On the Temporality of Emotions : An Essay on Grief, Anger, and Love 9780192591852, 9780198851165

Citation preview

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

On the Temporality of Emotions

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

On the Temporality of Emotions An Essay on Grief, Anger, and Love BERISLAV MARUŠIĆ

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Berislav Marušić 2022 The moral rights of the author have been asserted First Edition published in 2022 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2022943046 ISBN 978–0–19–885116–5 DOI: 10.1093/oso/9780198851165.001.0001 Printed and bound in the UK by TJ Books Limited Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

For Marko, Petra, and Niko

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

Contents Preface Acknowledgements

ix xiii

P ART I 1. The Puzzle of Accommodation

3

2. The Rationality of Emotion

27

3. In Defense of the Puzzle

61

4. Embodied Reason

93 PART II

5. Endless Love

127

Conclusion

163

References Index

169 179

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

Preface I feel vulnerable in writing this book. The book is an attempt to explain something that I find puzzling about emotional experience in general and, specifically, about the surprisingly rapid diminution of grief for my mother, who died unexpectedly in 2007 at the age of 55. It seems to me that the diminution of my grief was not a response to a change in my reasons for grief but that, nonetheless, the diminution was reasonable. In general, it seems to me that emotions can change in ways that do not respond to a change in our reasons for them but that such changes can, nonetheless, be reasonable. Yet how could this be?—This, in a nutshell, is the puzzle I aim to address in what follows. It is a puzzle about how to make sense of resilience—how to comprehend and to anticipate it in our self-consciousness. But I worry: Is my experience of grief universal enough, so that this book is not just about an idiosyncrasy of mine? Even if, as Stanley Cavell has said, philosophy is always autobiography, it surely isn’t just autobiography.¹ A related, but distinct and much bigger worry is that, to echo Iris Murdoch, in philosophy no less than in moral life, “the enemy is the fat relentless ego” (1970, 51), that is, “the intrusion of fantasy, the assertion of self, the dimming of any reflection of the real world” (58).² Is my puzzlement over the diminution of grief a morally objectionable assertion of self? Still, the very fact that one describes one’s own experience does not make one’s philosophy idiosyncratic, or an objectionable assertion of

¹ In A Pitch of Philosophy (Cavell 1994). ² Richard Moran writes, “[T]his ego is fat because in its self-satisfaction it stores up and accumulates its impressions, prejudices, and habits of thought and rather than risk finding them inadequate compared to the actual multiplicity of life, instead spreads this accumulated lifematerial upon experience, upon others, until this accumulation of its own personal history and culture is all that can fall within its range of vision” (2012, 194).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

x



self. I take it to be characteristic of a particular kind of philosophy—call it phenomenology—that it proceeds through a description of a personal experience that others can see themselves in. When a philosopher says “I,” others think “I,” just as when Descartes’s meditator says, “I,” his readers think “I.” At least this is what I hope for in describing my puzzlement over my course of experience. Yet I worry further: Have I really understood my own experience? Self-knowledge is hard, and relationships with mothers especially so. I am by no means immune to self-deception. Also, emotions don’t arise on their own but in concert with others, so that it is hard to identify them as clearly as I claim to do in this book. Moreover, they are dynamic; they vary greatly with the time of day, the level of blood sugar and fatigue, and, most importantly, with social interaction and conversation. Have I merely told a fantasy story here, which does not even reflect my own reality, never mind something universal enough to be properly regarded as an object of philosophical reflection? I also feel vulnerable, because I take as a point of departure the phenomenology of a particular emotional experience. I then seek to understand this experience neither by providing a theory of the emotions nor by studying emotions in an empirical or scientific way but rather by increasingly abstract and theoretical reflection on this experience. I count on readers to share, or at least to understand this experience. Yet I have been struck by how strongly and viscerally some people reject my reflections. Here is how an anonymous reviewer responded to my attempt in an earlier paper—now published pretty much in the form that the reviewer rejected it—to formulate the main question of this book: The paper is on a complex topic. Its treatment of the topic is superficial and largely confused. The author does a lot of opining about various matters related to grief; there is little in the way of significant philosophical work. The opining sometimes takes the form of the author’s quoting passages from some famous writer and then expressing agreement. The thought the author attributes to the writer is then used repeatedly later in the paper as if the author had established it rather than merely expressed agreement with it. Perhaps the author thinks

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi



xi

anything said by a writer with a reputation for profundity must be true. The same goes for the author’s use of empirical research. The author quotes from an empirical study on the duration of grief. But the author did not actually look at this study to see whether it was well-designed to support its conclusions, or even what the criterion was that the researchers used to measure how long grief lasts.

This is harsh. But to be honest, there are days when I, too, think that my treatment of the emotions is superficial and largely confused. But not on all days. And on one of those other days, I want to defend myself. (On the days that I do think it, I have not done much writing.) To start: What is “significant philosophical work” on the emotions? Whatever emotions may be, they have something to do with felt experience. Any reflection or theorizing on the emotions, which does not in some way speak to felt experience, fails to make contact with its topic. To ensure that my work at least makes contact with its purported topic, then, I start with felt experience. But not just any experience will do: I start with something puzzling about felt experience. I strive to explain the puzzle and then I do what, I take it, a philosopher does—I theorize it. I theorize it in my own words and on my own terms, in my own social, historical, and philosophical circumstances. I do not deny that I find myself in such circumstances; I do not pretend to express universal, timeless truths about the emotions that any sentient creature might experience. However, I do not theorize my social, historical, and philosophical circumstances (or my theorizing about such circumstances, or further theorizing about such theorizing—as long as the present remarks don’t count.) I simply want to try to understand my own felt experience in my own terms, and I hope that in doing so, I can say something that will resonate with the experience of others—or at least something that will strike others as interestingly wrong, so that it can bring into relief something about their own felt experience. Is that not significant philosophical work on the emotions? Three methodological assumptions frame my reflections: First, I begin with the phenomenology of emotional experience; second, I theorize this experience; and, third, I do not theorize the context in which this experience arises. The first assumption is a condition for the possibility

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

xii



of making contact with my topic. The third assumption is what I regard as the liberating method of analytic philosophy (or what remains of analytic philosophy). The second assumption requires some further elaboration: In theorizing my emotional experience, I inevitably abstract from the particulars of my situation. I abstract from the particulars of the relationships in which this emotional experience arises, and I abstract from other emotions that proceed in concert with the experience that is the theoretical focus. As I like to think of it, I stylize my emotional experience, thereby overemphasizing some aspects and deemphasizing others. (For example, I do not say anything about the relief over my mother’s death that accompanied my grief—as long as the present remark doesn’t count.) Such interpretive work is, I think, inevitable: There is no unvarnished description of emotional experience, much less an unvarnished theorizing of it. However, the stylizing gives rise to the two worries I identified at the outset: The worry that I may not be speaking for others and the worry that I may not even know my own emotional experience. What I do, to assuage these worries, is to look to others, some of them famous, who have identified what I take to be similar moments in their experience. For the rest, I hope for attunement with the reader. To invoke Cavell one more time: The philosopher . . . turns to the reader not to convince him without proof but to get him to prove something, test something, against himself. He is saying: Look and find out whether you can see what I see, wish to say what I wish to say . . . [T]he implication is that philosophy, like art, is, and should be, powerless to prove its relevance; and that says something about the kind of relevance it wishes to have. All the philosopher, this kind of philosopher, can do is to express, as fully as he can, his world, and attract our undivided attention to our own. (1969, 95–6)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

Acknowledgements I have incurred debts to many people in the course of writing this book. I have had the privilege of presenting my work to numerous audiences, and of receiving helpful suggestions, questions, and objections from a great many people, including many outstanding philosophers. I am humbled by the generous responses I have received, and I remain very grateful for them. I would like to acknowledge a profound influence on this book by Richard Moran, both through his book Authority and Estrangement, as well as our many conversations. I have learned a great deal from Matt Boyle, who inspired me to think philosophically about grief in the first place, and from Pamela Hieronymi, whose account of the mind frames much of my thinking. I am also especially grateful to Agnes Callard, Matthew Chrisman, Sanja Dembić, Eli Hirsch, David Hunter, Doug Lavin, Barry Maguire, Jennifer S. Marušić, Oded Na’aman, Sebastian Rödl, Amélie Rorty, John Schwenkler, Maura Tumulty, Aarthy Vaidyanathan, and Steve White. I am privileged to have been able to share my work, and engage in conversation, with these thoughtful, perceptive, and sympathetic philosophers. I am grateful to several reviewers for Oxford University Press, for extremely helpful comments on an early draft of the manuscript. Thank you to Daniel Friedman for his diligent research assistance and to Grace Garland for compiling an excellent index, as well as some thoughtful suggestions just in time for publication. Some of the material from this book has been published in earlier form in two articles, as well as a book in Croatian: ‘Do Reasons Expire?—An Essay on Grief ’ (Philosophers’ Imprint, 2018), ‘Accommodation to Injustice’ (Oxford Studies in Metaethics, Vol. 15, 2020), and Beskrajna Ljubav (Institut za filozofiju, Zagreb, 2021). I acknowledge the editors and reviewers for these publications. I thank Oxford University Press for permission to include material from ‘Accommodation to Injustice’ in

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/7/2022, SPi

xiv  this book and for permission to include material from this book in Beskrajna Ljubav. I gratefully acknowledge a sabbatical from Brandeis University, with funding from the Theodore and Jane Norman Fund for several small grants and a research leave, as well as sabbatical funding from the Humboldt Foundation and from the Humboldt Professorship of James Conant at the University of Leipzig, with special thanks to Jim. I am grateful to the community at the University of Leipzig for welcoming me once again, with special thanks to Sebastian Rödl and Andrea Kern. Finally, I am very grateful to my editor, Peter Momtchiloff of Oxford University Press, for encouragement, patience, and support. I dedicate this book to my three kids—Marko, Petra, and Niko—with, obviously, endless love.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

PART I

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

1 The Puzzle of Accommodation My mother died on November 30, 2007—suddenly and unexpectedly at the age of 55. In light of her death, I immediately experienced intense grief. And this seems as it should be: My reason for grief was that my mother had died, not exactly young, but too young. Indeed, if I had not experienced such grief, something would have been wrong with me. Contrast me with Camus’ character Meursault in The Stranger who, a day after his mother’s funeral, goes to the movies with a new love interest (1942/1988). Yet now, many years later, I experience hardly any grief at all. This, too, seems as it should be. In “Mourning and Melancholia,” Freud puts it with apparent simplicity: “[A]lthough mourning involves grave departures from the normal attitude to life, it never occurs to us to regard it as a pathological condition and to refer it to medical treatment. We rely on its being overcome after a certain lapse of time” (1917/1999, 243–4). In a similar vein, DSM-5, the contemporary American standard for classifying mental illnesses, states: “The dysphoria in grief is likely to decrease in intensity over days to weeks and occurs in waves, the so-called pangs of grief” (American Psychiatric Association 2013, 126, n.1). Yet I find the diminution and eventual cessation of my grief deeply puzzling. This chapter aims to articulate the puzzle.

1.1 The Puzzle: Diminishing Grief Here is a first pass at my puzzlement over the diminution of my grief: Plausibly, I grieved for a reason—that my mother had died. Her death did not change over time. However, it was not wrong for me to grieve less over time. Yet how could the diminution of my grief not be wrong, if my reason for grief did not change? On the Temporality of Emotions: An Essay on Grief, Anger, and Love. Berislav Marušic,́ Oxford University Press. © Berislav Marušic ́ 2022. DOI: 10.1093/oso/9780198851165.003.0001

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

4

   

This is the puzzle in a nutshell. As it stands, it is oversimplified. Let me unpack it a bit in a second pass. My grief was my response to my mother’s death: I grieved in light of her death. My response can be contrasted with a mere reaction like a fever, which I don’t have in light of anything. This means that my grief was responsive to a reason— however exactly we are to understand the reasons-responsiveness of emotions. Indeed, it was responsive to what was in fact a good reason for grief: In light of my mother’s death, my grief was fitting. If I had not grieved, I would have been like Meursault. Moreover, as time passed and as I grieved, the fact to which I was responding—the fact that my mother had died—did not change. And, to now remove the oversimplification, its significance did not change either, or at least not very much: I did not stop loving my mother. Admittedly, she played a less central role in my life.¹ For example, I no longer called her on Saturdays, I no longer planned my holiday travels around seeing her, and I no longer spent any time in her company. But I did not become unconcerned with her, as I might have become unconcerned with someone with whom I no longer shared an emotional bond. Indeed, I want to say, my love for her did not diminish at all, even as her death receded into the past. The ongoing significance of her death thus suggests that I continued to have good reason to grieve.² Nonetheless, it was not wrong for my grief to diminish. If it had not, I would have suffered from what DSM-5 classifies as “persistent complex bereavement disorder” (American Psychological Association 2013). I would have been a statistical abnormality. In common terms, I would have been stuck. And it is not wrong not to be stuck. In this way, we arrive again at my question: How could the diminution of grief not be wrong, if my reason for grief persists? How can I make

¹ See Nussbaum (2001, 80) for a thoughtful discussion of this. I address Nussbaum’s view in sections 3.2 and 3.5. ² Schönherr (2021) distinguishes between a metaphysical and a psychological problem concerning the reasonable diminution of grief. The metaphysical problem is that grief reasonably diminishes, even though the past facts, which constitute reasons for the diminution of grief, don’t change. The psychological problem is that grief reasonably diminishes, even though the continuing evaluative facts don’t change in proportion to the diminution. Since it is the continuing significance of loss that I take to be crucial for the puzzle, I take my puzzle to be the psychological problem of fitting grief. I discuss Schönherr’s view in section 4.3.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

5

sense of the diminution of my grief as reasonable, since it seems to be incongruous with my reasons?³

1.2 Three Guises of the Puzzle: Surprise, Anticipation, Retrospection I want to acknowledge: the puzzle is elusive. But it also arises in different guises, depending on where, in the grieving process (if we can call it that—more on that in due course⁴), one finds oneself and what one knows about grief. Here I want to describe three different ways in which the puzzle might strike one, or three different guises of the puzzle. When my mother died and I initially felt intense grief, it seemed to me that my grief would remain a deep part of my life. I was convinced, naïvely, that my life would always be infused with pain over her death. However, it was not so. I recovered from grief very quickly. Moreover, I was surprised by how quickly this happened and how thorough the recovery was.⁵—I’d like to think that this does not reflect a peculiar callousness on my part but is a common phenomenon. Empirical studies suggest that we typically come to terms with others’ deaths, and many other misfortunes, surprisingly quickly.⁶ Here is how George Bonanno, a leading researcher on grief, puts it at the opening of his book, The Other Side of Sadness: The good news is that for most of us, grief is not overwhelming or unending. As frightening as the pain of loss can be, most of us are resilient. Some of us cope so effectively, in fact, we hardly seem to miss ³ Although there is a large literature on grief in which some elements of the puzzle of diminishing grief are discussed, I take the puzzle I describe to be novel. Nussbaum (2001, ch.1) and Moller (2007) have been especially influential in my thinking. It is Nussbaum who first uses the term “diminution” (2001, 79–85). See also D’Arms and Jacobson (2010), Moller (forthcoming), and Na’aman (2021) for related discussion. ⁴ See especially sections 3.4, 3.5, and 4.2. ⁵ Some form of surprise or disorientation seems characteristic of grief. Augustine, famously says that, in his grief, he became a great riddle to himself (Augustine 2018, IV.9). For perceptive discussion of disorientation in grief, especially in the writing of C.S. Lewis (1961), see Atkins (2022a). ⁶ See, for instance, Bonanno et al. (2005) and the extensive references in Bonanno (2009), as well as the discussion in Moller (2007).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

6

    a beat in our day-to-day lives. We may be shocked, even wounded, by a loss, but we still manage to regain our equilibrium and move on . . . [Bereavement] is something we are wired for, and it is certainly not meant to overwhelm us. Rather, our reactions to grief seem designed to help us accept and accommodate losses relatively quickly so that we can continue to live productive lives. (2009, 7–8)

For what it’s worth, Bonanno’s claims are true to my experience. However, this is something that surprised me. I was surprised that only a few weeks after my mother’s death, I could lead my life more or less as I had before her death: I hardly missed a beat! I was also surprised that my grief seemed to disappear pretty much completely—just as Bonanno describes: The fact is that most of the time, there is no hidden grief. There may be lingering questions about the relationship, or changes wrought by the death may have to be dealt with, but usually when grief has come and gone, that’s it. Even if the anguish was short-lived, most of the time all that means is that the person has managed her or his grief effectively and is moving on with life. (22)

This, too, strikes me as true to my experience. But this, too, surprised me: It surprised me that there wasn’t hidden grief—or at any rate much less than I had initially believed there would be. This, then, gives rise to the puzzle: In my initial experience of grief, I naïvely expected my grief to continue, because I thought of my mother’s death as my reason for grief, and it seemed to me that my grief would continue for as long as her death continued to be such a reason—that is, as long as she continued to matter to me. This is why I was surprised at the rapid diminution and the eventual end of grief: I stopped grieving, even though she did not stop mattering to me. I acknowledge that the formulation of the puzzle in terms of a thought about reasons may be idiosyncratic.⁷ I find it almost irresistible to reach for the notion of reasons to describe my experience. I hope that others ⁷ Hence the worries I expressed in the Preface.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

7

can recognize the description as true to their experience even if they wouldn’t employ the notion of reasons straight away. Yet perhaps the puzzle could be formulated without immediately reaching for the notion of a reason. Here is how: Robert Solomon and Kathleen Higgins have said that grief is the continuation of love.⁸ I think that this captures my experience of grief quite well: I took grief to be the new form that my love for my mother had taken. And I thought I would feel this love for as long as I would live—or at least for longer than a few weeks. That is why I did not think that my grief would diminish so quickly. But it did diminish, and eventually it ended, and I struggle to understand how that can be, since, I’d like to think, my love continues.⁹ How could the end of my grief not be a failure of love? One might respond that there are other ways for love to continue, besides grief.¹⁰ And this is surely true: One’s response to a beloved’s death will involve many acts and activities besides grief, including practices of mourning. Nonetheless, this does not resolve my puzzlement, because it does not explain why the diminution and end of grief would not be a failure of love—a failure that, perhaps, we can compensate with other acts of love. Even if there are other ways for love to continue, the diminution and end of grief still may strike us as a failure of love.¹¹ I was surprised by the temporality of my grief because I was ignorant about grief. However, I probably should have known better, since there was plenty of evidence about how other people experience loss. For the less naïve, my puzzle will therefore arise differently. It will arise in anticipation of the diminution of grief. Indeed, it is the anticipation of the diminution of grief that makes the puzzle most vivid. There is something disturbing and jarring in the ⁸ See Higgins (2013, 159) and Solomon (2004, 80, 90; cf. 2007, 74), where Solomon credits Janet McCracken for this thesis (though McCracken, if I understand her correctly, rejects it as overly romanticized (2005, 152)). Higgins has done much to articulate the thesis in published and ongoing work (2013; 2019). ⁹ To articulate the puzzle through the expectation that love will continue, we’d have to understand why we would have such an expectation. This, I think, will eventually lead us back to the notion of reasons. See Chapter 5. ¹⁰ In an elegant essay, Ryan and Erica Preston-Roedder argue that there are other ways in which one can “stand in solidarity” with the deceased (2017, 96). ¹¹ In my home country, one can hire professional mourners, usually women, who are extraordinarily adept at expressions of grief. Might one compensate for the diminution of one’s grief by keeping a professional mourner in one’s permanent employment?

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

8

   

anticipation of the diminution or end of grief. Proust puts it well in a famous passage from Within a Budding Grove: [I]t was by way of consolation that my mind was offering to my heart a promise of oblivion which succeeded only in sharpening the edge of its despair . . . Our dread of a future in which we must forego the sight of faces and the sound of voices which we love and from which today we derive our dearest joy, this dread, far from being dissipated, is intensified, if to the pain of such a privation we feel that there will be added what seems to us now in anticipation more painful still: not to feel it as a pain at all—to remain indifferent . . . (1919/1998, 340)¹²

When we grieve, the thought that we will stop grieving strikes us as the thought that we will become indifferent to the dead. Anticipating the diminution of grief is thus anticipating a failure of love. To see this, contrast the Proustian thought with cases in which anticipating a change in feeling strikes us as welcome or—as I’d like to put it—as a return to reason. For example, I might, while on holiday, experience a crush on someone, all the while knowing that this is due to the experience of freedom on holiday or the fantasy of an unknown relationship. In anticipating a rapid diminution of the crush upon my return home, I will therefore look forward to such a change in feeling as a return to reason; I will regard it as the passing of an enchantment that temporarily blinds me to who truly matters to me. Or, to give another example, after a transatlantic flight, I might experience a gloomy feeling of detachment from the place I live. However, I know that this is due to jet lag, and I anticipate the passage of the gloominess as a way of reconnecting with reality. Crucially, however, anticipation of the diminution of grief is not like that. Quite to the contrary, anticipation of the

¹² I am indebted to Richard Moran and Nicholas Riggle for pointing me to this passage. Moran discusses it in his “Swann’s Medical Philosophy” (in preparation). Moller (2007, 312) also discusses the passage. He accepts Proust’s point and argues that our resilience in the face of loss is to be understood as a form of blindness to the significance of loss. For critical discussion of Moller’s reading of Proust, see Smuts (2016).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

9

diminution of grief strikes me not as a return to reason but as a failure to hold on to what I truly value.¹³ There are two ways to develop the Proustian thought: an epistemological and an axiological way. Dan Moller (2007) interprets the Proustian thought as the thought that anticipating the diminution of grief is anticipating a form of blindness to what one values—a failure to properly understand or apprehend what matters. Aaron Smuts (2016) argues that Moller misunderstands the Proustian thought in taking its significance to be epistemological. Smuts considers the continuation of the passage I quoted: for then our old self would have changed, it would then be not merely the charm of our family, our mistress, our friends that had ceased to environ us, but our affection for them would have been so completely eradicated from our hearts, of which today it is so conspicuous an element, that we should be able to enjoy a life apart from them, the very thought of which today makes us recoil in horror; so that it would be in a real sense the death of the self, a death followed, it is true, by resurrection, but in a different self, to the love of which the elements of the old self that are condemned to die cannot bring themselves to aspire. (340)

Smuts argues that the Proustian thought is not that one will no longer recognize what matters but that it will no longer matter. The worry is that one’s values will have changed so much that one will have become a different person. Understood in this way, the worry is axiological.

¹³ Michael Cholbi has raised the question of what is good about grief, given that grief tends to be so painful. He argues that it is grief ’s instrumental value for obtaining self-knowledge. Cholbi writes, “The pain of grief, I propose, is genuinely painful but one to which we can be attracted through a recognition that by fully engaging with that pain and with the cavalcade of other emotions we feel in the course of grief, we can enhance our knowledge of ourselves and our practical identities. Perhaps surprisingly, although grief is prompted by the deaths of others, it can put us in a position to relate to ourselves in more profound ways” (2017a, 102). In a similar vein, Sonja Rinofner-Kreidl argues that “in grieving we realize what matters in our lives” (2018, 95), and Michael Brady holds that emotions, in general, “reliably alert us to things that are relevant to what we in fact care about” (2013, 153). Though I agree that grief affords a unique opportunity to acquire self-knowledge, I do not see this as its main value. My answer to Cholbi’s question is that we value grief because it is a way of holding on to someone we love.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

10     Bracketing the question of how to correctly interpret Proust, how should we understand what is problematic in the anticipation of the diminution of grief?—It seems to me that both ways of understanding the worry can be taken seriously. They are not in competition with each other but suggest that there are two ways in which the rapid diminution of grief may seem problematic: It could be evidence of both blindness and callousness. Indeed, it seems to me that one would lead to the other. Both are failures of love. I hasten to add a clarification: Of course, we do not anticipate every change in feeling, even every reasonable feeling, with horror. For example, I once ruined my favorite shirt. However, I realized that, very soon, I would have a new favorite, and, in anticipating this change, I did not despair over becoming a different person. In my experience, there was no Proustian moment. This is because I did not identify with my liking of my favorite shirt. Plausibly, thus, the Proustian insight holds only of losses and our responses to them which, in an elusive sense, we identify with.¹⁴—In what follows, I propose to set aside cases like the ruined shirt: My puzzle arises in cases in which there is a Proustian moment. To sum up, the anticipation of the diminution of grief strikes me as problematic when we anticipate that our grief will diminish and yet it matters to us that our loss continue to matter. Such anticipation strikes me as problematic, even if we anticipate that our grief will not entirely conclude but merely lessen to a point at which it does not reflect the extent that our loss matters to us—or perhaps more precisely the extent to which it matters to us now that our loss should matter to us later. Let me now turn to a final way of bringing out the puzzle: the diminution of grief in retrospect. It seems to me that the diminution of

¹⁴ In an illuminating discussion, Smuts (2016) develops this line of thought in terms of Williams’s (1973a) notion of categorical desires. Categorical desires typically concern one’s ground projects—projects one identifies with. However, I think that the relevant sense of identification is not confined to categorical desires: One could have a Proustian moment in anticipation of the diminution of grief for a friend who is not deeply involved in one’s ground projects. This is why I prefer to put the point in terms of the notion of identification rather than categorical desires. The notion is well-known from the work of Harry Frankfurt (1971) and the large literature that it has inspired.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

11

grief in retrospect is in a sense unintelligible. To bring this out, I need to return to my favorite notion—the notion of a reason. I realize that when my mother died, I had very good reason to grieve. I also acknowledge that today, more than a decade after her death, I am not wrong not to grieve. But I find it puzzling why this should be so— since it does not seem that her death has stopped mattering, at least not to the extent that would be reflected in the almost complete cessation of my grief. The puzzle is that I struggle to identify the reasons in light of which the diminution of my grief would be intelligible. The main reason why my grief diminished so quickly seems to be that I was wired to move on. Intense grief is hard to bear and is a major disruption to life. If my initial grief had not diminished significantly and quickly, I would have gone to pieces. That is why it makes very good sense that I would be wired in a way that my grief would diminish. The trouble is that this is no reason for grief to diminish, since, at best, it is a reason of the wrong kind.¹⁵ I did not grieve because grieving was somehow good for me. I grieved because my mother had died. Considerations showing that it is good or bad for me to grieve are like considerations showing that it is good or bad to believe something: they may make sense of why someone believes something, but not in a way that renders the belief rationally intelligible to the believer. The goodness or badness of grief have something to do with me, the griever. But my grief was not about me; my grief was about my mother and the fact that she had died. That is why the goodness or badness of grief aren’t reasons that bear on whether to grieve—and, therefore, are reasons of the wrong kind. We can no more cease grieving because grieving is hard to bear than we can cease grieving because someone pays us a large sum to do so, or because someone threatens us with doom unless we do. The puzzle of diminishing grief arises because it is difficult to identify, in retrospect, the reasons in light of which it would make sense to grieve less. In an effort to explain why we grieve less, we inevitably seem to reach for the wrong kind of reasons. That is because the diminution of

¹⁵ See D’Arms and Jacobson (2000, 77) for a point in this vein about grief and Hieronymi (2005, 2013) for an account of the wrong kind of reasons that I find convincing. I discuss reasons of the wrong kind at length in section 2.4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

12     grief does not seem to be primarily a response to a change in the significance or value of the loss but is, rather, occasioned by the needs of the griever. Yet this makes it hard to understand the diminution of grief as a response to reasons—since the reasons for grief are not provided by the needs of the griever. To compare: If it were shown that we don’t believe something because it is bad for us to believe it, wouldn’t this reveal that we are, precisely, not responsive to our reasons—that we are unreasonable?

1.3 The Puzzle of Accommodation: The Example of Sandy Hook I want to acknowledge again: The puzzle is elusive. Relationships with mothers are complicated. (Mine certainly was!) And grief is often opaque. Some of it remains in the depths of the mind and resurfaces at odd moments, occasioned by a smell or a song or an interrupted sleep cycle. So, one might wonder: Is it really true that grief diminishes— since some of it persists? Isn’t it rather that grief changes, perhaps even grows?¹⁶ Indeed, isn’t this book evidence of the continuation of my grief and its growth? (It is not.) Alternatively, one might think that I have underestimated the change in significance that my mother’s death occasioned. Is it really true that my love for my mother continues more or less the same, despite the fact that she is dead? If it does, then our relationship nowadays is really low maintenance—and much more so than when she was alive! Could it be easier to love the dead than the living? These are difficult questions. Grief is complicated and idiosyncratic, at least in part because it depends on the history of an individual relationship.¹⁷ Rather than address these questions head-on, I want to defend the

¹⁶ As Kathleen Higgins suggested in personal correspondence. See also Higgins (2019, 325–6) for interesting reflections on how the relationship with the deceased may grow, following the “continuing bonds” view of bereavement (e.g. Klass, Silverman, and Nickman 1996). However, I am less sanguine about such growth than Higgins is. ¹⁷ See Rorty (1978) and Wollheim (1999), who emphasize the importance of the history of an emotion, though, I think, unduly at the expense of the emotion’s reasons-responsiveness.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.    

13

puzzle I sketched in terms of my experience of grief by locating it in a broader context of emotional experience. In this way, I hope to show that we can eschew some of the complications that arise from the particulars of my example and even from the particulars of the case of grief. We can see the puzzle as a general phenomenon about the temporality of emotions. To do this, I want to suggest (as Bonanno does) that the diminution of grief is only one aspect of our resilience and should be understood as part of the phenomenon that Piaget calls accommodation (1954).¹⁸ We accommodate ourselves to loss when we modify our cognitive structures so as to incorporate a new experience or piece of information. And accommodation is a general phenomenon: we accommodate ourselves not just to the deaths of our loved ones, but also to new places, new careers, new accomplishments, new relationships. Most importantly— and this will be a main concern in the later chapters—we accommodate ourselves to injustice. It is the change in our moral response to injustice that gives rise to the most pressing instance of the puzzle I have identified. In what follows, I will describe a particular example of this. I chose the example because it is the occasion on which the phenomenon of accommodation, and its anticipation, struck me in a particularly forceful way. (It is on this occasion that I applied the lesson I had learned from grieving for my mother.) But it is also a particularly important contemporary example. On December 14, 2012, a man shot twenty first-graders at Sandy Hook Elementary School in Newtown, Connecticut. Most of them were 6 years old, some already 7. He also shot six adults and eventually himself. These twenty children suffered a profound injustice. When I learned about the shooting at Sandy Hook, I had a very strong emotional response. I felt a combination of sadness, anger, and horror: I grieved for the murdered children, I was horrified by the fact that they

¹⁸ The provenance of my use of “accommodation” is thus different from Rae Langton’s in her John Locke Lectures entitled “Accommodating Injustice” (2015). She is concerned with the linguistic phenomenon of accommodation as identified by David Lewis (1979).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

14     were so young, I was angry at the shooter but also at a political environment that had allowed him to gain access to the weapons he had used, and I felt compassion for the parents. Perhaps, following Peter Strawson,¹⁹ I should describe my anger as indignation, since it concerns the injustice that befell someone else, rather than myself, but I don’t think this captures it. I was just plain angry—though the sadness prevailed. For the first two days, this emotional response pervaded my selfconsciousness and colored all of my experience. I listened to reports and stories about the kids on the radio, and I thought about them all the time. However, I knew—I had learned!—what would happen next. Very soon, in only a matter of days, the sadness, anger, and horror lifted, exactly as I anticipated. I stopped thinking of the shooting, the kids, and the parents, and I directed my attention elsewhere—to my own kids, my family, and my work (but also, I confess, to the stupid game, Candy Crush). In a word, I accommodated myself to what had happened. And I am sure that many people who were not personally involved in the events moved on in similar fashion. I don’t think that my experience illustrates my callousness, but it illustrates the resilience that many people have in the face of loss and injustice: they often respond to it very strongly, but then they move on—and they understand that this is how things go.²⁰ The puzzle I described concerning the diminution of my grief for my mother seems to me to arise similarly in the case of my emotional response to the murders of Sandy Hook—though there the diminution was significantly faster: I experienced sadness, anger, and horror for a reason—the reason being the injustice that the murdered children of Sandy Hook had suffered. This reason and its significance did not change ¹⁹ Strawson writes, “We are not now to discuss reactive attitudes which are essentially those of offended parties or beneficiaries. We are to discuss reactive attitudes which are essentially not those, or only incidentally are those, of offended parties or beneficiaries, but are nevertheless, I shall claim, kindred attitudes to those I have discussed. I put resentment in the centre of the previous discussion. I shall put moral indignation—or, more weakly, moral disapprobation—in the centre of this one” (1962/2008, 14–15). ²⁰ My experience may not illustrate my callousness, but it does illustrate my privilege: I can afford to accommodate myself in ways that those facing gun violence on a daily basis cannot. Accommodation can thus reflect and, indeed, contribute to the marginalization of the less privileged. There is much more that should be said about this than I do here. See section 4.2 for a bit more, where I indicate the need for an ethics of the emotions.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.    

15

at all over the couple of days in which I accommodated myself to what had happened. Yet nonetheless, it seems, it was not wrong for me to stop feeling sadness, anger, and horror in light of the events at Sandy Hook. At the very least, it was not wrong for me to feel much less sadness, anger, and horror than I felt initially. Yet how could the diminution of my emotional response not be wrong, if my reason for it did not change—if the injustice against the murdered children persists, unchanged, in its full significance? The diminution of my grief for my mother is one example of accommodation; the diminution of my emotional response to the injustice against the murdered children of Sandy Hook is another. If we look at these two examples together, I hope we can see that the phenomenon I take to be puzzling does not to depend on the idiosyncrasies of, for instance, my relationship with my mother. Rather, it is a general puzzle that stands in need of a general response: It is, I think, a puzzle about our resilience, something that we are wired for. Of course, this is not to say that there is no difference between the temporality of grief and the temporality of anger, nor that accommodation to loss proceeds in the same fashion as accommodation to injustice.²¹ Not at all! The crucial point is that in our emotional experience of grief and anger, there is such a thing as accommodation, which often gives rise to a diminution of the emotion that does not correspond to a diminution in the significance of its object. That is what I take to be puzzling when one seeks to understand it from the standpoint of one’s reasons. Let me add an important clarification. Both examples of accommodation I have presented are cases of a single episode of loss or injustice: my mother’s death and the murders at Sandy Hook. However, often loss and injustice consist in an ongoing series of episodes. For example, one’s loved one might suffer from gradually deteriorating health, so that one is confronted with her daily suffering over a prolonged period of time. Or, one might be the target of ongoing injustice, such as discrimination. Accommodation to such ongoing conditions may well be different than ²¹ For some phenomenologically perceptive discussion of the difference between the temporality of grief and the temporality of anger, see Rinofner-Kreidl (2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

16     accommodation to episodes of loss or injustice. When loss and injustice are ongoing, accommodation may be harder, because there is, to put it metaphorically, renewed fuel for grief and anger. However, even in those cases, there is, I think, a modicum of accommodation that remains puzzling. We get used to conditions of loss and injustice, even though the loss and injustice retain their significance. We learn to live with a new normal. Thus, even in cases of ongoing loss and injustice, we will face a version of the puzzle of accommodation. Let me conclude this discussion with a different, and rather theoretical way of motivating the puzzle. Peter Goldie says that in emotions, “intentionality and phenomenology are inextricably linked” (2002, 242).²² On the one hand, grief and anger are ways of apprehending loss and injustice. On the other hand, they have, at least some of the time, a distinctly felt quality. Insofar as they are ways of apprehending loss and injustice, they are like judgments, and the reasons for experiencing grief and anger just are the reasons for judging that one has incurred a loss or injustice with continuing significance. However, to conceive of grief and anger as nothing more than judgments, and of reasons for grief and anger as nothing other than reasons for judgment, does not seem to do justice to their felt quality, which is fleeting. Yet if emotions are conceived in terms of their felt quality alone, they are like pain or other sensations, which are not reasons-responsive. How, then, can we understand the diminution of grief and anger over time as reasonable—since reasons don’t seem to bear on the felt component of the emotions? The general problem behind the puzzle of accommodation is this: How are we to account for both component elements of grief and anger—their judgment-like quality and their felt quality—as being of a piece, that is, as elements of one single and unified mental state?²³

²² Goldie characterizes emotions as “thinking of with feeling” (2000, 19, 58). ²³ See here Jesse Prinz’s (2004, 18) instructive discussion of the Problem of Parts and the Problem of Plenty that any theory of the emotions must face. The Problem of Parts is to identify the essential parts of an emotion, and the Problem of Plenty is to explain how several essential parts make one coherent whole. My point here is a version of Prinz’s Problem of Plenty. For discussion of this issue in the context of grief, see Ratcliffe (2017).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.    

17

1.4 A Note on Methodology As noted in the Preface, not everyone who I presented the puzzle of accommodation to has been convinced that we are confronted with a genuine puzzle. Although I will offer some arguments in defence of the puzzle in subsequent chapters, I want to pause to acknowledge a methodological difficulty. My case for the puzzle importantly rests on two specific phenomenological observations. First, there is the phenomenological observation that my grief for my mother diminished fairly quickly, even though the significance of her death did not change correspondingly.²⁴ Second, there is the phenomenological observation, drawn from the example of Sandy Hook, that accommodation to injustice is of a piece with accommodation to loss and that, therefore, not only grief but other emotions—most importantly anger—diminish fairly quickly over time. But how am I to prove that these phenomenological observations are correct? For example, how am I to prove that I have correctly identified the scope of my grief—that my grief has really diminished rather than changed and even grown?²⁵ And what is grief anyway: a feeling (of pain),²⁶ a perception (of value perhaps),²⁷ physical arousal or an awareness thereof,²⁸ an occurrent judgment, a dispositional belief (about value perhaps),²⁹ a (narrative?) process,³⁰ or perhaps a syndrome of several of these components?³¹ Indeed, one might wonder, how can I put forward my phenomenological observations without an

²⁴ Higgins (2013) makes a compelling case against the disenfranchisement of grievers by social pressures to get over loss. (Compare also Srinivasan’s 2018 discussion of affective injustice, primarily in the case of anger.) My experience is not one of disenfranchisement by others, but of a surprising inability to carry on my grief. I don’t mean the present essay to be in any way a disenfranchisement of grievers. To the contrary, I have great admiration for those who grieve longer and deeper than I. See section 4.4. ²⁵ See Frijda et al. (1991) for an overview of the complexities of identifying the temporal scope of an emotion. ²⁶ James (1884), Damasio (1999). ²⁷ De Sousa (1987), Roberts (1988; 2003), Prinz (2004), Deonna (2006), Döring (2007), Elgin (2008). Johnston (2001) proposes to understand affect as a form of perception, and he follows Wollheim (1999) in taking affect to be a precondition for emotion. ²⁸ James (1884), Prinz (2004). ²⁹ Solomon (1976), Nussbaum (2001). ³⁰ Goldie (2012, ch.3), Na’aman (2021, 2020). ³¹ Scheffler (2010) and Wallace (2013) hold the view that valuing is a syndrome, and, plausibly, grieving is a way of valuing. However, see Prinz (2004, 18) for forceful criticism of the view that emotions have multiple components that are not further integrated.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

18     account of what grief is and—since I aim to speak about accommodation to injustice—also of what anger is? I do not have an account of grief and anger. Thus, I acknowledge a methodological difficulty in my approach: My argument rests essentially on certain phenomenological observations about my course of experience without a full-fledged theory of the emotions. And I cannot prove that these observations are correct—certainly not without an account of what grief and anger are. (But how would I prove its correctness, without, in turn, appealing to some phenomenological observations?)—Instead, I propose to enter the topic of emotions in a different way. Rather than develop yet another account of the emotions, as an addition to the vast array of options in the literature, I want to approach the topic through a specific aspect of the emotions—their temporality. I hope that this will provide a novel way of getting the emotions into view, and of articulating what understanding the emotions would require. In the course of my argument, certain features of the emotions—such as their reasonsresponsiveness and our distinctly first-personal understanding of them—will be prominent. But I will not offer an account of them: I simply hope that what I say here will shed light on the emotions in a perhaps less comprehensive but nonetheless illuminating way. Although my approach is distinctly phenomenological, it is also importantly limited: In particular, I do not propose to describe the full phenomenology of grief or the other emotions.³² Rather, I identify certain specific phenomenological observations and build on them. The reason I do so is that the phenomenology of grief will differ tremendously for different individuals and different losses—even to the point that there will be substantial disagreement over whether an emotional response to a specific loss constitutes grief. Furthermore, the phenomenology of anger, resentment, guilt, and other apprehensions of injustice is no less complicated and idiosyncratic, and I do not think I can come even close to doing justice to its complexities. Nonetheless, I am struck that there is a common element in these otherwise different ³² Some classic accounts of the phenomenology of grief are Lewis (1961), Cicero (2002, bk 3), Didion (2005), Oates (2011), and Augustine (2018, IV, 4–12; IX, 8–13). Some perceptive recent philosophical discussion of the phenomenology of grief includes Higgins (2013; 2019), Rinofner-Kreidl (2018), Ratcliffe (2020), and Atkins (2022a; 2022b).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.    

19

experiences—namely the phenomenon of accommodation. Thus, although I do not offer a full-fledged phenomenology of grief, I do think of phenomenology as my bedrock.

1.5 Accommodation as Mere Change One might still refuse to acknowledge the puzzle. One might insist: Why should we accept that grief and anger diminish over time? Isn’t that the foundational mistake—the sleight of hand that underlies the magic trick? Let me address this objection head-on. A straightforward way to resist the puzzle, even before it gets off the ground, is to say that accommodation should not be understood as the diminution of grief or anger but as a mere change in these emotions. If we can make sense of grief or anger as changed but persisting, we would not need to identify reasons in light of which they should diminish. Is it plausible to think that our emotional response to injustice persists, and that accommodation is merely a change? Several things may be taken to speak in favor of this view. First, it is plausible that emotions are not just feelings. Many emotions persist even when we don’t feel them—for example, our fear of death (Nussbaum 2001). On a plausible view, emotions are dispositions that persist and can manifest themselves in different feelings (and actions and desires) at different times (Wollheim 1999). For example, romantic love might manifest itself as an enamored feeling in the other’s presence, as a feeling of distress when the other suffers, as a feeling of anger when the other is insulted, as a feeling of jealousy when the other is with someone else, and as grief when the other dies. This would explain why accommodation might consist merely in a change rather than the diminution of emotion. (However, we would then still be left with a puzzle about feeling, if we took feeling to be reasons-responsive. I won’t pursue this line of thought.) A second thing that speaks in favor of the view that our emotional response to injustice persists is that we can experience it again at later times. For example, there are smells or songs that call forth—some would say “trigger,” as if I were a mechanism—grief for my mother. Also, when I focus my attention on my mother or on the Sandy Hook murders, I can

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

20     sometimes feel at least some of the sadness or anger I felt in the past. When I look at pictures, I usually can. And in some sense, my mother’s death remains with me all the time; it has become part of me. In light of these considerations, I think it is correct to think of accommodation as involving a change in emotional response. Indeed, to return to Piaget’s (1954, 350–7) use of the term, “accommodation” is to be contrasted with “assimilation”: When one comprehends an experience through existing cognitive structures, one assimilates the experience. When one adapts one’s cognitive structures to new experiences, one accommodates the experience. Thus accommodation, unlike assimilation, leaves us changed. This is an idea I mean to inherit from Piaget, though I don’t propose to delve further into his developmental psychology.³³ Nonetheless, I maintain, accommodation to loss and injustice involves not only change but also a diminution of the emotions to a quite significant degree—even if the emotion never goes away entirely. That is why I think that the puzzle persists. To see this, let us consider how we might anticipate the diminution of the feeling of anger. For example, recall a fight in which you felt angry because you thought you were wronged. How would you anticipate the end of the feeling of anger without anticipating a change in the underlying reasons—without an apology from the other or the recognition that your anger was misplaced? The thought that you won’t feel angry anymore will strike you as the thought that you will no longer be angry. If the person who wronged you said, “You won’t feel so bad about this tomorrow,” you’d be dismayed—even if you knew that the other was right. (Indeed, perhaps precisely because you knew it—because you knew that you wouldn’t be able to hold on to the anger!) Contrast this with a case in which you feel angry but realize that your feeling is excessive or misplaced: You realize that you had a bad day, or very little sleep the night before, but you can’t shake the angry feeling anyway. In this situation, the thought that your feeling of anger will pass ³³ Piaget uses the notions of accommodation and assimilation in his explanation of how young children acquire the concepts of self and of an objective world. The use of “accommodation” in the context of the emotions is thus much more general than Piaget’s original use. However, the term is standardly used in psychology (e.g. by Bonanno 2009) in the way I do here.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

21

would strike you as a success—a return to reason. This contrast suggests that the thought that your feeling of anger will pass, when you take your anger to be reasonable, will presumably not strike you as a success. I think that there is a straightforward explanation of this observation. A significant diminution in feeling implies a significant diminution in the emotion itself. If an emotion moves us less and less, then it is diminished. This is plausible if emotions are dispositions that are manifested in feeling and action. If we don’t feel very much and we don’t do very much, then (all else equal) the underlying disposition is weaker. If an emotion initially moves us all the time, in all aspects of life, but later moves us only very rarely, in very specific circumstances or only with an effort at focused attention, then it is plausible that it is diminished. This suggests that accommodation is not mere change. However, I think the suggestion that accommodation involves a change in emotions does contain two important insights: First, we do change. Accommodation is not assimilation. Second, the diminution of the emotions does not have to be complete for the puzzle to arise. Some residual grief or anger—indeed considerable amounts—may remain. Nonetheless, there is a puzzle as long as the diminution of the emotional response does not correspond to diminution in the significance of the loss or injustice to which it is a response. Thus, the puzzle of accommodation may arise even if our lives are profoundly changed and our pain remains deep. The puzzle remains so long as our grief and anger strike us as not deep enough in light of the loss or injustice that they are a response to.

