Oxford Studies in Philosophy of Religion Volume 10 9780192862976, 0192862979

Table of contents :
Cover
Oxford Studies in Philosophy of Religion: Volume 10
Copyright
Contents
Preface
The Sanders Prize in Philosophyof Religion
List of Contributors
1: And All Shall Be Changed: Virtue in the New Creation
I. The Terminal Conception of Virtue
Courage
Justice
Temperance
Prudence
II. The Christian Problem with the Terminal Conception
The Preliminary Problems
The Inverse Relation of Virtue and Age
Significance, Suffering, and Souls
Elegance Is Everything
Story One: That They Might Enter into Joy
Story Two: Away with These Instruments
Troubles in Platodise
The Inadequacy of Aristotle
Story Three: Approximations and Preconditions
III. A Workable Endurentist Account
Distributive Justice
Commutative Justice
Dulia
Gratitude
Truthfulness
Friendliness
Liberality
Courage
Cowardice
Fearlessness
Temperance
Prudence
Conclusion
Works Cited
2: Being Good and Loving God
1. Praisefulness, Thankfulness, and Contrition
2. The Moral Value of Praisefulness, Thankfulness, and Contrition
3. The Justification of Committing to God’s Existence out of Praisefulness, Thankfulness, or Contrition
4. Conclusion
References
3: How to be a Mereological Anti-Realist
1. Introduction
2. Non-Theistic Mereological Anti-Realism
2.1. Pearce
2.2. Kriegel
3. Theistic Mereological Anti-Realism
4. Motivations for Theistic Mereological Anti-Realism
4.1. Motivation 1: God is omnipotent
4.2. Motivation 2: Theistic mereological anti-realism potentially allows us to endorse a moderate answer to the special composition question
4.3. Motivation 3: Theistic mereological anti-realism allows us to explain mereological pairing relations
4.4. Motivation 4: Theistic mereological anti-realism provides novel resources for making sense of the general resurrection of the dead
4.5. Motivation 5: Theistic mereological anti-realism provides novel resources for responding to the problem of evil
5. Conclusion
References
4: Creativity in Creation
1. Introduction
2. Combination
2.1 Boden on Creativity
2.2 Creation in the World-Actualization Model
3. Exploration
3.1 Appreciating the Tension
3.2 Possible Dissolutions
4. Transformation
4.1 Creating Possibilities
4.2 Morals
5. Conclusion
References
5: Defining Religion
1. The Project: Where Angels Fear to Tread
2. Religiosity First
3. Religiosity
3.1. Community
3.2. Fundamentals of Faith
3.3. Imaginative Engagement
4. Religion Defined
References
6: Time and the Nature of the Atonement
1. Introduction
2. Key Theological Assumptions
3. Potential Roadblocks to the Debate
4. Existence Assumptions and Approaches to the Problem
5. Hypertime to the Rescue?
6. Consequence Removal Questions
7. Explicating the Metaphors: ‘Transfer’ Views
8. Solidarity Views and Stump’s Account
9. Divine Eternity
10. Eternalist A-Theory to the Rescue?
11. A Continuing Eucharist?
12. Taking Stock
References
7: Why the One Did Not Remain Within Itself
I. The Problem
II. Two Rejected Answers and Johnston’s Proposed Third Answer
A. Meliorism
B. Generosity
C. Manifestationism
III. Going Deeper
IV. The Principle of Sufficient Reason
References
8: God’s Perfect Will: Remarks on Johnston and O’Connor
References
9: Divine Self-Manifestation: Reply to O’Connor and Pearce
Real Contingency
Alethic Modality Refigured
The First Creation
10: To What Extent Must Creatures Return to the One?
11: God Does the Best: A Neoplatonic Meditation on O’Connor
References
12: The Contingency of Creation and Divine Choice
1. The Problem
2. Non-moral Divine Preferences
3. Contingency Regained
4. Concluding Remarks
References
13: Ineffability and Creation
References
Index

Citation preview

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

Oxford Studies in Philosophy of Religion

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

Editorial Board Andrew Chignell (Princeton University) Frances Howard-Snyder (Western Washington University) Klaas J. Kraay (Ryerson University) Sam Lebens (University of Haifa) Brian Leftow (Rutgers University) Christian Miller (Wake Forest University) Graham Oppy (Monash University) Eleonore Stump (St Louis University) Neal Tognazzini (Western Washington University) Peter van Inwagen (University of Notre Dame and Duke University) Linda T. Zagzebski (University of Oklahoma)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

Oxford Studies in Philosophy of Religion Volume 10 Edited by LARA BUCHAK AND DEAN W. ZIMMERMAN

1

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

3

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 © the several contributors 2022 The moral rights of the authors 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: 2019945436 ISBN 978–0–19–286297–6 DOI: 10.1093/oso/9780192862976.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY 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, 21/1/2022, SPi

Contents Preface The Sanders Prize in Philosophy of Religion List of Contributors

1. And All Shall Be Changed: Virtue in the New Creation

vii ix xi

1

Marshall Bierson

2. Being Good and Loving God

40

T. Ryan Byerly

3. How to be a Mereological Anti-Realist

83

Andrew Brenner

4. Creativity in Creation

120

Meghan D. Page

5. Defining Religion

145

Samuel Lebens

6. Time and the Nature of the Atonement

169

Amy Seymour

7. Why the One Did Not Remain Within Itself

234

Timothy O’Connor

8. God’s Perfect Will: Remarks on Johnston and O’Connor

248

Kenneth L. Pearce

9. Divine Self-Manifestation: Reply to O’Connor and Pearce

255

Mark Johnston

10. To What Extent Must Creatures Return to the One? Caleb Cohoe

270

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

vi



11. God Does the Best: A Neoplatonic Meditation on O’Connor

279

Katherin A. Rogers

12. The Contingency of Creation and Divine Choice

289

Fatema Amijee

13. Ineffability and Creation

301

Alexander R. Pruss

Index

311

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

Preface This is the tenth volume of Oxford Studies in Philosophy of Religion. As with earlier volumes, these essays follow the tradition of providing a nonsectarian and non-partisan snapshot of the subdiscipline of philosophy of religion. This subdiscipline has become an increasingly important one within philosophy over the last century, and especially over the past halfcentury, having emerged as an identifiable subfield at the same time as other emerging subfields such as the philosophy of science and the philosophy of language. This volume continues the initial intention behind the series: attracting the best work from philosophers well known for their contributions to the philosophy of religion, as well as including essays by philosophers working mainly outside this area when their interests intersect with issues in the philosophy of religion. This inclusive approach to the series provides an opportunity to mitigate some of the costs of greater specialization in our discipline, while at the same time inviting wider interest in the work being done in the subfield. Each volume typically includes an essay by the winner of the Sanders Prize in Philosophy of Religion, described within. L. B. & D. Z. Princeton, NJ; New Brunswick, NJ

