whether to satisfy or eliminate the desire.20 Now, each course of action requires further reasons. Admittedly, many times we do have reasons to prefer the satisfaction of our desires. Sometimes it may be simply easier to change the world than to get rid of our desires; desires can prove to be stubborn in the face of frustration. We may also have independent reasons to bring about p, as p may be itself good, apart from our desiring it (indeed, desiring the bodily disorder to cease seems to be


Hagit Benbaji

a candidate for such a motivated desire). Furthermore, as Jacobson notes, sometimes we have second-order attitudes in favor of keeping our desires, “since they are essential to who we are (and to who we wish to be)” (Jacobson 2018). All of these may constitute reasons to satisfy our desires rather than eliminate them. But they undermine the normative cum motivational role of the desire itself. On the disjunctive conception of desire, the desire alone (as Hume has said about beliefs), can never drive us to act. For the desire that p by itself does not constitute a reason to bring about p rather than eliminating itself. Only given other reasons, e.g., the second order desire not to rid of the desire that p, we can have a reason to bring about p. Hence, the disjunctive conception undermines the reason-giving force of desire altogether. Finally, let me point out that even if a frustrated desire were bad, it could not properly constitute the badness of pain. The disjunctive conception also involves a revision of our concept of pain. Pain motivates us to eliminate it regardless of what action we take towards the bodily disorder. The reason to eliminate the pain does not seem to be one alternative of a disjunctive reason: to eliminate either the bodily condition or itself. Pain gives us a reason to eliminate it. Even if a frustrated desire had been bad, it would not, then, be bad in the way pain is bad.

Conclusion I conclude that pains cannot give us both kinds of reasons, reasons to act on the world and to eliminate themselves. Since it seems to be a truism that pains give us reason to eliminate themselves, I deny that they also provide us with reasons to act on the world. But isn’t the claim that pains give us reasons to act on the world simply another truism? When your hand touches boiling water, don’t you pull away? In so far as this is an action, rather than a mere reflex, it is the pain that motivates you to pull your hand away, and thereby to eliminate the bodily disorder. And this seems to be a reason to act on the world. I agree that the pain motivates you to pull away your hand from the boiling water, but deny that this is a reason to act on the world. Rather, it is a reason to get rid of the pain itself, and to that end, “to change the world”. Pains are experienced as located—we talk of having the pain in the stomach, in the hand, etc. Given that we feel the pain “in this hand” (I put it in quotations marks because I don’t think that we conceive of the pain as literally located— we experience it as targeted at the hand), the pulling away is getting rid of the pain. The kind of acts that are considered to be “acts on the world”, pulling and pushing bodily parts, are, in fact, natural ways to get rid of the pain itself. The claim that pains motivate us to get rid of themselves, rather than to act on the world, provides a reason to go back to the traditional conception of pain as an inner experience that has no object independent of the experience. Although the inner view of pain has almost disappeared from the contemporary discourse on pain, it has been, throughout the history of

Pain: an attitude with two heads


philosophy, until very recently, not merely the prevailing view, but indeed the only one, and I believe that it still shapes our ordinary conception. The inner view raises a lot of questions, which should be elaborated on elsewhere, but it fits well with the conclusion of this chapter, that pains give us only reasons to get rid of themselves. The inner view may fit bodily pains, but it is totally misguided as a view of mental pains, which may raise a final objection to the conclusion of this chapter, that an attitude cannot give us both kinds of reasons, to change the world and to eliminate themselves. Mental pains, kinds of negative emotions, such as grief, sadness, anger, regret, shame, etc., are directed outside—the grief is about the loss, the anger is about the wrong doing, etc., and as such, they do give us reasons to change the world. But these emotions are also painful, and thus, seem to give us reason to get rid of themselves. Hence, mental pains seem to be a BMA—they seem to give us both kinds of reasons, to change the world and eliminate themselves. In reply, I deny that mental pains give us reasons to get rid of themselves. Although they are painful, they are not bad in the normative sense—they don’t give us reasons to eliminate them, rather, we have reasons to feel them—they are appropriate (which does not mean “good”). Bodily pains give us reasons to eliminate themselves, mental pains give us reasons to act on the world. They are all pains, but they are pains of different kinds.

Notes 1 This terminology is inadequate for two reasons. First, because I use the label “motivational” to mean that the attitude also justifies us acting outwardly. Second, because badness is also motivating. But this is just a label. 2 Bain argues that there is a gap between the claim that pain is intrinsically and noninstrumentally bad and the claim that pain is a motivating reason to eliminate itself. I shall come back to his view below. Badness is committed to both claims. 3 Or an illusion of badness, e.g., phantom pains. 4 A subjectively frustrated desire, since the experience can be inaccurate, that is, it can be an illusion of a bodily disorder. The desire is merely subjectively frustrated, because it is, as it were, already satisfied (there is no bodily disorder, so there is nothing to cease). 5 Klein’s view does not quite fit this schema, since he thinks pain’s content is entirely imperative, but that does not matter for the purposes of this chapter (anyway, the view must assume that the order to cease the bodily condition presupposes a perceptual experience of the bodily condition). 6 Although Klein doesn’t think pains represent a bodily condition, he does think the content is body-directed, since it is telling you what to do with and to your body. 7 A subjectively frustrated order, like a subjectively frustrated desire, is an order to cease the bodily disorder when the subject undergoes an illusion of it. Thus, phantom pain is a case of a subjectively frustrated order. The order is merely subjectively frustrated, because it is, as it were, already satisfied (there is no bodily disorder, so there is nothing to cease). This is different from Klein’s suggestion that phantom pains are especially awful since one cannot satisfy the command received. 8 Jacobson is the only exception that gives an argument to the intuition that a frustrated desire is bad.


Hagit Benbaji

9 Proponents of imperativism do not argue in this way, and they take their account not as an explanation of Badness. Rather, they seem to reject the intrinsic badness of pain: “No pain, however fleeting, abstract or undefined, is directing us to do something about itself” (Martinez 2015). Martinez’s view seems to be that what is bad is being commanded pointlessly. Therefore, I focus only on the first two theories in this chapter. 10 Putting aside the qualification regarding illusion of bodily disorder (see notes 3, 4, and 7). 11 One may object that if this were the view, then pain would not be bad provided we did what it told us to, e.g. take weight off our ankle, which seems incredible. But the same objection could be raised against the frustrated desire view—pain would not be bad provided we received what we wanted, e.g. take weight off our ankle. 12 Various objections to this formulation of the distinction have been raised; for a recent one see Shroeder (2012) and Hieronymi (2012). 13 Not all the considerations of the wrong kind fall under the category of incentives or prudential reasons. A funny joke can be morally cruel, thus providing us with a very good reason against laughing. 14 Although there is something odd about calling a reason to believe that not p a reason that the belief that p should disappear. Epistemic reasons are not reasons for eliminating or bringing about the belief as I argue in the text. 15 In some cases, the bad consequences that holding a particular belief might have may turn partly on the belief ’s content. If I am made sad by a belief that my mum has died, the content is playing a role in making me sad. 16 See Bain (2017) for this label. 17 Note that an additionalist view is intentionalist. It separates the painful quale from its motivational and normative aspects, but not the content from the phenomenology. 18 I doubt that such cases are conceptually possible, that is, cases of subjects who do not mind and even like the painful pain, rather than subjects who do not experience them as painful (asymbolia), or who like the pleasure caused by the suffering (masochists). Bodily pain is phenomenally bad, that is, bad in the experiential sense—it is bad in virtue of how it feels. Now surely X can be normatively bad for S though S may not care about it. X may be a wrongdoing, a torture, an injustice, etc. But it seems hard to conceive in what sense an experience is phenomenally bad, though the subject remains indifferent to it (or even likes it). 19 For a recent critique on the transparency of pain see Aydede (2019). 20 Bain raises a related objection: “we might especially balk at the … idea that your intrinsic desire for your mother not to be ill together with the fact that she is ill constitutes a non-instrumental reason on which you could act not only by taking steps to make her better but also by popping an anti-desire pill” (Bain 2017).

Acknowledgements This paper was supported by the Israel Science Foundation (grant no. 1168/ 2016). I am grateful for comments and discussion on this chapter to Dalia Drai and Hilla Jacobson. Special thanks are reserved for David Bain, who carefully read the chapter and made crucial substantive comments on almost every point.

References Armstrong, D. M. 1962. Bodily Sensations. London: Routledge and Kegan Paul. Aydede, M. 2019. Is the Experience of Pain Transparent? Synthese 196(2): 677–708.

Pain: an attitude with two heads


Aydede, M. and Fulkerson, M. Forthcoming. Reasons and Theories of Sensory Affect. In D. Bain, M. Brady and J. Corns (eds.), The Nature of Pain. Bain, D. 2003. Intentionalism and Pain. Philosophical Quarterly 53: 502–523. Bain, D. 2013. What Makes Pains Unpleasant? Philosophical Studies 166: 69–89. Bain, D. 2017. Why Take Painkillers? Nous 53(2): 462–490. Brady, M. 2015. Feeling Bad and Seeing Bad. Dialectica 69(3): 403–416. Byrne, A. 2001. Intentionalism Defended. Philosophical Review 110: 199–239. Byrne, A. and Tye, M. 2006. Qualia Ain’t in the Head. Nous 40(2): 241–255. Cutter, B., and Tye, M. 2011. Tracking Representationalism and the Painfulness of Pain. Philosophical Issues 21: 90–109. Cutter, B., and Tye, M. 2014. Pains and Reasons: Why It Is Rational to Kill the Messenger. The Philosophical Quarterly 64(256): 423–433. Hall, R. J. 1989. Are Pains Necessarily Unpleasant? Philosophy and Phenomenological Research 49: 643–659. Hall, R. 2008. If it Itches, Scratch! Australasian Journal of Philosophy 86(4): 525–535. Heathwood, C. 2007. The Reduction of Sensory Pleasure to Desire. Philosophical Studies 133: 23–44. Helm, B. 2002. Felt Evaluations: A Theory of Pleasure and Pain. American Philosophical Quarterly 39(1): 13–30. Hieronymi, P. 2005. The Wrong Kind of Reason. Journal of Philosophy 102: 437–457. Hieronymi, P. 2012. The Use of Reasons in Thought (and the Use of Earmarks in Arguments). Ethics. Jacobson, H. 2013. Killing the Messenger: Representationalism and the Painfulness of Pain. The Philosophical Quarterly 63(252): 509–519. Jacobson, H. 2018. Not only a Messenger: Towards a Hybrid Conative Representational Theory of Pain. Philosophical and Phenomenlogical Research. Klein, C. 2015. What the Body Commands: The Imperative Theory of Pain. Cambridge, MA: MIT Press. Korsgaard, C. 1996. The Sources of Normativity. Cambridge: Cambridge University Press. Martinez, M. 2011. Imperative Content and the Painfulness of Pain. Phenomenology and the Cognitive Sciences 10: 67–90. Martinez, M. 2015. Pains as Reasons. Philosophical Studies 172(9): 2261–2274. Parfit, D. 2001. Rationality and Reasons. In D. Egonsson et al. (eds.), Exploring Practical Philosophy, pp. 17–39. Burlington VT: Ashgate. Piller, C. 2001. Normative Practical Reasoning. Proceedings of the Aristotelian Society 75(Supp. 1): 195–216. Pitcher, G. 1970. Pain Perception. Philosophical Review 79: 368–393. Schroeder, M. 2012. The Ubiquity of State-Given Reasons. Ethics 122: 457–488. Tye, M. 1995. Ten Problems of Consciousness: A Representational Theory of the Phenomenal Mind. Cambridge, MA: MIT Press. Tye, M. 2006. Another Look at Representationalism about Pain. In M. Aydede (ed.), Pain: New Essays on Its Nature and the Methodology of Its Study. Cambridge, MA: MIT Press. Wittgenstein, L. 1953. Philosophical Investigations, trans. G. E. M. Anscombe. Oxford: Basil Blackwell.

Part III

The value of suffering


Suffering as transformative experience Havi Carel and Ian James Kidd

The ill person who turns illness into story transforms fate into experience; the disease that sets the body apart from others becomes, in the story, the common bond of suffering that joins bodies in their shared vulnerability. (Frank 1995, p. xi)

Introduction: transformative experience It is a cliché that suffering is transformative, a source of insight, wisdom, and understanding that can, or perhaps will necessarily, make one into a better person – one with a truer sense of what really matters in life which shows itself in the conduct of one’s life. But such claims, common as they are, are both complex and contested. Some find that experiences of suffering transformed them for the better, thereby endorsing Nietzsche’s claim that heroic coping with physical and psychological adversities ‘makes us deeper’ (Nietzsche 1974: preface §3). But others regard such claims as dangerous and deceptive, perhaps as failures to be truthful about the fact that suffering often takes more than it gives, or that it often gives nothing and takes everything, until – as Simone Weil graphically puts it – one is left ‘struggling on the ground like a half-crushed worm’, leaving them ‘in no state to help anyone at all, and almost incapable of even wishing to do so’ (Weil 2010, p. 39). Adjudicating such competing claims requires careful sensitivity to the subjective and situational complexity of such experiences, and alertness to the wider cultural contexts within which our conceptions of, and responses to, suffering are made. Barbara Ehrenreich is right to warn of a pervasive ‘ideology of positive thinking’ that encourages a relentlessly and dogmatically optimistic ‘bright-siding’ of human experience of adversity. In the case of illness, this can include the insistence that suffering always yields substantive moral goods, sufficient to nullify any sense of loss or regret arising from one’s suffering (Ehrenreich 2009; Kidd forthcoming). Such carefulness, sensitivity, and critical alertness is easier if one has robust conceptual resources on which to draw. In this chapter, we argue that the concept of transformative experience, developed by L. A. Paul, is one such


Havi Carel, Ian James Kidd

resource (Paul 2014).1 Her guiding claim is that there is a certain type of experience – or, perhaps, a set of experiences – that are distinctive in virtue of their doubly transformative capacity. They are, first, epistemically transformative, providing forms or degrees of knowledge that were previously unavailable and, more importantly, previously inaccessible, insofar as they depend on having the relevant experience (one of Paul’s examples is tasting the durian fruit). It is impossible to know, prior to tasting the fruit, whether one will love or hate it. Second, such experiences are personally transformative, fundamentally changing one’s values, preferences, and desires, and, therefore, transforming one’s identity in substantive ways. It’s not just what one knows and understands that’s substantially changed, but one’s wider structure of values, concerns, interests, and enthusiasms – or even, indeed, one’s entire character or subjectivity (ibid.). Paul argues that the concept of a transformative experience challenges certain culturally entrenched conceptions of the nature of human agency. Specifically, ones that tend to presuppose that our choices are typically performed through practices of rational appraisal and decision that proceed against the background characterised by, inter alia, agential autonomy, informed understanding, and situational stability. In the case of Paul’s interest in models of rational decision making, for instance, agents are conceived as prospectively identifying the possible outcomes of a decision, assigning values to them, and finally weighing those outcomes against her preferences, given their likelihood. Such a sanguine vision is challenged by the phenomenon of transformative experience. The person contemplating undergoing the experience may be quite different to the person who emerges from the experience. ‘We only learn what we need to know’, remarks Paul, ‘after we’ve done it, and we change ourselves in the process of doing it’ (Paul 2014, p. 4). If so, those conceptions of human agency will be significantly challenged by the phenomenon of transformative experience. Many experiences of suffering certainly share the sorts of features of interest to Paul. When one is suffering, for instance due to serious somatic illness, one can be significantly transformed by the illness experience. For example, one’s sense of their own body as naturally compliant and taken for granted can be profoundly changed by illness (Carel 2018). Similarly, the strength and stability of one’s capacities for practical and epistemic agency are often curtailed; one is less capable of autonomous agency and epistemic practice, and one’s bodily and social circumstances become unstable and frightening. This is a transformative experience of illness (Carel, Kidd and Pettigrew 2016). We suggest that suffering experiences can not only be transformative but can be – and often are – potentially more transformative. Before making that claim, though, we note that many persons will suppose that they have a fairly good idea about what this kind of suffering is like, even if they have not, themselves, actually experienced it. One might, for instance, extrapolate from minor bouts of illness, such as the flu, to more serious cases of illness – by extending the timescale, worsening the pain, and so on. Such imaginative

Suffering as transformative experience


exercises are actually far more limited than people suppose (see Carel 2016a, chs. 3–5), and they anyway ignore something emphasised by Paul: the intrinsic opacity of transformative experiences, the fact that we cannot know in advance what they will be like for us, or how they will affect our selfunderstanding, or even know whether whatever sense we do eventually make of them will be congruent with the sorts of understanding one currently enjoys. In Paul’s own words: The problem is pressing because many of life’s big personal decisions are like this: they involve the choice to undergo a dramatically new experience that will change your life in important ways […] But as it turns out […] many of these big decisions involve choices to have experiences that teach us things we cannot know about from any other source but the experience itself. (Paul 2014, p. 3) Paul’s own example, that of choosing to become a parent, has raised much interest because of her claim that this choice cannot be made rationally (e.g. Barnes 2015; Paul 2014, 2015a, 2015b; Harman 2015). This is a paradigmatic example of a transformative experience, although it is one of a largely positive nature. Earlier in her book, though, Paul mentions other types of negative and involuntary transformative experience, including ‘experiencing a horrific physical attack’, ‘having a traumatic accident’, ‘undergoing major surgery’, ‘experiencing the death of a parent’, and ‘experiencing the death of a child’ (Paul 2014, p. 16). Here we want to focus on these unelected transformative experiences, since we believe that most experiences of suffering tend not to be ones that the person chooses or decides to undergo, and so lie outside the framework developed by Paul, in which the emphasis on choice takes centre stage. Consider, for instance, how rarely one would ever want to elect to experience chronic illness, violent accident, trauma, separation, war, political upheaval, becoming a refugee, losing a loved one … the list is depressingly long and too many persons in our world are subjected to them, without ever choosing to undergo such experiences. If David Benatar is right, our unavoidable subjection to such awful experiences is constitutive of ‘the human predicament’ (Benatar 2017). While Paul focuses on voluntary (i.e. elected) and therefore positive transformative experiences, we suggest that many experiences of suffering come under the two categories we’ll develop, which we label non-voluntary and involuntary transformative experiences. Such categories can help us to better understand competing claims about the complexly valanced character of certain experiences of suffering. We also want to show that the epistemic and personal transformation that can be brought about by some suffering can be philosophically valuable in ways that don’t usually justify suffering but do allow us to see that it has intrinsic meaning stemming from the personal growth resultant from the experience. We elaborate on what we take to be


Havi Carel, Ian James Kidd

philosophically valuable about suffering experiences in the final section of this chapter. Two caveats before we begin. The first is that we do not accept that suffering should be endured unnecessarily or be actively sought or exacerbated in order to achieve these edifying effects (Kidd 2012; Carel 2016b). Suffering across its forms ought to be actively reduced, mitigated, and ameliorated, not glossed over, romanticised, or ‘bright-sided’. The ontological facts of human mortality mean that being ill is built into our natures, a constitutive feature of our mode of being. This constitutive mortality inevitably imposes significant limitations on the extent to which certain forms of suffering, such as that result from illness, could be substantially reduced (for instance, everyone gets sick, but not everyone has access to high quality healthcare, free at the point of use, that would mitigate their suffering). But there is a huge space between the exultant embrace of suffering urged by Nietzsche and a crass quietism that laments that suffering is inevitable and therefore only to be accepted. We can try to find ways of acknowledging the inevitability of suffering while still trying to find ways to cope with it that steer clear of both darkly pessimistic acquiescence and distorting ‘bright-siding’. Second, suffering does not always lead to personal growth. People who have suffered trauma often describe a world that has been permanently shattered. Some cases of suffering will cause irreparable damage that will not be a source of growth, but of permanent disablement and distrust. Holocaust survivors are a paradigm example of this, as are other sufferers of posttraumatic stress disorder. Some cases of suffering will be of the kind described by Arthur Frank (1995) as ‘restitution narratives’, where the endpoint of the narrative is one in which the person is somehow better off than she was prior to suffering. But other cases will be too chaotic, painful or traumatic to form a narrative at all, tending to destroy ‘the meaning-making and meaningsharing capacities of the victims of such harm’ – a grim state that José Medina calls ‘hermeneutical death’ (Medina 2017, 47). The violence of suffering, its eruption and destructive force cannot be understated; the ideas set out below about potentially edifying effects of suffering are in no way meant to belittle this.

The facts of life Human life takes many forms; but they share certain features. Often the features on which philosophers focus are positive – reason, autonomy, moral sophistication, virtue. But there are other, less attractive, features of human life. We are subject to a variety of interacting contingencies – bodily, social, and practical factors that can, and often do, fail us in deep and troubling ways. As embodied beings, we are destined to injury, illness, and ageing. As social creatures, ones disposed to form communities and engage in shared practices with others, many of us suffer abuse, exploitation, and oppression. As initiators of projects that aim to give order and meaning to our life, we face the prospect of their failing or

Suffering as transformative experience


going wrong in ways we did not expect and cannot correct. All of these can be amplified by chronic illness (see Kidd 2019). Gathering these together, we can say that three fundamental features of human life are contingency, vulnerability, and subjection. They are not episodic features of the lives of certain unfortunate human beings, but universal features of human life – they are existentialia, in Heidegger’s term, essential features of creatures with our ‘manner of being’ (Heidegger 1996). Alasdair MacIntyre expresses these features of the human condition: We human beings are vulnerable to many kinds of affliction and most of us are at some time afflicted by serious ills. How we cope is only in small part up to us. It is most often to others that we owe our survival, let alone our flourishing, as we encounter bodily illness and injury, inadequate nutrition, mental defect and disturbance, and human aggression and neglect. (MacIntyre 1999, p. 1) MacIntyre is concerned with how we can respond to significant life events, what virtues or qualities are displayed in these responses, and how an active awareness of these facts of life should transform our thinking: These two related sets of facts, those concerning our vulnerabilities and afflictions and those concerning the extent of our dependence on particular others are so evidently of singular importance that it might seem that no account of the human condition whose authors hoped to achieve credibility could avoid giving them a central place. (MacIntyre 1999, p. 1) We suggest that these features of human life indicate a ubiquity of suffering. If our epistemic and practical agency takes place against a complex background of contingency, vulnerability, and subjection, then few of us enjoy anything like freedom from suffering. Thus, it seems that many of the major experiences of the lives of most people are not elected – we do not select them but are subjected to them. Suffering experiences are no exception to that. In addition, a person might elect to seek out a positive transformative experience, such as having a child, only to find themselves subjected to an unelected type of transformative experience, such as grief about a miscarriage. We thus suggest that human agency is better conceived of as deeply conditioned by complexly dappled sets of circumstances, such as the constant possibility of our subjection to undesired situations, and the absence of optimal conditions for properly epistemically procedural deliberation. These features are evident in many cases of transformative experience, not all of which need or can be elected by the agent. These features are also present in all cases of suffering we could think of, although there may be others. The possibility of transformative experiences other than voluntary ones is a consequence of another ‘fact of life’, namely, the contingent, unchosen character


Havi Carel, Ian James Kidd

of most of the material and social conditions of an individual life. Many of these experiences are not, could not and would not be chosen. Note, too, that a person’s subjection to these sorts of experiences is shaped by social, economic, geographical, and cultural factors which vary within and across societies. Whether at the level of an individual life or the wider course of societies and cultures, far less of life is the outcome of careful deliberation and decision than is appreciated, not least since reflection on such pervasive contingency can induce a disturbing sense of vulnerability, vividly characterised by various existentialists (see Cooper 1999, ch. 8). Karl Jaspers, for instance, speaks of a ‘metaphysical fear’, arising during what he calls ‘boundary situations’ marked by acute personal challenge, which render null our typical values and expectations, forcing one to ‘decide for oneself how to respond’ (Jaspers 1932). A similarly acute sense of contingency and subjection flows through the writings of Maurice Merleau-Ponty: a person cannot ‘choose [themselves] from nothing at all’, in the sense of being able to ‘transform’ their values, commitments, and projects at will. For our reflections and choices always presuppose a ‘previous acquisition’, like a sense of what matters, something inherited from a surrounding world that we did not choose and over which we have little power (Merleau-Ponty 1962: 447, 452, 191ff.). The Buddhist tradition, too, emphasises the ‘dependent co-origination’ (paticca-samutpada) of things in the world, oneself included. Understanding the deep dependence and contingency that marks human life is an epistemic achievement, and honestly acknowledging it in one’s ways of living is a personal and moral achievement – a double feat that makes reflectively responding to the contingency of human life a ‘Noble Truth’ (see Harvey 2013, pp. 52–53). With this framework of the ‘facts of life’ in place, we now turn to study a specific type of transformative experiences, namely, those that are unelected.

Unelected transformative experiences The major life experiences we call attention to here as transformative experiences are primarily ones of suffering. These are transformative experiences that are negative, sprung upon us, and feared, even if we cannot avoid many of them. They are not captured by a voluntaristic and robustly active sense of agency, and this for several reasons. First, no reasonable person would willingly choose to undergo these experiences – for instance, to become a victim of violent crime, or to become seriously ill. These are highly negative experiences that may, of course, accrue some goods as a secondary outcome (Kidd 2012; Carel 2016a, ch. 6), but could not be desired by any informed competent person. Second, these sorts of experiences are not of a sort that people are invited to consider and choose to have. Trauma victims are victims in part because they have the experience forced upon them: they are subjected to the traumatic experience – a main reason trauma is so awful. This is a deep lesson depicted, argues Martha Nussbaum, by ancient Greek tragedy, which shows how people can be ‘ruined’ by ‘things that just happen

Suffering as transformative experience


to them’. Examples will include agents who experience awful ‘circumstances whose origin does not lie with them’, and the myriad of ‘things that they do not control’ – injustice, unfairness, everyday vices – which flow from the ‘ungoverned contingency [of] social life’ (Nussbaum 2001, pp. 25, 89). Take the case of Oedipus, who, by his efforts to avoid the prophecy about killing his father and sleeping with his mother, ends up fulfilling it. This illustrates a general truth about one possible source of suffering: any decision made in the thick of life could be thwarted, twisted, or otherwise frustrated by a variety of forces, many of them beyond our knowledge and control. As a rule, people tend to exaggerate the number of things that are ‘within our power’, a tendency lamented from the Stoics to contemporary existentialists (compare, for instance, Epictetus’s Enchiridion and de Beauvoir’s Force of Circumstance). We therefore suggest that the scope of transformative experiences should be expanded to capture cases where the experience (a) results from subjection rather than election and (b) is not one that a reasonable person could elect to undergo. We suggest that these kinds of transformative experiences are mostly ones of suffering. To Paul’s detailed analysis of one type of transformative experiences, which we will call voluntary, we add involuntary and non-voluntary transformative experience to create a broader taxonomy, more typical of human life and one that considers the ‘facts of life’ discussed above. All three types, we suggest, can be genuinely transformative, although our contribution is do develop the latter two types, given that the first type has been greatly discussed in the transformative experience literature. Consider the following examples: Non-voluntary At the age of 24, Primo Levi was arrested in Italy and sent to the Nazi concentration camp at Auschwitz on the German–Polish border, where he survived for twenty-two months. The experience left an indelible mark on Levi. The experience meets Paul’s dual criteria: it was epistemically transformative, as it taught Levi what starvation, imprisonment, torture, cruelty, total lack of control, severe illness, deprivation and degradation are. It was also personally transformative as he himself says: ‘I do not and cannot know what I would be today if I had not been in the camp […] this would be, precisely, a case of describing a future that never took place’ (Levi 1979, p. 397). Clearly, Levi’s experience was transformative, and not of a sort anyone would ever choose. Now consider a second example: Involuntary I see a young child run into the path of an oncoming car. Instinctively, I rush into the road, pick up the child, but the car hits me. The child survives unharmed, but I am severely injured, losing the ability to walk. In this case, I am causally responsible for the outcome, but did not intend


Havi Carel, Ian James Kidd it. The experience is also transformative: my new disability changes who I am, my future preferences, and what I know about myself and life. I would not have chosen to become disabled, but the choice to save the child was taken by me. I am transformed, but through an unintended consequence of a choice I make.2

We suggest that many, if not most, types of transformative experience are involuntary, imposed upon us by the contingencies of life, or non-voluntary (i.e. entirely unchosen). As described above, many of the situations in which people find themselves are ones that no one would choose. And yet, they can be and often are the rule rather than the exception. For example, most of us will die of a serious illness, and yet it is difficult, if not impossible, to imagine that a reasonable person, even in an epistemically ideal situation, could want to become seriously ill. Such contingencies are not rare and unusual cases but commonplace and a fundamental experiential currency through which we live. It may be helpful to take a life-cycle point of view here, which includes the different stages of life, from infancy and dependence, to adulthood and eventually dependence again to see how common these contingencies are. We suggest that this life cycle perspective can reveal susceptibility to transformative experience as belonging to ordinary lives and as ordinary modes of experience (cf. Carel 2016b). Given the framework developed so far, we suggest a double expansion of Paul’s notion of choice. First, many common and exemplary types of transformative experiences, such as those arising from serious illness, are not the outcome, and could not conceivably be the outcome, of a rational choice. Second, a coerced choice may be transformative even when devoid of all characteristics of choice. Certain choices can also be transformative, as noted by Ruth Chang (2015), such as Gaugin’s choice to abandon his family for the sake of his art. In short, coercive and unchosen situations may be transformative. Let us continue to develop our taxonomy, based on bivalent notions of epistemic transformation and personal transformation. We suggest that each of these modes of transformation can be either positively or negatively valanced, such that transformative experiences could take one of four general forms. Certain experiences will, of course, be too complex or indeterminate to be so easily classifiable. Moreover, we think there are certain ambivalent transformative experiences, for which it’s difficult to see whether the transformation was positive or negative. We therefore distinguish: 1. 2. 3. 4.

Positive epistemic transformation, positive personal transformation. Negative epistemic transformation, positive personal transformation. Positive epistemic transformation, negative personal transformation. Negative epistemic transformation, negative personal transformation.

Since we have developed a fuller account of these elsewhere, our focus is on forms 3 and 4, as those most salient to experiences of suffering (Carel and Kidd forthcoming).

Suffering as transformative experience


Positive epistemic transformation, negative personal transformation – in such cases one learns new things by having a new experience, but the personal transformation is negative. Consider again the earlier quotation from Primo Levi. During his imprisonment and torture, Levi learned everything there is to know about cruelty and degradation. He has certainly acquired deep knowledge of aspects of human existence he would not have known otherwise, so his experiences in the concentration camp were clearly epistemically transformative. Levi himself evaluates his experience, remarkably, at least partially epistemically positively, commenting that ‘in its totality this past has made me richer and surer’ (Levi 1979, p.398). He was also personally transformed by those experiences, but in ways that were deeply negative. Negative epistemic transformation, negative personal transformation – In these cases there is a loss of knowledge accompanied by a negative personal transformation. Take a person suffering from dementia: as her disease progresses, she remembers less and is less aware of her surroundings. She loses language and memory. She is confused and behaves erratically due to her neurodegenerative impairment. This is also a tragically negative personal transformation of a most radical sort. The past person is gradually erased and the new minimal person is reduced to the remains of the once vibrant person. We thus see that transformative experiences can be both negatively and positively valanced and can be both voluntary and non- or involuntary. We suggest that such an expansion of the notion of transformative experiences can be useful for capturing suffering experiences, to which we now turn.

Transformative experiences and suffering Our claim in this section is that experiences of suffering are very often cases of unchosen transformative experiences of either an involuntary or nonvoluntary character. They can also include cases of a voluntary transformative experience that goes awry, such that it turns into a non-voluntary transformative experience: think of the case of deciding to have a child but losing the pregnancy. Our starting point is Michael Brady’s definition of suffering, taken from his recent book, Suffering and Virtue: [A] subject suffers when and only when she has (i) an unpleasant experience, consisting of a sensation S and a desire that S not be occurring, and (ii) an occurrent desire that this unpleasant experience not be occurring. (Brady 2018, p. 55) On this account, suffering is composed of a sensation of some kind – which need not be pain – coupled to a desire that this sensation not be occurring. An experience of suffering is therefore a sensation plus a desire, such that ‘the unpleasantness of some experience is a relational property’ (ibid., p. 55). I am


Havi Carel, Ian James Kidd

suffering if there is a spreading ache in my lower back that’s sufficiently painful and distracting that I start to desire that the sensation ceases – so I might get up and stretch, or take a painkiller, or determine to tarry on but think about seeking physiotherapy. Transformative suffering naturally involves a range of unpleasant experiences, such as sensations of anxiety, dislocation, displeasure, uncertainty, vulnerability, and of course pain, plus the occurrent desire that these experiences not be occurring. In cases of unelected experiences of suffering, one might be painfully aware that one can do little, or nothing, to make the experience cease. Indeed, one reason many people would not elect to undergo an experience of suffering will be the knowledge that, once started, the experience either cannot be stopped (the case of torture, say) or that the experience would cause forms of harm that will persist even after the experience ceases (suffering physical mutilation). The question here is what do such unpleasant experiences contain in them that is transformative? The first clue can be found in Brady’s definition. Suffering is not simply an unpleasant sensation, but an unpleasant experience, such that the experience is coupled with a desire for it to cease. This points to the first feature of suffering that makes it transformative, namely, that it is an experience. As an experience, suffering gives us new information about what having certain sensations (e.g. pain, sadness, grief, despair, extreme fatigue, etc.) is like, while they stand in a particular relation to the desire that they stop. Merely having some sensation without the desire for it to cease will not constitute suffering. This stands to reason, as a sensation that may be unpleasant but that one is not moved to try to eliminate is not suffering, as Brady would claim. This is a further point in support of our view that unelected experiences can be and are transformative. We further suggest that the accompanying desire for its cessation points to the intensity, novelty and attentional focus which characterise suffering, each of which concern deep aspects of what it means to suffer. One feels disappointment when dropping one’s ice cream on the ground – although that feeling is unpleasant, it lacks the intensity typical of suffering (such disappointment won’t creep deep into our hearts, as the dull existential ache that will crumple every future moment of joy). One’s attention will also soon move away to something else – the sun sparkling on the sea, or a joke. Moreover, unless one is a small child, such disappointment is also unlikely to be novel – one will have dropped one’s food or spilled one’s drinks before, whereas everyday experiences of bodily health do not prepare one for the feeling of bodily doubt characteristic of chronic somatic illness (Carel 2016a, ch. 4). Experiences of suffering, then, are characterised by an intensity, novelty, and attentional focus that marks them out from other types of negatively valenced experiences. It is easy to see how the experience of a serious accident will be personally transformative. The intensity and novelty provide an experience unlike my previous ones, and indeed one that is likely to change my values and goals in a

Suffering as transformative experience


profound way. The structure of one’s experience – of time, social spaces, one’s body – all are transformed in fundamental and irreversible ways: one now suffers chronic back pain; one’s mobility is restricted; one’s sole focus is getting rid of the pain – and so on. The attentional focus demands that one give cognitive and emotional as well as practical attention to the accident and its consequences, thinking about it during one’s time in hospital, then again during rehab, then coping with an emerging realisation that the pain will become a permanent feature of one’s life. It is also easy to see how the experience is epistemically transformative. One learns new things about one’s body, how it behaves when damaged, and the variety of forms that experiences of pain can take. One learns things about hospitals, medical treatments, and the culture of modern healthcare and comes to understand the rhythms and rules of the health service. One also gains insights into the psychological reality of profound fear, grasps the complexity of what it means to understand one’s own mortality, and so on. Such epistemic goods could not be accessed before the experience, since they are premised upon changes to one’s lived experience – of, for instance, what it means to understand one’s own damaged body. We claim that it is in virtue of the three features – intensity, novelty, and attentional focus – that the accident is a transformative experience. Note these features are neutral – intensity can be good, too, as can novelty and attentional focus – and, moreover, that they help explain what makes an experience transformative. Where they are negatively valanced, they will also indicate cases of suffering.

What can suffering tell us about transformative experience? We want to conclude by returning to a feature of Paul’s original account, namely, her focus on elected transformative experiences. Although these are clearly genuine and important, our claim is that the complex dependencies and vulnerabilities characteristic of our natures as embodied social beings mean that subjection to unelected and involuntary forms of transformative experiences is far more characteristic of human life. Nowhere is this clearer than in experiences of suffering, which are, after all, prevalent to the point of ubiquity across the range of forms of human life – hence Benatar’s emphasis on the ‘awfulness’ built into ‘the human predicament’ (Benatar 2017). An important feature of negative experiences is that they are perhaps much more transformative than positive experiences, given that one also must make narrative sense of both the nature of the experience and its non-voluntary or involuntary nature. Such experiences might transform us in deeper ways than voluntary transformative experiences because we have to deal with the epistemic, personal and practical changes, as well as with their involuntary or forced nature. For example, when nine-year old Anna is forced to leave Germany to travel to Switzerland in 1933 because she and her family are Jewish and fear for their lives, she imagines this trip as a holiday, an adventure – in other words, as an elected experience (Kerr 2017). She not only needs to


Havi Carel, Ian James Kidd

adjust to life abroad, leaving her home and friends and so on, but also needs to make sense of her family’s escape due to imminent threat to their lives, which requires further cognitive and emotional labour from her, such as the acquisition and mastery of such complex concepts as injustice and antisemitism. It may be that positive narratives are less disruptive in this sense, since they entail less onerous and depressing moral, cognitive, and emotional labour (think, for example, of the ‘burnout’ that afflicts social justice activists, due to the physically and psychologically corrosive effects of a sustained engagement with feelings of anger, injustice, and other negative affects). A second reason for thinking that negative experiences are significant transformative experiences is that suffering transforms views about the structural context of the experience in ways that voluntary choices might not. A voluntary experience presupposes a social and practical context hospitable to exercises of self-directed agency, a structure of options and effective means for their realisation. But most experiences of suffering tend to be non-voluntary and involuntary, meaning that they are imposed upon the agent without their choosing them and invariably against their preferences. When an experience is forced upon us, that is itself part of its negativity – a sense that ‘I didn’t want this’, ‘it just erupted into my life’, and so on. When a person has consistently enjoyed good health, for instance, the sudden experiences of acute bodily transformation occasioned by a diagnosis of chronic illness comes as a shock – puncturing a comfortable sense of obliviousness and complacency in ways that are negatively personally and epistemically transformative. This suggests that by focusing on the voluntary the current literature on transformative experience neglects perhaps the greatest source of transformation, namely negative experiences. A final thought is that negative experiences – and especially suffering – can often have an edifying effect that also makes them particularly good vehicles of transformation. At its broadest, experiences of suffering are edifying when they afford opportunities for the cultivation and exercise of various positive attainments, such as moral virtues, wisdom, and a sense of existential perspective. Such edification can easily take trite forms, of course, like the Nietzschean sentiment (actually originating with Goethe) that ‘whatever doesn’t kill you only makes you stronger’, to the facile ‘bright-siding’ insistence that suffering always ends in triumph rather than despair, that battles are won but never lost, and so on. It is a deep truth that suffering often does not make us stronger or ‘deeper’, that one’s losses are not always followed by compensating gains, and that there are sometimes no ‘bright sides’, that what Jonathan Haidt (2006) calls posttraumatic growth is a possibility and not an inevitability, and so on (see Kirkengen 2010, Marmot 2016). Such truthfulness about the many ‘dark sides’ of experiences of suffering are still consistent, however, with a confidence that in certain cases and in the right context there can be philosophically valuable lessons that can be found in suffering. A properly balanced edificationist conception of suffering has the complex conditional form that certain experiences of suffering can afford certain goods and attainments for certain persons and under certain material

Suffering as transformative experience


and social conditions. The complexly conditional character of edification reflects what we earlier called ‘the facts of life’ – for instance, our dependence on other agents means that our access to the practical, moral, and emotional support cannot always be guaranteed. Many chronically ill persons report that, after their diagnosis, some friends stop calling, for instance, causing sadness and bitterness as well as removing a source of the comfort and support that a good friend ought to provide, an example of what Martha Nussbaum (2001) refers to as ‘the fragility of goodness’. The contingently available goods of edifying experiences of suffering can include a capacity to accept failure and disappointment, the ability to live in the present, to revise one’s life goals in the face of unwarranted calamity, and to co-exist with limitation, illness, disability, bereavement and loss (Carel 2016a, 2016b, 2018; Kidd 2012). Many of these concern the complex ability to develop authentic ways of appreciating and coping with one’s embodied vulnerability and subjection to materially and socially conditioned suffering and to bad luck. Since vulnerabilities and conditions are intrinsic to human life, and a main generator of the non-voluntary and involuntary transformative experiences described above. Although these abilities could be gained through other sorts of experiences, such as engagement with certain forms of literature, there are special reasons why they can be especially gained through first-person experiences of suffering. First, we may be unable to imagine the reality of a particular type of suffering without the requisite sorts of lived experience. Second, we may lack the empirical knowledge about the nature or character of an experience, especially if testimonial reports about it are hard to come by. Third, even if we suffered similarly in the past, we are notoriously bad at remembering negative events, and good at reconstructing our memories of events in distorting ways (Gilbert 2006). Here is an example: We may think we can reliably anticipate the misery of solitary confinement, by imagining how lonely we might feel and how bored we would become. Perhaps if we watch Marc Rocco’s Murder in the First we think we understand what it is like. In fact, the effects of solitary confinement include disruption of bodily sensory-motor functions, of a sort that a lay person would find hard to imagine and impossible to anticipate (Gallagher 2014). One cannot know in advance what sorts of transformations will or might follow from that sort of experience, however hard one might try. Many questions remain for further work on the relationship between suffering and transformative experience. Are there other features of suffering that further explain the link between suffering and edification? Why might pleasant experiences be less conducive to such edification? Are unelected and negative transformative experiences more central to human life than those that Paul discusses? We think so, since the conditions for voluntary transformative experience – such as socially facilitated agency – are less common and more fragile than one would like. If so, should we shift focus towards the negative and involuntary rather than the voluntary and positively life-changing experiences? Since we are dependent, afflicted, vulnerable creatures inhabiting


Havi Carel, Ian James Kidd

a deeply imperfect social world, what might be really primary is our susceptibility to negative involuntary and non-voluntary experiences that transform us personally and epistemically in disturbing ways. Suffering could therefore be a form of transformation imposed upon us by our fragile bodies, unjust societies, and the gritty contingencies of the world. Finding ways of appreciating and coping with that fact is a primary existential task for human beings.

Notes 1 The wider literature on the concept includes symposia in Res Philosophica 92:2 and Philosophy and Phenomenological Research 91:3, and several other writings (Pettigrew 2016; Talbot 2016; Shupe 2016; Cappelen and Dever 2017). 2 For an extreme account of such transformation by disability, see Jean-Dominic Bauby’s account of his stroke and the resulting ‘locked-in syndrome’ he suffered, described in his memoir, The Diving Bell and the Butterfly (Bauby 1998).

Acknowledgements We are grateful to David Bain and Michael Brady for inviting us to contribute to the volume and for providing helpful comments on the paper. Havi Carel is also grateful to the Wellcome Trust for funding her research for this chapter (grant number 103340).

References Bardi, A. et al. 2014. Value stability and change during self-chosen life transitions: self-selection vs. socialization effects. Journal of Personality and Social Psychology, 106. 1: 131–147. Barnes, E. 2015. What you can expect when you don’t want to be expecting. Philosophy and Phenomenological Research 91. 3: 775–786. Bauby, J.-D. 1998. The Diving Bell and the Butterfly. London: Vintage Books. Benatar, D. 2017. The Human Predicament: A Candid Guide to Life’s Biggest Questions. Oxford: Oxford University Press. Brady, M. 2018. Suffering and Virtue. Oxford: Oxford University Press. Campbell, J. 2015. L. A. Paul’s Transformative Experience . Philosophy and Phenomenological Research 91. 3: 787–793. Cappelen, H. and J. Dever. 2017. Empathy and transformative experience without the first-person point of view (a reply to L. A. Paul). Inquiry 60. 3: 315–336. Carel, H. 2016a. Phenomenology of Illness. Oxford: Oxford University Press. Carel, H. 2016b. Virtue without excellence, excellent without health. Aristotelian Society Supplementary Volume 90. 1: 237–253. Carel, H. 2018. Illness: The Cry of the Flesh, 3rd edition. London: Routledge. Carel, H. and I. J. Kidd. Forthcoming. Expanding transformative experience. European Journal of Philosophy. Carel, H., I. J. Kidd and R. Pettigrew. 2016. Illness as transformative experience. The Lancet 388. 10050: 1152–1153. Chang, R. 2015. Transformative choices. Res Philosophica 92. 2: 237–282.