1.6 Some Responses to the Puzzle There are many possible responses to the puzzle of accommodation. In this section, I will briefly describe some of them and I will then sketch the response I will go on to argue for in subsequent chapters. In this way, I will provide an overview of the central arguments in this book. First, there is the Hardline Response.³⁴ This response denies that accommodation is ever reasonable: Accommodation is yet another ³⁴ See section 3.1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

22     example of the many ways in which we fail to be rational. Grief in the face of loss, and anger in the face of injustice are rational for as long as the loss and injustice remain significant. Indeed, the only thing that determines the reasonableness of our emotional response is the significance of the loss or injustice—no matter when it occurred, how long we have grieved or been angry, or how much this costs us. Second, there is the Pragmatist Response.³⁵ This response explains the reasonableness of accommodation not in terms of the significance of the loss or injustice but in terms of the significance of grief or anger. Since grief and anger are taxing, accommodation is often reasonable even if the loss and injustice remain unchanged. Third, there is the Antirationalist Response.³⁶ This response denies that our emotional response is to be understood in terms of reasons. On this view, there really is no puzzle: our emotions simply unfold, often in predictable ways. Fourth, there are several versions of the Time-Sensitive Response.³⁷ This response maintains that accommodation is reasonable because the reasons for our emotional response are in some way or other timesensitive: they are either themselves time-sensitive or there is a timesensitive feature in virtue of which we have those reasons. For example, the reasons depend on our emotional distance to the loss or injustice, or they depend on the stage in a process of grief or anger that we find ourselves in, or the amount of psychological work we have completed. However exactly the details are to be spelled out, according to the TimeSensitive Response, accommodation can be reasonable. In contrast to all these responses I will ultimately defend a view on which the puzzle I described eludes a solution.³⁸ I will grant that accommodation can be reasonable but argue that, nonetheless, there remains something essentially incomprehensible in accommodation. Indeed, this incomprehensible element explains why one might feel surprise over accommodation, why one would be disturbed when anticipating accommodation, and why one can’t fully comprehend accommodation after it has occurred. ³⁵ See section 2.4. ³⁸ See Chapter 4.

³⁶ See sections 2.1–2.2.

³⁷ See sections 3.3–3.6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

23

Here is what I take to be the core obstacle to a fully satisfactory answer to the puzzle of accommodation: Our emotional response is directed outward—to loss and injustice. However, accommodation has to do with us—with our need to move on. That is why, even though accommodation is reasonable, we cannot comprehend its reasonableness in terms of reasons in light of which it would make sense to grieve less or be less angry. For, those reasons would have to show that the loss or injustice has lost its significance, which they do not—since they have to do with us rather than with the loss or injustice. Instead, the reasonableness of accommodation has to be understood differently: When we accommodate ourselves to loss or injustice, we understand that it is somehow all right for our emotional response to diminish; we just can’t say why. For example, in experiencing grief for my mother, my focus was on her, not on myself: In the first instance, I was pained over her death, not over what I had lost. In accommodating myself to her death, my grief diminished and eventually ended. However, this is not because my reasons changed; her death did not become insignificant. Nonetheless, it is somehow all right that I grieved less and eventually stopped grieving altogether—and I understand that it is somehow all right. It’s just that I can’t say why it’s all right. When I look for reasons, all I find are reasons of the wrong kind—reasons that have to do with me. Yet these are not reasons in light of which I can understand the diminution and end of my grief. I hasten to add a clarification: I have said that our emotional response is directed outward. However, this does not mean that our emotional response cannot have anything to do with us. Indeed, many losses are felt more dearly, and reasonably so, because we stand in a particular relation to the person who died. For example, I had reason to grieve for my mother, even though you did not. After all, she is my mother, not yours. Nonetheless, it is a mistake to take this to show that grief and other emotions are about us—that their focus is inward after all. This is clearest when we compare cases of partiality in emotions, such as my grief for my mother, with impartial emotional responses, such my sadness, anger, and horror over the murders at Sandy Hook. In both kinds of cases, the focus of the emotional responses is outward. But in cases that involve partiality, the relationship to the object of the emotional response affects the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

24     response’s reasonableness. This is something that stands in need of explanation: How can the relationship to the object of the emotional response affect the response’s reasonableness without compromising its outward-directedness? I will return to this question in due course. My argument, in a nutshell, will be that our relationship to the object of our emotional response is not a reason for our emotion but a background condition that our reasons depend on—and, moreover, a background condition that determines the significance of our reasons.³⁹ On the view I propose to defend, accommodation is reasonable, but its reasonableness is not comprehended in the comprehension of the reasons for our emotional response. Rather, we face irreconcilable double vision. The double vision arises from the fact that we cannot apprehend, at once, the object of our emotion together with empirical facts about the emotion, such as the fact (if it is one) that grief is a process. Nonetheless, in recognizing this double vision, we can come to see accommodation as reasonable, albeit without identifying reasons in light of which this would be so and, therefore, without resolving the double vision. This is why there remains an unreconciled moment in our emotional lives, and accommodation remains, in a salient sense, incomprehensible.

1.7 Outline Let me now outline how my argument will proceed. In Chapter 2, I will examine the rationality, or reasons-responsiveness, of grief and anger. In particular, I will explain why I take emotions to be reasons-responsive, and I will argue against the Antirationalist Response. I will consider the appeal to the wrong kind of reasons and argue against the Pragmatist Response. Finally, I will seek to clarify how to understand the objects of grief and anger. In Chapter 3, I will consider several possible responses to the puzzle of accommodation and argue that they are all inadequate. I will first argue against the Hardline Response, according to which time doesn’t matter at all. Second, I will consider whether acknowledgement of the shock and ³⁹ See section 2.6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. 

25

surprise that often accompanies our experience of loss and injustice can account for the temporality of grief and anger. I will then distinguish several versions of the Time-Sensitive Response, according to which reasons for emotions have a time-sensitive component. I will argue that they can’t adequately explain the anticipation of a change in our emotional response, nor can they make sense of this change in retrospect. This is because in their attempted explanations, they introduce a form of alienation, which leads us to lose sight of the object of our emotional response. In essence, if we are to understand our emotions in temporal terms, our attention is directed inward rather than outward—and this introduces an ineliminable form of double vision, since we cannot maintain the objects of our emotions in view at the same time. In Chapter 4, I will describe my preferred answer to the puzzle. I will argue that accommodation is reasonable, though we cannot identify reasons in light of which this would be so. Accommodation is a case of reasonableness without reasons. I argue for this by developing an account of pragmatic encroachment about reasons for emotion. Finally, in Chapter 5, I will present a contrasting case to grief and anger—the case of love. I will argue that the temporality of love is interestingly different from the temporality of grief and anger. In its self-consciousness, love is endless (much like my naïve experience of grief was). However, unlike in the case of grief and anger, this is something that is consistent with our empirical knowledge of love: Even if we know that love sometimes, all too often, ends, we can nonetheless be right to think of our love as endless, without any irrationality!

1.8 Conclusion Death and injustice are, in a salient sense, unacceptable.⁴⁰ Nonetheless, we realize that, in time, we will accept many deaths and much injustice, ⁴⁰ As the concluding words of her A Very Easy Death, Simone de Beauvoir writes, “All men must die: but for every man his death is an accident and, even if he knows it and consents to it, an unjustifiable violation” (1965, 106). This strikes me as true and important, even if one agrees with Williams (1973a) that eternal life is surprisingly unattractive.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

26     even if we don’t fully come to terms with them. We are, after all, resilient. Moreover, it is all right that this is so; it is all right that we don’t go to pieces. It is also a good thing that we are resilient, but that is not what makes it all right. Ultimately, I find something incomprehensible, and horrible, in its being all right, and I think that there are principled reasons why it is incomprehensible and horrible. In what follows, I will make my case for this.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

2 The Rationality of Emotion When we suffer a loss, we may experience grief in light of it. And when we suffer or witness an injustice, we may experience a combination of anger, resentment, indignation, sadness, or horror in light of it. The loss and injustice are our reasons for our grief and anger. However, we accommodate ourselves to loss and injustice: Our emotional response to loss and to injustice often diminishes fairly quickly, and sometimes even ceases altogether, despite the fact that the reasons for it don’t change correspondingly; the significance of the loss or injustice does not diminish in proportion, or even at all. Nonetheless such accommodation seems to be somehow all right: It seems not to be wrong for grief and anger to diminish. Indeed, it seems reasonable. This, I have argued, gives rise to a puzzle about accommodation. And the puzzle is quite general: it arises in similar fashion about other emotions, including guilt, delight, pride, and possibly regret. After all, we accommodate ourselves to the circumstances in which any such emotion is reasonable: We accommodate ourselves to the good and the bad. The aim of this chapter is to articulate and make plausible a crucial presupposition that the puzzle rests on: The view that emotions are in an important sense rational. They are rational in the sense that is opposed to arational, though, of course, they are often not rational in the sense that is opposed to irrational. Emotions are in the space of reasons: they are answerable to reasons, though we may not always have a good answer. I will mark this by saying that grief and anger are reasonsresponsive. In what follows, I will focus on grief and anger as my paradigms, but I hope that what I say about them carries over to other emotions (though love is different—as I explain in Chapter 5). I will proceed as follows: In section 2.1, I will clarify how to understand the claim that accommodation is (potentially) reasonable. In section 2.2, I will explain why I take emotions to be reasons-responsive. In On the Temporality of Emotions: An Essay on Grief, Anger, and Love. Berislav Marušic,́ Oxford University Press. © Berislav Marušic ́ 2022. DOI: 10.1093/oso/9780198851165.003.0002

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

28     section 2.3, I will respond to objections to this view. In section 2.4, I will explain why an appeal to considerations about the goodness or badness of grieving or being angry is an appeal to the wrong kind of reasons and, therefore, should be rejected. In section 2.5, I will consider whether considerations that don’t have to do with loss and injustice, such as the exigencies of life, can attenuate our reasons for grief and anger, and I will argue that this would, in effect, be a covert appeal to the wrong kind of reasons. Finally, in section 2.6, I will consider the objects of grief and anger and how our relationship to those objects bears on our reasons for grief and anger.

2.1 The Reasonableness of Accommodation I have formulated the puzzle of accommodation in terms of the claim that accommodation is, or seems, reasonable. I want to start by clarifying how to understand this claim. The clarification is necessary, because accommodation is not an emotion; rather, it is a change in emotion. Thus, there might seem to be a slide in my discussion from talking about the reasons for grief and anger to talking about whether there are reasons for a change in grief and anger. It would be like the slide from talking about reasons for belief to talking about reasons for changing one’s mind. Yet, plausibly, it is emotions, and, generally, states of mind, which are reasons-responsive and not changes in emotion, or changes in states of mind. Thus, to say that we have reason to change our mind is just to say that we have reason to adopt the state of mind we would have if we underwent the change. Here, then, is how we are to understand the puzzle of accommodation in light of this observation: The puzzle arises because we initially grieve or are angry in light of a loss or injustice, but we eventually come to be equanimous, or at least we approach equanimity, in a way that does not seem to be warranted by our reasons. The puzzle is how such equanimity can be reasonable if the reasons for grief and anger remain, at least to a large extent, unchanged. Therefore, the puzzle concerns accommodation not as a change in emotion but as an emotional state that one arrives at through undergoing the change.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.  -    

29

This clarification also helps bring out another important point. The puzzle of accommodation is not the puzzle of how we should respond to the fact that we are experiencing grief or anger—of whether we should seek to enable accommodation or to prevent it. The puzzle is not whether to take steps to maintain our emotional state or to extinguish it. This is, in effect, the question of how to manage our emotions—for example, whether to continually attend to the loss or injustice, so as to (seek to) prevent accommodation, or whether to distract ourselves (or take some pills!), so as to accommodate ourselves faster. This question may well be important in its own right, but it is not what gives rise to the puzzle. The puzzle is not what to do in light of loss or injustice over time but what to feel in light of it over time and what to think about the change in emotional experience. To confuse doing and feeling—to confuse will and emotion—is, as Anscombe evocatively put it, to confuse washing with soap (1978, 144).

2.2 The Reasons-Responsiveness of Grief and Anger I now turn to the fundamental assumption of the puzzle—that grief and anger are, in principle, responsive to reasons. In this section, I will clarify the assumption of reasons-responsiveness and defend its plausibility. A corollary of my argument will be that the Antirationalist Response to the puzzle (which denies that emotions are reasons-responsive) is mistaken. The assumption that emotions are reasons-responsive is plausible, because our emotions are not conditions that merely befall us, but they partly constitute our take on the world: In fear, we apprehend something as dangerous, in anger we apprehend something as unjust, and in grief, we apprehend something as a loss.¹ For instance, the apprehension that

¹ Robert Solomon writes “It is obvious that grief is suffering brought about by the recognition of a loss. Grief is not just a feeling or a Jamesian sensation caused by neurological or physiological changes in the body (although, to be sure, there always are such changes) but an engagement with the world, the recognition of a real loss in one’s world; and it consists, accordingly, of perspectives and ideas (that is, thoughts, perceptions, and memories)” (2004, 80). Here Solomon says that grief is caused by the recognition of loss and also that it is a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

30     is internal to grief is reflected in the fact that there is, in principle, something we could discover which would extinguish grief: In discovering that I hadn’t suffered a loss—that my mother wasn’t dead after all— I might therein stop grieving.² Grief is, in Thomas Scanlon’s well-known phrase, a judgment-sensitive attitude (1998).³ In this respect, grief differs from a fever: there is nothing I could discover, such that in discovering it, I would cease to have a fever.⁴ This observation reflects the fact that our emotions can be correct or incorrect. Fear is a correct response to danger, anger is a correct response to injustice, and grief is a correct response to loss. This is why emotions make sense to us in light of something that happens in the world—unlike, for instance, a fever, which cannot be correct or incorrect and which we don’t experience in light of anything. I propose to capture all these observations in terms of one criterion: When we experience an emotion, the question of why we experience it, posed in the specific sense in which it asks for our reasons, finds application. (This is a variation of Anscombe’s why-question.)⁵ This is recognition of loss. I think that, to maintain a conception of grief as an engagement with the world, we cannot see it as a causal consequence of our engagement with the world. Solomon’s second claim is the one I endorse. ² There might be inertia in grief, so that grief might persist even past the realization of its incorrectness. However, there needn’t be. In principle, grief could end with the realization of its incorrectness. And even if it does persist, it is fundamentally changed: At that point it should strike us as a condition that befalls us, rather than as our active take on the world. ³ See Solomon (1976; 2007), de Sousa (1987), Roberts (1988; 2003), D’Arms and Jacobson (2000; 2003), and Nussbaum (2001) for endorsements of the view that emotions are reasonsresponsive. ⁴ Cholbi also takes grief and other emotions to be a way of apprehending the world, however in a different way than I do. He writes: “A grief episode that contains moments of anger, fear, and joy (say) provides us evidence about what we find valuable or worthwhile. To learn that we feel anger when we ponder how the deceased person hurt us informs us about what we care about. For example, anger at recalling an instance for which the deceased person failed to attend an important event (a wedding or college graduation, say) tells us about the importance of that event in our lives and about our yearning to share it with others. Likewise for feeling fear, joy, or other emotions people experience in the course of grief. They too inform us about what and who we are, what we care about, and the place of particular others in our network of cares . . . Grief thus looks like our psyche’s way of instigating an emotional data dump” (2017a, 102; cf. 2018, 5). (For an interesting discussion of emotions as evidence, see also Cuneo 2006.) In contrast to Cholbi, I don’t think of emotions as evidence or data on the basis of which we infer what we value. Rather they are themselves ways of apprehending value (cf. Furtak 2018). Grief is not, in the first instance, evidence that I think of myself as having lost something or someone. Rather, it is itself a particularly vivid way of apprehending what I have lost. Treating grief as evidence of loss is possible; however, it constitutes alienation from the emotion. ⁵ Anscombe (1957/2000).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.  -    

31

so, even if we don’t have a good answer, or an answer at all.⁶ In contrast, a condition like a fever does not make such a question applicable: There are no reasons in light of which we suffer from a fever—though, of course, there are reasons why we suffer from the fever, explanatory reasons. Hence, the way in which we make sense of our emotions is different from the way in which we make sense of conditions that merely befall us, and this is because they are, in the relevant sense, reasonsresponsive.⁷ A further, related observation—though one I do not take to be as important—is that we criticize or praise others’ emotional responses. Camus’s character Meursault, for example, fails to properly respond to his loss, and others disapprove of him for that.⁸ However, I do not think that the felicity of criticism or blame is the main rationale for seeing grief as reasons-responsive. It is, rather, the applicability of the why-question, posed in the sense in which it asks for the reasons in light of which we grieve. This, in turn, may well be a condition for the appropriateness of criticism or blame.⁹ How are we to understand the reasons-responsiveness of the emotions? Although I will not propose a general account of this, I want to note that the reasons-responsiveness of the emotions should be distinguished from the reasons-responsiveness of action and may need to be

⁶ See Hubbs (2012) and Hieronymi (2014, 13–14). ⁷ The applicability of the Anscombean why-question also leads me to resist the view of emotions as perceptions. The question of why we grieve is quite unlike the question of why we perceive: the former asks for the reasons in light of which we grieve, whereas the latter, insofar as it is intelligible, asks for the point of looking at something. (Relatedly, see also Brady 2013, esp. 96–7, 112–16.) However, the issues here are complicated by the fact that there is cognitive penetration into perception (Siegel 2017) and that we can be alienated from perceptual experience (Tumulty 2020). Questions about the reasons-responsiveness of perception, though contiguous to the present topic, are beyond the scope of my argument. ⁸ Solomon argues, in light of Meursault’s example, that there is an obligation to grieve. He writes, “We are not just surprised when a person shows no signs of grief after a very personal loss. We are morally outraged and think much less of the person” (2007, 75; cf. 2004, 75–8). See also McCracken (2005). In contrast, Wilkinson (2000, 296–7) argues that we do not see a failure to grieve as a rational failure and, in light of this, concludes that grief is a non-rational response. I find Wilkinson’s argument unconvincing, but I think that only consideration of the Anscombean why-question brings out the shortfall of Wilkinson’s view. ⁹ Bittner argues that “there is nothing recommendable or praiseworthy in grieving. If you can do without it, do not think you rather ought to grieve” (Bittner 1992, 273). I strongly disagree: Meursault misses out on the reality of his loss. This is worthy of criticism.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

32     distinguished from the reasons-responsiveness of belief or judgment. Let me discuss each in turn. To distinguish the reasons-responsiveness of the emotions from the reasons-responsiveness of action, I want to note that emotions, unlike actions, are not conclusions of practical reasoning and are, therefore, not under our voluntary control.¹⁰ Something is a conclusion of practical reasoning if and only if it has a characteristic teleological structure; one reaches the conclusion in light of its being a means to an end or perhaps an end itself. This is true of actions: we conclude that we will do something in light of its being a means to an end or the achievement of an end. And this is clearly not true of our emotions: we do not experience grief or anger in light of their being a means to an end or an end itself: we are not sad or angry in order to achieve an end—even if, in fact, our grieving or being angry achieves something that is our end. Grief and anger, unlike actions, are not embedded in the teleological structure that is characteristic of practical reasoning. That is why their reasons-responsiveness differs from the reasons-responsiveness of action. (A corollary of this way of thinking about the nature of practical reasoning is that the Anscombean why-question defines the class of reasons-responsive phenomena in general, and not specifically the class of intentional actions. The further definition of intentional action is afforded by her characterization of the why-question in teleological terms, that is, her account of practical reasoning.) Must we also distinguish the reasons-responsiveness of emotions from the reasons-responsiveness of belief? This is a difficult question, insofar as it requires an account of the relation between emotion and belief. I do not propose to provide such an account, because I do not think that the puzzle of accommodation depends on a particular way of understanding the relation between emotions and beliefs.¹¹ All I will argue for here is

¹⁰ How exactly to understand the nature of practical reason is, of course, a large and controversial issue that is beyond the scope of the present argument. The view I sketch here is inspired by Anscombe (1957/2000), Hieronymi (2009), and Schwenkler (2019). For further articulation of my view of the nature of practical reason, see Marušicˊ and Schwenkler (2018; 2022). ¹¹ The dialectical position I take up here is exactly analogous to the dialectical position I took up in my discussion of puzzles concerning promises and resolutions to do difficult things: Those puzzles do not presuppose a cognitivist account of intentions, though they are particularly clear on such an account. See Marušicˊ (2015, esp. sect. 2.1 and ch.3).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.  -    

33

that the puzzle does not presuppose a “judgmentalist” account of the emotions, according to which emotions are, or constitutively involve, a corresponding belief.¹² Nonetheless, I maintain a cognitivist view of the emotion in the sense that I take emotions to be apprehensions of their objects—however the notion of apprehension is to be explicated.¹³ There may well be deep differences between emotion and belief.¹⁴ One important difference is that whereas both emotion and belief can be recalcitrant to reasons, the recalcitrance of emotion strikes us as more pervasive and arguably as less problematic.¹⁵ Richard Wollheim puts it well: Perhaps, if we are to think of some emotion of ours as altogether rational, we must think of its object as deserving it. But that is neither the norm that our emotions follow, nor one to which we think they should comply. In our emotional life, we do not always feel ourselves to have right on our side. (1999, 115)

¹² The best known judgmentalists, according to whom emotions involve or consist in evaluative beliefs are Solomon (1976) and Nussbaum (2001). ¹³ A recent view that I find generally plausible (though points of difference remain) is Rick Furtak’s (2018). He says the following about grief: “To perceive the loss of a beloved person just is to feel the emotion of grief, with its specific intentional content” (67). And he characterizes his overall view of the emotions as follows: “Emotions must . . . be understood as feelings through which we apprehend what matters to us. The embodied phenomenology of emotions is linked with the revelation of value or significance” (47). “The somatic agitation we feel when we are trembling with fear, for example, is not a mere sensation but a felt apprehension of danger. And to feel pangs of guilt is to undergo what might be called a thoughtful bodily agitation” (52). “This demands an account of human emotions, not as hybrids of rational judgment somehow conjoined with irrational corporeal commotion, but as feelings through which we apprehend what is significant to us” (52).—I return to the notion of embodiment in Chapter 4. ¹⁴ Maguire (2018) argues that there are no reasons for emotions but that emotions are assessed for fittingness. This is because the considerations that show an emotion to be fitting do not combine in the way that reasons for belief or intention combine. However, I don’t see talk of fittingness as an alternative to talk of reasons. Rather, it seems to me that considerations of fittingness could be understood as a particular kind of reasons, though they combine differently than reasons for other attitudes. Thus D’Arms and Jacobson write, “Crudely put, considerations of fittingness are all and only those considerations about whether to feel shame, amusement, fear, and so forth that bear on whether the emotion’s evaluation of the circumstances gets it right: whether the situation really is shameful, funny, fearsome, and so forth” (2003, 132). This is consistent with Maguire’s argument. ¹⁵ For discussion of emotional recalcitrance to reasons, see Rorty (1978), Greenspan (1981; 1988), Deigh (1994), Nussbaum (2001, 35–6), and D’Arms and Jacobson (2003). Once again, I am sympathetic to Furtak (2018, 56–63), despite differences over details.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

34     Wollheim is surely right that in our emotional life, we do not always feel ourselves to have right on our side. (Indeed, this can be true of belief: alienation from belief is also possible.) For example, we might experience sadness and think that we have no reason to be sad, we might experience fear and know that there is nothing to be feared, or we might experience envy and know that the other’s achievement would not really be of value to us. Even more often, we might experience an emotion very intensely—strong emotional responses to sports and games come to mind—and not take ourselves to be right in so doing. However, that we do not always feel right does not suggest that we do not aspire, in our emotional lives, to get things right.¹⁶ In experiencing an emotion, we recognize that the question of why we experience it finds application, even if we don’t think that we have a good answer. Richard Moran sums up the crucial point well: This is not to say that one normally arrives at one’s beliefs (let alone one’s fears or regrets) through some explicit process of deliberation. Rather, what is essential in all these cases is that there is logical room for such a question, about regret as much as about belief, and that the actual fear or regret one feels is answerable to such considerations. I may confess that my fear is beyond my control and that I can’t help being afraid of something where, by my own lights, there is nothing to be feared. But so long as I am to understand my condition as fear of any kind, even irrational fear, I cannot fail to accept the relevance, the force of the deliberative question “Is there anything to be feared here?” (2001, 63)

Moran’s central thought is that we think of our emotions as at least candidates for correctness, even if we do not always think of them as correct. This suggests that they are, in a salient sense, reasons-responsive. Indeed, even thinking of them as recalcitrant to reason suggests that ¹⁶ Wollheim (1999) argues that our emotions provide us with an attitude to the world, in contrast to beliefs, which give us a picture of the world. But it seems to me that an attitude is reasons-responsive: the question of why to have an attitude finds application in the relevant sense. Thus, I think that Wollheim’s view could allow that emotions are reasons-responsive, though perhaps not in the same way as beliefs. For a similar view to Wollheim’s, which nonetheless takes emotions to be reasons-responsive, see de Sousa (1987, esp. ch.7).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.  -    

35

emotions are reasons-responsive: Recalcitrance is a failure of sorts. In contrast, fevers are not recalcitrant to reason; they are altogether outside the space of reasons. Hence, the phenomenon of recalcitrance speaks in favor of the reasons-responsiveness of emotions, in the relevant sense that concerns me. How, then, are we to understand the reasons-responsiveness of the emotions? This is something that I would like to leave open, because I do not think that the puzzle of accommodation depends on the particulars of an account of the reasons-responsiveness of emotions. The puzzle arises on an account according to which emotions partly consist in judgments, but it also arises on an account according to which emotions don’t consist in, or involve, judgments, but the reasons-responsiveness of emotions is simply to be understood in terms of whether the emotion is a fitting appraisal of a situation. On such a view, the question is simply: Why is accommodation (in the sense of equanimity—of being accommodated) fitting after a certain amount of time has passed? I also want to leave open whether reasons-responsiveness is to be understood in terms of obligation or permissiveness, though I suspect that those are inadequate notions to capture the normative strength of reasons.¹⁷ Yet even if emotions are at most permissible and never required, the reasons-responsiveness of emotions has a temporal profile. For example, even if grief is always permissible, it seems plausible that one has reason to feel grief more strongly right after a loss and less strongly as time passes.¹⁸ Understanding the reasons-responsiveness of emotions in terms of (mere) permissiveness may obscure the puzzle, but it will not address it, as long as the temporal profile of the reasons for grief is not properly understood. This concludes my case for the reasons-responsiveness of emotions. To recap, emotions don’t merely befall us, like a fever or a stomachache. ¹⁷ The point is sometimes put in terms of “warranting” and “requiring” reasons (Abramson and Leite forthcoming). However, I am not at all sure that the notions of obligation and permission can be captured in terms of reasons (cf. Wallace 2019). See section 3.1 for slightly more on this. ¹⁸ Sometimes one may have reasons to pre-grieve a loss—though the strength of reasons for pre-grief seems to depend on the imminence of the loss and, therefore, to exhibit a temporal profile as well. Indeed, it seems to me that there can be a sense of relief when the loss occurs, insofar as one can finally grieve properly.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

36     Rather, they involve apprehension or understanding, even if not belief proper, insofar as the question of why we experience them, asked in the distinct sense in which the question asks for reasons in light of which we experience them, finds application. This may be so even if we realize that we don’t have a good answer to the question and, indeed, even if we realize that the emotion is recalcitrant. The applicability of this question is the decisive criterion of reasons-responsiveness. If this conclusion stands, then, as a corollary, the Antirationalist Response to the puzzle should be rejected.

2.3 In Defense of Reasons-Responsiveness One might still balk at my suggestion that emotions are reasonsresponsive. Here again is the unappreciative anonymous reviewer whose report I quoted in the Preface: The author’s characterization of grief as responsive to reasons is gibberish. Someone is responsive to reasons when what he or she does or believes is determined by the reasons for and against his or her doing it or believing it. The action or the belief is the outcome of a person’s making up their mind about what to do or to believe in view of the reasons favoring and opposing the action or belief. People don’t feel grief over the death of a loved one because they make up their minds about what to feel in view of reasons for and against grieving. Nor do we ever persuade people to feel grief over the death of a loved one by presenting them with reasons to feel grief that are stronger than reasons not to feel grief. To suggest otherwise is ludicrous.¹⁹

There are several possible ways of understanding the objection here. Let me take them in order. ¹⁹ I am not exactly grateful to the anonymous reviewer for their harshly formulated criticisms, though I have reason to be grateful, because they have given rise to considerable reflection and have helped me improve the present line of argument. I suppose my ingratitude is recalcitrant to reasons. But perhaps the reviewer will think that to speak of reasons for gratitude is also gibberish? Or do we make up our mind to be grateful?

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.    -

37

First, it might be thought that emotions are not reasons-responsive, because we never persuade people to have them. However, we certainly do present people with reasons not to feel grief or anger over something—we say things like, “Don’t be so upset: It doesn’t really matter that she forgot your ten-month anniversary. It doesn’t show that she doesn’t care about you!” Or, less often (perhaps out of deference or politeness), we might seek to persuade them to be upset. We might say, “Look, you hurt your father’s feelings. You should feel bad about that!” And sometimes we might even do something that resembles weighing reasons for and against sadness: “On the one hand, it is unfortunate that he forgot my birthday. On the other hand, he did seem quite contrite about it.” Thus, I do not see a decisive difference in our practice of giving and asking for reasons between emotions and other reasons-responsive attitudes.²⁰ A second interpretation of the objection focuses on the claim that we don’t make up our mind about what to feel. The force of the objection depends on what it is to make up one’s mind. There is clearly a sense in which we don’t make up our mind about what to feel: We don’t decide what to feel. However, in that same sense we also don’t decide what to believe. Nonetheless, this does not preclude belief from being reasonsresponsive (as the objector even grants). It seems to me that the appearance of plausibility in the objection trades on an equivocation between two senses of making up one’s mind. There is a practical sense of making up one’s mind, that is, the sense of forming an intention about what to do, and there is also a “settling” sense of making up one’s mind, that is, the sense of reaching a conclusion.²¹ In experiencing an emotion, we do not make up our mind in the practical sense, and this accounts for the plausibility of saying that we don’t make up our mind. However, this is not enough to show that emotions are not reasons-responsive. That is because in experiencing an emotion, we are ²⁰ Perhaps the force of the objection is that we don’t weigh reasons for and against emotions as Maguire (2018) argues. However, it seems to me that this merely shows a way in which emotions are a different kind of response to reasons than beliefs and intentions, not that they are not reasons-responsive. ²¹ See Hieronymi’s extensive work, which takes the notion of settling a question as fundamental (2005; 2006; 2009; 2011; 2013; 2021). My debt to her work goes much deeper than I can acknowledge here.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

38     settled about something, and so we make up our mind in the “settling” sense. And this is sufficient to establish reasons-responsiveness—as shown by the case of belief. After all, from the fact that we don’t make up our mind about what to believe, in the practical sense, it does not follow that we don’t make up our mind about what to believe, in the “settling” sense, nor does it follow that belief is not reasons-responsive. Or, to put the same point in slightly different terms: Just as we cannot infer that belief is not reasons-responsive from the fact that belief is not under voluntary control, we cannot infer that emotions are not reasonsresponsive from the fact that they are not under voluntary control.²² A possible source of confusion, which invites the objection under consideration, lies in the thought that we typically wouldn’t express our grief, by saying, “Therefore, I grieve.” Yet this thought rests on a misunderstanding concerning the expression of grief. That we don’t express grief in this way does not show that grieving is not a way of settling a question or of reaching a conclusion. By the same token, we do not express belief with the statement, “Therefore, I believe.” Nonetheless, believing is a way of settling a question or reaching a conclusion. We express such a conclusion by putting forward something as true. We say, “Therefore, p.” Similarly, we express the conclusion embodied in grief not with a statement concerning our grieving but rather with a characterization of something as a loss. The conclusion embodied in grief is expressed not by saying, “Therefore, I grieve,” but, “Therefore, I have suffered a terrible loss.”²³ In general, we express a mental state not through the description of that mental state but through the specification that the correctness condition of the mental state has been met. A third interpretation of the objection would insist that emotions are not reasons-responsive, because emotions cannot be conclusions of reasoning. This objection would depend on the assumption that if

²² For my preferred view of what it is to make up one’s mind in the “settling” sense, see Boyle (2011). Boyle’s characterization of making up one’s mind as an activity of reason could equally hold of belief and of emotion. ²³ Or perhaps grief is expressed through various actions, including mourning. This seems particularly plausible on the sort of Anscombean view that I am sympathetic to. Thanks to John Schwenkler for discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.    -

39

something is reasons-responsive, then it should, in principle, be possible for it to be a conclusion of reasoning. But why, exactly, should we think that emotions are not conclusions of reasoning? One possible reason to think this is because it is possible for reasoning to conclude in judgment without issuing in the respective emotion—so that, it appears, more than reasoning is required to get us all the way to an emotion. For example, I might conclude that I have suffered a loss, that the loss is significant, and that there are no other relevant circumstances, yet simply fail to feel grief. My reasoning would fail to issue in emotion, which would suggest that the emotion is not the conclusion of reasoning.—Yet the possibility of such emotional akrasia does not show that emotions are not conclusions of reasoning; it only shows that sometimes reasoning can fail to conclude in emotion. Again, there is a parallel with belief: Arguably, we might judge something without therein coming to believe it—without the judgment really sticking. However, the possibility of such doxastic akrasia does not show that belief is not reasons-responsive, nor that it is not a conclusion of reasoning when the belief does stick. Similarly, the possibility of volitional akrasia does not show that, when we are not akratic, our willing is not reasons-responsive. Nonetheless, there is something plausible in the claim that emotions are not conclusions of reasoning. For instance, it would be ludicrous to look for premises from which one would infer an emotion and to wonder whether the emotion follows from them. However, this does not seem to me enough to show that emotions are not reasons-responsive. After all, many of our beliefs are non-inferential—it would be ludicrous to look for premises from which we inferred perceptual beliefs, say—yet they are not, therefore, outside the space of reasons. For example, I might believe that there is a red cup on the table, because I see that there is one, but it would be implausible to say that I infer that there is a red cup on the table from the fact that I see that there is one. Nonetheless, my belief is reasons-responsive. For instance, it is responsive to misleading evidence concerning the color of the cup, such as information about nonstandard lighting conditions. And perhaps something similar is true of our emotions: They are not conclusions of reasoning, in the sense that they are not arrived at through inference, but they are nonetheless

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

40     reasons-responsive in that they are responsive to defeating reasons. For example, my grief for my mother would be responsive to considerations showing that she hadn’t died after all. Ultimately, however, I am not sure how exactly to respond to the objection on its third interpretation. I am unsure whether to insist that even non-inferentially arrived states of mind, including emotions, are conclusions of reasoning or whether to allow that non-inferential states of mind are not conclusions of reasoning but maintain that this does not show that they are not reasons-responsive. Either way, however, the fact that we don’t arrive at emotions through a process of inference, that we can’t identify premises from which we infer our emotions, and that it does not make sense to ask whether the emotions follow from those premises, does not show that they fail to be reasons-responsive, because that same observation does not show that our non-inferential beliefs fail to be reasons-responsive. This concludes my defense of the claim that emotions are responsive to reasons.

2.4 The Wrong Kind of Reasons I now turn to consider what the reasons for emotions are. In particular, I turn to the Pragmatist Response, according to which the badness of grief and anger are reasons against grief and anger, and, therefore, account for the reasonableness of accommodation. I will argue that this is an appeal to the wrong kind of reasons and that, therefore, the Pragmatist Response should be rejected. In formulating the puzzle of accommodation, I have suggested that the main reason why we accommodate ourselves to loss or injustice seems to be that we have to move on: Intense grief and anger are hard to bear, and we’d go to pieces if we didn’t accommodate ourselves. The trouble is that this is a reason of the wrong kind for accommodation: it does not bear on the object of our emotional response but has to do with the emotional response itself. To oversimplify somewhat, I did not grieve for my mother because grieving was somehow good for me. I grieved because my mother had died. Considerations showing that it is good or bad for

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

41

me to grieve are like considerations showing that it is good or bad to believe something: They may make sense of why I believe something, but not in a way that speaks to the why-question which asks for the reasons in light of which I believe. They are the wrong kind of reasons.²⁴ However, one might wonder whether I have prematurely ruled out those reasons as affording an answer to the puzzle.²⁵ One might think that emotions are subject to two dimensions of rational evaluation: First, reasons that are afforded by the object of the emotion and are apprehended in the experience of the emotion itself; and, second, reasons that have to do with the value of the emotion in the individual’s life. One might then think that the reasonableness of accommodation could be explained by appeal to the latter kind of reasons. If we were always and only sensitive to all the loss and injustice we suffer and witness, our lives would be terrible. This justifies a diminution in grief and anger, even though it does not constitute a reason of the right kind. Indeed, once we think of reasons in this way, focusing only on reasons that are afforded by the object of the emotion may strike us as myopic. Perhaps the solution to the puzzle of accommodation lies in adopting a pragmatist view of the rationality of emotions—much like a pragmatist view in the ethics of belief? (Hence—the Pragmatist Response.) However, I think that the Pragmatist Response should be rejected, because it involves, and invites, confusion. To explain why this is so, we first have to be clear about what the charge that something is a reason of the wrong kind consists in. Let’s start with belief. A wrong kind of reason for belief is a consideration that shows it worthwhile to believe p without showing p to be true. Given that truth is the standard of correctness for belief, we can generalize to other attitudes: A wrong kind of reason for an

²⁴ In her critique of anger, Nussbaum (2016) often appeals to reasons that, I think, are of the wrong kind. In this respect, I find her critique unsuccessful. However, sometimes her argument seeks to identify a conceptual incoherence in anger, which would avoid the present objection and strikes me as a more plausible case against anger. ²⁵ This is much in the spirit of the influential discussion in D’Arms and Jacobson (2000; 2009). Though D’Arms and Jacobson draw the distinction between the right and wrong kind of reasons for emotions, they regard the wrong kind of reasons as capable of justifying emotions. What such reasons don’t do, on D’Arms and Jacobson’s view, is show the emotion to be fitting. However, D’Arms and Jacobson argue, this just shows that there are different strands of evaluative thought, not all of which are concerned with morality. For further elaboration of such a view, see, e.g., Howard (2019) and Na’aman (2020; 2021).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

42     attitude (or against an attitude) is a consideration that shows it worthwhile to adopt (or not to adopt) an attitude without showing the attitude to be correct (or incorrect).²⁶ Applied to cases that are of present concern, a wrong kind of reason against grief is a consideration that shows it worthwhile not to grieve without showing grief to be incorrect— that is, without showing that what occurred is not a (currently significant) loss. A wrong kind of reason against anger is a consideration that shows it worthwhile not to be angry without showing anger to be incorrect—that is, without showing that what occurred is not a (currently significant) injustice.²⁷ From this characterization of the wrong kind of reasons, it follows that if something is a reason of the wrong kind, then, insofar as one understands it as such, one can’t adopt the relevant attitude in light of this reason. For example, if you understand a consideration as a wrong kind of reason for belief, then you can’t hold a belief in light of it. Why? Because in understanding it as a wrong kind of reason for belief, you take it to show that it is worthwhile to believe p without showing p to be true—but you can’t believe p in light of a consideration that you don’t take to show p to be true. To believe p in light of some consideration consists (at least partly)²⁸ in taking that consideration to show p to be true. Analogously, you can’t stop grieving in light of something that you understand to be a wrong kind of reason against grief, and you can’t stop being angry in light of something you understand to be a wrong ²⁶ As Alex Marmor points out to me, a wrong kind of reason is a bullshit reason. The bullshitter, as we know from Frankfurt (1988), is unconcerned with truth when making assertions and, generalizing the point, unconcerned with correctness conditions when performing speech acts or adopting attitudes. ²⁷ Trouble awaits if what it is to be a loss or injustice—and, generally, what it is to be a value—is to be understood in terms of reasons that justify certain responses to such values. This is the wrong kind of reasons problem for “reasons-first” approaches to value according to which values are to be understood in terms of reasons. But no such trouble affects my current argument, since I do not mean to presuppose a reasons-first approach. Indeed, I don’t mean to presuppose any particular metaphysics of value. For reasons-first or fitting-attitudes accounts, see especially Scanlon (1998) and D’Arms and Jacobson (2000) and for discussion of the wrong kind of reasons problem for these accounts, see Rabinowicz and RønnowRasmussen (2004), Olson (2004), and, again, D’Arms and Jacobson (2000). I concur with Hieronymi (2005) that the prospects for a solution to this wrong kind of reasons problem are dim. ²⁸ Perhaps more is required—such as doxastic enkrasia or at least the absence of doxastic akrasia. I omit the point for ease of exposition, but it should be understood as implicit throughout.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

43

kind of reason against anger. That is because to grieve or be angry in light of some consideration consists in taking that consideration to show that grief or anger is correct. But wait—how can I be so confident that you can’t do this? After all, people can do all sorts of things! Maybe some people can believe for the wrong kind of reasons, just as some people can wiggle their ears or even blush at will!²⁹—I am confident that this can’t be done because it is a matter of conceptual necessity. It is conceptually necessary that in adopting an attitude in light of a reason, we take the reason to show the correctness condition of the attitude to be satisfied.³⁰ To illustrate this, consider again the case of belief. Suppose you recognize something as a wrong kind of reason for believing p but you nonetheless believe p, allegedly in light of it. You effectively reason: “It’s good for me to believe p, so p.” But this is confused. That it is good for you to believe p does not show p to be true, and you understand this, since you recognize it as a reason of the wrong kind. That is why you cannot believe p for that reason. My argument for the claim of conceptual necessity is to point to the evident confusion in the thought, “It’s good for me to believe p, so p.” The evident confusion in this thought shows that we can’t believe p in light of a reason of the wrong kind.³¹ It might be retorted that people sometimes do believe for the wrong kind of reasons.³² You will hear someone say, “I believe it because it makes me happy,” or, “I believe it because otherwise I couldn’t go on.”— I think that people who say these things are confused.³³ Or, rather, they would be confused if, in making these claims, they were taking

²⁹ The example of blushing at will is from Williams (1973b). Williams uses the example to contrast the psychological impossibility of blushing at will with the conceptual impossibility of deciding to believe. The present point is more or less Williams’s, and I develop it in the way I take myself to have learned from Hieronymi (2006; 2011)—though some differences in formulation remain. ³⁰ Of course, “taking” here is not yet another, separate, attitude. Rather, I mean here to articulate the understanding that adopting an attitude in light of a reason consists in. ³¹ There is the special case in which the fact that it is good for me to believe p can be conclusive evidence for p. But in that case, I don’t recognize the reason as one of the wrong kind, since it is not. Rather, this is a case of self-fulfilling belief—a tricky matter in its own right. See Antill (2019) for an excellent discussion. ³² See especially McCormick (2018; 2019). ³³ D’Arms and Jacobson (2009) argue that there is a range of interesting cases in which it is not clear if people tacitly appeal to the wrong kind of reasons in justifying their attitudes. On

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

44     themselves to articulate the reasons in light of which they believe what they believe. For, these claims commit them to the evidently confused thought, “It’s good for me to believe p, so p.” However, we could alternatively understand these claims as providing causal explanations of belief, rather than specifying the reasons in light of which the beliefs are held. If understood in this way, these explanations would not be confused or problematic but could be understood as admissions of irrational belief. After all, it is perfectly clear that we believe many things because believing them makes us happy (and we believe many things because we are afraid!). But this does not show that it is possible to believe p in light of reasons that are of the wrong kind. Yet doesn’t my saying this amount to a concession that it is possible to believe p in light of reasons of the wrong kind, albeit confusedly? And doesn’t this disprove my claim of conceptual necessity?—That depends on whether confusedly believing something in light of a reason is a way of believing it in light of that reason, or whether it merely seems to be so, out of confusion. In general, it depends on whether confusedly reasoning is a way of reasoning. It seems plausible to me to restrict the notion of reasoning to non-confused reasoning and explain confusion derivatively, as something that is, for the subject, indistinguishable from reasoning. However, I don’t think that a substantial point depends on this: Even if I were to grant that it is possible to confusedly believe something in light of reasons of the wrong kind, the Pragmatist Response would have to countenance that accommodation is reasonable thanks to confusion. Surely that is not how we can explain its reasonableness, or make sense of it from the subject’s perspective? Let me now return to grief and anger. By analogy to the case of belief, I want to say: It is a confusion to think that we could stop grieving or being angry in light of considerations that show grief or anger to be bad for us. This would tacitly commit us to the thought, “Grief is bad for me, so nothing with current significance is lost,” or, “Anger is bad for me, so no injustice with current significance has occurred.” These thoughts are

D’Arms and Jacobson’s view, the normative force of their reasons is opaque: it is not clear if it derives from reasons of the right kind or of the wrong kind.—I want to add: It is opaque to people if they are confused. Such is the nature of confusion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.     