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

The Sanders Prize in Philosophy of Religion The Marc Sanders Prize in Philosophy of Religion is a biennial essay competition open to scholars who are within fifteen years of receiving a Ph.D. or students who are currently enrolled in a graduate program. Independent scholars may also be eligible and should direct inquiries to Dean Zimmerman, co-editor of Oxford Studies in Philosophy of Religion, at . The award for the prize-winning essay is $5,000. Winning essays will be published in Oxford Studies in Philosophy of Religion. Submitted essays must present original research in philosophy of religion. Essays should be between 7,500 and 15,000 words. Since winning essays will appear in Oxford Studies in Philosophy of Religion, submissions must not be under review elsewhere. To be eligible for the next prize, submissions must be received, electronically, by January 31st, 2023. Please send your submissions to the following email address: [email protected]. Refereeing will be blind; authors should omit remarks and references that might disclose their identities. Receipt of submissions will be acknowledged by e-mail. The winner will be determined by a committee of members of the Editorial Board of Oxford Studies in Philosophy of Religion and will be announced by late March. Inquiries should be directed to Dean Zimmerman at [email protected]. Marshall Bierson was the winner of the 2019 Prize. His essay, “And All Shall Be Changed: Virtue in the New Creation,” appears as Chapter 1 in this volume. Marshall is a Ph.D. candidate at Florida State University. The 2019 competition yielded three outstanding runners-up which are also published here: Andrew Brenner, “How to be a Mereological AntiRealist”; T. Ryan Byerly, “Being Good and Loving God”; and Meghan Page, “Creativity in Creation.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

List of Contributors Authors Fatema Amijee, University of British Columbia Marshall Bierson, Florida State University Andrew Brenner, Hong Kong Baptist University T. Ryan Byerly, University of Sheffield Mark Johnston, Princeton University Caleb Cohoe, Metropolitan State University of Denver Samuel Lebens, University of Haifa Timothy O’Connor, Indiana University Meghan D. Page, Loyola University Kenneth L. Pearce, Trinity College Dublin Alexander R. Pruss, Baylor University Katherin A. Rogers, University of Delaware Amy Seymour, Fordham University Series Editors Lara Buchak, Princeton University Dean W. Zimmerman, Rutgers University Managing Editors Isaac Wilhelm, National University of Singapore Chris Copan, Rutgers University Fatema Amijee [email protected] Marshall Bierson [email protected]

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 21/1/2022, SPi

xii

  

Andrew Brenner [email protected] T. Ryan Byerly t.r.byerly@sheffield.ac.uk Mark Johnston [email protected] Caleb Cohoe [email protected] Samuel Lebens [email protected] Timothy O’Connor [email protected] Meghan D. Page [email protected] Kenneth L. Pearce [email protected] Alexander R. Pruss [email protected] Katherin Rogers [email protected] Amy Seymour [email protected] Series Editors Lara Buchak [email protected] Dean W. Zimmerman [email protected] Managing Editors Isaac Wilhelm [email protected] Chris Copan, Rutgers University [email protected]

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

1 And All Shall Be Changed Virtue in the New Creation Marshall Bierson

Is virtue still necessary for human flourishing once humans are clothed in immortality? It seems that we most need the virtue of courage when our lives are endangered, we most need the virtue of justice when others’ lives are threatened, and we most need the virtue of temperance when faced with goods—like food and drink—that harm our health by overindulgence. But this raises a problem for the Christian ethicist: if virtue is most important when we are confronted with death, sorrow, crying, and pain, will virtue matter less once these former things are passed away? If love hath no greater manifestation, then has virtue peaked this side of eternity? And if not, what is the value of virtue in the new creation? This chapter develops in three parts. First, I present the worry that much of the value of the moral virtues terminates in the new creation. Second, I explain why Christians should be committed to the view that the value of the virtues endures. Third, I present a solution to the problem: I argue that the enduring value of the virtues is chiefly explained by the fact that virtue enables enjoyment of ideal conditions. I conclude, by explaining other applications of my thesis, that my solution has broad explanatory power and gesture at directions for future work.

I. The Terminal Conception of Virtue Martha Nussbaum and Bishop Joseph Butler both argue that human limitations play a central role in explaining the value of the virtues. Marshall Bierson, And All Shall Be Changed: Virtue in the New Creation In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Marshall Bierson 2022. DOI: 10.1093/oso/9780192862976.003.0001

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

2

 

Nussbaum, for instance, argues that human virtue is inseparable from “our finitude, and in particular our mortality.”¹ The general claim about finitude, even apart from the particulars of mortality, is compelling. For example, consider the widow’s offering in Luke 21. The widow is more generous than the wealthy giver because she gave out of her poverty. The rich must give far more than the poor to exhibit the virtue of generosity. In giving ten dollars, a person with ten dollars to his name is more generous than a person with a thousand dollars to his name. The general principle is simple: the wealthier people are, the more they must give in order to be generous. There is a finitude to wealth that makes generosity a possibility for all givers. Suppose there were an infinitillionaire; such a person could never be financially generous because no amount of giving could ever be financially sacrificial. Financial generosity is conditioned on financial finitude.² In turn, since other virtues play a role in human lives and since humans are finite creatures, it seems likely that in general human virtues are possible only for creatures with human limitations. Beyond making virtue possible, finitude also seems like a condition of the virtues being valuable to us.³ For instance, it is unclear what value intellectual virtues would have for an omniscient being. And why would an omnipotent person need the self-discipline to acquire moral, emotional, or physical strength? Most features of human finitude will remain in the new creation; resurrection makes us neither endlessly wealthy, omniscient, nor omnipotent.⁴ ¹ The Therapy of Desire, 226. While I focus on the version of Nussbaum’s argument presented in chapter 6 of Therapy of Desire, Nussbaum also argues for this conclusion in “Transcending Humanity” (pages 371‒8, esp. 374‒6), which is in turn an elaboration on the theme in Part III of The Fragility of Goodness. All Nussbaum quotes are from Therapy of Desire, though no doubt there are times when my exposition of Nussbaum is influenced by “Transcending Humanity.” ² This raises interesting questions about the generosity of God; however, I leave them aside for purposes of this chapter. ³ I am using “value” here as a catchall term to mean that we have reason to care about possessing the virtues. Different arguments in this chapter try to cash out value in different ways. I will distinguish these ways via the language of “necessary” and “important.” If a virtue is necessary for something, it means some good cannot be obtained without the virtue. If virtue is important to a good, it means that virtue is a central or major contributor to that good. Similarly, when I get to my own position, I begin to differentiate final and instrumental goodness, as well as intrinsic and extrinsic goodness (see footnote 16 for further clarification). ⁴ There may be a sense in which resurrection makes us omniscient/omnipotent. Augustine at times gestures at the thought that in fully knowing God, the creator of everything, we come to know everything ourselves. Such considerations, while interesting, are left to the side in this chapter.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

3

However, moving from a fallen to a perfected state does remove one central human limitation: our mortality. If the virtues, or their value, depend on human mortality, then the virtues, or their value, would terminate when we are raised to immortality in the new creation. This terminating conception of the virtues, I will argue, creates a troubling problem for Christian ethics. To see the force of the problem, consider the four cardinal virtues: courage, justice, temperance, and prudence.