Suffering as transformative experience


Cooper, D. E. 1999. Existentialism: A Reconstruction, 2nd edition. Oxford: WileyBlackwell. Ehrenreich, E. 2009. Smile or Die: How Positive Thinking Fooled America and the World. London: Granta. Frank, A. 1995. The Wounded Storyteller. Chicago, IL: University of Chicago Press. Gallagher, S. 2014. The cruel and unusual phenomenology of solitary confinement. Frontiers in Psychology 5: 585. Gilbert, D. 2006. Stumbling on Happiness. London: Harper Press. Haidt, J. 2006. The Happiness Hypothesis. New York: Arrow Books. Harman, E. 2015. Transformative experiences and reliance on moral testimony. Res Philosophica 92. 2: 323–339. Harvey, P. 2013. Dukkha, non‐self, and the teaching on the four ‘noble truths’. In S. Emmanuel (ed.), A Companion to Buddhist Philosophy. Oxford: John Wiley & Sons. Heidegger, M. 1996. Being and Time: A Translation of Sein und Zeit, trans. J. Stambaugh. New York: SUNY Press. Jaspers, K. 1932/1969. Introduction. In K. Jaspers, Philosophy, vol. I, trans E. Ashton, pp. 43–98. Chicago, IL: University of Chicago Press. Kerr, J. 2017. When Hitler Stole Pink Rabbit. London: Harper Collins. Kidd, I. J. 2012. Can illness be edifying? Inquiry, 55. 5: 496–520. Kidd, I. J. 2018. Adversity, wisdom, and exemplarism. Journal of Value Inquiry, 52. 4: 379–393. Kidd, I. J. 2019. Pathophobia, Illness, and Vices. International Journal of Philosophical Studies, 27. 2: 286–306. Kirkengen, A. L. 2010. The Lived Experience of Violation: How Abused Children Become Unhealthy Adults. Bucharest: Zeta Books. Levi, P. 1979. If This Is a Man & The Truce. London: Abacus. MacIntyre, A. 1999. Dependent Rational Animals: Why Human Beings Need the Virtues. London: Duckworth. Marmot, M. 2016. The Health Gap. London: Bloomsbury. Medina, J. 2017. Varieties of hermeneutical injustice. In I. Kidd, J. Medina and G. Pohlhaus (eds), The Routledge Handbook of Epistemic Injustice. London: Routledge. Merleau-Ponty, M. 1962. Phenomenology of Perception. London: Routledge. Nietzsche, F. 1974. The Gay Science, trans. W. Kaufman. London: Penguin Books. Nussbaum, M. 2001. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy, updated edition. Cambridge: Cambridge University Press. Paul, L. A. 2014. Transformative Experience. Oxford: Oxford University Press. Paul, L. A. 2015a. Transformative experience: replies to Pettigrew, Barnes and Campbell. Philosophy and Phenomenological Research 91. 3: 794–813. Paul, L. A. 2015b. What you can’t expect when you’re expecting. Res Philosophica 92. 2: 149–170. Paul, L. A. and K. Healy. 2018. Transformative treatments. Nous 52. 2: 320–335. Pettigrew, R. 2016. Review of L. A. Paul’s Transformative Experience. Mind 125. 499: 927–935. Shupe, E. 2016. Transformative experience and the limits of revelation. Philosophical Studies 173: 3119–3132. Styron, W. Sophie’s Choice. Talbot, W. 2016. Review of L.A. Paul, Transformative Experience. Analysis Reviews 76. 3: 380–388. Weil, S. 2010. Waiting on God. Abingdon: Routledge.


After motivational hedonism Feeling bad can be good | feeling good can be bad Colin Wayne Leach

It is widely believed in psychology and other behavioral sciences that feeling “good” serves a general psychological function of motivating us to continue the advantageous state of affairs that has led to our euphoria, whereas feeling “bad” functions to motivate us to discontinue the disadvantageous state of affairs that has led to our dysphoria (for reviews, see Buck, 1999; Frijda, 1986; Lazarus, 1991). Thus, the assumption is that feeling good functions to alert us to an advantageous state of affairs that we should approach (if we know what is good for us) and that feeling bad functions to alert us to a disadvantageous state of affairs that we should avoid (if we know what is bad for us). This view is often advanced today through an increasingly integrated approach to affect, emotion, and motivation that combines cognitive science, neuroscience, psychology, and behavioral economics (for a review, see Niedenthal & Brauer, 2012). It is rooted in an older perspective that can be called motivational hedonism (Moore, 2018) as it assumes that human motivation boils down to a basic desire to approach pleasure and avoid suffering (for discussions, see Bastian et al., 2014; Gray, 1990; Higgins, 1997). Motivational hedonism can be distinguished from ethical hedonism which is the view that only pleasure has value more generally (Moore, 2018). There can be little doubt that feeling bad about a state of affairs is an important basis for being motivated to discontinue it, and vice versa. However, what is generally true is not what is always true (McGuire, 1973). Indeed, what is generally true may give little indication to what is most interesting or important about phenomena. In the case of feeling bad, it is quite functional (i.e., efficient and useful affectively) that one is moved to end one’s affective state as quickly and easily as possible with little concern for the causes or consequences of one’s state. However, the risk to the individual in such motivational hedonism is that it may function to focus one narrowly on what is good for the self emotionally to the exclusion of what is good for the self ethically, socially, politically, or otherwise (for discussions, see Averill, 1982; Solomon, 1993; Tiedens & Leach, 2004). As such, the motivational hedonism in a functionalist perspective on affect leaves little room for the possibility that dysphoric states of affective suffering can sometimes have other value for the individual experiencing them.

After motivational hedonism


Of course, motivational hedonism also tends to neglect the implications that an individual’s dysphoric (or euphoric) state has for individual others, or for the greater good. In other words, presuming that it is functional and advantageous for individuals to approach pleasure and avoid suffering makes it difficult to see how such motivation may benefit or harm the others affected by the individual’s state (for discussions, see Averill, 1982; Tiedens & Leach, 2004). For instance, wanting to feel good, rather than bad, can affect our willingness to acknowledge our mistakes, tell the truth about our disappointments, or empathize with the suffering of others. As such, the perspective of motivational hedonism can obscure the aggregate-level implications that individual’s inter-related affective lives have for the social relations within which human beings are embedded (e.g., families, companies, cities, societies). Indeed, affect is not only an individual phenomenon. Its operation in dyads and larger social aggregates means that the value of feeling good or bad must also be examined at social levels of analysis if we are to properly understand it (Leach, 2016; Parkinson, Fischer, & Manstead, 2005; Tiedens & Leach, 2004). Instances in which feeling bad can do good offer unique insight into feelings, and into what is good and bad. The same is true of good feelings that do bad. Thus, in this chapter, I discuss my own and other’s work on social approaches to emotion to show that feeling bad can do us and others good (e.g., shame, righteous anger) and that feeling good can do us and others bad (e.g., schadenfreude). In the case of shame, for instance, this state of profoundly unpleasant self-criticism can move people to want to improve themselves and their social relations when the intense dysphoria is interpreted as a sign that improvement is necessary and possible. In schadenfreude, for example, the euphoria of feeling good about a bad event befalling another may debase one’s character in addition to harming one’s relation to the unfortunate. In shame, righteous anger, and schadenfreude, we see that the value of emotion (personal and social) has less to do with the hedonics of the state itself and much more to do with the psychological and social implications of the emotion as a way of “being in the world” (as Sartre would have it).

Motivational hedonism, and after It is important to start with the obvious (and necessary) point that the terms good and bad, positive and negative, are pregnant with normative meaning even when they are used to describe affective states (Solomon, 1993). Feeling bad about an event in one’s life suggests more than simple dysphoria. In addition to implying the (un)pleasant quality of affective experience, the terms good and bad, positive and negative, imply that affective states are good or bad for oneself (i.e., psychologically functional for the individual) and good or bad for others and/or the world in general (i.e., moral, ethical). Thus, it is widely presumed that feeling good is good for us, and often does good in the world (for discussions, see Frijda, 1986; Lazarus, 1991). This


Colin Wayne Leach

conflation of at least three different senses of good/positive or bad/negative feeling is likely due to the motivational hedonism assumptions implicit or explicit in much western theorizing of affect, emotion, and motivation (for discussions, see Arnold, 1960; Solomon, 1993). The motivational hedonism assumption is that people wish to approach pleasure and to avoid pain and suffering (for reviews, see Bastian et al., 2014; Gray, 1990; Higgins, 1997). This is facile functionalism at work, as we would be quite irrational creatures, the argument goes, if we didn’t approach what is good for us and avoid what is bad for us. And, if people are, by and large, ethically concerned, then the pleasure they wish to approach should be morally good and the suffering they wish to avoid should be morally bad. Numerous overarching theories of human nature can work to explain why feeling bad is bad for us and bad morally (for reviews, see Higgins, 1997; Lazarus, 1991; Solomon, 1993). Rational actors, evolutionary survivors, and self-esteem seekers should all end up with this confluence of affective valence, functionality, and ethics. And, to the degree to which it is assumed that biological, psychological, and social (moral) processes are intertwined in some single grand design which functions to unite human motivation, the confluence of affective valence, functionality, and ethics may be presumed as necessary and efficient (for discussions, see Barrett, 2006; Buck, 1999; Solomon, 1993; Tiedens & Leach, 2004). Despite the widespread influence of implicit or explicit motivational hedonism in theories of satisfaction and suffering (see Bastian et al., 2014; Higgins, 1997), it is clear upon reflection that these three senses of good-bad can and should be distinguished from one another. The simple truth is that terms like good and bad, positive and negative, are poor descriptions of the (un)pleasantness of affective experience, precisely because they imply something more normative than is necessary (Solomon, 1993). To “feel good” is to experience a pleasant state of euphoria, whereas to “feel bad” is to experience an unpleasant state of dysphoria. When considered simply in terms of affective (un)pleasantness, there is no necessary assumption that people generally wish to feel “good” or “bad” or that either state has consequences that are good or bad for the person, for others or for society at large. When good and bad are used simply as descriptive characterizations of an affective state of (un)pleasantness there are no functional or moral implications. Thus, it is more precise and more prudent to refer to this dimension of affective states as the (un)pleasantness of the experience, or what many behavioral science approaches to affect and emotion refer to as hedonic valence (for reviews, see Ortony, Clore, & Collins, 1988; Scherer, Schorr, & Jonhstone, 2001). After motivational hedonism, we are freer to consider the many other possible functions that (un)pleasant affective states may serve for individuals and for social relations (Leach, 2016, 2017). In terms of social relations, feeling good can be bad for relationships even if it is pleasant affectively. As I discuss below, the malicious satisfaction of schadenfreude is one example, whereby one takes pleasure in the misfortune of another. The pleasure of schadenfreude necessarily places one in an antagonistic relation with the less

After motivational hedonism


fortunate and thereby limits more pro-social states like empathy and compassion (e.g., Leach, Spears, Branscombe, & Doosje, 2003; for empirical reviews, see Smith, 2000; van Dijk & Ouwerkerk, 2014). Freeing our thinking about function from the narrow concern with individual benefit inherent to motivational hedonism also allows us to consider the way that unpleasant emotional states like anger and shame may serve to alert us to an unacceptable state of affairs that calls for action in an attempt at improvement (Leach, 2016). Rather than motivating avoidance of a disadvantageous state, emotions like anger and shame can focus us on altering the causes of our dysphoria. In other words, contra motivational hedonism, unpleasant states like anger and shame may encourage approach rather than avoidance. And, thus, their function for individuals may be to mobilize effort at changing circumstances, either by avoidance or approach. After motivational hedonism, we are freer to also consider the ethical or moral implications of (un)pleasant affective states in a way that goes well beyond what is good for the individual hedonically. In the case of schadenfreude, we may ask what implications these pleasant states have for the ethical codes by which we expect people to experience the world. Is it legitimate, kind, or decent to take pleasure in another’s suffering? (see Kristjánsson, 2006; Leach, Spears, & Manstead, 2015). Is it reasonable to be angry about believed injustice and to act in hostile opposition to the presumed unjust? (see Averill, 1982; Leach, 2008). By considering the moral good of affective states, we can move further beyond the individual and experiential concerns of motivational hedonism to examine what feeling bad and feeling good does ethically for the moral aspects of oneself and for the moral dimension of how one treats others and the world at large (for discussions, see Leach, Bilali, & Pagliaro, 2015; Solomon, 1993).

Feeling good can be bad The most prominent examples of pleasant affective states involve earned success, good fortune, or other events that benefit the self and perhaps others (Lazarus, 1991). Pride, joy, and satisfaction are all pleasant experiences because some benefit is bestowed (Leach et al., 2015). The pleasure of schadenfreude is thus noteworthy because one experiences satisfaction at a bad fortune rather than good fortune (Ortony et al., 1988; Smith, 2000). Schadenfreude Feeling good about something bad (that befalls another) is inherently malicious, at some level, as it implies dislike and ill will (Smith, 2000). Indeed, an antagonistic relationship with another that is marked by strong dislike facilitates the dramatic divergence in feeling that is satisfaction at another’s suffering (Leach et al., 2003; Smith, 2000). In schadenfreude, the value of (another’s) suffering is the hedonic pleasure it provides one who wishes to see


Colin Wayne Leach

another suffer (Leach et al., 2003). There are numerous reasons to want to see and enjoy another suffer (for a review, see van Dijk & Ouwerkerk, 2014). Most prominent among them are intense rivalry, a sense of inferiority, and a view of another’s misfortune as deserved comeuppance (Leach et al., 2003). Empirical research in psychology has shown that these subjective judgements can all contribute to various degrees to the pleasure that individuals take in the misfortune of a rival as long as the misfortune is not so extreme that it makes a little pleasure at a mishap seem indecent (for a review, see van Dijk & Ouwerkerk, 2014). Empirical research in psychology has also shown that the actual procedural fairness of the other party’s position or success has little do to with schadenfreude at their fall (Leach & Spears, 2008; Leach et al., 2015). Thus, as Nietzsche (1967/1887) suggested years ago, schadenfreude appears to have little to do with a genuine sense of morality. Instead, as Nietzsche suggested, schadenfreude has more to do with feelings about ourselves and the unfortunate than the apparent or believed deservingness of our fortunes (for reviews, see Leach, Spears, & Manstead, 2014; Leach et al., 2015; van Dijk & Ouwerkerk, 2014). Although “schadenfreude” is now used in English to describe satisfaction at seeing justice done to another who is seen as deserving punishment or some other adverse circumstance, such pleasure is more precisely described by the German term Genugtuung (Leach et al., 2014). The notion that people can gain “satisfaction” from seeing others “pay” for wrongdoing is long-standing in systems of criminal law and tort. In the German legal system, for example, Genugtuung is used to refer to a sense of satisfaction that the proper punishment has been meted out for wrongdoing (Prittwitz, 1999; Stoll, 1970). In many legal systems, punitive damages, restoration, or proper compensation, are designed to redress the harm of a crime by providing victims their satisfaction. We can see the attempt at formal justice here by the guiding principle that the victim’s satisfaction should be proportionate to the harm caused. More generally, justice may be done via the (formal or informal) punishment of a wrongdoer or by a popular recognition of the wrongdoer’s culpability. One may also feel Genugtuung about the vindication of a falsely accused, or maliciously maligned, rightdoer. It is in Genugtuung that we may see some psychological and social value in seeing others “suffer for their sins”, as the pleasure in Genugtuung is about the (retributive or restorative) justice accomplished by the suffering rather than being about the suffering itself. Thus, Genugtuung is a case where feeling good about something bad can be good. The pleasure of Genugtuung may be delivered by a kind of cosmic justice that can be thought of as the “hand of justice,” “karma,” “comeuppance,” or “just deserts”. Of course, one may also take pleasure when witness to authorities or groups punishing wrongdoers on behalf of the community at large. We can imagine that being pleased at seeing justice done serves to reinforce a concern for justice and perhaps even encourages faith in social systems of adjudication and punishment that mete out appropriate justice. This is an important question for future research.

After motivational hedonism


At some level, the emotion of Genugtuung has some similarities to the idea that the “altruistic punishment” of wrongdoers can be experienced as a form of psychological reward or satisfaction. For instance, Singer et al. (2006) found men who witnessed an unfair game player “punished” with shock showed activation in brain regions associated with reward processes. This psychological reward to seeing just suffering may reinforce within the individual a faith in the personal value of fair, or otherwise, moral action and also reinforce a faith in other’s accountability. It is possible that both effects on the individual may reinforce and facilitate cooperation and other ethically-guided action within social systems (Leach et al., 2015).

Feeling bad can be good Feeling good – pleasant affective experience – can be bad. As in the case of schadenfreude, pleasure may be malicious if it is satisfaction at another’s suffering as it sets one against others for petty and pernicious reasons. Feeling good about a bad thing befalling another can also be good, if it is an affirmation of a shared moral code that enables cooperation and trust as in the case of Genugtuung. Thus, consistent with the idea that the (un)pleasantness of emotion is good or bad in a way different from moral, or psychological, goodness, there is no simple one to one relation between feeling and function. This may be most dramatically demonstrated in the case of feeling bad being good (for oneself or others). The motivational hedonism underlying a great deal of thinking about emotion and action has wielded its influence most clearly in the understanding of dysphoric emotions like anger and shame. Because western psychology views pleasant and optimistic feeling and thinking about the self as a key sign of proper psychological function and well-being (for a discussion, see Taylor, Kemeny, Reed, Bower, & Gruenewald, 2000), unpleasant feelings about disadvantageous events such as tragedy or wrongdoing are widely viewed as maladaptive (for discussions, see Bonanno, 2004; Allen & Leach, 2018). No state of dysphoria illustrates this more clearly than the emotion of shame, which has long been considered among the most aversive and destructive emotions (for discussions, see Gausel & Leach, 2011; Leach, 2017). Another challenge to the motivational hedonism that suggests that feeling bad is bad may be found in the emotion of anger which was long considered aversive and thus a counter-productive basis of avoidant or destructive action (for a discussion, see Averill, 1982). When dysphoric emotions like anger and shame are considered outside the constraints of motivational hedonism, we may more easily imagine the many ways in which feeling bad can be good. That is, there are many ways in which unpleasant emotions, like anger and shame, may encourage individuals to think, feel, and act in ways that improve the self or the situation that has led to the dysphoria. In a recent review of research and theory on self-control, Inzlicht and Legault (2014) argue that emotional distress works as an alarm.


Colin Wayne Leach

When the alarm is heeded it orients people to expend effort at the regulation of their cognition, affect, and behavior to more efficiently and effectively exercise the self-control needed to address the alarming circumstance. Thus, emotions are indicative of one’s relationship to the world, to paraphrase Sartre (1948). Unpleasant emotions indicate an unwanted relationship to an unacceptable world and thus motivate efforts at altering the world, or one’s relationship to the world (Leach, 2016, 2017). This is one value of (emotional) suffering.

Shame Shame is widely considered one of the most unpleasant and aversive emotions that human beings can experience (for reviews, see Frijda, 1986; Lazarus, 1991; Tracy, Robins, & Tangney, 2007). This is because shame about one’s physical impairments, moral improprieties, intellectual failings, or social disgraces has long been thought to mark the body or the soul as abject (for reviews, see Gausel & Leach, 2011; Leach, 2017). As such, shame is often considered a sort of stigma that is burned deep into one’s psyche and is also apparent to others through visible signs of one’s wretchedness on the body or otherwise. In case abjection might not be noticed, there are many means by which people can be given “the mark of shame.” Examples include the branding of slaves, sartorial signs of heresy (like the scarlet letter or yellow star), and the dunce cap. Thus, there is a longstanding view of shame as marking individuals inside and out. More formally, we can think of this dual abjection as a view of shame as rooted in a stigma that damages one’s selfimage and one’s social-image (Gausel & Leach, 2011; Leach & Cidam, 2015). Indeed, in a great deal of work on shame, self- and social-image are conflated likely because damage to one is presumed to facilitate and reinforce damage to the other in a vicious cycle of internal and external processes of stigma (for reviews, see Gausel & Leach, 2011; Leach, 2017). In addition to a number of other problematic assumptions, the prevailing view of shame is rooted in the pervasive motivational hedonism that frames feeling bad as bad (for oneself, for others, and for the world at large). Indeed, if it is not escaped, the intense pain of shame is thought to eventually destroy the self in part or in whole, by leading to psychological disorder, self-harm, or even suicide. This is why shame is central to discussions in psychology of depression, low self-worth, and a wide variety of other “dysfunction” (for reviews, see Leach, 2017; Tracy et al., 2007). In “body shame” for example, psychologists imagine a state of deep hatred of the (physical) self that one wishes to obliterate by eating oneself to death or starving oneself to death. Or, in “gay shame,” psychologists imagine a psychological flight from the authentic self so as to avoid aspersions (from the self and others). Thus, here too, an external stigma from others is thought to be internalized and paralleled by an internal stigma of secret self-hate. If not internalized as self-destruction, the pain of shame is thought to be externalized as a rancorous hostility and hatred toward available scapegoats

After motivational hedonism


who can be made to suffer on one’s behalf. This externalization of profound suffering has been called humiliated fury or the shame-rage spiral and it is prominent in psychological theories of rancor, aggression, and violence (Gausel & Leach, 2011). It is not dissimilar to Nietzsche’s notion of ressentiment and more recent uses of this concept to explain revolution and other social upheaval from those lower in station or success in society (for discussions, see Leach, 2008; Leach & Spears, 2008). In the case of ressentiment, a sense of serious personal inadequacy is seen as so painful that an available target is blamed for one’s pain so that one may redirect one’s dysphoria by transforming it into a hostile and rancorous anger toward others (Leach & Spears, 2008). Given the common characterization of shame as a state of profound emotional suffering, we may wonder what value it could have for anyone. The truth is that the motivational hedonism view of shame leaves little room for positive (productive) potential in shame. A more social view of shame, found in social sciences like in sociology and anthropology, is less pessimistic. It focuses on the possible social function that shame alerts individuals that their reputation as a morally decent and capable member of the community is at risk (for a review, see Gausel & Leach, 2011). Based to some degree in primatology, the social view interprets shame as more of a signal of social abasement rather than as a psychological sign of self- abasement. Thus, like other primates, human beings may adopt a supplicant posture – lowered and contracted body; lowered, avoidant head and gaze; facial expression of sadness – to express appeasement. By submitting oneself to the judgment, or punishment, of others, the act of abasement expresses a recognition of failure and an interest in redress (Gausel, Leach, Vignoles, & Brown, 2012). The great apes do it. And, so can we. What is implied in this sort of social shame is that one believes that one can repair the social-image one has put at risk with one’s moral or other failure. To show the signs of appeasement is to expect, or hope, that the powers that be in the community recognize one’s contrite posture and take it as evidence of one’s willingness to make redress and to improve one’s one performance going forward (Gausel & Leach, 2011; Leach & Cidam, 2015). Recent empirical research in psychology has examined the emotional experience of shame rooted in a concern for protecting a damaged socialimage (for a review, see Gausel & Leach, 2011). Leach and Cidam (2015) conducted a quantitative synthesis of this small body of studies. They found that shame was moderately linked to effort at improving the social-image put at risk by a failure under circumstances that suggested that such effort might be effective. Indeed, shameful expressions of appeasement make little sense if one doubts that one can convincingly display appeasement or that the powers that be in one’s community will notice or accept such a gesture. For instance, some violations of community standards are so extreme that shameful abasement is not considered sufficient. Or, one’s pride may prevent convincingly contrite abjection.


Colin Wayne Leach

Although interesting and important, the socio-functional view of shame as appeasement leaves undisturbed the motivational hedonism view of shame. In the absence of a desire to win others over with shame, the common presumption is that a more internally-oriented shame is too psychologically devastating to be useful or sustainable. Because the motivational hedonism view of shame seems so clear about the lack of (individual) value in shame, over the last eight years my colleagues and I have developed an argument for its value to the individual (see Gausel & Leach, 2011; Leach, 2017; Leach & Cidam, 2015). Despite our departure from the prevailing view, we acknowledge that shame is often experienced as a debilitating state of profound selfcriticism that threatens to destroy the self and social relations that highlight the serious failure that caused the shame. In many instances, shame is destructive because there appears to be little option other than to escape the circumstances of one’s devastating failure. The key, however, is in understanding that it makes most sense to destroy oneself or one’s social relations when there appears to be little prospect of improvement. Running away, hiding, wishing to disappear are reasonable responses to shame about a failure that is catastrophic and thereby beyond repair. Catastrophic failure happens. And, one can give up on oneself. Some of even the most profound of failures, however, can be repaired. Thus, shame in such instances need not be considered as having little to no value to the self or to others. An alternative response to a deep and devastating failure is to take it as an occasion for a profound self-reproach also experienced as shame (Gausel et al., 2012). Rather than being rid of the self through self-annihilation, one may wish to be rid of the self through replacement. What I mean here is a sort of symbolic suicide whereby the intense dysphoria of shame moves one to wish for a better self in place of the failed self (Gausel & Leach, 2011; Leach, 2017). We can see the value in shame here because a less profound feeling would promote a less profound change in the self (Leach, 2017). If some aspect of the self is rotten to the core, a constructive response is to replace the core. The motivational hedonism view of shame does not allow for the important, if rare, possibility of self-transformation after failure. Yet, it is often only through profound failure that one can recognize a real need for profound change. Thus, my colleagues and I argue that when shame is experienced about a failed aspect of the self that seems improvable, the intense dysphoria of such self-reproach can fuel serious intent to jettison the failed self in favor of an improved version of the self. We can imagine shame about an alcoholic, criminal, or unfaithful self as encouraging deep self-criticism that creates the need for profound self-improvement. Outside of its potential social value, or societal/moral value, in regulating behavior (see Leach et al., 2015), shame has psychological value for the individual if it facilitates self-reform. A growing body of work in psychology now supports the idea that shame can be a productive – self-improving and/or pro-social – dysphoria (for reviews, see Gausel & Leach, 2011; Leach, 2017). For example, in Gausel, Vignoles, and Leach (2016), we asked university students to imagine

After motivational hedonism


betraying a friend’s trust or almost doing so. Imagining the actual moral failure led participants to expect to view themselves as suffering a serious moral defect and it increased feelings of shame. Those who felt more shame also reported greater contrition for the harm caused as well as a greater desire to make restitution and repair the damage done. In two studies by Gausel et al. (2012) we showed a similar pattern for Norwegian adult’s shame about recently documented eugenicist policies and other racism against “Gypsies” in Norway. It is important to note that this evidence does not mean that shame is always, or even often, predictive of constructive responses to moral or other failure. Indeed, a recent statistical synthesis of 90 study samples by Leach and Cidam (2015) showed the link between felt shame and constructive motivation and behavior to range from moderately negative to moderately positive. Because so much of the behavioral science literature endorses (implicitly or explicitly) the motivational hedonism view of shame, many published papers do not discuss the possibility that shame can be linked to a constructive approach orientation to failure (i.e., motivation, intentions, and actual behavior that approaches a failure in order to improve upon it). To skirt this potential bias in the published literature, in Leach and Cidam (2015) we used meta-analytic statistical techniques to re-estimate the links between shame and constructive approach orientation in individual samples from the published literature. This separated the empirical links observed in studies from the presumed conceptual link argued in the papers that reported the studies. Interestingly, in many papers that argued for a negative link between shame and constructive approach orientation, the link was actually positive or close to zero. Rather than arguing that shame is generally approach-oriented or avoidance-oriented, Leach and Cidam argued that the motivational implications of shame are modulated by the apparent possibility for improvement of the moral, achievement, social or other failure that precipitated shame. Thus, they examined each study in their analysis and gauged whether the context or method of the study implied that failure was improvable. For instance, some studies prevented participants from subsequently interacting with a person they had harmed and therefore precluded the chance of repairing the harm by apologizing or making other restitution. As expected, the improvability of the failure about which shame was felt had dramatic effects on the motivational implications of shame in the sample. In those samples where the failure appeared improvable, shame tended to have a moderate and positive link to constructive approach orientation such as working to improve on one’s failure or helping a person one had harmed. Where failure appeared not to be improvable, the link between shame and constructive approach orientation was moderately negative. In addition to offering an integrative explanation of what would otherwise be an inconsistent empirical literature, the Leach and Cidam analysis shows how the assumption that feeling bad is bad (for oneself and others) can obscure evidence of the opposite (for a general discussion, see McGuire, 1973).


Colin Wayne Leach

Righteous anger Although its hedonic reputation may be slightly better than that of shame, anger is widely viewed as a bad feeling with bad consequences. This is partly because anger is clearly an unpleasant affective state and thus the motivational hedonism view suggests that it is aversive. According to many approaches, the aversive agitation in anger leads to hostile antagonism of those seen as slighting the self (for reviews, see Averill, 1982; Frijda, 1986; Ortony et al., 1988). Nevertheless, Averill’s (1982) careful and comprehensive theoretical and empirical analysis of actual experiences of anger showed it to be more varied in its implications for the self and others. Although it is true that most people experience the impulse to aggress when angry, Averill found that few act on it. In fact, experiences of anger were just as likely to lead to selfreflection or attempts at reconciliation as to hostile antagonism of others. As in shame, the motivational implications of anger appear to have less to do with its unpleasantness and more to do with the believed chances of having a slight addressed (Leach, 2008, 2016). When one has good reason to believe that antagonistic or other protest of a slight is likely to lead to it being addressed, then one has a good basis for directing one’s anger outward in a constructive way designed to achieve acknowledgement, appeasement, or restitution (e.g., van Zomeren, Leach, & Spears 2012). When one’s own relative power or standing makes reasonable response to one’s demands unlikely, one may be forced to choose between “swallowing” one’s anger in an impotent rancor like that Nietzsche called ressentiment (Leach & Spears, 2008). Or, one may choose to express one’s anger symbolically in hopes that the disruption or destruction somehow punishes those by whom one has been wronged (see Averill, 1982; Leach, 2008). Thus, there is little reason to presume that because anger feels bad that it is bad for the self, for others, or for society at large. It is not enough to assume that the unpleasantness of anger “functions” to alert one to possible slights and to animate antagonistic action to defend one’s humanity. The unpleasantness of anger is one particular feature of the emotional state which cannot, in and of itself, suggest what one could or should do with one’s anger.

Conclusion Millennia after its early role in philosophical and other views of human nature, motivational hedonism remains alive and well in the prevailing functionalist views of (un)pleasant affective states. Although seemingly long removed from classical assumptions, like bodily humors of phlegm and bile, a contemporary biologism of brain and body presumes that the (un)pleasantness of feeling states determine their motivational implications because such states serve to orient us to the world in ways that are functional for our individual survival (for discussions, see Barrett, 2006; Buck, 1999). Pleasant states lead us to approach and unpleasant states lead us to avoid. Whether

After motivational hedonism


acknowledged or implicit, this motivational hedonism is a powerful and pervasive meta-theory of emotion that makes it much more difficult to theorize and study the potential value of (emotional) suffering. This is clear in the work on shame in psychology and cognate behavioral sciences (see Gausel & Leach, 2011; Leach & Cidam, 2015). It is also clear in a good deal of the work on anger as a presumed dangerous state of aversive agitation (Leach, 2008). I have argued that after motivational hedonism, we are freer to contemplate and to observe the many ways in which feeling bad can be good (and feeling good can be bad). Freed from motivational hedonism, we may interrogate the social, ethical, and political implications of affective states to wonder how they “function” to advance these aims. No less an authority than Martin Luther King, Jr. implored behavioral scientists to eschew motivational hedonism to help the world understand how the ethically-minded might avoid turning a blind eye to the sorry state of the most vulnerable in our societies (for a discussion, see Allen & Leach, 2018). King called for more, and more widespread, states of “creative maladjustment” throughout the world. He argued that our emotional suffering should continue as long as the suffering of others continued. For King, feeling bad in this way was necessary to doing good. With more serious consideration for the many functions that emotions may have for individuals and society we may more properly assess whether he was right. Given what I have shared here about shame, anger, and schadenfreude, I wouldn’t bet against him.

References Allen, A. E. & Leach, C. W. (2018). The psychology of Martin Luther King Jr.’s “creative maladjustment” at societal injustice. Journal of Social Issues, 74, 317–336. Arnold, M. B. (1960). Emotion and Personality (vols. 1 & 2). New York: Columbia University Press. Averill, J. R. (1982). Anger and Aggression: An Essay on Emotion. New York: Springer-Verlag. Barrett, L. F. (2006). Emotions as natural kinds? Perspectives on Psychological Science, 1, 28–58. Bastian, B., Jetten, J., Hornsey, M. J., & Leknes, S. (2014). The positive consequences of pain: A biopsychosocial approach. Personality and Social Psychology Review, 18(3), 256–279. Bonanno, G. A. (2004). Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events? American Psychologist, 59, 20–28. Buck, R. (1999). The biological affects: A typology. Psychological Review, 106(2), 301–336. Frijda, N. H. (1986). The Emotions. London: Cambridge University Press. Gausel, N. & Leach, C. W. (2011). Concern for self-image and social-image in the management of moral failure: Rethinking shame. European Journal of Social Psychology, 41, 468–478.


Colin Wayne Leach

Gausel, N., Leach, C. W., Vignoles, V. L. & Brown, R. J. (2012). Defend or repair? Explaining responses to in-group moral failure by disentangling feelings of shame, rejection, and inferiority. Journal of Personality and Social Psychology, 102, 941–960. Gausel, N., Vignoles, V. L. & Leach, C. W. (2016). Resolving the paradox of shame: Differentiating among specific appraisal-feeling combinations explain pro-social and self-defensive motivation. Motivation and Emotion, 40, 118–139. Gray, J. A. (1990). Brain systems that mediate both emotion and cognition. Cognition and Emotion, 4(3), 269–288. Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52(12), 1280–1300. Inzlicht, M., & Legault, L. (2014). No pain, no gain: How distress underlies effective self-control (and unites diverse social psychological phenomena). In J. Forgas & E. Harmon-Jones (Eds.), The Control Within: Motivation and its Regulation (pp. 115–132). New York: Psychology Press. Kristjánsson, K. (2006). Justice and Desert-Based Emotions. Aldershot: Ashgate. Lazarus, R. S. (1991). Emotion and Adaption. New York: Oxford University Press. Leach, C. W. (2008). Envy, inferiority and injustice: Three bases of anger about inequality. In R. H. Smith (Ed.), Envy: Theory and Research (pp. 94–116). Oxford: Oxford University Press. Leach, C. W. (2016). The meta-theory of examining emotion in social relationships. Psychological Inquiry, 27, 113–116. Leach, C. W. (2017). Understanding shame and guilt. In L. Woodyatt, E. Worthington, M. Wenzel & B. Griffin (Eds.), Handbook of the Psychology of Self-forgiveness (pp.17–28). New York: Springer. Leach, C. W., Bilali, R., & Pagliaro, S. (2015). Groups and Morality. In M. Mikulincer, P. R. Shaver, J. F. Dovidio & J. Simpson (Eds.) APA Handbook of Personality and Social Psychology, Vol. 2: Group Processes (pp.123–149). Washington, DC: American Psychological Association. Leach, C. W. & Cidam, A. (2015). When is shame linked to constructive approach orientation? A meta-analysis. Journal of Personality and Social Psychology, 109, 983–1002. Leach, C. W. & Spears, R. (2008). “A vengefulness of the impotent”: The pain of in-group inferiority and schadenfreude toward successful out-groups. Journal of Personality and Social Psychology, 95, 1383–1396. Leach, C. W., Spears, R., Branscombe, N., & Doosje, B. (2003). Malicious pleasure: schadenfreude at the suffering of another group. Journal of Personality and Social Psychology, 84, 932–943. Leach, C. W., Spears, R., & Manstead, A. S. R. (2014) Situating schadenfreude in social relations. In W. W. van Dijk & J. W. Ouwerkerk (Eds.), Schadenfreude: Understanding Pleasure at the Misfortune of Others (pp. 200–216). Cambridge: Cambridge University Press. Leach, C. W., Spears, R., & Manstead, A. S. R. (2015). Parsing (malicious) pleasures: gloating and schadenfreude about others’ adversity. Frontiers in Psychology, 6, 1–13. McGuire, W. J. (1973). The yin and yang of progress in social psychology: Seven koan. Journal of Personality and Social Psychology, 26, 446–456. Moore, A. (2018). Hedonism. The Stanford Encyclopedia of Philosophy (Winter 2018 2018 edition), E. N. Zalta (ed.). Retrieved from win2018/entries/hedonism.

After motivational hedonism


Niedenthal, P. M. & Brauer, M. (2012). Social functionality of human emotion. Annual Review of Psychology, 63, 259–285. Nietzsche, F. (1967[1887]). On the Genealogy of Morals (trans. W. Kaufmann and R. J. Hollingdale). New York: Random House. Ortony, A., Clore, G. L., & Collins, A. (1988). The Cognitive Structure of Emotions. New York: Cambridge University. Parkinson, B., Fischer, A. H., & Manstead, A. S. R. (2005). Emotion in Social Relations: Cultural, Group, and Interpersonal Processes. New York: Psychology Press. Prittwitz, C. (1999). The resurrection of the victim in penal theory. Buffalo Criminal Law Review, 3, 109–129. Sartre, J. P. (1948). The Emotions: Outline of a Theory (trans. B. Frechtman). New York: Philosophical Library. Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.) (2001). Appraisal Processes in Emotion: Theory, Methods, Research. New York: Oxford University Press. Singer, T., Seymour, B., O’Doherty, J., Stephan, K., Dolan, R. & Frith, C. (2006). Empathic neural responses are modulated by the perceived fairness of others. Nature 439(7075), 466–469. Smith, R. H. (2000). Assimilative and contrastive emotional reactions to upward and downward social comparisons. In L. Wheeler & J. Suls (Eds.), Handbook of Social Comparison: Theory and Research. New York: Kluwer Academic Publishers. Solomon, R. C. (1993) The Passions: Emotions and the Meaning of Life. Indianapolis, IN: Hackett. Stoll, H. (1970). Penal Purposes in the Law of Tort. The American Journal of Comparative Law, 18, 3–21. Taylor, S. E., Kemeny, M. E., Reed, G. M., Bower, J. E. & Gruenewald, T. L. (2000). Psychological resources, positive illusions, and health. American Psychologist, 55(1), 99–109. Tiedens, L. Z., & Leach, C. W. (2004) (Eds.), The Social Life of Emotions. New York: Cambridge University Press. Tracy, J. L., Robins, R. W. & Tangney, J. P. (Eds.) (2007). The self-conscious emotions: Theory and research (pp. 463–465). New York: Guilford. van Dijk, W. W. & Ouwerkerk, J. W. (Eds.) (2014), Schadenfreude: Understanding Pleasure at the Misfortune of Others. Cambridge: Cambridge University Press. van Zomeren, M., Leach, C. W., & Spears, R. (2012). Protesters as “passionate economists”: A dynamic dual pathway model of coping with collective disadvantage. Personality and Social Psychology Review, 16, 180–199.

10 From suffering to satisfaction Why we need pain to feel pleasure Brock Bastian

It’s fair to say that suffering has a bad name. Today, perhaps more so than any other time in history, we view suffering mostly as a negative experience. This is not to say that we simply recognize its innately anti-hedonic affective qualities, but that we see it as something that detracts from our quality of life in a more overarching sense – we have a suffering aversion. This way of thinking gives rise to the idea that if we could end all suffering the world, and our lives within it, would be all the much better. The project from which this volume resulted has attempted to provide a much-needed counter-point to this perspective, exploring the various nuances of suffering rather than casting it off as something to be avoided from the outset. In this chapter I will contribute to this shift in perspective by examining, from a psychological standpoint, the specific ways in which suffering can add value to our own lives and the lives of others. I will suggest that beyond simply being an additional source of value, from a psychological standpoint, suffering is itself a necessary experience in order that people can experience its opposite – pleasure, satisfaction, and meaning in life.

Challenging a narrow perspective on suffering? Before examining the benefits of suffering, I first want to examine what indeed the term ‘suffering’ means psychologically. Although I will leave the concise definitional work to the philosophers, here I want to examine the psychological underpinnings of the experience of suffering. I also want to examine some of the drawbacks of taking a narrow perspective on what it means to suffer. There are two ways that people frequently seek to distinguish their suffering; by referring to the intensity of an experience or to its causes. For example, in terms of intensity we might seek to distinguish suffering from mere unpleasantness. Yet, from a psychological perspective this distinction is less clear than it may seem. For example, the underlying neurological and psychological mechanisms involved in experiencing both a small amount or a great degree of physical pain are largely similar. Whether it is a pinprick or a broken arm, the hardware designed to detect these events is more or less the

From suffering to satisfaction


same. These similarities also extend to the various psychological responses that may arise. Although there are indeed some psychological responses that are unique to more extreme levels of suffering (e.g., post-traumatic stress or trauma more generally), there are also many that overlap with how people respond to unpleasant experiences across a wide range of intensities. For instance, inherent to most unpleasant experiences is the desire to avoid or escape the stimulus, even in those cases when this motivation is overcome (e.g., such as when eating chili). Just so, and as I will argue below, many of the benefits that may arise in response to small amounts of unpleasantness are just as likely to occur in the case of experiences that may be more commonly referred to as suffering. These same similarities are evident when we examine the various causes of suffering. For instance, although people may refer to both a broken leg and the loss of a loved one as painful, they tend to see these different pains as distinct – one is social and one is physical. Yet, as with suffering vs. unpleasantness, many of the same neurological and psychological mechanisms respond to both types of pain, and many of the same psychological responses may also be apparent (MacDonald, 2005). It is informative to consider how this tendency to look for differences in negative experiences compares with how we think about positive experiences. For instance, the comparison between pleasure and joy tends to be viewed as holding less informational value than is the comparison between unpleasantness and suffering. Knowing that something might bring me joy vs. pleasure is of less importance than knowing that something might cause me to suffer vs. feel unpleasant. The causes of pleasant experiences are also delineated with less importance. For instance, understanding the differences between the pleasures of sex vs. chocolate may be of less concern to me, than knowing the differences between the pain of emotional loss vs. physical injury. In short, we are more sensitive to the potential harm from negative experiences than we are to the potential gains from positive experiences, and therefore we seek to categorize negative experiences more carefully. This tendency to be more focused on the differences between negative experiences, relative to positive experiences, is apparent in the fact that we tend to have more words to describe negative vs. positive experiences. Studies have found that whether it is emotion words, emotion related personality traits, or words describing hedonic experience, we have at least twice as many words to describe the negative side of the spectrum (Averill, 1980; Rozin & Royzman, 2001). Why does it matter that we are more careful when describing the differences between negative experiences than when we describe the differences between positive experiences? When it comes to thinking about the role that various experiences play in our lives, focusing on the differences may be unhelpful. Take for instance marathons. These experiences clearly include unpleasantness, in fact they also involve significant levels of pain and suffering, yet people are more likely to say that they get pleasure from running


Brock Bastian

marathons, perhaps even joy, than they are to say that marathons cause them to suffer. They may acknowledge that running a marathon is very unpleasant and challenging, but they would not link this experience of unpleasantness or feeling of being challenged to the concept of suffering. This also means that they may fail to see links between the benefits that arise from suffering for the sake of a marathon and the experience of suffering more generally. In this sense, how we think about suffering often sells it short. Although suffering (broadly defined) is an aspect of daily living and may be a central component of many activities that we intentionally seek out and get pleasure from, we stop short of directly identifying the suffering in those activities. Rather we reserve the term ‘suffering’ for those extreme, chronic, or clearly harmful experiences in life. This leaves what I have referred to elsewhere as ‘pain blindspots’ (Bastian, 2018). We do not see the role that suffering plays in the experiences that we find rewarding, pleasurable, or meaningful. We also do not see the links between more intense negative experiences, the ones we might more commonly refer to as pain or suffering, and those mere unpleasant experiences that we might more readily understand as benefiting us. Yet, psychologically speaking, the differences are less than clear and the benefits that arise from these different negative experiences are not necessarily greater when they are only mild. For instance, the sense of achievement from running a marathon only increases as the level of effort, pain, and suffering endured to reach one’s goal increases. As I hope will become clearer, it is also the case that if people did not experience any pain or suffering as part of running marathons they would probably not get much enjoyment from them at all. The fact is that marathons are only rewarding because they are challenging, and they are only challenging because marathon runners have to suffer the pain it takes to run 40 kilometres. If we removed all suffering from marathons they would be neither challenging nor rewarding. It is interesting to note that a failure to see a link between suffering and pleasure or reward also leads us to view people who do report enjoyment from suffering as morally suspect. Most obviously, this occurs in the context of sexual masochism, frequently considered a form of deviant behaviour. Yet, as with marathons, research shows that many people seek out and enjoy innately negative experiences, such as sad movies, spicy foods, bungee jumping, and cold-water swimming. This behavioural tendency has been referred to as benign masochism, and research shows that people frequently get their peak level of enjoyment from these negative experiences just at the point where they can barely stand it anymore (Rozin, Guillot, Fincher, Rozin & Tsukayama, 2013). This suffering can be enjoyable, but it is not the case that the milder the suffering the better; people like their suffering to take them to a point which pushes their boundaries, challenges them, but also does not destroy or traumatize them. Interestingly, this appears to be a uniquely human tendency, with little evidence of animals developing a preference for stimuli which typically cause

From suffering to satisfaction


unpleasant experiences, such as spicy chili. Although dogs in Mexico will tolerate the spicy food disposed of by humans, they do not appear to develop a preference for it (Rozin et al., 2013). It is the human capacity for a metaexperience of ourselves, pushing against innately unpleasant experiences, that drives a sense of achievement and pleasure. While it may be tempting to say that it is not the suffering itself that people enjoy, it is also the case that if the suffering were not present, neither would be the rewards. While many may still prefer to reserve the term ‘suffering’ for more harmful and extreme negative experiences, the fact is that, at a psychological level, suffering is not clearly distinguishable from more mild forms of unpleasant experience, and furthermore, the benefits that may arise from unpleasant experiences are also apparent in cases of more extreme suffering. Drawing distinctions between different intensities and causes of suffering (in ways that are rarely applied to positive experiences) tends to blind us to the benefits that may be associated with suffering. Furthermore, when suffering is removed from many of the experiences that we find meaningful and rewarding, these same experiences become mundane, pointless, and unrewarding.