45

evidently confused. Hence, it is conceptually impossible for grief or anger to diminish in light of reasons that are of the wrong kind. It might be objected that even if I am right about belief, the argument does not carry over to grief and anger, because, as discussed earlier, emotions are not conclusions of reasoning. Indeed, it might be thought that the parallel between emotion and belief that I have been drawing on breaks down at this point.—However, the present point does not presuppose that emotions are conclusions of reasoning. It only requires that emotions be defeasible by reasons against having them. This was the point I made earlier in terms of non-inferential perceptual beliefs: It needn’t be that non-inferential perceptual beliefs are arrived at through reasoning. Even if they are not, they do remain defeasible by reasoning. For example, it needn’t be that I reason to the conclusion that there is a red cup before me when I see that there is, but my belief that there is a red cup before me is defeasible by reasons concerning non-standard lighting conditions. On the Pragmatist view that is under consideration here, invocation of pragmatic reasons against grief and anger would be exactly like invocation of reasons concerning non-standard lighting conditions in the perceptual case: They would be reasons that render reasonable the diminution of an emotion in the same way that the considerations about non-standard lighting conditions render reasonable the weakening or suspension of the belief concerning the red cup. This concludes my argument against the appeal to the wrong kind of reasons. My argument, in a nutshell, is that this appeal invites confusion. But the confusion is dispelled once we make fully explicit what the reasons are in light of which, allegedly, our grief and anger diminish. We can then see that they are no reasons at all. I have dwelled on the appeal to the wrong kind of reasons not only to make a negative point. The discussion can also shed light on how we should understand a crucial desideratum for a solution to the puzzle of accommodation: The solution should speak to the subject who experiences grief and anger and undergoes accommodation; it should furnish her with an understanding of a change in her emotion. I hold that the subject cannot understand accommodation as reasonable in light of considerations that show grief or anger to be bad for her, because she cannot accommodate herself to loss or injustice in light of those

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

46     considerations. Of course, she could explain her accommodation by appeal to the badness of grief and anger. Also, from a creatureconstruction perspective on herself (cf. Grice 1974), she could understand why she is wired in such a way that she would accommodate herself to loss and injustice: Grief and anger are hard to bear, and if she did not accommodate herself, she would go to pieces. But this explanation does not speak to the reasons in light of which she experiences her emotions, and so it cannot render the change in her emotions rationally intelligible to her. What we need for an adequate solution to the puzzle is to articulate the agent’s understanding that is involved in accommodation. What the Pragmatist Response, which appeals to the wrong kind of reasons, offers is confusion. But the challenge, as I see it, is to be able to say without confusion why it is all right for grief and anger to diminish if the loss and injustice to which they are a response remain significant. An appeal to the wrong kind of reasons has an illusory appeal.³⁴ The proponent of the wrong kind of reasons asks: What considerations count in favor of grief or anger? And she then finds that some considerations count in favor of grief or anger because they show the emotion to be correct, and some count in favor because they show the emotion to be good for the subject. She concludes that there are, therefore, two kinds of reasons that can justify grief or anger. In contrast, I have sought to articulate the subject’s own understanding that is partly constitutive of her emotional response. However, in experiencing an emotion, a subject apprehends the world in a certain way: Her attention is directed outward, to loss, injustice, and whatever else she may be emotionally responding to. She doesn’t look at herself from outside, as it were, and consider what sort of emotional response would be good for her. This is why the appeal to the wrong kind of reasons presupposes a form of alienation.³⁵ An appeal to the wrong kind of reasons will seem plausible so long as one does not address the puzzle of accommodation head-on, that is, in its first-personal guise. This is not to deny that there may be other ways to ³⁴ Here, again, I take myself to be following Hieronymi. See especially Hieronymi (2005; 2013). ³⁵ Why alienation? Because I relate to myself as I would relate to another (Lat: “alius”): I relate to myself as “someone who happens also to be me” (Moran 2001, 63).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. 

47

address the puzzle without speaking to the subject’s perspective. As Kieran Setiya (in p.c.) has put the point to me, it is common to think of there being more to reasonableness or rationality than responding to reasons. For example, there may be essentially diachronic elements to rationality, such as those articulated by Michael Bratman in his theory of intention (Bratman 1987). Nonetheless, I do not think that the focus on the subject’s perspective is mere myopia. The puzzle of accommodation poses a challenge to our self-understanding: How can we make sense of ourselves when we accommodate ourselves to loss or injustice? How can we make sense to ourselves? To say that the answer lies outside of anything we can articulate is, in effect, to concede the unintelligibility of our emotional response from the subjective perspective, even if, taking a theoretical view on ourselves, we could recognize accommodation as justified.³⁶ Indeed, the view I will eventually develop proceeds exactly along these lines (in Chapter 4). However, it is important to see that an inability to make sense of the puzzle without a theoretical view of ourselves—without alienation—must be regarded as a concession of unintelligibility.

2.5 Attenuation Before leaving the topic of the wrong kind of reasons, I want to identify various ways in which one might covertly appeal to reasons of the wrong kind, perhaps even without realizing it. Confusion can creep into our thought in subtle ways.³⁷ Indeed, it does so in what I think is the most popular response to the puzzle. The first thing that many people say in response to my puzzle is this: Life goes on! When your mother dies, or when something like Sandy Hook happens, you just have to move on. You have to take care of yourself and perhaps others, and you have to keep doing what you have ³⁶ Of course, the theoretical view of oneself is also a subjective perspective—though with the focus on oneself, rather than the bit of the world that the non-theoretical, engaged perspective is focused on. As we learn from Nagel (1986), there is no view from nowhere. ³⁷ We could regard this as another instance of what D’Arms and Jacobson (2009) might consider an interesting appeal to the wrong kind of reasons.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

48     been doing. Your children and your loved ones need you, and you owe it to yourself to live a happy life. All of this presents you with plenty of reasons to stop grieving or being angry! I think that, on a plausible interpretation, this line of thought involves a tacit appeal to the wrong kind of reasons. However, to substantiate this, we need to say more about how, exactly, this line of thought is to be understood. I think that it could be understood in terms of the notion of attenuation.³⁸ Let me explain: It is an important fact that reasons don’t have weight in isolation from other reasons but only as a whole. For example, the fact that a venomous snake is in front of you may make fear fitting. But the fact that the snake is paralyzed would attenuate that reason. The further fact that paralyzed venomous snakes carry a dangerous virus, which could paralyze you, may make fear fitting again. And the fact that you are vaccinated against the virus might again attenuate your reasons for fear. Only by considering all the reasons can it be determined whether fear is fitting or not. And it might be thought that accommodation is reasonable, because with the passage of time, our reasons for grief and anger are somehow attenuated by other things in our life. However, it should be clear that this line of thought contains a covert appeal to the wrong kind of reasons. Thus consider: What would attenuate my reasons to feel sadness, anger, and horror in the face of the Sandy Hook murders? The fact that I have to attend to my own children and carry on with my life is a reason of the wrong kind; it merely shows that it would be good for me not to feel sadness, anger, and horror, not that no injustice (with current significance) has occurred. Also, it is a reason that I had all along. Hence, it is not clear why it would make accommodation reasonable rather than show that my initial emotional response was unreasonable. The fact that I have to carry on may explain why I don’t feel sadness, anger, and horror—why my attention shifts from the murders to other things—but it does not provide a reason for the diminution of my emotional response. Indeed, whatever consideration would be of the ³⁸ See Dancy (2004) for the most thorough discussion of attenuation and related phenomena. For an argument that reasons for fear do not combine, and so attenuation may not be possible, see Maguire (2018). The present argument is compatible with Maguire’s, since I deny that attenuation helps explain the reasonableness of accommodation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. 

49

right kind to attenuate my reasons for sadness, anger, and horror is a consideration that would show that the injustice did not happen after all, or that it is not significant anymore. But there are no such considerations, and that is why there can’t be any attenuating reasons.³⁹ But perhaps there is a better argument for attenuation. Thus, consider the following transmission principle:⁴⁰ Transmission Principle: If φ-ing and ψ -ing are jointly impossible, then reasons for φ-ing are reasons against ψ -ing. The principle is plausible for actions: For example, if you can’t attend both Claire’s as well as Valentina’s birthday party because they are at the same time, then reasons for going to Claire’s party seem to be reasons against going to Valentina’s. But if the principle holds not just for actions but also for attitudes, then it may explain the reasonableness of accommodation. If we assume that it is jointly impossible to continually grieve or be angry, as well as, say, care for those we love, then reasons that speak in favor of caring for those we love are reasons against grieving or being angry.⁴¹ ³⁹ Oded Na’aman (in p.c) has suggested that attenuation of our emotional response to loss or injustice could be understood in a similar fashion as the diminution of regret. Harman (2009) and Wallace (2013) argue that unjustified actions may lead to a change in our reasons—insofar as they change what we desire or what we are attached to—so as to preclude later regret, despite the fact that one recognizes one’s earlier actions as unjustified. Could this serve as a model for accommodation?—I think not: In the sorts of examples that Harman and Wallace discuss, there is a change in reasons for regret. For example, the young mother who has a child at a very early age comes to love her child, and this precludes her from rationally regretting that she had her child so early, because it changes her reasons concerning the child. However, in the case at hand, no such change in reasons with regard to the objects of our emotional response is identifiable: For example, nothing about the murdered children of Sandy Hook, nor my relation to them, has changed that would explain why I would no longer have reason to feel sadness, anger, and horror. Nonetheless, even Harman and Wallace’s case remains complicated, because it is not clear whether the new reasons that the young mother acquires are reasons of the right kind against regret. For discussion, see Na’aman (2021, 259). I discuss Na’aman’s view in section 3.6. ⁴⁰ I owe the present line of thought, and the formulation of the transmission principle to Caleb Perl. Perl has suggested that the transmission principle could be derived from the instrumental principle. He writes (in p.c.), “if A-ing and B-ing are jointly impossible, then it seems like not-A-ing is like a necessary means to B-ing. Since reasons transmit to necessary means, reasons for B-ing would be reasons for not-A-ing. And reasons for not-A-ing look like reasons against A-ing.”—I concur with Perl’s reasoning. However, I think that it supports the argument I make in what follows, since the instrumental principle holds only of actions and, if applied to attitudes, runs afoul of the wrong kind of reasons problem. ⁴¹ I will assume that caring is an attitude, but nothing of philosophical significance depends on this assumption. One can substitute a different example, if one holds that caring is not an attitude.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

50     Yet however plausible the principle may be for actions, it is not plausible for attitudes, precisely because it covertly invokes the wrong kind of reasons. To see this, consider the case of belief. Suppose that believing p and believing q are jointly impossible. Crucially, however, we have to suppose that this is so not in virtue of the relation between p and q but in virtue of the relation of believing p and believing q. We have to imagine that believing p and believing q are jointly impossible, not because, say, we know p and q to be inconsistent, but because, as a matter of fact, we simply cannot hold both beliefs at the same time. If that is so, however, then we can see that reasons in favor of believing p can’t be reasons of the right kind against believing q—because they do not bear on the truth or falsity of q. They are, precisely, reasons of the wrong kind: They speak in favor of having an attitude without showing the attitude to be correct. One might wonder: Why must we assume in this example that believing p and believing q are jointly impossible in virtue of the relation between the belief states, rather than the propositions believed?—We must assume this, because the objection is meant to show how reasons that don’t bear on the objects of our emotions can nonetheless be the right kind of reasons against having those emotions: For example, reasons that don’t bear on whether we have suffered a loss—such as our need to take care of our children—nonetheless have to be reasons against grief, and hence the right kind of reasons.⁴² But if this argument is to work, then the joint impossibility of grieving and caring must be due to our psychological inability to maintain both attitudes, not to any relationship between the objects of those attitudes. Yet if that is correct, then the transmission principle will, in effect, invoke the wrong kind of reasons, because insofar as the reasons for caring for our children speak against grief, they do so without bearing on the correctness of grief. Reasons to care for our children don’t show that we have not suffered a loss, even if grieving and caring are jointly impossible.

⁴² Why is that? Because the present line of objection does not take issue with my characterization of the wrong kind of reasons but seeks to articulate how attenuation might work without running afoul of the wrong kind of reasons.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. 

51

I conclude that an appeal to the transmission principle cannot avoid covertly appealing to the wrong kind of reasons. It is, therefore, as plausible as the Pragmatist Response and should be rejected in light of the argument of the previous section. Let me consider a final attempt to appeal to attenuation without invoking the wrong kind of reasons.⁴³ Again, we start with the observation (which I shall grant for the sake of argument) that it is jointly impossible to grieve or be angry, and experience other emotions. To illustrate, we can imagine that shortly after my mother dies, my child is born. Whereas my mother’s death gives me reason to grieve, the birth of my child gives me reason to be joyful—and let’s say that grief is jointly impossible with joy. Now, one might wonder: Doesn’t the birth of my child affect my reasons to grieve, insofar as it confronts me with a new situation to which grief (alone) is not a reasonable response?— Admittedly, it is not clear what, exactly, the reasonable emotional response in these circumstances is, if grief and joy are jointly impossible. However, it might be thought, it is at least clear that a change in emotional response, which involves a diminution of grief, is reasonable. This is so, not through invocation of reasons of the wrong kind, but because in light of the events that have come to pass, we are now confronted with a wider set of circumstances. And isn’t this exactly the sort of thing that happens in life: We are confronted with other events, usually less momentous than the birth of a child, which provide us with reasons for other emotional responses? There is much I admire in this response. It clearly avoids appealing to the wrong kind of reasons, and it does justice to the phenomenology of emotional experience: We are always confronted with so many things, and we cannot possibly respond to all of them with the emotions that each of them, considered in isolation, would seem to warrant. Nonetheless, I do not think that this response adequately answers the puzzle of accommodation. If anything, it suggests that things are even more complicated than I have been suggesting. For starters, we are always confronted with a rich array of circumstances. If we really take all of them into consideration, then it is not clear ⁴³ I am grateful to Erasmus Mayr for this suggestion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

52     what our emotional response should ever be. For instance, if I had children before my mother died, why would this not have called into question my reasons for grief from the outset? After all, my having children remains something that gives me reason for joy continually.⁴⁴ In general, this response does not account for the temporality of grief and anger—of why they diminish in the way that they do. Moreover, this response does not account for the ways in which accommodation is reasonable in virtue of our coming to terms with loss or injustice. For instance, it does not take into account the fact (if it is one) that grief is a process of detachment.⁴⁵ Finally, even if this response can explain why my grief is, as it were, reasonably dislodged, it does not explain the reasonableness of accommodation. In particular, it does not explain why it is reasonable for me to be equanimous in the face of my mother’s death, rather than wavering between agony and joy. At best, this response can make sense of our being continually emotionally ambivalent. In short, this objection, if it were the last word on the matter, would make our emotional life thoroughly unintelligible. The way forward, I think, is to think more about the nature of the emotions, most importantly their functional roles and other “factual” aspects. I will turn to this in the next chapter. For now, however, I would like to address one further aspect of the rationality of emotions: the nature of their object.

2.6 The Objects of Grief and Anger My criticism of the Pragmatist Response rests squarely on the claim that pragmatic reasons against grief or anger don’t bear on the objects of those emotions. However, until now I have said fairly little, and assumed quite a bit, about what those objects are. In this section, I propose to clarify how to understand those objects.⁴⁶

⁴⁴ See Senior (2014). ⁴⁵ This will be a major theme in Chapter 3. ⁴⁶ Abramson and Leite formulate an analogous view about the objects of love, and they suggest in passing that such a view could hold for other emotions (2011, 681). See also Abramson and Leite (forthcoming).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.      

53

The objects of grief and anger are, in the first instance, persons: We grieve for someone or other and we are angry at someone or other.⁴⁷ For example, I grieved for my mother, and, in the case of Sandy Hook, I grieved for the murdered children and was angry at the shooter and at the political establishment at large. However, to say that grief and anger have objects in this sense—that they have concrete objects—is incomplete.⁴⁸ They also have what is sometimes called formal objects.⁴⁹ This is because when we grieve or are angry, we apprehend the objects of our grief and anger in a certain way.⁵⁰ We grieve for someone or other because they have died (or perhaps because something bad has happened to them), and we are angry at someone or other, because they have done something wrong or have caused a harm. (I don’t mean to assume that anger is always a moral emotion.) Indeed, it seems plausible to me that emotions are individuated in terms of the particular way in which we apprehend their objects; they are individuated in terms of their formal objects. It is here that we come to the formula that I have been drawing on: Grief is a response to loss, and anger is a response to injustice. In grieving, I apprehend something that is true of the concrete object of my grief: My mother has died. In being angry, I apprehend something that is true of my concrete object of anger: He murdered twenty firstgraders. It is this truth that I understand to be the reason for which I experience the emotion. Moreover, this truth has a valence. Thus, I apprehend the truth that my mother died with a different valence than the truth that a dictator died—and this valence is essential to the emotion. It is this valence that I mean to capture when I say that grief is a

⁴⁷ McCracken perceptively notes that grief, unlike other emotions, is “dedicatory”: “In a sense, a person who is grieving recognizes her grief as being not just about or of the object she has lost, but for the lost object: She experiences her grief as dedicatory, as an offering” (2005, 141–2). ⁴⁸ Deigh (1994) argues that it is a mistake to infer from the fact that emotions have intentional objects, that they have propositional contents. I propose no such inference: my point is that emotions have more than intentional objects. They have concrete and formal objects. ⁴⁹ For the notion of a formal object of an emotion, see Kenny (1963, ch.9). For discussion of the difference between the concrete and the formal object of an emotion, see Helm (2009, 251). ⁵⁰ To borrow a term from Roberts (1988; 2003) that I find illuminating, we have a particular construal of them. According to Roberts, a construal can fall short of belief (2003, e.g. 237).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

54     response to loss, and anger is a response to injustice: Both are responses to facts about their objects that have a certain evaluative significance. (One might wonder: what if they are not facts? What if the agent is mistaken? I have been speaking of “apprehending” truths, which is factive, but, surely, we can respond to what we take to be reasons, even if they are not, in fact, true.—This is an issue I want to set aside; it’s an issue that arises whenever we appeal to reasons. My focus here will be on the “good case,” and I will leave the explanation of the bad case for another occasion.⁵¹) On the view I prefer, grief and anger have persons as objects, but they are responses to more than just those objects: they are responses to those objects understood in a certain way—as loss, or as injustice. Those facts provide reason for the emotional response. I want to contrast my way of construing the objects of grief and anger with an alternative possibility. Michael Cholbi argues that the object of grief can’t be a particular person, but it has to be our relationship to the person as it once was.⁵² He writes, “Grief ’s object—what it is we grieve for as we experience anger, apathy, fear, etc.—is the loss of the relationship with the deceased as it was (where this also includes hopes or expectations as to how it might have been)” (2018, 12).⁵³ Here is Cholbi’s argument: When we grieve, we grieve a loss. But what sort of loss? Tempting though it is to say that grief ’s object is the deceased person, this cannot

⁵¹ I will thus assume what I take to be the central insight of disjunctivism in perception (e.g. McDowell 1986/1998) and non-conjunctivism in epistemology (e.g. Williamson 2000)—that our relation to the world and to others is prior in the order of understanding to failures of relating to the world and to others. We explain the latter in terms of a shortfall or privation in the former. For discussion of what to say about the “bad case,” see especially Dancy (2000) and Alvarez (2010). ⁵² See Wonderly (2016) for a similar view, which is developed in terms of the important psychological notion of attachment. Wonderly argues that “many of the self-regarding aspects of grief seem to track the griever’s loss of felt security” (238). Thus at least some central aspects of grief have their source in a relationship of attachment. Though I cannot do justice to Wonderly’s sophisticated view, I would hope to capture her insight by conceiving of attachment as a background condition for grief, as I will discuss shortly. ⁵³ In light of Cholbi’s formulation here, one might wonder if grief would be reasonable even in advance of the other’s death, since one can lose a relationship as it once was even before the other dies. In a more recent formulation, Chobli writes, “grief ’s formal object is the transformation of the bereaved’s relationship with the deceased” (2021, 60)—presumably the transformation of the relationship by the loved one’s death.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.      

55

be exactly correct. For while it is possible to feel sadness at any person’s death, grieving their death requires that the deceased stand in a minimally intimate relationship of some kind with those who grieve them. We characteristically grieve the deaths of spouses, siblings, parents, close relations, and friends. Conversely, claims to grief weaken as the intimacy of the relationship with the deceased weaken. (2017b, 285)

In light of the observation that the nature of one’s relationship (as it was) determines the valence of grief, Cholbi goes on to argue that grief is selfconcerning and agent-relative. He concludes: Given that grief is self-concerning and agent-relative but not directed at the goods that the deceased person provided us or the larger world, we are left with but one possibility: We grieve for the relationship we lose with the deceased person. That is, grief ’s object—what sustains a bereaved person’s attention throughout an episode of grief—is how her relationship is necessarily transformed by the other’s death. (2017b, 286)

Cholbi’s thought, in a nutshell, is that since our relationship to the dead, as it was, matters for grief, grief must be about that relationship. I don’t think that Cholbi is right about this. Here are three initial objections: First, it seems to me that we do experience grief even for those whom we don’t have an intimate relationship with—nothing more than a basic “moral relationship” in which we stand to all fellow human beings (Scanlon 2013, 90). For example, I did grieve for the murdered firstgraders of Sandy Hook, even though I did not have a relationship to any of them; I didn’t even know them. Second, if the formal object of grief is the relationship (as it once was), or the transformation of a relationship, how we can make sense of grief being for another person?⁵⁴ Wouldn’t we, if we were precise, have to say that we grieve for a friendship or a

⁵⁴ Consider again McCracken: “Grief is dedicatory; one experiences grief, not just as being occasioned by a loss, but as being in honor of the thing lost” (2005, 142). Surely, we would experience grief in honor of the dead, and not (just) in honor of the relationship we lost.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

56     relationship—as we would in the case of a breakup? But surely grief for the dead is different from grief (if it is that) over a breakup. Finally, if Cholbi is right, we would never grieve for the same things as others do. For example, since my relationship to my mother, as it was, is different from my sister’s relationship to her, as it was, we would each be grieving for different things—our grief would have different objects. But that is not how it felt at all; rather, it felt that we were united in shared grief— grief for our mother. Indeed, it is unclear to me how, on Cholbi’s view, we could make sense of shared grief. That is why Cholbi’s view strikes me as conceiving of grief in a way that is objectionably self-regarding.⁵⁵ However, I hasten to add that I don’t mean to deny that we couldn’t also grieve for the loss of a relationship (as it was), or even the deprivation that we will suffer—so that there may be a self-regarding element in grief.⁵⁶ Indeed, perhaps in some cases this is what we have most reason to grieve. However, it seems to me that in the central case, the primary object of grief is the other person and not our loss.⁵⁷ Interestingly, however, it strikes me that the primary object of grief is also not the loss for the dead.⁵⁸ In deploring my mother’s death, I don’t even primarily deplore that she is deprived of the goods of life. Rather, ⁵⁵ Cholbi embraces this objection. He explicitly says, “Grief is egocentric at heart” (2018, 11) and grants an objector that his view of grief is “self-centered” (19). For illuminating discussion of what is objectionable about self-concern in grief, see Atkins (2022b). ⁵⁶ See Rinofner-Kreidl (2018, 101) on “primary,” “secondary,” and “tertiary” losses. ⁵⁷ A particularly elegant example of self-regarding grief that is not objectionably selfregarding was suggested to me by Maura Tumulty: “Presumably, when spouses genuinely love each other, they both anticipate the end of their relationship upon the death of either as a loss. If I imagine the devastation I would feel were my husband to die—especially were he to do so ‘prematurely’—I am thinking about the objective loss to the world of this unique and kind person. But I am also thinking about the end of our marriage. And when I do so, I don’t think I have thereby turned my eyes towards myself in the psychologistic sense, or turned away from the object of my grief (my husband, and his death). That’s because the goodness of our marriage is (I think!) something he values, and so something whose end he would see (were he to know his death were imminent) as a loss. And in thinking about our marriage, I am indeed thinking about myself. In intimate relations—between family members, or spouses, or dear friends— when they go well, thinking about one’s self can be a way of continuing contact with the ‘right kind’ of reasons: with the reasons for grief.” I concur with Tumulty about this. But as she seems to grant here, such self-concern is secondary to grief for the objective loss. Also, it seems confined to examples of grief in relationships with people for whom we, too, played a significant role. ⁵⁸ In an earlier discussion of the objects of grief, I was not sufficiently clear about this (Marušicˊ 2018, 6). Thanks, also, to an anonymous reviewer for Oxford University Press for challenging me on what I had said. I am not sure, though, whether they would find the present position any more plausible.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.      

57

grief strikes me, in the first instance, as concerned with something that is not at all agent-relative but rather objective: I grieve in light of the fact that she is no longer in this world. Her death leaves a hole in the world, an irreplaceable absence. I do not conceive of this loss in an agent-relative way, neither relative to me nor to her, nor to the world as a whole. Indeed, it is in this way that I can make sense of grief for those who are not deprived of significant goods by their death (because they are old or very ill, say). The loss, in such cases, consists in the objective fact of their not being part of the world, not of the deprivation of goods that are of agent-relative value to them.⁵⁹ To illustrate this, compare grief for someone who has died with grief over the breakup of a relationship. In the case of the breakup, something like Cholbi’s view strikes me as being correct. In grief over a breakup, the primary object of grief is not another person—since the other is still alive and may not even have suffered much (especially if the other person initiated the breakup!). In grief over a breakup, the primary object is the loss of a relationship as it was, and possibly also the privation of goods for oneself. But this is in stark contrast to grieving for the dead—where the loss that we are concerned with is not self-regarding in this way. There is also a principled philosophical reason to resist Cholbi’s line of thought: He fails to distinguish objects of grief from background conditions. A background condition is something that explains why we have a reason, without being a reason itself. Applied to the present context, a background condition is something that explains why something is a loss or injustice with a particular valence, without itself being a reason for grief or anger. The idea of a background condition is first formulated by Philip Pettit and Michael Smith (1990) in the context of what they call “motivating ⁵⁹ It is sometimes held that the object of grief must be something about us, because we reasonably grieve for the elderly, but in virtue of being elderly, death does not deprive them of much, and so death is not really bad for them. Since our grief is reasonable, it must be the loss to us that constitutes our reason. However, this is not a good argument: If their death does not deprive them of much, then how could it deprive us of much? If life has nothing to give them, how could they have something to give to us? I think that this whole way of thinking is misconceived, because it is too consequentialist. Death is intrinsically and objectively bad, and not just because it is a privation of the goods of life. See Yourgrau (1987; 2019) for systematic development of such a view.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

58     reasons”—psychological states that afford a causal explanation of action. Pettit and Smith argue that desires are to be understood as in the background of deliberation rather than the foreground: They play a role in motivation without figuring in deliberation. Here is how they define the notion of a background desire: [A] desire is present in the background of an agent’s decision if and only if it is part of the motivating reason for it: the rationalizing set of beliefs and desires which produce the decision. A desire is present in the foreground of the decision if and only if the agent believed he had that desire and was moved by the belief that a justifying reason for the decision was that the option chosen promised to satisfy that desire. A desire may be in the background without in this sense figuring in the foreground. And equally a desire may be in the foreground without being in the background. (1990, 568)

On Pettit and Smith’s view, then, “the background-foreground distinction is a functional one” (568). They say, “A desire figures in the background of an agent’s decision making if and only if it plays the sort of role suited to producing choice, whereas it figures in the foreground if and only if it plays the role suited to engaging deliberation” (568). Applied to the present case, Pettit and Smith’s idea could be understood as follows: When I grieve, say, for my mother, my relationship to her plays different roles in motivation and in deliberation: It moves me to grieve without being that which my grief is about. Pettit and Smith’s notion of a background desire is introduced in the context of motivating reasons, that is, psychological states that cause actions or attitudes. But a structurally analogous view could be formulated about normative reasons. Here, for example, is Mark Schroeder’s account of desires as background conditions for normative reasons for action. On Schroeder’s view, a reason is a fact that helps explain why a particular action promotes an agent’s desire (2007, 29).⁶⁰ The desire itself is not a reason; rather it is a feature of the agent that makes it possible for ⁶⁰ See also Dancy (2000, esp. 127–8) and Keller (2013, esp. ch. 5).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.      

59

the agent to have a reason—since it is only in virtue of an agent’s having a desire that there can be facts that explain why an action promotes that desire. Schroeder illustrates the idea with the following example: There is going to be a party tonight, and Ronnie has a desire to dance. On Schroeder’s view, “the fact that there will be dancing at the party tonight is a reason for Ronnie to go there because he desires to dance. That Ronnie desires to dance is part of why this is a reason for him to go there” (2007, 23). But, Schroeder emphasizes, it is not his reason; rather, it’s a feature of Ronnie in virtue of which the fact that there will be dancing at the party is a reason for him to go. In this way, Schroeder suggests, a Humean theory of reasons can explain why in deliberation Ronnie will be concerned with what’s going on at the party, rather than with what’s going on with himself. I think that we should say something analogous about Cholbi’s argument: Our relationship to the dead (as it was) partly explains why we have normative reason to grieve, even though it is not the reason itself.⁶¹ Also, when we grieve, our relationship to the dead (as it was) may well play a motivational role for grief. However, it is a mistake to think that this makes grief into a concern with the relationship (as it was). In our experience of grief, we are primarily concerned with the dead, not with our relationship to them—even if the relationship is that in virtue of which we are so concerned. Relegating the relationship to the background also clarifies why, though grief is not primarily about the relationship, the relationship nonetheless matters a great deal. The relationship to the dead (as it was) at least partly determines the significance of our reasons to grieve. For example, it is in virtue of the fact that it was my mother who died that I have particularly strong reasons to grieve—more reason than a friend or a stranger. However, the relationship is not itself my reason; rather, it is that in virtue of which my reason has its strength or significance.⁶² To

⁶¹ Thus, he could be right that “grieving their death requires that the deceased stand in a minimally intimate relationship of some kind with those who grieve them” (2017b, 285): Such a requirement is satisfied even if the relationship is merely a background condition, and not a reason for grief. ⁶² The argument I present here parallels Keller’s (2013) criticism of the view that reasons of partiality are derived from relationships. Keller writes, “[Y]our reason to give special treatment

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

60     borrow the terminology from Jonathan Dancy, the relationship is an “intensifier” for my reason for grief (2004, ch.3, esp. 41–2). Another way to bring out this point is to compare (if I may) the sadness I felt for the murdered children of Sandy Hook with the grief that their parents must have experienced. No doubt, their grief was deeper and longer lasting than mine. And no doubt, they had many reasons to grieve that I did not have, since they shared their lives with their children. Nonetheless they did not (in the first instance) grieve for something altogether different than I did—for the loss to themselves, or for the relationship to their children. No, they were heartbroken over the murder of their children. And I, too, was heartbroken about that, however different my grief was, and however quickly I accommodated myself to what had happened. What this discussion brings out is that Cholbi is right that grief may involve a good deal of agent-relativity. However, this does not mean that grief is, after all, about us or about our relationships, past or present. Finally, what holds for grief holds for anger and other emotions. If grief is self-regarding, then so are all the emotions—since all emotions exhibit a great deal of partiality (a point I will return to),⁶³ and, therefore, a version of Cholbi’s argument can be put forward about each of them. But our emotional life is not all about us.

2.7 Conclusion In this chapter, I have clarified and defended the reasons-responsiveness of emotions, I have argued against an appeal to the wrong kind of reasons, and I have clarified how I propose to understand the objects of grief and anger. In this way, I hope to have made plausible that we are facing a genuine puzzle. In the next chapter, I turn to several possible responses to this puzzle.

to a friend arises from the value of your friend’s weal and woe for its own sake, and the situation within which that value comes to generate certain reasons for you is set by the background fact that the person shares with you a certain kind of friendship” (138). ⁶³ See section 3.3.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

3 In Defense of the Puzzle In this chapter, I will consider several possible responses to the puzzle of accommodation. In section 3.1, I will consider what I take to be the most important response and the response that, I think, is hardest to resist— the Hardline Response. According to it, accommodation to loss or injustice cannot be reasonable, unless the significance of the loss or injustice changes. I will argue that, although the response is quite attractive, it ultimately fails to do justice to our moral psychology and therefore cannot be acceptable. In the remainder of the chapter, I will then consider how to explain reasons for emotions as time-sensitive. I will take up five attempts at such explanation: In section 3.2, I will consider the reasonableness of accommodation as understood in terms of the shock and surprise in light of a loss or injustice. In section 3.3, I will consider whether it could be understood by analogy with the reasonableness of partiality. In section 3.4, I will consider it in terms of a process, specifically the process of detachment. In section 3.5, I will consider it in terms of the notion of psychological work. Finally, in section 3.6, I will discuss an appeal to background conditions for reasons. However, I will argue that none of these views adequately solve the puzzle of accommodation—mainly because they do not speak to the subject’s own perspective. I will conclude, in section 3.7, that the puzzle eludes a proper solution. However, I will contend that this is as it should be: We should acknowledge that accommodation is, in a sense, an unintelligible and unreconcilable moment in our emotional lives.

3.1 The Hardline Response The puzzle of accommodation arises because it seems plausible that accommodation is reasonable. This seems plausible, because there is a On the Temporality of Emotions: An Essay on Grief, Anger, and Love. Berislav Marušic,́ Oxford University Press. © Berislav Marušic ́ 2022. DOI: 10.1093/oso/9780198851165.003.0003

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

62      general expectation that we accommodate ourselves to loss and injustice, and because we often do so fairly quickly. However, it might be held that this expectation is mistaken. The fact that we accommodate ourselves to loss and injustice, and that we expect such accommodation, does not rest on reasons in light of which this would be reasonable. Hence, accommodation is, really, unreasonable after all. Though this is a hard line to take, there is something very plausible in this thought. Once we clearly reject an appeal to the wrong kind of reasons, it seems that there are no reasons left in light of which we could make sense of the reasonableness of accommodation. Indeed, it seems that accommodation reflects the way in which we are wired and really has nothing to do with loss and injustice. That is why it seems that as long as the significance of a loss or injustice persists, our reasons to grieve and be angry do so as well. Our reasons for grief just are our reasons for judging that we have suffered a loss with continuing significance or that an injustice with continuing significance has occurred. This would imply that in most cases, including my two main examples, we would have reason to grieve or be angry much longer than we actually do—possibly forever. If we failed to do so, we would simply be failing to adequately respond to our reasons. We would be irrational.¹ Indeed, we can think of this Hardline Response as the temporal counterpart to Peter Singer’s view in “Famine, Affluence, and Morality” (1972) that spatial distance does not affect our reasons to aid those in need. Presumably, Singer’s view applies to time as well as to space. Our temporal distance to others may limit the ways we can aid them, since it is not possible to send aid to the past, but our temporal distance as such does not seem to affect the intrinsic moral facts, and therefore the reasons that others’ plight presents us with, any more than our spatial distance does. Just as whether something is a loss or a harm does not

¹ Agnes Callard has argued for a version of the Hardline Response about anger, though she now allows that some considerations can change the significance of injustice and render the end of anger reasonable (Callard 2018). Moller (forthcoming) endorses the Hardline Response about grief and argues that our resilience could be understood as a form of blindness to value. D’Arms and Jacobson (2010) may perhaps be read as holding the Hardline Response. However, since they hold that reasons of the wrong kind can justify (D’Arms and Jacobson 2000; 2009), they could also be understood as endorsing a version of the Pragmatist Response instead. I discuss their view on the instability of affect in section 4.4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

63

depend on where it occurs, it does not depend on when it occurs. To echo Simone Weil: What’s time got to do with it?—“Comme si le temps faisait quelque chose à l’affaire”?² According to the Hardline Response, temporal distance matters only insofar as it affords a psychological explanation of our failure to grieve or be angry. Temporal distance is like spatial distance from those in need as Singer sees it: It explains our reactions but does not provide a reason for them. The Hardline Response may seem very demanding, as, indeed, the comparison with Singer’s view suggests. Yet how demanding it is will depend on how the strength of the reasons for grief and anger is taken to be: One could, variously, take the reasons to be warranting or requiring.³ Thus, a proponent of the Hardline Response could hold, alternatively, that our reasons always require us to experience grief or anger as long as the significance of the loss or injustice persists, or that we are always warranted to do so. The former would be much more demanding, as the latter would allow that we could altogether cease to experience grief or anger, without being unreasonable. Indeed, it might be held that there is, here, room for a great deal of equivocation, since the Hardline Response, understood in terms of merely warranting reasons, is really not a hard line at all. I confess to some uncertainty about how to best understand the Hardline Response. On the one hand, to say that according to the Hardline response we are required to experience grief or anger as long as the significance of loss or injustice persists, seems to construe the

² Maurice Schumann recalls Simone Weil saying: “How can we condemn the holocausts which are in preparation or are being perpetrated around us if we don’t condemn, or even acknowledge the holocausts as truths of the faith [i.e. the killings described in the Hebrew Bible] under the pretext that they occurred thousands of years ago, as if time made a difference to the matter?” (Kahn, ed. 1978, 25, translation and italics mine). (“Comment pouvons-nous condamner les holocausts qui se préparent ou qui se perpètrent autour de nous si nous ne condamnons pas, ou même si nous reconnaissons comme vérités de la foi les holocauses sous prétexte qu’ils se sont écoulés il y a un certain nombre de millénaires, comme si le temps faisait quelque chose à l’affaire?”). I owe the reference to Yourgrau (2010, 127). See Yourgrau (2010) for an account of Weil’s own suffering over temporally distant harms. ³ Here I am indebted to an objection by Doug Portmore. I borrow the terminology of “warranting” and “requiring” reasons from Abramson and Leite (forthcoming). Compare also the terminology of “insistent” and “noninsistent” reasons in Kolodny (2003, 163) who follows Kagan (1989, 378–81).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

64      response as so unrealistically demanding that it seems to be a straw man. On the other hand, to say that according to the Hardline response we are warranted but not required to experience grief or anger as long as the significance of loss or injustice persists, seems to me to construe the response as too undemanding. Ultimately, I suspect that neither interpretation is right, because, I suspect, both confuse the normative strength of obligation and permission with the normative strength of reasons. Plausibly, reasons—that is, considerations which count in favor of a conclusion—are neither as strong as obligations nor as weak as permissions (cf. Wallace 2019, 15). They are simply a different category of normative assessment.⁴ Nonetheless, I don’t want my reply to the Hardline Response to be contingent on this, surely controversial, view in the theory of reasons. This is why I propose to construe the Hardline Response as follows: The crucial point is that the strength of reasons for grief and anger does not vary with time, nor with anything else besides the significance of the loss or injustice. That is why, as long as the significance of the loss or injustice remain the same, the emotional response one will have reason for is also going to be exactly the same, and any change in emotional response will not be reasonable. In particular, a gradual diminution of grief or anger over time, as one has grieved or been angry, will be unreasonable, insofar as it is a change in emotional state for which there are no good reasons. What earns the Hardline Response its name is, therefore, its opposition to change. I think that there is something attractive about the Hardline Response: It is neat and clear and uncompromising. But I do not think that it can be right. Temporal distance does not merely afford a diminution in grief or anger; in many cases it makes it reasonable, or at least not wrong—not unreasonable.

⁴ Oded Na’aman has suggested to me (in p.c.) that this shows that emotions should be understood in terms of considerations of fittingness rather than reasons. However, I don’t see how the move to fittingness changes anything; considerations of fittingness, in the sense in which Na’aman understands the notion, are nothing other than reasons for emotions. The only point in denying this is to countenance wrong kind of reasons as reasons for emotions, which invites confusion (see section 2.4).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

65

Indeed, there is something wrong with being stuck or resentful. For example, there is something wrong with persistent grief. Persistent grief is distinguishable from (statistically) normal grief—and, interestingly, the main criterion is duration. DSM-5 states: “Persistent complex bereavement disorder is diagnosed only if at least 12 months (6 months in children) have elapsed since the death of someone with whom the bereaved had a close relationship . . . This time frame discriminates normal grief from persistent grief” (American Psychological Association 2013, 790). Persistent grief is a mental disorder of sorts.⁵ In contrast, normal grief is not a mental disorder. But if the Hardline Response were correct, persistent grief would not be a mental disorder but the rational response to loss.⁶ This observation suggests, to my mind, that the Hardline Response offers us an unrealistic moral psychology.⁷ The Hardline Response does not adequately take into account how grief and anger are actually experienced: it does not take into account their felt character, which is fleeting. Williams makes a point in this vein in a side remark in his famous “Moral Luck.” In arguing that the justification of moral approval depends on one’s perspective, Bernard Williams says, This is just one of the ways—the distancing of time is another—in which, if the moral sentiments are to be part of life as it is actually experienced, they cannot be modelled on a view of the world in which every happening and every person is at the same distance. (1981a, 37; italics mine)

Williams’s thought is something like this: If we are to understand the justification of moral emotions, we must consider them as they are ⁵ DSM-5 lists it among “Conditions for Further Study” (2013, 789–92) A revision of DSM-5, DSM-5TR, published in March 2022 as the present book was in production, includes the condition “Persistent Grief Disorder.” ⁶ Wilkinson (2000) argues that normal grief is a mental disorder on the grounds that it is not a rational (that is, in present terminology, a reasons-responsive) state. I hold that, because normal grief is reasons-responsive, it is not a mental disorder. ⁷ My argument that the Hardline Response offers an unrealistic moral psychology is indebted to Lucy O’Brien and Douglas Lavin’s discussion in “Living Historically” (unpublished). Unlike O’Brien and Lavin, however, I don’t take the articulation of a realistic moral psychology to solve (or dissolve) the puzzle.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

66      actually experienced—and they change significantly over time. The conclusion I propose to draw is that the reasons for human emotions have to be tailored to the moral psychology of human beings. And since in experiencing an emotion like grief, the temporal distance from a loss matters, whether the emotion is justified depends on its temporal relation to the loss. I can illustrate this objection to the Hardline Response through an illuminating example in Dan Moller’s paper, “Love and Death” (2007, 313–15).⁸ Moller asks us to consider two kinds of alien species—the Super-resilient and the Sub-resilient: [The Super-resilient] are like us except that members have no grief reactions at all to what would strike us as great tragedies . . . When their spouses drop dead in front of them, they shrug their shoulders and check what is on television. They . . . deny not caring for their loved ones; in fact, investigation reveals that they are willing to walk through fire for their husbands and wives, and generally show tremendous concern before their loss. It is just that afterwards adaptive mechanisms operate so as immediately to extinguish any feelings of distress. (313)⁹ The Sub-resilient are like us except that they never cease caring as deeply for their spouses as at the moment of death; the loss of that relationship is as deeply felt at half a century as it is at half an hour. The bereaved Sub-resilient are consequently extremely unhappy people who feel they suffer from a kind of never-healing open wound. (314)

I submit that the reasonableness of grief for members of these two species is different from the reasonableness of grief for a human being.¹⁰ ⁸ Moller’s remarkable paper has been influential in my thinking. For discussion of it, see also Vitrano (2013), Smuts (2016), Preston-Roedder and Preston-Roedder (2017), and Cholbi (2018). ⁹ Ratcliffe argues that it is “somehow incoherent” to say that someone could grieve for a very short time (2017, 157), and I suppose that he would reject Moller’s thought experiment on those grounds. However, I do not quite see why this should be incoherent: It seems to me that it is fairly easy to imagine the Super-resilient beings as subjects of science fiction. Can’t you imagine an episode of Star Trek featuring the Super-resilient? ¹⁰ If it is held that the mental states of the Super-resilient and the Sub-resilient can’t be grief, because mental states are constituted by normative features that govern them and those features differ for the mental states of the other two species, then the argument should be reformulated in terms of grief, grief*, and grief**. However, the point strikes me as merely verbal: For

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

67

A Super-resilient person who grieved like a statistically normal human being would strike members of her community as grieving unreasonably—as suffering from something like persistent grief. And a Sub-resilient person who grieved like a statistically normal human being would strike members of her community also as grieving unreasonably, but for the opposite reason—as suffering from an inability to appreciate the significance of loss in her life. In contrast, a human being who grieved in a statistically normal way will strike us as grieving reasonably: after all, that’s what statistically normal, reasonable grief is like.¹¹ To my mind, this suggests that the reasonableness of an emotion can only be understood against a background of a realistic moral psychology. To see this, just imagine three subjects—a Super-resilient, a Subresilient, and a human—whose reasons for grief are exactly matched; say, each of them loses their spouse after fifteen years of a happy marriage. (You can imagine them to be counterparts in different possible worlds.) And imagine that each of them grieves in exactly the same way, namely as a statistically normal human being would. Initially, they are despairing about their loss and are suffering a great deal; in the course of a couple of months they recover and return to baseline functioning; and in just over a year they remarry. For a Super-resilient, the experience of such grief would seem utterly unreasonable. Just imagine the Superresilient’s friends visiting on the eve of the spouse’s funeral for their weekly movie night. They would be quite astonished to find their friend in tears, unwilling to watch the rom com they had picked out a couple of weeks ago! Meanwhile, for a Sub-resilient, the experience of such grief would seem no less unreasonable. Although initially, on the eve of the spouse’s funeral, things would seem normal, the Sub-resilient’s friends would be quite appalled by their friend’s recovery, which would strike them as brutal callousness. In contrast, the human’s grief would be received in the way that statistically normal grief is typically received— as reasonable. I take this to suggest that we cannot abstract from example, it seems safe to speak of different kinds of chess—e.g. classical chess, Bughouse chess, Fischer Random chess—even though the different kinds are constituted by different, though related, sets of rules. ¹¹ This is not to identify, in general, the reasonable with the statistically normal: Surely unreasonableness is in many cases statistically normal.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

68      contingent, empirical features of the moral psychology of grief, if we seek an adequate understanding of the reasonableness of grief. In particular, how quickly we accommodate ourselves to loss—how resilient we are—is a contingent, empirical matter. A realistic moral psychology should do justice to such contingency, and a conception of the rationality of emotion should be sensitive to such a moral psychology. To say that we need a realistic moral psychology, however, is not yet to solve the puzzle of accommodation. We still need to understand how exactly reasons for emotions are sensitive, or even tailored, to our moral psychology. In particular, we still need to make sense of the anticipation of, and retrospection on, the change in emotion. For instance, when I grieve for my mother, I may know that my grief will diminish with time. I may know that, realistically, this is how things are for a human being like me. But that alone does not help me answer the why-question in light of which the diminution of grief would make sense. It does not help answer the Proustian worry that, with time, I will fail to love my mother. Why would I stop grieving if I continued to love? I conclude that the Hardline Response should be rejected on the grounds that it does not do justice to the moral psychology of grief. Our experience of grief is conditioned by the felt character of grief—and, generally, by our psychology, our physiology, our history, and our social circumstances. Somehow all this also affects our reasons for grief: the reasonableness of grief seems to be constrained by the moral psychology of grief. But it is a philosophical project to explain exactly how this could be—a project I will return to in Chapter 4. I now turn to a number of attempts to explain the time-sensitivity of reasons for grief and anger.