Courage Nussbaum argues that mortality is necessary for human courage by examining literary portrayals of immortal lives. She concludes that the gods of Greek mythology are particularly apt examples of an immortal human life and argues that they are rightly portrayed without courage. Since “courage consists in a certain way of acting and reacting in the face of death,” and since the Greek gods do not face death, they can at most have a “pale simulacrum” of courage in their attitudes towards pain.⁵ Now, Nussbaum’s conclusion is too strong. At most, Nussbaum shows that the Greek gods never act courageously, not that the gods are incapable of possessing courageous dispositions. Humans who never face danger can still be courageous if they would do the right thing were they, counterfactually, in dangerous situations. Similarly, immortal gods could be courageous if they would have risked their lives had they been mortal.⁶ However, this response only reveals how meaningless courage is to the gods. Our censure of cowardly gods amounts to the charge that they would act poorly were they beings radically unlike the beings they are— were they subject to limits to which they are not, and never will be, ⁵ The Therapy of Desire, 227. ⁶ One could salvage Nussbaum’s strong conclusion if acting virtuously is necessary to develop virtues. Because the gods cannot act courageously, they cannot develop courage and so can never be courageous. This move, however, means we cannot extend Nussbaum’s argument to the new creation. After all, even if we cannot develop new virtue in the new creation, we could still have the virtues we developed prior to our earthly deaths. Indeed, the fact that courage can only be developed here in the fallen world seems like an important premise for many versions of the soul-building theodicy.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

4

 

subject. But such a censure lacks bite. Just as a butterfly is in no way deficient in being unable to hunt with a spear, cowardice is in no way a deficit for gods who will never need courage. One might respond that the gods still need courage to face other risks. I am rarely, if ever, in life-threatening situations, yet there are plenty of cases in which my cowardice stops my pursuit of the good. My cowardice manifests not in the inordinate safeguarding of my life, but rather in the inordinate safeguarding of my reputation, position, and happiness. But even if certain ways of being immortal might still involve substantial danger, the many further perfections of the new creation make that an implausible response in the particular context of Christian paradise.

Justice Nussbaum also argues that there may be tension between the levity of immortal interests and the seriousness of our desire for justice. Political justice and private generosity are concerned with the allocation of resources like food, seen as necessary for life itself, and not simply for play or amusement. The profound seriousness and urgency of human thought about justice arises from the awareness that we all really need the things that justice distributes, and need them for life itself. If that need were removed, or made non-absolute, distribution would not matter, or not matter in the same way and to the same extent; and the virtue of justice would become optional or pointless accordingly. Aristotle is right to say that the idea of debates about justice and contract among the gods is a ridiculous idea . . . ⁷

It is bad to cheat in games, but that badness is trifling next to oppression of the global poor. Nussbaum’s thought is that the seriousness of justice assumes real need, which will not be present in paradise, and that this argument applies both to public justice and to private generosity. In Bishop Butler’s sermon on compassion, he echoes the latter application ⁷ The Therapy of Desire, 228.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

5

and suggests that the new creation is “to be considered as a foreign country” that contains “no follies to be overlooked, no miseries to be pitied, no wants to be relieved,” and where compassion “will happily be lost, as there will be no objects to exercise it upon.”⁸ Thomas Aquinas, in contrast to Nussbaum, thinks that justice is the virtue that most obviously persists in our future immortal state because our lives will still “be subject to God,” and subjection to a superior is a component of justice.⁹ Nussbaum and Aquinas are both deeply insightful thinkers, yet Aquinas seems to treat Nussbaum’s argument as obviously wrong. Who is correct? We can explain this disagreement by borrowing a distinction from Aquinas between commutative and distributive justice.¹⁰ Commutative justice governs the mutual dealings between private persons; distributive justice governs the dealings between persons as members of a single community and shared common good. Breaking a promise, for example, is a commutative injustice, while allowing another to die of poverty is a distributive injustice. It is both a commutative and distributive injustice to allow another to die of poverty by refusing to pay them for their labor. Aquinas argues that our rightful subjection to God’s authority would continue for eternity; such a duty falls under commutative justice. Nussbaum, on the other hand, explicitly limits her argument to “political” or distributive justice. As such, though Aquinas and Nussbaum likely disagree about the persistence of the virtue of distributive justice in the new creation, Aquinas’s argument does not suggest that Nussbaum made an obvious mistake. There are also reasons to think that part of the value of commutative justice terminates with death. Take as an example the commutative obligations to keep promises. Much human good hangs on our ability to rely on one another’s promises.¹¹ But this seems especially true when considered in light of our mortality. First, in trusting others we become vulnerable to consequences if the other person breaks their promise. If humans are mortal, then one consequence we can become vulnerable to ⁸ Fifteen Sermons Preached at the Rolls Chapel, 59. ⁹ Summa Theologiæ, I-II q. 67 a. 1. ¹⁰ II-II q. 61 a. 1. ¹¹ Anscombe, “On Promising and Its Justice, and Whether It Needs Be Respected In Foro Interno,” 73–4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

6

 

is death. Because death is a particularly serious consequence, that possibility means that more human good depends on promise-keeping than it would if death were not a possibility. Second, because we die, promising is one of the only ways to affect the world after our death. Thus, when expounding Anscombe’s account of promise-keeping, Philippa Foot uses death as a central illustration: “it is easy to see how much good hangs on the trustworthiness involved if one thinks, for instance, of the long dependency of the human young and what it means to parents to be able to rely on a promise securing the future of their children in case of their death.”¹² Death plays a significant role in increasing the importance of promise-keeping to human good. This suggests that some of the value of major aspects of commutative justice between human beings, such as promise keeping, is conditioned on mortality.

Temperance Temperance seems particularly important for mortal creatures in two ways. First, the import is explained partially by the impact of intemperance on human health. Gluttony risks death in a straightforward way. We require “moderation” or temperance as the “management of appetite in a being for whom excesses of certain sorts can bring illness and eventually death.”¹³ Second, temperance’s importance in a fallen world is greater than in a perfect paradise because occasions for intemperance may not exist in the new creation. We suggest to the intemperate that they avoid exposing themselves to temptation. For example, an American consumer is advised not to go grocery shopping while hungry because intemperance is less dangerous when there are few enticements to deviate from the good. This suggests temperance is less valuable in a world without temptations. Bishop Butler refers to a virtue’s loss of value when discussing patience in his Analogy of Religion, arguing “that there can be

¹² Natural Goodness, 45.