Why suffering is necessary for pleasure It is one thing to argue that suffering can be pleasurable in some contexts, but it is quite another to argue that suffering is in fact necessary for the experience of any pleasure at all. Yet, from a psychological standpoint there is good reason to think that eradicating all suffering may in fact serve to undermine, rather than promote, our wellbeing. This is not to say that all suffering all the time is the way forward either; as with anything moderation is needed. However, this is not the intuition that I am arguing against. It is the idea that the eradication of suffering would make the world a better place that I am challenging here. My argument mostly relates to one’s personal experience of suffering, and from this perspective it is important to note that what counts as too much or too little is likely to vary across contexts and individuals. People have different capacities to cope with negative experiences. As noted earlier, pleasure is sometimes maximized by taking ourselves to the very edge of what we can cope with. This points to the critical importance of finding a careful balance between one’s own capacities to cope and the demands of the experience we are confronting. This idea has been captured within the biopsychosocial model of challenge and threat, which predicts that a person will experience the positive benefits of being challenged when they perceive that their personal resources to cope outweigh the demands of a given situation (Blascovich & Tomaka, 1996). When situational demands outweigh personal resources, the experience of threat ensues. These different psychological states have different physiological and biological underpinnings, with the experience of challenge leading to the release of adrenaline and the experience of threat leading to the release of


Brock Bastian

cortisol, the latter of which has been linked to poor health outcomes across a range of factors (Dienstbier, 1989). Critically, however, suffering is not only associated with the experience of threat. We also cannot feel challenged without something to push against, and pleasure cannot provide this obstacle. It is pain and suffering (in all its intensities and varieties) which is necessary for the experience of feeling challenged. Suffering plays an important role in contributing to the good life, and the notion that by eradicating it completely we would be better off misses the role it plays in promoting pleasure, meaning, and wellbeing broadly understood. The psychology of adaptation The first reason that I believe we need suffering builds on the psychology of adaptation. Humans are highly adaptable to their environments, and this means that both the experience of pleasure and the experience of pain and suffering become less extreme overtime. This has allowed us to overcome, and thrive within, challenging environments. Yet, it also means that comfort and pleasure fade quickly from view if not also contrasted with something else. A good example is the experience of a hot spa. On a cold night, getting into a hot spa can be a highly pleasurable and rewarding experience. Overtime, however, the experience of pleasure dissipates as our bodies become accustomed to the new temperature. How could we get the experience of pleasure we had at first all over again? Well, generally, people jump into a plunge pool of cold water until shivering just a little and then get back into the spa. It is this contrasting experience which releases the same level of pleasure a second time. The same is true of many pleasures in life. We get more pleasure from food when hungry, from drinking when thirsty, and from rest when physically tired. If these drives or needs remain constantly satiated, there is little reward experienced from relieving them. It is when we first endure some level of discomfort or unpleasantness that we are able to reap the full benefits of a pleasurable and rewarding experience. The psychology of relief While the process of adaptation can lead to a pleasant experience becoming less intense and more mundane overtime, the experience of relief provides insight into why suffering has the capacity to increase pleasure. Opponent process theory observes that, broadly speaking, there is a pattern whereby a negative experience gives way to a sense of relief that in turn releases an experience of pleasure. Furthermore, the experience of pleasure tends to exceed the preceding hedonic state (Solomon & Corbit, 1974). This can be observed in the case of receiving bad news, which leads one to worry and experience a state of anxiety. If that news, however, turns out to be

From suffering to satisfaction


less negative than one first thought, the ensuing sense of relief is associated with a more positive hedonic state than that which was experienced prior to hearing the news in the first place. Another good example is the case of the ‘runners high’. People often report increased feelings of euphoria after they have engaged in strenuous exercise. Put simply, even though the process of exercising itself was unpleasant, upon completing this activity they feel better than they did before they started. Research has demonstrated that this opponent process is correlated with the release of opioids, providing an underlying biological mechanism via which the experience of suffering can lead to feelings of satisfaction (Boecker et al., 2008). Other researchers have identified that this is due to the shared neurobiology associated with the experience of pleasure and pain (Leknes & Tracey, 2008). Opioids are conducive to the experience of pleasure but are also activated in response to painful stimuli, because they are effective for pain management. When painful stimulation stops suddenly, people experience what has been referred to as an opioid overshoot. That is, they are no longer experiencing pain, but the brain is continuing to release opioids and this leads to a net increase in pleasure. Supporting this overall picture is other evidence showing that the emotional state of relief is not limited to the absence of negative affect but can be experienced as rewarding and includes the presence of positive affect (Franklin, Lee, Hanna & Prinstein, 2013; Leknes, Lee, Berna, Andersson & Tracey, 2011). Indeed, the experience of relief can even mean that a lessor experience of pain, following a more serve painful event, can feel pleasant (Leknes, Berna, Lee, Snyder, Biele & Tracey, 2013). The psychology of relief provides a mechanism for understanding why suffering may be important for boosting experiences of pleasure. Set in the context of what we know about adaptation, the role of contrasting hedonic experiences in boosting positive hedonic states seems especially necessary. It is also worth noting that according to opponent process theory, over repeated instances the negativity of suffering tends to reduce, whereas the resulting experience of pleasure tends to increase. This suggests that (within limits) adaptation in the context of contrasting experiences should lead to increased pleasure, whereas adaptation in the context of static hedonic states is just as likely to reduce pleasure. Peak experiences If you take a moment and recall the most memorable events of your life it is likely that there will be one thing in common. These memorable events will likely be characterized by high levels of affect. This may be highly positive experiences, but also highly negative experiences. Research on peak experiences has demonstrated that when we recall events our memory of those events tends to be shaped by the peak highs as well as the peak lows (Fredrickson, 2000; Fredrickson & Kahneman, 1993;


Brock Bastian

Kahneman, Fredrickson, Schreiber & Redelmeier, 1993). That is to say we are not very good at recalling the mean-level or absolute sum of an experience, or at least the mean-level/absolute sum does not factor heavily into how that event is constructed retrospectively. When it comes to what elements of an event give it meaning, or which events over a longer time span tend to give our lives meaning, it is the peaks that matter. When paired with an understanding of opponent processes, it would appear that the meaningful and memorable life is one composed of contrasts – peak moments of suffering followed by the ecstatic experience of relief. This is a question that we have begun to explore in terms of what factors make events feel meaningful. Moving away from a tendency to compare positive to negative experiences in their perceived meaningfulness, we asked the question of whether the meaningfulness of an event may be more to do with its extremity and levels of arousal, rather than its valence per se. By asking people to recall significant events over the past 12 months and to rate them on various factors, including their meaningfulness, we were able to address this question. Across three studies we found that people rated events that were either highly pleasurable or highly painful as more meaningful compared to events that were more neutral and less extreme. Our findings also revealed that both extremely positive and extremely negative events were rated as more intense, were contemplated more, and fostered social sharing with others, and these factors helped to explain why extreme events were considered to be especially meaningful (Murphy & Bastian, under review). Numbing pain also numbs pleasure The analysis thus far suggests that a failure to engage with suffering may be counter-productive to wellbeing, reducing satisfaction, pleasure, and meaning in life. There is still another side to this equation which is that our attempts to numb or avoid pain may also undermine pleasure. Quite apart from the fact that numbing pain would undermine the benefits of opponent processes, relief, and peak experiences, evidence suggests that it would also directly reduce our capacity to engage with pleasure itself. Evidence for this comes from a study examining the effects of the painkiller acetaminophen (paracetamol) on the rated pleasantness and unpleasantness of visual stimuli (Durso, Luttrell & Way, 2015). We might expect that if painkillers broadly reduce the unpleasantness of affective experiences, then they might also reduce the rated unpleasantness of visual stimuli. This is what the researchers found. When people had taken the painkiller they provided less extreme ratings of the unpleasantness of 40 different photos. Critically, however, they also provided less extreme ratings of the pleasantness of those same photos. The pain killers effectively narrowed the range of subjective affective experience, whereby people become numb to both pleasure and pain. These findings point to the shared underlying biology of pleasure and pain, suggesting that interrupting the experience of one also interrupts the experience of the other.

From suffering to satisfaction


Overall, then, I would argue that eradicating all suffering is unlikely to make our lives better. Furthermore, the irony is that without substantially changing our biological and psychological make-up, eradicating suffering is a near impossibility. Endless pleasure is itself a form of suffering. Whether that suffering is tied to over-indulgence, such as the feeling of sickness that ensues from too much chocolate, or the numbness that might result from enduring invariant hedonic states, pleasure quickly morphs into suffering, ensuring that we have the contrasts that we need while also effectively regulating our hedonic appetite.

Pathways from suffering to pleasure Moving beyond a hedonic analysis for why suffering is necessary for pleasure, there are serval other avenues through which suffering may enhance one’s satisfaction in life. These range from the capacity for suffering to fulfil justice motives, regulate other negative emotional states such as anger, build social connection, pro-social behaviour, and creativity within groups, and to promote resilience. My colleagues and I have explored several potential benefits of pain in detail elsewhere (Bastian, 2018; Bastian, Jetten, Hornsey & Leknes, 2014; Bastian, Jetten & Hornsey, 2014; Bastian, 2017; Leknes & Bastian, 2014). Below I review several areas in which there is new and emerging evidence. The justice of suffering The experience of suffering is frequently attributed with meaning. People may ask the question of why they are suffering or what it means that they suffer. The meaning of suffering can also be more intuitive, automatic, and embodied. In our work on the reasons why people seek out painful activities, we suggest that pain and suffering are closely tied to notions of justice. Indeed, if we are to bring justice to someone it is mostly through the experience of suffering in some form. This also suggests that one’s own experience of suffering may activate justice-related thinking. To test this, we have run several studies. Our first showed that when people are made to feel guilty by recalling a past immoral deed, they are more likely to seek out suffering by holding their hand in an ice-bucket for longer (Bastian, Jetten & Fasoli, 2011). Importantly, these individuals also rated the experience as more painful relative to those who were not made to feel guilty and held their hand in for a lessor amount of time – meaning our finding was not because the experience of guilt made them numb to the pain. So why would they seek out this type of unpleasant experience? We also found that compared to people who were not given the chance to experience pain, those who did, felt less guilty afterwards. Their experience of pain had provided a sense of justice, allowing them to feel better (and more moral) again. More recently we examined this effect in the context of keeping secrets (Slepian & Bastian, 2017). People often keep secrets because they wish to


Brock Bastian

avoid the just punishment that might prevail should they fess up to others about their misdemeanours. We predicted that in order to avoid the punishment of others, people are likely to undertake the punishment of themselves in order to maintain a sense that they are just and good. Across six studies we found that when people recalled a secret misdeed they were more likely to deny themselves pleasure (e.g., going out to dinner with friends, indulging in a small luxury) and to seek out pain (e.g., intense exercise, social isolation, setting difficult goals, or choosing a painful vs. non-painful task) compared to those who recalled a misdeed they had already confessed. This was especially the case for more significant secrets and was explained by self-reported feelings that one deserved to be punished. While suffering enforced on criminals provides a sense of justice within the community, it also appears to provide an internal sense of justice for the individual. Drawing on this same process, we wondered whether people who feel they don’t deserve to suffer – the saints rather than the sinners – might respond to their unjust punishment with a sense of entitlement (Bastian, Jetten & Stewart, 2013). Indeed, we found that when people viewed themselves as good (as opposed to guilty) they responded to the experience of pain by rewarding themselves with more sweets when offered or deciding to indulge themselves by taking a chocolate rather than a highlighter as a gift. Together these findings suggest that pain is experienced as the embodiment of punishment, activating higher-order cognitions about fairness and justice. It is through this avenue that pain has the capacity to restore justice or lead us to indulge ourselves a little ‘just reward’. Suffering as a salve for suffering While suffering can reduce guilt because it is linked to higher concepts of justice, it has also been shown to be effective in reducing negative affect more directly. This is cited as one (of many) reasons for why people engage in the practice of self-harm. Consistent with the work of Franklin et al. (2013), the pain associated with this experience has the capacity to down-regulate negative emotion and upregulate positive emotion, in part due to the psychology of relief. This understanding also suggests that those who self-harm may be well-advised to seek out pain in heathier and less harmful pursuits, such as through intensive exercise. While pain may downregulate negative affect in general, we also wondered whether there were particular negative emotions that were more likely to motivate people to engage with pain (Harmon-Jones, Summerell & Bastian, 2017). Drawing on the idea that under conditions of negative affect pain may represent an approach-avoidance conflict – people are motivated to avoid it due to its unpleasantness, but motivated to approach it due to its capacity to make them feel better – we predicted approach related negative emotions (such as anger) would be more likely to motivate engagement with pain than avoidance motivated emotions (such as anxiety). Indeed, this is what we

From suffering to satisfaction


found. After being induced to feel anger people increased their preference for painful activities, more so compared to people in a neutral condition or those who were induced to experience anxiety. Suffering gets us together and gets us going If suffering can make us feel better, can it also lead us to become better people? How do people respond to each other when they suffer together, and can the experience of suffering forge a sense of belonging and connection with others? There are many examples which suggest that people, at least sometimes, respond to traumatic events in this way. For instance, data shows that in response to the 9/11 attacks there was a spike in volunteering rates across America (Penner, Brannick, Webb & Conell, 2005). This was not only for crises-related targets, but for a range of targets, and not only in New York but nationwide. We wondered whether we might replicate this effect of shared suffering in an experiment where people experienced only mild pain compared to immense and traumatic suffering (Bastian, Jetten & Ferris, 2014). We had groups of volunteers come into our lab and they were either exposed to a set of painful tasks (putting their hands in ice-water and then performing leg squats) or non-painful tasks (putting their hands in room temperature water and then balancing on one leg). After completing these tasks in a group setting they were asked several questions related to how bonded they felt to the group, how much they felt a part of the group, and how much they trusted the other members of their group. We found that groups who had shared in a painful experience reported higher levels of bonding to their group than those who had shared a very similar, but non-painful activity. Critically, we ensured that each of the individuals in the groups did not know each other prior to the study. This meant that we were observing the formation of new social ties and these were accentuated when people experienced painful (versus non-painful) experiences together. Next, we wanted to know whether this effect might extend to cooperative and pro-social behaviour. To achieve this, we conducted the same experiment, but this time we had each of the groups participate in a cooperation dilemma. In this game they could choose a number between 1 and 7 across 6 rounds. If they chose a 7 they could get paid $7.80, but if someone else in their group chose a 1 they would only get $0.40. If they chose a 1, however, they would be guaranteed $4.20. This meant that choosing a higher number was the more cooperative and trusting option but would also expose the participant to the risk that someone else in their group might not choose the cooperative option, leaving them with a smaller reward. As in Study 1, we found that groups who had experienced pain together were also more likely to choose high numbers; they were more likely to cooperate with and trust each other. In our final study we wanted to create a situation that was more representative of how people might engage in a painful activity together in


Brock Bastian

everyday life. People frequently enjoy hot food, such as curries, as a social activity, and this represents a mild and everyday form of pain seeking behaviour. We wondered whether this might also have the same effect on cooperative behaviour. To this end, we asked people to eat a fresh birds eye chili (or as much of one as they could handle) in the same size groups as before. In the control condition we asked people to consume a butter scotch sweet – a similar activity, but more pleasurable than painful. As in the first two studies we used a cover story. Whereas in the earlier studies we told people that we were assessing their physical abilities (with ice-bucket and leg squat endurance), in this study we told them that we were conducting a consumer survey on different types of food. Again, we found that people who shared in a painful experience (this time the consumption of a hot chili pepper) were more likely to choose higher numbers, indicative of their willingness to cooperate with and trust the other group members, compared to those who shared a similar but non-painful experience (the consumption of a butter scotch sweet). In a more recent follow-up to these studies, we have examined whether the social bonding and group climate that emerges from a shared painful experience might also leverage other downstream outcomes (Bastian, Jetten, Thai & Steffans, 2018). Encouraging innovation within groups is a desirable outcome in many organizational contexts with research showing that one factor which undergirds innovation is a safe and supportive culture (Edmondson, 1999). For creativity to emerge within groups it is important people feel comfortable to express divergent and risky ideas, and this is quickly undermined when they feel they are likely to experience negative reactions from others. We therefore tested whether sharing a painful versus non-painful experience with others might engender more group-based creative outputs. To achieve this, we had groups of between three and five people first consume as many bird’s eye chillies as they could and then do the leg squat task in the experimental condition. In the control condition they ate a butterscotch and then were instructed to balance on one leg (as in the previous studies we had run). Next, each of the groups were given the task of coming up with as many different uses for a brick as they could. This is a classic task used to examine creative thinking, with the number of divergent and unique ideas coded for by the researchers (Guilford, 1967). Next, each of the groups were asked to design a poster using scrap magazines and other materials, with the instruction to make it a ‘work of art’. We video recorded all of these activities and then had independent coders rate each of the groups in terms of how supportive they were of each other. We also asked a different set of coders to rate their subjective impression of the creativity of each of the posters produced by the groups. First, we found that groups that shared a painful experience (versus a nonpainful experience) exhibited higher levels of social support. Second, this increased level of social support predicted the creativity (i.e., number of divergent and unique ideas) exhibited in the brick uses task, and also the

From suffering to satisfaction


independent ratings of poster creativity. These findings suggest that engaging in some suffering – whether that be extending outside of one’s comfort zones or embracing the possibility of difficulty and even failure – has the capacity to not only bring groups together, but also to improve their performance in various ways. The possibility that suffering might lead groups to respond in ways that enhance their likelihood of success is an idea that I have explored in collaboration with the anthropologist Harvey Whitehouse and others. In a recent paper reporting on a series of real-world experiments and mathematical modelling, we find that sharing painful experiences can produce ‘identity fusion’ – a visceral sense of oneness – and that this response motivates self-sacrifice for the group, including a willingness to fight and die for other members of that group (Whitehouse et al., 2017). We argue that this response, which is conditioned on experiences of shared suffering, emerged to help groups survive in highly threatening environments. Although the benefits of identity fusion may be less beneficial in more mundane contexts, our work shows that shared suffering has the potential to powerfully draw people together in ways that likely go beyond other types of shared experience. Suffering promotes resilience A final avenue through which I want to explore the idea that suffering has value in our lives is by examining the role of suffering in building resilience. Ultimately, if we are to live a happy life then we need the psychological architecture to maintain high levels of wellbeing. This means not only the ability to reap pleasure from life, but also to respond well to the many setbacks (Kalisch et al., 2017). Resilience is a term that has been used to describe the ability to respond well to stress and adversity in life. When people have high levels of resilience, they are more likely to take setbacks in their stride – to see them as a challenge. Furthermore, this capacity means that they are more likely to grow than wilt from these experiences. A lack of resilience, however, means they are more likely to feel threatened by adversity, to become traumatized by it, and in that way to wilt rather than grow. Resilience is that personal resource that can turn the experience of adversity into a challenge as opposed to a threat. The big question is how do people build resilience? Much of the work on resilience to date has viewed it as a character trait or individual difference – some people are just more resilient than others (Schultze-Lutter, Schimmelmann & Schmidt, 2016). While this is certainly true and may be useful for recruitment processes which aim to select individuals who cope well with stress, it is not especially helpful when it comes to understanding how to build or develop resilience. More recently researchers have called for a focus on the way in which dynamic processes underpin the capacity to adapt to adverse life circumstances (e.g., Kalisch et al., 2017).


Brock Bastian

A focus on how these dynamic processes unfold is something that my colleagues and I are becoming increasingly focused on. We are especially interested in the idea that an important building block for resilience is exposure to suffering. If resilience is like a muscle that needs to be developed, then it would be unlikely to become strong in contexts that are defined by pleasantness, ease, or comfort. Rather, to develop resilience, it is likely that we need to be exposed to pain, difficulty, and discomfort. We are especially interested in how this plays out in everyday experiences. As with our other work, which focuses on mild and mostly mundane experiences of pain, we are interested in whether resilience can be built through exposure to the day-to-day hassles and stressors in life. Our prediction here is drawn from previous work which has shown that more serious and traumatic events indeed have this capacity to build resilience and to produce positive outcomes. Beyond the rather large literature on posttraumatic growth (e.g., Tedeschi & Calhoun, 2004), other work has shown that past exposure to traumatic events can lead to positive improvements. For instance, research examining responses to the Virginia Tech campus shootings found that for some people there was a trajectory of improvement in response to this event – their mental health literarily improved (Mancini, Littleton & Grills, 2016). The reason for this, according to the authors, is consistent with the other evidence cited above that such experiences can lead to increases in social support. People who felt alone and depressed prior to the shootings experienced an improvement in their level of social connection as a result of the tragedy, and this in turn had mental health benefits. Having a supportive social network is important for resilience; however, other work suggests that intrapersonal processes may also be at play. For instance, researchers have asked people to report how many adverse life events they have been exposed to. Some of these are very significant, such as the loss of loved ones, war, or event personal assault. From this they calculated a person’s ‘cumulative lifetime adversity’ index and used it to predict a number of different outcomes. In one study, they asked participants to hold their hand in a bucket of ice-water (similar to our own methods) and found that people who had a moderate amount of lifetime adversity reported less pain and responded better to their pain. This was compared to people who had very little or very much lifetime adversity, these individuals reported more pain and responded less adaptively (Seery, Leo, Lupien, Kondrak & Almonte, 2013). They also found that people with a moderate amount (as opposed a little or a lot) of lifetime adversity also reported higher levels of wellbeing in life. Surveys conducted over the period of the 9/11 terror attacks also indicated that these individuals coped better with the associated stress and trauma (Seery, Holman & Silver, 2010). This work certainly suggests that some exposure to suffering is not only unlikely to undermine resilience but is necessary for its development. Providing further support to this idea is research using animal models which shows how exposure to some suffering builds the biological capacity to

From suffering to satisfaction


respond effectively to stress. Specifically, research suggests that when rats are exposed to stress early in life they develop larger adrenal glands. Theoretically, it has been proposed that large adrenal glands allow for a more efficient and adaptive response to threat and stress, but in a way that also allows for a faster return to baseline (Dienstbier, 1989). This is much like the example of a marathon runner whose high level of fitness means they are able to recover quickly. Being exposed to stress allows the physiological system to respond better to that stress in the future. A fascinating study on rats examined the prediction originally exposed by Seligman, which he later referred to as learned helplessness (Overmier & Seligman, 1967). In the original studies, Seligman had exposed dogs to an inescapable electric shock and then afterwards exposed them to the same shock but in a cage where they could jump over a barrier and escape. What the research found was that the dogs effectively gave up and even when the possibility of escape presented itself they did not take it. They had learned to be helpless in response to the shock. The follow-up study also exposed rats to an inescapable shock followed by a shock where they had the chance to escape. However, half of the rats were exposed to 14 days of cold swims and electric shocks prior to the test, whereas the other half were not. What this study found was that the rats that had been conditioned by repeated exposure to stressful events did not show the learned helplessness response, they escaped. The rats that did not have this exposure showed a similar learned helplessness response as did Seligman’s dogs (Weiss, Glazer, Pohorecky, Brick & Miller, 1975). Overall, the evidence suggests that exposure to suffering may be necessary for the development of resilience. This idea is not dissimilar to how we understand immunization. Exposing the biological immune system to a small amount of a pathogen triggers the system into building a more efficient response. Just so, exposing the psychological immune system to a moderate amount of stress, adversity, and suffering, provides the foundation on which more resilient responding can be developed.

Conclusion Suffering is everywhere! Such a statement can inspire a sense of hopelessness. Yet, I hope I have described a view of suffering that may allow for a different response. Suffering is everywhere, because it is hiding in many places where we often fail to see it. It is hiding in the activities that we find meaningful, the experiences that build community, and the events which bring us joy. Naively we might believe that we should eradicate suffering, but a broader view suggests that taken to the extreme this approach might be just a likely to undermine the very elements of these experiences that make them pleasurable, rewarding, and meaningful. Perhaps there is an imagined state in which humans could do without suffering, but until someone manages to biologically and psychologically


Brock Bastian

engineer this new form of human existence, suffering remains our friend. Without suffering we would cease to experience pleasure or have the capacity to extract meaning from life and it would derail a key pathway through which we are able to connect with others. If we did invent a human lifeform that could do without suffering, then we would also need to invent a context in which they were no longer exposed to adversity. Without this measure of protection, they would certainly crumble at the first hiccup, the first hassle, or first sign of pain. It is through suffering that we learn to become tough and resilient to the hiccups and setbacks that life inevitably throws us. If we are to cope well with suffering – even those chronic, harmful, traumatic, and undesirable forms of suffering that none of us would knowingly choose – then having a model which acknowledges its various positive sideeffects can only be helpful. A tendency to view suffering as simply futile will only make it worse. The irony of human experience is like that – the more we fear or try to avoid internal psychological events, the more powerful they become. Suffering has a bad name, but that is because suffering has been misunderstood. If it were only an experience which detracted value from our lives, then we could never explain why people seek out suffering. Moreover, with a broader view of suffering in hand we can see this behaviour occurs more frequently that we might at first realize. This is due to the many benefits it offers, and ultimately its necessary role in building pleasure and wellbeing in life. A life without suffering, in the end, appears not to be a very valuable life at all.

References Averill, J. R. (1980). On the paucity of positive emotions. In K. Blankstein, P. Pliner & J. Polivy (eds.), Advances in the Study of Communication and Affect Vol. 6. New York: Plenum. Bagozzi, R. P., Wong, N. & Yi, Y. (1999). The role of culture and gender in the relationship between positive and negative affect. Cognition & Emotion, 13, 641–672. Bastian, B. (2017). A social dimension to enjoyment of negative emotion in art reception. Behavioral and Brain Sciences, 40, E352. Bastian, B. (2018). The Other Side of Happiness: Embracing a More Fearless Approach to Living. London: Penguin UK. Bastian, B., Jetten, J. & Fasoli, F. (2011). Cleansing the soul by hurting the flesh: the guilt-reducing effect of pain. Psychological Science, 22(3), 334. Bastian, B., Jetten, J. & Ferris, L. J. (2014). Pain as social glue: shared pain increases cooperation. Psychological Science, 25(11), 2079–2085. Bastian, B., Jetten, J. & Hornsey, M. J. (2014). Gustatory pleasure and pain: the offset of acute physical pain enhances responsiveness to taste. Appetite, 72, 150–155. Bastian, B., Jetten, J., Hornsey, M. J. & Leknes, S. (2014). The positive consequences of pain: A biopsychosocial approach. Personality and Social Psychology Review, 18 (3), 256–279. Bastian, B., Jetten, J. & Stewart, E. (2013). Physical pain and guilty pleasures. Social Psychological and Personality Science, 4(2), 215–219.

From suffering to satisfaction


Bastian, B., Jetten, J., Thai, H. & Steffans, N. K. (2018). Shared adversity increases team creativity through fostering supportive interaction. Frontiers in Psychology, 9, 2309. Blascovich, J. & Tomaka, J. (1996). The biopsychosocial model of arousal regulation. In Advances in Experimental Social Psychology (Vol. 28, pp. 1–51). Boca Raton, FL: Academic Press. Boecker, H., Sprenger, T., Spilker, M. E., Henriksen, G., Koppenhoefer, M., Wagner, K. J., … & Tolle, T. R. (2008). The runner’s high: opioidergic mechanisms in the human brain. Cerebral Cortex, 18(11), 2523–2531. Dienstbier, R. A. (1989). Arousal and physiological toughness: implications for mental and physical health. Psychological Review, 96(1), 84–100. Durso, G. R., Luttrell, A. & Way, B. M. (2015). Over-the-counter relief from pains and pleasures alike: acetaminophen blunts evaluation sensitivity to both negative and positive stimuli. Psychological Science, 26(6), 750–758. Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44, 350–383. Franklin, J. C., Lee, K. M., Hanna, E. K. & Prinstein, M. J. (2013). Feeling worse to feel better: pain-offset relief simultaneously stimulates positive affect and reduces negative affect. Psychological Science, 24(4), 521–529. Fredrickson, B. L. (2000). Extracting meaning from past affective experiences: the importance of peaks, ends, and specific emotions. Cognition & Emotion, 14(4), 577–606. Fredrickson, B. L. & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65, 45–55. Guilford, J. P. (1967). The Nature of Human Intelligence. New York: McGraw-Hill. Harmon-Jones, C., Summerell, E. & Bastian, B. (2017). Anger Increases Preference for Painful Activities. Motivation Science, 4(4), 301–314. Kahneman, D., Fredrickson, B. L., Schreiber, C. A. & Redelmeier, D. A. (1993). When more pain is preferred to less: adding a better end. Psychological Science, 4 (6), 401–405. Kalisch, R., Baker, D. G., Basten, U., Boks, M. P., Bonanno, G. A., Brummelman, E., … & Geuze, E. (2017). The resilience framework as a strategy to combat stress-related disorders. Nature Human Behaviour, 1(11), 784–790. Leknes, S. & Bastian, B. (2014). The benefits of pain. Review of Philosophy and Psychology, 5(1), 57–70. Leknes, S., Berna, C., Lee, M. C., Snyder, G. D., Biele, G. & Tracey, I. (2013). The importance of context: when relative relief renders pain pleasant. PAIN, 154(3), 402–410. Leknes, S., Lee, M., Berna, C., Andersson, J. & Tracey, I. (2011). Relief as a reward: hedonic and neural responses to safety from pain. PloS One, 6(4), e17870. Leknes, S. & Tracey, I. (2008). A common neurobiology for pain and pleasure. Nature Reviews Neuroscience, 9(4), 314–320. MacDonald, G. & Leary, M. R. (2005). Why does social exclusion hurt? The relationship between social and physical pain. Psychological Bulletin, 131(2), 202–223. Mancini, A. D., Littleton, H. L. & Grills, A. E. (2016). Can people benefit from acute stress? Social support, psychological improvement, and resilience after the Virginia Tech campus shootings. Clinical Psychological Science, 4(3), 401–417. Murphy, S.C. & Bastian, B. (Under review). Emotionally extreme life experiences are more meaningful.


Brock Bastian

Overmier, J. B. & Seligman, M. E. (1967). Effects of inescapable shock upon subsequent escape and avoidance responding. J Comp Physiol Psychol, 63(1), 28–33. Penner, L., Brannick, M. T., Webb, S. & Connell, P. (2005). Effects on volunteering of the September 11, 2001, attacks: an archival analysis. Journal of Applied Social Psychology, 35(7), 1333–1360. Rozin, P., Guillot, L., Fincher, K., Rozin, A. & Tsukayama, E. (2013). Glad to be sad, and other examples of benign masochism. Judgment and Decision Making, 8(4), 439–447. Rozin, P. & Royzman, E. B. (2001). Negativity bias, negativity dominance, and contagion. Personality and Social Psychology Review, 5(4), 296–320. Schultze-Lutter, F., Schimmelmann, B. G. & Schmidt, S. J. (2016). Resilience, risk, mental health and well-being: associations and conceptual differences. European Child & Adolescent Psychiatry, 25, 459–466 Seery, M., Holman, E. & Silver, R. (2010). Whatever does not kill us. Journal of Personality and Social Psychology, 99, 1025–1041. Seery, M. D., Leo, R. J., Lupien, S. P., Kondrak, C. L. & Almonte, J. L. (2013). An upside to adversity? Moderate cumulative lifetime adversity is associated with resilient responses in the face of controlled stressors. Psychological Science, 24(7), 1181–1189. Solomon, R. L. & Corbit, J. D. (1974). An opponent-process theory of motivation: I. Temporal dynamics of affect. Psychological Review, 81(2), 119–145. Slepian, M. L. & Bastian, B. (2017). Truth or punishment: secrecy and punishing the self. Personality and Social Psychology Bulletin, 43, 1595–1611. Tedeschi, R. G. & Calhoun, L. G. (2004). Posttraumatic growth: conceptual foundations and empirical evidence. Psychological Inquiry, 15(1), 1–18. Weiss, J. M., Glazer, H. I., Pohorecky, L. A., Brick, J. & Miller, N. E. (1975). Effects of chronic exposure to stressors on avoidance-escape behavior and on brain norepinephrine. Psychosom Med, 37(6), 522–534. Whitehouse, H., Jong, J., Buhrmester, M. D., Gómez, Á., Bastian, B., Kavanagh, C. M., … & Gavrilets, S. (2017). The evolution of extreme cooperation via shared dysphoric experiences. Scientific Reports, 7, 44292.

11 ‘My horses and hogs and even everybody seemed changed’ Appreciating beauty in depression recovery Tasia Scrutton

Introduction Narratives of recovery from periods of mental suffering such as depression and melancholy will sometimes point to the positive benefits of having experienced mental suffering once the person has recovered from it. Commonly discussed benefits include increased compassion, understood both as insight into others’ pain and motivation to do something about it, traits such as trust and love that enable the capacity to live better with oneself and to have better relationships with others, and an increased recognition of the value of the things that matter most. For example, among the people who report experiencing a greater motivation to help others, Andrew Solomon writes that, if you have been depressed, ‘you lose some of your fear of crisis’, and become more likely to help people going through the same kind of experience: If you have been through such a thing, you cannot watch it unfold in the life of someone else without feeling horrified. It is easier for me, in many ways, to plunge myself into the sorrow of others than it is for me to watch the sorrow and stay out of it. […] Not interfering is like watching someone spilling good wine all over the dinner table. It is easier to turn the bottle upright and wipe up the puddle than it is to ignore what is going on. (Solomon, 2001, 499) Others report developing virtues that enable them to be happier and have better relationships with others. For example, Henri Nouwen says that following his experience of depression: Reading [my journal] makes me aware of the radical changes I have undergone. I have moved through anguish to freedom, through depression to peace, through despair to hope. [My depression] certainly was a time of purification for me. My heart, ever questioning my goodness, value, and worth, has become anchored in a deeper love and thus less


Tasia Scrutton dependent on the praise and blame of those around me. It also has grown into a greater ability to give love without always expecting love in return. (Nouwen, 2009, 97–98)

These accounts draw attention to what we might call broadly ‘moral’ characteristics: the person who has experienced depression becomes, as a result, more compassionate, more trusting, more loving, more hopeful, more insightful, less egotistical and/or more spiritual (see also Palmer, 2015). Literature that involves second-order theorising about these accounts, ranging in subject from pastoral care to analytic psychology to self-help books to Shamanic-influenced spirituality, has often focused on the image of the wounded healer as a theme that unites the moral virtues sometimes developed in the wake of mental suffering (Jackson, 2001). A less-theorised theme in accounts of recovery from mental suffering concerns primarily aesthetic rather than moral characteristics. As I will argue, people who have experienced mental suffering sometimes report a heightened appreciation of beauty, and especially of the beauty of the natural world, during and following recovery. In this chapter, I will explore this undertheorised theme in depression recovery accounts, looking at sources including published autobiographies that focus on depression and recovery from depression, the accounts William James discusses in talking about melancholy and religious conversion, and reports from the archives of the Alister Hardy Religious Experience Research Centre (RERC), which have recently been made available online. Many of the accounts I draw on are from people who identify (in diverse ways) as religious or in other ways spiritual. This might be for several reasons, some of which relate to the relationship between spirituality and natural beauty, and some of which relate to how and why I came across these accounts. In terms of the latter, this chapter is part of a larger project on religious and spiritual interpretations of depression and how those interpretations shape the experience of depression. Consequently, many of the sources I have read have been by people who identify, in different ways, as spiritual or religious. In terms of the former, a heightened appreciation of beauty may give rise to a religious or spiritual outlook, as is clear from some of the Jamesian and Alister Hardy accounts – and so it may be that there simply are a greater number of religious/spiritual accounts of this kind than non-religious or non-spiritual ones. In addition to this, it may be that people who are already religious or spiritual are more likely to articulate a heightened appreciation of beauty, because it confirms certain aspects of their worldview (for example, that the world is divinely created, or that the world is good). As a result, it may have a significance attached to it that people with a non-spiritual outlook do not attach to it. Finally – and more tentatively – it may be that people with a religious or spiritual worldview are more likely to have the experience of having a heightened appreciation of beauty following a period of mental suffering, because some religious and spiritual worldviews

Appreciating beauty in depression recovery


encourage people to find value in suffering and to focus on benefits accruing from it, and to perceive the world as beautiful and good. This outlook may make people more open to certain experiences, such as having a heightened appreciation of beauty. Whatever the truth of the matter, the focus of this chapter is not on the influence of religious and spiritual worldviews on the appreciation of beauty (see Wynn, 2018, for a discussion of this theme). Rather, my focus is on the idea that a heightened appreciation of beauty in recovery is relevant to the question of the value of suffering. This is because the heightened appreciation of beauty in recovery does not seem to be merely a return to the state that existed before the period of mental suffering, but something valuable that goes beyond it that the person did not have before. In a nutshell, my argument is this. When recovering or recovered from depression, as the sources I consider highlight, some people seem to have a heightened appreciation of beauty, both when compared to when they were depressed, and also when compared to their pre-depression state. I am not suggesting that mental suffering and recovery are necessary for every person to appreciate a heightened appreciation of beauty, but only that mental suffering and recovery seem to have caused a heightened appreciation of beauty among some people. By ‘appreciation of beauty’, I mean seeing or noticing beauty in things, and also having an affective response to it that has a positive valence (for example, experiencing wonder and joy in response to it).

Terminological issues As is now generally recognised in both the philosophical and the psychiatric literature, ‘depression’ and ‘melancholy’ are best regarded as cluster concepts and not (essentialist) natural kinds (see Haslam, 2002). I am treating depression and melancholy as roughly similar experiences such that aspects of them (such as how perception of the world is altered in them) can be discussed together. This is not because I think they are necessarily exactly the same experiences (which would suggest more of a natural kinds view) but because I think there is enough overlap between the various experiences they involve that people diagnosed with melancholy in the past would usually be diagnosed with depression today, and that the phenomena they describe are broadly similar. At times, I will also use the term ‘mental suffering’ as a general term for depression, melancholy, and related experiences. A benefit of this term is that it gets away from the medicalising connotations of the terms ‘depression’ and ‘melancholy’ (see Pipher, in Simon, 2014, 35–44; Swinton, 2015). However, as one of the aims of this research is to bring meaning-making and medical understandings of these experiences further into conversation with one another, I do not want to jettison terms such as ‘depression’ and ‘melancholy’ altogether. As the idea that depression can give rise to a heightened


Tasia Scrutton

appreciation of beauty is hopeful, and thus in some sense therapeutic, I want this research to be accessible to people who have the experiences our culture generally labels ‘depression’ and to people who (whether in medical or other ways) seek to help them by using different terms. In addition to this, many of the accounts I look at talk about ‘depression’ or ‘melancholy’, and so it is consistent with these people’s understandings of their experiences to use these terms.

Methodological issues I am taking the reports of the experiences of depression and melancholy recovery at face value, rather than treating them solely as narrative constructions indicative of the ways in which people have been taught to speak about depression and melancholy recovery. This is not to deny that people, including the people writing these accounts, do make sense of their experiences in ways that are in line with their culturally formed expectations about experiences and do articulate them in culturally formed ways. Instead, I suggest, we have no reason to think that, because of this, people’s reports of these experiences bear no relation to their experiences of things. Rather, people’s experiences include their socially influenced interpretations of those experiences such that, by and large, and unless we have particular reasons to think otherwise, we should think of such accounts as both culturally formed ways of describing and interpreting experiences, and also authentic reflections of what people’s experiences are. It may, of course, be that there are people who (for religious or other reasons) tend to make sense of experiences in terms of personal development, and these people are more likely to talk about these experiences (and perhaps also to have them). However, whether this is the case does not affect my argument that these experiences occur, and is left to one side. A further methodological issue concerns how I locate myself in relation to these experiences. My approach to these accounts is to act as an interpreter, and not to attempt to be an objective scientist. As a result, my account of these experiences is informed, or biased, by my own experiences. Having had some experience of depression myself, the interest in the topic of recovery and the appreciation of beauty emerged for me from my own experience. I have also reflected on my own experiences when thinking about the puzzles the accounts raise. I have tried to remain true to the reports of other people, and to take into account the diversity of experiences of mental suffering, such that I recognise that people’s experiences may differ from my own. Nevertheless, an experience-driven sense that depression can result in a heightened appreciation of beauty (after the event) drives this chapter, and there is probably some truth in Nietzsche’s remark at least here that a work of philosophy is an implicit ‘confession on the part of the author and a kind of involuntary and unconscious memoir’ (Nietzsche, 2003, 37). This does not worry me.

Appreciating beauty in depression recovery


Presuppositions I will presuppose that appreciating beauty is valuable, and having a heightened appreciation of beauty more so. We might think appreciating beauty is valuable for a number of reasons: ethically (it may mean we practice certain virtues by appreciating beauty, or, more instrumentally, mean we behave better towards the world); hedonically (experience of beauty gives us a certain kind of pleasure); therapeutically (experience of beauty is healing, and/or prevents us from being as harmed by stressors as we otherwise might); epistemically (experiencing beauty may show us something about the way the world really is); aesthetically (appreciation of beauty may be irreducibly good). The precise nature of the value of the appreciation of beauty of the people whose accounts are discussed here might be mentioned in passing in their accounts, but for the most part it is outside the scope of this chapter. In addition to presupposing that appreciating beauty is valuable, I will also presuppose that appreciating beauty is valuable whether it is ongoing (as in some of the accounts we shall see), or temporary or intermittent (as in others). Of course, the reasons for why the appreciation of beauty is valuable may alter, depending on (among other things) whether it is permanent, temporary or intermittent. For example, we might think that a permanent appreciation of beauty is more conducive to ongoing well-being by providing a psychological buffer to stress. In Rachel Carson’s words, ‘Those who contemplate the beauty of the earth find reserves of strength that will endure as long as life lasts’ (Carson, 1998, 100). On the other hand, having had a heightened appreciation at a particular time that did not last for ever might be sufficient to provide, as one person puts it, an ongoing sense of rebirth and of love and regard for life (Alister Hardy RERC, undated, 003641). The different reasons for why permanent, temporary, and intermittent heightened appreciation of beauty is valuable is a large topic outside the scope of this chapter.