3.2 Shock, Surprise, and Confounding Considerations I have argued that the primary object of grief and anger is another person. However, emotions are complex and not confined to one object. For example, when my mother died, I grieved not only for her: I also experienced shock, since her death was sudden; I experienced concern for my sister who is much younger and was bound to suffer a more

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. , ,   

69

significant impact; I worried about the arrangements that had to be made; and I was disoriented by the deep changes that I would have to make in my plans. I confess to also feeling some relief.—In the case of Sandy Hook, I grieved for the murdered children, I was horrified by the fact that they were so young, I felt deep disdain for the shooter, I felt a great deal of anger about the political situation, and I felt compassion for the parents. But I also felt shock and incredulity, an indistinct sense of guilt, a sudden worry for my own children, and a general sense of disorientation and unsettledness.¹² In both cases, my emotional response was complex: there was so much going on. I think that this richness of emotional experience makes the puzzle I sought to identify hard to recognize. It also invites the thought that accommodation could be understood in terms of elements of this rich experience that are not concerned with the primary objects of the emotions. Perhaps the reasonableness of accommodation could be explained by appeal to the waning shock and surprise over what happened, as well as the regaining of one’s orientation once plans get settled? This is one central idea in Martha Nussbaum’s account of the diminution of grief. Nussbaum poignantly and perceptively reflects on the diminution of grief over her mother’s death.¹³ I will quote Nussbaum at some length and then explain how I propose to understand her view. She writes, When I receive the knowledge of my mother’s death, the wrenching character of that knowledge comes in part from the fact that it violently tears the fabric of hope, planning, and expectation that I have built up around her all my life. But when the knowledge of her death has been with me for a long time, I reorganize my other beliefs about the present and future to accord with it. I no longer have the belief that I will see my mother at Thanksgiving dinner; I no longer think of the end of a busy

¹² See Atkins (2022a) for excellent discussion of the unsettledness of grieving. I cannot do justice here to her subtle and sophisticated analysis of “grief ’s wandering thought.” ¹³ This discussion is anticipated in Nussbaum’s The Therapy of Desire (1994, 375–89). See also her recent Anger and Forgiveness for the contrast between grief and anger (2016, 126). Cholbi (2018) also attributes considerable significance to the disorientation that comes with grief.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

70      day as a time when I can call her up and enjoy a long talk; I no longer think of a trip abroad as an occasion to buy presents for her; I no longer expect to make happy plans to celebrate her birthday. (2001, 80) I will still accept many of the same judgments—including judgments about my mother’s death, about her worth and importance, about the badness of what happened to her. But propositions having to do with the central role of my mother in my own conception of flourishing will shift into the past tense. By now, in August 2000, it is no longer as true of me as it was in 1992, that “my mother is an important element in my flourishing”; I am now more inclined to accept the proposition, “The person who died was a central part of my life,” and this judgmental change itself is a large part of what constitutes the diminution of grief. (82)

There are, as I read Nussbaum, two core considerations that make sense of the diminution of her grief: her adjustment to her mother’s death and her gradual detachment from her mother.¹⁴ The adjustment occurs when she recovers from the shock or surprise, and the detachment occurs as she changes her plans and expectations. I think that Nussbaum is right about both points. Accommodation to loss is, at least to some extent, reasonable insofar as one adjusts to the death of a loved one, and also insofar as, through the process of adjustment, the other comes to matter less. However, I don’t think that this is enough to explain the puzzle of accommodation. Concerning Nussbaum’s first point, even though the diminution of grief makes limited sense in light of our adjustment to the other’s death—especially when the death was unexpected—grief is not primarily a response to the surprise or suddenness of the other’s death, but to her death itself. That is why one’s adjustment does not significantly bear on one’s reasons for grief. To the extent that it bears on them, it does so insofar as grief is a response to the time or circumstances of the other’s death, rather than to the death itself.

¹⁴ I take Nussbaum’s saying that some of her value judgments change tense to reflect the fact that her mother comes to matter less. Of course, she comes to matter differently—but less.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. , ,   

71

Nussbaum’s second point—that over time the loved one matters less to us—may identify a good reason for the diminution of grief. Yet even if it does, this does not seem to me to be reason enough. For example, although it is true that my mother matters less to me today than she did a decade ago—to echo Nussbaum, she was a central part of my life, but she no longer is—my love did not disappear as quickly and as thoroughly as my grief. My grief started diminishing very shortly after her death and its diminution was rapid and pretty much complete. Yet I love her more and longer than is reflected in my grief. It is the discrepancy between the duration of grief and the extent to which the loved one matters to us that gives rise to the puzzle of accommodation—even if we acknowledge that over time the dead do, in fact, come to matter less. (The point seems quite disanalogous for anger, because injustice does not come to matter less—so that Nussbaum’s view would work even less well to explain the reasonableness of accommodation to injustice.) Indeed, it remains puzzling to me why Nussbaum’s second point— that her mother plays a less central role in her flourishing—would not constitute a reason for the intensification, rather than the diminution, of grief. After all, isn’t it a further loss—in addition to her mother’s being dead—that her mother no longer plays this central role? For what it’s worth, I have been struck by the thought that it should be distressing that my mother is no longer a central part of my life—for instance, when she missed the birth of her grandchildren. Why shouldn’t this provide further reasons for grief? It seems to me that it should—and yet I did not grieve. Nonetheless, my not grieving seems, somehow, all right.¹⁵ Perhaps something similar could be said about anger and injustice. For instance, if the collective anger over the murders of Sandy Hook has subsided, even though nothing has been done, is this not yet a further injustice? Why are we not further outraged by the collective accommodation to Sandy Hook and to the many other mass shootings? Surely, the significance of such injustice has not changed! I think that Nussbaum’s discussion is perceptive and sensitive. Nonetheless, I think she misses the heart of the puzzle. The diminution of shock and surprise, the adjustment of plans, and the gradual ¹⁵ I am indebted to Faye Halpern for discussion here.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

72      detachment from the lost object obscure the puzzle rather than solve it. They obscure it, because they make some sense of accommodation. But once we attend to the primary objects of grief and anger, we see that this is not enough. We need to account for the temporal profile of reasons for grief and anger.

3.3 Temporality as Partiality A first attempt to provide an explanation of the temporal profile of reasons for grief and anger takes a cue from Williams’s claim, quoted earlier, that “the moral sentiments . . . cannot be modelled on a view of the world in which every happening and every person is at the same distance” (1981a, 37). The place and time of the griever or the angry person, as well as other features of her, seem to play a crucial role in determining her reasons for grief or anger. For instance, I grieved intensely when my mother died, but the deaths of many mothers leave me cold. And that is all right: The fact that it was my mother who died, and that I love her, is partly what determines the significance of my reason for grief. Similarly, I am angry when my son is bullied at school, and significantly less angry, though not cold, when another child is bullied. And that, too, seems all right: The fact that it is my son seems to matter. The reasons for grief and anger at least sometimes exhibit partiality.¹⁶ In light of these observations, we could seek to understand the temporality of reasons for grief and anger by analogy to partiality: We might think that just as the personal relation between the griever and the dead affects the griever’s reasons, the temporal relation between the griever and the dead affects the griever’s reasons. By the same token, just as the personal relation between the angry person and the injustice affects the reasons for anger, the temporal relation does as well. Krister Bykvist explicitly argues for such a view:

¹⁶ See the discussion of Cholbi’s (2017b; 2018) view in section 2.6. For an excellent account of the reasons of partiality, see Keller (2013).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

73

How strongly we should react emotionally seems . . . to depend on temporal matters. For instance, we think it is fitting that the grief of a lost beloved softens with time. More generally, it seems fitting that the extreme horror we once felt towards some terrible massacre softens with time. Other things being equal, it is not fitting to feel the same intense emotion towards past sufferings that occurred thousands of years back in the past as we do towards some current suffering of the same severity. In all these cases, the degree to which it is fitting to positively respond to a state of affairs does not correspond to the degree to which it is good. How strongly one should favour an objectively valuable object depends on the “distance” between oneself and the object . . . [T]his distance has many dimensions, including modal distance, temporal distance, and “personal” distance. It is, therefore, all too crude to say that it is always fitting to feel more strongly about a better state of affairs or to be emotionally indifferent between states of affairs of the same value. (2009, 16)¹⁷

Bykvist’s aim, in making this argument, is primarily critical: He seeks to object to views that postulate a close relation between values and our reasons.¹⁸ However, Bykvist’s discussion also suggests that the griever’s temporal distance to a loss could be understood as akin to her “personal” distance—precisely as a form of partiality. I concur with Bykvist that there often is partiality in reasons for grief and anger. However, I don’t think that temporal distance is analogous to personal distance, or partiality. To show this, I will now consider three examples.

¹⁷ For a similar argument about blameworthiness, see Coleman and Sarch (2012). Like Bykvist, Coleman and Sarch observe that it is appropriate for blaming and related reactive attitudes to diminish over time, even though the blameworthiness of an act does not diminish, and they argue that this is a problem for “buck-passing” theories of blameworthiness. I think that they fail to see that this poses a deeper problem for our understanding of the temporality of blame, a problem that goes well beyond the buck-passing theories. I say more about their view in the Conclusion. ¹⁸ See especially Scanlon (1998) and D’Arms and Jacobson (2000) for influential contemporary accounts, as well as the extensive references in Bykvist (2009, 1, n.1).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

74      First, suppose we don’t learn about an injustice, such as the one that was perpetrated at Sandy Hook, for a certain period of time. For example, suppose you are camping in the woods for three weeks, deliberately taking a break from all news and all electronics. You return only to find that twenty first-graders have been murdered. The realization that it happened three weeks ago is no comfort whatsoever—despite the fact that after three weeks I have already accommodated myself to the same injustice. In contrast, the realization that it wasn’t your child who was bullied, or that it wasn’t your mother who died, will ease your anger or grief, and reasonably so. This suggests that there is an important disanalogy between temporal distance and partiality.¹⁹ Here is another example that brings this out:²⁰ Imagine you are in a car accident with your mother. You survive, but your mother does not. However, you spend ten years in a coma. When you wake up, you are informed that your mother has died. You are not relieved, and have no reason to be relieved, to hear that the accident happened ten years ago. The mere passage of time makes no difference to your reasons. In contrast, if, after initially being told that your mother had died, you learn that it was someone else’s mother after all, you will be relieved and, plausibly, you have reason to be relieved. Again, temporal distance is not like partiality. Finally, consider a real example—Oliver Sacks’s harrowing case history of Clive Wearing, an English musician and musicologist, who suffers from, as Sacks puts it, “the most devastating case of amnesia ever recorded” (2007).²¹ This amnesia prevents Wearing from engaging in temporally extended activities like grief. Sacks describes

¹⁹ Perhaps matters are different if one learns of, e.g., a massacre that happened 500 years ago. (Weil would deny this, of course.) Perhaps one would then experience a diminution in one’s sadness or anger. However, this may be because the large temporal distance makes one less partial to the victims, or perhaps because such large time scales do make a difference to one’s reasons somehow. Crucially, however, the temporal distance has to be very large, and much larger than is required for accommodation.—Here, I am grateful to John Schwenkler. ²⁰ I owe the example to Eli Hirsch. ²¹ Thanks to Jeremy Fantl for pointing me to Sacks’s case.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

75

a conversation with Deborah Wearing, Clive Wearing’s wife, who had written a memoir: When I asked Deborah whether Clive knew about her memoir, she told me that she had shown it to him twice before, but that he had instantly forgotten. I had my own heavily annotated copy with me, and asked Deborah to show it to him again. “You’ve written a book!” he cried, astonished. “Well done! Congratulations!” He peered at the cover. “All by you? Good heavens!” Excited, he jumped for joy. Deborah showed him the dedication page: “For my Clive.” “Dedicated to me?” He hugged her. This scene was repeated several times within a few minutes, with almost exactly the same astonishment, the same expressions of delight and joy each time. Clive and Deborah are still very much in love with each other, despite his amnesia. (Indeed, Deborah’s book is subtitled “A Memoir of Love and Amnesia.”) He greeted her several times as if she had just arrived. It must be an extraordinary situation, I thought, both maddening and flattering, to be seen always as new, as a gift, a blessing.

Eventually, Sacks concludes: It has been twenty years since Clive’s illness, and, for him, nothing has moved on. One might say he is still in 1985 or, given his retrograde amnesia, in 1965. In some ways, he is not anywhere at all; he has dropped out of space and time altogether. He no longer has any inner narrative; he is not leading a life in the sense that the rest of us do. And yet one has only to see him at the keyboard or with Deborah to feel that, at such times, he is himself again and wholly alive. It is not the remembrance of things past, the “once” that Clive yearns for, or can ever achieve. It is the claiming, the filling, of the present, the now, and this is only possible when he is totally immersed in the successive moments of an act. It is the “now” that bridges the abyss. (Sacks 2007; italics mine)

Wearing’s deeply distressing case shows that if we have less reason to grieve over time, it is not because our loss is at a greater temporal

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

76      distance; it is because we have already grieved. Grieving requires something that Wearing does not have—a persistent conception of the past as well as a continued and continuously remembered experience of grieving. Thus, again, temporal distance alone is quite unlike partiality.²² A lesson to draw from these examples might be that what accounts for the reasonableness of accommodation is not temporal distance alone. They suggest that the relevant sort of distance is not merely temporal but consists in a change in us which happens to take time. Call this “emotional” distance.²³ We can now wonder: Could accommodation be understood in terms of “emotional” distance, whatever exactly that is, so that just as “personal” distance makes a difference to our reasons for action, “emotional” distance makes a difference to our reasons for grief and anger? In this way, we could attempt again to explain the reasonableness of accommodation by analogy to partiality. I don’t think that this will work, though it is difficult to argue for this without a better sense of what emotional distance is. To suggest why it won’t work, let me sketch a fourth example, as a variation on the first, albeit reframed in terms of emotional distance instead of temporal distance. Suppose my mother dies, and I spend several months grieving during which I gain emotional distance. I then have a car accident. I wake up with amnesia, which extends over my mother’s death as well as the grieving period. I am told that my mother died and also that I have spent several months grieving for her. I am no more relieved about the information about emotional distance than, in the first example, about the information about temporal distance. The mere fact that I have grieved does not make a difference to my reasons, even if I know this fact. What this shows is that emotional distance is unlike partiality, because whatever exactly it is, it is internal to the consciousness of grief or anger. Grief and anger seem to be like a process, and the reasons for our ²² As a corollary, this suggests that the puzzle of accommodation is distinct from Lucretius’s puzzle about why we fear death but are unconcerned about pre-natal non-existence. Lucretius’s puzzle has to do with the futurity of death and depends essentially on the nature of time. The puzzle of accommodation is not primarily concerned with the nature of time since it is not through the passage of time alone that accommodation is rendered reasonable. ²³ For the notion of “emotional distance” I am indebted to Doug Portmore (in p.c.).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

77

emotional response depend on where, in this process, we are. Accommodation seems to be something that happens as we undergo the process. However, sometimes the process does not even start as time passes (as illustrated by the first three examples), and sometimes the process has to restart (as illustrated by the fourth). In these ways, emotional distance is quite unlike personal or temporal distance: It has an altogether different form. That is why the analogy between the reasonableness of accommodation and partiality does not hold up. But this brings us to a different idea: perhaps the notion of a process holds the key to the puzzle of accommodation.²⁴

3.4 The Process View In this section, I will consider the possibility that the reasonableness of accommodation could be understood in terms of the successful completion of a process of accommodation. I will frame the discussion in terms of the example of grief, though it should be, in principle, possible to tell a corresponding story about anger. And if it is not, then the Process View is unpromising as an account of the reasonableness of accommodation to injustice. It is plausible to think of grief as a process—for instance, a process through which we heal from a psychological injury.²⁵ Alternatively, we could think of grief not as the healing process itself but as a manifestation of it—a manifestation of our “emotional immune system” (much like a fever is a manifestation of our physiological immune system), which regulates our emotional response to loss.²⁶ Alternatively, and perhaps even more plausibly, we could think of grief as an experience of a psychological injury concomitant to a healing process: as the healing

²⁴ Indeed, perhaps emotional distance has to be understood in terms of a process, so that, in my fourth example, one no longer has emotional distance. If so, then my objections to the appeal to emotional distance will be my objections to the Process View. ²⁵ This might seem especially plausible if it is held that grief has stages—for instance, the five stages posited by Elisabeth Kübler-Ross (1969): denial, anger, bargaining, depression, and acceptance. However, see Bonanno (2009, 20–1) for criticism. ²⁶ See Gilbert et al. (1998) for an account of the “emotional immune system,” which regulates our response to loss. I owe the reference to Moller (2007, 305).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

78      process progresses, we feel less grief—just as when we heal from an injury, we feel less pain. Either way, the duration of grief would be determined by the duration of the healing process. This would suggest that accommodation to loss is reasonable when the healing process is complete.²⁷ Call this the Process View. I do not think that the Process View affords an adequate solution to the puzzle of accommodation—essentially because it does not adequately address the first-person perspective. To think of grief as a healing process involves a distinct kind of alienation: it requires taking a detached, clinical view of ourselves.²⁸ When we view grief as a healing process, we turn our attention away from our loss, to ourselves. But this distorts our grief. When I grieve, my attention is directed to my loss, not myself. When my mother dies, I think, not, “This is hard for me,” but “No! She is dead.” In grief, I am transparent: I am the pained apprehension of loss.²⁹ But when I take a detached view of grief, and apprehend that I am suffering from a psychological injury, my attention shifts from my loss to—myself. I think about the fact that I am beset by grief. I can then understand my grief as an emotional response to a psychological injury that I have to live with. I can see my grief as posing a problem for me— something I need to cope with or manage. However, this is an alienated view—a view in which I no longer attend to my reasons for grief but to myself. The Process View cannot offer an adequate solution to my puzzle, because if, as a griever, I think of my grief as a process that I am undergoing, I am no longer attending to what my grief is about. I don’t attend to the reasons in light of which I grieve, and that is why I do not accurately comprehend why my grief should diminish.

²⁷ For versions of the Process View, see Solomon (2004), Goldie (2012, ch.3), Ratcliffe (2017), who offers trenchant criticism of Goldie’s view, Bagnoli (2018), and Na’aman (2020; 2021). I discuss Na’aman’s view in section 3.6. ²⁸ Here I follow what I take to be essentially Moran’s argument in Authority and Estrangement (2001). See also his (in preparation) “Swann’s Medical Philosophy.” For a different sense of alienation that nonetheless has certain affinities with the present discussion, see D’Arms and Jacobson (2010). ²⁹ Brady (2013, 81), following Salmela (2011), denies that emotions are typically transparent. For an excellent defense of transparency, see Johnston (2001), following Buber (1965).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

79

Here is how the point becomes clearest: If the diminution of grief consists in the healing of a psychological injury, then the end of grief is something that we should look forward to. The thought that we will heal should strike us as a relief. But this is not how we feel about the diminution of grief, when we anticipate that the loss will continue to matter to us. We don’t look forward to the end of grief but are dismayed by it. Well, well, perhaps this is too high-minded! Perhaps we are not pure Proustians, and we do look forward to the end of the pain of grief.—Fair enough. However, when we grieve, we do not merely feel pain, and we don’t merely, or mainly, look forward to the end of the pain. We don’t merely regard grief as a malady that befalls us.³⁰ In grieving, we also apprehend our loss, and insofar as we do, we understand our grief as a response to a reason. In this respect, grief differs from a physiological injury: unlike an injury, grief involves understanding. But the Process View does not speak to what we understand in grieving. As far as the Process View is concerned, all that matters is that we heal; it does not matter whether we are healing from (as it were) a rational psychological injury, nor whether the psychological injury (as it were) continues to be rational. Indeed, this very formulation brings the distortion to light: It makes no sense to speak of injuries as rational or irrational—and so, on the Process View, the rationality of grief falls out of the picture. This is not to deny that the Process View contains an insight: When we suffer a loss, we presumably do undergo a healing process. And it is plausible that the duration of grief is reflected in the completion of this process. Someone who is interested in our well-being—a bookie, say, or a doctor, or the human resources department in our company (who thinks of us as a resource!)—will be interested in how the healing process is coming along: How much we are suffering, whether we can get back to work, whether we will need to go on medical leave, and so on. And, as grievers, we may be interested in these things too. However, all of this requires an alienated perspective—a perspective in which the grief is the topic of reflection and not the loss, to which our grief is a response. To ³⁰ Compare here Moran’s illuminating article on Proust, entitled “Swann’s Medical Philosophy” (in preparation).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

80      the extent that we view ourselves as undergoing a process, we become alienated from our grief: We detach the state or activity of grieving from the object of grief. This shows that the Process View does not provide an adequate solution to the puzzle of accommodation to loss: It does not adequately address how to anticipate the diminution of grief, nor does it help us understand, in retrospect, the reasons in light of which our grief diminished. The Process View leaves us, as it were, in a state of double vision: As grievers, we at once apprehend our loss but also apprehend that we are undergoing a healing process. However, we can’t quite seem to be able to hold them in view together: We can’t seem to reconcile these two perspectives.³¹ One might think that this is so because we have worked with too crude a conception of the process. After all, grief involves understanding in a way that a healing process does not. Perhaps if we could describe the process as itself involving understanding, we could eliminate the double vision. The crucial thought would be that the process of grief is an intelligent process. This brings us to the next possible response to the puzzle of accommodation, which conceives of the process in question in more sophisticated terms: Perhaps, the process of grief is psychological work?

3.5 Grief as Work In this section, I will consider whether understanding grief as psychological work could explain the reasonableness of accommodation. Importantly, the view that grief is psychological work is concerned with the emotion of grief, not activities that constitute the activity of mourning. According to the Work View, the emotional response to loss is itself best understood as a form of work—of coming to terms with loss. The Work View is thus a variant of the Process View, which conceives of ³¹ Thomas Nagel deploys the metaphor of double vision to characterize an “insoluble conflict between subjective and objective conceptions of the same thing” (1986, 86). My invocation of the same metaphor is much in the spirit of Nagel’s thought. For further discussion of double vision, see section 4.1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

81

the process in question as an activity. (As noted in the previous section, if the same line of thought does not carry over to anger, the Work View won’t afford a solution to the puzzle of accommodation to injustice.) Consider two prominent formulations of something like the Work View:³² In “Mourning and Melancholia,” Freud understands grief as the activity of “reality-testing” through which one’s libido detaches itself from the “lost object.” He writes, “In mourning time is needed for the command of reality-testing to be carried out in detail, and . . . when this work has been accomplished the ego will have succeeded in freeing its libido from the lost object” (1917/1999, 252). “[W]hen the work of mourning is completed the ego becomes free and uninhibited again” (245).³³ In a similar vein, though without the Freudian theoretical commitments, Nussbaum argues, [T]he experience of mourning is in great part an experience of repeatedly encountering cognitive frustration and reweaving one’s cognitive fabric in consequence. I find myself about to pick up the telephone to tell [my mother] what has just happened—and then see before me that image of her lying in the hospital bed, with the tube coming out of her nose. In every area of my life in which she has played a part, I find myself expecting her to appear—and I then must work to cut short and to rearrange these expectations. (2001, 80–1; italics mine)

³² Plausibly, Cholbi’s view of grief is a version of the Work View. He says, “Grief is a purposedriven activity wherein our emotional attention to the death of another has a point” (2018, 12). Rinofner-Kreidl also speaks of grief as involving “achievements” (2018, 93), “tasks,” and “purposes” (99). Another view, according to which emotions are active processes, is Brady’s (2013). ³³ Freud repeatedly stresses that he doesn’t know the “economic means” by which this work is carried out (245, 255). He offers merely a conjecture: “Each single one of the memories and situations of expectancy which demonstrate the libido’s attachment to the lost object is met by the verdict of reality that the object no longer exists; and the ego, confronted as it were with the question whether it shall share this fate, is persuaded by the sum of the narcissistic satisfactions it derives from being alive to sever its attachment to the object that has been abolished. We may perhaps suppose that this work of severance is . . . slow and gradual” (255). Bonanno rejects the conception of grief as work, because he objects to Freud’s account of grief on the grounds that going through the memories of the dead would perpetuate grief, since it would lead those memories to dominate our consciousness (2009, 17–18). However, I see this as an objection to the particulars of Freud’s account of grief as work, not to the general idea that grief consists in psychological work.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

82      We can set aside the particulars of Freud’s and Nussbaum’s views for present purposes: We can simply take it that what matters is that grieving is an active process, which takes time and during which we accomplish something: we accomplish the detachment from the person or object we have lost.³⁴ The Work View suggests that, if accommodation is reasonable, this is so not in virtue of the passage of time alone but in virtue of the activity of grieving, which takes time.³⁵ Although there is some plausibility in the thought that grief consists in psychological work, ultimately I think that this is a distorted view. In the end we will be left with the same kind of double vision as on the simpler Process View. Here is why. We think of work as a goal-directed activity that aims at change. But we do not think of grief as aimed at change. More precisely, in grieving, we do not think of our grief as aimed at change—at the detachment from the lost object, say. To illustrate: When we work in the garden with the goal of clearing weeds from the flower beds, we continue (if all goes well) until the weeds are cleared, and then we stop. Perhaps we subsequently adopt a new goal, or we change our focus altogether. But as we work, the thought that we will work until all the weeds are cleared is not, in principle, disconcerting—though we might feel overwhelmed at the sight of an overgrown garden. There is nothing problematic as such about the thought that our work will come to an end once our task is complete. And if grief consists in psychological work, then matters should be the same with grief. But they are not—as the Proustian observation makes vivid. When we pull out weeds from the garden, then we (quite literally!) remove the reasons we have to be working—the weeds that grow in the garden. However, when we grieve, we don’t therein produce a change; we don’t

³⁴ In her recent Anger and Forgiveness, Nussbaum presents a less active view of grief: “[W]orking through grief is something that simply happens as life goes on: new ties replace the old, the world revolves less around the departed person” (2016, 126). However, it seems to me that working through grief, insofar as it is really work, is precisely not something that simply happens but something that we do. ³⁵ Indeed, something like this lies behind the notion of Vergangenheitsbewältigung—a very prominent notion in public discourse in Germany (which, via Adorno (1959/1977) goes back to Freud). The German verb “bewältigen” signifies something one would do with a task. Man bewältigt eine Aufgabe.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

83

remove our reasons to grieve.³⁶ Indeed, in grieving, “there is nothing which can be done” (Gustafson 1989, 469).³⁷ Of course, there is much that we can do when we grieve. We can pay respect to the dead, we can acknowledge our loss, and we can manage our grief.³⁸ Much of this is done through various activities of grieving—such as sitting Shiva—and various ways of coping. But these activities are not to be confused with the emotion of grief. These activities may be considered work, but grief, the emotion, is not. And the point holds true for anger as well: We may be motivated by anger to do various things—yell, take revenge, or take a pill. But these activities are not to be confused with the anger itself. Anger is not action.³⁹ There is a principled reason why this is so: Work, unlike grief and anger, is subject to the will. Thus, we can apprehend the temporal limitations of work, because we set them. But we cannot apprehend the temporal limitations of grief and anger, because we don’t set them. For example, when we work in the garden with the goal of clearing weeds from the flowerbeds, we will continue, if all goes well, until the weeds are cleared. But that is because we have set out to clear the weeds from the flowerbeds: we have set that as our goal and, in so doing, we have set the endpoint of our activity. Since it is up to us to clear the weeds from the flowerbeds, we can decide whether to do so, when to do so, and for how long to do so. But whether we grieve, when we grieve, and for how long we grieve is not in the same way up to us. Similarly, whether we are angry, when we are angry, and for how long we are angry is not in the same way up to us. In this respect, grief and anger are like belief—a persistent state or activity that constitutes a response to the world, rather than a goaldirected activity that aims at change.⁴⁰ That is not to say that we are

³⁶ I am indebted to Arden Koehler for these formulations. ³⁷ Gustafson writes, “the peculiar strength and depth of the feeling of grief might well be explained by the absence of any rational motive force and the absence of associated action” (469). Unlike Gustafson, however, I do not conclude that grief is essentially irrational. For extensive criticism of Gustafson, see Cholbi (2017b). ³⁸ See Preston-Roedder and Preston-Roedder (2017) for thoughtful discussion. ³⁹ Callard (2018) is particularly clear about this, though Na’aman (2020) suggests that her own proposal falls short of providing an alternative. ⁴⁰ See Boyle (2011) for an illuminating account of belief as a persistent activity.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

84      delivered to grief and anger—that they just happen to us. Grief and anger are active insofar as they are our response to reasons. They are also activities or states of ours that we can apprehend and manage.⁴¹ But the understanding involved in grief and anger is not goal-oriented and thus does not include a temporal limitation in its self-consciousness. Thus, the end of grief and anger are not accomplishments, and we do not look forward to them as accomplishments.⁴² A corollary observation is that if we were to understand our grief or anger as work, we would take them to be responsive to reasons that show grieving or being angry to be worthwhile: When we work, we understand ourselves to be responding to reasons that show our work to be worthwhile. But when we grieve and when we are angry, we do not understand ourselves as responding to reasons that show our grief or anger to be worthwhile. Grieving and being angry, unlike working, do not involve apprehending a teleological structure of means and ends—even if, as a matter of fact, grieving and being angry are a means to an end, that is, the end of coming to terms with loss and injustice. Emotions are neither an exercise of the will nor are they responsive to practical reasons. Hence, the Work View is mistaken. Nonetheless, there is an insight in the Work View, at least as a view about grief. The insight is that grief is (concomitant to) a process through which we come to terms with a loss. But this insight has to be captured differently. Grief is best understood as akin to a judgment with a temporally limited functional role. However, the functional role is not part of the content of grief.⁴³ The functional role of grief is like the functional role of an overconfident judgment during a race: It contributes to one’s ends, because it can help one win the race, but not through ⁴¹ See Hieronymi (2006) on managerial control and Tumulty (2020, ch.2) on different ways of managing one’s mental states. ⁴² I thus also disagree with Solomon, who writes, “[W]hat happens in grief (and many other emotions) isn’t just consumption or relief but satisfaction, that is, the fulfillment of the desires that constitute the emotion—and, I would add, the fulfillment of the emotion itself” (2004, 95). I do not think there is such a thing as satisfaction in the completion of grief. Freud asks, “Why . . . after [normal mourning] has run its course, is there no hint in its case of the economic condition for a phase of triumph? I find it impossible to answer this question straight away” (Freud, 1917/1999, 255). I venture to say: Because normal mourning does not aim at an accomplishment, there is no sense of accomplishment at the end of it, and hence no triumph over a job well done. ⁴³ See section 4.2 for further discussion of grief ’s functional role.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

85

the content of the judgment. The functional role of grief—be it detachment from the lost object or healing from a psychological injury—is essentially separated from the understanding involved in grief. Thus, we are back to double vision: In grieving, we apprehend our loss. And in apprehending ourselves as grieving, we apprehend ourselves as going through a process in the course of which we detach ourselves from the lost object. But we have not found a way to unify the two perspectives. How to go forward? Well, perhaps we don’t need to unify the two perspectives. Perhaps we can relegate one of them to the background. Perhaps we can relegate the process or activity through which accommodation becomes reasonable to the background just as, in Chapter 2, I proposed we relegate the personal relationship to the objects of grief and anger to the background. Let us, then, return to the idea of background conditions.

3.6 The Backgrounding View As discussed in section 2.6, background conditions are features of an agent which put her in a position to have reasons but are not themselves reasons. On Schroeder’s view, a reason for action is a fact that helps explain why a particular action promotes an agent’s desire (2007, 29).⁴⁴ For example, Ronnie has a reason to go to a party, because he desires to dance and there will be dancing at the party. His reason to go to the party is that there will be dancing there, and this is his reason in virtue of the fact that he desires to dance. Applying this idea of background conditions to reasons for emotions, one could then seek to explain the reasonableness of the diminution of grief and anger in terms of them. I will call this the Backgrounding View. In recent work, Oded Na’aman has defended just such a view. He writes, “With the idea of background conditions in hand, we can acknowledge that even while the facts that constituted reason for an emotion persist, their status as reasons might change due to changes in background conditions” (2021, 257). On Na’aman’s view, whether one has reasons ⁴⁴ See also Dancy (2000, 127–8).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

86      for an emotion is a function of both the considerations that constitute those reasons and one’s place in a process, which could, but need not, be identified with the emotion. He says, “The process is the background condition for the reason that makes the emotion fitting and the progression of the process explains why a fact that was once a reason is no longer so” (2020, 2427). For example, on Na’aman’s view a background condition for having reason to grieve is that one has recently learned of a loss and that one is still early in the process of grief. However, once one has grieved for a while and one has accommodated oneself to one’s loss, one no longer has reason to grieve, not because the consideration which constitutes the reason changes—the facts about the loss still obtain—but because the background conditions change: one completes the process of grief.⁴⁵ Thus, on Na’aman’s view, emotions like grief are self-consuming. He writes, the fact that an emotion has persisted for some time might itself render its dissipation fitting and its continuation unfitting. Like fire, which can be the cause of its own expiration, it is part of the rational structure of certain attitudes that they consume themselves: the longer they endure the less fitting they become. (2021, 251)

This straightforwardly explains why the reasonableness of accommodation has something to do with us—specifically with where, in the process of grief or anger, we are—without invoking the wrong kind of reasons. Thus, Na’aman’s view may seem to afford a solution to the puzzle of accommodation. Before we can draw on Na’aman’s view to formulate a solution to the puzzle, however, we need to resort to an important assumption that Na’aman is not officially committed to—namely that facts about our resilience affect the fittingness of grief.⁴⁶ Thus, we would need to ⁴⁵ The same point holds for anger and its cognates, including resentment: “That I resented the injustice for a long time might be part of the rational explanation of the fitting diminution of my resentment without being the reason (or the fact that constitutes the reason) for my current resentment. The fact that constitutes the reason for whatever resentment I still harbor remains the injustice done to me; it constitutes a reason for milder resentment because I have already resented the injustice intensely in the past” (Na’aman 2021, 252). ⁴⁶ Here I am grateful to Na’aman for helpful correspondence.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

.   

87

establish the claim that the reasonableness or fittingness of grief and the other emotions has to be sensitive, or tailored, to the contingent moral psychology of those emotions—of how they are actually experienced. As I have already argued in response to the Hardline View, this strikes me as a plausible assumption. Also, it seems to me that Na’aman’s account is made to vindicate this assumption; it seems to me that the Hardline View has no need for process rationality—since the view is opposed to the thought that undergoing a process could render a change in emotion reasonable. That is why I propose to proceed with this assumption. Nonetheless, I think that the Backgrounding View does not offer a proper solution to the puzzle, because it does not remove the double vision I described earlier. To see this, I would like to bring out a significant disanalogy between background conditions for reasons for action and background conditions for reasons for emotions. Let us return to Schroeder’s example of Ronnie, who has a reason to go to a party, because he desires to dance and there will be dancing at the party. Now consider a change in background conditions of Ronnie’s reason to go to the party. Suppose that Ronnie sprains his ankle, and, consequently, no longer desires to dance. The facts about the party have not changed: It is still true that there will be dancing at the party. However, this is no longer a reason for Ronnie to go, since he no longer desires to dance. But it will be perfectly clear to Ronnie why he no longer has a reason to go to the party: It is because he doesn’t want to dance anymore. There is no longer any point in going to the party; nothing speaks in favor of going to it. Thus, in understanding his reason to go to the party, Ronnie understands the background condition for having this reason.⁴⁷ In contrast, when, in light of my mother’s death, I have experienced grief for a couple of months and then stop doing so, it is not clear to me why I should stop. I am still confronted with the fact that my mother died, and this is still a significant loss. Why would this not speak in favor

⁴⁷ Ronnie could have a reason to go to the party without realizing it—since the fact that there is dancing at the party could explain why going to the party would promote a desire of his even if he doesn’t know that there is dancing at the party. Indeed, as Schroeder emphasizes, it could be that Ronnie has a reason only if he doesn’t know about the party, if he desires to be surprised (2007, 33). My point is not to deny this. My point is that if Ronnie understands his reason to go to the party, he understands the background condition for it.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

88      of grieving? If grief had a point, the point of grieving would be exactly the same!⁴⁸ In the case of Ronnie, the consideration that constituted a reason for going to the party—that there will be dancing—has lost its attractiveness, because Ronnie doesn’t want to dance anymore. In contrast, the consideration that constituted my reason for grief—that my mother has died—has not lost its significance. Hence, my progression in the process of grief does not make the diminution of my grief intelligible. At least it doesn’t do so without my coming to consider the empirical fact of my resilience, the empirical fact that grief is a process, and the empirical fact that I have made progress in the process of grief—that is, without returning us to the double vision of the Process View. I want to be clear about what, exactly, I mean to be arguing here: I do not mean to deny that emotions are to be understood in terms of process rationality, as Na’aman argues. Indeed, ultimately, I don’t mean to deny that, in some way or other, we could account for the reasonableness of accommodation by appeal to Na’aman’s view. What I do mean to deny is that backgrounding the process helps resolve the double vision. To understand accommodation as reasonable, one has to get into view certain empirical facts about the emotions, and this requires a theoretical perspective on oneself. This perspective eludes integration with the subjective perspective of those very emotions. The point is perhaps clearer if put in this way: In understanding my reasons for grief, I do not therein understand the extent of my resilience. In understanding that my mother has died, I do not know myself to be human rather than Super-resilient or Sub-resilient. Indeed, my understanding of my reason for grief leaves it open whether I am more like the Super-resilient or the Sub-resilient. This is exactly why the rapid diminution of my grief—the fact of my resilience—could surprise me, and why, once I was in a position to anticipate such rapid diminution, I was horrified by the thought that it will come to pass.

⁴⁸ As I argued earlier in response to the Work View (section 3.5), grief and other attitudes are not embedded in a teleological structure in the way that actions are. That is why, whereas the reasonableness of an action is understood in terms of its point, the same is not true of grief.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. :     

89

Na’aman says, [I]f one were to return to one’s emotional and functional baseline without passing through the valley of grief, so to speak, then one would not arrive at the fitting end-state of the process even if there is no other difference between one’s current state and the fitting end-state of the process. (2021, 263)

Even if this is right about human beings, it is not true of the Superresilient. Thus, Na’aman expresses an empirical fact about how human beings experience grief—an empirical fact that is contingent on our moral psychology. Yet how long the process of human grief is, what shape or stages it has, and what unifies it into a single process are all matters of empirical fact. And an understanding of these empirical facts is not contained in an understanding of our reasons for grief. This is why understanding the reasonableness of accommodation gives rise to double vision—an apprehension of the nature of loss or injustice alongside an apprehension of ourselves as empirical creatures who grieve and are angry in certain specific ways.

3.7 Conclusion: In Defense of the Puzzle In concluding, I would like to put forward a lesson that I wish to draw from the rejection of the various attempts to solve the puzzle of accommodation that I discussed in this chapter. The lesson is not that these attempts are wrong; the lesson is, rather, that these attempts do not address the puzzle in a fully satisfying way. Rather, they help reveal an essentially unreconcilable moment in our emotional lives—our inability to apprehend, at once and in one self-consciousness, the objects of our emotions and the empirical nature of those emotions. This is not to say that either of those must remain unknown to us. Double vision is not partial blindness: It is not as if we can only attend to the objects of our emotions, thereby entirely forgetting about the empirical nature of those emotions, only to immediately forget the objects altogether and regard our emotions as conditions from which we suffer. We can apprehend

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

90      both at once, but we cannot fully integrate what we so apprehend. As we focus on one thing, the other escapes into the background of our attention. Hence, we are left with double vision. It seems to me that recognition of this unreconcilable moment in our emotional lives should not come as a surprise to us. Indeed, it is already evident in the Proustian thought that anticipation of the end of grief is horrifying, especially once it is contrasted with examples in which anticipation of the end of an emotion strikes us as a return to reason.⁴⁹ Already in this Proustian thought, we can recognize the tension between the apprehension that is internal to an emotion like grief and the empirical knowledge we have of ourselves as grievers. It strikes me as a mistake to expect that this tension should be resolved. The Proustian thought is not mistaken! It also strikes me as a mistake to expect that it should be the task of philosophical reflection to resolve the tension. After all, why should we think that we can be fully reconciled with ourselves, or that philosophy would be able to afford us such reconciliation? It would be to conceive of our emotional life as too neat and of philosophy as too comforting. Yet this is not to deny a role for philosophical reflection. The arguments I put forward are not, as an unsympathetic reader has put it, a defense of “the null hypothesis.”⁵⁰ Rather what I seek to do is to identify a new task for philosophy. What I hope to accomplish in what follows is to provide a principled explanation for why the double vision we encountered in this chapter remains persistent, and why, therefore, the puzzle of accommodation eludes a solution proper. This explanation will draw on what I take to be three central lessons from this chapter: First, reasons for grief and anger have to be sensitive to our moral psychology or else they presuppose an unrealistic view of these emotions. Second, the tailoring of reasons to our moral psychology has something to do with us, rather than with the passage of time alone. For instance, it is because we

⁴⁹ See section 1.2. ⁵⁰ Another unsympathetic reader writes: “There are some aspects of our emotional lives that cannot be given philosophical illumination, and so are not suitable objects of philosophical inquiry. To the extent that [the argument] is convincing that the diminishing is incomprehensible, it also makes plausible the thought that diminishing is not something for philosophers to focus on”.—Socrates be damned.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

. :     

91

undergo a psychological process that accommodation is reasonable, not merely because time passes. Third, since grief and anger are concerned with loss and injustice, and not with us, the psychological process doesn’t figure in the reasons for grief and anger themselves but is somehow in the background. The challenge is to understand how, once it is relegated into the background, the process puts us in a position to make sense of the reasonableness of accommodation, without resolving the double vision and therein explaining too much. The challenge is to understand ourselves as unreconciled—of course without thereby achieving reconciliation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 12/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

4 Embodied Reason In this chapter, I will present my preferred response to the puzzle of accommodation. The response has two components. First, it aims to provide a principled explanation of why the double vision we encountered in the last chapter remains persistent and why, therefore, the puzzle of accommodation eludes a proper solution. Second, it aims to explain why, nonetheless, accommodation could be reasonable, albeit without resolving the double vision. Here, in a nutshell, is the core of my response: The double vision arises from the fact that we cannot apprehend, at once, the object of our emotion together with empirical facts about the emotion, such as the fact, if it is one, that grief is a process. This double vision is thus a reflection of the twofold reality of an emotion—its subjective and its objective reality. Its subjective reality is the apprehension that constitutes the emotion, such as the apprehension of loss in the case of grief and the apprehension of injustice in the case of anger. Its objective reality is the particular way in which the emotion is realized in us—its specific psychological, physiological, historical, and social reality, such as the fact, if it is one, that grief is a process. It is because these two realities cannot be integrated into a single viewpoint, a single moment of emotional selfconsciousness,¹ that the double vision persists. Nonetheless, in recognizing the objective reality of the emotions, we can come to see accommodation as reasonable, albeit without identifying reasons in light of which this would be so and, therefore, without resolving the double vision. This is why there remains an unreconciled moment in our emotional lives, and accommodation remains, in a salient sense, incomprehensible. ¹ Why self-consciousness and not simply consciousness? Because for human beings, consciousness is a form of self-consciousness. As Sartre puts it, “any positional consciousness of an object is at the same time a non-positional consciousness of itself” (1943/2018, 11). See Boyle (2018; forthcoming) for discussion of this Sartrean idea. On the Temporality of Emotions: An Essay on Grief, Anger, and Love. Berislav Marušic,́ Oxford University Press. © Berislav Marušic ́ 2022. DOI: 10.1093/oso/9780198851165.003.0004

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

94   I will proceed as follows: In section 4.1, I will consider in some detail why the double vision is irresolvable. In section 4.2, I will explain how this affords an explanation of the reasonableness of accommodation, though without identifying reasons in light of which it would be reasonable. In section 4.3, I will discuss in some detail such reasonableness without reasons; I will present it as an instance of pragmatic encroachment—the view that practical considerations can come to bear on reasons for attitudes without themselves being reasons. In section 4.4, I will argue that, nonetheless, there remains something incomprehensible in our emotional experience: Even if accommodation is reasonable, this does not remove the double vision but leaves us with an unreconciled moment in our emotional life. In section 4.5, I will consider the objection that my account commits an is-ought fallacy. In section 4.6, I will conclude by contextualizing the view I put forward in terms of Jean-Paul Sartre’s discussion of bad faith.