¹³ Nussbaum, The Therapy of Desire, 228.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

7

no scope for patience” in the new creation “when sorrow shall be no more.”¹⁴ It is the same with temperance.

Prudence Nussbaum does not discuss the virtue of prudence because her argument is that it is impossible for immortal creatures to possess certain virtues. Prudence is possible for immortal gods—even immortal creatures can make better or worse choices—so Nussbaum does not explore it. However, we can make the case that prudence would be less valuable in eternity, even if not impossible. First, there are prima facie reasons to think prudence is less significant if mistakes do not result in death and suffering. We often commend prudence in times in which a bad decision could result in terrible consequences. For example, I am more concerned with a presidential candidate’s wisdom than with the wisdom of the child who rakes my leaves. This difference is explained precisely by the dire consequences of a political leader’s failure. We care most about prudence in high-stakes contexts. Prudence is therefore more valuable when mistakes cause greater harm. This is significant because it seems like there are few, if any, high-stakes contexts we would face in the new creation. After all, no one will suffer, much less die, because of any mistake I make. Death and suffering will not follow imprudent decisions when death and suffering are no more. Second, practical wisdom is the virtue by which we select means to the attainment of our rightful ends. Prudence is important this side of eternity because we are unlikely to obtain our good without it. But if the new creation entails that we have achieved our ultimate good—worshipping in the bodily presence of God—then prudence is no longer as valuable.¹⁵ ¹⁴ The Analogy of Religion, 148. ¹⁵ A final worry you might have about the value of prudence is that prudence acts as a master virtue. It is important because without prudence none of the other virtues help us achieve our ends. Thus, the courageous person may risk her life to save a stranger, but only if she can correctly reason about what heroic acts will likely succeed. In this way, if the other virtues are of

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

8

 

II. The Christian Problem with the Terminal Conception Can the Christian simply accept the terminal conception of the virtues? After all, we accept a terminal understanding of other important and useful skills. We admire those who make the effort to become doctors and anesthesiologists, despite our awareness of the lower value those skills will have in a world without sickness and pain. Plenty of skills, from firefighting to grave robbing, are only valuable given our current mortality. Why not append virtue to the list? Indeed, even if we accept that the virtues have an extrinsic value that depends upon our mortality, we can still understand the virtues as having final value as an ultimate end of mortal human life.¹⁶ There is remarkable final value in Bach’s concertos, but that value depends on the extrinsic facts of human auditory processing. If there were no rational creatures with auditory senses, then the production of beautiful music would not be an important end of human lives. No one thinks the dependence of music’s value on auditory processing makes it less important as an end of auditory life. Thus, the extrinsic dependence of virtue’s value on mortality is even less a problem for that value than is the extrinsic dependence of medicine’s value on death. You might think that medicine is both extrinsically valuable and also merely instrumentally valuable as a means to keeping people from pain and death. By comparison, virtue could be extrinsically valuable and yet still a final end of human life.

less value, then prudence too is of less value. The reason I do not consider this argument in the chapter is because it would be redundant. If I can successfully answer the objections to courage, justice, and temperance, then this objection to prudence will be resolved as well. ¹⁶ Following Christine Korsgaard in “Two Distinctions in Goodness.” I am using “extrinsic” as a contrast word with “intrinsic” and “final” as a contrast word with “instrumental.” Something is extrinsically good if its goodness depends on something external to itself; it does not entail that the object is only valuable as a means to some further end. For example, I might find the possession of a treasured letter finally valuable, not just instrumentally valuable for containing useful advice, because it was the last letter my deceased parent sent me. While the letter is valuable for its own sake, its value depends on extrinsic facts and relations (my father dying when he did, who wrote the letter, etc.). In contrast, something is instrumentally valuable if it is not valuable for its own sake, but only as a means to a further end. It is not an end-in-itself.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

9

The Preliminary Problems While we can accept that the virtues are extrinsically valuable final ends, there are three reasons Christians should insist on an endurentist understanding of the virtues. The Inverse Relation of Virtue and Age First, the terminal conception could entail that our reasons to inculcate virtue taper off with age. I would advise students not to study musical composition if I knew that all humans would go deaf in a year. Similarly, I could advise those near death not to lay up for themselves virtues on earth, where germs and storms destroy and where murderers break in and kill, if I knew that they were near a future without death or pain. However, the idea that we have less reason to be concerned with our character as we age should strike us as both profoundly counterintuitive and unrecognizable as a principle of Christian ethics. Significance, Suffering, and Souls Second, much of Christian tradition has taught that the presence of suffering is, at least in part, a divine grace ordained for the development of virtue. Christian philosophers, for instance, have argued that soulbuilding is one of the chief explanations for why God allows evil. And even without endorsing a soul-building theodicy, the pastoral thought that one can achieve comfort by recognizing the refining power of suffering is prominent in Christian teaching. Understanding suffering as a means to the development of virtue plays an important role in enabling the Christian to see even suffering as a part of divine grace. If virtue’s value depends upon suffering and death, suffering and death cannot be understood as gracious means to virtue. A coach might put her athletes through a grueling practice. If the players ask why it is so difficult, then she could rightly respond that the grueling nature of the practice is important for equipping them with certain physical and mental dispositions. However, this is only a good response if those mental and physical dispositions are valuable for competitions. If these dispositions are not helpful for external success, but rather merely help one perform better in grueling workouts, then the athletes are left

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

10   without a good justification for the workout’s difficulty.¹⁷ A similar explanatory circle exists in the view of earthly suffering as a divine grace for purposes of character refinement if it turns out virtues are extrinsically valuable only in light of our earthly suffering. Elegance Is Everything Finally, on the terminal understanding of the virtues, we are left with a less elegant overall Christian account of the world. For most Christian accounts, both salvation and sanctification follow upon faithful conversion, but if virtue is primarily valuable because we will die, then once we are saved the value of sanctification begins to disappear. God grants us courage to face death, but only after we have acquired a faith that assures us death has lost its sting. God grants us temperance, but he grants it only to those who will be raised with an imperishable body. On this account, God holding out sanctification till after justification is like a doctor refusing to prescribe any symptomatic treatment until the cure has taken place. We lose our explanation of why sanctification is so central to the Christian life. We are ultimately destined for a life without suffering and death, and our ultimate good is found on that distant shore. We will spend an eternity in a perfected state and only a brief time in a fallen world. Why then does it seem so important for Christianity that we become equipped to handle this life well? If the virtues are ultimately important for our lives in the new creation, then it is easy to explain the Christian focus on virtue. But if the virtues are primarily valuable for our

¹⁷ You might think the way certain people work out is a clear counterexample to my claim here. While many work out either to improve performance in competition, or to improve physical health, others seem to simply pursue their own excellence in the workouts—take those who try to beat their personal records in running—and seem to do so just for its own sake. This can plausibly show that exercise can, at times, be a final, rather than instrumental, good. However, I do not think it is a plausible candidate for an intrinsic good. To see why, just note how bizarre the behavior would be if it was a practice that we did not think tied to any good outside the practice—for example, someone who works diligently to coat different shapes with lemon juice so he can become faster and faster at coating shapes with lemon juice. This behavior is unintelligible in a way that fanatical exercise is not, precisely because the lemon coater is pursuing a skill that does not seem to relate to any external good at all. Perhaps exercise is an end in itself, but it is only an end in itself because physical excellence is in turn tied to other forms of human well-being.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

11

lives now, then we lack a compelling explanation to counter Nussbaum’s conclusion that the need for virtue expires with us. There are at least three stories one might tell to reconcile Christianity with the terminal conception. All three fail.