Appreciating beauty in recovery from periods of mental suffering States such as joy, happiness, sadness and depression are usually thought of as affective rather than perceptual states: as things that describe how we feel rather than describing how we see the world. Nevertheless, when we look at people’s descriptions of various states, a common feature of both positive and negative states is an altered perception of the world (see Wynn, 2013; Ratcliffe, 2015). Thus, William James argues that one of the changes in both melancholy and recovery from it is the ‘objective change which the world often appears to undergo’ (James, 1906, 243). James focuses on accounts of religious melancholy, which he regards as one of several types of melancholy. These experiences are followed by a recovery which takes place in the context of religious conversion, which James regards as ‘one out of many ways of reaching unity’ and thus of remedying the ‘inner incompleteness’ and


Tasia Scrutton

reducing the ‘inner discord’ we find in melancholy (James, 1906, 174). James’s account suggests that a changed perception of the world in recovery from melancholy will be true both of people who recover from melancholy through religious conversion, and also people who recover from melancholy in other ways, though the former are the focus of James’s discussion. In conversion, the perception of the world is positively altered as a result of the new, nonmelancholic affective state. James cites a number of accounts as evidence, of which the following are a few: The appearance of everything was altered; there seemed to be, as it were, a calm, a sweet cast […] in almost everything. God’s excellency, his wisdom, his purity and love, seemed to appear in everything; in the sun, moon, and stars; in the clouds and blue sky; in the grass, flowers, and trees; in the water and all nature […] And scarce anything, among all the works of nature, was so sweet to me as thunder and lightning; formerly nothing had been so terrible to me. Before, I used to be uncommonly terrified with thunder, and to be struck with terror when I saw a thunderstorm rising; but now, on the contrary, it rejoices me. (Edwards, cited in James, 1906, 243)

The very heavens seemed to open and pour down rays of light and glory. Not for a moment only, but all day and night, floods of light and glory seemed to pour through my soul, and oh, how I was changed, and everything became new. My horses and hogs and even everybody seemed changed. (Anon, cited in James, 1906, 245) It was like entering another world, a new state of existence. Natural objects were glorified, my spiritual vision was so clarified that I saw beauty in every material object in the universe, the woods were vocal with heavenly music. (Anon, cited in James, 1906, 244) […] everything looked new to me, the people, the fields, the cattle, the trees. I was like a new man in a new world. (Bray, cited in James, 1906, 244) I cannot explain how I felt. It was as if I had been in a dark dungeon and was lifted out into the light of the sun. (Anon, cited in James, 1906, 251) Accounts of a change in the person’s perception of the material world during recovery from a period of mental suffering are not limited to the experiences James finds in cases of religious conversion. For example, in an

Appreciating beauty in depression recovery


1888 letter to the French painter Émile Bernard, Van Gogh gives advice on Bernard’s melancholy (something Van Gogh himself had been afflicted with). Among other things, Van Gogh consoles his friend, ‘But what I wanted to say is this: after the period of melancholy is over you will be stronger than before, you will recover your health, and you will find the scenery round you so beautiful that you will want to do nothing but paint’ (Van Gogh, 1888). A more recent account comes from a contributor to the Alister Hardy RERC who writes that, following the sudden death of her husband: ‘The shock and deep depression took some years to live through, but gradually my mind became conscious of […] a more acute sense of the beauties that are around us and a thankfulness which I was not aware of before’ (Alister Hardy RERC, undated, 000435). Another contributor to the Alister Hardy archive writes that she experienced ‘a sense of meaning and wholeness’ and ‘a sense of joy’ in the aftermath of deep depression, and that ‘this joy comes almost always in connection with some growing thing’: in winter and early spring the growing bulbs will do it. In early April this year I walked with my dog in a country lane (filled actually with rubbish: decaying mattresses etc.) & had the most clear sense of this wholeness: the trees were in early leaf, intensely green & the sky was of a rich blue that I can never recall seeing before […] This whole experience was repeated for a few moments for a period of three or four days. I have never before had such a continuous feeling of joy, love and serenity. (Alister Hardy RERC, undated, 003156) This writer relates that she regards the depression as the ‘price’ of the intense joy. In these accounts, the new perception of the material world occurs after the experience of mental suffering. Some other writers report a new perception of the material world at the time of the experience of depression, which they regard as a hiatus in the mental suffering, and also, in some cases, as the beginning of recovery from it. For example, Kimberlee Conway Ireton writes that: In the midst of that time [of depression], which I mostly remember as if it were shrouded in thick, dark clouds, I can recall one moment when these clouds parted and I was able to see the reality beyond the one in which I was trapped. It happened, of all places, in the kitchen. I was washing a bunch of Swiss chard in the sink when suddenly I became aware of how beautiful it was – the crinkly green leaves with their bright red veins, the thick yet silky texture of the leaf as I gently pulled apart each fold to wash inside it, the way the leaves glistened in the sunlight slanting through the kitchen window […] Time seemed to stop – or at least cease to matter – as I wondered at the beauty of the chard […]


Tasia Scrutton It didn’t instantaneously end my depression and bring me to a place of joy. But it stirred my desire to love. It enticed me to notice and pay attention to the world around me. And at a time when I felt hopeless, this moment of mystery gave me hope that there is more to life – my life, the life of the world – than usually meets the eye, or the ear, or any of my physical senses. In the moment when the veil parts, we see the not-yet now, we glimpse the mystery and beauty at the heart of all that is, we see things as they really are and not as they usually appear. (Ireton, 2008, 113, 114)

Lauren Slater reports sudden and striking changes in perception, and especially to her attention to the colour of the sun and fruit, at the beginning of her recovery from depression: The purple silk of a plum. Sun on a green plate. Over the next few days, even in the thicket of obsessions, these genuine moments occurred – perhaps they always had but I had never noticed them or given them their value – split-second snappings of the shutter, the click of freedom. Then closed. (Slater, 1998, 125) Another contributor to the Alister Hardy research archive reports, in the context of having experienced both ongoing background depression and also grief at the death of her husband, an experience in which her grief and depression were transformed: […] without Robert the years stretched ahead in an endless meaningless succession; yet I knew at first hand that ‘God’ is true, and feels like love. Robert was safe. And for the first time in my life I experienced, for a matter of weeks, what it is to be absolutely free. (From fear? From Time?) One day I saw some beech leaves in silhouette against a clear sky: I stood and stared; their shapes were unbelievably perfect. From that moment I was dazzled by the sheer beauty, ingenuity and variety of forms in nature; they were everywhere I looked – in growing things, insects, birds, animals: all intensely significant, as though reflecting something more than themselves. (Alister Hardy RERC, undated, 003469) Although she doesn’t use the language of recovery, this contributor describes this state as being characterised by ‘an astonishing synthesis of serenity and grief ’, contrasted with the more anxious and listless grief that came before. Another contributor to the Alister Hardy archive speaks of a period of mental suffering in the context of difficult significant life changes, in which he has two experiences in which the world is transformed: ‘The very bricks and metal around me glowed with life’; the colours ‘were almost luminous, the

Appreciating beauty in depression recovery


drabbest of buildings shone with light and were alive’ (Alister Hardy RERC, undated, 003641). These experiences were accompanied by a ‘tremendous sense of serenity’ and filled him with an enormous sense of love for the people around him. On the second occasion, the intense perceptual experience or ‘glow’ lasted the whole day and part of the next. When it left him, the sense of rebirth and of love and regard for life remained. There are both similarities and differences between these accounts. Some of them describe an ongoing or permanent appreciation of beauty (e.g. Alister Hardy RERC, undated, 000435), while for others the peculiar sense of beauty is intense and short-term, but may give rise to an ongoing appreciation of life (e.g. Alister Hardy RERC, undated, 003641). For still others the sense of beauty is intermittent (e.g. Alister Hardy RERC, undated, 003156). The Jamesian accounts in particular note a sense of the ‘newness’ or ‘cleanness’ of the material world (James, 1906, 243), while the more recent accounts – perhaps reflecting differences in narrative conventions, conceptual categories or linguistic preferences of different periods – use more often the language of the ‘acuteness’ or ‘intensity’ of the experience, or sense of the ‘reality’ of the world (Alister Hardy RERC, undated, 000435, 003156; Ireton, 2008, 113, 114). For most of the accounts, the intentional object seems to be the whole world, though several also note having their attention drawn to specific objects – intriguingly, often ones that are notably quotidian or mundane (the ‘crinkly green leaves’ of Swiss chard; bricks and metal, grass, insects, cattle, horses and hogs). And the intentional objects seem overwhelmingly to involve (what is likely to be construed as) nature: only report 003641 of the Alister Hardy archive speaks of metal or bricks, and his context seems to be urban. Interestingly none of the accounts I encountered talk about works of art. What seems to me to unite these accounts is a heightened appreciation of beauty, which involves both noticing beauty and responding to it with joy, gratitude and wonder. Thus, the first contributor to the Alister Hardy archive writes that she has ‘a more acute sense of the beauties that are around us and a thankfulness which I was not aware of before’ (Alister Hardy RERC, undated, 000435). Van Gogh writes from his own experience that, following the experience of melancholy, his friend will find the scenery around him ‘so beautiful’ that he will only want to paint (Van Gogh, 1888). One of the people cited by William James says that he ‘saw beauty in every material object in the universe’, and James, who himself experienced melancholy, regards these experiences as characterised by a ‘beautiful newness’ (James, 1906, 244, 243). Ireton characterises her experience both as awareness and wondering at the beauty of the chard, which she regards as glimpsing the ‘beauty and mystery’ at the heart of all that is (Ireton, 2008). Another of the contributors to the Alister Hardy archive writes that she was dazzled by the ‘sheer beauty’ of nature, and another speaks both of the experience itself as beautiful, and of the beauty of the people who were on the bus with him (Alister Hardy RERC, undated, 003469, 003641). Beauty is responded to with wonder and joy: thunder becomes a source of ‘rejoicing’; there is ‘a sense of


Tasia Scrutton

joy’; a ‘wondering’ and a ‘moment of mystery’; an experience of being ‘dazzled’; a ‘sense of rebirth and love of life’ (Edwards, cited in James, 1906, 243; Ireton, 2008, 113; Alister Hardy RERC, undated, 003469, 003156, 003641). Very basically then, both particularly noticing beauty and responding to it with a high level of joy and wonder seem common to these people’s experiences of depression and melancholy recovery. It seems reasonable to refer to these experiences as a heightened appreciation of beauty. Although this topic is under-theorised, this conclusion seems to have some support in a psychology study. In a 2006 research project, Christopher Peterson, Nansook Park and Martin E. P Seligman investigated whether there was any evidence for the claim that people develop positive character traits, including appreciation of beauty, following episodes of both ‘physical’ and ‘psychological’ illness. In order to achieve this, they recruited 2,087 adults to fill in an online questionnaire about subjective wellbeing and character strengths (such as appreciation of beauty, courage, kindness, love, and hope). Participants began by reporting the degree to which statements applied to themselves (e.g. ‘At least once a day, I stop and count my blessings’), where these statements were thought by the researchers to reflect particular character strengths (in the above example, thankfulness). Participants were then asked about life satisfaction via a self-report questionnaire where they rated whether they agreed or disagreed with statements such as ‘The conditions of my life are excellent’. Finally, participants were asked whether they had ever experienced a serious physical or psychological problem, whether they had ever recovered for it, and what kind of problem it was (e.g. arthritis, cancer, depression, schizophrenia). By introducing this question at the end of the questionnaire (and with no alternative to go back and change answers to previous questions), the authors hoped to ameliorate the possibility that participants are primed to report increased well-being or character strengths in light of common ‘survivor’ memes (Peterson, Park and Seligman, 2006, 18; see Frank, 1997). Peterson, Park and Seligman found that around 51% of participants reported having had a psychological problem. For those individuals who had recovered from a psychological problem, an appreciation of beauty was somewhat higher than for people without a history of psychological problems (for individuals who recovered from physical problems, it was only slightly higher). The most common psychological problem was depression, which was reported by 27.9% of the overall group of participants, and so represented the largest number of people who reported having had psychological problems, though the data analysis shows only the relationship between appreciation of beauty and psychological problems in general, rather than depression in particular. Peterson, Park and Seligman’s study seems to support and supplement my argument by pointing to a far larger group of people who report this kind of phenomenon. Of course, because it doesn’t distinguish between depression and other psychological problems, the study is only suggestive for positing a relationship between depression recovery and appreciation of beauty.

Appreciating beauty in depression recovery


Furthermore, by virtue of its quantitative nature, Peterson, Park and Seligman’s study is relatively limited with respect to what it can tell us about the quality of people’s experiences of beauty, or what the appreciation of beauty involves.1 Yet, taken alongside one another, their study and my more qualitative exploration suggest that we have some reason to think that at least some people recovering from depression or melancholy experience a heightened appreciation of beauty.

Two objections Two objections are worth considering here.2 First, that the heightened appreciation of beauty may in fact suggest misdiagnosed bipolar disorder, rather than recovery from unipolar depression – with the heightened appreciation of beauty being part of the manic or hypomanic phase of the bipolar cycle. This possibility is worth taking seriously, because bipolar disorder is sometimes misdiagnosed as unipolar depression, and because if the heightened appreciation of beauty is in fact part of a manic/hypomanic episode, this may affect whether and how we regard it as valuable (though it doesn’t entail that we regard it as not having a positive value). Second, that at least some of the intentional objects (perhaps Swiss chard, or bricks and metal) are not in fact beautiful, and so the person’s recovery mood is misleading or epistemically vicious in making her think that something is beautiful that is not.3 In response to the first objection: i While claims have been made that Van Gogh, for instance, did have bipolar disorder, it seems unlikely that all the people writing these accounts have bipolar disorder. Ireton, and the woman experiencing longterm depression following her husband’s death, for example, are writing significantly after the depressive episode, and what they describe following it does not sound similar to a manic/hypomanic episode. ii In addition, joy and wonder at beauty do not seem to be common characteristics of mania/hypomania. While a popular perception of mania/ hypomania is that these states involve large amounts of pleasure, in fact they are often characterised by irritation, agitation, sleeplessness, pressure to keep talking, and racing thoughts, which are not especially associated with pleasure, and may even have a negative valence. While excessive involvement in ultimately costly activities that are typically pleasurable (e.g. over-spending, sexual promiscuity) are symptoms of mania/hypomania, these typically pleasant experiences are not always experienced as pleasant during manic/hypomanic episodes. So manic and hypomanic episodes do not seem particularly likely to account for the experiences of joy and wonder these accounts describe. iii Where there does seem some scope for seeing these experiences in manic/ hypomanic terms (e.g. Alister Hardy RERC document 0003641 reports bright or luminous colours, which is not a DSM symptom of mania or


Tasia Scrutton hypomania but which is noted by Mind as a possible experience associated with it), we might want to say, given the sense of serenity and love that accompanied and continued beyond it, the experience was as valuable as the experience following unipolar depression recovery – even if we also wanted to say that other symptoms of mania/hypomania and the experience of bipolar more generally are not desirable. If this were the case, in order to include this as an example of the value of suffering, it would also be important that the appreciation of beauty during mania/ hypomania were causally related to the prior depressive episode. There is not scope to argue for that claim here, since this chapter concerns unipolar rather than bipolar depression – but suffice it to say that (i) and (ii) seem sufficient in defending my claim that the heightened appreciation of beauty described in the accounts I have discussed are best characterised collectively as occurring in the context of recovery from unipolar depression.

What might we want to say in response to the second objection – the concern that some of the accounts involve a faulty attribution of beauty to non-beautiful objects? This would limit the possible value of the aesthetic experience to, say, therapeutic or ethical rather than epistemic goods (it means the experience would not tell us anything about the way the world actually is, even if it enabled us to live better in it). Perhaps more seriously, it’d mean that the experience is epistemically vicious, making us perceive, and encouraging us to believe, that something is beautiful when in fact it is not. Subjectivists, objectivists, and inter-subjectivists about beauty will respond to this objection differently. Subjectivists will see the concern as void: if something is beautiful to me, then it is beautiful. Objectivists will have a harder time, since it may be that Swiss chard is not in fact beautiful. However, we don’t have a clear way of knowing what things are objectively beautiful, even if we think there is a fact of the matter – and so the possibility that Swiss chard is not beautiful is difficult to discern either way. In this case, the burden of proof seems to be on the objector to argue that Swiss chard is not beautiful, since this does not seem obvious to me. Finally, if we are ‘inter-subjectivists’ – if we think we can speak of a fact of the matter about beauty, but that this is grounded by the consensus of a wider community rather than having an extra-human objective basis – then, again, the burden of proof seems to be on objectors to say that there is a consensus that Swiss chard and bricks and metal and so on are not beautiful. This does not clearly seem to be the case.

Conclusion: the value of suffering In this chapter, I have defended a relatively simple and modest claim: that, in depression recovery, some people experience a heightened appreciation of beauty. How does this relate to the theme of this book – the value of suffering? In at least some, if not all, of the recovery descriptions we have

Appreciating beauty in depression recovery


looked at, the change in recovery seems to be a heightened appreciation of beauty that goes beyond the way beauty was experienced in the pre-depression state. As one of the contributors to the Alister Hardy archive puts it, ‘gradually my mind became conscious of a more acute sense of the beauties that are around us and a thankfulness which I was not aware of before’ (Alister Hardy RERC, undated, 000435). Slater says that while moments involving seeing sun on a green plate or the purple silk of a plum may have occurred before, she had ‘never noticed them before or given them their value’ (Slater, 1998, 125). Ireton speaks of her experience of being struck by the beauty of the Swiss chard in terms of glimpsing the beauty and mystery at the heart of all that is, and of seeing things as they really are, rather than as they usually appear (Ireton, 2008, 113–114). That the recovery state is not simply a reversion to a ‘normal’ appreciation of beauty is arguably also implicit in the accounts which focus on remarkably mundane things as being beautiful. A sense of the everydayness of the objects that are transformed into beautiful things pervades the recovery accounts: rather than a particularly beautiful mountain or sunset or seascape, it is Swiss chard and ‘my horses and hogs and even everybody’ that seem changed, and insects whose beauty is dazzling (Ireton, 2008, 113; James, 1906, 245; Alister Hardy RERC, undated, 003469).4 The striking perception of the beauty of objects usually regarded as very ordinary also suggests that there is something extraordinary about this appreciation of beauty – that it is a heightened appreciation of beauty rather than a return to a normal appreciation of beauty. This resonates with James’s conclusion that the recovery state is not a ‘mere reversion to natural health’ but a ‘deeper kind of conscious being than he could enjoy before’ (James, 1906, 156).

Notes 1 The ‘appreciation of beauty’ part of the questionnaire is very general: participants are asked to rate their ‘appreciation of beauty and excellence: noticing and appreciating beauty, excellence, or skilled performance in all domains of life’. 2 Thanks for these objections to participants at the SLU ‘Happiness and Wellbeing’ workshop, and the Ethics and Aesthetics research centre at the University of Leeds, where versions of this chapter were presented. 3 I’m indebted to Gerald Lang for this objection. 4 This may be important for the idea that the appreciation of beauty is important for ethical, and especially ecological, issues. Supporting the idea that appreciation of beauty is ethically and ecologically important, Yuriko Saito writes that ‘Without […] aesthetic attraction and emotional attachments cultivating a respectful attitude toward the land would be, if not theoretically impossible, a hardsell, psychologically and pragmatically’ (Saito, 2007, 71). However, objectors to this idea point out that people’s ecological actions are often in fact unhelpfully skewed by their aesthetic predilections – for example, members of the public will give money to support animals they regard as beautiful such as leopards, but not to support insects, which are often not regarded as so beautiful but which are essential to ecosystems. The problem here does not seem to me to be so much that the appreciation of beauty is ethically unhelpful, as that people tend to have too narrow a sense of what is


Tasia Scrutton

beautiful. If insects, for example, were also regarded as beautiful, these would be better cared for and supported. These accounts suggest that the appreciation of beauty can be broadened to include quite mundane objects. Furthermore, it suggests that experiences of mental suffering may be instrumental in facilitating this appreciation of the beauty of a wider range of objects. I think this might be worth exploring more, but I don’t have space to do it here.

Acknowledgements Thanks go to the John Templeton Foundation, and the SLU Happiness and Wellbeing Project, which supported my research of this topic. Thanks also go to the Centre for Ethics and Aesthetic Research Centre at the University of Leeds, and to a Value of Suffering Workshop at the University of Glasgow, where versions of this chapter were presented, and helpful feedback given. Particular thanks are due to Gerald Lang, Sam Clark, Simon Hewitt, Jennifer Corns, Michael Brady, Mark Wynn and Ian Kidd for great comments and insights that helped improve this chapter.

References Alister Hardy RERC. Undated. Alister Hardy RERC archive database. Retrieved from ne-archive. Carson, Rachel. 1998. The Sense of Wonder. New York: Harper & Row Publishers. Frank, Arthur. 1997. The Wounded Storyteller: Body, Illness and Ethics. Chicago, IL: University of Chicago Press. Haslam, Nick. 2002. Kinds of kinds: a conceptual taxonomy of psychiatric categories. Philosophy, Psychiatry and Psychology 9. 3, 203–217. Ireton, Kimberlee Conway. 2008. The Circle of Seasons: Meeting God in the Church Year. Downers Grove, IL: InterVarsity Press. Jackson, Stanley W. 2001. The wounded healer. Bulletin of the History of Medicine, 75. 1, 1–36. James, William, 1906. The Varieties of Religious Experience: A Study in Human Nature. Pennsylvania, PA: Pennsylvania State University Press. Nietzsche, Friedrich. 2003. Beyond Good and Evil (trans R. J. Hollingdale). London: Penguin Books. Nouwen, Henri. 2009. The Inner Voice of Love: A Journey through Anguish to Freedom. London: Darton, Longman and Todd. Palmer, Parker. 2015. ‘Welcome to the human race’: an interview with Parker J. Palmer on the topic of depression. Retrieved from (accessed 9 August 2016). Peterson, Christopher, Park, Nansook, and Seligman, Martin. 2006. Greater strengths of character and recovery from illness. The Journal of Positive Psychology 1. 1, 17–26. Ratcliffe, Matthew. 2015. Experiences of Depression: A Study in Phenomenology. Oxford: Oxford University Press. Saito, Yuriko. 2007. Everyday Aesthetics. Oxford: Oxford University Press.

Appreciating beauty in depression recovery


Simon, Tami (ed.). 2014. Darkness before Dawn: Redefining the Journey through Depression. Boulder, CO: Sounds True. Slater, Lauren. 1998. Prozac Diary. London: Penguin Books. Solomon, Andrew. 2001. The Noonday Demon: An Anatomy of Depression. London: Vintage Books. Swinton, John. 2015. Theology or therapy? In what sense does depression exist? Philosophy, Psychiatry and Psychology 22. 4, 295–298. Van Gogh, Vincent. 1888. Letter from Van Gogh to Émile Bernard, Arles, c. 18th June 1888. Retrieved from (accessed 5 August 2016). Wynn, Mark. 2013. Renewing the Senses: A Study of the Philosophy and Theology of the Spiritual Life. Oxford: Oxford University Press. Wynn, Mark. 2018. Aesthetic experience and spiritual well-being: locating the role of theological commitments. International Journal for Philosophy and Theology 79. 4, 397–409.

Part IV

The normativity of suffering

12 Hedonic rationality Jennifer Corns

Introduction Philosophers traditionally recognize two domains of rationality: practical and epistemic. Epistemic rationality is traditionally concerned with the rationality of our doxastic states, i.e., beliefs and judgments. It has seemed obvious that we sometimes have reasons to believe or judge that something is the case, and that an agent is sometimes rational or irrational in virtue of their doxastic states. Practical rationality is traditionally concerned with the rationality of our actions. It has seemed obvious that we sometimes have reasons to do or, at least, intend to do things, and that an agent is sometimes rational or irrational in virtue of what they do or intend to do. Reasons to believe and do have thus been well recognized and debated. In this article I aim to make plausible that there is at least one other domain of rationality yet to be discussed: hedonic rationality. Hedonic rationality is concerned with the rationality of the hedonics (i.e., the pleasantness or unpleasantness associated with a wide range of mental phenomena). I aim to make plausible that we sometimes have reasons to feel pleasantly or unpleasantly, and that an agent is sometimes rational or irrational in virtue of the hedonic tone of their mental states. For establishing this domain, I focus on negative hedonic tone (i.e., unpleasantness).1 Three preliminary clarifications are in order. First, unpleasant mental episodes are complex, and I am here only focused upon the negative hedonic tone associated with these episodes as against any other features that the mental episodes may have. Negative hedonic tone is unpleasantness—the unpleasantness that may be a feature of, e.g., experiences of pain, tiredness, hunger, disappointment, and so on. So, while an unpleasant state or process may have many other rationally assessable features, I am interested in whether any mental phenomena, and in turn any person, is rationally assessable in virtue of the associated unpleasantness. Second, I do not intend to focus on whether unpleasantness ever provides reasons, but instead on whether it is ever responsive to reasons. Recent debates have focused on, but only on, the former.2 Though I am strongly inclined to think that unpleasantness is at least sometimes reason-providing, I mean to here focus on whether negative


Jennifer Corns

hedonic tone is ever reason-responsive. Whether, for instance, one ever has reasons for the unpleasantness of a headache, regardless of whether or not one also thinks the unpleasantness of the headache provides a reason to, say, take an aspirin. Third and finally, in establishing this novel domain, I intend broad ecumenism. In particular, accepting hedonic rationality need not turn on one’s preferred hedonic theory. It should be uncontroversial that unpleasantness is associated with a wide range of mental states and I intend neutrality about the controversies concerning both the nature of that association and the nature of the unpleasantness. In particular, I intend neutrality between positing negative hedonic tone as a feature of independently characterized state types (e.g., thoughts or emotions) as against instead positing negative hedonic tone as a sui generis state (i.e., a distinct feeling of unpleasantness). I will thus move fluidly between referring to the unpleasantness of a state and an unpleasant state. Hedonic rationality is thus intended to be compatible with hedonic adverbialism, feeling-tone theories, desire theories, evaluativism, and so on. My use of the term “hedonic tone” to refer to the pleasantness and unpleasantness associated with a wide range of mental phenomena no doubt suggests my own sympathies, but one need not share them to accept the arguments or the existence of the domain they are intended to support. As discussed below, I likewise intend neutrality concerning theories of reasons or rationality. Relatedly, I aim to show that hedonic rationality is plausible whether one focuses on the reasons for the associated hedonics of an episode or the rationality of the person in virtue of those associated hedonics. Similarly, I intend neutrality concerning representationalism (strong, weak, or neither), internalism versus externalism (for both reasons and motivation), and further foundational questions in epistemology and metaphysics (e.g., epistemological foundationalism or ontological dualism). It is a virtue of hedonic rationality that it may be embraced by such a broad church. As with epistemic or practical rationality, it is expected that a theory of hedonic rationality will be enriched by one’s further commitments. The plan for establishing this domain is as follows. First, I distinguish the proposed hedonic rationality from the rationality of the emotions. Though instructive, the rationality of the emotions is orthogonal to hedonic rationality and the two should not be conflated. I then address methodological problems determining whether any mental phenomenon is rational and outline my ecumenical approach. Next, I offer three case descriptions that support hedonic rationality and which are bolstered by the use of three positive indicators of rationality drawn from my earlier argument. I then discuss four seemingly negative indicators for hedonic rationality cited earlier, and I treat these as objections and argue that they instead highlight choices for enriching one’s theory of hedonic rationality, rather than undermining its plausibility. Finally, I draw some conclusions.

Hedonic rationality


Hedonic rationality and the rationality of the emotions The last twenty years have seen increased attention to the emotions in ways that complicate the traditional idea that rationality is exhausted by the practical and the epistemic. Do we have reasons for our emotions and is an agent ever rational or irrational in virtue of their emotions? Opinions in this growing and increasingly sophisticated literature diverge. The rationality of the emotions, when granted, is posited as being in virtue of an emotion’s posited doxastic component, motivational aspect(s), intentionality (i.e., feeling towards), or some combination. Important work concerning the rationality of the emotions includes, for instance, Michael Brady (2013), Ronald de Sousa (1987), Nico H. Fridja (2007), Peter Goldie (2012), Patricia Greenspan (1988), and Robert C. Solomon (2006). Emotions and hedonics, however, are distinct mental phenomena that arguably doubly dissociate. A wide range of mental phenomena—including bodily sensations, thoughts, and desires—have associated hedonics.3 So, for instance, the thought that I am having dinner with a friend this evening is, we might colloquially say, a pleasant thought. Less controversially, being hungry and being tired are unpleasant. Despite their associated hedonics, one may resist classing these as emotions. On the other hand, emotional episodes arguably dissociate from hedonics. Surprise, it seems, may be sometimes pleasant and sometimes unpleasant. Grief may sometimes be neither pleasant nor unpleasant, but instead hedonically blank or neutral. So again, though emotions are hedonic paradigms, emotions and their paradigmatically associated hedonics arguably doubly dissociate. Even if one resisted these dissociations, the rationality of the emotions is distinct from the rationality of their hedonics—even if we grant that the latter are necessary or essential to the emotions. Emotions and emotional episodes are complex and no proponent of the rationality of the emotions argue for the rationality of affect, or hedonics, as such—instead, they focus on some other feature or combination of features.4 Michael Brady (2018), as a clear instance, argues that while emotional types of suffering, like grief and rejection, are reason-responsive, physical types of suffering, like pain and hunger, are not. It is thus ruled out that it is the common negative hedonic tone which is rational—both types of experiences are unpleasant forms of suffering, but the physical forms are nonetheless arational. One may thus hold that the emotions are sometimes rational while denying that hedonic tones ever are. One may, for example, allow that while grief is sometimes rational or irrational, the unpleasantness associated with grief never is. Indeed, one might hold that emotions like grief are rationally assessable precisely because one holds that there is more to emotional episodes than associated pleasant or unpleasant feelings. At the same time, should one be persuaded of hedonic rationality, one may nonetheless resist that emotional episodes as such are ever rational. One might hold that there are never reasons for grief as such, though there are reasons for the hedonics (at least


Jennifer Corns

paradigmatically) associated with grief. The questions are thus orthogonal. Though instructive, existing arguments for the rationality of the emotions are not arguments for hedonic rationality and the arguments for hedonic rationality below are not arguments for the rationality of the emotions. Whether hedonic rationality is distinct from the rationality of the emotions has theoretical implications. Briefly consider two. First, hedonic rationality—as distinct from the rationality of the emotions—has important implications for contemporary debates concerning the nature of the emotions. Consider recalcitrant emotions. When you believe that you are not in danger but feel fear, for example, your feeling of fear is a recalcitrant emotion. Recalcitrant emotions have been sometimes offered as an objection to both (neo)cognitive and feeling theories of emotions: the former because they render recalcitrant emotions too irrational, and the latter because they have seemed unable to account for any irrationality whatever. If hedonics is itself rationally assessable, however, then these objections are too quick. Not only the rationality of the emotion, but the rationality of the hedonics, will have to receive an account. The rationality or irrationality of affective states will have to be generally reconsidered if it turns out that hedonic tone is itself rationally assessable. If there is a neglected, distinctive domain of hedonic rationality, emotions may not only be epistemically or practically rational (or both), but hedonically rational. As noted above, no one currently takes such a position. As a second and broader theoretical implication, notice that countenancing hedonic rationality appears to significantly extend the scope of rationality. There are traditional arguments concerning whether and how the space of reasons might extend to perceptual states. If unpleasantness is sometimes rational and irrational, might the space of reasons extend even beyond perception to hedonics? How might accepting this broadening of the space of reasons change how we think about the architecture of the mind or nonhuman animals? These questions pertain to the rationality of hedonics in particular, and not only the complex, emotional episodes with which they are paradigmatically associated. As such, they extend rationality to a wider range of episodes and creatures. In summary, whether the emotions are rationally assessable is orthogonal to whether hedonics are rationally assessable. Advocates of the rationality of the emotions do not advocate, and sometimes implicitly reject, hedonic rationality—a distinctive domain of rationality with distinct implications.

An ecumenical methodology To establish hedonic rationality, it is helpful to ask how we determine whether any target mental phenomena is reason-responsive. T. M. Scanlon might be thought to have provided the most direct answer. Scanlon (1998) offers an influential account that characterizes reason-responsive mental state types as judgment-sensitive attitude types. This doesn’t get us far

Hedonic rationality


for present purposes, however, since we can now simply ask: how do we determine when a mental state type is a judgment-sensitive attitude? Scanlon himself admits that this characterization is explanatorily empty (ibid., p. 20). We get something that appears more helpful when Scanlon goes on to characterize judgment-sensitive attitudes (p.20): The judgment-sensitive attitudes are attitudes that an ideally rational person would come to have whenever that person judged there to be sufficient reason for them and that would, in an ideally rational person “extinguish” when that person judged them not to be supported by reasons of the appropriate kind. (Scanlon 1998, p. 20) But this is not in fact helpful at present. Judgment-sensitive attitudes are here characterized as what the ideally rational person would have when they judged that there were reasons for them. So unless we already know which state-types have reasons for them, this characterization won’t help.5 Though Scanlon does not offer any criterial test for judgment-sensitive attitudes, he nonetheless offers examples of state types that are and state types that are not judgment-sensitive attitudes, i.e., reason-responsive state types (ibid.). “Mere feelings” are, of course, in the latter camp. Hedonics are arguably thus summarily dismissed. Why? Perhaps Scanlon thinks that which state types are reason-responsive, like reasons themselves, are primitive. Scanlon earlier says (ibid., p. 17) that he will “take the idea of a reason as primitive. Any attempt to explain what it is to be a reason for something seems to me to lead back to the same idea …” If it is supposed that it is obvious to determine, yet difficult to explain, what reasons are, perhaps the same is being supposed of which mental phenomena are reason-responsive. Perhaps, that is, Scanlon thinks that which mental state types are reason-responsive is not amenable to an explanation that is more basic than the ruling itself. The determination of which states are reasonresponsive is, like reasons, taken as primitive. This interpretation is suggested by Scanlon’s noting that judgment-sensitive attitudes are those for which “it makes sense” to demand reasons. From Scanlon then, rather than a direct answer to the question of how to determine whether any mental state type is ever rational, we seem to get new questions: Is it obvious that unpleasantness is never rational or irrational? Does it make sense to demand reasons for the unpleasantness associated with a state? In the following section, I offer three cases intended to answer these flat-footed questions: a case of rational unpleasantness, irrational unpleasantness, and unintelligible unpleasantness. The question of whether it is intelligible to talk about one’s reasons to feel unpleasant is addressed by the descriptions of these first two cases, and stands in contrast with the unintelligibility described in the third. Despite others apparently thinking it simply obvious that hedonics are


Jennifer Corns

never rational or irrational, I offer these cases to show that unpleasantness is sometimes rational, sometimes irrational, and where—even when the hedonics is unintelligible—it is legitimate to ask for reasons. Insofar as the reader finds the descriptions of these cases obvious and to make sense, it should be equally obvious and intelligible that unpleasantness is sometimes rational or irrational. If we can, however, we should go further than claims about what’s obvious and examples which merely make sense. In continuing ecumenism, I deploy seven dominant indicators of reason-responsiveness from the literature. My use of the word “indicator” is intended to flag neutrality. The goal is to establish the plausibility of hedonic rationality independent of any preferred theories of the nature of reasons and rationality—independently of which, if any, of the following indicators are accepted. There are three indicators which I will argue bolster the case for hedonic rationality. These are: 1. 2. 3.

There are considerations that speak in favor of X or that some agent takes to speak in favor of X. X is a mental phenomenon that is made intelligible by other mental phenomena and that makes other mental phenomena intelligible. X coheres with one’s other mental phenomena.

All three might be either individually or jointly taken as sufficient conditions for X to be rational or, depending on your preferred way of speaking, for an agent to be rational in virtue of X. While I remain neutral about which, if any, of these indicators we should accept as sufficient, using any or all of these three indicators bolsters the case for hedonic rationality. There are four indicators which may initially seem to undermine the case for hedonic rationality. These are: 1. 2. 3. 4.


is voluntary. is not a (mere) perceptual experience. has intentional, representational content. has an aim.

All four of these indicators might be either individually or jointly taken as necessary conditions for X to be rational or, again, for an agent to be rational in virtue of X. While I remain neutral about whether we should accept any of these indicators as necessary, I take these as objections later in this chapter and argue that none undermine hedonic rationality. My three-pronged ecumenical approach is therefore to establish (i) that it is natural and intelligible to talk about reasons for negative hedonic tone and the rationality of an episode or agent in virtue of negative hedonic tone, (ii) there are indicators of rationality in the literature that bolster these descriptions, and (iii) indicators of rationality from the literature which may seem to undermine them in fact do not.

Hedonic rationality


The cases In this section, I offer one case each of rational, irrational, and unintelligible negative hedonic tone. I first offer these cases, described as seems natural to me, briefly bolstering them with indicators of rationality from the literature. Case 1 Rational unpleasantness: Grieving Glenda For 25 years, Glenda was married to her beloved husband George. They had a quiet married life. Neither the best nor the worst of men, George worked at the bank. He took out the trash on the occasional Wednesday and brought Glenda flowers on the occasional Friday. Glenda and George fought only occasionally—and occasionally about the frequency with which trash disappeared and flowers appeared. Earlier today, George unexpectedly died in a car accident. Home from the hospital, Glenda wanders into her kitchen. Her head swims with thoughts of her future life without George and the arrangements that must be made. She leans on the kitchen counter to steady herself and glimpses the trash. It is full and, peeking out, are the remains of last Friday’s flowers. For the next hour, many of Glenda’s perceptions, perceptual beliefs, feelings, thoughts, and imaginings are unpleasant.6

Grieving Glenda, I take it, has good reasons for the unpleasantness associated with her states—the unpleasantness that she feels is rational. She has good reasons not only for her grief, but good reasons for the unpleasantness associated with her states. This description is bolstered by at least three indicators of rationality from the literature: 1. 2.


There are considerations that speak in favor of the unpleasantness—applying this indicator as used in, e.g., Scanlon (1998). The unpleasantness is made intelligible by the agent’s other mental states—applying this indicator as used to evaluate the rationality of beliefs and desires as in, e.g., Davidson (1963) or Helm (2001). The unpleasantness coheres with the agent’s other mental states—applying this indicator as used in discussion of practical and epistemic rationality as in, e.g., Smith (1995) or Parfit (2001).

Case 2 Irrational unpleasantness: Careful Clara Clara has always been careful. She double-checks that the coffee maker is off before leaving and triple-checks the locks before bed. Clara likes carnations. She keeps seven blue and seven white in each of three silver vases, placed carefully throughout her apartment. Clara is a good accountant: her numbers


Jennifer Corns go in rows. Her figures go in columns. Earlier today, Clara invited her friend Cloe over for the evening. Cloe has a key and lets herself in, armed with the surprise of fresh carnations. Clara comes home, chats with Cloe, and leans back contentedly, steadying herself on her kitchen counter. She glimpses the nearest silver vase. It holds seven blue and eight white carnations. For the next hour, many of Clara’s perceptions, perceptual beliefs, feelings, thoughts, and imaginings are unpleasant. If you asked her why her the mental episodes are so unpleasant for her, she’d lament that it was “because of the damned flowers.”

Careful Clara, I take it, does not have good reasons for the unpleasantness associated with her states—the unpleasantness that she feels is irrational.7 This case description is likewise bolstered by the three positive indicators of rationality from the literature, applied as in that literature, above: there are considerations that Clara takes to speak in favor of the unpleasantness, the unpleasantness that she feels is irrational insofar as it is made unintelligible by and clashes with her other mental states. The unpleasantnesses featured in these cases are plausibly described as rational and irrational respectively; the agents’ feeling of unpleasantness is rational or irrational, and in virtue of this unpleasantness the agent is rational or irrational.8 If that is right, then it is plausible that there are at least some cases of rational and irrational unpleasantness. One might initially worry, however, that Clara’s unpleasantness is not really irrational because, after all, she has a putative reason to feel unpleasant: namely, that there are an uneven number of blue and white carnations in her vase. Addressing this worry requires a terminological detour. It is important for present purposes to keep clear what Clara has and what she does not have—however we label her holdings. One might say that Clara’s unpleasantness (and in turn, Clara) has a bad reason and lacks a good reason. Eschewing all talk of “bad reasons” and insisting that reasons are always good, one might instead claim that what Clara has is a merely apparent reason, but that (qua associated hedonics, at least) she lacks any genuine reason. Yet alternatively, one might prefer to mark this distinction by noting that what Clara has is (again merely, if one felt strongly) an explanatory reason, but what she lacks is a normative reason. Nothing below turns on which terminology one adopts, and I will sometimes use all three in my continued attempt to be ecumenical. This ecumenism is legitimate for present purposes, since which of these terminological sets one adopts is largely determined by which indicators of rationality one adopts, and my present goal is to make hedonic rationality legitimate whichever indicators one favors. That is, I aim to make plausible that however one slices the normative pie, negative hedonic tone looks to be sometimes rational if anything is: there are sometimes normative reasons, stroke genuine reasons, stroke good reasons for unpleasantness. This is, I stress, not to say that I don’t myself think there are better or worse ways of slicing. Only my views on that matter aren’t

Hedonic rationality


important for present purposes. Nor are yours. Important for present purposes is that the hedonics associated with Clara’s states, while admitting of a reason-based explanation, is irrational. Both the rational unpleasantness of Glenda and the irrational unpleasantness of Clara should be contrasted with a case that lacks any reason-based explanation. Case 3 Unintelligible unpleasantness: Bizarre Betty9 Betty likes kung fu. And comedy. Accordingly, once a month, Betty watches Jackie Chan’s The Legend of Drunken Master. She knows all the lines and all the moves. She laughs throughout the film and always walks away with uplifted spirits. Today, however, Betty’s Drunken Master viewing experience is not the joy that it normally is. She sits in the same position, eats the same snacks, and watches Jackie chug the same booze, but it’s not pleasant. Indeed, it’s all rather unpleasant. From frame to frame, her heart grows heavier. The movie ends, and Betty feels bad. For the next hour, many of Betty’s perceptions, perceptual beliefs, feelings, thoughts, and imaginings are unpleasant.

The case as described does not offer even an apparent reason for Betty’s unpleasantness; watching the film causes Betty’s felt unpleasantness, but for what reason? Going further, we might ask why this cause is even a cause. If the cause is as described, the associated hedonics are bizarre. As bolstered by some indicators of rationality from the literature (intended to be employed, again, just as they are applied elsewhere in discussion of the traditional domains): there do not seem to be any considerations that speak in favour of the unpleasantness of Betty’s subsequent episode or, indeed, even any that Betty (even mistakenly) takes to speak in favor of the unpleasantness, and the unpleasantness that she feels does not seem made at all intelligible by her other mental states, such that it is difficult to evaluate whether it coheres or clashes. Upon encountering Betty in real life, we would undoubtedly attempt to help Betty fill in the rest of the case (i.e., to help her figure out her associated hedonics). We would attempt to figure out why the cause is a reason. Failing to identify any such reason, we would try to identify some other, arational, cause. This search would be appropriate since watching kung fu is hard to make sense of as a cause of unpleasantness unless it is also a reason for unpleasantness—though, of course, if it is a reason for unpleasantness, it is quite easy to make sense of it as a cause. If we cannot make sense of her movie-watching as a reason for Betty’s unpleasantness, then it is hard to make sense of it as a cause either. Watching the movie appears to cause the negative hedonics associated with Betty’s states, but it is not clear how or why it might


Jennifer Corns

actually be the cause. And so, as described and unlike grieving Glenda’s and careful Clara’s, bizarre Betty’s negative hedonic tone remains unintelligible. The descriptions of these cases support the plausibility of the reasonresponsiveness of negative hedonic tone. Grieving Glenda’s unpleasantness is rational: she has good reasons for the negative hedonic tone associated with her states and indeed, all things considered, she should feel unpleasant.10 Careful Clara’s unpleasantness is irrational: the reasons for her negative hedonic tone are not good/normative/genuine and indeed, all things considered, she should not feel unpleasant. These two cases are to be contrasted with the case of bizarre Betty, whose unintelligible unpleasantness has a posited cause hard to make sense of as a cause because hard to make sense of as a reason. I find these descriptions of the cases the most natural. I think they show that it makes sense to talk about reasons for unpleasantness and that it is sometimes obvious that the hedonics associated with our inner episodes—and not only the episodes they are associated with—are rational. It makes sense to talk about reasons for a state to feel unpleasant or (depending on your preferred theory) to feel a distinctive unpleasant state. Furthermore, three indicators from the literature, applied just as they are applied in that literature, support these descriptions. There are reasons that speak in favor of, or that an agent takes to speak in favor of, the negative hedonic tone associated with their states. Hedonic tones make intelligible and are made intelligible by other mental phenonemena and, likewise, cohere with other mental phenomena. Should one think any of these indicators are sufficient for rationality, one should conclude in favor of hedonic rationality.

Objections In the previous section, I offered cases that I think are intelligibly and naturally described as rational, irrational, and rationally unintelligible negative hedonic tone: grieving Glenda, careful Clara, and bizarre Betty. I offered three indicators from the literature that support the appropriateness of those descriptions: there are considerations that sometimes speak in favor of unpleasantness, unpleasantness is sometimes made intelligible by one’s other mental states, and unpleasantness can sometimes clash, and sometimes cohere, with one’s other mental states. One might object, however, that it is easy to talk about mere causes as if they are reasons. Easy, that is, to describe things (in this case mental, hedonic things—whether theorized to be states or features of states) as if they are responding to reasons, when they are only the effects of causes. I have attempted to describe my controversial cases and bolster them with indicators in the same way that the non-controversial cases, e.g., beliefs and actions, have been described and bolstered. The objector who thinks these indicators merely seem to bolster the case for the rationality of hedonics think something is missing such that they do not in fact provide support. That is, the objector must think there is something more to rationality such that the

Hedonic rationality


considerations seeming to speak in favor of unpleasantness are not rational considerations, the clashes are not rational clashes, or the intelligibility is not the right kind of intelligibility. What more might be required? Not yet discussed are the other indicators of reason-responsiveness that might seem to indicate that hedonic tone is never rational or irrational. In this section, I consider four indicators of rationality as objections to hedonic rationality. Again: I do not intend to advocate any one of these indicators, but rather to be as ecumenical as possible. Indeed, I take these to be the strongest objections to hedonic rationality. Though space precludes full discussion, my goal in this section is to make plausible that these objections can ultimately be fully met—how they are met depending on one’s further theoretical commitments. As elaborated below, I contend that all four objections ultimately open up avenues for future research, rather than close down the present one. Objection 1 Voluntariness: Rational states are voluntary, and negative hedonic tone is not voluntary If something is reason-responsive, then it is appropriate to evaluate—and so sometimes praise or blame—an agent in virtue of that something. Submitting an agent to such rational assessment, however, may seem appropriate only if the something for which the agent is being criticized is voluntary. This objection, in short, is an ought-implies-can principle, applied to hedonics. Rationality is a normative notion. Among other things, that means if negative hedonic tone is ever rational or irrational, then it must make sense that one (at least sometimes) ought or ought not feel unpleasant. But, so the objection goes, associated hedonics are involuntary, so such ought-talk is senseless. The hedonic tone associated with one’s states is never voluntary, and so it is never rational or irrational. The ought-implies-can principle requires interpretation and is controversial. Applied to rationality, we may ask: which, if any, “cans” are appropriately required for the oughts of rationality? This also is controversial. It is not clear what, if any, sort of voluntariness should be required for rational assessment. Clearing up this controversy, however, is unnecessary for present purposes since beliefs are paradigmatically rational but involuntary. If beliefs are appropriately, and indeed paradigmatically, assessed for rationality despite being involuntary, then whether or not hedonic tone is voluntary seems irrelevant for hedonic rationality. In response, one might press the objection by claiming that beliefs are never rational or irrational: properly speaking, it is judgments—and not beliefs— that are reason-responsive. While it might be right to distinguish between judgments and beliefs at least partly on the basis of voluntariness, a position that thereby denies the rationality of the latter is extreme. We commonly


Jennifer Corns

speak, both in everyday and theoretical contexts, of it being rational or irrational to believe something, and of the formation of beliefs as a response to reasons. The contrary, at any rate, is fringe enough that I will not further pursue it here (cf. Nickel 2010).11 The most promising development of this indicator, it seems to me, is instead an identification of a distinctive sort of voluntariness relevant for rational assessment. Thus, Helm (2001) argues for a distinctive type of “rational control,” characterized as, for instance, “control by virtue of an appreciation of reasons.” Similarly, Scanlon (1998) claims that while attitudes like beliefs and intentions are not (voluntary) actions, we can nonetheless be responsible for them in the sense required for rationality. He writes: Because “being responsible” is mainly a matter of the appropriateness of demanding reasons, it is enough that the attitude in question be a judgmentsensitive one—that is, one that either directly reflects the agent’s judgment or is supposed to be governed by it. For this reason, one can be responsible not only for one’s actions but also for intentions, beliefs, and other attitudes. That is, one can properly be asked to defend these attitudes … (Scanlon 1998, p. 22) Scanlon thus makes room for the rational assessment of involuntary states like beliefs, for which the agent is—we might say—rationally responsible. Space precludes discussion of the adequacy of these accounts or further elaboration of their application to negative hedonic tone. Notice, however, that the use of either rational control or rational responsibility as an indicator for reasonresponsiveness looks to bolster the case for the rationality of hedonics. As described: negative hedonic tone is at least sometimes reflective of judgments and at least sometimes results from an appreciation of reasons. Glenda’s negative hedonic tones, for example, reflect her appreciation of her beloved husband having died, the flowers being from him, the trash being his duty, and so on.12 Most important for present purposes is that without the development of some distinctively rational sort of voluntariness—rational control, rational responsibility—the objection does not press. If we required voluntariness for normativity, beliefs would never be rational or irrational, but beliefs are a paradigmatically reason-responsive state-type. Rather than undermining the plausibility of the hedonic rationality, this objection merely highlights the importance of an independently-needed account of rational responsibility. Objection 2 Perception: Negative hedonic tone is a type of perceptual experience, and since perceptual experiences are never rational, neither is negative hedonic tone Just as many think it obvious that feelings are never rational, many think it obvious that perceptual experiences are not either. In denying the

Hedonic rationality


reason-responsiveness of perceptual experiences (in particular, as against the reason-responsiveness of perceptual beliefs), Richard Heck Jr., for example, writes: My perceptions, in particular, are something with which I just find myself “saddled.” They are merely caused: They are not something I can justify; but nor are they something that it makes sense to ask me to justify (or to expect me to revise). (Heck 2000, p. 522) Similarly, consider Michael Brady: We should not be tempted here to propose that we should understand the concept “red” in terms of the rational appropriateness of what people see, or in terms of responses to reasons to see, simply because we do not assess seeing red in terms of its rational appropriateness; to see something as red is not a response to reasons. (Brady 2013, p. 153) These are two clear statements of the nigh-universal view that perceptual experiences are never responses to reasons. Since felt unpleasantness is a perceptual experience, so this objection goes, it must never be reason-responsive either. In response, notice first that this objection only presses if negative hedonic tone is granted to be a type of perceptual experience. As introductorily noted, unpleasantness may be thought of as either an integrated feature of many experience types—perceptual experiences among others, such as beliefs—or as a distinct feeling state (e.g., a sensation or some other sui generis state type). The former model would still seem to leave at least some negatively hedonic toned experiences unscathed by the present objection. The latter model would rescue hedonic rationality from standing or falling with the rationality of perceptual experiences. But to be ecumenical we should consider the possibility that negative hedonic tone is a type of perceptual experience. In doing so, we should then ask for the reasons to hold that perceptual experiences are never responsive to reason. There seem to be two beyond the sociological. First, one might think that talking about reasons for a perceptual experience is flatly unintelligible. I admit that it may normally be odd to talk about one’s reasons to, e.g., hear a high-pitched sound—but perhaps it’s just normally odd to talk that way. One explanation of the oddity may be reliability. It may normally be odd to talk about any reliable causes as reasons without some good reason to do so—but that doesn’t mean that there aren’t any and that there aren’t sometimes good reasons to talk about them. Moreover, the cases in the previous section intelligibly describe reasons for negative hedonic tone. Insofar as the objection concerns


Jennifer Corns

what it makes sense to say, the previous section should establish that it makes sense to talk about reasons for negative hedonic tone. Finally and regardless, ordinary language, as always, needs regimenting, and is unlikely to settle the issue. Second and more theoretically significant, one might be tempted to think it is important that some mental phenomena stand in a causal and not rational relationship with the world. One might think that such states are needed to serve as a reason-providing foundation for the rest of one’s epistemic edifice. Even if one requires this, however, there seems no reason to think that all perceptual experiences need be foundational in this manner. As an example of this line of resistance, Jona Vance (2013), for instance, has argued that some perceptual experiences are cognitively penetrated such that the penetrated state is rationally assessable in virtue of its penetrator. Important for present purposes is that if Vance is right, then some perceptual experiences are reason-responsive. Vance rightly denies that what is true for some cognitively penetrated perceptual experiences must be true for all of them, much less all perceptual experiences. Thus, about visual experiences in particular he writes: We do not tend to talk of subjects as having reasons for seeing things as they do. In many—perhaps most—cases of visual experience, one does not see things as being a certain way for a subjective reason. … Arguably in most cases of vision, reasons do not figure in the process. However, my claim that visual states are often not grounded by reasons is compatible with the claim that they sometimes are. Sometimes a subject sees things as being a certain way for a (subjective) reason. (Vance 2013, p. 19) Notice that even if Vance’s cases of reason-responsive perceptual experiences are granted, it does not follow that all perceptual experiences are reasonresponsive. The foundationalist may require that some perceptual experiences are not reason-responsive without requiring this of them all. Perhaps some are and some are not. Cognitive penetration aside, a version of Vance’s move is appropriate in response to the present objection. If one holds with our objector that negative hedonic tone is a type of perceptual experience, the plausibility of hedonic rationality should prompt one to reconsider the epistemic unity of perceptual experiences. Theoretical simplicity does not trump epistemic reality. In sum: one may respond to objection 2 either by rejecting that negative hedonic tone is a type of perceptual experience, or by rejecting that perceptual experiences are never reason-responsive. Both disjuncts are avenues for further research.