4.1 Unreconcilable Double Vision Grief and anger have a twofold reality. On the one hand, they are ways of apprehending loss and injustice: Put crudely, they are about loss and injustice, rather than about us. On the other hand, they have something to do with us: they have an empirical reality that is tied to our being—our psychology, physiology, biology, history, and social circumstances. In a slogan (echoing Sartre),² grief and anger are embodied and situated. When we seek to understand the reasonableness of accommodation, this twofold role of grief and anger gives rise to double vision: On the one hand, in experiencing grief or anger we are apprehending a loss or injustice. On the other hand, in our effort to understand the temporality of grief or anger, we have to think about the embodied and situated reality of those emotions. But because grief and anger are not about this reality, our apprehension of the temporality of grief and anger is at odds ² Sartre holds that freedom is always in situation and that our embodiment is a crucial element of our situation. If we identify freedom with reason, as I like to do, we get the slogan that reason is situated and embodied. This is one of several ways in which the argument of this chapter is inspired by Sartre. I leave the exegesis of Sartre’s actual view for another occasion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

95

with our apprehension of the objects of grief and anger. For instance, in grieving we can either apprehend our loss, or we can apprehend ourselves as undergoing a process of detachment from the lost object. But we cannot do both at once: as we attend to one thing, the other recedes into the background. This is not to say that it disappears from our selfconsciousness altogether: We suffer from double vision, not toggling.³ But why should this be so? Why couldn’t we, at once, apprehend our loss and our grief at the same time? Surely, we can attend to more than one object at once!—The difficulty is that one of the objects we would be attending to happens to be ourselves attending to something else—and that is not just another object among others that exists independently of our self-consciousness. In particular, understanding the temporality of our emotions requires taking a step back from grief—so that grief becomes an object of awareness, rather than the awareness itself. But this stepping back refocuses the original awareness of loss—it transforms it into an object of self-consciousness and therein distorts the grief.⁴ To illustrate this, I would like to consider, for comparison, the double vision in a well-known example from Richard Moran’s Authority and Estrangement—the case of the rakehell, adapted from Kingsley Amis’s That Uncertain Feeling. Moran writes, There’s a well-known line from a novel by Kingsley Amis that concerns a married man with family, who at one point in the story spends an evening at a nightclub with another woman he knows from work. As he sneaks back home after the encounter, he describes himself in his guilty reflections as “feeling a tremendous rakehell, and not liking myself much for it, and feeling rather a good chap for not liking myself much for it, and not liking myself at all for feeling rather a good chap.” (Moran 2001, 175; cf. Amis 1956, 93)

Let us consider the rakehell’s emotional experience. First, he feels guilty. And this is a redeeming fact about him: At least he has this much moral ³ Thanks to Maura Tumulty for this great expression. ⁴ Readers of Sartre will recognize here one possible way to characterize the movement from non-positional, pre-reflective consciousness to positional, reflective, self-consciousness (esp. in The Transcendence of the Ego (1937/2004) and the introduction to Being and Nothingness (1943/2018)). For incisive discussion of Sartre on this topic, see Boyle (2018; forthcoming).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

96   sense—to feel guilt over a betrayal. However, the rakehell himself cannot get this fact into view without stepping back from his guilt and thinking about himself, rather than the object of his guilt—the betrayal—that his guilt is an apprehension of. In stepping back from his guilt, his attention thus shifts from his betrayal to himself; the betrayal recedes into the background. And this, as a matter of fact, is a self-serving attempt at redemption. However, the rakehell himself cannot get this fact into view without stepping back from the sense of redemption. When he does, and when he recognizes how self-serving he is, his attempt at redemption recedes into the background. It is now a fact about him that he has enough self-knowledge to recognize that he is being self-serving. However, the rakehell himself cannot apprehend this fact, without yet again distorting his emotional response.—With enough imagination, we can think of further iterations of this dynamic. Following Moran, I think that the crucial insight here is not that there is no termination point—that the possibility of stepping back and assessing one’s emotional response is always open. This is certainly true, and it is significant: It shows that one cannot get a fully objective view of oneself. (Getting such a view presumably requires apprehending all the relevant facts about oneself as objects of one’s self-consciousness, but the act of apprehending something about oneself is a subjective fact that is not an object of its own apprehension.) However, the more important insight that, with Moran, I wish to draw from the case of the rakehell is that at least in some cases one cannot, at once, get into view the object of one’s emotional response together with that emotional response. In general, Moran’s insight is that, at least when it comes to certain forms of emotional self-consciousness, one cannot integrate one’s subjective perspective with an objective view of that very subjective perspective. Hence, one is stuck with—double vision. The double vision is a structural feature of subjective self-consciousness: subjective self-consciousness cannot apprehend itself as an object of self-consciousness without distortion.⁵

⁵ It is not entirely clear what, on Moran’s view, accounts for the failure of integration. At one point he suggests that this is because “an emotional attitude constitutes something closer to a total orientation of the self, the inhabiting of a particular perspective” (2001, 181). Lucy O’Brien

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

97

It is this structural feature of self-consciousness that lies at the heart of the puzzle of accommodation. Grief and anger are forms of selfconsciousness: they are ways of apprehending certain conditions in the world—loss and injustice. But these forms of self-consciousness are embodied and situated: They have, as it were, an empirical side to them.⁶ Features of their empirical side account for their temporality. But these features cannot become the objects of self-consciousness without changing what self-consciousness is of—and this provides the distortion. To put it metaphorically: These features are features of one’s lens and are therefore not in view of the lens. However, when the lens turns upon itself, the focus changes, and the objects that were in focus before recede, blurry, into the background. It might seem that suffering from such double vision leaves us in an unstable predicament, and very far from understanding accommodation as reasonable.⁷ That is because it might seem that we toggle between perspectives—that we have an unstable, constantly shifting conception of, on the one hand, seeing accommodation as unreasonable and, on the other hand, seeing the emotions as arational conditions that simply befall us. From the subjective standpoint of grief or anger, when we only have the loss or injustice in view, accommodation will strike us as unreasonable (if we can think about it at all), precisely because we cannot identify reasons in light of which it would be reasonable. Meanwhile, from an objective, theoretical standpoint on the emotions (which is still a subjective perspective albeit on a different object), we will see our grief and anger as changing in accord with what we know about our moral psychology—such as that we are undergoing a process and, therein, detaching ourselves from the lost object. And it might seem that we suggests that Moran’s position should, rather, be understood as that “[i]t is taking an attitude towards ourselves in the way we might take an attitude towards another that is the problem, not the fact that we have a mixture of emotions, nor that we have emotions as the objects of other emotions” (2003, 387–8). Both interpretations are ways of further articulating what I take to be the core problem: the impossibility of integrating a subjective perspective with an objective perspective on that very subjective perspective. ⁶ It is an interesting question, which I must leave for another occasion, whether there are forms of self-consciousness that are not embodied, or that don’t depend on embodiment in the way that, I argue here, emotional self-consciousness does. For a little bit more on this question, see the Conclusion, esp. n.2. ⁷ I am grateful to Agnes Callard and Sebastian Bender for probing questions that helped me think through this part of the argument.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

98   will be left toggling between these two forms of self-consciousness. How could this account for the reasonableness of accommodation? However, this way of describing our situation misconstrues the significance of the metaphor of double vision. Double vision is not toggling. In suffering from double vision, we are not torn between two ways of thinking of our emotional experience, perhaps even unaware that we are constantly shifting perspectives. Rather, when in recognizing the twofold reality of the emotions we suffer from double vision, we come to understand something about our emotional experience, and this understanding arrests the toggling. We come to understand that what is available from the subjective perspective of the emotion is not the entire story. There is more to grief and anger than the apprehension of loss and injustice, because these emotions have an empirical reality. The crucial thing, therefore, is that double vision, once we understand it as such, is itself a form of self-consciousness. It is not toggling—but a stable form of self-consciousness, stable in its potential instability. This does not resolve the double vision; it does not integrate the subjective perspective of the emotions with an objective view of them (or a subjective perspective on them). However, it does save us from a restless shifting between perspectives. That is why, I hold, the double vision is, itself, a source of insight about our condition—an insight that, I will now argue, accounts for the reasonableness of accommodation.⁸

4.2 The Reasonableness of Accommodation What, exactly, is this insight? My main thought is this: The reasonableness of grief and anger depends in part on the way these emotions are realized in our moral psychology, and we can, therefore, learn about it empirically—through experience with the particular ways in which the emotions are so realized. In particular, we learn empirically that we ⁸ In characterizing what he calls “double vision” in the context of skepticism, Nagel writes, “[The problem] of insoluble subjective-objective conflict arises when we succeed in constructing an objective conception of something and then don’t know what to do with it because it can’t be harmoniously combined with the subjective conception we still have of the same thing” (1986, 87). The sort of double vision Nagel describes is also not toggling.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.    

99

accommodate ourselves to loss and injustice and also that it is reasonable to do so—though we cannot integrate this empirical insight about reasonableness with the subjective standpoint of our emotions. In this section, I will explain and defend this thought. To make my main thought plausible, let me return to a phenomenological observation about my experience of grief for my mother. When I grieved, I understood that I was grieving, and I understood my reasons for grief. However, as my grief diminished, I was surprised by my resilience. I did not anticipate that my grief would diminish and end in the way it did. Thus, in understanding my reasons for grief, I did not therein understand the extent of my resilience. In understanding that my mother had died, I did not know myself to be human rather than Superresilient or Sub-resilient.⁹ What I did not understand, in grieving, was the objective side of the emotion. This is something I had to learn empirically—through experience with grief or perhaps through observation. (Another thing I learned empirically, through my experience of grief, is that grief comes in waves, the so-called pangs of grief. This, too, strikes me as a contingent, objective feature of grief that is not understood in understanding the reasons for grief.) I think that, so far, this is not particularly controversial: It is just a particular way of saying that emotions have an objective, empirically knowable reality. The crucial point is that what we empirically learn, when we learn about that objective reality, itself bears on our understanding of the reasonableness of our emotions. In experiencing the diminution and end of my grief, I also learned that this is somehow all right. I understood that I am resilient, and I understood that my resilience is relevant to the question of whether grief is reasonable. I did not think of persistent grievers as more reasonable than I, and I did not long to be a persistent griever. I regretted that I could not grieve longer and deeper, and I also regretted that it is all right not to grieve longer and deeper. I continue to have great admiration for those who are capable of grieving longer and deeper—a point I will return to—but, having learned about grief, I do not see the diminution and end of my grief as a rational failure. ⁹ Cf. Moller (2007) and section 3.6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

100   In general, I want to suggest, we learn about the reasonableness of our emotions at the same time, and in the same way, as we learn about the empirical nature of our emotions. When we learn how anger is experienced, we also learn about how anger is reasonably experienced. Emotions are embodied and situated, but so is emotional reason. That is why, when we recognize the objective, empirical reality of the emotions, we can also come to see accommodation as reasonable. It should be clear that this thought was already implicit in my rejection of the Hardline Response, and specifically in my discussion of Moller’s example of the Super-resilient and the Sub-resilient. There, I argued that to understand the reasonableness of grief, for human beings and for other species alike, we need a realistic moral psychology: We need to take into account how grief is actually experienced and what role it actually plays in a subject’s life. Whether a Super-resilient, a Sub-resilient, and a human being are grieving reasonably is not just a function of the objects of their grief, but it also depends on how grief is realized in their form of life—it depends on their moral psychology. This is so, because emotions are forms of embodied and situated reason. On my proposal, whether accommodation is reasonable is thus to some extent an empirical question, to be settled by consideration of the objective nature of the emotions, that is, by developing a realistic moral psychology. Here is a way to make this point particularly clearly. Suppose that an emotion—let’s continue with grief as the example—has a specific functional role. Perhaps grief has, as Freud thought, the functional role of getting our libido to detach itself from a lost object (1917/1999).¹⁰ It seems plausible that this feature of grief not only determines how and when (and, in a non-Anscombean sense, why) we experience grief but also whether and when it is reasonable to do so. In particular, it is reasonable to experience grief while one is in the process of detachment, and it is no longer reasonable to do so, once the process is completed. Indeed, we could see the process as a background condition for having reason to grieve.¹¹ However, because grief is not about this functional

¹⁰ See section 3.5.—This is as good a place as any to hide the little piece of information that my mother was a Freudian psychoanalyst. ¹¹ See section 3.6 and Na’aman (2021).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.    

101

role but about loss—and because in grieving, our attention is on the dead and not (primarily) on ourselves—the bearing of the functional role of grief will not be understood in our experience of grief. It will be an external insight that we acquire independently of our understanding of the reasons for grief.—That is why it makes sense that I was surprised at the diminution of my grief. At this point, one might wonder why I am so tentative about the functional role of grief. Indeed, one might ask, perhaps with some impatience: “Where is your account of the moral psychology of grief and anger? Tell us all about the functional roles! Indeed, if the reasonableness of accommodation is an empirical matter, where are your empirical studies?” In response to these questions, let me clarify how I propose to understand the burdens of proof in the present context. I don’t see it as my task here to provide a full, empirically informed moral psychology of grief and anger. I see my task as more specific—as addressing the puzzle of accommodation. The puzzle confronts us with a “How possible?” question, a question that arises because it seems ordinarily plausible that accommodation is reasonable yet it also seems impossible for it to be so, since we don’t seem to be able to identify the reasons in light of which it would be so. Once we understand that accommodation can be reasonable even in the absence of reasons, but merely in virtue of the fact that emotions are forms of embodied reason, the obstacle to the ordinary thought is removed. In this way, the philosophical task is accomplished: The “How possible?” question is answered, and the ordinary intuition that accommodation is reasonable is no longer threatened.¹² Thus, I can leave it to others to work out the empirical nature of the moral psychology of grief and anger. I understand the present task quite like the task of answering the skeptical challenge: The challenge confronts us with the question of how it is possible to know things through perception, since, for all we know, we could be systematically hallucinating, and, therefore, the

¹² See Cassam (2007) for discussion of “How possible?” questions and the removal of obstacles as a way of answering them.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

102   mind-independent world could be radically different or even non-existent. Once the skeptical challenge has been answered, the philosophical task is accomplished, and we can embrace the thought that we are in a position to acquire perceptual knowledge. We can do this without empirical research on perception, and we can remain agnostic about most scientific and many philosophical questions about the nature of perception. But perhaps this is too quick: Even if, to answer skeptical worries, one need not develop a full-fledged theory of perceptual justification, one might think that more needs to be said about the topic of the emotions. The emotions, unlike perceptual beliefs, have an ethical dimension. In particular, it might be held that even if accommodation to loss is reasonable, accommodation to injustice can never be. To address this line of thought, let me consider Martin Luther King Jr’s words in the following forceful and beautiful passage: Psychologists have a word which is probably used more frequently than any other word in modern psychology. It is the word “maladjusted.” This word is the ringing cry out of the new child psychology—“maladjusted.” Now in a sense all of us must live the well-adjusted life in order to avoid neurotic and schizophrenic personalities. But there are some things in our social system to which I’m proud to be maladjusted and to which I call upon you to be maladjusted. I never intend to adjust myself to the viciousness of mob rule. I never intend to adjust myself to the evils of segregation or the crippling effects of discrimination. I never intend to adjust myself to an economic system that will take necessities from the masses to give luxuries to the classes. I never intend to adjust myself to the madness of militarism and the self-defeating effects of physical violence. And my friends, I call upon you to be maladjusted to all of these things, for you see, it may be that the salvation of the world lies in the hands of the maladjusted. The challenge of this hour is to be maladjusted. (2007, 326)¹³

¹³ I owe the passage to Michaela McSweeney. See McSweeney (forthcoming) for discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.    

103

Alas, I wish I was more maladjusted! There are many things in our social system to which maladjustment seems to be a fitting response. Segregation, discrimination, income inequality, militarism, and physical violence are among many other sources of injustice that remain pervasive. Wouldn’t it be reasonable to be maladjusted in light of them? Indeed, wouldn’t it be unreasonable to be well-adjusted in light of them? It is important to note that the sort of maladjustment King is concerned with goes beyond an emotional response: Most importantly, it concerns action. (King speaks of the intention not to be adjusted to injustice; plausibly, it is actions that are the objects of intentions, not emotions.) Moreover, the relevant action will plausibly pertain not just to injustice but also to our emotional response to injustice: It will be action that strives to disrupt equanimity over injustice, raise awareness of injustice, and inspire outrage against it. None of that is at issue in my argument. I do not countenance the reasonableness of accommodation to injustice in our actions, including the actions we take concerning our emotions. Yet is emotional accommodation to injustice reasonable? Is being emotionally well-adjusted to injustice reasonable?—I am unsure what to say. On the one hand, I have great admiration for Dr King. I wish I was emotionally more maladjusted to injustice, and I certainly don’t think that being maladjusted to it is unreasonable. On the other hand, I do think that, for human beings like us, accommodation is an empirical fact; it is a feature of our resilience. Furthermore, I think that it is reasonable to accommodate oneself to loss and injustice at least to some extent.¹⁴ For instance, it is reasonable that I do not continue to be consumed by sadness and anger as I was in the days after the murders at Sandy Hook. Thus, I do think that the Hardline Response about anger is mistaken. Indeed, this strikes me as consistent with King’s claim: He

¹⁴ I have been privileged not to suffer significant injustice in my life. Thus, I am especially worried about the blind spots of my perspective and about my ability to speak for others. Here I wish to acknowledge the limitations of my perspective. For a little bit more discussion, see the Preface. I also want to note that recognition of the resilience of the oppressed requires judiciousness, as praise for the resilience of the oppressed can serve to reinforce oppression.— I am grateful to Grace Garland for discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

104   calls for us to be maladjusted, not shattered. Even maladjustment requires some accommodation. Nonetheless, I am not in a position to say how much accommodation is reasonable. What is needed, to answer this question, is an ethics of the emotions.¹⁵ This would be akin to an ethics of belief, insofar as it would be distinct from an ethics of action: An ethics of belief is concerned with the question of what to believe, not of what actions to undertake in order to induce belief.¹⁶ Similarly, an ethics of the emotions would be concerned with the question of what to feel, or what to emotionally experience, not of what actions to undertake with respect to our emotional responses. Its chief difficulty would be to explain how ethical expectations could come to bear not only on what we do, but what we feel or emotionally experience, without committing the moralistic fallacy (D’Arms and Jacobson 2000). I think that this can be done, but it is beyond the scope of the present essay to do so. Without a proper ethics of the emotions, all I can say is that some accommodation to injustice is reasonable, but I will leave unanswered the difficult question of how much.

4.3 Reasonableness without Reasons On the account I have put forward, accommodation is reasonable, but we (that is, the subjects undergoing accommodation) are not in a position to identify reasons in light of which this would be so. In this section, I want to consider how, as subjects, we are to understand such reasonableness without reasons. The synoptic characterization of the reasonableness of accommodation that I have used in my initial formulation of the puzzle is that we accommodate ourselves to loss and injustice and that we find that this is somehow all right. I do think that this captures the phenomenology of ¹⁵ For a beginning, see Srinivasan (2018). ¹⁶ This, of course, requires a resolute rejection of pragmatism in the ethics of belief. Such a rejection is of a piece with a rejection of the appeal to the wrong kind of reasons (see section 2.4).—For an attempt to formulate the beginning of an ethics of belief in this vein, see Marušić and White (2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

105

our emotional self-consciousness quite well. However, we may want to know more: How are we to respond to the question of why we are equanimous in light of a loss or injustice, especially if we ask it ourselves? And how are we to think of our reasons for grief or anger—i.e. the loss or the injustice? Can we still recognize them as reasons for grief or anger? To articulate the subject’s comprehension (or lack thereof) of reasonableness without reasons, let me first draw a contrast with akrasia. On the last page of his famous essay, “How Is Weakness of the Will Possible?,” Davidson articulates the subject’s comprehension of akrasia by saying that in the subject’s action there is “something essentially surd” (1970/ 1980, 42). The subject does not understand the reason for her akratic action. Or, rather, she understands her reason, but she does not understand why she did not act on what she thought of as a better reason. This failure of comprehension is, on Davidson’s view, a mark of the irrationality of akratic action. Quite like Davidson, I think of reasonableness without reasons as involving a certain failure of comprehension. Indeed, I find Davidson’s description of the akratic’s self-understanding an apt characterization of accommodation as well: In accommodation there is something essentially surd. What is essentially surd is that the subject cannot identify reasons in light of which accommodation would be reasonable. Instead, the agent seems to be in a position to identify reasons in light of which it is unreasonable. In contrast to Davidson’s understanding of akratic action, however, the accommodated subject is not irrational. The failure to identify reasons is, in this case, not a mark of irrationality but rather indicates that the force of the subject’s reasons is opaque.¹⁷ When I am equanimous in light of my mother’s death, I do not think that I have reason to grieve yet fail to understand why I don’t grieve; rather, I understand my equanimity as reasonable yet fail to understand what reasons there might be for it. The opacity of the rational force of reasons is, as it were, a point reflection of akrasia: Both are failures of comprehension, but one is

¹⁷ See Enoch (2014) and D’Arms and Jacobson (2014) for insightful discussion of the opacity of the normative force of reasons.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

106   understood as rational, whereas the other is understood (or is at least thought of)¹⁸ as irrational. To further make sense of this account of the reasonableness of accommodation, I now want to explain it as a version of pragmatic encroachment. This is the view that practical considerations bear on reasons—in the present case reasons for emotions—without themselves being reasons.¹⁹ In putting forward this explanation, I will draw on an analogy that Kate Nolfi uses to illustrate her version of pragmatic encroachment. Indeed, I find Nolfi’s view a quite illuminating counterpoint to my account of the reasonableness of accommodation. (Unlike classic pragmatic encroachment, Nolfi’s view does not appeal to stakes as the pragmatic considerations that bear on reasons for belief. Rather, she takes the pragmatic considerations to be provided by the functional role of belief. I take this to be a virtue of her view, since the notion of stakes remains quite unclear. In general, I hold that the very notion of pragmatic encroachment should be understood as a family of views according to which any practical considerations could bear on reasons, not just stakes. It seems to me that this remains insufficiently clear in the classic formulations of the view.) In presenting her account of the norms of belief, Nolfi draws an analogy between beliefs and maps. She then argues that beliefs, like maps, have the functional role of putting us in a position to achieve our ends: they are action-guiding. To perform this function, beliefs, like maps, must reflect what is true. However, good mapmakers will introduce a certain amount of distortion to account for human cognition. Nolfi suggests that by the same token good belief-regulating systems may introduce a certain amount of distortion as well: If the cartographer’s goal is to create a map of the New York subway that facilitates human beings’ successful navigation of the city by way

¹⁸ Thus allowing for the possibility of rational akrasia. ¹⁹ Pragmatic encroachment is often formulated as a view about knowledge, but it is usually also taken to be a view about the reasonableness of (all-out) belief (though typically not credences). For classic formulations of pragmatic encroachment, see Owens (2000, ch.2), Fantl and McGrath (2002; 2007; 2009), Stanley (2005), and Hawthorne and Stanley (2008). For recent discussion that has helped my thinking, see Schroeder (2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

107

of the subway, it makes sense that she will begin with information about the geography of the city, the city’s subway routes, the locations of subway stops, which exceptionally common destination points or widely recognized landmarks are easily accessible from particular stops, points where transfers are possible, etc. The cartographer will then construct her map on the basis of this information, introducing distortions as necessary in order to ensure that the map will be a useful tool for human navigators. Similarly, I suggest that ideal cognition with respect to belief regulation will begin with information about reality and construct our belief corpus on the basis of this information. (2018a, 195)

However, then, Nolfi adds, ideal cognition with respect to belief regulation will also introduce distortion which will be conducive to human beings’ capacity to achieve their ends: For instance, it will introduce, as empirical research has shown, some positive illusions and optimism bias. Nolfi’s crucial idea, for present purposes, is that norms of belief are tailored to the functional role of human cognition, which is to guide action.²⁰ Her analogy to maps is particularly illuminating: It is plausible that maps are simplified with a view to their functions. For example, the map of the New York City subway expands the distances between stations on Manhattan and condenses distances in the boroughs that are further out, because the distances between stations are much shorter on Manhattan than far out. If the map were not designed that way, either the stations on Manhattan would blur together or the map would be so huge as to be unhandy. Also, there are many lines on Manhattan that are close together but that are distinct from each other—most prominently the express and the local lines—but on the map they are drawn as one line. Finally, stations often have complicated pathways on different levels for transferring from one platform to another, but on a map the stations are represented as dots. All of this distortion and simplification is, of course, reasonable given the representational function of a subway map, in contrast to, for example, a cartographically correct map for urban planning. ²⁰ Nolfi develops this view in a series of papers, including (2018a; 2018b; 2021).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

108   Nolfi’s suggestion is that our beliefs can be understood as analogues of the map: Just as the functional role of the subway map is to put us in a position to reach our destinations, the functional role of our beliefs is to put us in a position to achieve our ends. And sometimes, as an empirical matter of fact, our ends are best achieved if the belief-regulating mechanisms introduce a certain amount of distortion: a little bit of extra optimism, say, or some positive illusions about ourselves. Although this distortion may lead to false beliefs, these beliefs are perfectly reasonable—or epistemically flawless, as Nolfi puts it—insofar as they fulfill their function: they put us in a position to achieve our ends. Nolfi’s view is a version of pragmatic encroachment: It countenances that practical considerations bear on our reasons for belief, albeit without being reasons themselves. The practically justified distortion that is introduced into our belief-regulating processes is behind the lens; indeed, it is a property of the lens. We find ourselves having positive illusions about ourselves, but we don’t understand them as such—and so we don’t have to explain their reasonableness in terms of reasons which would show such illusions to be conducive to our ends. (This implies that we have to think of the map analogy as one in which the map does not contain the proviso “not to scale.”) However, this is not to say that the distortion is somehow mysterious. After all, subway maps do not lead us to believe that subway stations really are very large red dots. Rather, understanding the distortion requires an empirical insight about maps—an understanding of what subway maps are that is not itself part of what is understood in understanding the map. Analogously, understanding the distortion that our belief-regulating processes may introduce is not itself represented by the belief-regulating system; rather it is an empirical insight about beliefregulating systems. In this way, features about us can affect what is represented in our beliefs without featuring as reasons. I have some reservations about Nolfi’s account of the norms of belief. My main objection is that her form of pragmatic encroachment is reflectively unstable: recognition of the truth of her view, if it is one, is in tension with the truths that would be so recognized. Even if it is true that practical considerations bear on the reasonableness of belief in the way that Nolfi suggests, recognizing this in deliberation is incompatible

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

109

with their so bearing. This, too, can be brought out by consideration of maps, albeit by a disanalogy between maps and belief-regulating systems. Learning that a map introduces distortion does not lead us to impugn the map. However, learning that a belief-regulating mechanism introduces distortion does lead us to impugn the relevant beliefs. Or, to put it even more clearly, suppose that Nolfi is right that, in virtue of their fulfilling belief ’s functional role, some false beliefs are epistemically flawless.²¹ This is reflectively unstable: Learning that a belief is false reveals a decisive flaw in the belief, even if the belief is ideally suited to fulfill its functional role, and even if, had we not learned this, it would have been epistemically flawless. This reveals, I think, a problem for Nolfi’s view.²² In short, I think that she lacks what I have identified as double vision—an empirical view of our belief-regulating mechanisms that we hold alongside the beliefs that we consider to be the outputs of those very mechanisms. Yet whether we will end up with double vision or with something less reconciled is unclear to me. This is a question I am happy to set aside.²³ Despite my reservations about Nolfi’s account of the reasonableness of belief, I think that her discussion of maps aptly illustrates my account of

²¹ See especially Nolfi (2021). ²² Classic pragmatic encroachment (see note 19), which is formulated in terms of stakes, also strikes me as potentially reflectively unstable. To see this, suppose we accept pragmatic encroachment, and we find ourselves moving from a high-stakes situation to a low-stakes situation. In the high-stakes situation, our evidence does not suffice for rational belief, because the stakes are high, though in the low-stakes situation that same evidence does suffice. Since we accept pragmatic encroachment, we now must explain our situation as follows: Earlier, we didn’t rationally believe the same thing that we now rationally believe because the stakes were higher. Yet this seems problematic; it seems like an invocation of the wrong kind of reason. (This is so even if, in the absence of accepting pragmatic encroachment, we can avoid rational instability, as Schroeder (2018) suggests. What generates the problem for pragmatic encroachment is acceptance of pragmatic encroachment.) What a proponent of classic pragmatic encroachment should welcome is an appeal to irresolvable double vision—a theoretical perspective from which we understand that, given the kinds of finite agents we are, stakes bear on what it is rational to believe, and an engaged perspective from which we only consider our evidence. This would leave her in a similar position as I am advocating for understanding the reasonableness of emotional accommodation. ²³ In The View from Nowhere, Nagel stresses that not all conflicts between the subjective and the objective perspectives will lead to double vision. Sometimes the objective perspective will correct the subjective one (1986, 87). I suspect that this may be so if we come to understand that some of our belief-regulating mechanisms introduce distortion: In that case, we would come to understand our beliefs as unreasonable.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

110   the reasonableness of accommodation. What the map analogy brings out is how features of the means of representation can affect the reasonableness of representation without being themselves represented. For example, the need for a subway map to be handy will affect how the distances between stations are represented—and reasonably so—without representing that very need. And the view I put forward earlier is that things are analogous with the emotions. There are features of the emotions that can affect how it is reasonable to represent, or apprehend, the objects of the emotions that are not themselves objects of the emotions. For example, our need to detach ourselves from someone who has died may affect how we apprehend loss—and reasonably so—without itself being the object of grief. The need to detach may make grief reasonable, and the cessation of the need, due to successful detachment, may, in turn, make equanimity reasonable. In this way practical matters would affect, or encroach upon, the reasonableness of emotional experience without themselves being objects of this experience. Of course, the analogy between maps and emotions is not perfect: A significant difference is that the latter are dynamic, whereas the former are static. To improve the analogy, we would have to imagine maps that, due to our practical needs, are changing and fading over time, and that this is what introduces the distortion. Yet in the age of smart phones, this is surely not hard to do. We can imagine a smart phone map which represents things in a way that helps us achieve our aims, and what is represented changes with our aims. In this way, one and the same representation could be appropriate or inappropriate, depending on what our aim is. (If we are looking for a place to eat, we don’t need to know where the laundromats are.) In this way, considerations about us would affect the appropriateness of representations without themselves being represented. Once we conceive of my account of the reasonableness of accommodation as an instance of pragmatic encroachment, we can give an answer to the question of what happens to our reasons through accommodation. Once we have accommodated ourselves, loss and injustice still constitute reasons for grief or anger; however, their force or significance is now different, in virtue of the practical considerations that serve to determine that significance. We still have the same reasons to grieve or be angry;

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

111

however, in virtue of, say, completing the emotional process of detachment, or satisfying the functional role of the emotion, they are no longer forceful. (To draw an analogy with classic pragmatic encroachment,²⁴ it’s as if we had entered a high-stakes context in which, even though the rational credence that p stays the same, the stakes make it the case that this rational credence is no longer sufficient for outright belief that p or for acting on the premise that p.) It is in this way that we can conceive of a change in reasonableness without a change in reasons. Yet one might insist: How is this possible? Have I not somehow tacitly appealed to the wrong kind of reasons after all?²⁵ Indeed, it seems that I must have done so. For instance, Julius Schönherr (2021) interprets my view, formulated in earlier work, in exactly this way. Schönherr quotes the earlier formulation of my view: “over time, as we grieve, it becomes not wrong to grieve less. However, I also think that there is no good way to understand this. When we try to understand it, all we find are reasons of the wrong kind” (2021, 243, quoting Marušić 2018, 16).²⁶ He then goes on to say, “Of course, if grief remains eternally fitting, it must, if attenuated at all, be attenuated for the wrong reasons” (243; italics are Schönherr’s). Yet I want to deny this. Not only is it not true that it must be so; it is not so at all. I deny emphatically that the diminution of grief, or accommodation in general, is rendered reasonable by reasons of the wrong kind.²⁷ My view is that we cannot identify reasons in light of which the diminution of grief would be reasonable, because there are no such reasons. Rather, there can be a change in reasonableness without a change in reasons. Because this is so, there is no way to understand accommodation in terms of reasons—and there is, therefore, in our emotional life something essentially surd. The difference between the sort of view that Schönherr countenances and the view I aim to defend is the same as the difference between pragmatic encroachment and pragmatism.²⁸ On a pragmatic encroachment view, practical considerations bear on what reasons we have, ²⁴ See references in n.19. ²⁵ I am grateful to several people for pressing different version of this objection. ²⁶ See also Marušić (2020). ²⁷ For my rationale, see section 2.4. ²⁸ All pragmatic encroachers, I know, including the authors listed in n.19 above, as well as Nolfi, clearly distinguish their views from pragmatism.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

112   without being reasons themselves. For example, practical considerations may determine how good our evidence has to be to constitute sufficient evidence for belief (Owens 2000, ch.2). By the same token, considerations about the functional role of grief may determine how long it is reasonable to grieve. Yet these are not reasons against grief: Considerations about the functional role of grief do not show that a loss with continuing significance has not occurred. This is why my view does not appeal to the wrong kind of reasons.

4.4 The Surdness of Accommodation One might wonder whether my explanation of the reasonableness of accommodation in terms of pragmatic encroachment is consistent with my insistence that there is something surd or incomprehensible in accommodation, that there is an unreconcilable moment in our emotional lives, and that the double vision we have concerning our emotions is irresolvable.²⁹ In identifying practical considerations as encroaching upon reasons for emotions, by determining their force or strength, haven’t I eliminated the surdness, resolved the double vision, and reconciled our emotional lives? In general, why doesn’t our empirical understanding of the reasonableness of accommodation put us in a position to understand accommodation as reasonable in terms of our reasons—especially if what we come to understand is that the force of our reasons changes?³⁰ The aim of this section is to answer these questions. Very often, what we empirically learn about the emotions helps us make sense of our emotional experience in terms of our reasons. For example, when we learn about fear, we also learn about reasonable fear: We learn not only how fear is felt and manifested but also how much fear is reasonable, and in light of which dangers it is reasonable. When we learn about amusement, we understand both how much amusement is reasonable in light of which circumstances, and we learn how long

²⁹ See sections 4.3, 3.7, and 4.1 for the respective formulations. ³⁰ I am grateful to Richard Moran for pressing me on this question in particularly clear fashion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.    

113

reasonable amusement is. There is no double vision about amusement: The apprehension that is internal to amusement, such as the appreciation of a joke, is not at odds with our sense of how long it is reasonable to be amused or to laugh. There is no Proustian moment about a joke— no horror in anticipation that one will reasonably cease to be amused by it, even though the joke will continue to be funny. That is because, in learning about amusement, we learn that reasonable amusement fades very quickly, and we integrate this into our experience of the amusing. Why should things be different with grief and anger? Indeed, the point could be brought out vividly as follows. Earlier I drew on the metaphor of a lens. I said that the empirical reality of grief and anger is comparable to the features of a lens that are not in view of the lens. Moreover, when the lens turns upon itself, the focus changes, and the objects that were in focus before recede, blurry, into the background. This, I argued, gives rise to double vision.—Na’aman replies perceptively: “When I see the world through my dark sunglasses I don’t actually think the world turned dark, or that evening has fallen suddenly. My awareness of the color of my lenses informs my understanding of the images I see and of what they represent.”³¹ Why, then, shouldn’t we say that, just as we are aware of the darkening effect of sunglasses in apprehending the world, we can be aware of the empirical reality of grief and anger in apprehending loss and injustice? This a very difficult question, and I confess to some uncertainty about what to say in response. The best answer I can give is to point to the phenomenology of grief and anger and, specifically, to the force of the puzzle of accommodation in its first-personal guise: Anticipation of the end of grief or anger just is at odds with our understanding of the reasons for grief and anger. I want to insist on this as a datum about our emotions—as the Proustian insight: It is horrible that grief and anger diminish and that they do so reasonably. Indeed, that it is reasonable for grief and anger to diminish is what makes it even more horrible. The thought that I am emotionally akratic would be less hard to bear than the thought that I can’t even be expected to hold on to the loss or injustice.

³¹ In personal correspondence, for which I am very grateful.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

114   Since I take this to be an insight about the phenomenology of grief and anger, it strikes me as a condition on the adequacy of an answer to the puzzle of accommodation that it preserve the insight. Surely, Proust did not overlook something—some reasons that he failed to get in view? (Also, surely there is no way for the rakehell to have a fully integrated view of his betrayal, his guilt, and what that says about his character?) I acknowledge, however, that this is a place where some readers might part ways with me, because they differ deeply in their comprehension of the phenomenology of emotional experience.³² However, if Proust’s insight is to be preserved, we should not take the empirical insight we gain about the reasonableness of accommodation to be comprehensible in terms of our reasons. In anticipating the diminution of grief or anger, we are rightly horrified that we will accommodate ourselves (even if the horror is mixed with relief). A crucial element of what is horrifying is precisely the realization that, given the kinds of creatures we are and given the moral psychology of our emotions, such accommodation is reasonable. Proust’s insight teaches us that we must hold on to the double vision, the surdness, and the unreconciled moment, lest we conceive of our emotional life as too neat.³³ Let me now return to the issue of pragmatic encroachment. It might seem that characterizing my view as a form of pragmatic encroachment itself eliminates the Proustian insight, because it suggests that practical considerations can bear on the strength of our reasons without being reasons themselves. Thus, provided we understand our reasons, and their strength, we will be able to understand accommodation as reasonable in light of our reasons.—However, I do not think that we are forced to this conclusion. I think we can maintain that the strength of our reasons remains opaque to us. Perhaps we realize that our reasons lose their force as we complete an emotional process or as the functional role of the emotion is satisfied. However, this does not put us in a position to

³² Eli Hirsch (2018, 12), following William James (1902/1958, 326), draws the distinction between sick-souled and healthy-minded philosophers and attributes considerable significance to it. I think this is a moment in which one would have to be sick-souled to accept my argument. The healthy-minded might well beg to differ. ³³ Another example of the sort of philosophical reflection that leaves us unreconciled is Jay Wallace’s The View from Here (2013) for which I have great admiration.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.    

115

properly understand ourselves—any more than recognizing the reason for which we acted akratically puts us in a position to properly understand ourselves. We can look at it this way: People are akratic; I am akratic. People binge on Netflix shows late at night when they know they should go to sleep, because they give in to the allure of just one more episode. Indeed, I binge on Netflix shows late at night when I know I should sleep, because I give in to the allure of just one more episode. This is all too comprehensible. Yet from the agent’s point of view—from my point of view, as I am about to press, “play next episode” (never mind that if I waited just ten seconds, the next episode would start on its own!)—it is incomprehensible why I should do so, given that I know that I should go to sleep. By the same token, we can say: Reason is embodied and situated. Emotions are processes; they have functional roles. Those bear on how emotions are reasonably experienced. For example, when one has undergone the process of grief, and when one has detached oneself from the lost object, one’s reason to grieve no longer has force. This is all too comprehensible. Yet from the agent’s point of view—from my point of view, as I am equanimously contemplating my mother’s death or the murder of twenty first-graders—it is incomprehensible why I should not be breaking down in sorrow. We can understand the reasonableness of accommodation in a structurally analogous way to how we understand the reasons for akratic action: from the outside. From the outside, we understand why I binge on Netflix; I am akratic. Similarly, from the outside I understand why I accommodate myself to loss and injustice: Emotional reason is embodied. However, from the inside, I fail to understand why I am not moved by what I judge to be the better reason—why I don’t go to sleep— and I fail to understand why the loss or injustice is not a better reason than I take it to be. Such an outside view often comes naturally to us, and we may even forget about the engaged, inside view altogether. We may toggle between the views, perhaps even in ways that serve the ends of our lower selves. However, once we have understood the double vision, we will arrive at the view I have been seeking to articulate: Accommodation is reasonable, though we cannot identify reasons in light of which this would be

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

116   so—reasons that we could comprehend from the engaged perspective in which our attention is directed outward, to loss or injustice. There is, therefore, in our emotional life something essentially surd—something that must leave us unreconciled. Indeed, there is a further unreconciled moment, concerning the fact that there is an unreconciled moment in our lives. As I noted earlier (in section 4.2), I continue to have great admiration for those who are capable of grieving longer and deeper than I am, and especially those who remain outraged over injustice. I regard Simone Weil and Martin Luther King Jr as heroes. Also, I am always especially content when, unexpectedly, on a particular occasion, my grief returns, or when I manage to feel at least some of the outrage that the current social injustice warrants.³⁴ It is in those moments that I feel connected to reality—connected to the reality of loss and injustice from which my resilience reasonably but regrettably protects me most of the time. Nonetheless, I do not think it would be reasonable to feel grief and anger all the time. I do not think it would be reasonable to suffer persistent grief as DSM-5 classifies it. And I do not think it would be reasonable to be a relentlessly angry person, despite the reality of injustice in the world. Recognizing the reasonableness of accommodation no more reconciles me with loss and injustice than it reconciles me with that very unreconciledness. I may not be maladjusted, but at least I am neurotic.

4.5 An Is-Ought Fallacy? Let me now turn to an objection that some readers might have had for a while. One might think that the view I have put forward commits an isought fallacy: It infers from how we actually grieve to how we ought to grieve or, rather, how it is reasonable to grieve. Additionally, or alternatively, one might wonder why considerations about the moral psychology of the emotions affect the strength of reasons ³⁴ The contentment, of course, sullies the grief, though I am relieved to have enough selfknowledge to realize this, all the while finding myself appalled at such relief.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.  - ?

117

rather than merely providing an exculpation: For example, why say, as I have done, that accommodation is reasonable rather than say that it is irrational but excusable? In general, why is the reasonableness of grief and anger not entirely determined by the significance of loss and injustice, whereas consideration of the empirical reality of grief and anger affords exculpatory considerations? It is these questions that I want to address in the present section. As a first response to these objections, I want to draw on what I take to be an insight in Sartre’s account of freedom—namely that our body and, generally, our situation is not a limitation of our freedom but a condition for it. We are not first free, only to choose our body and our situation. Rather, our freedom is embodied: There is no freedom apart from our embodiment and our situation.³⁵ Therefore, embodiment and, generally, our situation cannot be limitations on our freedom. As it is with freedom, so it is with reason, especially emotional reason: Our embodiment, and specifically the way in which grief and anger are embodied and situated in our lives, is not a limitation on their reasonableness but a condition for it. We do not first grieve, only to be impeded in our grief by our body and situation. Rather, grief itself is an embodied and situated apprehension of loss.³⁶ That is why the reasonableness of grief cannot be understood apart from the way in which grief is embodied and situated. And the same is true for anger. In light of this Sartrean thought, I propose a first response to the objections under consideration: Saying that the empirical reality of grief and anger affects the reasonableness of grief and anger is not to commit an is-ought fallacy, because the empirical reality of those emotions is a condition for the very subject matter of the ought claims, or, rather, the claims about reasonableness. The point of the Sartrean thought is that

³⁵ Sartre writes, “the body cannot be distinguished from the for-itself ’s situation, since, for the for-itself, to exist and to be situated are one and the same” (1943/2018, 417). “[M]y finitude is a condition of my freedom, for there is no freedom without choice and just as the body conditions consciousness as a pure consciousness of the world, it also makes it possible, even in its freedom” (440). ³⁶ See Furtak (2018, esp. 68–72).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

118   the very topic of normative assessment is of a piece with, and not separable from, the factual reality of what is assessed. There is no factvalue dichotomy as would be required for the objection to get off the ground. Furthermore, we can see why the empirical reality of grief and anger should not be understood as providing exculpatory conditions for their diminution. The empirical reality of grief and anger is not an impediment to grieving or being angry which would furnish us with an excuse. Because it is a condition for grief and anger, it conditions what these emotions are and also when they are reasonable. To make this abstract argument more concrete, I would like to consider how, following Justin D’Arms and Daniel Jacobson, one might approach the puzzle of accommodation. One might hold that accommodation is one of many “obscuring factors, which generate or suppress emotional responses in ways that do not reveal the agent’s underlying sensibility” (2010, 598), and this would explain “the instability of affect” (598). It is in this way that D’Arms and Jacobson propose to understand why we don’t continue to laugh at jokes that, in fact, remain funny even once our amusement is no longer reasonable. They write: The tendency of amusement (and other emotions) to fade with repetition, for example, is not itself an expression of the sense of humor (or other sensibilities) but a feature of our dispositions that does not reveal anything about our values. This point is tacitly acknowledged by common sense when we do not treat a failure to be amused, on the tenth hearing, as grounds for doubting that the joke that amused us the first nine times is funny. We can grow desensitized to other sentimental values too upon overexposure. When we become used to even outrageous behavior, it gradually ceases to actually outrage us. No doubt this tendency has useful consequences, since perpetual outrage, amusement, disgust, or fear quickly grows counterproductive. In such cases, where one’s dispositions to amusement stop reflecting one’s sense of humor due to overexposure to the object, we can say that repetition is an obscuring factor with respect to a joke’s funniness. (2010, 597)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.  - ?