Story One: That They Might Enter into Joy First, a Christian could argue that we can explain the value of the virtues, even as we approach death, because they are an entry condition into paradise. Neglecting virtue as one gets older would be like neglecting SAT prep as one enters junior year and justifying this choice by pointing out you won’t need to take the SAT in college. It is important that we have our souls refined, not so we can be virtuous in paradise, but so we can deserve to enter paradise at all. Even sola-thumping Protestants will acknowledge that only those with a regenerate faith enter salvation. The virtues this side of mortality, according to this story, are entry conditions for obtaining everlasting life and have no value afterwards. However, a picture on which the virtues are valuable as entry conditions renders those very entry conditions arbitrary. Consider two possible organizations: Mensa and Pensa. Both Mensa and Pensa have a strict IQ entry requirement; only someone deemed a genius on certain standardized tests can join. The difference between Mensa and Pensa is that Mensa meetings consist of intellectual discussions and puzzles, while Pensa is a billiards club where members spend most of their time discussing celebrity gossip. To enjoy and contribute to the Mensa meetings a certain intelligence is actually required whereas the same is not true of Pensa. If you asked someone why he was studying so hard for IQ tests, he could explain that he was trying to get into Pensa. However, that explanation would only answer your first question by raising a far more mystifying one: why does Pensa insist on such an entry condition at all? Trying to ground the value of the virtues in eternity by reference to entry conditions does not advance the debate in any meaningful way. Unless we have some reason to think these are fitting entry conditions (based on the value meeting those conditions will have after entry) we will not have an adequate explanation of the value of the virtues. Any

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

12   appeal to some good-making feature of demanding virtue, for instance the good of divine justice, would circularly assume the virtues are ultimately valuable and therefore a fitting object of demand.

Story Two: Away with These Instruments The Christian might next object that it is only the lower instrumental value of the virtues that passes away with our mortality. The final value of the virtues endures, and it is that final value that should chiefly motivate our concern with virtue. After all, if virtues were only instrumentally valuable, then those who are secure might be justified in remaining cowardly—a clearly unacceptable conclusion. The fact that the virtues are less instrumentally valuable in the new creation should not concern us any more than the fact that courage is less instrumentally valuable to those who are physically safe. The concern over a loss of instrumentality is irrelevant, this story argues, because it was never the instrumental value of the virtues that mattered, and so it was never their instrumental value we should have looked to in order to vindicate our endurentist conception. The problem with this second story is we could reconstruct Nussbaum’s challenge at another level. Rather than saying the virtue of courage loses its value because it is no longer instrumentally useful, we could instead demand an explanation for why courage has final value for a creature who has no need of courage. How might our imagined interlocutor answer this demand for an explanation of the final value of the virtues? Those who hold this view could provide two sorts of explanations. First, they could provide a Platonic explanation according to which virtues correspond with a metaphysically independent goodness, and thus in acquiring these virtues we more fully participate in the good. The Christian version of this Platonism would maintain something like the following: God himself possesses these virtues, though perhaps only analogically, and thus when we acquire these virtues we grow closer in nature and unity to God.¹⁸ These Platonic arguments treat the goodness ¹⁸ Arguments of this sort can be found in Adams’s Finite and Infinite Goods, 13–14; and Alston’s “What Euthyphro Should Have Said” (that piece by Alston is also extremely similar to what he says in “Some Suggestions for Divine Command Theorists”).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

13

of virtue as final. The extent to which you are courageous is the extent to which you resemble God, and the extent to which you resemble God is the extent to which you are good. Second, they could provide an Aristotelian explanation according to which the virtues are the excellencies of a distinctively human life because they are the dispositions which are essentially involved in our flourishing. A human is both a rational creature and a creature that experiences fear. Thus, for humans to be able to thrive consistently, we must find a way for our fears to cooperate with our rational judgments. The cooperation between our reason and fear is the virtue of courage. Rosalind Hursthouse, for instance, explains the value of the virtues by arguing that they are the most reliable bet for how to achieve happiness (though the virtues are neither necessary, nor sufficient for that end). Philippa Foot, in contrast, explains the value of the virtues by appealing to the Aristotelian necessities that accompany our form of life. Judgments like “the good human keeps her promises” are just like judgments of the form “a good bobcat breeds in spring.” They are a way of picking out a distinctive form of logical thought that attaches to our understanding of life-forms. What is distinctive in these Aristotelian arguments is that they treat the final goodness of the virtues as an extrinsic feature of the virtues. Courage is valuable because courage happens to be a part of the good life. However, for another form of life courage would not be a virtue at all. Neither the Platonic nor the Aristotelian versions of this second story will be able to answer Nussbaum’s argument that the value of virtue disappears in immortality. Troubles in Platodise There are two problems with Platonic attempts to secure the final value of the virtues in resurrected life by appealing to divine nature. First, if we cannot secure the value of the virtues in a mortal human’s life, a fortiori we cannot secure an account of what courage is in the divine life. If it is hard to understand what relevance courage could have for a creature who can never die, it will be even harder to understand what relevance courage could have for an impassible deity. The Platonic response is intended to show that we can still value the virtues in the new creation