Hedonic rationality


Objection 3 Intentionality: Rationality requires representational, or at least intentional, content and negative hedonic tone lacks such content It is widely accepted that only mental phenomena with intentional content is reason-responsive. Like voluntariness, intentional content is often thought to be not merely an indicator of reason-responsiveness, but a necessary condition. Let us grant for the moment, as needed for this objection, that negative hedonic tone does not have intentional content. As originally described, Glenda was rational and Clara was irrational in virtue of negative hedonic tone. If negative hedonic tone is granted to lack intentional content, then an objector pressing this third objection might attempt a re-description of the cases offered in the previous section. This re-description would need to identify some other mental phenomenon with intentional content; one might think there must be some such content, and so some other mental phenomenon, in virtue of which it even makes sense to make these ascriptions. Any appearance to the contrary is explicable by illicit trading between Glenda and Clara’s complex states and their association hedonics.13 One may, however, resist any attempted re-description of the cases by focusing on any one particular state which is hypothesized to have associated negative hedonic tone. Consider, for instance, Glenda’s and Clara’s respective perceptual beliefs about flowers. Clara believes that she sees seven blue and eight white carnations, and Glenda believes that she sees last Friday’s discarded flowers. What is irrational of Clara’s, but rational of Glenda’s, is the negative hedonic tone associated with these otherwise equally rational states: the difference in rationality is down solely to the associated negative hedonic tone. A more promising intentionalist re-description that one might offer would involve making explicit some presumed other of Clara’s or Glenda’s representational states. So, for instance, one might re-describe Clara’s case by explicitly adding a belief, e.g., the belief that there being eight flowers in the vase instead of seven is worth getting upset about. It is this evaluative belief, the objector might claim, that is the source of Clara’s irrationality—and this evaluative belief has intentional content. I agree that such a belief would, also, be irrational. But now imagine the case without any such presumed irrational beliefs—as originally intended. Surely this can be done. To return to the example: you ask Clara: “Is this worth getting upset about?”, and she sincerely responds that it is not. I see no independent reason no reason to think the irrationality must eventually bottom out in some originally-unmentioned intentional state. Our objector might keep offering further beliefs, and I will keep Clara rebuffing. Imagine an exasperated Clara: “I know I shouldn’t feel this way!” Perhaps more compellingly, consider that no intentional representational states will suffice for the objector’s needed re-description of


Jennifer Corns

Glenda’s rationality. Adding more states will not shift the locus of rationality: there is reason for the unpleasantness that Glenda feels, whatever the other states for which she also has reasons. Attempts to re-describe the cases such that the locus of rationality is shifted from the unpleasantness to a distinct representational state in the neighborhood thus fail. In Clara’s case, any attempted shift can be credibly resisted. In Glenda’s case, adding further states cannot yield a shift. In both cases, it is the negative hedonic tone itself that is rational or irrational. Notice now that this indicator not only fails to undermine hedonic rationality, but bolsters the case for hedonic rationality if negative hedonic tone is taken to have intentional content. According to the strong representationalist, for example, unpleasantness—like any other mental phenomenon that we ever experience—is reducible without remainder to the possession of representational content. If negative hedonic tone just is an evaluative belief, for instance, then unpleasantness is reducible to some representational content or other and this objection does not press. Moreover, one needn’t be a strong representationalist to deny that unpleasantness lacks representational resources. On my preferred view, for instance, negative hedonic tone is best characterized using a quality space theory, according to which it has representational qualitative character but not content. Is this representational enough to avoid the present worry? In continuing ecumenism, however, it is worth probing the claim that intentional content is necessary for reason-responsiveness. Why, that is, think that rationality requires intentional content? The answer, I think, is correctness.14 Rationality requires correctness conditions and representational states have correctness conditions; they are correct when they represent the world as it is and incorrect insofar as they do not. Thus Sabine Döring (2008), for instance, holds that rationality requires representational (intentional) content because rational assessability just is (a type of) assessability for correctness. She writes: Let us be clear that to be capable of playing a rationalizing role and of entering rational conflicts, both sense perceptions and emotions must have an intentional content of a certain kind. They must be about the world and represent it as being a certain way so as to be assessable for correctness, i.e., for adequacy to the actual state of the world that they purport to represent. (Döring 2008, p. 3, italics added) Here again: representational content is required for rationality, because correctness conditions are required for rational assessment—according to Döring, both reason-provision and the reason-responsiveness which is of present interest. The assumption seems to be that it is only representational content that can be so assessed.

Hedonic rationality


Similarly, according to Douglas Lavin (2004), correctness is key to understanding any normative activity whatsoever. Thus he writes, “On any account, reasoning is activity governed or guided by norms, rules, standards, or principles, and so the very idea of this activity must contain the distinction between correct and incorrect application of them” (ibid., p. 425). Here again, the idea is that rationality requires correctness conditions; to be rational a mental phenomenon must be assessable for its correct or incorrect adherence to the norms that govern it. I take the underlying argument for this indicator, then, to run roughly as follows: 1. 2. 3. 4. 5.

Rationality is normative. Normativity involves evaluations of correct and incorrect adherence to the norms in a domain. So, correctness conditions are necessary for rationality. States with representational contents are the only mental phenomenon that can be evaluated for correctness. So, representational content is necessary for rationality.

Notice that even if premises 1 and 2 are granted, premise 4 might be resisted. If there are mental phenomena lacking representational content that can be evaluated for correctness, then we need some other reason for the almost universally accepted conclusion. The upshot being that any legitimate demand for representational content is satisfied by correctness conditions. If one insists on premise 4, then this demand may be satisfied if negative hedonic tone can be reduced to something suitably representational, but it will also be satisfied if norms governing hedonic tone are otherwise secured. I am willing to grant that hedonic rationality requires evaluations of correct or incorrect adherence to norms in the relevant domain—and, indeed, I think the cases as described and discussed in the previous two sections support their existence. Glenda’s negative hedonic tone is correct. Clara’s negative hedonic tone is incorrect. Glenda should feel unpleasant. Clara should not. Because I think that plausible, I think it plausible that at least one of the following are true: (A) Negative hedonic tone is reducible to something representational such that its correctness conditions are given by the represented portion of the world. (B) There are some standards of correctness—some norms—governing at least some non-representational mental phenomena. In continuing ecumenism, I will not here argue for A or B. Both disjuncts are widely accepted and the current objection is no objection for anyone accepting either.


Jennifer Corns

One might, however, still worry. Thus far I have had nothing to say about what the hedonic norms might be. Space obviously precludes satisfying discussion of the nature of norms and their application to hedonics. Instead, I will consider only one further indicator stemming from the nature of normativity that might be thought a particular problem for hedonics. Objection 4 Aim: Rationality requires an aim, and negative hedonic tone has no aim Rationality’s normativity requires an end: an aim, function, purpose, or point (henceforth, just “aim”). Without an aim, the normativity of hedonics appears unintelligible. Without it, we seem left without any means of evaluation—none of the correctness conditions that we were granted are needed, no sensible evaluations, no normativity whatsoever. Aims are broadly employed as an indicator of rationality. Consider the two traditional domains. In the epistemic domain: beliefs, or judgments, aim at truth. In the practical domain: actions, or intentions to act, aim at their successful completion or fulfillment—though the details of this aim, and when it has been accomplished, are controversial. Beyond these two tradition domains, emotions perhaps aim at understanding (cf. Brady 2013). What is the analogue in the hedonic domain? If pleasantness or unpleasantness is ever rational or irrational, one might think, then it must have an aim. The need to identify the aim of negative hedonic tone, in particular, may be particularly pressing if one thinks it is always intrinsically bad. Negative hedonic tone might always provide reasons—namely, reasons for the agent to act in such a way that it, or its cause, ends. In order to be responsive to reasons, however, then there must at least sometimes be reasons for negative hedonic tone. There must sometimes be something good, purposeful, or correct about unpleasantness: some aim that associated negative hedonics sometimes achieves. Consideration of everyday cases supports the plausibility of a hedonic aim. The cases and indicators discussed in the previous section may themselves be taken as reasons for thinking that negative hedonic tone must have an aim, though we may not yet know it. More directly, our everyday practices evidence that we sometimes judge that one should feel unpleasant. We criticize each other for negative hedonic tone and for its lack. These practices indicate our tacit commitment to hedonics having an aim, our commitment to the idea that there are norms according to which it is sometimes correct and sometimes incorrect to feel unpleasant. Helm (2001) notes that though we give ourselves and each other a wide berth concerning our pleasant and unpleasant feelings, we do sometimes criticize. He gives two examples: when we tell children that something doesn’t really hurt and when we criticize others for some deviant sexual pleasures (ibid., p. 96). Setting the case of pleasure aside, I think Helm is right that children need to be taught—and we teach them—when they ought, and

Hedonic rationality


ought not, feel unpleasant. As Elgin (2007) points out, we have a robust practice of criticizing each other for affective overreacting and underreacting. This evaluative practice supports the idea that not only do we evaluate negative tone, we evaluate it for degree. Our evaluations of our children, ourselves, and each other thus reflect our commitment to the idea that hedonics has an aim. Even if we can’t yet clearly articulate what it is. These practices might be mistaken, but current ignorance is not a good enough reason to think so. Instead, I think we should take these practices as some evidence that there are hedonic norms and correctness conditions. Identifying a hedonic aim that explains our evaluative practices requires empirical inquiry into what negative hedonic tone, and hedonic tone more generally, does. The role that hedonics plays in, for instance, attention, learning, motivation, problem-solving, social stability, and moral development. Even if such empirical inquiry will not be sufficient to settle the normative question of the aim of hedonics, it should be agreed by all to be a crucial foundation. The necessary empirical inquiry into the nature and function of hedonics is currently underway. All four indicators which might have initially seemed to undermine hedonic rationality, instead highlight the need for ongoing research without undermining the plausibility of hedonic rationality. First: even if it is granted that negative hedonic tone is not voluntary, its rationality may nonetheless be vindicated by an independently-needed account of a distinctively rational control or responsibility. Second: negative hedonic tone may either not be granted to be a type of perceptual experience, or, if it is, some circumscribed set of perceptual experiences may nonetheless plausibly be deemed rational. Third: the requirement for representational content may be satisfied either by developing a suitably representational account of negative hedonic tone or by identifying correctness conditions for non-representational mental phenomena. Fourth: we should grant that hedonic rationality requires a hedonic aim, but our evaluative practices support the existence of such an aim and the research needed for its articulation is ongoing.

Conclusion I have argued for the plausibility of hedonic rationality. It is natural and intelligible to talk about reasons for negative hedonic tone, there are positive indicators of rationality taken from the literature that bolster cases so described, and seemingly negative indicators of rationality from the literature do not undermine them. I close by briefly noting some implications. While some implications for philosophical theorizing have already been noted above, hedonic rationality may also have practical implications. For instance, hedonics may be more (and more directly) susceptible to rational interventions than we thought. Unpleasantness may be yet more responsive to reason-giving then is allowed by current models that take hedonics as always


Jennifer Corns

a merely causal response to stimulation of various kinds. This is not to deny the importance of structural and chemical interventions for the treatment of some conditions that wreak havoc on associated hedonics—any more than it is to deny the importance of structural and chemical interventions for the treatment of some mental illnesses that wreak havoc on doxastic states. On the other hand, it may be inappropriate to intervene to eliminate negative hedonic tone in more instances than we thought. If negative hedonic tone is sometimes a rational occurrence that has an aim, extinguishing it may sometimes be inappropriate. If some negative hedonic tones are appropriate and useful for some purpose, then it will sometimes be inappropriate, and destructive of that purpose, to intervene. Consider most broadly and finally a general theoretical implication. One might think that accepting the reason-responsiveness of unpleasantness undermines the very distinction between rational and arational mental phenomena. One might have thought that if any mental phenomenon is merely caused and not responsive to reason, it is hedonics. Hedonics, that is, may have been our paradigm of an arational mental phenomenon. If even hedonics are rational, however, then the very distinction between rational and arational mental phenomena may be less interesting than we thought. If a very paradigm of arationality is accepted as sometimes rational, then the very distinction between rational and arational may matter little. This is not to suggest that it would follow that there is no distinction, but rather that the nature of the distinction is different—and perhaps different such that its explanatory purposes are more restricted—than many have thought. Ironically then, one of the reasons it might matter whether unpleasantness is sometimes rational and not merely caused is that it does not matter much whether anything is sometimes rational and not merely caused. More research exploring these and other implications is required. So too and as noted throughout, the further development of a theory of hedonic rationality will depend on one’s other theoretical commitments concerning the nature of hedonic tone, representation, mental ontology, reasons, and rationality. This ecumenism is a feature and not a bug of the present account as acceptance of the two traditional domains of rationality is likewise independent of these further commitments. I have argued that hedonic rationality is a distinctive domain of rationality that is evidenced in our everyday practices but as yet unacknowledged in the philosophical literature. Moreover, I have highlighted some practical and theoretical implications of countenancing this novel domain. In recent years, our attention has extended beyond the traditional domains to consideration of the rationality of emotions. It is time to consider hedonic rationality.

Notes 1 I favor a view of hedonic tone that characterizes pleasantness and unpleasantness on a qualitative continuum, such that what is true of the reason-responsiveness of

Hedonic rationality

2 3 4


6 7 8 9 10




negative hedonic tone will hold of positive hedonic tone, but I will not argue for this here or further discuss positive hedonic tone. The points in the text are independent of any (a)symmetries between positive and negative hedonic tone, though a full account of hedonic rationality would have to address these. See for example Bain (2013) and Cohen and Fulkerson (2014). Again and throughout: I intend ecumenism concerning the nature of this association. Bennett Helm (2001) may seem an exception. He develops a domain of rationality that he sometimes calls “affective rationality” and other times a “rationality of import.” The norms in this domain govern affective states—states that feel good or bad, pleasant or unpleasant. Nonetheless, on Helm’s view affective states are essentially both motivational states—and so subject to the norms of practical rationality—and doxastic states—and so subject to the norms of epistemic rationality. Affective rationality is then only a novel domain of rationality in virtue of being a hybrid between the traditional two: practical and epistemic rationality. These traditional domains, though both in play, are all that is in play. An agent is still rational or irrational in virtue of their actions, or intentions to act, and doxastic states. We still only have reasons for doxastic states and action (or, at least, intentions to act). In reading Helm we are challenged to combine these two domains—not to discover an independent third. The novel hedonic rationality here proposed is instead focused squarely on hedonics, i.e., the unpleasant or pleasant aspect itself and as irreducible to a motivational cum doxastic state. Though Scanlon does not offer all of them as such, I have culled nine characterizations offered throughout his discussion (Scanlon 1998). I do not discuss any of them as criterial, however, since none will adequately serve this purpose and, I take it, Scanlon has not so intended them. Notice again: it is not some particular emotion (i.e. grief) that is here in focus—though one may ask about the rationality of that, along with any of Glenda’s beliefs or actions (or intentions to act). If one is inclined to argue that the negative hedonics associated with Clara’s states are instead rational, then one is anyway granting that hedonics are rationally assessable and can come up with a distinct example. Again: whether we take the unpleasantness to be a distinctive episode or a feature of other episodes. I am thankful to Robert Cowan for suggesting this type of example. Note that the claim that Glenda should, all things considered, feel unpleasant, is intended to be compatible with the claim that feeling unpleasant is intrinsically bad or always bad for creatures like us. What I think the case of Glenda shows is that one should, all things considered, feel unpleasant despite the badness—even as the utilitarian, for instance, thinks that one should, all things considered, kill one to save five despite the badness involved in killing the one. If judgments are taken to be a type of action and a strong version of the voluntariness objection is taken on board, then the norms of epistemic rationality will presumably be a subset of those of practical rationality. This, I take it, is a neoKantian line according to which theoretical rationality falls squarely within the scope of practical rationality. On this approach, hedonic rationality would be a sub-type of practical rationality. I cannot pursue such a course here, but I do not think it incredible that if epistemic rationality is a subset of practical rationality, so too is hedonic rationality. Similar comments apply if one holds, conversely, that practical rationality is a subset of epistemic rationality. Notice that whether there are reasons for Glenda’s associated hedonics is a distinct question for whether there are reasons for Glenda’s grief. Even if we answer affirmatively to both of these questions, the relevant reasons offered may differ.


Jennifer Corns

13 A proponent of the rationality of the emotions who denies hedonic rationality seems likely to object in this way. I am grateful to Robert Cowan and Michael Brady for discussion on this and related points. 14 In the discussion that follows “correctness” may be changed to “accuracy” without loss or change to the argument.

References Bain, D. 2013. “What makes pains unpleasant?” Philosophical Studies, 166: 69–89. Brady, M. 2013. Emotional Insight. New York: Oxford University Press. Brady, M. 2018. “The rationality of emotional and physical suffering,” in D. Bain, M. Brady, and J. Corns (eds.), The Nature of Pain: Unpleasantness, Emotion, and Deviance, pp. 81–94. London: Routledge. Cohen, J. and Fulkerson, M. 2014. “Affect, rationalization, and motivation,” Review of Philosophy and Psychology, 5(1): 103–118. Davidson, D. 1963. “Actions, reasons, and causes,” The Journal of Philosophy, 60: 685–700. De Sousa, R. 1987. The Rationality of Emotion. Cambridge, MA: MIT Press. Döring, S. 2008. “Conflict without contradiction,” in G. Brun, U. Kuenzle, and D. Dogouglu, (eds.), Epistemology and Emotions, pp. 83–104. Farnham: Ashgate. Elgin, C. 2007. “Emotion and understanding,” in G. Brun, U. Kuenzle, and D. Dogouglu, (eds.), Epistemology and Emotions, pp. 33–50. Farnham: Ashgate. Fridja, N. H. 2007. The Laws of Emotion. Hillsdale, NJ: L. Erlbaum Associates. Goldie, P. 2012. The Mess Inside: Narrative, Emotion, and the Mind. New York: Oxford University Press. Greenspan, P. 1988. Emotions and Reasons: An Inquiry into Emotional Justification. New York: Routledge. Heck Jr., R. 2000. “Nonconceptual content and the ‘space of reasons’,” The Philosophical Review, 109: 483–523. Helm, B. 2001. Emotional Reason: Deliberation, Motivation, and the Nature of Value. New York: Cambridge University Press. Lavin, D. 2004. “Practical reason and the possibility of error,” Ethics, 114: 424–457. Nickel, P. J. 2010. “Voluntary belief on a reasonable basis,” Philosophy and Phenomenological Research, 81(2): 312–334. Parfit, D. 2001. “Rationality and reason,” in D. Egonsson, J. Josefsson, B. Petersson and T. Rønnow-Rasmussen (eds.), Exploring Practical Philosophy: Essays in Honour of Ingmar Persson, pp. 17–39. Farnham: Ashgate. Scanlon, T. 1998. What We Owe to Each Other. Cambridge, MA: Belknap Press of Harvard University Press. Smith, M. 1995. “Internal reasons,” Philosophy and Phenomenological Research, 55: 109–131. Solomon, R. C. 2006. True to Our Feelings: What Our Emotions Are Really Telling Us. New York: Oxford University Press. Vance, J. 2013. “Emotion and the new epistemic challenge from cognitive penetrability,” Philosophical Studies, August. doi:10.1007/s11098-11013-0181-z.

13 The agony of reason The unsteady bond between suffering and human rationality Matthew Fulkerson and Jonathan Cohen

Other creatures live in the world more or less as Nature presents it to them; and they react to it more or less directly, albeit sometimes with remarkable sophistication. In contrast, we human beings live to a significant degree in the worlds that our brains remake—though brute reality too often intrudes. (Tattersall 2012, xiv)

On the face of things, it would seem that suffering can have significant value for creatures like us. It is overwhelmingly plausible from the point of view of both naive reflection and broadly empirical investigation that sensory/ psychological suffering is a necessary condition on a certain kind of fitnesspreserving causal motivation, and that it assists and supports rational activity. On such a view, suffering is a primary source of “motivating reasons,” viz., practical reasons that guide a subject’s actions.1 Thus, when Lucy the distracted machinist strikes her thumb hard with a ball-peen hammer, or when Luke the graduate student is compelled to overcome his shyness to lead a seminar discussion, it is extremely natural to describe the situations by saying that their forms of suffering provide Lucy and Luke with motivating reasons to act in certain ways. (For Lucy perhaps pain gives a reason to tend to her injured thumb; for Luke perhaps the suffering caused by social embarrassment gives a reason to focus his attention on the intellectual material at hand.) These, like similarly ordinary descriptions of everyday cases, motivate treating the picture of suffering as reason-conferring as a default starting place. This picture gathers support from the observation that suffering and reasons hang together in surprisingly durable ways, even in cases where one might have expected the two to come apart. Thus, for example, it is interesting that those incapable of physical suffering (e.g., pain asymbolics and those with severe leprosy) tend to ignore proposed replacement harm signals (say, ringing bell sounds or flashing lights) unless the intensity of these signals is increased to a point that induces genuine suffering: it would appear that signals that fail to induce suffering just do not compel motivationally/rationally in the way that suffering does (Auvray et al. 2010; Brand and Yancey 1993).2 Or, again, it is interesting that canonical descriptions of learned helplessness


Matthew Fulkerson, Jonathan Cohen

(e.g., in clinical depression or as a result of uncontrolled stress) involve both a lack of affect and a lack of motivation (Abramson et al. 1978). In a similar vein, it is interesting that cornered prey animals both freeze (stop being compelled to act) and release pain-masking endogenous opioids (thereby presumably mitigating suffering) just at the time when suffering ceases to be adaptively or motivationally useful (Amit and Galina 1986).3 As we say, we are inclined to accept such pretheoretical and broadly empirical considerations at face value, and to accept the picture of suffering as reason-conferring that they appear to support.4 However, accepting this picture invites a concern that, to our knowledge, hasn’t been discussed elsewhere, over the nature of and limits to the interaction between suffering and rationality. This concern will be the focus of this chapter. One way to bring out the disconnect starts with the observation that states of suffering play a positive rationalizing role in our actions—they are reasonconferring—although they seem immune or resistant to rational considerations—they are not reason-responsive. Or at least, if they are reason-responsive, then they are reason-responsive to a more limited extent than other of our reason-giving states. We seem rationally motivated by our states of suffering, but our states of suffering do not seem sensitive to our broader rational circumstances in the same way and to the same extent. Now, suffering is plausibly not unique in this respect. Many states, events, facts, and states of affairs can plausibly serve as reasons for a subject without themselves being responsive to reason (again, depending somewhat on one’s account of reasons). For instance, if Theo is thirsty and heads to the fridge because he knows there is water there, there’s a good, if minimal, sense in which the water’s being in the fridge counts as a reason for his action; but of course, the water’s being in the fridge is completely immune to influence by Theo’s broader reasons and motivational profile. If pointing to this were all we had in mind, the observation would be of limited interest. But the case of suffering is plausibly different from and more interesting than the case of the water’s being in the fridge (etc.) in at least two respects. First, the rational role of suffering for subjects like Lucy and Luke is significantly more extensive than that of water’s being in the fridge for Theo. Second, there is a sense in which states of pain and suffering are both internal to and part of a subject’s own experience. In cases where Theo is unaware that there is water in the fridge, there’s a good sense in which that fact, though a reason for him, is no part of his psychological economy. In contrast, there is no question that felt pains and other states of suffering are psychologically available to the subjects in whom they occur. They are, like subjects’ beliefs and desires, a central part of their current psychological profiles. And yet, like things typically outside of our psychologies, states of suffering seem somehow resistant to our broader rational concerns. While we obviously don’t have complete control over all of our mental states and experiences, suffering is among the most salient and intense of those states, and plays a more significant motivating role than other typically uncontrolled mental states. It

The agony of reason


hardly needs to be said that states of suffering play a role far more central than other typical external entities (which, after all, on most views get their motivating force only after uptake into a subject’s psychological economy).5 The claim with which we began, that suffering confers (practical) reason, can be understood as the claim that suffering plays a central and often beneficial role in practical deliberation for the purposes of action (broadly construed). It is this very fact that makes the disconnect between suffering and reason so pressing. Roughly, the worry is that, if suffering does confer practical reason, and thereby comes to be integrated in critical ways with practical rationality in creatures like us, there seem nonetheless to be significant limits to that integration. In particular, when suffering becomes (practical) reason-conferring for a subject (when it enters into what Sellars 1956 famously calls the “space of reasons”), it nonetheless continues to behave in ways that distinguish it from other objects of rationality, and make it appear more like (yet nonetheless still distinct from) a kind of non-rational, outside influence. In this sense, suffering seems to play a dual (partly rational, partly not rational) role in the mental lives of creatures like us. Interestingly, and as we shall argue below, this duality seems peculiar to our sorts of mental lives: it likely does not occur in the mental lives of psychologically less or more rationally sophisticated creatures. Before we move on, it will be useful to limit the scope of discussion by mentioning (so as to set aside) a couple of aims that we do not have in the present chapter. First, while we take the disconnect between reason and suffering that we are highlighting to be interesting and worthy of discussion (partly because of the surprising predictions and explanations that, we argue below, it makes available), we do not offer it as a necessary or sufficient condition for the occurrence of suffering. Indeed, our position is that the disconnect is itself a further consequence—and one that can have serious negative effects on the creatures in whom it arises—of an antecedent state of primary suffering for which we have no analysis to offer in the present chapter. Second, nothing we say here is an attempt to explain the inherent badness of such suffering, nor to say anything directly about how states of suffering motivate our behavior. We set aside for now the issue of what makes suffering bad (but see Aydede and Fulkerson 2018), and assume that suffering has an antecedent negative valence. Our concern is to explore the nature and consequences of the disconnect that, we claim, arises for states that have this negative valence. Having so qualified our aims, here is our plan for what follows. We’ll begin our discussion by attempting to more precisely characterize the disconnect discussed above, and arguing that it has serious negative psychological consequences. Next we’ll ask how best to understand this disconnect. We’ll argue against several attempts to explain it away reductively in terms of more familiar rational psychological pathologies, and put forward an alternative descriptive conception that treats the disconnect as a reflection of our


Matthew Fulkerson, Jonathan Cohen

peculiar, partly but not wholly integrated, rational psychology. Finally, we’ll conclude.

A disconnect between suffering and action To bring out the salient aspects of the disconnect between suffering and our rational psychologies, it will be useful to consider the contrast between the two following cases of practical deliberation, one of which involves suffering, and the other of which does not. Beach You are deciding whether to undertake the action of going to the beach for the afternoon. You desire an afternoon of relaxation, and you believe that going to the beach will achieve that end. This belief–desire pair, which does not involve suffering, and which we will call R+, is a reason for you, and one that speaks in favor of the action of going to the beach. In addition, you have a looming deadline for submitting a paper, and you desire to fulfill this obligation. You believe that an afternoon spent working will achieve this end. This second belief–desire pair, which also does not involve suffering, and which we will call R−, is also a reason for you, and one that speaks against undertaking the action of going to the beach. In this situation you have two competing (non-suffering) reasons, R+ and R−, pointing in opposite directions with respect to the action; your decision how to act is an exercise of your rationality, pitting reason against reason. If one desire is stronger than the other, and no other considerations arise (it doesn’t start to rain, and you don’t get an extension on your submission), the rational thing for you to do would be to satisfy that stronger desire (again, on a rationalizing but not necessarily on a justifying account of action explanation). Assuming your desires are equally strong, and no other competing considerations arise, you might proceed by an explicit process of rational deliberation, comparing the relative utilities of the outcomes, looking for alternative ways of satisfying both desires (writing at the beach!), or reassessing your desires in light of the new evidence (it may rain, better to hold off on the beach day). At some point, this process of deliberation will end. In this case, you come to the considered view that the reason R+ outweighs R−, and so decide, decisively, by this exercise of rational deliberation, to go to the beach. At that point, after having decided in favor of undertaking the action, what can we say about your view toward the reason R− that spoke against that action before you had undertaken it? Of course, R− may continue to be a pro tanto reason against acting. After all, you still have a deadline coming up, and a desire to fulfill your obligation to submit your paper. You also still have the beliefs about (at least) one possible course of action that would fulfill that

The agony of reason

255 −

desire. Now, if your decision was made simply because your desire in R was not as strong as the one in R+, then you are likely happy with your choice and will proceed directly to the beach without looking back. If the decision was more difficult, you might still feel some pull toward the afternoon of work, even as you head off to the beach. That said, it is notable that the exercise of your rationality by which you weighed R− against other reasons had the net effect of making R− cease to be an all things considered guide to action. The pull that it exerts is sensitive to your other concerns and decisions, such that your decision to act on the basis of one consideration seemingly weakens the pull of your reasons. After all, if you are on your way to the beach, you’re not acting on R−. After deciding, you can consider and reflect consciously on R−, and even recognize clearly its rational and motivational force, but it is no longer playing any active role in the explanation of your present actions (or if it is, that activity is influenced by and sensitive to the reasons working in the opposite direction). You have, by a successful exercise of rationality, converted R− from a reason that is potentially action-guiding to a reason that is not action-guiding. Vaccination A new vaccine has just hit the market. It is very effective at preventing a range of disorders, and thereby positively increases your long-term health. It requires that the patient remain completely still for two minutes while the drug is administered via an intravenous needle. If the patient moves too much or the needle is removed early, the procedure must be repeated from the beginning. You are in the doctor’s office and, just before the shot is administered, are deciding whether to go through with the procedure. You have a longstanding desire to avoid serious illness and you believe (correctly) that getting the vaccination will help you achieve this end. You thus have one reason R*+ speaking in favor of the action of remaining still and going through with the procedure. On the other hand, getting a shot hurts, and you also have a longstanding desire to avoid pain whenever possible. You believe avoiding the shot will help you achieve this goal. You thus have a reason R*− speaking against undertaking the action. Now, since you are already in the doctor’s office, it is likely that your desire to have good health is stronger than your desire to avoid pain, and so the overall rational thing for you to do is to go through with the shot. But there is pain involved, so you’re having a much more difficult time sitting still while the nurse prepares the needle. As before, you embark on explicit deliberation, comparing likely outcomes, thinking of alternative considerations, or other ways to have it both ways. As before, you have two reasons, R*+ and R*−, pointing in opposite directions with respect to the action; your decision how to act is presumably an exercise of your rationality, pitting reason against reason. Suppose that you come to the considered view that, despite the pain,


Matthew Fulkerson, Jonathan Cohen

the reason R*+ outweighs R*−, and so decide, decisively, by this exercise of rationality, to remain still and go through with the shot. At this point, after having decided in favor of undertaking the action, what can we say about your view toward the reason R*− that spoke against that action before you had undertaken it? From the point of view of rationality, it seems that this case is just like Beach. As before, R*− plausibly continues to be a pro tanto reason against acting. As before, through the exercise of your rationality by which you weighed R*− against other reasons you can make R*− cease to be an all things considered guide to action. While you still have the desire to avoid pain, and you still believe moving away will help you achieve that goal, you have decided on a different course of action. As before, after having decided, you can consider and reflect consciously on R*−, and even recognize clearly its rational and motivational force, without being tempted to change your mind about the decision, and without taking it as a guide for your action. You have, by a successful exercise of rationality, converted R*− from a reason that is potentially action-guiding to a reason that is not action-guiding. Thus, it would seem that the case of Vaccination and the case of Beach are, from the point of view of rationality, structurally analogous. That analogy notwithstanding, it seems that there is an important difference between the cases. Namely, in the case of Vaccination, unlike the case of Beach, where the reasons did not involve suffering, there is a way in which the motivational force of R*− persists, or resists being overwhelmed by rational control. R*− continues to exert control over our actions even after rational deliberation has done its work.6 Even if we manage to stay still, doing so requires continual executive control. We have to, as it were, “fight through” the pain in order to execute the action we have settled on—even after having settled on it.7 It would appear that the action-guiding force of the suffering-involving reason R*− in Vaccination survives the exercise of rationality—and in this sense is interestingly disconnected from many of our broader reasons—in a way that the force of the non-suffering-involving reason R− in Beach does not. Nor is this an artifact of this particular case; we all have endured pains and suffering that continued to impact our actions despite strong rational considerations acting in the opposite direction. An ordinary example of this kind involves spicy food. Capsaicin works by binding at relatively low temperatures to TRPV1 receptors in the nose, mouth, and skin that are ordinarily activated only by high temperatures; it thereby tricks these receptors into producing signals of heat at temperatures much lower than ordinarily necessary for their activation. But learning that the TRPV1 signals present after eating modestly spicy food are false positives—that there is no harm in the offing—does nothing to lessen the associated suffering, or, more importantly, to reduce its motivational force.8 In what follows, we want to argue that this disconnect holds significant lessons about the nature and role of suffering in the mental lives of creatures like ourselves.9

The agony of reason


The agony of disconnect One of the reasons that the disconnect between rationality and suffering we’ve identified bears interest is that, we claim, it has the potential to initiate novel forms of sensory and psychological suffering, and thereby can significantly impede the overall well-being of individuals subject to it. It also seems to reveal something about the particular nature and strength of our rational lives in an especially illuminating way that promises interesting theoretical implications. In this section, we’ll focus on the first claim; we’ll return to the second below. Human suffering is often rooted in bodily reactions, but it is also often mediated by the unique rational perspective we can take with respect to that suffering. Thus, even a brief examination of the literature on pain and other forms of human suffering suggests a strong role for top-down cognitive effects on our pain experiences. For instance, anticipation effects play a role in mediating how much pain we feel; social context can influence pain judgments; and will power is inversely correlated with pain tolerance (Ossipov et al. 2010; Silvestrini and Rainville 2013). One area of investigation that most clearly reveals the potential for the kind of unique suffering we propose here are cases of chronic pain—in particular, chronic pain arising from central and peripheral neuropathy. Neuropathy can cause a wide range of painful sensations depending on the type of nerve affected and the nature of the damage. These can be intense and unpleasant tingles, shock sensations, burning, chills, and severe aches. The nerve damage can arise from many different causes, including diseases like diabetes, from direct trauma to the nerves, and can even be inherited. Neuropathy is remarkably and frustratingly difficult to treat effectively. One of the most insidious aspects of chronic pains arising from neuropathy is their pervasive psychological impact. In many cases, the nerve damage itself raises few additional health concerns.10 Subjects who suffer this sort of pain know that it is “just” pain, and that it does not pose (in itself) any additional health risks. Such pains do not serve any protective or informative end, and subjects can be aware of this very fact. And yet, the pains still play a powerful and relatively immutable role in motivating and guiding behavior.11 At once she feels the immense pain, while also knowing that it is not a meaningful signal, not serving any larger purpose beyond the pain itself. This disconnect, we believe, manifests itself in the torrent of harmful psychological and physiological reactions that are known to affect chronic pain sufferers, including depression, immune deficits, and other disorders. And this is to say that, at least for creatures in whom suffering gives rise to the disconnect we’ve identified, this primary suffering can produce yet further, secondary suffering. Moreover, and like the primary suffering, secondary suffering is also unlikely to be assuaged by more, or more conclusive, rational argument: it’s hard to see how supplementing rationality in these ways could dampen our reaction to the suffering, given that primary suffering is not under rational control.12 It


Matthew Fulkerson, Jonathan Cohen

is this set of well-known secondary reactions that constitute (one variant of) the novel form of suffering we believe arises from our peculiarly human perspective. One place where there is considerable evidence of just such secondary effects is in cases of chronic pain.13 Despite being often defined in temporal terms (e.g., as pains that persist for three or more months) what critically distinguishes chronic from other pain is not simply a matter of time elapsed, but the close association between chronic pain and other negative states like depression, learned helplessness, and immune system weakness. But researchers have found that chronic pain also has important links to particular types of rational appraisal. Thus, Casey et al. (2008) report that depressive symptoms and negative beliefs (primarily constancy and permanence beliefs: constant negative thoughts about the pain, combined with the belief that the pain would never go away) are a more significant factor in the development of chronic pain than were previous instances of trauma, duration of acute pain, intensity of suffering episodes, and baseline pain beliefs: Baseline depressive symptoms and pain permanence beliefs were the most powerful predictors of chronic disability, uniquely accounting for nearly half of the variance predicted by the full model. Depressive symptoms and uncontrollability beliefs may lead to passive coping and avoidance, thereby exacerbating disability. (Casey et al. 2008, 75) It would seem, then, that what best explains the difference between acute pains (which are often manageable) and chronic pains (which are often debilitating) are the negative beliefs a subject has about the pain. What we have said above about rational disconnect is useful in this context because it offers an account, not otherwise addressed in the scientific literature we are reviewing, of the source of such negative beliefs. Namely, if we are right that subjects undergoing such primary suffering experience a direct, ongoing, and uncontrollable conflict between their overall rational considerations and the practical influence of the intense signals of sensory pains, it is unsurprising that this should result in negative thoughts, especially of the constancy and permanence of the primary states of suffering. For those with neuropathy, the pain signals are constantly intruding: what to eat, what to wear, where to sit—every decision runs through their pain. This is because they have painful experiences, and the strong motivation to remove these experiences is part of what it is to have these experiences at all. No outside reason will mitigate this form of motivation unless it also removes or lessens the experiential qualities associated with the pains (which is why, for instance, being distracted or attending to other stimuli can indeed have a positive impact on suffering episodes). We can understand how this directly leads to negative thoughts

The agony of reason


about the nature and status of the pains. It is even worse in those cases where the pain is known to be giving a false signal, since it is not signaling any immediate danger. If there is no damage or potential damage being signaled, then there is nothing to heal or correct in order to make it go away (the fact that central neuralgias are also immune to most forms of pain treatment also cannot help). We are instead left only with the self-motivating experience. When suffering and our other reasons come into conflict, the states of suffering are sticky—recalcitrant to our broader reasons and concerns, constantly intruding on our thoughts, impeding our actions, and functioning as motivating reasons despite our awareness that they ought not. Sometimes (rarely) we are able to suppress the pains, and inhibit them for short durations, by convincing ourselves that the pains are actually good for us, or at least serve some positive purpose. This often weakens or pushes to the side the very experiential qualities motivating us in the first place (as above). But this requires extreme effort, and we are rarely up to the task, especially when the pains are severe or prolonged. This disconnect, we propose, leads to increased traumatic stress reactions, prevalence of negative beliefs, and feelings of helplessness and depression. We thus suggest that there is a compelling reason for caring deeply about the identified disconnect between sensory pains and our broader rational capacities: such a disconnect is likely a central contributing factor in novel and debilitating forms of suffering beyond those found in the initial pain state.

Alternative diagnoses So far we have argued that there is a peculiar disconnect between suffering and rationality in our psychologies and that this can have serious harmful consequences. Before we go on to offer our own characterization of the disconnect in terms of the way that suffering fails to fit comfortably into our broader rational capacities as other reasons do, we want to rule out several putative explanations for the disconnect—viz., explanations that attempt to reduce what is going on in cases like Vaccination to some more common, and well-investigated, phenomenon. As we see it, there are five principal explanatory alternatives that need to be set aside: explanations in terms of akrasia, non-intentional action, alief, further deliberation, and failures of deliberative decisiveness. Obviously this list is not exhaustive, but we think these are the most important alternatives. Akrasia The problem of akrasia has a long history in philosophy, and has been treated by most authors as a problem about the possibility of weakness of will.14 Here is a typical example.


Matthew Fulkerson, Jonathan Cohen Akrasia Suppose Corine judges that it would be best for her to eat a healthier diet, and she thereafter avoids excess sugars and fats. She does this because she has reasoned that this is, all things considered, what she has most reason to do. It’s not difficult to imagine that she firmly sets her will to the task, and throws out all of her sweets, orders healthy options when dining out, etc. It’s also easy to imagine that her resolve fails when faced with temptation. On an outing, she cannot resist the temptation to join when her friends all order chocolate sundaes. Even though she knows it’s not the right thing for her to do, and still, it seems, believes this, she breaks down and eats the sundae anyway.