119

Especially in light of their claim about outrage, it might be that D’Arms and Jacobson would endorse the view that accommodation, whether occasioned by repetition or not, is often an obscuring factor.³⁷ However, I do not think that accommodation is plausibly regarded as an obscuring factor. That is because if it were an obscuring factor, then someone who did not accommodate herself—a persistent griever, say, or a deeply resentful person—would have a more reasonable response to loss or injustice than someone who accommodated herself in the (statistically) normal way. This person would see the relevant value more clearly, since, precisely, their view would not be obscured. Indeed, a Subresilient person would be most reasonable since, by hypothesis, the Subresilient continually grieve their losses. But that is implausible. Indeed, its implausibility is even starker when it comes to jokes: Someone who continued to laugh at a funny joke with the same enthusiastic, undiminished amusement for an extended period of time, or who repeatedly laughed at the very same joke, would not exhibit a greater sensibility than most of us. (I say this as someone who does laugh at jokes a little bit too long!) Indeed, they would exhibit a very odd sense of humor—perhaps a juvenile sense of humor.³⁸ When it comes to amusement, we are like the Super-resilient grievers! This suggests that accommodation should not be understood as an obscuring factor but must be understood as part of one’s sensibility. We don’t first experience grief or anger after which our emotional response is obscured as we accommodate ourselves. Rather, accommodation is part of the empirical reality of grief and anger. Or, to put it in the terms of D’Arms and Jacobson: Accommodation is partly constitutive of our sensibility, because how we accommodate ourselves to loss and injustice is as much a reflection of our sensibility as how we ³⁷ In effect, their view might thus be taken to be a version of the Hardline Response. However, as I note in section 3.1, since D’Arms and Jacobson hold that reasons of the wrong kind can justify, this may oversimplify their view, and they may not, in fact, endorse the Hardline Response (2000; 2009). Also, they offer two qualifications of the remark I quoted which, as far as I can tell, don’t bear on the present point, but may be intended to do so: they say that, on their view, repetition is not always an obscuring factor, and they say that sensibility is not static but changes over time. In the end, I am not sure what, exactly, they would say about the puzzle of accommodation. ³⁸ Indeed, D’Arms and Jacobson accept that a juvenile sense of humor is a distinct kind of sensibility (2010, 595). So, if they agree that persistent amusement, or amusement to a repeated joke (that one has not forgotten in the meantime), is a mark of juvenile humor, they would have to concede that accommodation is not merely an obscuring condition.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

120   experience grief or anger in the first place.³⁹ Of course, things could have been different; there could be different forms of life with a different sense of humor for whom amusement plays a different role. For them, a different form of amusement would be appropriate, a different form with longer duration. I have offered the Sartrean thought that the empirical reality of the emotions is not a limitation but a condition of their reasonableness as a first response to the objection under consideration. There is, nonetheless, an important truth that the objection concerning the is-ought fallacy reveals, and this requires some refinement to my response. Our empirical reality and our situation are changeable:⁴⁰ Thus one might hold that if an empirical feature of our embodiment or our situation conflicts with our reasons, we should change or resist that empirical feature. For example, suppose we are amused by a particular prejudice, or by a perception of physical pain or a weakness of some sort. And suppose that this is not obviously a mistake: our amusement is not over something that is not funny. Perhaps our amusement is even reasonable.⁴¹ Nonetheless, we could take the fact that our amusement is conditioned by our embodiment and our situation—and surely amusement over, say, certain kinds of physical pain is so conditioned—to be something that we should resist or change. We could think: We shouldn’t find it funny when people are clumsy, or stumble, or fail to understand something that seems so obvious to us. (Think: Scale error in 2-year-olds. It is very funny!)⁴² We could then take various steps to change what we find funny. And by the same token we might think of accommodation as something that we should resist or change. Indeed, arguably, the Stoics thought we should all be Super-resilient. The crucial point is that from the mere fact that a

³⁹ Cf. Na’aman (2021, 256–7). ⁴⁰ I am grateful to an anonymous reader for Oxford University Press for raising this line of objection, and for extensive helpful comments about it. ⁴¹ I concur with D’Arms and Jacobson in their rejection of moralism about humor, according to which the fact that it is morally wrong to be amused about something shows that amusement is not fitting. This would be to commit to the moralistic fallacy (D’Arms and Jacobson 2000), which is a version of the appeal to the wrong kind of reasons. Still, we might see moral considerations as providing us with reasons to change our sense of humor—to change our sensibility. ⁴² https://www.youtube.com/watch?v=OtngSHtz-cc

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.  - ?

121

certain emotional response is typical for someone in our situation, it does not follow that it is reasonable, or that we should consider it reasonable in a way that we should be content with having it. This is a different way in which one might raise the objection that my view commits the isought fallacy. My response to this way of understanding the objection is, again, primarily dialectical. As noted earlier, I do not take myself to have shown that accommodation is reasonable; rather, I take myself to have addressed the question of how accommodation could be reasonable, given that we do not seem to be in a position to identify reasons in light of which it would be so. Once the obstacle to seeing how accommodation could be reasonable is removed—that is, once we accept the possibility of reasonableness without reasons—we can revert to the original intuition that accommodation is reasonable. I do not propose to explain which empirical features of grief and anger—whether their functional roles, or the fact, if it is one, that they are processes—would account for their reasonableness. Saying this is compatible with granting that there is a gap between is and ought—that from the mere fact that we typically experience an emotion it does not follow that our experience is reasonable. The point is just that we should not conceive of the reasonableness of emotional experience entirely apart from how they are actually experienced and that, given the original intuition that accommodation is reasonable, our inability to identify reasons in light of which it would be so does not preclude us from taking the intuition to be correct. To this, one might want to reply: Why shouldn’t we strive to become more like the Super-resilient or, alternatively, more like the Sub-resilient? Why shouldn’t we seek to change the empirical reality of grief and anger? However, these questions are only indirectly questions about the reasonableness of emotions; they are primarily questions about the reasonableness of actions concerning our emotions, which should be clearly distinguished. They are questions of will, not of emotion.⁴³

⁴³ See section 2.1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

122  

4.6 A Sartrean Conclusion In this chapter, I have argued that emotions have a twofold reality: They are, at once, reasons-responsive apprehensions of the world as well as contingent, empirical conditions of ours that depend on our embodiment and, in general, on our situation. They are forms of embodied reason. This twofold reality of the emotions affords two kinds of misrepresentation. One could either deny that emotions are reasons-responsive or one could deny that they are forms of embodied and situated reason, that is, deny that the ways in which they are embodied and situated bears on their reasons-responsiveness. This, in effect, is what the Antirationalist Response and the Hardline Response do. These two responses thus make mistakes that are structurally analogous, insofar as they misunderstand something about the very nature of the emotions. Indeed, I want to suggest, these two responses correspond to the two paradigmatic kinds of Sartrean bad faith.⁴⁴ Sartre famously describes two paradigmatic ways of being in bad faith.⁴⁵ The first is to treat oneself as an object or, to put it in Sartre’s terms, to identify oneself with one’s “facticity”—like a gambler who thinks that the fact that he’s always abandoned resolutions to stop gambling before shows that he will likely gamble again, because he is, after all, a gambler. The second is to ignore one’s facticity and identify oneself entirely with one’s freedom—like a gambler who takes his gambling history to be irrelevant to the question of whether he will gamble in the future. Sartre’s concern, in discussing bad faith, is primarily with agency.⁴⁶ However, we can readily see that both kinds of bad faith are available with regard to our emotions. We can treat our emotions as objects too: We do so when we take a theoretical view of them—when we see them as conditions that befall us or as processes that we undergo. In so doing, we ⁴⁴ For discussions of grief in a Sartrean vein, see Rinofner-Kreidl on Sartrean bad faith (2018, 206–7), as well as Solomon (2007, 77; 2004, 80) and Higgins (2019, 319) on Sartre’s phenomenological reflection on the nature of absence. For a related discussion in the context of Merleau-Ponty’s work, see Ratcliffe (2020). ⁴⁵ See Sartre (1943/2018, pt.1, ch.2) and Moran (2001, esp. 77–83). The example of the gambler occurs in the context of anguish. I use it here, as is often done, in place of Sartre’s highly problematic example of the homosexual. ⁴⁶ See Marušić (2013; 2015) for extensive discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.   

123

deny or misrepresent our freedom. Of course, this is not the freedom we enjoy in action; it is not freedom of the will. But it is freedom nonetheless, because our emotions don’t just happen to us but are our active responses to reasons. Indeed, this is especially plausible if to be free just is to be reasons-responsive (Wolf 1990). Yet we can also exhibit something like the second kind of bad faith: We can disregard the facticity of our emotions. We do so when we deny or misrepresent the way in which our emotions are embodied and situated, and, specifically, when we deny or ignore the bearing of their empirical reality on their reasons-responsiveness. In effect, this is what the Hardline Response does. In misrepresenting the empirical reality of grief and anger, it indulges in an unrealistic sense of freedom. On Sartre’s view, as I understand it, the dilemma of bad faith is intransigent: We cannot comprehend ourselves as freedom and facticity at once. Our freedom and our facticity do not allow for reconciliation; the best we can do is to coordinate them.⁴⁷ This is why we suffer from ineliminable double vision: We apprehend ourselves as free and, therein, attend to the world and respond to our reasons—be it our reasons for action or emotion or belief. And we apprehend ourselves in situation, as creatures with a particular psychology, physiology, history, culture, and social circumstances, who, as an empirical matter of fact, act, feel, and believe in certain ways and for certain purposes—and therein exhibit a kind of alienation from ourselves. But we cannot bring these two views into one reconciled self-consciousness: whenever we attend to one of them, the other is, as it were, blurry in the background. I conclude, therefore, that the puzzle of accommodation is a particular manifestation of general predicament—namely that we are free yet material, rational yet embodied. The puzzle of accommodation is one of the many ways in which it is puzzling how we could be—as we surely are—subjects and objects at once. ⁴⁷ Sartre writes: “What unites these various aspects of bad faith? It is a certain art of forming contradictory concepts, i.e. concepts in which an idea and the negation of that idea are united. The underlying concept generated in this way makes use of the twofold property of human beings, of being a facticity and a transcendence. These two aspects of human-reality are, in truth—and ought to be—capable of being validly coordinated” (1943/2018, 98–9). See Marušić (2015, sect. 6.4.2) for some further discussion of this passage, though my present way of thinking about the dilemma of bad faith differs somewhat from the earlier one.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

PART II

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

5 Endless Love It is not an empirical question, for me, how long I will love my children. Indeed, this is not a question for me at all. My love for them is endless. And this is not because I am a particularly good parent, which I may well not be, but because this is what love for children is like. However difficult such love may be, a parent’s love is endless. It is not an empirical question, for me, how long I will love my spouse. Indeed, this is not a question for me at all. My love for her is endless. And that is not because I am a good spouse, which I may well not be, but because this is what married love is like. In loving my spouse, the question of how long I will love her is not open. That is not because the question would be out of place altogether: Love is in time; it is temporally extended. It is because, in its self-consciousness, married love is endless. That is to say, in loving, I implicitly understand my love to have no end. Many people solemnly promise that they will love someone for the rest of their life. The phrase in a Catholic wedding—a phrase I spoke some years ago—is, “I will love you all the days of my life.” One might wonder: What is their evidence for such a prediction? Isn’t such a claim perhaps irrational?—Yet this is not an empirical prediction. Since the question of whether you really love someone is not separate from the question of how long you will love them, the claim about the duration of love is not separate from the claim about the love. To say that you will love someone all the days of your life is to say nothing more than that you really, maturely love the other. Nonetheless, the problem with this line of thought is apparent: There are, as an empirical matter of fact, innumerable purported counterexamples to the claim that love—even real, mature love—is endless. Especially when it comes to “romantic” love, hearts are broken, love is betrayed, and people grow tired of each other. Love sometimes, all too often, ends. On the Temporality of Emotions: An Essay on Grief, Anger, and Love. Berislav Marušic,́ Oxford University Press. © Berislav Marušic ́ 2022. DOI: 10.1093/oso/9780198851165.003.0005

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

128   Love is eternal while it lasts, as Vinicius de Moraes quipped.¹ The selfconsciousness of love as endless may strike us as an illusion. This, then, confronts us with a puzzle. At the risk of being misleading, I will call it the puzzle of endless love. The puzzle is this: How can we understand our love as endless, and do so without being irrational, if, empirically, we know that it may not be so? And if the self-consciousness of love is endless, how, if at all, can we anticipate or even think about a possible end to love?² These formulations of the puzzle are meant to be analogous to the formulations of the puzzle of accommodation. However, unlike the puzzle of accommodation, which concerns the diminution of grief and anger, the present puzzle concerns the continuation of love. What is puzzling is how, once we have learned about the many ways in which love might end, we can still understand love as endless, and reasonably so. As in earlier chapters, my discussion will proceed from a phenomenological observation, namely that in its self-consciousness, love is endless—that is to say, that, in loving, we understand (or at least think)³ of our love as endless. This seems to me to be the case when it comes to different kinds of love, including mature love for a spouse and love for one’s children, and extending to some familial relations and friendships. I take it that this observation reveals something deep and interesting

¹ Moraes is known as the author of the lyrics for The Girl from Ipanema. As far as I understand, the quip derives from the last line of his Sonnet of Fidelity, which ends with the exhortation, addressed to love, to “be infinite while it lasts” (available at https://allpoetry. com/Sonnet-of-Fidelity). Interestingly, Moraes’s actual line is deeper than the quip, since it suggests a longing for an experience of the kind of love that, in its self-consciousness, is endless, all the while conceding that it is an illusion. ² In their article “Love and Time,” Aaron Ben-Ze’ev and Angelika Krebs (forthcoming) also consider whether love could be understood as endless (or eternal). They conclude that it can, albeit they do so on empirical grounds. I am less sanguine that a purely empirical consideration of the question can be satisfactory: To understand the temporality of love, we must understand love’s “inner temporality”—a notion I will explain shortly. Nonetheless, I hope the reader will indulge my quoting the concluding words of their article: “The taste of love is marvelous when love is fresh; however, love may become more nutritious with time. As with food, both taste and nutrition are important, but if you wish to live longer, or to have a longer romantic relationship, the value of nutrition cannot be ignored. Thus, romantic love is not necessarily best when it is fresh” (forthcoming, 19). ³ I add the parenthetical to account for the “bad case,” that is, the case that is posterior in explanation. (More on that shortly.) By “understanding,” I mean the sort of implicit, unreflective understanding that Sartre would call “non-positional consciousness” (1943/2018, esp. the introduction). See Boyle (2018; 2019; forthcoming) for illuminating explanation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.     

129

about love—though, of course, various kinds of love differ in various other ways. (Obviously, love for a spouse is different from love for children!) In what follows, I will provide a rationale for this observation when I explain what I shall call the “inner temporality” of love. Nonetheless, in this chapter, no less than in previous chapters, I will have to count on readers to share, or at least sympathetically comprehend, this observation. Those who don’t do so, or who think that the selfconsciousness of love is yet another example of human irrationality and delusion, may regard what follows as an articulation of an amusing, or bemusing, romantic rationalism. I will proceed as follows: In section 5.1, I will clarify the puzzle I just sketched. In section 5.2, I will explain what I mean by “inner temporality”—the observation that we understand the duration of a mental state through an apprehension of the reasons for being in that mental state. In sections 5.3 and 5.4, I will consider two ways of explaining the inner temporality of love that would account for why, in its selfconsciousness, love is endless. In section 5.5, I will consider how to reconcile the inner temporality of love with its “outer” temporality— with empirical reflection on love, especially if love is understood as an instance of embodied reason akin to grief and anger. I will argue that even though empirical reflection on love reveals ways in which love might fail, commitment and trust can provide a scaffolding that precludes an anticipation of the ways in which love might fail. Even if love sometimes fails, this is not something a lover can bring into view without a problematic form of alienation.

5.1 The Puzzle of Endless Love I have claimed that in its self-consciousness, love is endless. The question of how long we will love is not separate from the question of whether we really love. In affirming that we love, we understand, or think that, our love will have no end. It is this phenomenological observation that drives the puzzle; the thought that love might end needs no defense. In this section, let me consider in a bit more detail why the observation may seem plausible.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

130   To feel the force of the observation that, in its self-consciousness, love is endless, consider thinking: “I love you, but how long will my love last?” The question, if imagined as a genuine question, is in some way infelicitous. Asking “How long will my love last?” is in tension with affirming “I love you.” It is a bit like thinking, “I believe p, but will I still believe it tomorrow?,” which, in turn, is a bit like thinking, “I believe p, but is p really true?”—since the possibility of (rationally) changing my mind about p tomorrow seems to concede that p might not be true after all. We can make sense of the question of how long our love will last, much like the question of how long our belief will last, as a question that is asked from outside, by someone else, and is therefore comprehended apart from the affirmation of love. The question concerning the duration of love is a question that a disinterested observer might ask, and we can ask the question ourselves to the extent that we take on the position of an observer of our own mental state.⁴ Let me add some further clarification. Despite affirming one’s love, one might wonder how long the feeling of love will last—the feeling of amorousness or infatuation that is characteristic of fresh love. Yet this is not what gives rise to the puzzle of endless love. Such a question may be somewhat odd: When one is fully engrossed in the feeling of fresh love, asking this question reveals that the feeling is already past its first bloom. However, it is nonetheless intelligible to wonder how soon the freshness of one’s love will change—how soon it will mature. Yet the puzzle of endless love is not a puzzle about the freshness of love or the duration of infatuation. By the same token, there may be other forms of love—one-sided love, unrequited love, broken-hearted love, betrayed love, and perhaps uncommitted love—which may not be, in their self-consciousness, endless. Indeed, one may well as the question how long such love will last. Yet the puzzle of endless love, as I conceive of it, is not primarily about such cases. Rather, the puzzle arises about the “good case”

⁴ Here, as everywhere, my thinking is indebted to Moran (2001).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.     

131

(to borrow a term from epistemology)⁵—a kind of love that we may sometimes feel for another person, a kind of love that is uncompromised, mature, committed, and trusting. (Kierkegaard describes married love as endless in this sense;⁶ this strikes me as apt, but I think that other forms of love that are not “romantic” can be endless in the relevant sense as well.) I think that in its self-consciousness, uncompromised, mature, committed, and trusting love is endless as such love. My aim is to make sense of why we would apprehend such love as endless and why this may be reasonable. Eventually, in section 5.5, I will also say something about the “bad case.” Finally, I shall leave open whether the arguments extend to love that has, as its objects, something other than a person; my interest here is interpersonal love—an emotional relation between subjects.⁷ I have introduced the puzzle of endless love as a new puzzle. However, one might wonder whether it really is distinct from the topic of the preceding chapters—whether it is not just a version of the puzzle of accommodation. One might wonder whether the puzzle of endless love is, quite like the puzzle of accommodation, primarily a reflection of a tension between the standpoint that is internal to an emotion and a ⁵ I borrow talk of the “good case” and “bad case” from Williamson (2000), who uses it in the context of skepticism (cf. McDowell 1982/1998; 1986/1998). The “good” case is good, not so much because it is preferable but because it is primary in the order of explanation. For Williamson, the “good case” is one in which we have knowledge, and it is prior in the order of explanation (hence his slogan, “Knowledge First”) to the “bad case” in which we are deprived of knowledge. The methodology of analyzing a phenomenon in terms of the good case and understanding the bad case as a case of privation, rather than analyzing a phenomenon in terms of the bad case plus the satisfaction of various other conditions, strikes me as especially apt in an effort to understand interpersonal phenomena—since the bad case is often bad in virtue of there not being a genuinely interpersonal relation. ⁶ See The Esthetic Validity of Marriage in his Either/Or: “The faithful romantic lover waits, let us say for fifteen years; then comes the moment that rewards him . . . A married man is faithful for fifteen years, and yet during these fifteen years he has had possession . . . [The married man] has not fought with lions and trolls but with the most dangerous enemy, which is time. But now eternity does not come afterward, as for the knight, but he has had eternity in time, has preserved eternity in time” (Kierkegaard 1843/1988, 138). ⁷ One might object that love is not an emotion. Indeed, it might be thought that my puzzle actually illustrates this: Emotions, especially if understood as feelings, diminish. The fact that love does not diminish over time shows that love is not a feeling and, therefore, not an emotion.—I don’t mean to insist that love is an emotion. Indeed, I would have no grounds for such insistence since I don’t offer a theory of the emotions. Yet even if it is granted that love is not an emotion, it is still intelligible and worthwhile to ask: Why is love, in its selfconsciousness, endless, and how could this be reasonable? If the answer is that love is not an emotion, we might want to know what, then, makes it different from the emotions.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

132   theoretical standpoint on that emotion from which that emotion is considered empirically. Indeed, it might be thought that the tension is the same as the tension that gives rise to the puzzle of accommodation: After all, asking how long one’s grief or anger will last, while grieving or being angry, may seem infelicitous in the same way in which asking how long one’s love will last is. This may be taken to suggest that, in their selfconsciousness, grief and anger are also endless. It is true that there is a common element in the puzzle of accommodation and the puzzle of endless love. In its naïve self-consciousness, that is, before we learn more about it empirically, grief may strike us as endless. That is why it is so apt to think of grief as the continuation of love. However, at this point the puzzles diverge. In the case of grief, we have to understand how, in light of what we empirically learn about the emotion, the naïve self-consciousness of grief as endless is mistaken. In the case of love, we have to understand how, in light of what we empirically learn about the emotion, the self-consciousness of love as endless could be correct. It is this difference that I will eventually seek to explain. Before I am in a position to do that, however, I first would like to offer an explanation of why, in its self-consciousness, love would strike us as endless in the first place.

5.2 Inner Temporality To get a better understanding of why, in its self-consciousness, love should be endless, let me consider in this section the temporality of some other mental states—in particular, of pain and belief—in order to bring out what I call the inner temporality of a mental state. In due course, I will then return to love. Consider the realization that you are afflicted by a discomfort in the abdominal area—say, an irritation of the gallbladder. It is one thing to know that you are suffering abdominal pain and another thing to know how long it will last. In general, the temporality of pain is not internal to the self-consciousness of pain: Experiencing pain, and understanding that one is in pain, leaves open how long the pain will last. We only learn

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.  

133

about its duration once we classify the pain and perhaps come to understand its causes on empirical grounds. We can contrast pain with rational belief. When we rationally believe something, the question of how long we will believe it is, for us, not an open, empirical question. This is because, when we rationally believe something, any temporal limitation we anticipate is reflected in the content of what we believe, rather than in our understanding of the temporality of the mental state of believing. For instance, we would conclude reasoning in believing, “It is sunny today,” rather than in concluding that today, we will spend (some of) the day believing, “It is sunny.” Only in special cases, which I will address shortly, does the question concerning the duration of belief arise apart from the content of what is believed, and in those cases a belief ’s rationality is compromised. An explanation for this can be given by appeal to the transparency of belief—the observation that, in believing something, our attention is on the content of our belief, rather than on the psychological fact that we have this belief. To develop this explanation, let me consider Moran’s account of transparency. Moran explains transparency as a relation between questions: Questions about belief are transparent to questions about their contents. That is to say, “a first-person present-tense question about one’s belief is answered by reference to (or consideration of) the same reasons that would justify an answer to the corresponding question about the world” (2001, 62). We can illustrate this with a well-known example from Gareth Evans (1982, 225): One’s belief is transparent when one settles the question of whether one believes that there will be a third world war by considering the question of whether there will be a third world war and, in particular, by considering the reasons that show it to be true or false that there will be a third world war. Transparency, as Moran explains it, is not a necessary truth but a normative requirement. On Moran’s view, transparency “looks more like a kind of normative ideal” (62).⁸ He writes,

⁸ See Boyle (2019; forthcoming) for development of Moran’s position in the spirit of Sartre (1943/2018), which I find congenial.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

134   The vehicle of transparency . . . lies in the requirement that I address myself to the question of my state of mind in a deliberative spirit, deciding and declaring myself on the matter, and not confront the question as a purely psychological one about the beliefs of someone who happens also to be me. (2001, 62 3)

When a person’s self-attribution of a particular belief fails to conform to the transparency condition, they are, in a salient sense, alienated from what they think. This sense of alienation reflects the fact that transparency represents a standard of rationality for reflective creatures like us: If we self-attribute a belief merely as an empirical hypothesis about ourselves, rather than on the grounds for actually so believing, the belief is detached from its grounds and, in that way, fails to be rational.⁹ This account of transparency is well-suited to explain why, when we rationally believe something, any temporal limitation we anticipate is reflected in the content of our conclusion rather than in our understanding of the temporality of the belief itself: When a belief is transparent there is no room for a question about the temporality of the belief that is separate from a question about the temporality of its content. If, in settling whether we believe something, we consider only what is true, then the temporality of our mental state—a psychological fact about us— is simply not a topic of our reflection. Rather, our belief reflects what we take to be so. And what is so is—endlessly so. In its self-consciousness, belief is endless, because truth is eternal. The self-consciousness of belief is a reflection of the eternity of truth, which belief is an apprehension of. This is why, when we rationally believe something, the question of how long we will believe it is, for us, not an open, empirical question. If the question were to be open, we would be in violation of transparency, and our belief would be irrational. Let me consider two objections—one from the philosophy of language, the other from epistemology: First, does my argument rest on an eternalist, as opposed to temporalist, conception of propositions (which I take to be the contents of belief and bearers of truth values)? Eternalism is the view that true propositions are eternally true, whereas ⁹ For further discussion of its irrationality, see Marušić and White (forthcoming).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.  

135

temporalism is the view that propositions are time-neutral. To illustrate: On an eternalist view, when one believes (something that would be expressed by saying)¹⁰ “It is sunny,” the proposition believed has to be supplemented with a time, a place, and a possible world, so that what one, in fact, believes is an eternal truth. In contrast, on a temporalist view, the proposition believed is merely, “It is sunny,” and the context supplies time, place, and a possible world, so that the proposition believed is not endlessly true but is only true in a context of utterance (or, possibly, of assessment).¹¹ Does my account of the endlessness of belief still work even on such an account of propositions?—The issue is more complicated than I can do justice here. However, I think that the argument I gave succeeds even on a temporalist view of propositions. This is because when one believes something like, “It is sunny,” one implicitly understands that what one believes is true in one’s particular context. It is not as if one merely believed that it is sunny, while leaving it to whatever context one happens to find oneself in to determine, from outside of one’s understanding, what place, time, and possible world one’s belief is about.¹² One implicitly understands that one’s belief concerns a particular place and time, as well as the actual world—or else one couldn’t, in believing something, take it to be true; one would have to first find out about the place, time, and world that one’s belief is concerned with. Thus, even if features of the context don’t supplement the proposition, they are nonetheless understood in understanding one’s belief.¹³ Hence, even if propositional truth is not eternal, truth in a context is eternal, and, since truth in a context is within one’s understanding, it accounts for the self-consciousness of belief as endless. Therefore, the shift from eternalism to temporalism does not affect my argument—though the argument is more straightforward on an eternalist view of propositions.

¹⁰ I will omit the parenthetical for ease of exposition in what follows. ¹¹ For discussion of contexts of assessment, and, generally, of the semantic issues I treat here only very briefly, see MacFarlane (2014). ¹² Here I am grateful to Sebastian Rödl for discussion. ¹³ This is another place where Sartre’s notion of non-positional consciousness could be illuminating. Once again, see Boyle (2018; 2019; forthcoming).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

136   I turn to the second objection. One might deny that, in its selfconsciousness, rational belief is endless, because one might hold that it is possible rationally to believe something yet expect that one will receive new evidence and, therefore, anticipate changing one’s mind. Furthermore, whether one will receive new evidence about a proposition is an open, empirical question. Thus, if one takes one’s rational belief to be responsive to evidence, it should be possible for it to be an open, empirical question how long one will rationally believe something to the extent that it is an open, empirical question what new evidence one will receive. Yet I do not think that this is right. If we are rational, expectations concerning new evidence do not inform our understanding of the temporality of our belief but rather inform the strength of our doxastic attitude: If we expect new evidence on a question to emerge, the rational doxastic attitude will be to withhold judgment, or believe something to a lesser degree, until the evidence has come in. To the extent that we think the evidence is out on a question, to that extent we can’t have an all-out rational belief on the matter.¹⁴ This is why asserting, “p, but I might change my mind tomorrow” is infelicitous. I concede that this is a substantial claim in epistemology and there is room for argument. However, since I think that the analogy between belief and love holds only in certain respects, I won’t pursue the point but will rather return to the topic of love. It is not straightforward to extend the explanation of why, in its selfconsciousness, belief is endless to an explanation of why, in its selfconsciousness, love is endless. There are significant points of disanalogy. For one thing, it is not clear what the analogue of truth is in the case of love. For another thing, it is not clear what the reasons for love are. (I will return to both points in due course.) However, there are also three less obvious and more interesting points of disanalogy. ¹⁴ An argument for this can be given by appeal to van Fraassen’s controversial principle Reflection (1984; 1995), according to which one’s current belief should be constrained by the beliefs one believes one will have in the future. Otherwise, van Fraassen argues, one is vulnerable to a Dutch Strategy. Though the appeal to the Dutch Strategy is an appeal to the wrong kind of reasons (Kelly 2002), and though van Fraassen’s own formulation of Reflection is vulnerable to counterexamples, it seems plausible that a revised version of the principle is defensible. For state-of-the-art discussion, see Briggs (2009).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.  

137

The first point of disanalogy could be captured in terms of a slogan: Love is commitment in good times and in bad; belief is commitment only in good times.¹⁵ The crucial point is this: Whereas it is reasonable for belief to dissipate in light of countervailing evidence, it is reasonable for love to continue, despite the analogue of countervailing evidence. When times are bad, that is, when there seem to be reasons for belief to dissipate, one should let go of one’s belief. In contrast, when times are bad, that is, when there seem to be reasons for love to dissipate, one should hold on to one’s love. This is why the slippage of love could rightly be felt with apprehension, whereas one can’t rightly feel the same in light of the slippage of belief.¹⁶ A second important disanalogy is this: A belief that, in its selfconsciousness, is endless concerns a truth that will come to be in the past: When today I believe that it is sunny, and I see no end to my belief, the belief will concern a fact about the weather today, not tomorrow. Tomorrow, it will be a belief about the past. In contrast, when I see no end to love, the love I foresee concerns the beloved not only today but also tomorrow. Tomorrow it will be a love of the same person tomorrow. Thus, even if love and belief are enduring, belief concerns a temporalized proposition,¹⁷ whereas love concerns a temporally continuing person. A third disanalogy, connected to the first two, is that the selfconsciousness of love includes a sense of love’s duration in a way that the self-consciousness of belief does not. In loving, we implicitly understand something about the duration of our love. As Sandy Diehl puts it, “love presents itself as persisting into the future, or involves a projection of oneself into the future, whereas belief does not.”¹⁸ This is not to say that, in its self-consciousness, belief is not endless or that it is atemporal: The question of how long we will believe something, when we believe it, is not open. However, in the case of love, unlike in the case of belief, it makes sense to say something about the duration of one’s love—to say ¹⁵ I owe the latter slogan to Baron Reed. ¹⁶ Perhaps matters are different with faith, which is not my topic. ¹⁷ Or, if you are a temporalist, a proposition at a time. ¹⁸ In personal correspondence, for which I am grateful. In a similar vein, Sartre writes, “It is not enough to say, in effect, that a love ‘has’ a future, as if the future were external to the object it characterizes: rather, the future is part of the organized form of the flow that is ‘love,’ because its being in the future is what gives love its meaning as love” (1943/2018, 236).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

138   that one will love the other all the days of one’s life—and, in that way, to characterize one’s love (as mature, committed, trusting). And this may be a reflection of the fact that love is commitment in good times and in bad, and that it is a relation to a temporally continuing person. In light of these disanalogies, I propose to extract only a minimal conclusion from my consideration of belief: The case of belief reveals what I shall call the inner temporality of a mental state. The inner temporality of a mental state is not apprehended as an empirical insight about that mental state; rather it is apprehended in the apprehension of the object of that mental state and the reasons for which one is in the mental state. Thus, I shall argue, the inner temporality of love is apprehended in the apprehension of the beloved and the reasons for loving them. And this, in turn, accounts for why, in its self-consciousness, love is endless.

5.3 The Inner Temporality of Love I: Love is Rationally Self-Propelling There are two ways to explicate the inner temporality of love so as to explain why, in its self-consciousness, love is endless. Although I was initially attracted to the first way (Marušić 2021), I will argue here that it is unsatisfactory, and I will defend the second way. In this section, I will articulate the view that, in its self-consciousness, love is endless because love is rationally self-propelling. In the next section, I will then turn to my preferred view, which is that, in its self-consciousness, love is endless because love is the apprehension of an individual who is endless. I turn to my first proposal. One might think that the inner temporality of love, quite like the inner temporality of belief, is apprehended in the apprehension of the reasons for loving. One could further hold that we understand love to be endless, because, in loving, we understand that our love can rationally sustain itself over time. In a slogan: Love is rationally self-propelling. This thought presupposes that love is rational, in the sense of reasonsresponsive. To conceptualize love in this way is, however, controversial.¹⁹ ¹⁹ Frankfurt (2004) denies that love is reasons-responsive. Velleman (1999), Kolodny (2003), Jollimore (2011), and Abramson and Leite (2011; forthcoming) affirm it. With a heavy heart

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

139

The main consideration that I take to speak in favor is that it is apt to ask why one loves someone, in the sense in which the question asks for reasons rather than merely for an explanation. Love is not like a fever—a condition that merely befalls us—but is rather an intelligent attitude, a way of apprehending someone or something in the world.²⁰ In fact, this is the criterion I put forward earlier in my general account of the rationality of emotion (section 2.2). The main consideration that speaks against the rationality of love is that, even if the why-question finds application in the relevant sense, one is often at a loss as to how to answer it.²¹ It is hard to identify what the relevant reasons would be. It is hard to say why one loves someone, and even the things that one can say don’t seem quite right: It does not seem that the beloved’s qualities are really the reasons for love.²² For example, if you love someone for their sense of humor, then why not love everyone else with a similar sense of humor, and why not love others with an even better sense of humor even more?²³ If you love someone for the exact combination of qualities they exhibit, your interest in that very combination seems more like a kink than love (Velleman 1999, 370). Also it makes the continuation of love through change unintelligible, and it seems to leave open the possibility of rationally loving someone’s Doppelgänger—a distinct person who has all the same qualities. Indeed, to conceive of love as reasons-responsive seems objectionably economic. It seems, as a friend put it to me, “ridiculous and faintly odious.” (I suspect some understatement on his part.)

I must part ways with Sartre who says that love is a state and likens it to an illness and is something that “happens to me” (1943/2018). However, he also says this: “A love, or an undertaking, is the organized unity of these three dimensions [i.e. a Past, a Present, and a Future]” (236). I think his conception of love as an ‘undertaking’ or a project is in tension with his conception of love as something that happens to us. My present view is to be taken in the spirit of his remarks about the former, not the latter. ²⁰ A friend objects that sometimes love is a fever. I reply: In that case it is a rational fever. ²¹ Frankfurt puts forward a different consideration against the reasons-responsiveness of love. He writes, “Love is not a conclusion. It is not an outcome of reasoning, or a consequence of reasons” (Frankfurt 2006, 25). For discussion of the same sorts of considerations about reasoning in the context of grief and anger, see section 2.3. What I say there about grief and anger carries over to love. ²² For criticism of the quality theory, which I draw on here, see especially Velleman (1999) and Kolodny (2003). For a sophisticated defense of the quality theory, see Jollimore (2011). ²³ Jollimore (2011, ch.1) offers an excellent discussion of this and related problems for the “quality” view of the reasons for love.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

140   For now, let us set these concerns aside; I will return to them in due course. It is instructive to see that conceptualizing love as reasonsresponsive makes available a way of explaining why love is in its selfconsciousness endless. The key idea is that love is rationally selfpropelling: In loving someone, we have reason to continue loving them. This is why, in apprehending our reasons for loving, we apprehend something about the duration of our love, and we do not stand in need of empirical information about how long our love will last. In this way, love differs from a condition like pain, which in itself provides no indication of its temporality. Moreover, we can also comprehend love’s inner temporality as different from belief and from the other emotions: What is distinctive of our apprehension of the inner temporality of love is that love has the rational resources to sustain itself over time. In this way, even though love and belief are ongoing, love, unlike belief, remains a relation to (someone in) the present rather than coming to be, like belief, a relation to (a proposition concerning) the past. In what follows, I want to describe two accounts of love and its reasons that one could draw on to advance the thesis that love is rationally selfpropelling. However, I will then explain why I find them unsatisfactory. The two accounts in terms of which one could explain love as rationally self-propelling are the following: First, one could take love to be a reason for itself. Second, one could take love to be generating reasons for itself. We can find the first line of thought in Niko Kolodny’s theory of love, and the second one in Kate Abramson and Adam Leite’s.²⁴ On Kolodny’s view, love of a person consists in rationally valuing a relationship together with the person one is relating to. Kolodny writes, “love consists (a) in seeing a relationship in which one is involved as a reason for valuing both one’s relationship and the person with whom one has that relationship, and (b) in valuing that relationship and person accordingly” (2003, 150). He furthermore argues that a reason for love is, precisely that very relationship. He writes, My claim, then, is that the presence of an established, ongoing friendship or romantic relationship—understood, in part, as a history of ²⁴ A third account of love as rationally self-propelling is hinted at in Nehamas (2010, ch.3).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

141

shared concern and activity, and, in part, as one’s friend’s or lover’s present disposition to perpetuate this concern—can be a reason for one’s present concern, a concern that constitutes the relationship going forward. (163)

This implies that love is a reason for itself: Valuing a relationship, together with the person one is relating to, itself constitutes a strong reason for continuing to so value. Kolodny writes, The history of an established relationship with a person, and the fact that she continues to reciprocate one’s feelings, . . . constitute a normative reason for one’s present emotional vulnerability. Why does it make sense for you to be so concerned, here and now, about what happens to that woman in particular? . . . Because she is your friend or life partner. (163)

Kolodny’s view offers an elegant solution to the problem of the constancy, or continuation, of love: Even if the person we are relating to loses their attractive qualities, our relationship with them is ongoing, and we thus retain our reason for loving them. Following Kolodny, we could say that love is rationally self-propelling, insofar as it consists in the persistence of a valuable relationship—and love is, in its self-consciousness, endless insofar as one understands oneself to be in such a relationship and understands the relationship to provide one with reasons to continue loving. Kate Abramson and Adam Leite (2011; forthcoming) also offer an account of the reasons for love that affords an understanding of love as rationally self-propelling. However, unlike Kolodny, they don’t take love to consist in valuing a relationship. Instead, they argue that love consists in “an affectionate attachment to another person” (2011, 677), and they argue that one acquires new reasons to love through being in the state of having such an attachment.²⁵ They write,

²⁵ For a sophisticated conception of “romantic” love in terms of attachment, see Wonderly (2016). Though I cannot do justice to Wonderly’s account, it seems to me that the attachment

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

142   Insofar as I love people who are worth loving, I gain genuine new reasons to seek their company, express affection for them, pursue their welfare, and the like. (16) Some of these new reasons are state-based. For example, love may give me a new sufficient reason to do my beloved a favor. Some are relationally—based—reasons which arise only in the context of a loving relationship. My reasons for marking a friend’s birthday, for instance, arise partly from the fact that he would be hurt if I, his friend, didn’t. (17)

Abramson and Leite emphasize that the reasons one has for continuing to love someone are not the same as the reasons for coming to love them in the first place. Rather, love generates new reasons for loving. For example, they point out that “the fact that someone who is in a reciprocal love relationship with you would be badly hurt if you ‘traded up’ is itself a relational reason not to do so” (13). Hence, on their account, even though love is not a reason for itself, as it is on Kolodny’s view, love generates reasons for itself. This is another way to understand love as rationally self-propelling, and, therefore, the inner temporality of love as endless. Abramson and Leite’s account of the reasons of love is more inclusive than Kolodny’s, insofar as they countenance all sorts of reasons for love—including reasons having to do with the beloved’s qualities.²⁶ Insofar as it is more inclusive, their view thus does a better job of explaining features such as love at first sight or the beginning of love. she speaks of characterizes merely certain relationships of love—most prominently the love of an infant for a parent and in some instances the love of a “romantic” partner. However, it does not seem to me to be true of love as such. ²⁶ A significant difference between Kolodny’s and Abramson and Leite’s views concerns the strength of the reasons that love generates. On Kolodny’s view, love is an insistent reason for itself; that is to say, loving renders the absence of love inappropriate (2003, 163). In contrast, on Abramson and Leite’s view, the reasons for love that love generates are merely non-insistent or, in their terminology, “warranting” rather than “requiring” reasons; that is, these reasons justify love but do not imply that the absence of love is unjustified (2011, 4).—My hesitation to take a stand on this issue stems from the suspicion that this distinction covertly introduces the notions of permissibility and obligation into the domain of reasons. However, reasons—that is, considerations which count in favor of a conclusion—are neither as strong as obligations nor as weak as permissions, but are a different category of normative assessment altogether (cf. Wallace 2019, 15). See also sections 2.2 and 3.1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

143

Nonetheless they, too, can explain the constancy of love in terms of the reasons that are generated in a particular loving relationship. However, I do not propose to adjudicate here between Kolodny’s and Abramson and Leite’s views. Rather, I will argue that neither account provides a satisfactory explanation for the thesis that love is rationally selfpropelling. The problem with Kolodny’s account should be familiar from the earlier discussion of the rational significance of relationships.²⁷ Like Cholbi’s account of the reasons for grief, Kolodny’s account of the reasons for love confuses a background condition for a reason with the reason itself: The presence of a relationship is no more a reason for love than the loss of a relationship is a reason for grief. Indeed, the presence of a relationship is even less a reason for love than the loss of one is a reason for grief, since the loss of a relationship may be a secondary reason for grief, whereas the presence of a relationship does not strike me as even a secondary reason for love. The significance of relationships is that they are conditions in virtue of which we have reasons; they are not reasons themselves. Moreover, plausibly, relationships bear on the strength and significance of our reasons: they can be intensifiers in Dancy’s sense (2004, ch.3, esp. 41–2; cf. Jollimore 2011, 114). In his trenchant criticism of the view that relationships constitute reasons of partiality, Simon Keller brings out this point in an intuitive way: [A] person who characteristically thinks of her relationships when she acts well toward others is not someone you would want as a friend or a loved one. A friend who is always thinking of improving your friendship, a colleague whose main concern is with the value of collegiality, a parent who thinks mainly of how important it is to have a good relationship with his child—all of these characters are annoying to have around, and all of them seem to be missing what really matters in their relationships. In a relationship with such a person, you may feel that he cares less for you than for his relationship with you. He cares less for you yourself than for a role that he wants you to fill. (2013, 63)

²⁷ See section 2.6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

144   The lesson from Keller’s observations is that what really matters in a relationship is not the relationship but the person one is relating to. Thus, it is a mistake to regard the relationship as a reason for various actions and attitudes, including for love, rather than the person one is relating to.²⁸ It might be thought that this is an unfair criticism of Kolodny, since Kolodny allows that love is concerned not just with the relationship but also with the beloved. He writes, “On my proposal, love has a single ground (one’s relationship) but two foci (one’s relationship and one’s relative)” (2003, 154).—Yet, however nice it is to be a focus of love, I object that it is not enough. In denying that the beloved is the ground of one’s love, Kolodny renders the beloved in a salient sense unimportant. If the beloved is not the ground for one’s love, then why isn’t the focus on the beloved a misdirection—a distraction from the real reasons? Contrary to Kolodny, I hold that the beloved is all important. The reason I love my spouse is not, as Kolodny says, that she is my spouse—that we are in a functioning marriage which I intrinsically value—but that it is she, the individual, the person. I may cite the fact that she is my spouse to explain why she is my reason. Also, the relationship may be an intensifier for my reasons to love her. However, it is not my reason for loving her; she is.²⁹ I turn now to Abramson and Leite’s view. Their view is pluralistic insofar as it countenances all sorts of reasons for love. Nonetheless, something strikes me as missing from their account: Their account of the reasons of love is still too inconstant. Their view implies that love is rationally self-propelling, because love generates reasons for itself. The inner temporality of love thus rests not on an understanding of the reasons for loving, but on an anticipation of the reasons that one will have at some future time—reasons that will be generated in due course.³⁰ Yet I object that this, too, is not enough. I understand my love for my ²⁸ Keller’s argument is a variant of Williams’s “one thought too many” argument (Williams 1981b, 18–19), which gives rise to subtle and complicated issues. Keller considers them in detail in advancing five arguments against the view that relationships constitute reasons of partiality (2013, ch.3). I would largely follow Keller, though I don’t have space here to go through all the arguments. ²⁹ Again, see Keller (2013, ch.3, esp. 56–64). ³⁰ I am indebted to Nora Kreft for this point.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

145

spouse, or my children, to be endless, not because I anticipate—would it be through an inductive inference?—that I will have reason to love them, but because I already have all the reasons I need. There is no need to wait for the reasons for my enduring love to emerge, nor is their emergence hostage to the contingencies of the future.—Once again, I want to say, the reason for loving my spouse, now and in the future, is that it is she. I need nothing more, and nothing less. The view that emerges, then, is the view that our reason for love is the other individual—and all that matters is their individuality.³¹ As long as they remain the individual they are, however they change and whatever relationships they enter or exit, we will have (defeasible)³² reason to love them. This much is needed to make sense of the persistence of love through changes in qualities and of the independence of the reasons for love from the relationship one is in. Indeed, if love is endless, then what else could be a reason for love than the individual—since everything else is bound to change?