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

14   even if Nussbaum’s argument succeeds in showing the virtues lose their instrumental value. However, if Nussbaum’s argument is successful, it will undermine one of the key premises of this Platonic response— namely, that God is courageous. Thus, the Platonic response of virtues being finally valuable because they make us more like God will succeed only if we already have good reason to doubt Nussbaum’s conclusion, rendering it unhelpful as a way to reconcile Nussbaum’s conclusion with Christianity.¹⁹ There is one way this argument could still advance the debate. Some Platonists might reject Nussbaum’s arguments for the strong conclusion—that it is impossible for us and for God to possess the virtues in the new creation—but might still accept my arguments for the weaker conclusion—that the virtues lack instrumental value in the new creation. Those in that position could argue God possesses the virtues as a brute fact of his nature and use that brute fact to explain the final goodness of our possession of the virtues in the next life. Even if we make courage brutely good, and bite the bullet on an explanatory picture where cowardice could have been the brute good instead, this solution still will not work. To see why, we need to turn to my second argument. There are presumably many ways we could resemble God, some of which are more and less relevant considering the sort of creatures we are. If there were a species that never experienced fear, perhaps a super-evolved species of ant, the virtues of that creature would probably resemble different features of God. Such creatures, which never see their personal good as separable from the good of the colony, might be able to resemble the unity of life and purpose that God partakes in within the trinity, and so would have the virtue of “trinitarian ¹⁹ Robert Adams, a defender of the sort of Christian Platonism I am suggesting, does suggest two reasons to doubt Nussbaum’s conclusion in his book Finite and Infinite Goods. First, he argues that the virtues only need involve “fragmentary resemblance” and so for those virtues that “seem to depend on our finitude, and . . . our mortality,” we need only resemble God in “what we care about, and the strength and effectiveness of our caring” (31). Second, he suggests that there might be “analogues in God of some of the limitations of human life.” For one thing, God may have taken on finitude in the incarnation, and even beyond the incarnation “the act of creating might carry with it a liability to frustration.” As I discussed in Part I of this chapter, I am already inclined to reject Nussbaum’s strong conclusion that finitude is necessary to possess, rather than value, the virtues. I expect something like Adams’s arguments succeed as a response to Nussbaum, and thus will focus my attention on the second issue with the Platonic solution.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

15

cooperation” rather than courage. Within the Platonic story, we still need to explain why it is bad for us to lack courage, and not bad for the ants to lack courage. The most natural explanation is that courage plays a positive role in our form of life but does not in the ant’s form of life. To put this problem another way, note that the virtues are only one of many ways that we can resemble God. God is a great artistic creator, and certainly the production of great art is a final, and likely even intrinsic, good. Nevertheless, I should be more concerned that I develop temperance than that I develop great artistic skill. Similarly, God has a deep understanding of the underlying nature of our world, and certainly the careful study of scientific reality is a final, and likely even intrinsic, good. Nevertheless, I should be more concerned that my children develop into just people than that they develop into great scientists. The virtues are a special subset of all the ways we can resemble God. They are not optional for a human life like other finally good resemblances. Thus, the Platonist must provide an explanation for this special significance of the virtues, and the only plausible explanation is that virtues are necessary for a good human life in a way those other excellencies are not. Thus, if Nussbaum’s argument shows those virtues are not necessary once death is no more, the Platonic rebuttal cannot differentiate the non-optional virtues from other ways we can resemble God in the new creation.²⁰ ²⁰ While I was revising this chapter, I discovered that in A Theory of Virtue, Robert Adams defends a Platonic account of the sort I am sketching and explicitly extends moral virtue to include “concern for quality in intellectual and aesthetic pursuits for their own sake” (20). He acknowledges that good taste is a “minor” moral virtue, but it is a moral virtue nonetheless (21). Adams acknowledges that there is an asymmetry between the standard moral virtues (like justice) and aesthetic virtues. You can be morally good without good taste, though you cannot be morally good without justice—despite good taste and justice both being moral virtues (21‒2, 36). Put another way, deficiencies in some virtues constitute vice (e.g., a deficiency in self-control constitutes the vice of incontinence), but deficiencies in other virtues do not (37). Adams explains this asymmetry by arguing that our interests in cultural goods, such as philosophy or art, require us to relate well to other people (because our interest in those subjects involves a desire that people participate in those goods)—those with aesthetic virtues don’t just value the existence of great art, but great art witnessed and enjoyed by people (45‒6). This asymmetry entails that one cannot properly love art without properly loving people who enjoy the art. One can, however, properly love people without properly loving the art those people enjoy (46). This solution is inadequate, and Adams even seems to acknowledge as much (47). It may be true that good taste requires justice while justice does not require good taste. But surely the fact that injustice vitiates taste does no explanatory work in explaining why justice is a non-optional

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

16   The Inadequacy of Aristotle It is easier to see why the Aristotelian approach to the virtues will not provide a way to counter Nussbaum’s conclusion. While Aristotelian accounts maintain the final value of the virtues, they acknowledge that virtue’s value is extrinsic and depends on the role those virtues play in human lives.²¹ Courage, says the Aristotelian, is a finally good feature of a human life—even if that very courage leads to a painful death—because courage somehow relates to life’s flourishing. Courage would be finally valuable, the story goes, only because courage makes a human life go well. However, if Nussbaum is right, courage only makes a human life go well because we are mortal creatures. Thus, if we are no longer mortal in the new creation, then this radical sort of change seems to alter what sort of dispositions have final value for our lives. Our Aristotelian interlocuter might counter that we remain human even in the new creation. And thus, distinctively human virtues are still final goods, even if we happen to be in a situation where they don’t benefit us (just as temperance can be a finally good virtue, even when stranded on a desert island where it serves no role in preventing overindulgence). There are two problems with this rejoinder. First, what constitutes a virtue can change across periods of a creature’s life. The dispositions that constitute excellence in infancy, for example polyphasic sleeping, do not constitute excellence in adulthood. The sort of metamorphosis we undergo in death and resurrection is at least as significant as the change we undergo in puberty. Second, if we are holding our nature constant across death, then why does what makes a pre-death life go well define the final goods for a post-resurrection life? Rather, shouldn’t a post-resurrection life define the final goods for a pre-death life? After all, it is only in the new creation that we achieve our final telos,

virtue. If good taste were entirely independent of justice, it would still be the case that injustice and cowardice are moral vices. Justice and cowardice are vices in their own right and on their own. Thus, the dependency of good taste on justice cannot explain the asymmetry. Adams concludes that even if his explanation is incomplete, other accounts “are no better off ” (47). As such, I will return to this asymmetry in footnote 27 after I have presented my own solution. ²¹ Hursthouse, On Virtue Ethics, chaps 8 & 9; Foot, Natural Goodness, 44–7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

17

which seems to suggest we should look to our post-resurrection life in order to identify the finally good virtues.