These cases are common, and have raised philosophical worries since antiquity (hence their name). How is this case any different from the disconnect with rationality manifested in the suffering-involving case of Vaccination we considered earlier? Can we explain our inability to sit still when faced with actual (or impending) pain as nothing other than a certain kind of weakness of will?15 We cannot. There are several respects in which the disconnect present in Vaccination comes apart from that present in Akrasia, at least as the latter is usually understood. First, canonical descriptions of weakness of will involve a subject choosing freely to do B even though she judges A to be the better option. Thus, the disconnect between deliberation and action manifested in Akrasia is a singlestage problem resulting from a gap between the beliefs/judgments and the desires involved. In contrast, in Vaccination and similar suffering-involving cases, there is a further disconnect, over and above the disconnect between the first-order gap between deliberation and action. After all, when the needle hits the arm, it is not as though the subject is giving in to her other desire, and thereby acting against her better judgment. Indeed, typically she is actively fighting against the alternative movement, trying tenaciously to avoid flinching. The disconnect we’re interested in occurs even when the agent does not give in, and does not even have a desire to do so. A second difference between Vaccination and Akrasia is that the gap involved in the former but not the latter crucially implicates a reason that is not itself responsive to our broader reasons. In the cases of Beach and Akrasia, the alternative reasons between which we deliberate (to have a good time/ to get necessary work done; to eat healthily/to enjoy a chocolate sundae with friends) are reason-responsive, and so fit more easily into the subject’s overall psychology. What makes the suffering-involving disconnect in Vaccination so trenchant is just that at least one of our reasons—viz., that involving suffering—is to a significant extent not responsive to the rational influence of our other reasons. A third difference between Vaccination and Akrasia builds on the second. It is plausible that the subject in Vaccination who gives in to her suffering

The agony of reason


reason against what she takes to be her all things considered interest is less blameworthy than the subject in Akrasia who gives in to her non-suffering reason against what she takes to be her all things considered interest. We typically think of subjects as responsible or blameworthy for their weakness of will, and feel guilt or shame when we ourselves exhibit this phenomenon. But we have these reactions to a significantly lesser extent in cases involving suffering. When we give in to our suffering, it is generally thought that the suffering mitigates our responsibility. Hence, action is not as freely chosen in such cases—presumably just because, as noted, suffering reasons are not themselves reason-responsive, so significantly less under rational control.16 Beyond these differences between the sort of disconnect manifest in Vaccination and that manifest in Akrasia, there is a further reason for not wanting to explain our disconnect in akratic terms. Recall that we introduced our disconnect, above, by contrasting a case where it does arise (Vaccination) from a case where it does not (Beach). Given that there is this contrast, it is reasonable to demand of a proposed reductive explanation of our disconnect that it explain why the disconnect arises in the one sort of case but not the other. But on the (plausible) assumption that non-suffering reasons like those figuring in Beach are no less susceptible to akrasia than suffering reasons such as those figuring in Vaccination, it’s doubtful that an akratic explanation of the disconnect could meet this demand: on its face, it would seem that an akratic account of the disconnect would be no less applicable to cases involving non-suffering reasons (e.g., Beach) than to cases involving suffering reasons (e.g., Vaccination).17 We take these considerations to limit considerably the attractions of understanding our disconnect in terms of akrasia.18 Non-intentional action In arguing for a distinction between the suffering-involving disconnect in Vaccination from the disconnect in Akrasia, we contended that the former (unlike the latter) should not be seen as a rational interplay between mutually influencing reasons, in so far as suffering reasons are not typically responsive to our broader reasons. But this contention invites the thought that perhaps the suffering-guided actions under consideration in suffering cases are, because only to a limited extent subject to reason, not intentional at all. Thus, in the case of Vaccination, for example, perhaps the flinch or movement performed when the needle hits the arm is a kind of reflex that occurs without our sanction. It is extremely plausible that there are pain-induced reflexes that shouldn’t count as intentional actions. When we touch a hot pan, we might jerk our hand back automatically without (in some sense) deciding to perform that action. However, we don’t think the action in Vaccination and similar cases should be understood in this way. For the actions undertaken in our sorts of cases are flexible, actively performed actions that we recognize and treat as


Matthew Fulkerson, Jonathan Cohen

our own. They can be, and often are, controlled and managed, though sometimes imperfectly (cf. Schroeder 2005 on Tourette’s). In fairness, there is a sense in which suffering (e.g., pain) overwhelms us in these non-reflexive cases. But this point requires care. Suffering does not literally force us to move in particular ways in such cases. Rather, it overwhelms us in making us decide to act. What is forced isn’t the physical movement or action (as in a reflex), but our executive control concerning the decision to act. This is true even of slight pains, which, like more severe pains, do not force us to move our bodies, but are still a form of suffering and still have motivating force. They have this force, as we’ve said, because of the way they (and other pains) are presented to us. In severe cases, these experiential pains can become decisive. Though, in such cases, we cannot help deciding to act, the flexibility exhibited by the ensuing actions suggests that the latter are intentional in ways that reflexes are paradigmatically not.19 Another way to see this point is to note that, in many cases of suffering disconnect, the actions we undertake are highly complex and deliberate. For instance, one of the authors of this chapter sometimes suffers from extreme itchiness on the hand from an old chemical burn. Despite knowing that scratching and cold water in the long term make the itching worse (by irritating the skin and drying it out), this author sometimes finds the immediate relief provided by scratching and cold water irresistible. The scratching actions and the walk over to the faucet to turn on the water, etc., are extended in time, controlled, and seemingly voluntary, although the executive decision to act in these ways—though they are known to be all things considered suboptimal—is not.20 These sorts of actions involve a kind of disconnect, but not, we claim, as a result of the actions failing to be intentional. Something similar is going on in Vaccination: if the subject decides to move because of the pain, it need not be a reflexive jerk, but could easily be a complex action involving bodily movements and verbal requests. There’s a further reason for us not to accept the present suggestion that the suffering-guided actions under consideration in suffering cases are nonintentional. Treating the actions resulting from suffering as non-intentional because not subject to reason is at odds with the supposition that suffering is an element of the space of reasons—that it provides reasons for the actions in which it eventuates—with which we began. Giving up that supposition is not an alternative solution to our puzzle, but an abandonment of it, and would seem to fly in the face of the robust intuitive and theoretical support that, as we observed at the outset, the supposition appears to enjoy. We are, therefore, inclined to seek other avenues of response. Alief In a pair of influential papers, Gendler (2008a, 2008b) proposes the existence of a cognitive state of “alief.” Roughly, aliefs are tacit and automatic belieflike states that guide behavior despite being at odds with both rational norms

The agony of reason


and the explicit beliefs of the subject in whom they occur. Some simple cases from Gendler’s many putative examples of aliefs are: (i) the state of the man suspended safely in an iron cage above a cliff that causes him to tremble, (ii) the state of the person who has knowingly and deliberately set her watch five minutes fast that causes her to rush, even though she knows that her watch is inaccurate, and (iii) the state that causes a hungry person to shrink from eating fudge shaped like dog feces (Gendler 2008b). While there is much to say about both the proposed examples and the proposed category of alief, the notion is important in the present context because aliefs appear to share the crucial features of states of suffering we are pointing to: they are motivating and guide action, yet seem relatively immune to our broader rational concerns. Rational argument (whether supplied by the subject or another) is singularly ineffective against the trembling man’s trembling, the watch advancer’s rushing, and the fudge abstainer’s reluctance: the relevant aliefs persist despite countervailing rational considerations, and continue to guide action in directions subjects know goes against their best interests. In short, then, it appears that aliefs display something closely akin to the rational disconnect we have claimed attends states of suffering. It may be tempting, therefore, to suggest that our suffering disconnect can be reductively explained in terms of alief. Unfortunately, however, we don’t see how such a reductive explanation could go. For, on the one hand, states of suffering can’t be identical with aliefs: among other differences, states of suffering are (especially) consciously accessible and negatively valenced, while, ordinarily, aliefs are neither. And on the other hand, it won’t help to hold that states of suffering are merely associated with (/give rise to) numerically distinct aliefs, such that it is not the state of suffering, but the concomitant alief, that is the true locus of rational disconnect. For, even assuming there were such an association between suffering and alief, it remains true that states of suffering themselves (whether or not this is also true of any concomitant states) do elude rational control. As such, we cannot envisage any promising way of explaining away our disconnect in terms of alief. Spike in desire, new deliberation Perhaps what happens in the cases of apparent suffering disconnect is that there are really two separate steps of deliberation, and each is internally consistent. The idea is that we may have misjudged matters in our initial deliberation and come to a new and different practical conclusion that incorporates new information we hadn’t initially realized we needed. Thus, perhaps the best description of the case of Vaccination takes this form. Perhaps we decided at time t1 to stay still on the basis of the best available evidence (it was, at t1, what we had most reason to do). But by t2, perhaps the epistemic situation had changed—we had come to know just how painful that needle was really going to be (or how much it was going to bother us, etc.), and staying still was no longer what we had most reason to


Matthew Fulkerson, Jonathan Cohen

do. So we engaged in a second, distinct act of deliberation, and, based on the new evidence, simply changed our mind in a rational way. The idea, then, is that once we get the new information, once we are exposed to the feeling, we quickly (maybe very quickly) engage in a new round of practical deliberation and come to a new conclusion: we should move away from that needle, and fast (or, maybe we’re not that worried about our overall health after all, etc.). If so, then there isn’t anything like partial integration or a distinct kind of reason, and there is no real disconnect to be accounted for; instead there are just two distinct evidential circumstances and an ordinary, rational change in our decisions based on the new evidence.21 There are reasons for believing that the suffering disconnect we’ve pointed to is not best understood as involving two distinct acts of deliberation with different evidential circumstances. Crucially, if there really were new, separate deliberation, then ceteris paribus the subject should, after feeling the pain, stick with and accept the conclusion of that later round of deliberation— which means sticking with the rejection of the conclusion of the initial round of deliberation. But, though there are cases that take that form, that doesn’t seem to be the general description of our sorts of suffering cases. In many cases, including cases like Vaccination, the subject does not disavow the original practical conclusion. Rather, she persists in believing that the choice she had initially made remains all things considered best. She does not prefer the deviating action or change her mind or succeed in undermining her other rational concerns. Indeed, a common occurrence is for someone to flinch once or twice, but to keep trying again and again to fulfill the original settled decision. This “fighting through” is an important marker of the kind of disconnect we’ve been highlighting. None of these features are consistent with the pattern of reassessment and new deliberation suggested by this alternative. Failure of deliberative decisiveness A yet further possible redescription construes the alleged disconnect in suffering cases as a failure of deliberative decisiveness.22 Here the story would be that, contrary to our initial description of the cases of Beach and Vaccination as alike from the point of view of rationality, there is a characteristic failure of rational decisiveness in Vaccination and other cases involving suffering (and not elsewhere) which has the result that suffering – involving lesser reasons (such as R*− in Vaccination, but unlike R− in Beach) don’t cease to be actionguiding. This failure comes about because practical deliberation becomes Hamlet-like, unable to come to the point of a decisive verdict about how to act, and instead continues to oscillate between reasons when it is applied to suffering-involving reasons.23 Unfortunately, there are two serious defects with the Hamlet proposal, when construed as a diagnosis of the asymmetry. The first is that it only pushes back, but does not resolve, the explanatory question posed by the disconnect we have observed in suffering-involving reasons. Whereas we

The agony of reason


began by asking what it is about the suffering-involving reason R* that (unlike the non-suffering-involving reason R*) makes it retain its action-guiding force despite the exercise of rationality, adopting the Hamlet diagnosis trades that initial question for the new question of what it is about the sufferinginvolving reason R*− that (unlike the non-suffering-involving reason R*) permits practical rationality to reach a decisive terminus when applied to it. Since we don’t see the second question as any easier than the first, we don’t see that much has been gained in passing from the one to the other. The second defect of the Hamlet proposal is that it appears to have dubious entailments. Specifically, because it locates the asymmetry in a difference of the efficacy of practical reason, the Hamlet proposal would seem to predict that the asymmetry would be lessened or eliminated in subjects suffering from deficits of practical reasoning (e.g., juveniles, psychopaths) who (because of their deficits) are less successful at engaging their practical reason even in cases where their reasons are not suffering-involving. Because such subjects are less likely to engage the processes that, on the proposed diagnosis, risk turning asymmetrically Hamletian just when suffering is involved, we should expect not to see an asymmetry between their responses to suffering-involving and non-suffering-involving reasons. Alas, there is no reason to believe that this prediction is correct.24

Ambivalence though partial integration The shortcomings of the options considered in the preceding sections suggest strongly that the disconnect we’ve highlighted between states of suffering and rationality cannot be adequately accounted for by reducing it to other, better known, forms of rational disconnect. If so, the disconnect we have identified is a distinct syndrome with its own psychological profile. In this section we attempt to characterize that profile, argue that it reveals something peculiar about our kinds of minds, and address remaining taxonomic worries. Partial integration and experiential motivation We have claimed that states of suffering are, while sources of rational motivation, only partially connected with our rationality. But we now want to refine this claim. As we see it, states of suffering are rationally motivating in two distinct ways—and only one of these forms of rational motivation gives rise to our disconnect. What are these two forms of motivation that attach to states of suffering? First, when we undergo pains or other states of suffering, we are rationally motivated by these states to reduce, overcome, or eliminate the underlying distal cause of our suffering. Thus, for example, when you undergo a pain caused by touching something hot, you are rationally motivated to stop touching whatever it was that caused the pain. But, additionally, when we undergo pains or other states of suffering, we are rationally motivated by


Matthew Fulkerson, Jonathan Cohen

these states to reduce or eliminate the suffering experience itself—even if we know that doing so does nothing at all to correct or remove the underlying causes of the pain or other suffering. This is why, for example, we often select, use, and regard as effective medical treatments for pain that reduce our experiential suffering without treating or doing anything to address the underlying cause of the suffering.25 Thus, states of suffering, in a way that distinguishes them from other reason-providing states, partly motivate us to act in a way directed on our very experiential states themselves. They signal reasons for us to act that are partially generated by, and directed on, their own experiential qualities.26 In so doing, such experiential states bring along their own reasons, simply by their very presence. They are, as we might say, experience-directed motivators. 27 Noting that states of suffering are experience-directed motivators in this sense allows us to locate more precisely the source of the rational disconnect that arises for these states. Namely, we want to suggest that the disconnect is a feature of the experience-directed motivation, rather than the non-experience-directed motivation, provided by states of suffering. To see this, consider poor Lucy once again, who now feels a sharp, consistent pain in her leg. From one perspective, the non-experience-directed motivating reason the pain provides for Lucy—say, a reason to visit the doctor and undergo treatment to address the underlying cause of the pain— seems reasonably well-integrated with (i.e., not disconnected from) her rationality. This non-experience-directed motivating reason arises because Lucy’s state of suffering, like any state that plays a role in rational practical reflection, provides her with actionable information about the distal world. Lucy’s pain enables her to note that there is something amiss, and to undertake rationally well-advised, practical steps to address it. Significantly, and as in other cases of non-suffering practical deliberation like Beach, her state of suffering provides reasons and motivates in ways that can easily influence and be influenced by her broader, objective reasons. New information or evidence, or competing objective reasons, can influence, boost, or suppress her sufferingprovided non-experience-directed motivation, just as they do in non-suffering cases like Beach. In short, the non-experience-directed motivation provided by Lucy’s suffering in this case seems no less fully integrated with her broader rational life than are motivations in typical, non-suffering, cases of practical rational reflection. From another perspective, however, there is an experience-directed motivating part of the suffering that does not integrate in this way with Lucy’s broader reasons. It doesn’t matter what Lucy’s all-things-considered reasons might be, what her plans for the day might be, or what she would rather be doing. When she feels the sharp, intense pain in her leg, she is immediately motivated to alleviate it (i.e., the experience) in some way. This motivating reason, directed at the suffering itself, is not as easily and fully sensitive to Lucy’s other reasons. Suppose Lucy is informed by her (trustworthy, knowledgeable, and reliable) doctor that there is nothing physically wrong with her

The agony of reason


leg, nothing amiss that needs to be addressed or treated. Even so, she still has reason to act, she is still motivated to alleviate the painful experience. So long as she undergoes it, her painful experience is likely to retain its status as something that provides her with this second type of experience-directed rational motivation.28 The observations that states of suffering are motivating in both experiencedirected and non-experience-directed ways, and that the experience-directed motivations they provide (as opposed to the non-experience-directed motivations they provide) are often imperfectly integrated with broader rationality do not, by themselves, go terribly far in explaining the original disconnect between suffering and rationality. For they leave open both why states of suffering have this dual structure and why experience-directed motivators fail to integrate smoothly with our broader rational concerns.29 However, our observations do allow a more specific description of the initial disconnect. Namely, they allow us to say that the rational disconnect exhibited by states of suffering traces to their associated experience-directed motivations (a kind of motivation that is connected with their felt nature), as opposed to their associated non-experience-directed motivations (a kind of motivation that is connected with their informativeness about their distal causes). Suffering without a felt component would not, it seems, be rationally disconnected in the way that it seems to be. Suffering in minds like ours If we are right that the rational disconnect observed in states of suffering is a result of the dual, partly rationally integrated and partly rationally non-integrated character of the motivations they provide, then this suggests that the disconnect should only arise in creatures whose psychologies exert both rationally integrated and rationally non-integrated pulls on states of suffering. Of course, suffering plays a powerful psychological (and, presumably, adaptively advantageous) role in simple creatures just as it can for us. But the present thought is that the powerful psychological role suffering plays in simple creatures need not run through rationality. When a simpler creature feels pain it is only motivated by the experience-directed motivating aspects of the experience. As far as we know, rabbits and birds and fish, for instance, don’t endure pain and suffering on the basis of broader rational reflection. Indeed, part of what we take their relative simplicity to consist in is that such creatures don’t seem to weigh reasons in this way at all. Thus, though it is plausible that simpler creatures avoid sensory pains, it is unclear that they would endure sensory pains that they recognized as in the service of some overall good reason, or that they would be more upset about pains that they deemed not in the service of some overall good reason.30 If this is right, it is tempting to think of the psychological role of suffering in such creatures as running through only one of the two distinct pathways that are operative in us: they only have experience-directed reasons. But if there is only this single


Matthew Fulkerson, Jonathan Cohen

pathway, there is no possibility of conflict, hence no possibility of the kind of ambivalence to which we find ourselves vulnerable. Suffering will be simply, straightforwardly, and unambivalently motivational for such creatures, as it is not for us.31 On the other hand, it is also possible to imagine a kind of hyper-rational creature in whom suffering-involving reasons are treated exactly like nonsuffering reasons in being completely under fully integrated rational control. Such creatures would not need experientce-directed motivating states in order to act in appropriate ways: they would, unlike us, be strongly motivated solely in virtue of the broader practical reasons. In addition, we can imagine that such creatures would cease to have a negative reaction to suffering-involving reasons like that in Vaccination once the latter had been mitigated by the successful exercise of rationality, since (by hypothesis) there would be no motivational remainder. There would just be the informational signal that something bad was going on (and this signal would be appropriately discounted when that information was deemed incorrect). Plausibly, suffering would be motivational in these hyper-rational creatures in just the way (whatever that is) that non-suffering reasons can be motivational to us in typical cases. But, again, this relegates the motivational force of sufferinginvolving reasons in such creatures to just one of the dual pathways available in our own psychologies, and so removes the possibility of the kind of clash to which we are susceptible. Again, suffering will be simply, straightforwardly, and unambivalently motivational for such creatures, as it is not for us. The peculiar ambivalence toward suffering that we experience can only arise because our psychologies combine both the systems for responding to suffering present in simpler creatures and the flexible, rational system for responding to reasons present in the imagined hyper-rational creatures, and because these distinct systems can come into conflict. Moreover, and significantly, when they arise, these conflicts cannot be assuaged by rational argument. You may have good reasons for believing that an episode of suffering is all things considered worth undergoing for some greater good, and even that it is not associated with any significant harm. But none of this makes your suffering lose its motivational force. That such conflicts can arise and persist in creatures like us, with highly flexible cognitive systems with capacities for introspection and self-reflection, means that the automatic functioning of our (generally low-level) suffering system is for us not just an adaptive key to survival (surely it is, as in the rest of the animal kingdom) but also, and additionally, a point of worry and social anxiety caused by its unsteady connection to reason. In this sense, the role of sensory and social suffering for us, while adaptively connected with motivation and reason, is different from—and, as we have argued above, carries significant further negative consequences for well-being relative to—the role it has in either simpler creatures who only respond to occurrent negative stimuli or in imagined hyper-rational creatures who can reason away the motivational force of their suffering.

The agony of reason


Taxonomy In the absence of an uncontroversial account of what suffering amounts to, we have had no choice in the foregoing but to constrain our story about the disconnect between suffering and rationality by appeal to paradigm cases. That is, we have aimed for an account that predicts a disconnect in what we take to be paradigm instances of suffering (e.g., Vaccination) and that predicts no disconnect in what we take to be paradigm non-instances of suffering (e.g., Beach). However, as Rawls (1971, I.4, I.9) famously observed, theoretical equilibrium works in the other direction as well: general principles designed to accommodate cases can themselves suggest revisions to the intuitive taxonomy of cases, particularly once we move away from the most central paradigms. We believe that what we have said about the disconnect between suffering and rationality has a number of taxonomic consequences that fall under this general umbrella. First, and as we’ve emphasized throughout, the story we’re telling is designed to extend uniformly to the many disparate forms of sensory and non-sensory suffering—including (for example) pain, shyness, hunger, thirst, and certain pathologies of deliberation. The commonality potentially revealed by so extending our story suggests that there may be principled theoretical grounds for counting all of these as instances of suffering, and indeed that there is an interesting theoretical continuity behind what might otherwise seem a dauntingly heterogeneous category (cf. the unity worry mentioned in note. 4). Second, as we argued earlier, the very same story extends to chronic pains and psychological compulsions (and underwrites needed explanations of some of the most significant negative consequences of these disorders). And this suggests that there is something explanatorily useful (besides mere continuity) to be gained from construing the latter as instances of suffering. A further consequence of what we have said that may initially seem less appealing is that it threatens to extend the category of suffering to even very minor annoyances.32 For example, it seems that our disconnect with rationality could arise even for my mild state of annoyance on hearing elevator music—that state might give me a reason to direct my attention to something (anything) other than the music, but does not diminish in the face of whatever reasons I bring to bear (telling myself that the annoyance is very slight, that it will soon pass, etc.). Still, despite its patterning with our other cases, it might seem overblown to describe this minor annoyance as an instance of suffering. Our response to this consequence is to accept it. In our view, the cost of revising our intuitive taxonomy of suffering to include minor annoyances is outweighed by the view’s explanatory benefits. And in any case, nothing we have said precludes acknowledging differences in intensity, duration, import (etc.) among instances of suffering, or, therefore, from distinguishing minor from major instances of the category.


Matthew Fulkerson, Jonathan Cohen

Conclusion That our reactions to suffering embody conflict, and hence ambivalence, in the way we have been trying to bring out, is, it seems to us, an interesting feature of our mental lives. Of course, the idea that there is a conflict between rational and arational elements of our psychologies is a standard trope of modernity. For this reason alone, it would not be surprising to find some sort of conflict between our broader rationality and suffering falling out of views on which the latter fails to provide reasons. What we have been suggesting, however, is that psychological conflict arises as well on the opposite supposition that suffering does provide reasons, and in multiple ways. Indeed, the conflict is in a way more pressing on the latter view. For a conflict between reasons and non-reasons might seem to be a clash between elements too disparate in kind to be mutually constraining—perhaps in that case the best response would be to resort to a coin-flip. In contrast, if, as we are suggesting, the conflict arises within the space of reasons, and between two different kinds of reason, it is to that extent easier to see the elements as commensurable and mutually constraining. Perhaps this explains why the conflict we experience in such cases is such a powerful source of anxiety.

Notes 1 Practical reasons are often contrasted with “justifying reasons,” those considerations that justify or give good reason for an action, whether the agent acts on those considerations or not. While we believe that suffering provides both motivating and justifying reasons for action, in what follows we will focus only on the motivating reasons provided by suffering. That said, we do not wish to commit to any particular account of action explanation or motivating reason in what follows, since we believe the interest and importance of the disconnect we highlight below is independent of such controversies. Thus, we will sometimes characterize motivating reasons in terms of instrumental belief–desire pairs, and other times we’ll talk about states of suffering themselves as motivating reasons, as best smooths overall exposition. Readers with more committed views are invited to substitute their preferred ways of describing reasons as appropriate. If, for example, one prefers not to talk about a state of suffering itself as a motivating reason, one may take our use of such states/facts as a shorthand to pick out the genuine motivating reason—e.g., to pick out the set of motivating attitudes typically caused by that state (cf. Smith 1994). In the case of pains, these would be those beliefs and desires intrinsic to typical painful experiences. 2 Interestingly, the suffering caused by the alarms does not motivate subjects to avoid the genuine harms indicated by the alarms; instead, they strongly motivate behavior to shut off or remove the alarm signal. This reflects a crucial ambiguity in the motivating role of suffering itself to which we will return below. 3 Similar effects have been found in humans; see e.g., the discussion of the human pain inhibitory system in Hardcastle (1999, 132–134). 4 In saying that the picture of suffering as reason-conferring enjoys support both from our pretheoretical picture of the world and broadly empirical investigation, we do not mean to say that there are no puzzles or worries that can be raised against it. For example, if you are inclined to think about pain and/or certain forms of suffering in causal terms, sympathy to certain views about the relation

The agony of reason






between causes and reasons might make you doubt that pain or suffering could be a bona fide reason. (We address this latter worry in Cohen and Fulkerson 2013.) One might also worry about whether the category of suffering is sufficiently unified to support any interesting and generalizations at all. Indeed, some have argued that the heterogeneity of sensory pain threatens the prospects for systematic theorizing about pain (Heathwood 2006; Smuts 2010); and those worries are presumably more pressing as applied to the yet (far) more inclusive category of suffering. (We return to this worry below.) In fact, states of suffering may not be unique in this respect even among states that are internal to subjects’ psychologies: arguably some perceptual states and noninstrumental desires are reason-conferring but not reason-responsive. (Thanks to David Bain for this observation.) Michael Brady (personal communication) suggests that this characterization might not be true of at least some cases of emotional suffering, like suffering from guilt, shame, or regret, which he thinks depend on a cognitive evaluation and therefore can be subjected to rational control—e.g., when I am rationally convinced to give up the belief that I am responsible for the situation about which I feel guilt/shame/ regret. We have three points to make. First, if Brady’s cases of suffering are indeed subject to rational control, they are plausibly special—viz., unlike other instances of suffering—in just this way, which would mean that our story remains intact as a characterization of the general case. Second, Brady’s claim that states of emotional suffering are susceptible of rational control is controversial; in this connection, it is worth noting that rational cognitive behavioral therapy and related interventions that bring rational considerations to bear appear to be effective with respect to a wide range of problems (especially when combined with medical and other forms of treatment), but have been found to produce more limited therapeutic effects on states of emotional suffering in the general population (cf. Hofmann et al. 2012, 436; Davidson et al. 2006, 7). Third, even if rational considerations can remove states of emotional suffering that depend on cognitive appraisal, it’s certainly possible to deny that, in such cases, the states of suffering themselves (as opposed to the cognitive states on which they depend) are under the influence of rationality. At a minimum, then, we deny that Brady’s cases impugn the story we are telling. A similar finding shows up in neuroscientific models of conflicts in decisionmaking between abstract long-term goals (say, a desire for health) and immediate short-term affective response (say, the pleasure of eating ice cream) (Teuscher and Mitchell 2011; Bechara et al. 1994, 1999; McClure et al. 2004; Ballard and Knutson 2009). As many of us know all too well, the immediate pleasure of ice cream ordinarily exerts a much stronger pull on action than the abstract long-term goal of health (MacKillop et al. 2011). To be sure, we sometimes succeed in acting in accord with our abstract long-term goals. But when we do, this is not because our top-down long-term reasons gain in strength; rather, they appear to triumph by suppressing their short-term affective competitors (Benedetti et al. 2006, 2010). This is consonant with our general picture in suggesting that top-down reason has a limited effect on suffering, and only by removing it. (To be fair, the same idea also cuts against our picture somewhat by suggesting a mechanism by which our disconnect can be resolved; but because these cases of successful top-down mediation appear to be limited, effortful, and temporary, we are inclined to treat them as exceptions that prove our more general rule.) A slightly more obscure example of this sort involves the “thermal grill” demonstrated by Thunberg (1896)—an alternating pattern of warm (40°C) and cool (20°C) bars that produces an experience of intense burning to the touch. Subjects know that the temperature range on the bars of the grill is well within the limits of safety, and can verify this by touching the warm and cool bars separately without undergoing any suffering. Yet the burning sensation produced by touching a



10 11




Matthew Fulkerson, Jonathan Cohen

larger region of the grill is so intense that subjects are unable to keep their hands on it. (For a simple model of the mechanism underpinning our response to the thermal grill, see Craig and Bushnell 1994.) Objection (pressed on us by Dana Nelkin): Our presentation of the disconnect, which rests only on an intuitive understanding of pain and a few examples (as opposed to any systematic account of pain) may crucially conflate features of pain with those of numerically distinct states or processes. If, for example, it is not pain but some concomitant state that is reason-conferring, then there is no puzzling disconnect, for there is no one entity that is both reason-conferring and reasonresistant. Response: First, just to be clear, the puzzle arises for states of suffering generally, rather than exclusively for pains, and this point already makes it somewhat unlikely that the puzzle depends essentially on differences in the theoretical understanding of the subclass of pains. Moreover, as we have argued elsewhere (Cohen and Fulkerson 2013), proponents of different accounts of pain—say, attitudinal theories (Feldman 2002; Heathwood 2006), evaluationist theories (Helm 2002; O’Sullivan and Schroer 2012; Bain 2013), functionalist theories (Aydede 2014; Clark 2005; Fulkerson 2014), and imperativist theories (Klein 2007; Hall 2008; Martínez 2011)—can and should accept the claim that pains confer reasons as a prima facie datum which it is incumbent on their theories to explain. If so, any adequate account will have to cite some theoretical element (whether numerically identical to the pains themselves or not) that makes pains reason-conferring, and with respect to which a version of our puzzling disconnect arises. We can ask: why is it that that element (whatever it is) makes the pains with which it is connected reason-conferring and extensively integrated with rationality, but does not also make those pains susceptible to rational mitigation in the way that non-suffering reasons seem to be? Neuropathic pain is by no means unique in this respect; many pains are caused by conditions that have little immediate health impact other than the pain. Note that this has a structure similar to that of compulsions. A subject suffering from severe compulsion might recognize the meaninglessness of the compulsive act, and its all-things-considered emptiness as a reason, and nonetheless feel compelled to carry out the act. That compulsion exhibits this structure is, we suggest, a reason for thinking of compulsion, despite the typical absence of concomitant pain, as a species of human suffering (cf. McKay and Dennett 2009, 499–500). This consideration invites worry about a looming regress: if reflection on the resistance of n-ary suffering to rationality causes n + 1-ary suffering, then there will be a regress if reflection reveals the latter to be itself resistant to rationality. Luckily (from the point of view of avoiding regress), there are limits to ordinary reflection: it would seem (at best) unusual to reflect on the relations between rationality and n-ary suffering for very many values of n. Wherever reflection ends, the present threat of regress ends with it. A good (if now dated) discussion of these cases can be found in Hardcastle (1999). In addition, the more recent biopsychosocial conception of pain takes seriously the pervasive influence of social and psychological factors on pain experience and management (for a good introduction to this perspective, see the essays in Hadjistavropoulos and Craig 2004). An exception to this tradition is Holton (1999), who argues that akrasia is not a weakness of will, but an over-readiness to reconsider one’s intentions. On Holton’s view, cases of akrasia are in fact instances of the failure of deliberative decisiveness we consider (and which, we argue, are also distinct from our cases of disconnect) below. Since we’ll be considering that further alternative below, and so will eventually provide reasons against both as reductive explanations of our disconnect, our treatment of akrasia as a weakness of will can be treated as merely an expository convenience, rather than as our taking any official stand between Holton’s

The agony of reason

15 16




20 21

22 23


and the majoritarian story qua conception of akrasia (or, for that matter, as committing to any particular account of akrasia). As one might expect, Beach can easily be recast as such an example. Having distinguished in these ways between the kind of disconnect present in Vaccination and the kind present in Akrasia, we note that it may, nonetheless, be possible to describe particular cases—perhaps even the sundae-eating case we labeled “Akrasia”—on the latter model. On this description, overeating is more like acting under compulsion than a weak-willed choice, and patterns with Vaccination rather than Akrasia on the dimensions just considered: it involves fighting against rather than giving in to an opposed reason; the triumphant reason is not reason-responsive; and because it is not fully under rational control, our succumbing to it seems less apt for blame and responsibility. But the availability of a non-akratic/compulsion understanding of this and other cases traditionally labeled as akratic does not undermine our distinction between the disconnect present in Vaccination and the disconnect present in Akrasia. Rather, it reinforces our point that there are two quite different models here—and reveals that there is controversy about which of them is the best description of individual cases. Might the proponent of the akratic explanation answer our demand by supplementing her account with some more specific further reason why suffering reasons are especially apt targets for akrasia? Perhaps she could; but if so, then we suggest that it is that specific further reason, whatever it is, and not akrasia itself, that explains the disconnect. To be clear, our claim is certainly not that the suffering-involving disconnect we’ve highlighted can’t be thought of as a form of akrasia in any sense at all. But we hope it is clear that even if suffering does amount to a kind akrasia, it is a special form of the phenomena, with a range of distinctive features, that therefore deserves its own explanation. We take no official stand here on the complicated and controversial matter of how to distinguish intentional from unintentional action. For some influential views on this matter, see Anscombe (1957); Davidson (1963); Goldman (1970); Searle (1983); Dretske (1988); Velleman (1992); Setiya (2007). The case is, also, importantly different from cases of OCD or addiction, though in these cases, too, we might also find the very form of disconnect we’re interested in; see note 11. In conversation, Rashida Ahmad has suggested that perhaps there is only a difference here between occurrent versus predictive acts of deliberation. This is partially right, in the pain case, as we’ve said: perhaps part of the motivating reason is only present when the pain is occurrent. But our puzzle can’t be fully accounted for in terms of an occurrent/predictive difference. First, our disconnect happens even when the suffering and the alternative are both occurrent (as in Vaccination). Second, if this were the right diagnosis for the disconnect, we should expect nonsuffering versions of the disconnect trading only on the occurrent/non-occurrent distinction. We cannot think of any. Thanks to Fiona Macpherson for pressing us to consider this possibility. Another possibility might fall between the alternatives considered in the “Spike in desire, new deliberation” and “Failure of deliberative decisiveness” sections (many thanks to Dana Nelkin for suggesting this possibility). On this alternative, a subject might have multiple, alternating spikes in deliberation, leading to something that looks like a failure of deliberation, even though at any one time the subject is completely committed to one alternative or the other. We do not think this proposal is a live option. For one thing, it cannot be applied to cases of disconnect that involve a sustained, occurrent state of suffering. When we exert mental effort to overcome the pain in Vaccination, we are not changing back and forth between the various options. Second, it is difficult to make rational sense of a psychology








Matthew Fulkerson, Jonathan Cohen

that endogenously switches between alternative options in this way: it would be very unstable, at the very least. In fairness, there is some evidence that suffering may be decreased in certain populations—e.g., subjects with attention deficit hyperactivity disorder, or patients with frontal lobe damage (including damage resulting from lobotomies) of the kind described by Damasio (1994)—that are canonically described as involving deficits in practical reasoning. But we take this point not to undercut what we say in the main text. First, the deficits in these subjects appear to result from excessively Hamletian practical reasoning: these are not cases where subjects fail to engage a system that risks turning Hamletian, but cases where subjects do engage a system that does turn Hamletian (unusually often). As such, they are not the kind of subjects who fall under the prediction of the Hamlet diagnosis to which we are objecting. And, second, even an observed symmetry in these subjects’ responses might not reflect a genuine symmetry in their reactions to suffering-involving and non-suffering-involving reasons: for there is reason to believe that such subjects experience less suffering than others (Damasio 1994, 60), hence have fewer or less powerful suffering-involving reasons. This is not always the case, of course. Many analgesics also help with the underlying conditions. Ibuprofen, for instance, reduces inflammation, which is often a contributory cause of pains (especially muscle aches). Still, we take the ibuprofen primarily because it alleviates the painful experience, and only indirectly because it reduces inflammation. Though we say that this feature distinguishes states of suffering from other reasonproviding states, there is room for argument here about just how far it extends. Thus, while the feature at issue does not seem to be exemplified by ordinary doxastic states such as beliefs, it does seem sometimes seem to attach to desires. Suppose you strongly desire chocolate milk, and therefore have a pro tanto reason to acquire chocolate milk. But suppose the doctor has ordered you not to satisfy this desire; the sugar is too dangerous to your health. In this case, you now might have an additional reason directed at the desire itself. Having a strong desire that cannot be satisfied now gives you a reason to remove the desire itself. Interestingly, strong unfulfilled desires are deeply unpleasant (e.g., in the context of romantic love); and their sharing this feature might therefore provide a principled reason for counting them as states of suffering. (Reasons of space prevent us from taking up here issues about the connections between unpleasant desires and suffering; we hope to return to these matters on another occasion.) We hasten to add that nothing about the claim that states of suffering are experiencedirected motivators is intended to commit to a particular metaphysical account of pains or suffering. On the contrary, most theories of pains agree that pains are intrinsically motivating, and often give us reasons to act against their very presence or experiential qualities (see Cutter and Tye 2011 for the rare view that denies such experientially directed reasons, and recent debates about whether it is rational to “kill the messenger”; Cutter and Tye 2011; Bain 2013; O’Sullivan and Schroer 2012). Of course, as we’ve noted, there can be top-down influences on many forms of suffering. But the fact that these potential influences are difficult to induce and sustain, cognitively draining, and limited in effectiveness supports our claim in the main text. That we have left these questions unanswered would be a shortcoming in what we say if, like the suggestions canvassed above, we were attempting to explain away the disconnect reductively in terms of other phenomena. To be clear, we believe there are answers to such questions—indeed, we believe that any adequate account of suffering will ultimately have to provide some account of such features (Aydede and Fulkerson 2014). But since we don’t need to settle these issues to secure what

The agony of reason


we are claiming is a datum—viz., that there is a disconnect between states of suffering (however construed) and our broader rational lives, there’s no reason for us to take sides in these controversies here. (Cf. Aydede and Fulkerson 2018 for an attempt to build an adequate metaphysics of pain from the dual motivational structure we note here.) 30 These points need to be treated separately from cases of classical conditioning, in which even very simple organisms can be trained to endure a small shock in order to receive a reward. Such conditioning can be explained through simple causal learning mechanisms that require no rationality. 31 That said, there could be a distinct kind of conflict between reactions to forms of primary suffering (say, between reactions to thermal pain and thirst) that could arise even in the absence of reflection and reasoning, hence in simpler creatures in whom the motivational role of suffering is exclusively experience-directed. We take this point not as contravening our contention that suffering is unambivalently motivational (in our sense) for such creatures, but as showing that they may have multiple unambivalently motivational states that pull in different, and possibly opposed, ways. Indeed, that is plausibly true both for these simpler creatures and for more psychologically complicated creatures, such as ourselves, who are additionally susceptible to the particular kind of ambivalence about suffering that we have highlighted. 32 Thanks to Michael Rescorla for raising this issue.

Acknowledgements We are grateful to Marilyn McCord Adams, Craig Agule, David Bain, Brock Bastian, Michael Brady, David Brink, Jennifer Corns, Matthew Hanser, Ross Hetherington, Tom Johnstone, Fiona Macpherson, Alisa Madrigal, Manolo Martínez, Dana Nelkin, Michael Rescorla, Abraham Sapié Córdoba, Assaf Weksler, and Aaron Zimmerman for discussion and comments on earlier versions of this material. We have also benefited from discussions of this material in presentations at the University of Glasgow and the University of California, Santa Barbara, and on the Value of Suffering blog (www. valueof This chapter is fully collaborative; the authors are listed anti-alphabetically.

References Abramson, L. Y., Seligman, M. E., and Teasdale, J. D. (1978). Learned helplessness in humans: critique and reformulation. Journal of Abnormal Psychology, 87(1), 49. Amit, Z. and Galina, Z. H. (1986). Stress-induced analgesia: adaptive pain suppression. Physiological Reviews, 66(4), 1091–1120. Anscombe, G. E. M. (1957). Intention. Harvard University Press, Cambridge, MA. Auvray, M., Myin, E., and Spence, C. (2010). The sensory-discriminative and affective-motivational aspects of pain. Neuroscience and Biobehavioral Reviews, 34(2), 214–223. Aydede, M. (2014). How to unify theories of sensory pleasure: an adverbialist proposal. Review of Philosophy and Psychology, 5(1), 119–133. Aydede, M. and Fulkerson, M. (2014). Affect: representationalists’ headache. Philosophical Studies, 170(2), 175–198.


Matthew Fulkerson, Jonathan Cohen

Aydede, M. and Fulkerson, M. (2018). Reasons and theories of sensory affect. In D. Bain, M. Brady, and J. Corns (eds.), The Nature of Pain. Routledge, New York. Bain, D. (2013). What makes pains unpleasant? Philosophical Studies, 166(1), 69–89. Ballard, K. and Knutson, B. (2009). Dissociable neural representations of future reward magnitude and delay during temporal discounting. Neuroimage, 45(1), 143–150. Bechara, A., Damasio, A. R., Damasio, H., and Anderson, S. W. (1994). Insensitivity to future consequences following damage to human prefrontal cortex. Cognition, 50 (1), 7–15. Bechara, A., Damasio, H., Damasio, A. R., and Lee, G. P. (1999). Different contributions of the human amygdala and ventromedial prefrontal cortex to decision-making. The Journal of Neuroscience, 19(13), 5473–5481. Benedetti, F., Amanzio, M., Vighetti, S., and Asteggiano, G. (2006). The biochemical and neuroendocrine bases of the hyperalgesic nocebo effect. Journal of Neuroscience, 26(46), 12014–12022. Benedetti, F., Carlino, E., and Pollo, A. (2010). How placebos change the patient’s brain. Neuropsychopharmacology, 36(1), 339–354. Brand, P. and Yancey, P. (1993). Pain: The Gift Nobody Wants. HarperCollins, London. Casey, C., Greenberg, M. A., Nicassio, P. M., Harpin, R. E., and Hubbard, D. (2008). Transition from acute to chronic pain and disability: A model including cognitive, affective, and trauma factors. Pain, 134(1–2), 69–79. Clark, A. (2005). Painfulness is not a quale. In M. Aydede (ed.), Pain: New Essays on its Nature and the Methodology of its Study. MIT Press, Cambridge, MA. Cohen, J. and Fulkerson, M. (2013). Affect, rationalization, and motivation. The Review of Philosophy and Psychology, 4(4), 1–16. Craig, A. D. and Bushnell, M. C. (1994). The thermal grill illusion: unmasking the burn of cold pain. Science, 265, 252–255. Cutter, B. and Tye, M. (2011). Tracking representationalism and the painfulness of pain. Philosophical Issues, 21(1), 90–109. Damasio, A. (1994). Descartes’ Error: Emotion, Reason and the Human Brain. Avon, New York. Davidson, D. (1963). Actions, reasons, and causes. Journal of Philosophy, 60(23), 685–700. Davidson, K., Norrie, J., Tyrer, P., Gumley, A., Tata, P., Murray, H., and Palmer, S. (2006). The effectiveness of cognitive behavioral therapy for borderline personality disorder: results from the Borderline Personality Disorder Study of Cognitive Therapy (BOSCOT) trial. Journal of Personality Disorders, 20(5), 450–465. Dretske, F. (1988). Explaining Behavior: Reasons in a World of Causes. MIT Press, Cambridge, MA. Feldman, F. (2002). The Good Life: A defense of attitudinal hedonism. Philosophy and Phenomenological Research, 65(3), 604–628. Fulkerson, M. (2014). The First Sense: A Philosophical Study of Human Touch. MIT Press, Cambridge, MA. Gendler, T. S. (2008a). Alief and belief. Journal of Philosophy, 105(10), 634–663. Gendler, T. S. (2008b). Alief in action (and reaction). Mind and Language, 23(5), 552–585. Goldman, A. I. (1970). A Theory of Human Action. Princeton University Press, Princeton, NJ. Hadjistavropoulos, T. E. and Craig, K. D. (eds.) (2004). Pain: Psychological Perspectives. Psychology Press, New York.

The agony of reason


Hall, R. J. (2008). If it itches, scratch! Australasian Journal of Philosophy, 86(4), 525–535. Hardcastle, V. G. (1999). The Myth of Pain. MIT Press, Cambridge, MA. Heathwood, C. (2006). The reduction of sensory pleasure to desire. Philosophical Studies, 133(1), 23–44. Helm, B. W. (2002). Felt evaluations: a theory of pleasure and pain. American Philosophical Quarterly, 39(1), 13–30. Hofmann, S. G., Asnaani, A., Vonk, I. J. J., Sawyer, A. T., and Fang, A. (2012). The efficacy of cognitive behavioral therapy: a review of meta-analyses. Cognitive Therapy and Research, 36(5), 427–440. Holton, R. (1999). Intention and weakness of will. The Journal of Philosophy, 96(5), 241–262. Klein, C. (2007). An imperative theory of pain. The Journal of Philosophy, 104(10), 517–532. MacKillop, J., Amlung, M. T., Few, L. R., Ray, L. A., Sweet, L. H., and Munafò, M. R. (2011). Delayed reward discounting and addictive behavior: a meta-analysis. Psychopharmacology, 216(3), 305–321. Martínez, M. (2011). Imperative content and the painfulness of pain. Phenomenology and the Cognitive Sciences, 10(1), 67–90. McClure, S. M., Laibson, D. I., Loewenstein, G., and Cohen, J. D. (2004). Separate neural systems value immediate and delayed monetary rewards. Science, 306(5695), 503–507. McKay, R. T. and Dennett, D. C. (2009). The evolution of misbelief. Behavioral and Brain Sciences, 32(6), 493. Ossipov, M. H., Dussor, G. O., and Porreca, F. (2010). Central modulation of pain. Journal of Clinical Investigation, 120(11), 3779–3787. O’Sullivan, B. and Schroer, R. (2012). Painful reasons: representationalism as a theory of pain. Philosophical Quarterly, 62(249), 737–758. Rawls, J. (1971). A Theory of Justice. Harvard University Press, Cambridge, MA. Schroeder, T. (2005). Moral responsibility and Tourette syndrome. Philosophy and Phenomenological Research, 71(1), 106–123. Searle, J. (1983). Intentionality. Cambridge University Press, Cambridge. Sellars, W. (1956). Empiricism and the philosophy of mind. In H. Feigl and M. Scriven (eds.), Minnesota Studies in the Philosophy of Science, volume 1, pages 253–329. University of Minnesota Press, Minneapolis, MN. Setiya, K. (2007). Reasons Without Rationalism. Princeton University Press, Princeton, NJ. Silvestrini, N. and Rainville, P. (2013). After-effects of cognitive control on pain. European Journal of Pain, 17(8), 1225–1233. Smith, M. (1994). The Moral Problem. Blackwell, Oxford. Smuts, A. (2010). The feels good theory of pleasure. Philosophical Studies, 155(2), 241–265. Tattersall, I. (2012). Masters of the Planet: The Search for Our Human Origins. Macmillan, New York. Teuscher, U. and Mitchell, S. H. (2011). Relation between time perspective and delay discounting: a literature review. The Psychological Record, 61, 613–632. Thunberg, T. (1896). Förnimmelserne vid till samma ställe lokaliserad, samtidigt pägäende köld-och värmeretning. Uppsala Läkfören Förh., 2(1), 489–495. Velleman, J. D. (1992). What happens when someone acts? Mind, 101(403), 461–481.

14 Some paradoxes of pain for rational agency Marilyn McCord Adams

Bio-evolutionary ambivalence Our material world is an ambivalent environment for living things generally and human beings in particular. It is the world in which we have evolved and to which our coping capacities are adapted. And it is a world that is dangerous to our health, a world in which we all individually and perhaps even our kind specifically will eventually be destroyed. It is remarkable that the material stuff of our world should “cook up” structures that can host life—even more amazing, support rational agency and personal life. And these structures are fragile, more and less vulnerable to damage and destruction by the same material forces that gave rise to them in the first place. Pain systems are likewise ambivalent. On the one hand, pain networks serve a positive function, insofar as they help protect us individually and our kind specifically against threats to our survival. Pain has a cognitive dimension, insofar as it alerts us to threats to our individual and collective survival, normally and naturally in the form of bodily dysfunction (e.g., the appendix rupturing) and/or mismatches with our proximate environment (e.g., our finger resting on the hot stove). Moreover, pain goes beyond the merely cognitive to motivate: normally and naturally, pain is unpleasant, although there are rare conditions in which individuals feel pain but aren’t bothered by it. Normally and naturally, pain will not go away unless and until we do something to address the conditions to which it points (e.g., unless we remove the finger from the stove or have an appendectomy). Together, these two features of pain—the cognitive and the motivational—combine to educate higher animals, provoking them to activate instincts and evoking their capacities for developing survival skills and strategies. Natural selection favors individual specimens that are better able to do this. Specimens are the more robust, the more they are able to take of what the environment dishes out and still engage in the natural functions of their kind. On the other hand, too much of the wrong kind of pain temporarily or permanently destroys the individual’s ability to function (e.g., the throes of fullon stomach flu, or the pain of a hunter’s bullet ripping through a deer’s flesh). Whether by itself or together with the dysfunctions and misfits it signals, pain often ends the individual’s life.