5.4 The Inner Temporality of Love II: Love as the Apprehension of an Individual This brings me to my second proposal for understanding why, in its selfconsciousness, love is endless. The proposal is this: Love is the apprehension of an individual, and an individual is endless. Indeed, in this respect, the analogy with belief holds, however imperfect it may be in other respects—since in its self-consciousness belief is endless, because it is an apprehension of truth, and truth is endless. The self-consciousness ³¹ My view is much in the spirit of Velleman (1999) and Setiya (2014). Velleman holds that the reason for love is another’s personhood, which he understands in Kantian terms, as the rational will, i.e. “the intelligible essence of a person” (1999, 344). In a similar vein, Setiya (2014) holds that another’s humanity is a sufficient reason for love. I am inclined to agree with Setiya that personhood in the Kantian sense may be too demanding a conception of the reasons of love: It makes it hard to understand, as rational, one’s love for, e.g., infants (cf. Setiya 2014, 262). But it also strikes me that humanity might be objectionably concerned with the human species. This is why I settle for the notion of individuality here. However, I remain open to fleshing out the substance of individuality in various ways. I want to second Setiya, who says, “my view is that whatever property gives us moral status of the sort that commands respect, it is this property that justifies love” (2014, 262). ³² I return to the issue of defeasibility in section 5.5.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

146   of love is, ultimately, the apprehension of another person in a loving way—much in the way that the self-consciousness of belief is the apprehension of truth in an affirming way. Thus, we can understand the inner temporality of each mental state as a comprehension of the temporality of the object of the mental state. In the case of belief, what accounts for the inner temporality of belief is the transparency requirement: first-person, present-tense questions about what one believes are to be answered by considering the corresponding questions about what is true. By analogy, I will have to establish that love, too, is transparent. This is the crucial issue, to which I now turn. There is a way in which love is transparent, though its transparency differs from belief. To bring this out clearly, however, is a delicate matter. In support of the phenomenology, I would like to start with the following passage from Martin Buber (which I owe to Mark Johnston):³³ When I was eleven years of age, spending the summer on my grandparents’ estate, I used, as often as I could do it unobserved, to steal to the stable and gently stroke the neck of my darling, a broad dapple-grey horse. It was not a casual delight but a great, certainly friendly, but also deeply stirring happening. If I am to explain it now, beginning from the still very fresh memory of my hand, I must say that what I experienced in touch with the animal was the Other, the immense otherness of the Other, which however did not remain strange like the otherness of the ox and the ram, but rather let me draw near and touch it. When I stroked the mighty mane, sometimes marvelously smooth-combed, at other times just as astonishingly wild, and felt the life beneath my hand, it was as though the element of vitality itself bordered on my skin, something that was not I, was certainly not akin to me, palpably the other, not just another, really the Other itself; and yet it let me approach, confided itself to me, placed itself elementally in the relation of Thou and Thou with me. The horse, even when I had not begun by pouring oats for him into the manger, very gently raised his massive head, ears flicking, then snorted quietly, as a conspirator gives a signal ³³ Buber (1965, 11), quoted in Johnston (2001, 202).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

147

meant to be recognizable to his fellow conspirator; and I was approved. But once—I do not know what came over the child, at any rate it was childlike enough—it struck me about the stroking what fun it gave me, and suddenly I became conscious of my hand. The game went on as before, but something had changed, it was no longer the same thing. And the next day, after giving him a rich feed, when I stroked my friend’s head he did not raise his head. I considered myself judged.

Buber’s perceptive description of his relation to his horse, his friend, brings out what I take to be the transparency of love: Love, which Buber is describing here, is first and foremost concerned with the other, the beloved, and not with oneself.³⁴ Indeed, concern with oneself spoils love, as Buber brings out in describing the shift in consciousness from the horse to his hand. Concern with oneself is, to echo Murdoch, “the intrusion of fantasy, the assertion of self, the dimming of any reflection of the real world” (1970, 58). It is because the young Buber is conscious of such an assertion of self that he considers himself judged.³⁵ Let me develop these observations through David Velleman’s reflections on love in his remarkable essay “Love as a Moral Emotion” (1999). There, Velleman writes, I am inclined to say that love is . . . the awareness of a value inhering in its object; and I am also inclined to describe love as an arresting awareness of that value. This description of love seems right, to begin with, as a piece of phenomenology, just as the conative analysis of love seems implausible, to begin with, on phenomenological grounds. Love does not feel (to me, at least) like an urge or impulse or inclination toward anything; it feels rather like a state of attentive suspension, similar to wonder or amazement or awe. (1999, 360) ³⁴ The other’s otherness, which Buber perceptively remarks upon, is an important topic, which I cannot do justice to here. See Sartre (1945/2018, pt.3, ch.3) for some interesting, though, I think, needlessly pessimistic, observations. ³⁵ Instructively, Johnston describes “regarding one’s affective engagements as primarily mental, rather than forms of openness to how things are,” as “the pornographic attitude” (2001, 203). The young boy’s sense of being judged strikes me as an apt characterization of the self-consciousness of such an attitude.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

148   I concur with Velleman’s description of the phenomenology: Love strikes me, too, as an arresting awareness of the beloved, quite like wonder, or amazement, or awe. And this helps bring out the way in which love is transparent: Love is fundamentally about the Other, not about oneself, such as one’s own states of affect or sensation.—Love is about the other in the way in which belief is about truth. There are, however, some instructive difficulties in Velleman’s characterization of love. The first is that it brings into sharp relief the puzzle of endless love: Even without a deficiency in attention, it is typically very hard to remain arrestingly aware of one object. We are distractable creatures. Hence, it is hard to see how, if we think of love as a form of awareness or attention, we could see it as enduring, not to say endless. The second difficulty is that the conception of awareness suggests too perceptual a conception of love. To bring this out, consider the Anscombe-inspired why-question:³⁶ The question “Why are you aware of this?” is like the question “Why do you perceive this?,” and unlike the questions “Why are you amazed by this?” and “Why are you in awe of this?.” The sense in which there can be good reasons to be amazed by something or in awe of something is very different from the sense in which there can be reasons to be aware of something or to perceive something. When we cite reasons for amazement and awe, we give reasons in light of which amazement and awe are reasonable, that is, considerations which show the object of amazement and awe to be amazing and extraordinary. When we cite reasons for awareness, we either offer a causal explanation (e.g. “It was a loud noise”) or a reason for action, specifically the action of directing attention (e.g. “I wanted to find out about it”). The difficulty for the awareness view is that love is much more like amazement and awe (and respect)³⁷ than like awareness. This returns us to the idea, discussed in the previous section, that love, like awe and amazement, is reasons-responsive. And reasonsresponsiveness entails transparency: If a state is reasons-responsive, ³⁶ Here I am drawing on the discussion in section 2.4. ³⁷ Velleman writes, “I regard respect and love as the required minimum and optional maximum responses to one and the same value” (1999, 366). Respect is, plausibly, a reasonsresponsive attitude. This suggests that Velleman’s view should be understood along the lines of the view I develop in what follows, rather than in terms of awareness.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

149

then it constitutes a take on its object. That is why we can answer questions about that state by considering the corresponding question about its object and, specifically, the reasons for responding to the object in such a way that our so responding constitutes being in the relevant mental state.³⁸ Of course, now we are back to the difficult question of how to understand the reasons for love. Yet now we also have the materials to answer it: We should think of the person we love as, at once, a reason for love and also as our love! In this way, we can do justice both to the transparency and the reasons-responsiveness of love. Even if this may seem strange, it is very much part of ordinary thought. To illustrate: We might address, as well as think of, the people we love as, “My love!” (English), “Mon amour!” (French), “Mi amore!” (Italian), “Mi amor” (Spanish), or “Ljubavi moja” (Croatian)³⁹—and therein identify the object of love with our love. This account of the reasons for love also reveals these reasons to be distinctive. To bring out their distinctiveness (and to address the concern that conceiving of love as reasons-responsive is “faintly odious,” as my understated friend put it), I would like to consider in some detail the following objection: One might think that the individual cannot be a reason for love, because this would run afoul of the universality of reasons.—Let me first develop the objection and then formulate my response.⁴⁰

³⁸ We can develop the argument in the following way: It seems plausible that there could be reasons of the wrong kind for love, that is, reasons which show it good to love someone without showing love to be correct. For example, one might find oneself in a loveless but functioning marriage, and the fact that the marriage is functioning might be a reason in light of which it would be good to love one’s spouse without being a reason to love them. (Apparently, being in a functioning marriage significantly improves one’s life expectancy.) The advantages of being in a functioning marriage speak in favor of being in the state of love without showing love to be correct—without showing the other to be lovable. And if there can be reasons of the wrong kind for love, then love is in a salient sense transparent: One can answer the first-person presenttense question about love by attending to the right kind of reasons in light of which one loves. ³⁹ It might seem that German speakers are not romantic: “Meine Liebe” does not mean the same as “My love.” However, this does not reveal German speakers’ blindness to love but rather reflects the fact that “meine Liebe” means “my dear.” Even in German one can identify one’s beloved as one’s love, for example in the phrase, “Du bist die Liebe meines Lebens!” ⁴⁰ I am grateful to Sebastian Rödl for pressing a version of this objection upon me and for helpful discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

150   Suppose I suffer from friendlessness: My life would be better off if I had a loving friend. As a rational agent, I could then calculate how to best relieve my friendlessness—what sort of individual would fit the job description. I make a list: My friend should have (a) a good sense of humor, (b) children, so we always have something to talk about, and (c) an interest in fishing. After some searching, I find that Sam has all three features (as well as no objectionable features), and I come to love Sam as a friend. Now, one might point out that this conception of reasons is inadequate: Since anyone with features (a) (c) (and no objectionable features) would fit the job description, this conception of reasons does not identify Sam, the individual, as my reason for love. Yet if it did—and now we can see the force of the objection—then the reasons for loving an individual are not considerations in light of which I could come to love them, since to have the reason, I would already have to love the individual. The reasons for love are not considerations in light of which I could relieve myself of friendlessness. My response to the objection is to deny that all reasons must be universal in the sense at issue. In response to a similar objection, Troy Jollimore instructively argues that “not all valuings are like the valuings of the rational economic agent” (Jollimore 2011, 23). Reasons are economic insofar as they abstract from the individuality of the object that they are about.⁴¹ But reasons for love are not economic. My reason for loving Sam is not like my reason to relieve myself of friendlessness. A reason for love is not the fact that someone fits a job description, or fulfills a functional role, or makes for a successful relatum in a relationship. My reason for loving Sam is—Sam. There is no universal description of qualities that can replace the proper name in the statement of my reason to love Sam. Nonetheless, I want to hold on to the thought that reasons for love are reasons—albeit non-economic reasons. My rationale for this is by now familiar: Love is not passive in the way in which a condition that simply befalls us is passive. Love is unlike a fever or abdominal pain. The

⁴¹ Christopher McMahon writes, “[I]f a free-enterprise system is to be economically efficient, consumers and firms must be ‘anonymous’. That is, the proper names of consumers and firms (and the products of firms) must play no role in decisions to buy or sell” (1981, 269).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.      

151

question “Why love x?” finds application in the sense in which it asks for reasons in light of which we love x, rather than for a causal explanation or an explanation of action. Indeed, when we love someone, we should be able to say some nice things about them.⁴² (My spouse is smart, funny, thoughtful, patient, considerate, and beautiful!) Appealing to the beloved’s attractive qualities makes sense of love in a different way than one would make sense of a fever or abdominal pain: It shows the beloved to be lovable. Nonetheless, the appeal to the beloved’s qualities is not an appeal to economic considerations. Love is not meritocratic; it does not have to be earned. For example, although I love my spouse because she is smart, funny, thoughtful, patient, considerate, and beautiful, I do not therein have reason to love someone else more, who would surpass her in these qualities.⁴³ It is essential that it is her intelligence, her funniness, her thoughtfulness, her patience, her considerateness, and her beauty which make rational sense of my love. The appeal to her qualities, that is, to her, the individual, is ineliminable in a proper understanding of my reasons for love. Indeed, her attractive qualities embody her individuality, and, in that way, they make it possible for me to see her, the individual, as my reason for love.⁴⁴ Her qualities make her concrete, visible, and, hence, someone with whom I can stand in the relation of love.—This is why I concede and even welcome the observation that reasons for love are not universal, and not economic, without being forced to deny that love is reasons-responsive.⁴⁵ ⁴² However, see Yao (2020): Sometimes we may need recourse to grace. Yet even in such cases, our love is intelligible in a way that distinguishes it from a condition that merely befalls us. ⁴³ Something that defies my imagination. ⁴⁴ I take myself to have learned this from Jollimore (2011), though I think my view is more resolutely rationalist than his. Jollimore proposes to understand love as “something in between” rational and arational—as a form of vision (2011, esp. 23ff.). I have already discussed (in the context of Velleman’s view, but also in Chapter 2) why I don’t think of love as a form of vision. Once we accept the applicability of the why-question as the criterion for reasonsresponsiveness, I do not think we need to concede that love falls short of reasons-responsiveness at all. ⁴⁵ If we follow Velleman in thinking that “respect and love [are] the required minimum and optional maximum responses to one and the same value” (1999, 366), we will see that reasons for respect are non-economic as well: The reason to respect someone is that they are—that very individual. This, in turn, vindicates the Kantian idea that others figure in our thought in a fundamentally different way than mere means: our fundamental relation to others is not economic!

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

152   This concludes my account of the inner temporality of love in light of which we can explain why, in its self-consciousness, love is endless. I end the section with a final objection. One might say the following: Even if the argument I have put forward were correct, wouldn’t it be a mistake to think of an individual as endless? Indeed, isn’t death the end of an individual? To see how to resist this objection, we need to remind ourselves of an important metaphysical insight about death—an insight that Socrates expresses in his famous contention, expressed by Plato in Phaedo (115c d), that he cannot be buried: One remains an individual even in death. (The individual is not to be identified with the corpse—and hence the individual is not buried.) As Palle Yourgrau aptly puts it: “There simply is such a person as Socrates, and nothing, not even his death, can ever erase this fact. (If the bad news is that you’re going to die, the good news is that you won’t ‘disappear’—i.e., become nothing.)” (1987, 89 90).⁴⁶ Even after death, an individual retains a reality of sorts—a reality that is reflected in the fact that we can stand in various relations to them. For example, after the death of our beloved, we can miss them, long for them, grieve for them, and—love them. Indeed, as has been my theme in the first part of the book, we can, perhaps naïvely, regard grief as the continuation of love. For this to be possible, the dead must continue to be individuals, in some sense of “being.” And that is the sense in which individuals are endless.⁴⁷ One might now renew the objection, though from the other side, concerning the endlessness not of the beloved but of the lover: One might insist that even if there is a sense in which an individual is endless and this accounts for why we can love them past their death, surely it is a mistake to think of our love as endless since we can’t love them past our death. I confess to some uncertainty about how to respond to this objection. One possibility is to hold that just as existence is not required to be loved, ⁴⁶ For an elaboration of the metaphysics of death that this claim presupposes, see Yourgrau (1987; 2019). ⁴⁷ Yourgrau distinguishes “being” in this sense from existence and argues that existence is a property that the dead lose in dying—that dying is, precisely, the loss of the property of existence (1987; 2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.     

153

it is not required to love. On this line of thought, love is like individuality—an essential property that one does not lose in death. Yet despite my exuberant romantic optimism, I find it hard to square this claim with the thought that love is something that we do—that it is an active response to reasons. Can the dead be rational and irrational? Sadly, it seems to me that a better response to the objection makes a concession. To specify the required concession, I would like to return to the idea of transparency. I argued that the inner temporality of a mental state is understood in the apprehension of the object of the mental state, as well as the reasons in light of which one is in that mental state. Crucially, the temporality of our mental state, considered as a psychological fact about ourselves, is not a topic of that mental state: Love is about the beloved, not about us. This is why, in our apprehension of the temporality of love, our own finitude does not show up. We do not see our end as an end to our love—because we are not the subject of our love. Indeed, the same is true of belief: We do not see our death as the end of our belief. Thus, I must concede that, in this respect, there is something about the temporality of love that is missing “from within.” The inner temporality of love does not exhaust our understanding of the temporality of love. Yet now one might worry that this concession opens the door to regarding the inner temporality of love as much more limited than I have suggested—indeed even as a mere illusion. Are there other things about ourselves besides our death that we don’t see “from within,” in the apprehension of our beloved?—To answer this question, I now turn to consider how to reconcile the inner temporality of love with what we learn about love empirically, or “from outside.”

5.5 Reconciling Inner and Outer Temporality In Chapter 4, I argued that emotions are forms of embodied reason. They have an empirical side, which can affect their rationality. To complement the terminology of this chapter, we could say that emotions have an “outer” temporality, that is, that there is a way of understanding their temporality from a theoretical, empirical standpoint on them. My

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

154   argument concerning grief and anger has been that, in the case of these emotions, inner and outer temporality are irreconcilable: We suffer from a form of double vision. What we learn about grief and anger empirically corrects their inner temporality. “From within,” grief may strike us as the continuation of love, but we learn empirically that it is not: We learn that our love outlasts our grief. Nonetheless, we cannot fully reconcile this empirical insight with the standpoint of grief itself. In light of these arguments, one might wonder: Isn’t love, just like grief and anger, a condition of embodied reason? If so, then why don’t the empirical considerations about love affect our understanding of its reasonableness in ways that parallel the case of grief and anger? Indeed, can’t we suffer from the same kind of double vision in love as we do in grief or anger? And couldn’t what we learn about love empirically correct the self-consciousness of the inner temporality of love? Here is one particularly unfussy way to develop this line of thought. In advancing what he calls the “Bachelor Argument” against marriage, Moller reasons as follows: [I]n countries where getting a divorce is not difficult and doesn’t carry much social stigma, a great many people choose to get a divorce. In nofault-divorce societies (where a divorce can be obtained without proof of abuse or adultery) the rate is typically over 40%, and frequently over 50%. Presuming that people who choose to get a divorce have found that the feelings they originally had for each other dissipated, it seems safe to assume these divorce rates indicate that love rather often does not survive the passage of time. (Moller 2003, 80–1; italics added)

Of course, we can grant that such statistics should be regarded with caution: They are not all the relevant information about our love that we have. Furthermore, it is frequently the same people who get remarried and redivorced, which may account for the high divorce rates. Finally, it may be that love lasts past divorce, so that the assumption concerning the dissipation of love is mistaken. Nonetheless, Moller’s reflections bring out that the question of whether one’s love will last could be considered from an empirical, statistical point of view. Why wouldn’t reflection on the statistics call for a correction of the inner temporality of love?

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.     

155

The same question could be raised in a more elaborate way so as to tie it closely to the discussion of grief from the earlier chapters. In my earlier argument, I have made much of Moller’s description of the Superresilient and the Sub-resilient grievers. Recall that the Super-resilient are like us except that members have no grief reactions at all to what would strike us as great tragedies. (2007, 313) The Sub-resilient are like us except that they never cease caring as deeply for their spouses as at the moment of death; the loss of that relationship is as deeply felt at half a century as it is at half an hour. (314)

I have argued that, to properly understand the reasonableness of grief, we have to take into account the empirical insight that we are neither Superresilient nor Sub-resilient. And this is an empirical insight: We don’t know our resilience from the inside, in understanding our grief or anger, but we have to learn about it through experience. To fail to acknowledge this is to indulge in an unrealistic sense of freedom—to exhibit a form of bad faith. Indeed, it is this empirical insight that corrects our understanding of the inner temporality of grief. Yet why not say the same thing about love? Why doesn’t what we empirically learn about love correct our apprehension of the inner temporality of love? We could imagine equivalents to the Super-resilient and the Sub-resilient for love—the Super-loving and the Sub-loving. The Super-loving would be like the Sub-resilient, in that “they never cease caring as deeply for their spouses,” and their love “is as deeply felt at half a century as it is at half an hour” (Moller 2007, 313). Meanwhile, the Subloving would be like the Super-resilient in that their love is extremely short-lived. While it lasts, their love would be as strong and as profound as ours, and naïvely they might think of it as eternal. Echoing Moller, the Sub-loving would be “willing to walk through fire for their husbands and wives, and generally show tremendous concern before [their love ended]. It is just that . . . adaptive mechanisms operate so as to . . . extinguish” love after a brief period of time (Moller 2007, 313). Alas, one could easily imagine an environment in which brief, intense, eternal-seeming bouts of love would be adaptive!

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

156   Yet, arguably, we are neither Super-loving nor Sub-loving. We are not Sub-loving, because typically our love is not extinguished after a brief period of time. And we are not Super-loving, because our love changes in important ways. Nonetheless, one might insist: Knowing the differences between the Super-loving, the Sub-loving, and ourselves is empirical knowledge. How, then, are we to account for the difference between love, on the one hand, and grief and anger, on the other? One seemingly promising answer may readily come to mind: Love is ongoing. In contrast, the examples of grief and anger I have focused on are responses to a single episode of loss or injustice: my mother’s death and the murders at Sandy Hook. Perhaps the difference in temporality of these emotions is due to the difference in temporality of their objects? I confess to some uncertainty about how to both understand this suggestion and respond to it. First, how to understand it? It is, of course, true that my mother’s death, and the murders at Sandy Hook, are in the past. They have happened. However, their significance is ongoing. Unlike my formerly favorite shirt, which I do not care about anymore so that my ruining it is no longer a source of frustration, I do continue to care about my mother. Moreover, even more clearly, the injustice of the murders at Sandy Hook remains as significant as ever, and its significance will not end. This makes it difficult to understand how there is supposed to be a contrast between grief and love. Indeed, the thought that grief is the continuation of love suggests that the temporal structure of their objects is the same since the objects are the same: They are individuals. Second, as I have suggested in Chapter 1, even when conditions of loss or injustice are ongoing in the sense that they continue to happen, we (arguably) accommodate ourselves to them. We come to live with a “new normal.” The reoccurrence of loss or injustice will often move us less, even though it is no less significant. Thus, even if we acknowledge a difference in temporal structure between single episodes of loss and injustice and ongoing conditions, it is not clear how exactly this can account for the difference between grief or anger, on the one hand, and love on the other. We accommodate ourselves even to ongoing loss or injustice. Why should it be different with love?

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.     

157

I think the answer lies not in an account of the temporal structure of the objects of grief or anger and of love, respectively, but in the nature of the emotion. There is a profound difference between love and the other emotions. In the good case, grief and anger diminish and, arguably, end. In contrast, love grows.⁴⁸ When we find that someone continues to love another person, we are neither surprised nor do we think of them as stuck. We don’t think with wonder, “He still loves her, even after five years!.” But we do think with concern, even if the concern is seldom expressed, “He is still intensely grieving for her, even after five years!.” The crucial difference between grief and love is that the continuation of love needs no explanation, whereas the end of love does needs an explanation. When love ends, we suspect that something happened— even if not betrayal, then at least negligence or interference. This observation suggests the following way to reconcile the inner and outer temporality of love—the self-consciousness of love and our empirical knowledge of love: Love is defeasibly endless. This is because love grows, provided nothing compromises, stifles, or extinguishes it. Therefore, as long as we don’t have any particular reason to think that our love will be compromised, stifled, or extinguished, what we learn empirically about love is consistent with the self-consciousness of love as endless (with the exception of the point, noted earlier, that our love ends with our death). To see why such a reconciliation should be plausible, recall that even if it is granted that love is a condition of embodied reason, this does not, by itself, give rise to puzzlement. Rather, what gives rise to puzzlement is our inability to integrate inner and outer temporality—our inability to integrate the subjective perspective of an emotion with an objective perspective on that very perspective.⁴⁹ In the present context, this would be the inability to integrate what we apprehend in loving with what we empirically learn about love. Yet if, as an empirical matter of

⁴⁸ For a summary discussion of empirical research, see Ben-Ze’ev and Krebs (forthcoming). I make my claim not based on empirical research but, as before, on phenomenological grounds. ⁴⁹ Again, compare Nagel: “[The problem] of insoluble subjective-objective conflict arises when we succeed in constructing an objective conception of something and then don’t know what to do with it because it can’t be harmoniously combined with the subjective conception we still have of the same thing” (1986, 87). Cf. section 4.1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

158   fact, love (defeasibly) grows, then what we empirically learn about love is consistent with its inner temporality. Of course, even in the case of love, there is something like accommodation.⁵⁰ Mature love differs from fresh love. A feeling of infatuation that often characterizes fresh love gives way to a greater connectedness that leaves room for other aspects of life. However, crucially, accommodation is not a diminution of love but its growth. Indeed, we are not just unlike the Sub-loving whose love evaporates quickly, but we are also unlike the Super-loving whose love does not change. We are the Super-plus-loving, whose love increases and deepens over time. Thus, in the case of love, the rational and the empirical, the subjective and the objective, do not give rise to ineliminable double vision. Yet there is one important wrinkle in this argument. I have said that love grows, provided nothing compromises, stifles, or extinguishes it.⁵¹ The end of love stands in need of explanation. Now, one might wonder why the possibility of such an adverse turn of events does not stand in the way of a reconciliation of inner and outer temporality. Suppose, for example, that one suffers a particularly egregious betrayal by one’s beloved. It might be that, in such a case, one’s love will be extinguished, and its end is reasonable. And isn’t it an empirical question whether one will suffer such a betrayal?—In general, if love is defeasibly endless, don’t we still face the problem of defeating conditions?⁵² To vindicate the reconciliation of inner and outer temporality in the face of this objection, I want to identify two conditions for love that would allow us to understand love as endless—two conditions that are necessary for a “good case” of love, that is, the kind of love that, in its selfconsciousness is reasonably endless. These conditions are (1) commitment and (2) trust. When one has the relevant kind of commitment and trust, the possibility of defeating conditions will not transform the question concerning the duration of one’s love into an empirical

⁵⁰ See Rorty (1986) for delightful discussion, despite our differences about the end of love (404). ⁵¹ Once again, I recommend Yao (2020), who defends the possibility of grace in the face of such adverse conditions. ⁵² Kolodny (2003, 164–6) as well as Abramson and Leite (2011, 688; forthcoming, 9) stress that love is a source of defeasible reasons for love.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

.     

159

question, because the commitment and trust already constitute a negative answer to the question of whether the defeating conditions will obtain. Though the objection arises from a consideration of a possible betrayal by one’s beloved, I want to start my argument by considering the anticipation of a corresponding consideration from the perspective of the one who loves. For instance, in conceiving of my love as endless, how can I be confident that I won’t betray the other person and, in this way, extinguish our love? Notably, if I did, I would be failing to adequately respond to reasons—at the very least the reasons afforded by my beloved but also other reasons, including moral reasons against betrayal. The first crucial point is that it is not an empirical question for me whether I will be rational or moral. This is because irrationality and immorality are not things that happen to me, like conditions from which I suffer, but they are characteristics of things I do. And the question of whether I will do something is not settled through empirical consideration of whether this will come to pass but through consideration of the reasons in light of which one would do it. What settles, for me, whether I will betray my beloved is a question of commitment, not an empirical prediction of my behavior.⁵³ Needless to say, this argument could be expanded considerably.⁵⁴ In particular, more would have to be said about one’s understanding of the difficulty of upholding a commitment—the difficulty which, for an impartial observer is grounds for prediction that one may not follow through. However, what is important for present purposes—that is, to address the objection under consideration—is not so much betrayal and other compromising conditions in one’s own love, but such conditions in the other person’s love. It is those that seem to give rise to empirical doubts about the duration of one’s own love. This brings me to the second crucial point: In responding to someone’s commitment, we can trust them—that is, we can take their ⁵³ Indeed, my argument concerning the inner temporality of love and of other attitudes already anticipated this conclusion: Insofar as we are rational, the question of whether we have an attitude is settled by consideration of the reasons in light of we hold the attitude, not by empirical reflection on ourselves. ⁵⁴ For my book-length version of it, see Marušić (2015).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

160   commitment, understood just as such rather than as a predictor of their behavior, to settle the question of what they will do.⁵⁵ For example, if your beloved promises to you to be faithful, you can take their word for it. This is why empirical reflection on the occurrence of defeating conditions—on the likelihood that the other will egregiously betray you—is at odds with the trust that is required for love. In trusting the beloved, the question of whether the defeating conditions will obtain is settled: they won’t! Hence, even if love is (only) defeasibly endless, commitment and trust preclude a concern with defeating conditions. This is why the mere possibility of defeat does not transform the question of what will happen into an empirical question. I hasten to add a clarification: One might wonder whether this concluding argument carries more weight than I seem to attribute it. Isn’t the appeal to commitment and trust alone sufficient to account for why, in its self-consciousness, love is endless?—Yet this misconstrues the significance of my argument. The appeal to commitment and trust is merely meant to address the concern about defeaters; it is not part of my positive case for the endlessness of love. This is because love is distinct from commitment and trust. My aim in this chapter is to give an account of why love would be in its self-consciousness endless, that is, why, in loving, we understand our love to last. This account is not provided by an appeal to trust and commitment. Rather, the appeal to trust and commitment merely provides the scaffolding which explains why, in some cases, we need not be concerned with certain specific ways in which love might end. And in other cases—the love for one’s children is a case in point—even such scaffolding seems unnecessary: For what it’s worth, I can envision no defeaters for loving my children. My love for them is unconditional. I can perhaps bring out the point in this way: Even unrequited love, broken-hearted love, or tragic love may sometimes strike us as endless. Such a love can linger in ways in which its end is not in sight: We may continue to be in awe of the beloved individual. However, in those kinds of cases, the prospect of conditions which would extinguish our love remains a live issue, and reasonably so. For example, when suffering ⁵⁵ See Marušić (2017) for further discussion of the distinction between trusting someone and making predictions about someone in light of their commitments.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

. 

161

from a broken heart, I can wonder how long my pain will last, and I can take some comfort in the thought that the view from within love—the view of my broken-heartedness as endless—is likely mistaken. In general, even in the “bad case,” love may, in its self-consciousness, seem endless. However, in such a case, the question of whether something will, in due course, extinguish it can be an open, empirical question—and that is why I do not propose to defend, as reasonable, the endlessness of such love. Nonetheless, I hope that such cases help make clear that it is not commitment and trust which account for the endlessness of love, but, rather, love’s inner temporality.

5.6 Conclusion In his famous and often quoted first letter to the Corinthians, Saint Paul wrote, Love is patient, love is kind. It does not envy, it does not boast, it is not proud. It does not dishonor others, it is not self-seeking, it is not easily angered, it keeps no record of wrongs. Love does not delight in evil but rejoices with the truth. It always protects, always trusts, always hopes, always perseveres. Love never fails. (1 Corinthians 13)

Here, Saint Paul is not making a—we might add, poorly informed— empirical characterization of love. He is also not merely articulating an ideal that love aspires to, the love of saints or even angels. Rather, he articulates what is internal to the self-consciousness of love. I would like to share Saint Paul’s optimism about love, although I find it hard to say what he says without the sort of faith that he can draw on. At the very least, however, I share his optimism about the duration of love. “Love never fails,” he says. Of course, love does fail, only all too often. Saint Paul was no fool. He must have realized that. However, insofar as we love, and insofar as our love is committed and trusting, it cannot think of itself as possibly failing. And insofar as we understand its grounds—the beloved individual—we can, without irrationality, understand our love as endless.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

Conclusion We are, at once, subjects and objects—subjects, because we are conscious, have a point of view, have an inner life rich in phenomenology, and are in possession of freedom and reason—and objects because we are made of flesh and blood (and guts), and we are, to borrow Moore’s phrase, “to be met with in space.”¹ Crucially, we are subjects and objects at once: Our subjectivity is not separable from our objectivity. This, itself, is a persistent source of philosophical puzzlement. Perhaps in no other aspect of our being is the thought that we are subjects and objects at once more compelling than in our emotional life. When it comes to our emotions, there is no neat separation of body from reason, of sensation from understanding. Our emotions are embodied forms of understanding. To deny this is, alternatively, either to confuse emotion with mere sensation or to confuse emotion with mere judgment. Emotions are not mere sensations, because they are responsive to reasons. Yet emotions are also not mere judgments, because the characteristic ways in which they are embodied and situated shapes their reasonableness.² For example, the reasonableness of grief is plausibly conditioned by the functional role that it plays in our life. Yet since grief is not about this functional role, an aspect of its reasonableness remains

¹ This is Moore’s criterion for being an external object in his “Proof of an External World” (Moore 1939). Moore attributes the phrase to Kant. ² Perhaps I should say that their embodiment and situatedness shapes the rationality of emotions differently than its embodiment and situatedness shapes the rationality of judgment. It is an interesting question, which I leave for another occasion, whether and how our embodiment shapes our reasons for judgment and belief. For instance, one could hold that what it is rational to believe depends on a subject’s stakes or on the functional role of belief—and that matters would be different for creatures who were differently situated or constituted. (See section 4.3 for discussion.)—Alternatively, one could develop the point like this: It is a contingent fact about us that we forget, and this fact seems to be conditioned by our embodiment and our situation. Does the extent of our forgetfulness affect the rationality of belief and judgment? On the Temporality of Emotions: An Essay on Grief, Anger, and Love. Berislav Marušic,́ Oxford University Press. © Berislav Marušic ́ 2022. DOI: 10.1093/oso/9780198851165.003.0006

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

164  comprehensible only from a theoretical standpoint on the emotion. From the inside, the diminution of grief remains incomprehensible. In this book, I have urged that we should accept such incomprehensibility and live with an unresolved moment in our emotional life. To explain the diminution of grief and other emotions as altogether reasonable would be to conceive of our emotions as too neat; to deny the aptness of rational explanation altogether would be to conceive of our emotions as too dumb. As the puzzling subject-objects that we are, we must acknowledge puzzling aspects of our being. Needless to say, to acknowledge them does not, in itself, render them intelligible. Yet it also does not leave things just where they were. I venture to say that it can bring us to a new place of understanding—an understanding of the limitations of our intelligibility—and, at least in my case, to an ambivalent acceptance that the diminution of grief and anger may be reasonable. Indeed, once we recognize limitations on our understanding, and hold on to the continuing sense of puzzlement, we can perhaps recognize similar moments of incomprehensibility in our emotional lives elsewhere. A particularly relevant example is the puzzle of forgiveness (Kolnai 1973, Hieronymi 2001; cf. Derrida 2001): In forgiving we have to, at once, maintain that a wrong has been committed (or else there is nothing to forgive) and that this is somehow all right (because that is what forgiveness consists in). How, then, can forgiveness be reasonable? Indeed, when forgiveness strikes us as reasonable, usually the only reasons that we can identify are reasons of the wrong kind. This is so because forgiveness involves forgoing resentment, yet forgiveness does not undo the wrong to which we are responding with resentment (or else there is nothing to forgive). That is why, when forgiveness strikes us as reasonable, we can usually only identify reasons of the wrong kind— reasons that bear on what we have reason to do. However, reasons for action are the wrong kind of reasons for a change in emotion such as forgoing resentment. For instance, we might forgive someone because we need to get on with our lives. However, getting on with our lives is no more a reason to forego resentment than it is to stop grieving; it is a reason of the wrong kind. Hence, there is a direct analogy between the puzzle of accommodation and the puzzle of forgiveness. In light of this, I want to suggest that just like the puzzle of accommodation, the puzzle

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi



165

of forgiveness eludes a proper solution. Rather than seek to solve it, we should acknowledge an unreconciled moment in our emotional life.³ Indeed, there is another way in which forgiveness and accommodation are similar: They both have a temporal structure. When we suffer an injustice, we are often not prepared to forgive right away; the injustice may strike us as unforgiveable. But with time, as we accommodate ourselves to injustice, what seemed unforgiveable often seems to become forgiveable. And it seems reasonable for this to be so: It seems to be a mistake to forgive something right away, even though it may not be a mistake to forgive it after a while despite the fact that the significance of the wrong has not been undone. This suggests that accommodation itself may play a role in forgiveness: At least sometimes, accommodation seems to be a pre-condition for forgiveness. This further supports the thought that in forgiveness, as in accommodation, there is an unreconciled moment in our emotional life. Of course, maintaining that there is an unreconciled moment in our emotional life raises its own philosophical problems. Particularly pressing questions concern the relation between emotional attitudes and moral facts. Here is a particularly clear instance of this: In their interesting article “Blameworthiness and Time” (2012), Jules Coleman and Alexander Sarch advance an objection against what they call “reactive attitudes accounts” of blameworthiness.⁴ According to such accounts, there is a close connection between emotional attitudes and moral facts. In particular, “blameworthiness is to be explained by the blame responses being correctly held, felt, or otherwise directed toward one” ³ Hieronymi (2001) has been influential in my thinking about the puzzle of forgiveness. Her own solution builds on a particular conception of resentment. She holds that “resentment protests a past action that persists as a present threat” (546). Forgiveness, which consists in forgoing resentment, suspends the protest and accepts the damage caused by the wrong. My worry is that this solution does not avoid appealing to the wrong kind of reasons. That is because protesting is an action and, therefore, the question of whether to protest is a practical question. Considerations that show it worthwhile to cease protesting could make it reasonable to cease protesting. However, considerations that show it worthwhile to forgive cannot make it reasonable to forgive (if forgiving involves forgoing resentment). They would be reasons of the wrong kind.—In any case, however, the issue deserves more thorough discussion than I can offer here. I merely want to suggest that the proper response to the puzzle of forgiveness may be not to offer a solution but to live with an unreconciled moment in our emotional life. ⁴ The inspiration for such accounts is Strawson’s seminal “Freedom and Resentment” (1962/ 2008). The classic defense is Jay Wallace’s Responsibility and the Moral Sentiments (1994), which is also Coleman and Sarch’s main target.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

166  (2012, 103). However, they object that “the appropriateness of the blaming emotions is sensitive to the passage of time in ways that the blameworthiness of agents for what they have done is not” (105).⁵ They suggest that the time-sensitivity of blaming emotions constitutes “an argument for statutes of limitations that grounds them in our blaming practices and goes far beyond the epistemic and instrumental considerations that are generally thought to underwrite such limitations” (106). Yet if, indeed, blameworthiness and reasonable blaming responses come apart, as Coleman and Sarch argue, then why should crimes be prosecuted only as long as it is reasonable to blame the perpetrators, rather than as long as the perpetrators are blameworthy? Why should the fact that a crime leaves us cold, and reasonably so, mean that it should be beyond the reach of the law? In general, how does our moral emotional response, even if it is reasonable, matter, if it does not reflect the moral facts?—It seems to me that Coleman and Sarch reveal not so much a problem for reactive attitudes accounts of moral facts but rather a problem in our understanding of our own blaming responses. Indeed, a way to raise the puzzle of accommodation is to ask whether it is reasonable to blame the blameworthy, once we are reasonably equanimous towards the wrong they have committed. And it is hard—but I think necessary—to acknowledge here a limitation of intelligibility in our emotional response towards those moral facts.⁶ All this is true when it comes to grief and anger, and many other emotions. However, in a perhaps unexpected turn of the argument, I consider love to be an exception (or, as I like to think, though I have not argued it here, the exception). In the case of love, unlike in the case of grief and anger, I am sanguine about a reconciliation between the ⁵ As I stress in Chapter 3, it is not the mere passage of time, but the use we make of time to accommodate ourselves, which matters. This complicates Coleman and Sarch’s argument, as it suggests that the argument may not bear directly on statutes of limitations, as they contend. ⁶ It is instructive to consider Jay Wallace’s own thought as it develops from Responsibility and the Moral Sentiments (1994) to The View from Here (2013). It seems to me that in the later book, Wallace acknowledges an unreconciled moment in our emotional life that is not unlike the one I advocate here. On Wallace’s account, our view “from here”—that is, our subjective perspective—precludes us from viewing the world, including the moral and other evaluative facts, in terms of their full-fledged objective significance. In particular, there are certain conditions that we can recognize as regrettable without being able to reasonably regret them. This, I would add, leaves us with a form of double vision. Indeed, the echoes of Nagel’s The View from Nowhere (1986) pervade Wallace’s later book, including its title.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi



167

subjective and the objective. This is because what we learn about (real, mature) love from an empirical point of view is not inimical to its subjective apprehension. I am so optimistic about love, because I see it as the only emotional attitude that, over time, is not hostage to our embodiment and situation. Unless it is extinguished, love reasonably persists, indeed grows, even as our qualities and circumstances change. In time, love lifts us out of our embodiment and situation, and it allows us to share our life with another subject.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

References Abramson, Kate and Adam Leite. (2011). ‘Love as a Reactive Emotion.’ The Philosophical Quarterly 61, 673–99. Abramson, Kate and Adam Leite. (forthcoming). ‘Love, Value and Reasons.’ In The Oxford Handbook of Philosophy of Love, ed. Christopher Grau and Aaron Smuts. Oxford: Oxford University Press. DOI: 10.1093/oxfordhb/9780199395729.013.7 Adorno, Theodor. (1959/1977). ‘Was bedeutet. Aufarbeitung der Vergangenheit?’ In Gesammelte Schriften, Vol. 10: Kulturkritik und Gesellschaft. Frankfurt: Suhrkamp. Allais, Lucy. (2014). ‘Freedom and Forgiveness.’ In Oxford Studies in Agency and Responsibility, Vol. 2: ‘Freedom and Resentment’ at 50, ed. David Shoemaker and Neal Tognazzini. Oxford: Oxford University Press. Alvarez, Maria. (2010). Kinds of Reasons: An Essay in the Philosophy of Action. Oxford: Oxford University Press. American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders: DSM-5. Washington, DC: American Psychiatric Association. Amis, Kingsley. (1956). That Uncertain Feeling. New York: Harcourt Brace. Anscombe, G.E.M. (1957/2000). Intention. Cambridge, MA: Harvard University Press. Anscombe, G.E.M. (1978). ‘Will and Emotion.’ Grazer Philosophische Studien 5, 139–48. Antill, Gregory. (2019). ‘Evidence and Self-Fulfilling Belief.’ American Philosophical Quarterly 56, 319–31. Atkins, Ashley. (2022a). ‘On Grief ’s Wandering Thought: A Philosophical Exploration.’ Pacific Philosophical Quarterly 103, 95–117. Atkins, Ashley. (2022b). ‘On Grief ’s Sweet Sorrow.’ European Journal of Philosophy 30, 3–16. Augustine. (2018). The Confessions, trans. Sarah Ruden. New York: Penguin Random House. Bagnoli, Carla. (2018). ‘Emotions and the Dynamics of Reasons.’ The Journal of Value Inquiry 52, 347–63. Ben-Ze’ev, Aaron and Angelika Krebs. (forthcoming). ‘Love and Time.’ In The Oxford Handbook of Philosophy of Love, ed. Christopher Grau and Aaron Smuts. Oxford: Oxford University Press. DOI: 10.1093/oxfordhb/9780199395729.013.14 Bittner, Rüdiger. (1992). ‘Is it Reasonable to Regret Things One Did?’ The Journal of Philosophy 89, 262–73. Bonanno, George. (2009). The Other Side of Sadness: What the New Science of Bereavement Tells Us about Life after Loss. New York: Basic Books.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