Story Three: Approximations and Preconditions A Christian might finally object that the pre-death virtues could simply be effective approximations or preconditions of some other dispositions we will require in the new creation. Butler suggests, “there can be no scope for patience, when sorrow shall be no more: but there may be need of a temper of mind, which shall have been formed by patience.”²² On this view, patience is a virtue needed for this life, but patience bears a striking resemblance to “patience 2.0,” some unknown disposition that will be needed for the next. And because we are not yet in the next life, the development of patience here and now may be the best way to develop patience 2.0. Although Butler never extends this treatment to other virtues, such an extension seems plausible. On Butler’s view, we have an endurentist conception of virtue, but a terminal conceptional of the particular virtues like courage and temperance that we inculcate here and now. The trouble with this solution is that it is blurry—and bad vision is a notorious problem for needle-threading. If courage is too similar to courage 2.0, then Nussbaum’s argument will apply to courage 2.0 as well. If courage is too dissimilar to courage 2.0, then courage 2.0 is not a plausible explanation of the value of courage. Either way, Nussbaum’s argument presents a problem. For Butler’s solution to succeed, we must, for each original virtue, postulate a virtue 2.0 at just the right conceptual distance. This solution is vague to the point of being unhelpful. By hypothesis, we cannot give much explanation of what the virtue 2.0s are like, since were they the sort of thing we could currently understand then presumably God would have directly instructed us in the development of virtue 2.0 during this life. We are left with a dubious punt to the unknown as a solution to Nussbaum’s challenge.

²² The Analogy of Religion, 202.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

18   Our first story attempted to make plausible a Christian terminal conception of the virtues by explaining virtues as entry conditions. Our second story attempted to vindicate an endurentist conception of the virtues by grounding the value of the virtues in something that will persist even through death. Our third story tried to split the difference by distinguishing our terminating mortal virtues and our enduring heavenly virtues. Unfortunately none of these approaches, at least on its own, is quite satisfactory.

III. A Workable Endurentist Account Virtues involve complex dispositions governing both how we act and what we enjoy.²³ Yet most discussions of the virtues focuses on how they dispose us to act. Virtuous actions, in turn, seem to make the biggest difference in conditions of suffering and need. For example, the generous tend to give more to others than the selfish do, and generous giving tends to be most needed when others are in conditions like poverty. Virtues however dispose us to more than merely actions. In particular, the virtuous differ from the vicious in what they enjoy.²⁴ The virtuous tend to enjoy the good features of the world, while the vicious enjoy very particular deviations from the good. For example, a generous friend will tend to rejoice over news of my promotion. In contrast, an envious friend may grow bitter upon hearing the same news. While the benefits of virtuous action are most clear in non-ideal conditions, the benefits of virtuous enjoyment are clearest in good, or ideal, conditions. The virtuous are disposed to enjoy the good, and so we

²³ And no doubt more besides. ²⁴ An analysis of enjoyment, while important to complete this account, is outside the scope of this chapter. I expect we have a pretty similar shared intuitive sense of enjoyment as more than merely felt pleasure (of the sort you can produce by electrocuting a rat’s brain) and including something like the welcome entering into activity. Most, it seems, describe states of flow as enjoyable but not really hedonistically pleasurable. I expect there is some weightier way to explain enjoyment in terms of the role it plays in flourishing, but since I’m currently as confused about what it is to flourish as I am about what it is to enjoy, for now it does not seem like a particularly helpful way to explain the concept.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

19

should expect a correlation between goodness and the enjoyment of the virtuous. It is not surprising that when philosophers like Nussbaum look for the value of the virtues in the new creation, they look for potential need of virtuous action. That is the natural place to look, because in our world the advantage of virtue is most clearly seen in action. For example, if you think about the primary evils caused by injustice, you rightly focus on the horrors done to those who suffer injustice, rather than the inability of the unjust to enjoy certain pleasures.²⁵ Further, because the virtuous are pained more by news of injustice than are the vicious, the prevalence of ill-news in our world means the virtuous are often pained disproportionately as a consequence of their justice. However, while it is tempting to look for advantages of virtuous action in the new creation, that’s the wrong place to look. In the new creation, the unjust will no longer abandon others to hunger and chains; when earth is made as heaven, the coward will not stand idle before bullying and harassment; and indeed, upon reaching that distant shore, the glutton will no longer indulge in injurious consumption, even if it is an injury done only to himself. The new creation differs from our world in being ideal, and actions tend to be most needed in non-ideal conditions. If we instead look to the difference that virtue makes in what we enjoy, it becomes clear why the virtues are enduringly important in the new creation. The ideal conditions of the new creation are good conditions that can only be enjoyed by the virtuous. Consider envy as an example. The envious person desires what is properly someone else’s. My desire that a thief return my bike does not constitute envy, nor does the impoverished’s desire that the wealthy send them aid. Desire for possession only rises to the level of envy when you desire what should be someone else’s.²⁶ In the new creation everything will be justly distributed,

²⁵ This focus on the evils that befall those who suffer, rather than do, injustice is the morally correct focus to have. First, because we have meaningful influence over the material conditions of those who suffer in a way we do not over the character of the vicious. Second, because God has disclosed himself as identifying, in a special sense, with those who suffer injustice—with the poor, the widow and the imprisoned. Thus, even if you agree with Plato’s arguments in The Gorgias and The Republic that it is better to suffer than to do injustice, still you should focus your assistance on those who suffer injustice. ²⁶ DeYoung, Glittering Vices, 43.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

20   as such the envious will be eternally dissatisfied while the generous will be perpetually content. We can generalize the problem into the dilemma of the vicious. In the new creation, everything will be as it ought. And while the virtuous are disposed to enjoy that way of being, the vicious will desire things be different. Either the vicious cannot shape the world after their desire, and thus are left with ever-present dissatisfaction and unhappiness, or the vicious can shape the world after their image (in certain highly limited ways) and to the extent that their will is efficacious they will be driven out of unity with the good activity of salvation. Thus, the value of the virtues in the new creation is simple: the virtues are simply those dispositions of character that render enjoyable one’s participation in the activity of salvation in ideal conditions.²⁷ Each cardinal virtue is necessary to enjoy one of the ideal features of the new creation.

²⁷ In footnote 20 I argued that Adams’s Platonic account cannot explain the asymmetry between the non-optional dispositions like courage and the optional dispositions like good taste. At this point, you might worry that my account has the same explanatory problem. Surely those with poor artistic taste or poor scientific wonder will be unable to enjoy certain ideal features of the new creation (such as the new creation’s ideal beauty). To explain this asymmetry, we need to note the difference between a lack of a particular enjoyment and finding something positively unenjoyable or distressing. If I lack good taste, I may lose out on some enjoyment of the beauty of the new creation. But there is no positive distress that pollutes my enjoyment of other goods. Consider an example. I have little architectural appreciation. Thus, if I am worshipping in a cathedral, I may miss out on the enjoyment of the beautiful building. In contrast, my architecturally talented friend may be sensitive to the majesty of the building and so enjoy something I miss. Nevertheless, my lack of positive enjoyment does nothing to prevent my enjoyment of other good things. Now consider a contrasting example. Suppose my friend is impatient. If we are worshipping in a cathedral, my friend will find the service positively unenjoyable. It will be unpleasant to be there for so long. This unpleasantness will, in turn, infect other sources of potential enjoyment. My impatient friend will likely be unable to even enjoy the architecture because boredom drives out such enjoyment. My position is that the virtues are necessary to enjoy the ideal conditions of the new creation. If you lack the virtues, you don’t just miss out on one source of enjoyment. Instead there is something positively unenjoyable or distressing about the state that you are in. In contrast, other excellences like good musical taste might give me further or unique sources of enjoyment, but the lack of that excellence does not make anything unenjoyable. Thus, we can explain the asymmetry between virtues and other excellencies by which we resemble God. Now, poor aesthetic taste can make things unenjoyable. For example, those without an appreciation of good wine often positively dislike wine. Still, it will not be the case that those who dislike wine are unable to enjoy paradise. After all, the new creation is unlikely to be systematically characterized by the forced drinking of wine. The virtues, unlike a taste for good wine, respond to those ideal conditions which pervade the new creation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