Some paradoxes of pain for rational agency


The special case of rational animals Where lower forms of life are concerned, it is difficult to distinguish plants from animals, and mere biochemical tropisms from aversive sensation. What does seem likely is that the experience of pain varies qualitatively with the animal’s cognitive capacities. Rational animality such as is certainly found in human beings (and may or may not be approached in dolphins and some kinds of apes) brings along a distinctive enhancement in cognitive capacities, including a significant upgrade in symbolic functioning. This means that any assessment of the interface between pain and rational agency has to consider not only the concrete sensation of pain, but also its symbolic significance. For rational agents, pain has unconscious and/or conscious symbolic punch: not only does it signal bodily dysfunction and environmental misfits; it also signifies that the individual in one degree or another falls short of being a perfect specimen, that the individual is not only vulnerable but mortal. Burning pain not only warns us to take our finger off the hot stove; it is also one face of death! For rational as for other higher animals, pain also has social significance. How much there is to you as a person is positively correlated with high pain tolerance and wide-ranging strategic facility that works out ways, if not completely to avoid, to live through and/or with pain. Persons of substance are able to “take it” and “carry on” doing their duties and meeting the expectations of their several and various social roles. Extreme and disabling pain keeps the individual from being socially useful. Prolonged disability risks social marginalization or abandonment. Think of how Native Americans used to leave the old, who were no longer able to walk or ride a horse, to starve to death! Disability meant it was “their time to go.” Think of the way modernday families abuse elders and/or warehouse them in shabby and unsafe “facilities.” Think of the callousness with which politicians raid pension funds or threaten to cut off retiree benefits, all to avoid increasing the tax burden on those who are still contributing socially! Disabling pain is a face of social death, even as it is also a sign of dust to dust returning. Thus, when we ask how pain affects rational agency, it is mostly misleading to limit attention to the concrete sensation of pain to the exclusion of its attendant symbolic values and meanings. In what follows, I want to sketch how pain broadly construed gives rise to paradoxes for rational agents, for their functioning as moral agents on the one hand and as meaning-makers on the other.

Moral agency, double-bound Two-tiered morality For present purposes, we may distinguish two tiers of morality: call them “tribal” morality and “high” morality. Tribal morality maps evolutionary roots up into the personal.


Marilyn McCord Adams

Individuals have an obligation to fulfil their social roles and to limit ingroup hostilities in the interest of tribal survival. But tribal morality does not sponsor much in the way of out-group altruism, except where dictated by outgroup alliances that benefit the tribe. The obvious reason is “Darwinian”: resources are and are perceived to be scarce. What they eat comes off of our plate! We may well be in competition with out-groups for the turf and stuff that afford the necessities of life. By contrast, high morality moves in two ways beyond Darwinian drives towards individual and species-preservation. First, it universalizes concern beyond the tribal to include all human beings. Second, high morality insists that human beings are to be honored as ends in themselves and not merely as socially useful or as means to someone else’s ends. At the very least, high morality lays down dictates of minimal decency: not to treat other human beings in such a way as seriously to compromise their capacity to engage in characteristically human functions.1 Dictates versus concrete experience Either way, rational moral functioning demands that agents take into account how bad things are—circumstances, actions and/or their consequences—in deliberating about what to do. Moral agents are obliged to take into account any pain and suffering of human beings that would be attendant on a proposed course of action, the worse the harms, the weightier the consideration. But—especially where pain and suffering are concerned—one cannot appreciate how bad they are without being able imaginatively to represent them to oneself—i.e., without first-person experience or first-hand empathetic engagement. Here there are analogies between pain and color. No matter how many color facts a totally blind person may know, s/he is missing something important (perhaps what is most important) about color because s/he cannot imagine how it looks or what it is like to see it. Likewise, no matter how much neuro-biology the analgesic may know, s/he can’t know how it feels to be in pain. Fully to assess the significance of what s/he is contemplating, the rational agent needs imaginatively to represent to her/himself what it would be like to experience pain of the sort her/his action would occasion. Imaginative representation is key for rational deliberation because it reacquaints the agent with the motivational dimension of pain. Feeling or empathetically re-presenting intense pain, one experiences the motivational urgency to make it stop. This is a vivid and concrete measure of how bad it is to experience pain of that kind. Attention to abstract moral dictates will not be enough, because—among other things—what abstract moral dictates abstract from is pain’s motivational sting!2 Moral competence requires rational agents to cultivate sensitivity and a capacity to empathize with the pain and suffering of others. Not doing so is characteristic of the callous who in consequence deliberate under conditions

Some paradoxes of pain for rational agency


of culpable ignorance. One can, of course, know what types of action are harmful in the abstract, and can acquire head-knowledge in an ethics class that torturing babies is entirely off-limits. But experience shows that abstract knowledge is necessary but not usually sufficient as a guide to practical action, because—minus re-acquaintance with the motivational sting—agents find it easy to rationalize accepting the severe pain and suffering of others as the price of avoiding some other evil or securing some great enough good. Thus, when reports leaked out about Abu Ghraib and practices at Guantanamo Bay, Americans were invited to believe that torture is not utterly off limits, but is justifiable when it promotes national security. Alternatively, authorities indulged in semantics: the pain and suffering caused detainees was not really torture, because water-boarding was an allegedly effective means to getting information that saved American lives. Likewise, dropping atomic bombs on Hiroshima and Nagasaki was not really a crime against humanity because—despite the fact that heat from the explosions evaporated human beings—the attacks hastened the war’s end and, once again, saved American lives. Offensive warfare, pre-emptive strikes were “justified” in Iraq, because they were supposed to protect American and British interests. Military jargon carries abstraction to extremes with its sterilized references to “collateral damage”! Leaving the theater of war for the church, we have seen how the refusal of empathetic engagement with sex-abuse survivors, made it easy for Roman Catholic bishops to pursue policies of cover-up and reassignment of predator priests, all as an expression of clerical collegiality and in the interests of protecting the institution’s reputation. Tribal morality in state and church, all around! Moreover, the requirement of first-person experience and/or first-hand empathetic engagement is not a one-time thing. Precisely because pain, especially severe pain, is a face of social ruin and death, we instinctively turn away from it, experience difficulty in “sticking with it.” Vivid awareness of the urgency of getting the pain to stop, fades quickly. Recall how Job’s friends begin by sitting with him in silence for seven days, empathetically engaging and mourning his condition. But when Job begins to accuse God of injustice, his friends fear being held guilty by association with a blasphemer, beat a retreat into theological theory, and increasingly blame the victim. Many of the Roman Catholic bishops who participated in the cover-up of clerical pederasty must have had pastoral experience of pain and suffering like enough to what the victims of predator priests were going through. Some generals in the Pentagon will have seen front-line action and will have had personal experience of what “collateral damage” concretely means. Most of them are able to regard ghastly human pain and suffering with a cool and calculating eye, only because their closerange empathetic engagement is not fresh enough. Once again, such ignorance is culpable. Moral agency is supposed to deliberate from a realistic appreciation of the harms, and this requires repeated close-range empathetic engagement.


Marilyn McCord Adams

Information versus distortion We all know, however, how experience of intense pain and suffering at least distorts our perspective. The worst types of pain and suffering overwhelm us to the extent of wiping out—whether temporarily or permanently—our ability to function as rational agents altogether. Hunger-crazed mothers in the siege of Jerusalem, starving deportees in Australia resorted to cannibalism. For the soldier whose limbs have just been blown off, for the post-operative heart surgery patient writhing in pain, reason is—as Aristotle would say—“bound.” In the Bible story, Job’s reason is not bound, but his reasoning is distorted. Job’s suffering makes him an expert on how intense the torment is for him. But Job’s suffering combines with the abstract moral theory he shares with his friends—the act consequence principle: good for good, evil for evil—to caricature who God is and what Divine policies are. Job accuses God of wrong, slaying the innocent with the guilty; of being more a chaos monster than a creator, of being generally incompetent to govern the world. Prisoners waterboarded or tortured on the rack normally and naturally demonize their tormenters. Released survivors may for years hold onto the conviction that nothing is too bad for their captors. Alternatively, the tortured may become willing to do anything—including betray their deepest loyalties—to get the pain to stop. Likewise, those hospitalized with excruciating pain may demand a lethal dose of morphine. A moral paradox So there is a paradox about the moral competency of rational agency where intense pain and suffering are concerned. To be competent to calculate, the rational agent has to appreciate how bad the harms—the pain and suffering—are. Full appreciation requires first-hand empathetic engagement, which imaginatively represents the dire state of those who absorb the harms and reacquaints the perceiver with its motivational sting. And empathetic engagement—the deeper it is, the more like first-person experience—not only informs, but distorts the agent’s judgment. In the worst case, it overwhelms rational functioning altogether so as to make it temporarily or permanently impossible. Intense pain and suffering double-binds moral functioning. Either the rational agent remains disengaged and so lacks the information needed to take the worst with full moral seriousness; or the rational agent does empathize with the result that her/his rational functioning is distorted and/or disrupted. Experience-grounded deontology To avoid the moral paradox of pain, what we need is a way to capture the motivational sting of the experience, without being so caught up in the experience as to be unable to think straight. Deontological principles claim to

Some paradoxes of pain for rational agency


carry motivational urgency, because they assert themselves as binding, no matter what. What if we used the immediate aftermath of first-person experience or first-hand empathetic engagement to capture the content of what we have learned in fresh deontological principles about the sorts of pain and suffering that it is impermissible to impose as costs on others? Holocaust memorial museums forward the slogan “never again!” to send the message that genocide and its attendant de-humanizing practices are everywhere and always forbidden. Geneva conventions about the treatment of prisoners of war, the wide-ranging and eclectic Universal Declaration of Human Rights, attempt to lay down what we might call “a deontology of human decency.” General principles might be “don’t do anything to degrade another human being,” “treat every human being decently,” “do not enslave another human being,” “organize society in such a way that everyone has a decent standard of living.” Particular principles might include “except very temporarily as part of a medical procedure aimed to benefit the patient, do not do anything to a human being that disables her/his human functioning”; “do not starve, dehydrate, expose to the elements (life-threatening heat or cold), or torture another human being.” Useful as these initiatives may be, experience shows that they do not dissolve the paradox. Abstract deontological principles assert their categorically binding force, but far from all moral agents credit their unqualified claims. Categorical imperatives assert, but they do not reliably sting! Many argue that no ethical principles have deontological force, that none is unexceptionable, because ethical principles rest on a consequentialist foundation. What acts are right or wrong to do, is a function of their consequences. It is easy to turn consequentialist calculations into rationalizations for treating other people indecently. Nations do this when they claim the right of self-defense and go to war, which involves a posture of accepting degrading pain and suffering for home troops as well as enemies, all for the glory of the fatherland. Experience shows the fragility of high morality. Once collectives turn tribal, genocidal pogroms seem justifiable as means to self-preservation and flourishing. In the twentieth century, think not only of Nazi Germany, but of erstwhile Yugoslavia and Rwanda. Widely publicized moral disasters such as these do remind human societies of just how tribal they are. Sometimes recently engaged ghastliness drives nations to go beyond giving lip-service to deontological moral principles, to pass laws that impose penalties (e.g., it is now a crime in Germany publicly to deny that the Holocaust happened; economic embargoes were imposed on South Africa under apartheid). In a tribal species such as ours, tough legal sanctions may do more than high morality to give dictates motivational sting. Complementary distortions? Where intense pain and suffering is concerned, each human perspective is distorted. But they are distorted in different ways, and many such perceptions


Marilyn McCord Adams

may also tell some truth. Job is wrong that God is a chaos-monster, but right that his pain has brought him to the breaking-point. Job’s friends are right that God is just and that Job’s restoration lies in turning to God, but wrong when they suggest that his suffering can’t be all that bad, or that his sin must be very great to deserve such misfortunes, or that talking back to God will bring down final ruin. Persons in great physical pain are in no condition to make important financial decisions, much less choose among the medical options that are available to them. Physically sound persons making up living wills are not fully in touch with how bad certain bodily states can be, but they have their wits about them, can read abstract accounts and weigh options. Friends or relatives with power of attorney have the advantage of empathetic engagement, but may still be in a position to make discriminations and apply directives. Perceivers close to the pain experience the strength of its motivational sting. Those at greater or lesser remove remain able to place the pain and suffering in a context that includes other factors (especially including the patient’s overall or long-term benefit) that have some claim to motivate us. Pooling perspectives and separating the wheat from the chaff may furnish a way forward, albeit an imperfect way forward.

Meaning-making, stimulated and stalemated Evolution results in organisms furnished with structures and powers that in fact promote individual and species-survival. The anti-teleology of some scientific methods forbids us to say that the animals developed these structures and powers in order to survive. Rather, we are told, individuals chanced to come equipped with them and survived to reproduce themselves as an efficientcausal result.3 By contrast, rational agents are purpose-driven meaning-makers, who organize their activities and projects around worthwhile goals. If meaning-makers weave their lives into a tapestry of narratives, pain and suffering are plot complicators. Within limits, we value plot complications, because necessity is a mother of invention, a prod to creative problem-solving. Meeting challenges is important for personal development, because it brings out more of what there is to a person. We may find “fair-weather” persons shallow or boring; those who have coped with significant obstacles more creative and fascinating. Indeed—and this has evolutionary roots—the more pain and suffering they have lived through, worked around, or triumphed over, the more we admire them. But the worst pain and suffering can interrupt, whether temporarily or permanently disrupt attempts to make positive sense of one’s life. Acute pain and suffering (as occasioned, e.g., by an attacking heart or a spastic colon) can “bind reason” and so make rational activity impossible. Simone Weil defines the category of affliction, which she understands as severe chronic pain that constantly confronts its victims with a sense of themselves as defective specimens, not capable of normal functioning and/or of doing things

Some paradoxes of pain for rational agency


they consider meaningful. Affliction convinces its victims that they are not the kind of thing that could be valuable to themselves or others and so fills them with self-loathing (Weil 1951: 117–136). Then there are persons with fatal diseases that predictably inflict excruciating pain at the end. When the end draws near, they sometimes conclude that life is not worth living anymore and long to be euthanized. Naturally produced severe and ghastly pain can be socially alienating to the extent that it convinces victims or their erstwhile associates that they are no longer useful to anybody. But the negative social meaning of severe and ghastly pain is amplified when it is deliberately inflicted by others. Think of the cruel and unusual torments of war and political prisons: for example, starvation, dehydration, and sleep-deprivation; brutal beatings; medical experimentation; torture on the rack or by hanging from the wrists; incineration. Degrading treatment that aims at the breakdown of personality sends victims the message that they are subhuman—worse than dogs, so much garbage—and leaves them unable to make any positive sense of their lives. Some emerge able to reason but unable to experience anything as valuable enough to make life worth living. Some come out convinced that who and what they have done or been in dire circumstances puts them beyond redemption. Still others are so traumatized that they lose their capacity to narrate their lives into stable or meaningful stories (e.g., soldiers returning from war with PTSD) or even to manage the ordinary tasks of daily life. So, for human beings in this world, pain is both a plot complicator and a plot stopper. Pain and suffering may be a condition of the possibility of having an interesting rather than a trivial life. And too much of the wrong kind of pain in the wrong conditions can seemingly destroy any possibility of positive meaning. Pain can disable us from pursuing goods. Pain can keep us from finding previous goals worth caring about. Pain can cut us off from activities and social relationships. And pain can keep us from thinking about much of anything. The ambivalence of pain is treacherous insofar as the same pain can bring on our finest hour or become the occasion of our deepest ruin. Paradoxically, the very worst pain and suffering is what most cries out to be made sense of, and the very worst pain and suffering is what stumps, stalemates, and/or destroys our capacity to make positive sense of anything at all. Some but only some of these effects submit to human antidotes. Rational agents are meaning-makers. But human beings are also social animals, and both high morality and tribal morality encourage social solidarity. Individuals may be meaningful because others count them meaningful, even when they themselves do not think so, whether because they are too young (infants and toddlers), too senile (Grandpa with Alzheimer’s), too mentally deranged (a schizophrenic cousin), too physically sick (persons with MS, Lou Gehrig’s disease, or advanced cancer), or too depressed. People visit the sick to demonstrate that—despite their present inability to function—they are still valued members of their social circle. High morality would send the message: “we love you for yourself and not just for what you can do for us.” Tribal


Marilyn McCord Adams

morality is usually a subtext: “Don’t let the side down! You owe it to us to hang in there, to get through this and return to business as usual!” Social solidarity also takes the form of providing and seeing recovering patients through rehabilitation programs; of changing their living arrangements to be friendlier to their condition (e.g., low counters for people confined to wheel chairs; communal living with meals and medical care laid on for the ageing); in the right moments, offering counsel, helping those who have suffered to make sense of their experiences and appreciate the positive significance of their lives. Because meaning-making is partly shared, individuals don’t have to take the lead at every juncture.

God and the paradoxes of pain The God of traditional Christianity is not metaphysically material. The Divine nature is not bodily and so includes no networks of pain nerves to alarm it against bodily dysfunction and threatened bodily damage. Yet, because the God of traditional Christianity is Spirit, God is personal: a rational agent who acts for reasons, a purpose-driven meaning-maker who organizes Divine activities around chosen projects and worthwhile goals. The God of traditional Christianity is the creator and providential governor of this material world, which God has chosen on purpose and with the purpose of entering into wholesome relationships with rational animals, with material persons such as we human beings are. Whether or not the God of traditional Christianity is held to have moral obligations to creatures,4 God’s advertised positive interest in human persons seems to signal that Divine agency—in choosing among possible worlds and providential plans—counts the costs in pain and suffering to any and all rational (indeed any and all sentient) creatures involved. If God is a purposedriven rational agent, do the above-discussed paradoxes of pain vex Divine agency as well? Omniscience and the moral paradox of pain To act as paradigm rational agency with full knowledge might seem easy for God, because the God of traditional Christianity is omniscient. True, where intense pain and suffering are concerned, every human perspective is somehow distorted—whether that of those who stick to abstract principles; those who empathize with reserve; those who empathize with abandon; and those who are in the throes of pain themselves. But isn’t this merely a limitation of human consciousness, which is causally so configured that intense sensation obstructs rational functioning? What if some other sort of perceiver had enough cognitive power to keep pain and its motivational sting vivid, while exercising practical reason to the full? Wouldn’t the Christian God turn out to be such an “ideal observer” who is able to weigh courses of action with full appreciation of any pain and suffering they would involve? In fact, saying

Some paradoxes of pain for rational agency


“yes” is non-trivial, because it requires a fresh understanding of what the doctrine of Divine omniscience includes. Abstract and propositional? The God of traditional (i.e., of patristic and scholastic) Christian theology is immutable and impassible. Duns Scotus holds that God understands all creatable natures and God knows all true propositions. But insofar as Divine knowledge abstracts from sensory “feels” and qualia, this sort of Divine omniscience would have God operating on the basis of abstract principles. God would know that pain of a high intensity normally and naturally motivates the person who suffers from it to do something to get it to stop. But the God of patristic and scholastic Christian theology would not be acquainted with the “ouchy”—i.e., with the motivational sting of that pain, the factor that shows how bad the pain really is. Divine sensitivity? Taking many pages from the process philosophy of Alfred North Whitehead, Charles Hartshorne (1948, 1984) rejects this classical conception of Godhead in favor of panentheism, according to which everything other than God exists in God. Panentheism contrasts with pantheism insofar as pantheism identifies God with everything that is. But—as Hartshorne sees it—parts are essential to their wholes. If everything were God, then God could not exist without each and everything that does, did, or will exist. But Hartshorne thinks that creatures are contingent, and that—while God could not have existed without any creatures at all—God could have existed without any individual or collection of individuals in particular. Those that do exist contingently are thus in God and constitute God’s body. Revising the classical conception, Hartshorne conceives of omniscience in terms of sensitivity. Divine omniscience is not only propositional. God experiences all of our experiences and feels all of our feelings, most notably, our pain and suffering. Thus, Hartshorne’s God knows something that Scotus’s God is missing about what God is risking when God persuades the world to evolve in the direction of greater complexity and intensity. Nevertheless, Hartshorne insists, Divine cognitive capacity is so great that having all of those feelings and experiences does not interfere with, but rather furnishes material for God’s rational creativity that persuades creation into ever more perfect configurations. A God who feels all of our feelings and suffers with us goes a long way in the direction of empathy. But Divine omniscience means that Divine consciousness includes many other contents, alongside the pain and suffering we experience in this present life, not least, the many excellences that go with Godhead itself. We may well wonder: even if Hartshorne’s God feels our pain, does Hartshorne’s God really appreciate what it is like for us to feel our pain.


Marilyn McCord Adams

We experience the pains as overwhelming, but God does not and cannot experience them as interrupting Divine rational functioning. Divine omnisubjectivity Linda Zagzebski (2008, 2013, 2016) tries to meet precisely this challenge, when she proposes that Divine omniscience includes Divine omnisubjectivity. God not only feels my feelings. God knows what it is like for me to feel my feelings. God knows what it is like for me to be me, without ever losing sight of the fact that God is not me. And so on for each and every actual and possible sentient creature: God knows what it would be like for that creature to have this experience, without losing sight of the fact that God is not that creature and that God does not own that experience. Zagzebski sees Divine knowledge of our subjectivity as analogous to our empathetic knowledge of one another. Where human empathy is concerned, she endorses a “two-ego’s” imaginative representation theory, according to which when X empathizes with Y’s experience, X consciously produces a copy of what Y is experiencing and remains aware that X’s copy is a copy of what Y is experiencing. X thereby knows what it is like for Y to have the experience, while keeping straight in X’s mind that X is not Y and that the experience copied belongs to Y and not to X. Conscious and deliberate imaginative representation is supposed to give the empathizer-ego cognitive access to the other’s experience, while preserving the awareness that the two egos and their experiences are distinct. Where God is concerned, the difference is that God is immediately aware (in medieval jargon, has an intuitive cognition) of what it is like for me to be me and what it is like for me to feel intense pain. God’s knowledge does not involve any act of Divine imagination that represents what it is like for me to be me or what it is like for me to feel intense pain. Moreover, God’s knowledge is complete. Zagzebski reckons that for God to be fully informed in creating and governing the world, Divine omnisubjectivity would need to encompass, not only what it is like for me to feel this pain, but what it would be like for each subject to have each of the experiences of which it is capable. Whether or not such knowledge would interfere with immutability and impassibility as traditionally conceived, Zagzebski is assuming that Divine omnisubjectivity would not interrupt or upset, but rather would be an important asset in the exercise of God’s creative rationality. Undoubtedly, such intuitive cognition would allow Zagzebski’s God to be much better informed than we are about what harmful outcomes might cost others. Moreover, Zagzebski’s “two-ego’s” model would allow God to see how suffering undoes the created person’s subjectivity, while leaving the Divine subjectivity intact. But the latter may well cause us to wonder whether Zagzebski’s God really knows how bad the worst suffering is for us. A big part of what is so bad about intense suffering is the experience of falling apart without a back-up stabilizer. To the extent that the Divine subjectivity is not and

Some paradoxes of pain for rational agency


has not been because it metaphysically cannot be undone, God remains out of touch with some of the worst aspects of our experience. Nor would Incarnation allow Divine subjectivity to fill this informationgap. Incarnation allows the Divine person metaphysically to own a particular human nature. Incarnation puts Zagzebski’s “two-egos” into one metaphysical supposit. Even if Christ’s human subjectivity does come undone in the throes of crucifixion, Christ’s Divine subjectivity will remain an intact observer. The Divine person will experience what it’s like for us in his human nature, but his Divine subjectivity will be none the better informed.5 Divine ideal observer? Even if Divine omnisubjectivity would not include all of the knowledge that first-person experience would give us, Zagzebski’s God would have intuitive cognition of what pain does to us and of its motivational sting. Relative to the moral paradox of pain, God would be much better informed than we are in framing God’s own providential plans. Would an appeal to a Divine Ideal Observer also open a better way around the moral paradox of pain for God’s human devotees? Perhaps. Hitler knew in the abstract that applying phosphorus to the skin in medical experiments or marching children into fiery ditches were beyond the pale in traditional morality. But Hitler was callous. He countered with an ideology of “kein Mitleid.” To be fully competent morally, we need a kind of knowledge that packages cognitive contents (which acts are to be pursued or avoided, which harms are not to be inflicted) with motivational sting. Think of the biblical Abraham or Jacob. Experience of God commanding or counselling this or that course of action, carried its own motivational force. “The fear of the Lord is the beginning of wisdom” (Proverbs 9:10). Like most human groups before and since, ancient Israelis were tribal, not motivated by high morality to empathize with the plight of immigrants and resident aliens. The Bible’s God substitutes the motivational sting of Divine authority—not of merely human legal and political institutions—for the motivational sting of perceived pain and suffering, in God’s largely vain attempt to get people to do the right thing. Omnisubjectivity and the problem of evil To the extent that omnisubjectivity puts Divine agency in an advantageous position relative to the moral paradox of pain, it digs God deeper into the problem of evil. If God acts with intuitive cognition of what it would be like for Holocaust prisoners to feel the pain of their torment, and yet chose to let it happen anyway, doesn’t that prove that God is like Hitler—certainly not loving; more likely, callous or cruel? In the second half of the twentieth century, the problem of how God and evil can co-exist in the same world received a lot of attention among analytic philosophers of religion. Interestingly, many defenders of their compossibility


Marilyn McCord Adams

took a page from Job’s friends. Because it would be too scary to think that God exists but is an evil demon, or that God exists but is callous or cruel, they insist that there are morally sufficient God-justifying reasons for whatever evils actually exist. Where evils themselves are concerned, most such God-defenders bracket their own experience of evil’s motivational sting the better to retreat into abstractions, and attempt generic explanations of evils— say, evils as consequent upon created free choices, evils as necessary for cosmic excellence, evils as required for an apt environment for soul-making, evils as an occasion for being of use. Alternatively, they appeal to ignorance and insist that things would look to us pretty much the same on the ground, whether God had reasons or not. What such God-defenders mostly do not do is allow what they learn from empathetic engagement of the worst pain and suffering, to call into question the adequacy of their proferred generic sufficient reasons (cosmic excellence, free will, soul-making, etc.) and/or to challenge their assumption that means-end reasoning (e.g., God permitted the Holocaust in order to allow Nazis wide-scope free will) can justify the worst.6 Divine meaning-making pain paradox? Meaning-making is an essential function of rational agents. For us, the meaningmaking paradox is that the pain required for some interesting meaning-making is apt to destroy our meaning-making capacities altogether. Moreover, the very same pain and suffering can teeter-totter us between soaring achievement and total ruin. The worst that we can suffer, be, or do, is what most cries out for redemption into positive significance. And the worst that we can suffer, be, or do, is what may not only stump but also smash our meaning-making capacities. Divine purposes participate in a version of this paradox. The God of traditional Christianity is a purpose-driven meaning-maker. Because God is not the world and—pace Hartshorne—the world is not in God, Divine providence can be seen as imposing purposes on this world from the outside. God’s aim in allowing or nudging matter to evolve structures that can host personal life, is that the material world should also mean something from the inside. God wants there to be insider meaning-makers who see and value God’s world in distinctive ways and seek to make more of it (by composing symphonies, writing poetry, building cities or at any rate a better mouse trap). But Divine purposes seem self-defeating if God creates us to make positive sense of our world in which pain and suffering are both necessary for significant positive purposes and yet destructive of our capacity to make meaning at all—in which the very worst is both the pain and suffering that most demands to be made sense of and the pain and suffering that wrecks us as meaning-makers. The paradox intensifies, because God wants us to be realistic meaningmakers. Some individuals may be lucky enough to sail through life avoiding first-hand experience of the worst. But God calls on us human beings to widen our horizons beyond the individual and the tribal: to try to make sense,

Some paradoxes of pain for rational agency


not only of our own individual lives or of the destiny of our race or nation, but of the human predicament in this world generally. This assignment requires us to take into account human beings of “all sorts and conditions”— like Job, to face up to the worst that human beings can suffer, be, or do. But—as before—to be fully realistic about pain and suffering requires either first-person experience and/or first-hand empathetic engagement. And—as before—opening ourselves to either risks breakdown. Happily, Divine agency has more resources for resolving the meaningmaking paradoxes than we do, most importantly, that God is a being a greater (a more excellent or worthy) than which cannot be conceived. What God is, is immeasurably valuable. Apt relation to great goods is life-enhancing. Traditional Christian theology holds: apt and intimate access to the Good that God is, is immeasurably good for us. This means that God has the capital to compensate human beings for the pain and suffering of this present life. What Divine omnisubjectivity knows is not only what it would be like for us to experience ordinary pain and suffering and what it would be like for us to experience the worst; but also what it would be like for us to experience the best, to enjoy God’s intimate presence and friendship. So far as the problem of evil is concerned, it helps to think that—whatever reasons Divine providence has for making this material world and permitting (or nudging) it to evolve structures that host personal life—God would not have set human beings up for the sometimes seemingly ruinous sufferings of this world, if God did not know how to make it up to them with infinitely more than they could ask or imagine. To cut through the meaning-making paradoxes, to make a success of God’s own project of populating the world with insider meaning-makers, Divine providence will have to go further by putting insider meaning-makers in a position to make positive sense of the human predicament in this world as we find it, and to find positive significance in their own individual lives. Though much better resourced than human comforters, some of God’s strategies are the same. First, God answers pain-induced despair with gestures of solidarity. God becomes an insider meaning-maker. God the Son becomes Incarnate, takes to himself a particular human nature, in that human nature attempts to respond to a Godward calling and in consequence tastes a representative sample of the worst that human beings can suffer, be, or do. If excruciating pain does not at every moment wipe out, it mightily interferes with rational functions. Symbolically, crucifixion adds insult to injury: from the Roman point of view, declaring that the victim is not fit for polite society; from the standpoint of Deuteronomic law, asserting that the crucified is cursed, cut off from God and the people of God. Crucifixion was meant to put an end to Jesus’ Messianic pretensions, to shatter the positive purposes that centered his human life. If Divine omnisubjectivity furnishes God with intuitive cognition of what it is like for a human being to suffer the torments of crucifixion, Incarnation does not (as above) so much add information as give God ownership: the


Marilyn McCord Adams

pains and torments of Christ crucified within the frame of his human nature, are God’s very own. Divine solidarity cancels curse (if God takes God’s stand with the cursed, then the cursed are not cut off from God after all) and insists that—however disabling—nothing we can suffer, be, or do can separate us from the love of God. Solidarity in suffering is also a manifestation of Divine good will, because it shows that God does not ask more of us than God does of Godself. Divine solidarity both asserts and confers positive meaning on suffering human beings. Indeed, because God is infinitely more than we can ask or imagine, Divine solidarity pays suffering human beings immeasurable honor. Divine solidarity also catches up the worst that individuals can suffer, be, or do, into their relationship with God by turning the worst into shared predicaments, episodes and seasons which God and human beings got through together. Nevertheless, if human beings are meant to be insider meaning-makers, it is not enough for God to value them and find meaning in their lives. Suffering human beings must be brought to a point where they themselves can make positive sense of their lives, become so convinced of the positive significance of their lives as a whole as to become glad to have lived them. Here Divine omnipotence can outdo merely human friends and compatriots. God can restructure the material world into a less dangerous environment. God can also heal traumatized and damaged meaning-making capacities, and—figuratively as well as literally—restore the dead to life. God the Creator fully understands human nature; Divine omnisubjectivity is the ultimate in therapeutic insight. God will know how—through labyrinthine paths and individual syllabi—to win the trust of human beings who have suffered the worst and to coach them into finding positive significance even in the very worst experiences of their lives. Because of who and what God is, God is able to be immeasurably good to everyone, and God’s project of a material world with insider meaning-makers, can be successful after all.7

Summary conclusion Pain produces paradoxes for human rational agency, because two of its essential functions—moral deliberation and meaning-making—require an appreciation of how bad pain is. Prominent among the factors moral agents must take into account in assessing alternative courses of action, is the pain and suffering they would cause. Honest meaning-making is realistic about persons’ life experiences. How good or bad they were affects what sense can be made of them and bears on our judgments about whether human life in this world is worth living. But fully to appreciate how bad pain and suffering are, requires first-person experience or first-hand empathetic engagement. And the limitations of human psychology mean that the worst pain and suffering disrupt the exercise of human reasoning—whether by distorting our perspective, by stalemating our problem-solving, or by disabling our ability to calculate altogether. Where

Some paradoxes of pain for rational agency


possible, withdrawal into abstractions preserves calculative capacities but distorts perspectives by leaving us culpably ignorant of what we are reasoning about. The result is that competent human rational functioning is compromised in the very cases where it is most important to get it right. Solving the moral paradox of pain for human beings requires finding a way to preserve calculative capacities while still confronting us with motivational sting. Humanly speaking, this challenge can be only imperfectly met by division of labor. Pooling our distorted perspectives and letting them correct one another, helps. Letting fresh experience and empathetic engagement inform deontological principles and attaching graduated legal sanctions substitutes the motivational sting of conscience or threatened punishment for that of experienced pain itself. First-person experience of the worst pain and suffering at best stumps and in extreme cases wrecks the individual’s ability to make positive sense of their lives. Social solidarity that counts a person meaningful, attempted rehabilitation, relocation into more hospitable living conditions, and counselling—in ordinary time, where small, medium, and large evils are concerned, all of these help individuals recover a conviction that their lives are worth living. But post-Holocaust narratives serve to document how they are not enough to stand up against the worst that we can suffer, be, or do. Not only are individual participants stumped and disabled. Friends and onlookers also find it reasonable to conclude that participation in the worst ruins human lives. Generalizing, the fact that human beings are radically vulnerable to the worst in this world, makes it reasonable to conclude that human being in this world can as easily be a curse as a blessing. Where Divine rational agency is concerned, the “size-gap” makes all of the difference. For starters, Divine psychology is more capacious. Even if Divine omnisubjectivity does not entirely escape the moral paradox of pain, it still leaves God much better informed than we are about the significance for others of the harms our actions may cause. More importantly, what God is, enables God to escape the meaning-making paradox and to solve the problem of evil. For God is infinitely more than we can ask or imagine, an incommensurate good, apt relationship to which is incommensurately good for human beings. Divine solidarity with human beings in the worst that we can suffer, be, or do, lends a dimension of positive significance to the worst, by catching it up into the warp and woof of an individual’s relationship with God. According to traditional Christianity, Divine powers to relocate us, to rehabilitate and coach our meaning-making capacities are what will make it possible for those who experience the worst to recognize that the worst was not a plot stopper, but only a plot complicator after all.

Notes 1 This sort of distinction is noted by Richard Swinburne in The Evolution of the Soul (Swinburne 2007: 220–230; by Kathleen Taylor in Cruelty: Human Evil and the





5 6 7

Marilyn McCord Adams

Human Brain (Taylor 2009: 36–41); and by Paul Bloom in Just Babies: The Origins of Good and Evil (Bloom 2013: 6 and passim). Appeals to sensory or phenomenological “feels” or qualia, form the basis of many critiques of philosophical attempts to reduce the mental to the physical and/or to identify the mind with the brain. See, for example, Jackson (1986), and more recently Nagel (2012, especially chapters 3 and 5). For an extensive evaluation and critique of such arguments, see Pereboom (2011). Despite Pereboom’s judicious cautions, I remain sympathetic to the argument from qualia against reductive materialism in philosophy of mind. Space does not permit me to defend position here. See Thomas Nagel’s Mind and Cosmos for an interesting and provocative attempt to resist this verdict (Nagel 2012, esp. chapters 3–5). He argues that the facts of human consciousness, cognition, and rational purpose-driven valuations, invite a reconsideration of natural teleology. Most medieval theologians (e.g., Anselm, Scotus, Ockham) make a point of denying that God has obligations to creatures, but most analytic philosophers of religion take for granted that because God is personal (and therefore rational), God’s perfect goodness must be moral goodness. For a sample of my attempts to problematize the latter assumption, see my “Ignorance, Instrumentality, Compensation, and the Problem of Evil” (Adams 2013). For a discussion of the metaphysics of Incarnation, see my Christ and Horrors: the Coherence of Christology (Adams 2006: 108–143). For some scathing critiques of such approaches, see D. Z. Phillips’s The Problem of Evil and the Problem of God (Phillips 2004). I elaborate on these ideas about God and suffering at some length in my books Horrendous Evils and the Goodness of God (Adams 1999) and Christ and Horrors (Adams 2006).

Acknowledgements Thanks for helpful comments go to Robert Merrihew Adams and to participants in the “Suffering and Reason” conference, held at the University of Glasgow, 4–6 July 2014.

References Adams, Marilyn McCord. (2006). Christ and Horrors: the Coherence of Christology (Cambridge: Cambridge University Press). Adams, Marilyn McCord. (1999). Horrendous Evils and the Goodness of God (Ithaca, NY: Cornell University Press). Adams, Marilyn McCord. (2013). “Ignorance, Instrumentality, Compensation, and the Problem of Evil.” Sophia 52(1), 7–26. Bloom, Paul. (2013). Just Babies: The Origins of Good and Evil (New York: Broadway Books). Hartshorne, Charles. (1948). The Divine Relativity: A Social Conception of God (New Haven, CT: Yale University Press). Hartshorne, Charles. (1984). Omnipotence and Other Theological Mistakes (Albany, NY: University of New York Press). Jackson, Frank. (1986). “What Mary Didn’t Know.” Journal of Philosophy 83(5), 291–295.

Some paradoxes of pain for rational agency


Nagel, Thomas. (2012). Mind and Cosmos: Why the Materialist Neo-Darwinian Conception of Nature is Almost Certainly False (Oxford: Oxford University Press). Pereboom, Derk. (201). Consciousness and the Prospects of Physicalism (Oxford: Oxford University Press). Phillips, D. Z. (2004). The Problem of Evil and the Problem of God (London: SCM Press). Swinburne, Richard. (2007). The Evolution of the Soul: Revised Edition (Oxford: Clarendon Press). Taylor, Kathleen. (2009). Cruelty: Human Evil and the Human Brain (Oxford: Oxford University Press). Weil, Simone. (1951). “The Love of God and Affliction.” In Simone Weil, Waiting for God, trans. Emma Craufurd with introduction by Leslie A. Fiedler (New York: Harper Colophon Books). Zagzebski, Linda. (2008). “Omnisubjectivity.” In Oxford Studies in Philosophy of Religion, vol.1, ed. by Jonathan Kvanvig (Oxford: Oxford University Press). Zagzebski, Linda. (2013). Omnisubjectivity: A Defense of a Divine Attribute, The Aquinas Lecture 2013 (Marquette, WI: Marquette University Press). Zagzebski, Linda. (2016). “Omnisubjectivity: Why It Is a Divine Attribute.” Nova et Vetera 14(2), 435–450.

15 Suffering as a virtue Michael Brady

Nearly everyone agrees that physical pain is bad. Indeed, if anything merits the status of a platitude in our everyday and philosophical thinking about value, the idea that pain is bad surely does. Equally, it seems clearly true that negative emotions – like despair, loneliness, grief, disappointment, guilt, shame, lovesickness, and the like – are all bad as well. We are strongly inclined to pity and feel sorry for those who experience pain and negative emotion; we are motivated, at least some of the time, to do what we can to alleviate such things. Now I think that physical pain, and certain kinds of negative emotional experience, constitute forms of suffering.1 What all such states have in common is a component of negative affect or unpleasantness. But a state’s being unpleasant doesn’t suffice for suffering. Things can, after all, be mildly unpleasant. Instead, it is plausible to think that pains and negative emotions constitute suffering insofar as they are suitably intense, or insofar as we have a suitably strong desire that they cease. In a slogan: suffering is unpleasantness that we mind. Given this, it is plausible to claim that suffering is bad: that those who suffer should be pitied, that we should do what we can to alleviate suffering. This view of the value of suffering is, of course, compatible with another: that suffering can have significant instrumental or extrinsic value. What is much more controversial – indeed, might strike some as paradoxical – is the idea that suffering can be good in itself or have intrinsic value. It is this controversial view that I aim to defend in this chapter. I want to argue against what common sense and philosophical tradition (from Aristotle to Mill and beyond) would have us believe, and propose that suffering can be a particular kind of intrinsic good. Suffering, I want to say, can be a virtuous motive, and dispositions to suffer can constitute virtues of our physical and emotional systems. Or so, at least, I hope to convince you.

A virtue-theoretical perspective The claim that suffering can have value shouldn’t strike us as bizarre. For pretty much anything can have value, given the right kinds of circumstances. Nor, when we reflect for a moment, will the claim that certain forms of suffering are essential to the development and expression of virtue strike us as

Suffering as a virtue


implausible. Suffering is vital for the development and expression of virtues that constitute strength of character – a point that is prominent in Stoic and Christian writers, but also, perhaps most famously, in Nietzsche.2 Without suffering, the thought goes, we would be unable to cultivate and express courage, fortitude, resilience, and patience. And without these qualities of character, it is highly unlikely that we could live a good or flourishing life, given how central strength of character is to accomplishments and activities that constitute a good human life. Suffering also has instrumental value in the cultivation and exercise of virtues that are associated, not with strength and health, but with illness and disease, what we might call virtues of vulnerability.3 The idea that illness provides the opportunities for the cultivation and exercise of an important class of virtues – and thus has edificatory value – has been a recent theme of work by Havi Carel and Ian James Kidd, who examine and highlight suffering’s role in promoting creativity, humility, and intimacy.4 Moreover, insofar as illness and disease, dependency and infirmity are characteristic of human lives – in the sense that most of us grow old, and most of us thereby experience the deterioration of our physical and emotional faculties – it can be argued that suffering associated with disorder, malfunction, and infirmity in general is necessary if one is to possess and display the widest range of human virtues.5 In addition, suffering has clear instrumental value in the development and cultivation of moral virtues, and in particular the executive virtue of wisdom. This idea is central to the “virtue solution” to the problem of evil, but has a good deal of plausibility even if we set theodicy aside.6 Without suffering, there would be little need for compassion, good faith, sympathy, self-sacrifice, and the highest forms of love. Suffering is essential for moral development – which begins with fear of punishment, and essentially involves feelings of guilt, remorse, shame – and is positively correlated with those elements or components which constitute wisdom, such as understanding, humility and compassion (again), good decision-making and advice-giving, and reflectiveness.7 At the social level, moreover, suffering – in the form of judicial punishment and the imposition of hard treatment – is essential for social justice; for the strengthening of loving relationships; and for the communication of virtue to others, which is why suffering plays a central role in initiation rites and ceremonies. The willingness to suffer displays appropriate levels of courage and fortitude, but also commitment and loyalty to some group – virtues which stand both the individual and the group in good stead. In all of these ways, therefore, suffering has significant instrumental value, in the cultivation and expression of traits that are vital to the flourishing of our lives, both individual and collective. Why think, therefore, that suffering can be good in itself, or can have intrinsic value? Why push the line that suffering can itself be virtuous, and not just important for the development and cultivation of other virtues? In my view there is a simple and compelling reason to accept this view: namely, that


Michael Brady

forms of suffering are, in the right circumstances, appropriate or fitting responses to various kinds of disvalue, where what makes them appropriate or fitting is that they enable us to deal with such disvalue in the best way possible. This suggests that forms of suffering can themselves be virtuous motives, at least given an account of virtue as that which enables us to deal appropriately or excellently with important objects and events. This is the kind of picture of virtue proposed by Martha Nussbaum, as a development of Aristotelian thought.8 Nussbaum notes that Aristotle’s methodology for enumerating the virtues is to first “isolate a sphere of human experience that figures in more or less any human life, and in which more or less any human being will have to make some choices rather than others, and act in some way rather than some other”.9 Nussbaum continues: Aristotle then asks: What is it to choose and respond well within that sphere? What is it, on the other hand, to choose defectively? The ‘thin account’ of each virtue is that it is whatever it is to be stably disposed to act appropriately in that sphere.10 Virtues, in other words, are stable dispositions that enable us to respond appropriately in important spheres of human experience. These experiences will include the class or category of things which can be viewed as the formal objects of suffering, both physical and emotional: such things as bodily damage, lack of food, lack of sleep, frustrated expectation, loss of loved ones, social disgrace, and moral wrongdoing. My thought is that there are ways of responding well and responding badly when faced with these things, and that suffering, in the right way and in the right conditions, is part of what it is to respond well. If so, then a plausible picture of what virtues are supports a virtue-theoretical perspective on suffering. The idea that suffering is a fitting response to important spheres of our existence, and that enables us to deal appropriately with them, constitutes an intuitive reason in favour of viewing suffering as virtuous. A more detailed account will emerge in the next section, where I’ll examine the nature of virtue more closely, and show how forms of suffering can constitute virtue’s motivational component. In the final section I’ll defend my view against two criticisms.