170  Bonanno George et al. (2005). ‘Grief Processing and Deliberate Grief Avoidance: A Prospective Comparison of Bereaved Spouses and Parents in the United States and the People’s Republic of China.’ Journal of Consulting and Clinical Psychology 73, 86–98. Boyle, Matthew. (2011). ‘ “Making Up Your Mind” and the Activity of Reason.’ Philosophers’ Imprint 11, 1–24. Boyle, Matthew. (2018). ‘Sartre on Bodily Transparency.’ Manuscrito 41, 33–70. Boyle, Matthew. (2019). ‘Transparency and Reflection.’ Canadian Journal of Philosophy 49, 1012–39. Boyle, Matthew. (Forthcoming). Transparency and Reflection. New York: Oxford University Press. Brady, Michael. (2013). Emotional Insight: The Epistemic Role of Emotional Experience. Oxford: Oxford University Press. Bratman, Michael. (1987). Intention, Plans, and Practical Reason. Cambridge, MA: Harvard University Press. Briggs, Ray. (2009). ‘Distorted Reflection.’ The Philosophical Review 118, 59–85. Buber, Martin. (1965). Between Man and Man. New York: McMillan. Bykvist, Krister. (2009). ‘No Good Fit: Why the Fitting Attitude Analysis of Value Fails.’ Mind 118, 1–30. Callard, Agnes. (2018). ‘The Reason to be Angry Forever.’ In The Moral Psychology of Anger, ed. Owen Flanagan and Myisha Cherry. London: Rowman and Littlefield. Camus, Albert. (1942/1988). The Stranger, trans. Matthew Ward. New York: Vintage. Cassam, Quassim. (2007). The Possibility of Knowledge. Oxford: Oxford University Press. Cavell, Stanley. (1969). ‘Aesthetic Problems of Modern Philosophy.’ In Must We Mean What We Say? Cambridge: Cambridge University Press. Cavell, Stanley. (1994). A Pitch of Philosophy: Autobiographical Exercises. Cambridge, MA: Harvard University Press. Cholbi, Michael. (2017a). ‘Finding the Good in Grief: What Augustine Knew that Meursault Could Not.’ Journal of the American Philosophical Association 3, 91–105. Cholbi, Michael. (2017b). ‘Grief ’s Rationality, Backward and Forward.’ Philosophy and Phenomenological Research 94, 255–72. Cholbi, Michael. (2018). ‘Regret, Resilience, and the Nature of Grief.’ Journal of Moral Philosophy 16, 1–23. Cholbi, Michael. (2021). Grief: A Philosophical Guide. Princeton, NJ: Princeton University Press. Cicero, Marcus Tullius. (2002). Tusculan Disputations 3, in Cicero on the Emotions: Tusculan Disputations 3 and 4, trans. Margaret Graver. Chicago: The University of Chicago Press. Coleman Jules and Alexander Sarch. (2012). ‘Blameworthiness and Time.’ Legal Theory 18, 101–37. Cuneo, Terence. (2006). ‘Signs of Value: Reid on the Evidential Role of Feelings in Moral Judgment.’ British Journal for the History of Philosophy 14, 69–91.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi



171

Damasio, Antonio. (1999). The Feeling of What Happens: Body and Emotion in the Making of Consciousness. San Diego, CA: Harcourt. Dancy, Jonathan. (2000). Practical Reality. Oxford: Oxford University Press. Dancy, Jonathan. (2004). Ethics without Principles. Oxford: Oxford University Press. D’Arms, Justin and Daniel Jacobson. (2000). ‘The Moralistic Fallacy: On the “Appropriateness” of Emotions.’ Philosophy and Phenomenological Research 61, 65–90. D’Arms, Justin and Daniel Jacobson. (2003). ‘The Significance of Recalcitrant Emotion (or Anti-Quasijudgmentalism).’ In Philosophy and the Emotions, Royal Institute of Philosophy Supplement 52, ed. Anthony Hatzimoysis. Cambridge: Cambridge University Press. D’Arms, Justin and Daniel Jacobson. (2009). ‘Wrong Kind of Reason and the Opacity of Normative Force.’ Oxford Studies in Metaethics, Vol. 9, ed. Russ Shafer-Landau. Oxford: Oxford University Press. D’Arms, Justin and Daniel Jacobson. (2010). ‘Demystifying Sensibilities: Sentimental Values and the Instability of Affect.’ In The Oxford Handbook of Philosophy of Emotion, ed. Peter Goldie. Oxford: Oxford University Press. D’Arms, Justin and Daniel Jacobson. (2014). ‘Wrong Kind of Reasons and the Opacity of Normative Force.’ In Oxford Studies in Metaethics 9, ed. R. ShaferLandau. Oxford: Oxford University Press. Davidson, Donald. (1970/1980). ‘How Is Weakness of the Will Possible?’ In Essays on Actions and Events. Oxford: Oxford University Press. de Beauvoir, Simone. (1965). A Very Easy Death, trans. Patrick O’Brian. New York: Pantheon. de Sousa, Ronald. (1987). The Rationality of Emotion. Cambridge, MA: MIT Press. Deigh, John. (1994). ‘Cognitivism in the Theory of Emotions.’ Ethics 104, 824–54. Deonna, Julien. (2006). ‘Emotion, Perception and Perspective.’ Dialectica 60, 29–46. Derrida, Jacques. (2001). On Cosmopolitanism and Forgiveness, trans. Mark Dooley and Michael Hughes. London: Routledge. Didion, Joan. (2005). The Year of Magical Thinking. New York: Random House. Döring, Sabine. (2007). ‘Seeing What to Do: Affective Perception and Rational Motivation.’ Dialectica 61, 363–94. Elgin, Catherine. (2008). ‘Emotion and Understanding.’ Epistemology and Emotions, ed. Georg Brun, Ulvi Doğouğlu, and Dominique Kunzle. London: Ashgate. Enoch, David. (2014). ‘A Defense of Moral Deference.’ The Journal of Philosophy 111, 229–58. Evans, Gareth. (1982). The Varieties of Reference. Oxford: Oxford University Press. Fantl, Jeremy and Matthew McGrath. (2002). ‘Evidence, Pragmatics, and Justification.’ The Philosophical Review 111, 67–94. Fantl, Jeremy and Matthew McGrath. (2007). ‘On Pragmatic Encroachment in Epistemology.’ Philosophy and Phenomenological Research 75, 558–89. Fantl, Jeremy and Matthew McGrath. (2009). Knowledge in an Uncertain World. Oxford: Oxford University Press. Frankfurt, Harry. (1971). ‘Freedom of the Will and the Concept of a Person.’ The Journal of Philosophy 68, 5–20.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

172  Frankfurt, Harry. (1988). ‘On Bullshit.’ In The Importance of What We Care About: Philosophical Essays. Cambridge: Cambridge University Press, 117–33. Frankfurt, Harry. (2004). The Reasons of Love. Princeton, NJ: Princeton University Press. Frankfurt, Harry. (2006). Taking Ourselves Seriously and Getting it Right. Stanford CA: Stanford University Press. Freud, Sigmund. (1917/1999). ‘Mourning and Melancholia.’ In The Standard Edition of the Complete Psychological Works of Sigmund Freud, Vol. 14, trans. and ed. James Strachey. London: The Hogarth Press, 243–58. Frijda, Nico, Batja Mesquita, Joep Sonnemans, and Stephanie van Goozen. (1991). ‘The Duration of Affective Phenomena or Emotions, Sentiments and Passions.’ International Review of Studies on Emotion, Vol. 1, ed. K.T. Strongman. New York: Wiley and Sons. Furtak, Rick. (2018). Knowing Emotions: Truthfulness and Recognition in Affective Experience. New York: Oxford University Press. Gilbert, Daniel, et al. (1998). ‘Immune Neglect: A Source of Durability Bias in Affective Forecasting.’ The Journal of Personality and Social Psychology 75, 617–38. Goldie, Peter. (2000). The Emotions: A Philosophical Exploration. Oxford: Oxford University Press. Goldie, Peter. (2002). ‘Emotions, Feelings and Intentionality.’ Phenomenology and the Cognitive Sciences 1, 235–54. Goldie, Peter. (2012). The Mess Inside: Narrative, Emotion, and the Mind. Oxford: Oxford University Press. Greenspan, Patricia. (1981). ‘Emotions as Evaluations.’ Pacific Philosophical Quarterly 62, 158–69. Greenspan, Patricia. (1988). Emotions and Reasons: An Inquiry into Emotional Justification. New York: Routledge, Chapman and Hall. Grice, H. Paul. (1974). ‘Method in Philosophical Psychology (From the Banal to the Bizarre).’ Proceedings and Addresses of the American Philosophical Association 48, 23–53. Gustafson, Donald. (1989). ‘Grief.’ Noûs 23, 457–79. Harman, Elizabeth. (2009). ‘ “I’ll Be Glad I Did It”: Reasoning and the Significance of Future Desires.’ Philosophical Perspectives 23, 177–99. Hawthorne, John and Jason Stanley. (2008). ‘Knowledge and Action.’ The Journal of Philosophy 105, 571–90. Helm, Bennett. (2009). ‘Emotions as Evaluative Feelings.’ Emotion Review 1, 248–55. Hieronymi, Pamela. (2001). ‘Articulating an Uncompromising Forgiveness.’ Philosophy and Phenomenological Research 62, 529–55. Hieronymi, Pamela. (2005). ‘The Wrong Kind of Reason.’ The Journal of Philosophy 102, 437–57. Hieronymi, Pamela. (2006). ‘Controlling Attitudes.’ Pacific Philosophical Quarterly 87, 45–74. Hieronymi, Pamela. (2009). ‘The Will as Reason.’ Philosophical Perspectives 23, 201–20.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi



173

Hieronymi, Pamela. (2011). ‘Believing at Will.’ In Belief and Agency, ed. David Hunter. The Canadian Journal of Philosophy, Supplementary Volume 35, 149–87. Hieronymi, Pamela. (2013). ‘The Use of Reasons in Thought (and the Use of Earmarks in Arguments).’ Ethics 124, 114–27. Hieronymi, Pamela. (2014). ‘Reflection and Responsibility.’ Philosophy and Public Affairs 42, 3–41. Hieronymi, Pamela. (2021). ‘Fairness, Sanctions, and Condemnation.’ Oxford Studies in Agency and Responsibility, ed. David Shoemaker. Oxford: Oxford University Press. Higgins, Kathleen. (2013). ‘Love and Death.’ In On Emotions: Philosophical Essays, ed. John Deigh. New York: Oxford University Press. Higgins, Kathleen. (2019). ‘Putting the Dead in their Place.’ In Philosophy of Place: A Cross-Cultural Dialogue, ed. Roger Ames and Peter Herschock. Honolulu: University of Hawai’i Press. Hirsch, Eli. (2018). Radical Skepticism and the Shadow of Doubt. London and New York: Bloomsbury. Howard, Christopher. (2019). ‘The Fundamentality of Fit.’ Oxford Studies in Metaethics, Vol. 14, ed. Russ Shafer-Landau. Oxford: Oxford University Press. Hubbs, Graham. (2012). ‘Answerability without Answers.’ Journal of Ethics and Social Philosophy 7, 1–16. James, William. (1884). ‘What is an Emotion?’ Mind 9, 188–205. James, William. (1902/1958). The Varieties of Religious Experience. New York: The New American Library of World Literature. Johnston, Mark. (2001). ‘The Authority of Affect.’ Philosophy and Phenomenological Research 63, 181–214. Jollimore, Troy. (2011). Love’s Vision. Princeton, NJ: Princeton University Press. Kagan, Shelly. (1989). The Limits of Morality. Oxford: Oxford University Press. Kahn, Gilbert, ed. (1978). Simone Weil. Philosophe, historienne, et mystique. Paris: Aubier Montaigne. Keller, Simon. (2013). Partiality. Princeton, NJ: Princeton University Press. Kelly, Thomas. (2002). ‘The Rationality of Belief and Some Other Propositional Attitudes.’ Philosophical Studies 110, 163–96. Kenny, Anthony. (1963). Action, Emotion, and Will. London: Routledge. Kierkegaard, Søren. (1843/1988). Either/Or: Part II. Princeton, NJ: Princeton University Press. King, Martin Luther, Jr (2007). ‘The Christian Way of Life in Human Relations.’ In The Papers of Martin Luther King, Jr., Vol. IV: Advocate of the Social Gospel, ed. Clayborne Carson et al. Berkeley and Los Angeles, CA: University of California Press. Klass, Dennis, Phyllis Silverman, and Steven Nickman, ed. (1996). Continuing Bonds: New Understandings of Grief. Philadelphia, PA: Taylor and Francis. Kolnai, Aurel. (1973). ‘Forgiveness.’ Proceedings of the Aristotelian Society 74, 91–106. Kolodny, Niko. (2003). ‘Love as Valuing a Relationship.’ The Philosophical Review 112, 135–89.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

174  Kübler-Ross, Elisabeth. (1969). On Death and Dying. New York: Simon and Schuster. Langton, Rae. (2015). Accommodating Injustice. The John Locke Lectures, delivered at Oxford University. Oxford: Oxford University Press. Lazarus, Richard. (1991). Emotion and Adaptation. New York: Oxford University Press. Lewis, C.S. (1961). A Grief Observed. New York: Harper Collins. Lewis, David. (1979). ‘Scorekeeping in a Language Game.’ Journal of Philosophical Logic 8, 339–59. MacFarlane, John. (2014). Assessment Sensitivity: Relative Truth and its Applications. Oxford: Oxford University Press. Maguire, Barry. (2018). ‘There Are No Reasons for Affective Attitudes.’ Mind 127, 779–805. Marušić, Berislav. (2013). ‘Promising against the Evidence.’ Ethics 123, 292–317. Marušić, Berislav. (2015). Evidence and Agency: Norms of Belief for Promising and Resolving. Oxford: Oxford University Press. Marušić, Berislav. (2017). ‘Trust, Reliance and the Participant Stance.’ Philosophers’ Imprint 17, 1–10. Marušić, Berislav. (2018). ‘Do Reasons Expire?—An Essay on Grief.’ Philosophers’ Imprint 18, 1–21. Marušić, Berislav. (2020). ‘Accommodation to Injustice.’ Oxford Studies in Metaethics, Vol. 15, ed. Russ Shafer-Landau. Oxford: Oxford University Press. Marušić, Berislav. (2021). Beskrajna Ljubav. Zagreb: Institut za filozofiju. Marušić, Berislav and John Schwenkler. (2018). ‘Intending is Believing: A Defense of Strong Cognitivism.’ Analytic Philosophy 59, 1–32. Marušić, Berislav and John Schwenkler. (2022). ‘Agency and Evidence.’ In Routledge Handbook of the Philosophy of Agency, ed. Luca Ferrero. London: Routledge. Marušić, Berislav and Stephen White. (2018). ‘How Can Beliefs Wrong?—A Strawsonian Epistemology.’ Philosophical Topics 46, 97–114. Marušić, Berislav and Stephen White. (Forthcoming). ‘Disagreement and Alienation.’ Philosophical Perspectives. McCormick, Miriam. (2018). ‘No Kind of Reason Is the Wrong Kind of Reason.’ In Believing in Accordance with the Evidence: New Essays on Evidentialism, ed. Kevin McCain. Dordrecht: Springer. McCormick, Miriam. (2019). ‘Can Beliefs Be Based on Practical Reasons?’ In Well Founded Belief: New Essays on the Epistemic Basing Relation, ed. Adam Carter and Pat Bondy. London: Routledge. McCracken, Janet. (2005). ‘Falsely, Sanely, Shallowly: Reflections on the Special Character of Grief.’ International Journal of Applied Philosophy 19, 139–56. McDowell, John. (1982/1998). ‘Criteria, Defeasibility and Knowledge.’ Reprinted in: Meaning, Knowledge and Reality. Cambridge, MA: Harvard University Press. McDowell, John. (1986/1998). ‘Singular Thought and the Extent of Inner Space.’ Reprinted in: Meaning, Knowledge and Reality. Cambridge, MA: Harvard University Press. McMahon, Christopher. (1981). ‘Morality and the Invisible Hand.’ Philosophy and Public Affairs 10, 247–77.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi



175

McSweeney, Michaela. (forthcoming). ‘Maladjustment.’ Moller, Dan. (2003). ‘An Argument against Marriage.’ Philosophy 78, 79–91. Moller, Dan. (2007). ‘Love and Death.’ The Journal of Philosophy 104, 301–16. Moller, Dan. (forthcoming). ‘Love and the Rationality of Grief.’ In The Oxford Handbook of Philosophy of Love, ed. Christopher Grau and Aaron Smuts. Oxford: Oxford University Press. DOI: 10.1093/oxfordhb/9780199395729.013.35 Moore, G.E. (1939). ‘Proof of an External World. Proceedings of the British Academy 25, 273–300. Moran, Richard. (2001). Authority and Estrangement: An Essay on Self-Knowledge. Princeton, NJ: Princeton University Press. Moran, Richard. (2012). ‘Iris Murdoch and Existentialism.’ In Iris Murdoch, Philosopher, ed. Justin Broackes. Oxford: Oxford University Press. Moran, Richard. (in preparation). ‘Swann’s Medical Philosophy.’ Murdoch, Iris. (1970). The Sovereignty of Good. London and New York: Routledge. Na’aman, Oded. (2020). ‘The Fitting Resolution of Anger.’ Philosophical Studies 177, 2417–30. Na’aman, Oded. (2021). ‘The Rationality of Emotional Change: Toward a Process View.’ Noûs 55, 245–69. Nagel, Thomas. (1986). The View from Nowhere. New York: Oxford University Press. Nehamas, Alexander. (2010). Only a Promise of Happiness: The Place of Beauty in a World of Art. Princeton, NJ: Princeton University Press. Nolfi, Kate. (2018a). ‘Why Only Evidential Considerations Can Justify Belief.’ In Normativity: Epistemic and Practical, ed. C. McHugh, J. Way, and D. Whiting. Oxford: Oxford University Press. Nolfi, Kate. (2018b). ‘Another Kind of Pragmatic Encroachment.’ In Pragmatic Encroachment, ed. B. Kim and M. McGrath. New York: Routledge. Nolfi, Kate. (2021). ‘Epistemically Flawless False Beliefs.’ Synthese 198, 11291–309. Nussbaum, Martha. (1994). The Therapy of Desire: Theory and Practice in Hellenistic Ethics. Princeton, NJ: Princeton University Press. Nussbaum, Martha. (2001). Upheavals of Thought: The Intelligence of Emotions. Cambridge: Cambridge University Press. Nussbaum, Martha. (2016). Anger and Forgiveness: Resentment, Generosity, Justice. New York: Oxford University Press. Oates, Joyce Carol. (2011). A Widow’s Story: A Memoir. New York: Harper Collins. O’Brien, Lucy. (2003). ‘Moran on Agency and Self-Knowledge.’ European Journal of Philosophy 11, 375–90. O’Brien, Lucy and Douglas Lavin. (Unpublished). ‘Living Historically.’ Olson, Jonas. (2004). ‘Buck-Passing and the Wrong Kind of Reasons.’ The Philosophical Quarterly 54, 295–300. Owens, David. (2000). Reason without Freedom. London and New York: Routledge. Pettit, Philip and Michael Smith. (1990). ‘Backgrounding Desire.’ The Philosophical Review 99, 565–92. Piaget, Jean. (1954). The Construction of Reality in the Child. New York: Routledge. Preston-Roedder, Ryan and Erica Preston-Roedder. (2017). ‘Grief and Recovery.’ In The Moral Psychology of Sadness, ed. Anna Gotlib. London: Rowman and Littlefield.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

176  Prinz, Jesse. (2004). Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford University Press. Proust, Marcel. (1919/1998). In Search of Lost Time II: Within a Budding Grove, trans. C.K. Moncrieff, Terence Kilmartin, and D.J. Enright. New York: Modern Library. Rabinowicz Wlodek and Toni Rønnow-Rasmussen. (2004). ‘The Strike of the Demon: On Fitting Pro-Attitudes and Value.’ Ethics 114, 391–423. Ratcliffe, Matthew. (2017). ‘Grief and the Unity of Emotion.’ Midwest Studies in Philosophy 41, 154–74. Ratcliffe, Matthew. (2020). ‘Towards a Phenomenology of Grief: Insights from Merleau-Ponty.’ European Journal of Philosophy 28, 657–69. Rinofner-Kreidl, Sonja. (2018). ‘Grief: Loss and Self-Loss.’ In Emotional Experiences: Ethical and Social Significance, ed. John Drummond and Sonja Rinofner-Kreidl. London: Rowman and Littlefield, 91–120. Roberts, Robert. (1988). ‘What an Emotion Is: A Sketch.’ The Philosophical Review 97, 183–209. Roberts, Robert. (2003). Emotion: An Essay in Aid of Moral Psychology. Cambridge: Cambridge University Press. Rorty, Amélie. (1978). ‘Explaining Emotions.’ The Journal of Philosophy 75, 139–61. Rorty, Amélie. (1986). ‘The Historicity of Psychological Attitudes: Love is not Love which Alters not when it Alteration Finds.’ Midwest Studies in Philosophy 10, 399–412. Sacks, Oliver. (2007). ‘The Abyss.’ The New Yorker, 24 September. http://www. newyorker.com/magazine/2007/09/24/the-abyss Salmela, Mikko. (2011). ‘Can Emotion be Modelled on Perception?’ Dialectica 65, 1–29. Sartre, Jean-Paul. (1937/2004). The Transcendence of the Ego, trans. Andrew Brown. New York: Routledge. Sartre, Jean-Paul. (1943/2018). Being and Nothingness: A Phenomenological Essay on Ontology, trans. Sarah Richmond. London: Routledge. Scanlon, T.M. (1998). What We Owe to Each Other. Cambridge, MA: Harvard University Press. Scanlon, T.M. (2013). ‘Interpreting Blame.’ In Blame: Its Nature and Norms, ed. D. Justin Coates and Neal A. Tognazzini. New York: Oxford University Press. Scheffler, Samuel. (2010). ‘Valuing.’ Equality and Tradition: Questions of Value in Moral and Political Theory. New York: Oxford University Press. Schönherr, Julius. (2021). ‘Two Problems of Fitting Grief.’ Analysis 81, 240–7. Schroeder, Mark. (2007). Slaves of the Passions. Oxford: Oxford University Press. Schroeder, Mark. (2012). ‘The Ubiquity of State-Given Reasons,’ Ethics 122: 457–88. Schroeder, Mark. (2018). ‘Rational Stability under Pragmatic Encroachment.’ Episteme 15, 297–312. Schwenkler, John. (2019). A Guide to Anscombe’s Intention. New York: Oxford University Press. Senior, Jennifer. (2014). All Joy and No Fun: The Paradox of Modern Parenthood. New York: Harper Collins.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi



177

Setiya, Kieran. (2014). ‘Love and the Value of a Life.’ The Philosophical Review 123, 251–80. Siegel, Susanna. (2017). The Rationality of Perception. Oxford: Oxford University Press. Singer, Peter. (1972). ‘Famine, Affluence and Morality.’ Philosophy and Public Affairs 1, 229–43. Smuts, Aaron. (2016). ‘Love and Death: The Problem of Resilience.’ In Immortality and the Philosophy of Death, ed. M. Cholbi. London: Rowman and Littlefield, 173–88. Solomon, Robert. (1976). The Passions. New York: Anchor Press, Doubleday. Solomon, Robert. (2004). In Defense of Sentimentality. New York: Oxford University Press. Solomon, Robert. (2007). True to Our Feelings. New York: Oxford University Press. Srinivasan, Amia. (2018). ‘The Aptness of Anger.’ The Journal of Political Philosophy 26, 123–44. Stanley, Jason. (2005). Knowledge and Practical Interests. Oxford: Oxford University Press. Strawson, Peter. (1962/2008). ‘Freedom and Resentment.’ In Freedom and Resentment and Other Essays. New York and London: Routledge. Tumulty, Maura. (2020). Alien Experience. New York: Oxford University Press. van Fraassen, Bas. (1984). ‘Belief and the Will.’ The Journal of Philosophy 81, 235–56. van Fraassen, Bas. (1995). ‘Belief and the Problem of Ulysses and the Sirens.’ Philosophical Studies 77, 7–37. Velleman, David. (1999). ‘Love as a Moral Emotion.’ Ethics 109, 338–74. Vitrano, Christine. (2013). ‘Love and Resilience.’ Ethical Perspectives 20, 591–604. Wallace, R. Jay. (1994). Responsibility and the Moral Sentiments. Cambridge, MA: Harvard University Press. Wallace, R. Jay. (2013). The View from Here. New York: Oxford University Press. Wallace, R. Jay. (2019). The Moral Nexus. Princeton, NJ: Princeton University Press. Wilkinson, Stephen. (2000). ‘Is ‘Normal Grief ’ a Mental Disorder?’ Philosophical Quarterly 50, 289–304. Williams, Bernard. (1973a). ‘The Makropulos Case: Reflections on the Tedium of Immortality.’ In Problems of the Self. Cambridge: Cambridge University Press. Williams, Bernard. (1973b). ‘Deciding to Believe.’ In Problems of the Self: Philosophical Papers 1956–1972. Cambridge: Cambridge University Press. Williams, Bernard. (1981a). ‘Moral Luck.’ In Moral Luck: Philosophical Papers 1973–1980. Cambridge: Cambridge University Press. Williams, Bernard. (1981b). ‘Persons, Character, and Morality.’ In Moral Luck: Philosophical Papers 1973–1980. Cambridge: Cambridge University Press. Williamson, Timothy. (2000). Knowledge and its Limits. Oxford: Oxford University Press. Wolf, Susan. (1990). Reason within Freedom. New York: Oxford University Press. Wollheim, Richard. (1999). On the Emotions. New Haven: Yale University Press.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 13/7/2022, SPi

178  Wonderly, Monique. (2016). ‘On Being Attached.’ Philosophical Studies 173, 223–42. Yao, Vida. (2020). ‘Grace and Alienation.’ Philosophers’ Imprint 20, 1–18. Yourgrau, Palle. (1987). ‘The Dead.’ The Journal of Philosophy 84, 84–101. Yourgrau, Palle. (2010). Simone Weil. London: Reaktion Books. Yourgrau, Palle. (2019). Death and Nonexistence. New York: Oxford University Press.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

Index For the benefit of digital users, indexed entries that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. Abramson, Kate 35n.17, 52n.46, 63n.3, 138n.19, 140–3, 142n.26, 144–5 accommodation as accomplishment 82–4, 84n.42; see also process, grief as as managing emotions 5–6, 29, 78 as mere change 19–21, 24–5 as obscuring 118–20, 119n.38 collective 71 comparison to forgiveness 164–5; see also forgiveness incomprehensiveness, unintelligibility of 10–11, 22–6, 46–7, 51–2, 61, 93–4, 112, 115, 163–4 irrationality of 62, 79, 105–6, 116–17 to injustice 13–16, 19–23, 27–9, 41, 45–7, 71–2, 86n.45, 93, 102–4, 116, 156 to ongoing conditions 4, 15–16, 156 reasonableness of 22–4, 28–9, 40–1, 44, 46–7, 49, 51–2, 61–2, 69, 76–7, 80–1, 86–8, 90–1, 94–5, 103–4, 106, 110–12, 115–16 adjustment 70; see also maladjustment agency 85, 105, 122–3 agent-relativity of grief, see grief akrasia 39, 42n.28, 105–6 alienation 24–5, 46, 78, 123, 129, 134, 147n.35 ambivalence 51–2, 164 amusement 112–13, 118–21, 120n.41 anger collective 71

reasons-responsiveness of, see reasons-responsiveness of emotions temporality of 15, 17–18, 25, 72, 94–5, 97, 153–4, 156 kinds of reasons for 16, 27–8, 32, 37, 40–5, 63, 72, 76, 83–4, 86n.45, 97–8, 104–5, 110–11; see also reasons anonymous reviewer, see reviewer Anscombe, G.E.M. 29, 38n.23; see also why-question (Anscombean) anticipation of grief diminution axiological vs epistemological 9 as a failure of love 7–8 as a form of blindness to what one values 8–10, 62n.1 as indifference 8 Antirationalist Response 22, 24, 29, 36 assimilation 20–1 Atkins, Ashley 5n.5, 18n.32, 56n.55, 69n.12 attachment 54n.52, 141 attenuation 48n.38, 49n.39, 51–2 attitudes doxastic 136 emotional 96n.5, 167 kinds of reasons for 41–3 vis-à-vis emotions 34n.16; see also emotions Augustine 5n.5, 18n.32 background conditions, desires 23–4, 57–9, 85–7, 100–1, 143 Backgrounding View 87–9

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

180  bad faith 94, 122–3, 155 belief analogy with maps, see maps (analogy with beliefs) as making up one’s mind 36–8 ethics of 41, 104 functional role of 106–9 inferential vs non-inferential 39–40, 45 in light of reasons 42–3 inner temporality of, see inner temporality of love irrational 43–4, 134 pragmatist view of, see Pragmatist Response truth standard of 41–2 beloved, see love Bender, Sebastian 97n.7 blame, appropriateness of 31, 165–6 Bonanno, George 5–6, 5n.6, 13, 20n.33, 77–8, 81n.33 Boyle, Matthew 38n.22, 83n.40, 93n.1, 95n.4, 128n.3, 133n.8, 135n.13 Brady, Michael 9n.13, 31n.7, 78n.29, 81n.32 Bratman, Michael 46–7 breakup (relationship) 55–7 Buber, Martin 78n.29, 146, 146n.33, 147, 147n.34 burden of proof 101 Bykvist, Krister 72–3, 73nn.17,18 Callard, Agnes 62n.1, 83n.39, 97n.7 Camus, Albert 3, 31 Cavell, Stanley ixn.1, xii children, see relationships (significance for grief, love) Cholbi, Michael 9n.13, 30n.4, 54–7, 54n.53, 56n.55, 59–60, 66n.8, 69n.13, 72n.16, 81n.32, 83n.37, 143 Coleman, Jules 73n.17, 165–6, 165n.4, 166n.5 cognitivist view of emotions, see emotions coming to terms with 51–2, 80–1, 84; see also process, grief as

commitment 129, 137–8, 158–61 confusion 43n.33, 44, 47; see also reasons creature-construction perspective 45–6 Dancy, Jonathan 48n.38, 54n.51, 59–60, 85n.44, 143 D’Arms, Justin 5n.3, 11n.15, 30n.3, 33nn.14–15, 41n.25, 42n.27, 43n.33, 47n.37, 62n.1, 73n.18, 78n.28, 104, 105n.17, 118–20, 119nn.37,38, 120n.41 Davidson, Donald 105–6 Deigh, John 33n.15, 53n.48 Descartes, René x desire, see background conditions, desires detachment 51–2, 61, 70–3, 82, 84–5, 94–5, 100–1, 109–10; see also process, grief as Diehl, Sandy 137–8 diminution of grief ix anticipation of, see anticipation of grief diminution as accommodation, see accommodation in retrospect 10–11, 24–5, 68, 79–80 surprise at 5n.5, 22, 24–5, 88 double vision 24–5, 80, 82, 84–5, 87–91, 93–4, 97–8, 108–9, 112–13, 115–16, 123, 153–4 irreconcilable 97–8 vs toggling, see toggling DSM (Diagnostic and Statistical Manual of Mental Disorders) 3–4, 65, 116 duration, see temporality embodied, situated emotion 33n.13, 94–5, 97, 100, 117, 123, 153–4, 163–4 freedom 117 love 129, 154, 157–8 reason 100–1, 115, 123

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

 emotions as apprehensions of their objects 32–3, 95–6, 153; see also objects of grief, anger cognitivist view of 13, 20, 32–3 complexity of x, 18–19, 68–9 empirical nature of, see moral psychology ethics of 104 felt quality of xi, 16 first-personal understanding of 18, 46–7, 78, 113 functional role of 49, 84–5, 100–1, 110–12, 114–15, 121, 150, 163–4; see also belief judgment-like quality of 16, 84–5, 163–4 judgmentalist view of 33 reasons-responsiveness of, see reasons-responsiveness of emotions vs conditions that befall us 29–31, 35–6, 97–8, 122–3 vs feelings 19–21 emotional distance, see temporality emotional reason 100–1, 115, 117 endlessness of love 25, 128–9, 131–2, 138, 141, 144–6, 160–1 comparison with belief 134–8 defeasible 157–60 of the individual 152; see also love epistemology 130–1, 134–6 equanimity 28, 35, 51–2, 103–6, 109–10, 115; see also accommodation ethics of belief, see belief ethics of emotions, see emotions Evans, Gareth 133 excusable, exculpatory (accommodation) 116–18 fact-value distinction, see is-ought fallacy facticity 122–3 Fantl, Jeremy 74n.21, 106n.19 fear 19, 29–30, 30n.4, 34, 48, 112–13

181

feelings, see emotions first-personal understanding of emotions, see emotions fittingness (of emotions) 4, 33n.14, 35, 48, 73, 85–7 forgiveness 164–5, 165n.3 Frankfurt, Harry 10n.14, 77n.26, 138n.19, 139n.21 Freud, Sigmund 3, 81–2, 84n.42, 100–1 friendship, see relationships (significance for grief, love) functional role of emotions, see emotions of belief, see belief Furtak, Rick 30n.4, 33nn.13,15, 117n.36 gambler 122 goal-directedness, see process, grief as Goldie, Peter 16, 16n.22, 17n.30, 78n.27 grief agent-relativity of 55–7, 60 as the continuation of love 7, 128, 132, 139, 152–4, 156 fitting, see fittingness (of emotions) kinds of reasons for, see reasons normal vs persistent, see persistent grief object of, see objects of grief, anger obligation to grieve 7n.8, 31n.8 permissibility of 35, 63–4 reasons-responsiveness of, see reasons-responsiveness of emotions self-consuming 85–6 self-regarding 54n.52, 55–7, 56n.57, 60 shared 55–6 guilt 95–6, 114 Gustafson, Donald 82–3, 83n.37 Halpern, Faye 71n.15 Hardline Response 21–2, 61, 67–8, 86–7, 100, 103–4, 119n.37, 122–3 Harman, Elizabeth 49n.39

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

182  Hieronymi, Pamela 11n.15, 31n.6, 32n.10, 37n.21, 42n.27, 43n.29, 46n.34, 84n.41, 164–5, 165n.3 Higgins, Kathleen 7, 7n.8, 12n.16, 17n.24, 18n.32, 122n.44 Hirsch, Eli 74n.20, 114n.32 indignation 14–15 individual beloved, see love infatuation 130, 158 inferential, non-inferential belief; see belief injustice (accommodation to), see accommodation inner temporality of love 128–9, 138–45, 154, 157–8, 160–1 vs of belief 133–4, 138 vs of grief 155 vs of pain 132–3 intention 37–8, 76, 103 is-ought fallacy 94, 159–61 fact-value dichotomy 117–18 Jacobson, Daniel 5n.3, 11n.15, 30n.3, 33nn.14–15, 41n.25, 42n.27, 47n.37, 62n.1, 73n.18, 78n.28, 81n.33, 104, 105n.17, 118–20, 119nn.37,38, 120n.41 James, William 29n.1, 114n.32 Johnston, Mark 17n.27, 78n.29, 146, 146n.33, 147n.35 joy 51–2 judgmentalist view of emotions, see emotions Kant, Immanuel 145n.31, 151n.45, 163n.1 Keller, Simon 58n.60, 59n.62, 72n.16, 143–4, 144nn.28–29 Kierkegaard, Søren 130–1, 131n.6 kinds of reasons, see reasons King, Martin Luther, Jr. 102–4, 116 Koehler, Arden 83n.36 Kolodny, Niko 63n.3, 74n.19, 139n.22, 140–3, 142n.26, 144, 158n.52 Kreft, Nora 144n.30

Lavin, Douglas 65n.7 Leite, Adam 35n.17, 52n.46, 63n.3, 138n.19, 140–5, 142n.26 Lewis, C.S. 5n.5, 18n.32 love about the Other 148; see also Otherness beloved (individual) 7, 138–9, 142–4, 147, 151, 160–1 endless; see endlessness of love fresh 128n.2, 130, 158; see also infatuation growth of 157–8, 167 inner temporality of, see inner temporality of love married 127, 130–1, 131n.6 mature 127–31, 137–8, 158, 166–7 romantic 19, 127–8, 128n.2, 130–1, 140–1, 141n.25 self-consciousness of 25, 127–30, 132–4, 138, 141, 145–6, 152, 154, 157, 160–1 self-propelling 138, 140–2, 144–5 unconditional 160 unrequited 130–1, 160–1 vis-à-vis awe, amazement 147–9 loved ones, see relationships (significance for grief, love) Maguire, Barry 33n.14, 37n.20, 48n.38 maladjustment 102–4, 116 maps (analogy with beliefs) 106–10 Mayr, Erasmus 51n.43 McCracken, Janet 7n.8, 31n.8, 53n.47, 55n.54 McSweeney, Michaela 102n.13 mental illness, disorder, see persistent grief Meursault (character) 3–4, 31, 31nn.8–9 misrepresentation (of emotions) 122–3; see also bad faith Moller, Dan 5nn.3,6, 8n.12, 9, 62n.1, 66, 66nn.8–9, 77n.26, 99n.9, 100, 154–5 de Moraes, Vinicius 127–8, 128n.1 moral emotions 65–6, 147, 166

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

 moral psychology 61, 65–8, 65n.7, 86–7, 89–91, 97–101, 114, 116–17 Moran, Richard ixn.2, 8n.12, 34–5, 46n.35, 78n.28, 79n.30, 95–6, 96n.5, 112n.30, 122n.45, 130n.4, 133–4, 133n.8 motivating reasons, see background conditions, desires moving on 5–6, 11, 14, 23, 40–1, 47–8; see also Pragmatist Response Murdoch, Iris ix, 147 Na’aman, Oded 5n.3, 17n.30, 41n.25, 49n.39, 64n.4, 78n.27, 83n.39, 85–9, 86nn.45,46, 100n.11, 113, 120n.39 Nagel, Thomas 47n.36, 80n.31, 98n.8, 109n.23, 157n.49 new normal 16, 156; see also accommodation Nolfi, Kate 106–10, 107n.20, 109n.21, 111n.28 non-inferential belief, see belief normal (statistically) grief, see persistent grief norms of belief 106–9 Nussbaum, Martha 4n.1, 5n.3, 17n.29, 19, 30n.3, 33nn.12,15, 41n.24, 69–72, 69n.13, 70n.14, 81–2 O’Brien, Lucy 65n.7, 96n.5 objects of grief, anger 24–5, 40–1, 53–7, 53n.49, 68–9, 79–80, 93, 109–10 Otherness 82n.34, 146–7 pain 5–6, 8, 9n.13, 16, 21, 79, 120–1, 132–3, 150–1 pangs (of grief) 3, 33n.13, 99 partiality 23–4, 54–5, 60, 75–7, 143 perceptual justification 45, 101–2 persistent grief as mental illness, disorder 4, 65, 116 as reasonable 65n.6, 112–13 vs normal (statistically) grief 65–7

183

personal distance, see temporality Pettit, Philip 57–9 phenomenology as methodological bedrock x–xi, 17–19 of emotional experience 16, 51, 99, 104–5, 113–14, 128–9, 148, 163 philosophical reflection (role, scope of) x–xi, 90–1, 114n.33 Piaget, Jean 13, 20, 20n.33 Plato, see Socrates Portmore, Douglas 63n.3, 76n.23 practical reasoning 32, 84; see also reasons pragmatic encroachment 25, 94, 106, 108–12, 109n.22, 114–15 Pragmatist Response 22, 40–1, 44–5, 51–2 pre-grief 35n.18 Preston-Roedder, Erica 7n.10, 66n.8, 83n.38 Preston-Roedder, Ryan 7n.10, 66n.8, 83n.38 Prinz, Jesse 16n.23, 17nn.27–28,31 process, grief as as a matter of empirical fact 87–9, 93; see also moral psychology goal-directed, teleological 32, 82–4, 88n.48 healing 77–80, 84–5 process rationality 86–8 Process View 77–8 Propositions (eternalist vs temporalist view) 134–5 Proust, Marcel 7–10, 8n.12, 10n.14, 68, 79, 82–3, 90, 112–15 psychological injury 78–9, 84–5 vs physiological injury 77–9 psychological work 22, 61, 80, 83–5; see also process, grief as temporal limitations of 83–4 rakehell 95–6, 114 Ratcliffe, Matthew 16n.23, 18n.32, 66n.9, 78n.27, 122n.44

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

184  reality-testing 81 reasons attenuating reasons, see attenuation confused vs non-confused 43–5 defeating 39–40 embodied, see embodied, situated inferential vs non-inferential, see belief normative 58–9, 141 of the right kind 41, 43n.33, 48–50, 56n.57, 149n.38 reasonableness without reasons 25, 94, 111–12, 121 space of reasons 27, 34–5, 39–40 temporality of, see temporality warranting vs requiring 35n.17, 42n.26, 63–4, 63n.3 reasons-responsiveness of emotions 4, 24, 31–2, 35–6, 60, 122–3 in comparison to reasonsresponsiveness of action 32 in comparison to reasonsresponsiveness of belief 32–3 of grief, anger 29, 32 of love 138–40, 139n.21, 149, 151n.44 vs permissiveness, obligation 35, 63–4, 142n.26 vs recalcitrance to reason 33–6 reconciliation 90–1, 123, 166–7; see also double vision of inner and outer temporality 153–4, 157–9 unreconcilable, irreconcilable 24, 61, 89–90, 93–4, 112, 114–16, 164–6, 166n.6 redemption 95–6 Reed, Baron 137n.15 regret 34, 49n.39, 99 relationships (significance for grief, love) children 47–8, 50–2, 60, 127–9, 143, 160 friendship 10n.14, 55–6, 59n.62, 128–9, 140–3, 146–7, 150

loved ones 13, 36, 47–8, 54n.53, 70–1, 142–3 mother x, 7, 12, 23–4, 53–4, 58–60, 69–70, 72, 74 spouse 56n.57, 127–9, 144–8, 149n.38, 151 relief xi, 35n.18, 68–9, 79, 114, 116n.34 resilience ix, 5–6, 13–15, 25–6, 67–8, 86–8, 99, 103–4, 116, 155; see also Sub- and Super Resilience respect 145n.31, 148, 148n.37, 151n.45 pay respects 83 revenge 83 reviewer (mean) x–xi, 36, 36n.19, 56n.58 Rinofner-Kreidl, Sonja 9n.13, 15n.21, 18n.32, 56n.56, 81n.32, 122n.44 Roberts, Robert 17n.27, 30n.3, 53n.50 Rödl, Sebastian 135n.12, 149n.40 Rorty, Amélie 12n.17, 33n.15, 158n.50 Sacks, Oliver 74–5 Saint Paul 161 Sarch, Alexander 73n.17, 165–6, 165n.4, 166n.5 Sartre, Jean-Paul 93n.1, 94–5, 94n.2, 95n.4, 117–18, 117n.35, 122–3, 122nn.44–45, 123n.47, 128n.3, 133n.8, 135n.13, 137n.18, 138n.19, 147n.34; see also situation (Sartrean) Scanlon, Thomas 29–30, 42n.27, 55–6, 73n.18 Schönherr, Julius 4n.2, 111–12 Schroeder, Mark 58–9, 87n.47, 106n.19, 109n.22 Schwenkler, John 32n.10, 38n.23, 74n.19 self-consciousness ix of belief 134–5, 137–8, 145–6 of emotions 89–90, 93n.1, 95–8, 104–5, 123 of love, see endlessness of love of pain 132–3 Setiya, Kieran 46–7, 145n.31 shock 5–6, 55, 61, 71–2; see also surprise Singer, Peter 62–3

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/7/2022, SPi

 situated reason, emotion, see embodied, situated situation (Sartrean) 94n.2, 117, 117n.35, 120–2, 167; see also embodied, situated skepticism 98n.8, 101–2, 131n.5 Smith, Michael 57–9 Smuts, Aaron 8n.12, 9, 10n.14, 66n.8 Socrates 90n.50, 152 Solomon, Robert 7, 7n.8, 17n.29, 29n.1, 30n.3, 31n.8, 33n.12, 78n.27, 84n.42, 122n.44 de Sousa, Ronald 17n.27, 30n.2, 34n.16 space of reasons, see reasons spouse, see relationships (significance for grief, love) stakes 106, 109n.22, 110–11; see also pragmatic encroachment Stoics 120–1 Strawson, Peter 13–14, 14n.20 Sub- and Super-resilience 66–8, 88–9, 99–100, 119–21, 155; see also resilience Sub- and Super-love 155–6, 158 subject-object dilemma 123, 163–4 surprise 5–7, 22, 55, 61, 71–2, 88, 99–101 surprise at grief diminution, see diminution of grief teleological structure, see process, grief as temporal distance, see temporality temporality as partiality 75–7; see also partiality inner, see inner temporality of love of reasons 72 temporal profile of emotions 35, 71–2 temporal vs emotional distance 22, 76–7, 77n.24 temporal vs personal distance 72–3, 76 temporal vs spatial distance 62–3 vis-à-vis endlessness, see endlessness of love

185

time-sensitivity of emotions 61, 68, 165–6; see also temporality Time-Sensitive Response 24–5, 117 toggling 95, 97–8, 115–16 transparency of belief 133–4, 146 of love 146–9, 153 trust 129–31, 137–8, 158–61 truth 41–2, 50, 53–4, 134–6, 145–6, 148, 161 Tumulty, Maura 31n.7, 56n.57, 84n.41, 95n.3 unreconcilable, see reconciliation van Fraassen, Bas 136n.14 Velleman, David 138n.19, 139, 139n.22, 145n.31, 147–8, 148n.37, 151nn.44–45 Wallace, R. Jay 17n.31, 35n.17, 49n.39, 63–4, 114n.33, 142n.26, 165n.4, 166n.6 warrant, see reasons waves (of grief), see pangs (of grief) weakness (of will) 39, 105, 115 Wearing, Clive 74–6 Wearing, Deborah 74–6 Weil, Simone 62–3, 63n.2, 74n.19, 116 Wilkinson, Stephen 31n.8, 65n.6 Williams, Bernard 10n.14, 25n.40, 43n.29, 65–6, 72, 144n.28 why-question (Anscombean) 30–1, 31nn.7,8, 32, 40–1, 68, 139, 148, 151n.44; see also Anscombe, G.E.M. Work View 80–2, 84–5, 88n.48; see also psychological work Wollheim, Richard 12n.17, 19, 33–4, 34n.16 Wonderly, Monique 54n.52, 141n.25 Yao, Vida 151n.42, 158n.51 Yourgrau, Palle 57n.59, 63n.2, 152, 152nn.46–7