A neo-Aristotelian account of virtue In order to present a virtue-theoretical account, I should first say something about the nature of virtue. Heather Battaly writes: “Virtues are qualities that make one an excellent person.”11 These qualities have certain features. As Julia Annas writes, “A virtue is a lasting feature of a person, a tendency for the person to be a certain way … It is active: to have it is to be disposed to act in certain ways.”12 Such qualities are, moreover, positively related to valuable goals or ends. Robert Adams identifies virtue “with persisting

Suffering as a virtue



excellence in being for the good”. This is an intentional link. But another is causal. As John Greco and John Turri put it, “A virtue is a stable and successful disposition: an innate ability or an acquired habit, that allows one to reliably achieve some good.”14 Let us see these ideas in greater depth by focusing on one of the most influential accounts of the nature of virtue in the recent philosophical literature, one that is developed and expressed by Linda Zagzebski in her 1996 book Virtues of the Mind. Zagzebski develops her account by focusing on central and traditional ideas surrounding virtue. The first, which we have already seen, is that virtue is an excellence. The second is the idea that virtue “is a property that we attribute to the person in a deep and important sense”.15 Virtues, on this view, are stable traits or qualities or characteristics of people such as “courage, generosity, compassion, justice, honesty, wisdom, temperance, and selfrespect”.16 Moreover, such traits are not “natural” but are instead “acquired”. She writes: “virtue … is an excellence, but not every excellence is a virtue”. Rather, “a virtue is an acquired human excellence”, and “the way it is acquired” is something “intrinsic to the nature of virtue”.17 For this reason, virtues are things for which we are responsible and praiseworthy, while vices are blameworthy. Two features are particularly important in Zagzebski’s account of virtue, for my purposes. The first is that virtue has a motivational aspect. Zagzebski thinks that virtues are partly constituted by feelings. This falls short of the claim that virtues are identical with feelings; here Zagzebski agrees with Aristotle about the need to keep the two distinct. However, although virtues are not identical with feelings, “almost every writer on the moral virtues has connected them with feelings”.18 On Zagzebski’s account, we can understand feelings as a kind of motive: A motive is a force acting within us to initiate and direct action … On one extreme, there are motives that are regarded as almost completely physiological, such as hunger, thirst, and fatigue. These motives are feelings, although they are not emotions. On the other extreme, there is the alleged pure motive of duty, which may not have a “feel” to it at all. The motives in the large area in between are emotions of various sorts.19 Zagzebski continues and describes motives in more detail, writing: A “motive” in the sense relevant to an inquiry into virtue is an emotion or feeling that initiates and directs action towards an end. Motives are connected with virtues in that virtuous persons tend to have certain emotions that then lead them to want to change the world or themselves in certain ways.20 So the compassionate person is someone who is motivated by feelings of compassion to help others, the honest person is motivated by a love of the


Michael Brady

truth, the courageous person is motivated by feelings of courage to stand firm in the face of danger, and so on. Each virtue, for Zagzebski, “involves a disposition to have the characteristic emotions that direct action in a particular direction”.21 The second important element in Zagzebski’s account of virtue is a “success” component: on her account, a virtue must reliably lead the subject to achieve the relevant end. As she puts it: Virtue possession requires reliable success in attaining the ends of the motivational component of the virtue. This means that the agent must be reasonably successful in the skills and cognitive activities associated with the application of the virtue in her circumstances. A person of virtue, among other things, understands some aspect of the world very well. A courageous person is good at understanding how to evaluate the level of danger in a situation, understands the consequences of various courses of action, and knows which dangers are worth facing in a certain manner and which are not. … Virtue, in short, involves knowledge and understanding of the world in this applicable area, both in general and in the particular case.22 So reliable success would seem to require that the subject have the relevant epistemic values – of knowledge and understanding – as well as the motivation needed to bring about the values at which the virtue aims. This is what virtues are, on a traditional neo-Aristotelian understanding. But give this picture, a good case can be made that forms of suffering are essential constituents of virtue, and hence human excellences, insofar as they constitute the relevant feelings. Recall that Zagzebski includes forms of suffering (though not under that description) such as hunger, thirst, and fatigue among the class of feelings from which we can identify virtuous motives. Given this, there should be no reason in principle for us not to add feelings of physical pain and forms of mental and emotional suffering. If so, however, then different forms of suffering, as feelings of the right kind, constitute virtuous motives, and a disposition to have such feelings is a part of virtue itself. For, clearly, feelings of hunger and thirst and fatigue pain can help us to be reliably successful in achieving valuable goals or ends. In the case of hunger, this will be to eat nutritious food. In the case of thirst, to drink suitable liquids. In the case of fatigue, to sleep. In the case of pain, to avoid noxious stimuli and protect the body so as to facilitate repair. More obviously, given Zagzebski’s stress on emotions as motives, we can include forms of emotional suffering in this picture: feelings of guilt, shame, disgust, and fear can alert us to the presence of the relevant values, and motivate appropriate behaviour (apologies, atonement, distancing oneself from contamination and danger, respectively). As a result, the feelings associated with physical and emotional suffering would seem to be in the relevant category of motives, and can help us to reliably achieve important goals or ends.

Suffering as a virtue


The general picture of how forms of suffering can be intrinsically valuable that I want to propose is, therefore, as follows. Virtues are dispositions to respond appropriately in important spheres of human experience. Such dispositions are, in part, dispositions to have certain motives which, when combined with the right kind of understanding, enable the subject to be reliably successful in bringing about certain valuable goals or ends. Forms of suffering can be part of the class of virtuous motives, alongside feelings of compassion, benevolence, and the like. And since virtuous motives like compassion, benevolence, and the like are, we can assume, intrinsically valuable, then forms of suffering which constitute virtuous motives are intrinsically valuable as well.

Objections and replies There are doubtless many objections that might be raised against this account of the value of suffering. In this section I’ll consider, and respond to, the two that I think are most pressing. 1.


On Zagzebski’s account, virtues are character traits and acquired human excellences which have to be developed, rather than being natural or innate features of people. This means it is doubtful that the feelings of hunger, thirst, tiredness, and pain can constitute virtuous motives, since these feelings and motives are not acquired, not developed, and not part of our character or person in any deep sense. We do not, for instance, learn to feel hunger and tiredness by the process of “habituation”, by doing what the virtuous person does and as a result by feeling what the virtuous person feels. Instead, these feelings – like our visual and auditory experiences – occur naturally and unbidden in us. As a result, such feelings fail to reflect who we are in any deep sense. It is thus doubtful that these feelings can constitute virtuous motives on Zagzebski’s reading. Even if a prima facie case for viewing suffering as intrinsically valuable can be made along the lines above, the idea that suffering has value in this way is inconsistent with something that seems much more plausible, and supported by common sense and philosophical theorising. This is the thought with which this chapter began: that forms of suffering are all prima facie intrinsically bad. Any positive case that I’ve made for thinking that suffering can be virtuous would seem to be undermined, therefore, by this platitude about the value of suffering.

Let us take these objections in turn. The claim that many forms of physical suffering are not the kinds of feelings that can play a role in virtuous motivation depends upon our acceptance of a traditional understanding of the nature of virtue, as character traits that are deep, acquired, and things for which we can be held responsible. This criticism will not prove to be devastating, however, if we have reason to think that the traditional understanding of virtue needs overhauling and the


Michael Brady

category of virtue expanding. And that is precisely what many philosophers have come to conclude. A number of important thinkers have proposed that there can be natural or innate or hard-wired excellences that constitute virtues. This is because we can identify certain stable and persisting dispositions to believe and to act that are generally reliable in bringing about certain valuable ends – dispositions that thus share core characteristics with the traditional virtues, but are best understood as faculties rather than traits of character. Such a view of virtue has been most prominent in virtue epistemology, as a development of the reliabilist tradition of thinking about knowledge and justification, and has been proposed and defended by John Greco and Ernest Sosa, among others. On this approach, “faculty virtues” include things like “perception, intuition and memory”, and can be contrasted with the more traditional Aristotelian virtues – known as “trait virtues” – like “conscientiousness and openmindedness”.23 In virtue epistemology, the idea of faculty virtues is supported by the need for a virtue-epistemic account that can accommodate different kinds of knowledge. As Greco and Turri put it: [I]t is plausible that a complete epistemology must feature both facultyvirtues and trait-virtues. Faculty-virtues seem indispensable in accounting for knowledge of the past and the world around us. Trait-virtues seem equally indispensable in accounting for richer intellectual achievements such as understanding and wisdom, which may presuppose knowledge, but which may also exceed it.24 I think that we can understand physical suffering on virtue theoretical lines, by identifying forms of physical suffering with the motivational elements in faculty virtues. In short: forms of physical suffering constitute virtuous motives of important faculties or systems, and play essential motivationalepistemic roles in bringing about valuable goals. To see this possibility, let’s look more closely at what a faculty virtue is. Ernest Sosa identifies faculties as dispositions or abilities to attain certain goals or accomplishments.25 In a similar vein, John Greco writes that a faculty “is a power or ability or competence to achieve some result”.26 And: a subject S “has an ability to achieve result R in conditions C if and only if, across the range of possible worlds where S is in C, S achieves R in C with a high rate of success”.27 What kinds of things can we identify as faculties or abilities in this sense? Sosa notes that “the most general faculties traditionally recognized are … perception, introspection, and memory”.28 Greco’s examples are “sight, hearing, introspection, memory, deduction and induction”.29 A faculty is, then, an ability, capacity, power, or function. What of faculty virtue? For Sosa, a faculty will be virtuous in light of “its performance of powers”:30 it must demonstrate the ability to attain the set of accomplishments – i.e. to do the things it does – to a high degree. And what this means, for Sosa, is “that it outperform feasible competitors” in its “delivery” of the

Suffering as a virtue 31


relevant goals or ends that constitute the accomplishments. We thus define faculty in terms of the ability or power to accomplish things, and define faculty virtue in terms of a faculty that does better in bringing about these accomplishments than feasible competitors. On this account, faculties like sight count as a virtue, at least in the “broader sense of virtue … in which anything with a function – natural or artificial – does have virtues. The eye does, after all, have its virtues, and so does a knife. … And if we include grasping the truth about one’s environment among the proper ends of a human being, then the faculty of sight would seem in a broad sense a virtue in human beings”.32 Let us assume, for the moment, that faculty virtues are genuine forms of virtue. On my view, it makes good sense to expand our category of faculties to include those that govern various sorts of bodily harms and seek to serve various sorts of bodily needs. The idea of an expanded class of faculties is not implausible. In addition to the five traditional sensory faculties, there are others that make up the somatosensory system. These include faculties or systems for temperature (thermoreception), the position and movement of our body (proprioception), and balance (equilibrioception). Moreover, there seem to be systems that govern other forms of physical suffering: we have systems governing hunger (the neuroendocrine system), fatigue (the sleep-wake homeostatic system), and thirst (the renin-angiotensin system). If so, then there is the possibility that these faculties could also constitute faculty virtues, since they can all be understood in terms of powers or abilities to attain sets of accomplishments relative to circumstances, and can plausibly outdo feasible competitors in bringing about these accomplishments. Finally, we might try to identify a faculty or system that governs our painful experiences. Such an identification won’t be simple; many people think it is a mistake to identify the pain system with the nociceptive system, since there are good reasons to resist the idea that nociceptors are pain receptors. After all, pain is a complex experience, and seems to involve, and be generated by, input from sensory, cognitive, and emotional elements. Nevertheless, if we follow Greco and Sosa in thinking about faculties in terms of abilities or powers to achieve certain results or accomplishments, and can identify distinctive accomplishments associated with pain – such as threat detection, damage avoidance, and repair33 – then we might still speak of a pain system or faculty in terms of the ability or power to achieve these.34 Whatever story we ultimately tell about how we should understand such a system or faculty (or complex network of sub-systems and sub-faculties), the feeling of pain will be essential to its operation. This is because feelings of pain will be much more motivationally effective in achieving these goods or accomplishments than feasible competitors like judgements or beliefs about bodily danger and injury. As any smoker or drinker or glutton will attest, judgements or beliefs – indeed, knowledge – that one is damaging one’s body in various ways are very often motivationally ineffective in getting us to do what we ought to do. Not so with physical pain, experience of which prioritises and motivates avoidance and repair much more effectively.


Michael Brady

If so, then the pain “system” or “faculty” outperforms feasible competitors when it comes to attaining the relevant set of accomplishments, at least in the relevant circumstances or under normal conditions. It does so because it involves a disposition to generate painful experiences, and so this disposition constitutes a virtue of this system or faculty. Since, in other words, it is the feelings of pain that reliably alert the subject to the relevant damage, then these feelings constitute virtuous motives in these conditions. And what is true of pain is, plausibly, true of other forms of physical suffering: the feeling of extreme coldness reliably alerts us to temperatures that threaten bodily damage and motivates us to seek warmth, and thus facilitates the proper functioning of the thermoreceptive system; it is because we feel cold that we know that we should move and are motivated to seek warmth, and thus secure the goals at which the system aims. By the same token, the feeling of hunger alerts us to the need for nutrition and motivates us to seek food, and thus facilitates the proper functioning of the appetitive system; it is because we feel hungry that we are aware of this lack and are motivated to seek food, and thus secure the goals at which this system aims. If so, then dispositions to physically suffer can constitute virtues of the relevant faculties or systems. At this point Zagzebski and other traditional theorists might raise two further objections. The first is that virtue epistemologists in the reliabilist camp are simply mistaken to claim that a faculty such as sight is a virtue, even in a broad sense. This is because our visual system or faculty is to be identified as the thing that has a function, and hence the thing that can have virtues, rather than as a virtue itself. So the faculty of sight is no more a virtue than a knife is a virtue; instead, virtues are those features that enable our visual system, and such things as knives, to perform their function well. Now perhaps Sosa and other virtue reliabilists can claim that trait virtues are, properly speaking, things like good eyesight, good memory, et al., which do seem to be genuine excellences of a person, and so enable a person to perform various accomplishments (such as recognizing medium-sized objects in the near distance, or recalling recent events) reliably. But – and here is the second objection – such things are still too far away from our traditional understanding of virtue to be plausible. Even if we admit that virtues can be innate and not acquired, and can be reliably connected to the achievement of some good, still faculties like good eyesight and good memory don’t qualify as virtues because they lack the essential element of right feeling. If we think, as Zagzebski does, that virtues are partly constituted by motives, then we might maintain that faculty virtues such as good eyesight and good memory are simply not virtues, although they are – like good looks and a ready wit – ways in which someone can be excellent and admirable. Whatever the merits of this attack on the kind of faculty virtue identified by Sosa and Greco, it is clear that it does not apply to the kind of faculty virtues in which forms of physical suffering play an essential role. This is because the feelings of pain and other forms of physical suffering are motives, and indeed this is what enables the relevant systems or faculties to perform their vital

Suffering as a virtue


functions. If so, then the traditionalists will need a different reason to rule out the possibility that there are genuine faculty virtues governing pain, thermoreception, fatigue, hunger, et al. The dialectical situation thus favours extending the range of virtues to include faculty virtues of this kind. For my opponents will have to argue, first, that the kinds of virtues identified by virtue reliabilists such as Sosa and Greco are not virtues. Even if they can come up with a convincing argument here, grounded in the idea that such things as good eyesight and good memory lack a distinctive motivational component, such an argument will clearly not apply to the kinds of virtues which have forms of physical suffering as motives. As a result, the case for including forms of physical suffering in the class of virtuous motives is starting to look rather strong. Even if I have convinced you that dispositions to suffer physically can constitute faculty virtues, and forms of physical suffering can constitute virtuous motives, the second objection still looms large. For the idea that physical or emotional suffering can be a virtuous motive clashes with the common-sense and intuitively plausible account of the disvalue of suffering. For commonsense and philosophical theorising maintain that forms of suffering are all prima facie intrinsically bad. The idea that forms of suffering constitute virtuous motives thus suggests that such feelings are both intrinsically good and intrinsically bad – on the assumption that virtuous motives are intrinsically good. Since it is impossible for such motives to be both intrinsically good and intrinsically bad, then we should reject the virtue-theoretical framework that encompasses suffering as a virtuous motive. The best way to respond to this criticism is to embrace what seems like an impossibility, and allow that suffering can indeed be intrinsically good and bad. In order to make this plausible, I’ll have to more carefully specify the bearers of the relevant value than I have to this point. So what I should say is that an experience that is intrinsically bad can nevertheless be part of a relation that is intrinsically good. Suffering, as an experience, can thus be intrinsically bad, but also be an essential part of a relation that is intrinsically good. Strictly speaking, therefore, what is good isn’t suffering in isolation, but suffering in relation to some fitting or appropriate object. This is all very abstract, but hopefully an illustration will help to clarify what I mean. As we have seen, common sense and philosophical theorizing maintains that certain forms of emotional suffering – such as remorse, shame, and disappointment – are intrinsically bad. Shame certainly seems a horrible thing to experience, and is to this extent something that we have prima facie reason to think is a bad thing. However, shame when one has done something shameful would seem to be a good thing. Here the claim about value isn’t about feelings of shame in isolation; instead, it’s a claim about the relation of such feelings to something else that is intrinsically bad, namely behaviour that is shameful. The thought is that an intrinsically bad mental state – such as shame – can also be intrinsically good when it is directed towards something else that is intrinsically bad – like shameful behaviour.


Michael Brady

The idea that negatively valenced attitudes can be positively valenced in relation to certain objects has been developed in depth by the virtue theorist Thomas Hurka, although he notes that this approach to value is apparent in the works of Aristotle, Brentano, Rashdall, Moore, Chisholm, and Nozick. Hurka’s account begins by distinguishing two senses of “intrinsic value”. On a strict view, “a state’s intrinsic goodness can depend only on its intrinsic properties, that is, properties it has independently of any relations to other states”.35 However, he notes: A looser view equates intrinsic goodness just with non-instrumental goodness, or with that portion of the overall goodness of the world that is located in or assignable to a particular state. It is the state’s own goodness, whatever its basis, rather than some other’s. Unlike the strict view, this looser one allows a state’s intrinsic goodness to be affected by its relational properties.36 If we adopt Hurka’s looser understanding of intrinsic goodness, we can allow that forms of suffering – which are in a strict sense intrinsically bad – can be intrinsically good when directed towards something that is itself intrinsically bad. To accept this is to accept what Hurka calls a “recursive characterization of good and evil”. This characterization involves a number of clauses. For our purposes, two are particularly important. The first is a recursion clause “about the intrinsic goodness of a certain attitude to what is good, namely, loving it, or, more specifically, loving for itself what is good, (LG): If x is intrinsically good, loving x (desiring, pursuing, or taking pleasure in x) for itself is also intrinsically good.”37 The second is a recursion clause about “hating” evil: “(HE) If x is intrinsically evil, hating x [desiring or pursuing x’s not obtaining or being pained by x’s obtaining] for itself is intrinsically good.”38 Hurka continues: “(HE) makes it intrinsically good for B to be sympathetically pained by A’s pain – to feel compassion for A’s pain – or to desire or try to relieve it. Even if B’s compassion has no further effects, it is something good in itself.”39 If Hurka is right, then it would seem that suffering of various kinds – both physical and emotional – can constitute forms of hating evil, in the sense described above, and as a result can constitute intrinsically valuable attitudes. As a result, my account can avoid the second criticism. This is because the negative feelings that constitute forms of suffering are in a strict sense intrinsically bad, but are, when directed towards things that are themselves intrinsically bad, in a looser sense intrinsically good. There is thus no contradiction in maintaining that suffering is both always intrinsically bad and, when it constitutes a virtuous motive, intrinsically good. Consider an example: remorse feels terrible, and is to this extent an intrinsically bad state for us to be in. But remorse is also a way of hating or being pained by one’s own moral wrongdoing, and so counts as a form of hating evil. On this account, therefore, remorse is intrinsically bad because of how it feels, but also intrinsically good because it is a form of hating evil.

Suffering as a virtue


There is a final objection that can be raised against the idea that suffering can be a virtuous motive. For it might seem that a world in which there is more virtue is intrinsically better than a world in which there is less. And this means that we should prefer a world in which bad things happen and people suffer as a result, to a world in which bad things don’t happen and people don’t suffer. But this is absurd. If so, then we have good reason to deny that suffering, when virtuous, is intrinsically good. This response is initially compelling. But it should, I think, be rejected. For we can acknowledge the intrinsic value of virtuous motives, and so acknowledge the fact that suffering can have this kind of value, while also maintaining that the value of the virtue is in most cases less than the value of the evil to which the virtue is a response. To accept this would be to accept an instance of what Hurka calls the “comparative principle, (CP), about attitudes and their objects: The degree of intrinsic goodness or evil of an attitude to x is always less than the degree of goodness or evil of x.”40 He continues: According to this comparative principle, the intrinsic goodness of loving a good is always less than that of the good, as is the intrinsic evil of hating it. … Similarly, the evil of loving an evil is always less than that of the evil, as is the goodness of hating the evil.41 As Hurka notes, this kind of move was in fact made by G. E. Moore in response to similar worries. Moore writes: There seems no reason to think that where the object [of an attitude] is a thing evil in itself, which actually exists, the total state of things is ever positively good on the whole. The appropriate attitude towards a really existing evil … may be a great positive good on the whole. But there seems no reason to doubt that, where the evil is real, the amount of this real evil is always sufficient to reduce the total sum of value to a negative quantity.42 If this is right, then the claim that suffering is intrinsically valuable when and because it constitutes a virtuous motive does not imply that we should prefer a world in which bad things happen and people suffer as a result, to a world in which bad things don’t happen and people don’t suffer. For it is plausible to suppose that the intrinsic badness of an existing evil always, or least typically, outweighs the intrinsic goodness of the appropriate attitude towards it. So even if all people responded to evil in the world with the appropriate negative attitudes – even if, that is, everyone hated what was evil whenever they encountered it – this would still be a world in which the intrinsic badness of the evil is greater than the intrinsic goodness of the virtuous attitudes towards it. From the standpoint of intrinsic value, therefore, the virtuousness of suffering is not sufficient to justify the presence of evil to which it is a response: the intrinsic goodness of pain does not outweigh the disvalue of


Michael Brady

bodily damage; the intrinsic goodness of remorse does not make up for the disvalue of moral wrongs. A final point needs to be stressed here. For virtues are not simply intrinsically valuable motives. They are, in addition, of great instrumental value, since they facilitate the appropriate behavioural response in important spheres of human experience, and in so doing enable us to secure a range of vitally important goods. Once we factor the instrumental goodness of suffering into our equations, it is much less obvious that the goodness of suffering is always swamped by the badness of that to which suffering is a response. Conversely, a case can be made that we are significantly better off all-things-considered as a result of suffering, given the vital role that suffering plays instrumentally. For instance, a life without the emotional suffering of remorse would be a life in which we are unreliable at knowing when we acted wrongly, and – even if we do know this – unmoved to apologize and make appropriate reparations. It would be a life in which our moral relations fracture and our reputation is sullied, a life in which we are mistrusted and shunned and unloved by others. In this sense emotional suffering, in the form of remorse, is necessary if we are to remain respected and trusted enough to function properly as a social being. Particular instances of remorse might well be intrinsically bad overall. But we are significantly better off as a result of this kind of negative emotional experience.

Conclusion In this chapter I have argued for a provocative claim: that forms of suffering can, in the right circumstances, be intrinsically valuable. This is because forms of suffering can constitute virtuous motives, or the motivational components of faculty virtues. Since such motives are appropriate or fitting responses to a range of disvalues – they are forms of hating what is bad or evil – then these motives are, on a plausible understanding of intrinsic value, themselves intrinsically good. So suffering isn’t just instrumentally or intrinsically valuable, therefore. Forms of suffering can be good in themselves.

Notes 1 This view is developed in detail in the first two chapters of my book Suffering and Virtue (Brady 2018). The present chapter addresses issues about the intrinsic value of suffering that I discuss in chapter 3 of that book. 2 For detailed discussion, see Brady (2018, chapter 4). 3 Ibid. 4 See Carel (2013) and Kidd (2012). 5 Illness might not by itself be necessary for cultivating the virtues of vulnerability – although it certainly can help with this. But some form of deterioration of arguably necessary if we are to develop these virtues. Illness is thus one of a larger class of conditions that facilitate creativity, humility and intimacy. 6 The “virtue solution” is a common theme in religious thinking about ethics. The basic thought is that suffering is necessary for the development of a range of moral

Suffering as a virtue


8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

35 36 37 38 39 40

41 42


and other virtues, and that a world with such virtues is better than a world without. This is meant to provide good enough reason for God to create a world with suffering, and so answer the question of why there is evil in a world created by a loving God. For a discussion of empirical work relating emotions and moral development, see Prinz (2007, ch. 1). There is a wealth of empirical work detailing the relations between suffering, adversity, and wisdom. See for instance Weststrate & Glück (2017, p. 800). In Nussbaum (1988). Ibid., p. 35. Ibid. Battaly (2015, p. 5. Annas (2011, p. 8. Adams (2006, p. 11. Greco & Turri (2015). Zagzebski (1996, p. 85). Ibid., p. 86. Ibid., pp. 105–6. Ibid., p. 128. Ibid., pp. 129–30. Ibid., p. 131, emphasis original. Ibid., p. 132. Ibid., p. 134. Greco & Turri (2015). Ibid. In “Knowledge and Intellectual Virtue” (1985), reprinted in Sosa (1991). Greco (1993, p. 520). Greco (2000, p. 212). Sosa (1991), p. 225. We might also include the faculty of reason, which is “a single faculty with sub-faculties … [of] intuitive reason and inferential reason.” Greco (1993, p. 520). Sosa (1991), p. 227. Ibid., p. 227. Ibid., p. 271. See Grahek (2007). Thanks to Jennifer Corns for pressing me – not for the first time – to be clearer on this issue. I fear that what I say here won’t suffice to address her worries about the legitimacy of talking about pain systems or faculties. But it’s the best I can do in this present chapter. Hurka (2001, p. 6). Ibid. Ibid., p. 13. Ibid., p.14. Ibid., p. 17. Ibid., p. 133. It’s actually to accept a weaker version of this, since there could be cases where the values of the attitudes might outweigh the value of their objects. I might take delight in a small evil, and this taking delight is worse than the evil itself. Or I might take great delight in a good deed, and the goodness of my delight is greater than the goodness of the deed. So it might very well be possible for a virtuous response to evil to outweigh the evil itself. Nevertheless, we can doubt that this is usually the case. Thanks to Jennifer Corns for pushing me to be clearer about this. Ibid., pp. 133–4. Moore (1903, p. 219). Hurka comments: “If the value of a virtuous attitude were greater than that of its object … the combination of pain and compassion for it


Michael Brady

would be on balance good. But this is intuitively unacceptable. The compassion is indeed good, and makes the situation better than if there were only pain and no compassion, but it cannot outweigh or justify the pain. If so – if a combination of pain and compassion for it is always on balance evil – the goodness of virtuously hating an evil must be less than the evil of its object” (ibid., p. 135).

Acknowledgements Many thanks to David Bain, Havi Carel, Ian James Kidd, Christian Miller, and especially Jennifer Corns for very helpful feedback on earlier versions of this chapter.

References Adams, R. (2006), A Theory of Virtue, Oxford: Oxford University Press. Annas, J. (2011), Intelligent Virtue, Oxford: Oxford University Press. Battaly, H. (2015), Virtue, Cambridge: Polity Press. Brady, M.S. (2018), Suffering and Virtue, Oxford: Oxford University Press. Carel, H. (2013), Illness (2nd edition), Durham: Acumen Publishing. Grahek, N. (2007), Feeling Pain and Being in Pain, Cambridge, MA: MIT Press. Greco, J. (1993), Virtues and Vices of Virtue Epistemology, Canadian Journal of Philosophy 23: 413–432. Greco, J. (2000), Putting Skeptics in Their Place, Cambridge: Cambridge University Press. Greco, J. & Turri, J. (2015), Virtue Epistemology, in The Stanford Encyclopedia of Philosophy, E. N. Zalta (ed.). Retrieved from istemology-virtue. Hurka, T. (2001), Virtue, Vice, and Value, New York: Oxford University Press. Kidd, I. (2012), Can Illness Be Edifying? Inquiry 55(5): 496–520. Moore, G. E. (1903), Principia Ethica, Cambridge: Cambridge University Press. Nussbaum, M. (1988), Non-relative virtues: an Aristotelian approach, Midwest Studies in Philosophy 13, 32–53. Prinz, J. (2007), The Emotional Construction of Morals, New York: Oxford University Press. Sosa, E. (1991), Knowledge in Perspective: Selected Essays in Epistemology, Cambridge: Cambridge University Press. Weststrate, N. & Glück, J. (2017), Hard-earned wisdom: explanatory processing of difficult life experience is positively associated with wisdom, Developmental Psychology 53(4): 800–814. Zagzebski, L. (1996), Virtues of the Mind, Cambridge: Cambridge University Press.


abasement 187 abjection 186–7 Abraham the patriarch 289 absent suffering 44–6 abstractions 27, 280–3, 286–7, 290, 293 Abu Ghraib 281 activism 176 acute pain 184–5, 258 Adams, Robert 298–9 adaptation, psychology of 198–9 additionalism 151–2 affect 56; at the social level 181 affective attitude 5, 91 affective construal 3, 25–6, 32–3; see also negative affective construal affective states 6, 103–6, 115–16, 181–5, 190; rationality or irrationality of 232 affliction 284–5 affordances 3, 25–8 agency, human 166, 169 agential mental processes 3, 38, 46–52 akrasia 259–61 aliefs 262–3 alienated desires 151 alleviation of suffering see relief anaesthetics 70 anger 183, 185, 190–1 anhedonia 32 animals: behaviour of 252; experiments on 206–7; moral status of 75; rationality of 279 Annas, Julia 298 appeasement 187–8 Aristotle and Aristotelian thought 282, 296, 298–9, 306 assimilation of mental states 49–50 asymbolia 45, 56, 60, 63–4, 70, 85, 90, 93 128, 251

attitudinal pain and suffering 2–3, 20, 24–5, 29–33, 85 attitudinal-representational theory (ART) 7, 123–30, 139–41, 150–3, 156 “attitudinal views” 125–7 Auschwitz 171 Averill, J.R. 190 aversion 3, 23–5, 32–3; to suffering 194 “avoidance behavior” (Bain) 147 Aydede, M. 253 bad motivational attitudes (BMAs) 145–59 badness 5, 86, 89–92, 128, 145, 148, 151–8, 253, 296, 301 Bain, David 8, 108–9, 127, 147–55; co-editor . Barrett, Linda Feldman 104 Battaly, Heather 298–9 beauty, appreciation of 212–23; value of 215; when recovering from mental suffering 215–21 Beauvoir, Simone de 171 beliefs as distinct from desires 156 Benatar, David 167, 175 Benbaji, Hagit 8 biopsychosocial model of challenge and threat 197 bipolar disorder 221–2 blameworthiness 260–1 Block, N. 79, 132, 134 bodily condition 126–8, 145, 151–2, 157–9 bodily feelings 3–6, 31–2, 104–17 bonding 203–4 boredom 22, 32 “boundary situations” in a person’s life 170



Brady, Michael 2, 19–22, 30, 33, 173–4, 231, 241; author of Chapter 15 and co-editor breathlessness 31 Brentano, F. 80, 306 Buddhism 32, 44, 170 Bursack, C.B. 45 Byrne, A. 127 Carel, Havi 31, 297 Carson, Rachel 215 catastrophic failure 188 categorical imperative 283 central neuralgias 259 challenge 197–8, 284 Chang, Ruth 172 changing the world 148–59; see also under world Christianity 75, 286–93, 297 chronic pain 257–8, 269 Cidam, A. 187, 189 Clark, A. 57 cognate accusatives 76–7 cognitive capacities of animals 279 cognitivism 128 Cohen, J. 44, 132, 135 coldness, feeling of 304 Coleman, Sam 4, 41 color: analogies with pain 280; experiences of 59, 132–5, 138–9 color qualia 62–3 comparative principle (CP) (Hurka) 307 compassion 5, 75, 83–4 con-attitudes 147–8, 153 conative attitudes 7, 124, 127, 130, 139 conditioned taste aversion (CTA) 136–7 conflicting reasons, argument from 146–53; applications of 150–2 consciousness 56–8, 60, 64–5; and suffering 69–71; and the unconscious mental life 41–2 constitution relation 65 construal of a situation 3, 25–32; in abstract terms 27; affective or practical 25, 29, 32; see also affective construal containment thesis 6, 104, 106, 112, 115 control, sense of 24, 38 “Coolman” 155 cooperative behaviour 203–4 “core affect” approach 6, 104–6, 110, 112, 116 Corns, J. 44, 49; author of Chapter 12 and co-editor cortisol 197–8

creativity 204–5 Crisp, R. 65–6 crucifixion 291 “Darwinian” resources 280 death, thinking of 39 decisiveness 264 demanding desideratum (DD) 67 dementia 173 deontological principles 282–3, 293 depression 21, 28, 32, 211–14, 220–1, 258 Descartes, René 71 desire-like attitudes (DEp) 7–8, 123–5, 139–40 desire theory 4, 60, 66–7 desires: characteristics of 128; as distinct from beliefs 156 direct and indirect suffering 55–6, 68–9 disability 279 disconnect between rationality and suffering 253–69; putative explanations for 259; taxonomic consequences of 269 disjunctive conception of desire 157–8 disruption model of suffering 3, 38–41; benefits of 42–6, 48–51 distorted reasoning and judgement 282–6 distraction 60, 67 distress caused by pain 44 Divine agency 286, 291; see also God; moral agency division of labor 293 Döring, Sabine 244 Duns Scotus, John 287 dysphoria 180–5, 188 ecumenical methodology 232–8, 248 edifying experieces and edificatory value 168, 176–7, 297 Ehrenreich, Barbara 165 elevator music 269 Elgin, C. 247 emotional suffering 300, 305; need for 308 emotions 6, 26–8, 110–11, 181–6; negative 296; rationality of 230–2; recalcitrant 232; theory of 28 empathy and empathetic engagement 280–2, 286–8, 291–3 Epictetus 44, 171 epistemology 302 equilibrium, mental 4, 40, 46, 49–50 ethical principles 283 euphoria 182

Index euthanasia 285 Euthyphro dilemma 86, 90, 147 evaluative conditioning (EC) 137 evaluativism 7, 123–32, 137–9, 147–55 evolutionary theory 27 experience-directed motivators 266–7 experience of suffering 166–7, 305; benefits gained from 211–12 experiential representations 126, 128 experiential sense of suffering 20 explanatory priority thesis 6, 104, 106, 112–15 facts of life 168–70, 177 faculties and faculty virtues 302–5, 308 failure, constructive orientation to 188–9 false signals 259 falsidical experiences 7, 133–6, 139 fear 26–7 feeling good and feeling bad 180–6, 189–91 feelings 300 Feldman, F. 24, 78, 85 “fighting through” 264 first-order desire (FOD) 147–50 “flourishing” 28, 138 foundationalism 242 Frank, Arthur 165, 168 Franklin, J.C. 202 Fulkerson, M. 44, 253 functionalism 126, 180, 182 Garcia effect 137 Gauguin, Paul 172 Gausel, N. 188–9 Gendler, T.S. 262–3 genocide 283 Genugtuung 84–5 Germany 283 Gibson, J.J. 25 goal-directedness 24 God of traditional Christianity 286–93 Goethe, Johann Wolfgang von 176 Greco, John 299, 302–4 Greek tragedy 170 grief 3, 21–3, 28, 39–41, 46, 49–51, 231–2; acceptance of 49 Guantanamo Bay 281 guilty feelings 201–2 gustatory experiences 7, 123, 130–4, 140–1 Haidt, Jonathan 176 hallucinatory pain 94–5


“Hamlet” proposal 264–5 happiness 32 Hardcastle, V. 71 Hardy Alister see Religious Experience Research Centre Hartshorne, Charles 287, 290 hating evil (HE) 306–8 Haybron, Dan 32 hearing loss 30 Heathwood, Chris 23, 62 Heck, Richard Jr. 241 hedonic changes 136–7 hedonic dimension of suffering 6, 21 hedonic rationality 229–48; case for 234; development of theory 248; objections to 239–47; practical implications of 247–8 hedonic tone 232, 239; see also negative hedonic tone hedonism, ethical or motivational 180; see also motivational hedonism Heidegger, Martin 169 Helm, B. 240, 246 heterogeneity problem (HP) 4, 56, 61–3, 78, 85 Hick, J. 22–3, 33 higher-order desire theory (Brady) 33 higher-order properties 68–9 Hiroshima 281 Hitler, Adolf 289 holistic unpleasantness 3–4, 40–2, 45, 67–8 Holocaust, the 168, 283, 289–90 homeostasis, mental 4, 49 honoring of human beings 280 hot spas 198 Hufendiek, Rebekka 26 “human predicament” 167 Hume, David 158 hunger 85, 90, 304 Hurka, Thomas 306 Hyman, J. 82 hyper-rational creatures 268 hypnosis 5, 70 Ideal Observer, God as 289 identity fusion 205 identity-theories of mind 94 imaginative representation 280, 288 immunization see vaccination imperativism 125, 147–8: theories of pain 48 Incarnation 289–92 incorrigibility of pain 5, 93–5



“inner view” of pain 158–9 instantiation theory of suffering (ITS) 68–71 instrumental value of suffering 297, 308 intensity of an experience 194 intentionality and intentionalism 65–6, 78, 125–7, 147, 243–4 intrinsic goodness or badness 296, 301, 305–8 introspection 57–60, 63 intuitive cognition 288–91 Inzlicht, M. 185–6 Iraq 281 Ireton, Kimberlee Conway 217–23 Jacobson, H. 148, 150, 156–8 Jacobson, Hilla 7 James, William 212, 215–16, 219, 223 Jaspers, Karl 170 Jesus Christ 75, 289, 291 Job 281–4, 289–91 joy 195 judgement-sensitive attitudes 233 justice: concern for 184; of suffering 201–2 justifying reasons 8, 155 Kauppinen, Antti 2–3, 19 Kidd, James 297 King, Martin Luther Jr. 191 Klein, C. 37, 44, 48, 63, 69 Koffka, Kurt 25 Korsgaard, C. 154 Kriegel, U. 65–6 Kripke, S. 94–5 Kübler-Ross model of grief 39, 49 Lavin, Douglas 245 Lazarus, R.S. 27 Leach, C.W. 187–9 “learned helplessness” 207, 251–2 Legault, L. 185–6 leprosy 251 Levi, Primo 171, 173 Lewin, Kurt 25 life-cycle perspective 172 life experiences 292 lifetime adversity 206 “linkage problem”’ (Feldman) 78, 85 living wills 284 Loesser, J.D. 75 loneliness 23 loss, sense of 87, 90

love and loviing relationships 3, 40, 67, 297 loving good (LG) 306 McClelland, Tom 3–4, 67–9 MacIntyre, Alasdair 169 marathons 195–6, 207 masochism 196 Massin, Olivier 5, 66–9 meaningful activities 200, 207 meaning-making 284–6, 290–3; Divine and by insiders 290–2; paradox of 290–3 Medina, José 168 melancholy 213–16, 219–21 mental disruption 3–4, 37–52; not synonymous with suffering 40–1 mental pain and suffering 67, 77, 79, 159, 213 Merleau-Ponty, Maurice 170 messenger objection 146 Mill, John Stuart 296 Mind 221–2 Moore, G.E. 306–7 moral agency 279–84, 292 morality 182–5, 279–80, 297 “motivating reasons” 8, 153–6 motivation 23–4, 299–308; effectiveness of 303; virtuous 305–8 motivational hedonism 180–91; assumption of 182 motivational “sting” 280–90, 293 Nagasaki 281 naive realism 89–90 Native Americans 279 natural selection 278 nausea 56 “navigation” of suffering 50 negative affective construal 2–3, 20, 28–32 negative beliefs and thoughts 258–9 negative experiences 172–6, 195–7 negative hedonic tone 223–30, 235–48; examples of 235–8 neuropathy 257–8 Nietzsche, Friedrich 165, 168, 176, 184, 187, 190, 214, 297 nociceptors 303 Noordhof, P. 82 normative requirements 154 Norway 189 Nouwen, Henri 211–12 Nussbaum, Martha 170–1, 177, 298

Index objectivism 139 Oedipus 171 omnipotence, Divine 292 omniscience, Divine 287–8 omnisubjectivity, Divine 288–9, 292–3 opioids and opioid overshoot 199, 252 opponent process theory 198–9 out-groups 280 pain 37–8, 29–32; bodily location of 80–2, 86, 95, 158; cognitive aspects of 278; concepts of 95, 158–9; as distinct from suffering 3, 5, 20–3, 30–1, 42–5, 76–7, 82–5, 88, 90, 93, 95; experience of 79, 92, 95, 280, 284; and guilt 201; intensity of 37; motivational dimension of 201–3, 278, 280; numbing or avoidance of 200; paradox of 94; perception of 154–6; pleasant or unpleasant 7, 41, 108–12, 123–8, 278; reactions to 44, 88–91; social significance of 279; suffering account of 85–6, 91, 95; symbolic significance of 45, 279 “pain blindspots” (Bastian) 196 pain research 56 pain system 303–4 painfulness 4, 56–70, 92, 123–8, 145–8, 151–2; as a quale 57–61; and suffering 66–9; unconscious 64 “paininess” 4, 59, 64–6 painkillers 8, 145–7, 200 pantheism and panentheism 287 paracetamol 200 Park, Nansook 220–1 partial cognitive penetrability 44 patients’ suffering 75 Paul, A.L. 165–7, 171–2, 175 peak experiences 199–200 perceptual account of suffering 86–7, 90–1 perceptual experience 108–11, 114, 240–2 pervasiveness 2–3, 20, 22, 28–32 Peterson, Christopher 220–1 phantom pain 94–5, 155 phenomenology 4, 26, 43, 51, 57–8, 65–7, 108, 110, 124, 152 philosophy of mind 5, 75 physical pain and suffering 4, 20, 55, 77, 296, 302–5 Pitt, D. 57–8 pity 26 Plato 147


pleasure: attitudinal and sensory 78, 85; comparison with joy 195; experience of 198–200, 221; pathways to and from suffering 201–7; taken in other people’s misfortune 183–4 positive experiences 172–5 post-traumatic stress 49–50, 168, 195, 285 powers of attorney 284 “problem of evil” 289–93 proprioceptive experiences 109–12 psychc affirmation of one’s life 32 punishment 185, 202, 297 pyschogenic cases 155 qualia (qualitative character) 4, 56–60, 70 qualia view (QV) 60–3 Rachels, S. 64 Ratcliffe, Matthew 28 rational and irrational mental phenomena, distinction between 248 rational appraisal 166–7 rational deliberation 280–6, 290–3 rationality 256–8, 263–7; of beliefs 240; epistemic, hedonic and practical 229–30; see also hedonic rationality Rawls, J. 269 “reactive account” 76, 95 reason-conferring view of suffering 252–3 reasons for action, content-related or attitude-related 150 reasons for belief, epistemic or pragmatic 149–50 recovery from mental suffering 212–21 “recursive characterization of good and evil” (Hurka) 306 referred pain 94 reflexes, pain-induced 261 regular perceptual experiences (REp) 124 relationalism 135, 138–9 relief, psychology of 198–9 relief of suffering 21, 29, 32, 44, 50 Religious Experience Research Centre (RERC) 212, 217–20, 223 remorse 43, 306, 308 repentance 33 representationalism 126, 129–39, 230, 244–5 research on suffering 1–2 resilience 205–8 responding well 298, 301, 308 ressentiment 187, 190



retributive justice 75 Roberts, R.C. 27 Rocco, Marc 177 role-functionalist analysis of mental properties 69 role-playing 69 Roman Catholic bishops 281 Russell, James 104–5 Sartre, Jean-Paul 26, 181, 186 Scanlon, T.M. 232–3, 240 schadenfreude 181–5 Scheler, M. 83–4 Schiavo, Terri 70 Schilder, P. 93 scientific methods 284 scope challenges 129–30 scratching actions 262 second-order attitudes 158 “secondary pain effect” and “secondary unpleasantness” 40–1 secrets, keeping of 201–2 self-harm 202 self-transformation after failure 188 Seligman, Martin E.P. 207, 220–1 Sellars, W. 253 sensory pain (s-pain) 56–65, 85, 258 sensory suffering 3, 20–3, 29–32, 268 shame 181–90, 305; motivational implications of 189; socio-functional view of 187–8; value to the individual 188 shameful behaviour 305 shared experience 203–5 Sidgwick, H. 61 Siegel, Susanna 25 sight, faculty of 302 simple creatures 267–8 Singer, T. 185 “skilled” and “unskilled” suffering 4, 50 Slater, Lauren 218, 223 slight pain/suffering 91 Smith, Adam 83–4 social alienation 285 social justice 176, 297 social relations 181–2 social roles of individuals 280 social suffering 268 social support 206 solidarity: Divine 292–3; social 285–6, 293 solitary confinement 177 Solomon, Andrew 19, 28, 211 somatosensory system 303 Sosa, Ernest 302–4

South Africa 283 spicy food 85, 196–7, 204, 256 spirituality 212 states of suffering 263–7; sticky 259 Stengel, E. 93 stigma 186 Stoicism 44, 51, 297 “Strangelove” 155 strength of character 297 struggle in suffering 46 subjectively-frustrated desires 8, 24, 128, 130, 148, 156–7 subjectivity, human and Divine 288–9, 292–3 subtraction problem (Wittgenstein) 146 success in attaining desired ends 300–1 suffering: as an active struggle 38; as an attitude 76–8; benefits gained from 196–201, 207–8; causation of 69, 266; characteristics of 2, 20, 24–5; commonality between different experiences of 3, 43; definitions of 22, 173; direct and indirect 55–6, 68–9; as distinct from what is suffered 42, 68; diversity in experiences of 2, 21, 269; extent of 37, 42, 44; function or role of 4, 38, 48–50, 252–3, 267; getting people together and getting them going 203–5; of an infant or an animal 51; intensity, novelty and attentional focus of 174; motivational force of 268; need for 194, 197–201, 297, 301; primary or secondary 257–8; reduction or elimination of 265–6; as a salve 202–3; at a social level 297; sought or merited 20–1, 33, 49; sources of (as distinct from ways of) 21; use of the term 196–7; value and disvalue of 50, 222–3, 296–8, 305, 308 “suffering from” various conditions 21, 77 Svenaeus, Fredrik 31 Tattersall, I. 251 Teroni, Fabrice 6 theories of mind 94 “thick” and “thin” values 111–15 tinnitus 82–4 top-down cognitive effects 257 torture 75, 87, 91, 281–5 tracking theory of representation 30 trait virtues 302, 304 transformative experiences 165–78; epistemic or personal 172–3;

Index involuntary or non-voluntary 172, 176; positive or negative 172–3; and suffering 173–8 trauma and traumatization 170, 203, 206, 259, 285 tribal morality 279–86, 289 Turri, John 299, 302 “two-ego’s” theory 288–9 Tye, M. 81, 127 ubiquity of suffering 169 unconscious suffering 56–7, 71 uniformity assumption 129–30, 141 unity of all suffering 30 Universal Declaration of Human Rights 283 unpleasantness 2–3, 21–4, 30–3, 40–2, 45–52, 61, 68, 174, 296; rationality of 233–9; unintelligible 237–8; see also negative hedonic tone urges, mental 4, 47–52; resistance to 47–8 utilitarianism 75 vaccination 207, 255–6, 259–64, 268 valence 7, 110, 114, 129–32, 138–41, 87, 103–17, 172–3, 182; variability of 130, 132, 140 value feelings 92–3 value-plurality hypothesis (VPH) 132–4, 139–40 value-uniqueness hypothesis (VUH) 132–4, 139–40 Van Gogh, Vincent 217, 219, 221


Vance, Jona 242 veridical experiences 7, 128, 130, 133–6, 139–40 Vignoles, V.L. 188–9 virtue: acquisition of 299; different types of 302; moral 297; nature of 298–304; and suffering 298, 300, 307 virtue epistemology 302 virtue reliablists 304–5 virtuous motivation 305–8 voluntary element in suffering 4, 46–7 volunteering 203 vulnerability 279; virtues of 297 wanting, two senses of 23–4 warfare 281 Watzl, S. 48 Weil, Simone 165, 284 Whitehead, Alfred North 287 Whitehouse, Harvey 205 wisdom 297 Wittgenstein, Ludwig 146–8 the world: altered perceptions of 215–17; reasoning or feelings directed at 8, 22, 33, 145–6, 149–50; see also under changing the world wrong kinds of pain and suffering 285 yearning 24 yoga 55 Young Casey, C. 258 Zagzebski, Linda 288–9, 299–300, 304