21

Distributive Justice There is also a distinctive way in which only the distributively just will be able to enjoy the new creation. Consider a stereotypically unjust person in paradise (a harrowing of heaven, if you will). While racism, for instance, would not perfectly pollute all sources of happiness, the racist person would be unable to enjoy fully the just equality of the new creation. Of course, there are many ways racism can manifest, but let’s start by considering a stereotypical self-professed white supremacist. No matter how much he receives, the racist person would be endlessly troubled that he is not receiving more, in either resources or status, than the otherwise equal person of a different race next to him.²⁸ A white supremacist does not merely desire the happiness or status of white people, but a happiness or status greater than that shared by black people. This sort of racism involves a desire for a wrongful superiority and hierarchy. This desire makes equality itself feel like insult and degradation. Two features will further exasperate the dissatisfaction. First, the white supremacist will likely see that he is not even treated the same as the people of other races. The consistent testimony of scripture is that the new creation will be characterized, not by perfect equality of status, but by an inversion of the status of those on earth—the first shall be last. The racist would likely have a lowly position in the divine kingdom, magnifying the feeling of indignity, ill-use, and dissatisfaction. Second, the racist person will not just be united in a hierarchy with those he considers unworthy. Instead, he will be bound together with them in an insoluble metaphysical unity, and the racist person finds interaction with people of other races contaminating. This attitude manifests itself in resistance and opposition to various types of social and legal unions, such as interracial marriage. However, a unity far more radical than marriage is involved for all people in the new creation.

²⁸ You might object that as long as your good is satisfied, racism would not cause you to resent the equal good of another. However, it seems in general this is not how we respond to goods. I can be perfectly content with my salary until I realize someone putting in less work in the vineyard is making the same amount as myself.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

22   According to Christian eschatology the saved will all be united as the body and bride of Christ. Of course, racism may manifest in forms other than outright white supremacy. Many of the advocates of apartheid in the Dutch Reform tradition in South Africa, for instance, seemed to couch their racism as a sort of spiritual condescension. They were not “sullied” by being around black people, rather they thought it was part of their higher and noble calling to condescend, and, by their assistance, to empower the black community to be self-sufficient. Yet this attitude too involves a desire that the world be different from how it will be in the new creation. Indeed, anything we can describe as “racism” will involve a preference that the world be structured in some unjust way, and as such, the racist person will find something unsatisfying in a world set finally right.²⁹ We are now in a position to identify the dilemma of the vicious in the context of distributive justice. The new creation will be characterized by a rightful hierarchy. The unjust will not desire to enter into that hierarchy. We are left with two possibilities. It may be that the racist person will be unable to separate himself from the radical unity and rightful equality of the new creation, in which case he will be eternally in a state he finds insulting and demeaning. Or else it may be that the racist person will be able to separate himself from the radical unity and the rightful equality of the new creation, in which case he will resist full participation in his ultimate good—participation as a member of Christ’s body. With the example of racism in mind, we can generalize the reason for thinking that being distributively just is necessary for full enjoyment of the new creation. The external distribution of goods and status in the new creation will be the rightful distribution. Distributive injustice

²⁹ There might, of course, be a merely intellectual racism. So we can imagine someone who believes in the inferiority of one race, not from a disposition to look down on others or from a disposition of prejudice, but just because of a reasoned response to a biased set of available data. It seems plausible, that if that intellectual racism does not infect the person’s will, such that he desires that inequality, that such a person could fully enjoy the new creation because he would abandon his racist beliefs immediately upon seeing the evidence provided by God’s kingdom. I expect that is right, and that it might show that intellectual racism that is not itself a result of biased evidence gathering (perhaps the sort of thing that could happen for some people 300 years ago) might not indicate a failure of justice proper (though of course such mistaken beliefs can have terrible consequences and so everyone should always be vigilant for where their reasonable beliefs might be false).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

23

causes the vicious to desire unjust distributions because they are motivated by favoritism. This will either cause the unjust to refuse to participate in their own place in the new creation or will lead to participation eternally marred by dissatisfaction.

Commutative Justice It is difficult to explain commutative justice as important for enjoying participation in the rightful hierarchy of heaven. The difficulty stems from the fact that commutative justice does not have the same sort of unity as distributive justice. Commutative justice, because it concerns what we owe to one another as private citizens, is difficult to delimit. I owe obligations of filial piety, of promise-keeping, of respect to excellence, of gratitude, of friendliness, of liberality, and the list can go on. Even Aquinas seems to have had a difficult time organizing the components of justice.³⁰ While many of these annexed virtues are clearly important for enjoyment of the new creation (such as deference to rightful authority), other related virtues may seem unnecessary (such as promise-keeping). In this section, I will focus on those components of commutative justice which seem to less naturally translate to the new creation. Thus, I will set aside virtues like deference to God’s commands, and instead focus on those virtues which chiefly govern our interactions with other people: dulia, gratitude, truthfulness, friendliness, and liberality.³¹ Dulia Aquinas contrasts the virtues of dulia and obedience with the vice of disobedience.³² It is easy to see why the vice of disobedience would be a problem in the perfect hierarchy of the new creation, and thus why the virtue of obedience would be necessary to fully enjoy salvation. Here on earth, being disobedient may be beneficial, such as when those in authority command poorly. However, in the new creation the disobedient will only ³⁰ Summa Theologiæ, II-II q. 80‒122. ³¹ II-II q. 102‒119. ³² Dulia can be roughly defined as right respect and esteem of worthy persons, such as saints.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

24   receive perfect commands that conduce perfectly to their good. Either the disobedient will grudgingly follow the commands, always resenting the authority and restriction, or they will break the commands, and so take actions that undermine their good. The obedient, in contrast, will be able to participate in the best activities of the new creation and do so happily. We can say something equally compelling about dulia. Contrast two characteristic ways people respond to those they consider their superiors. First, consider the adoring fan, overjoyed to be in the presence of someone she respects and admires. There is almost nowhere else the devoted fan would rather be than in the presence of the person she adores. Second, consider the resentful critic. This critic is bitter and selfdeceived, insisting, against all the evidence, that she is “just as good” as her superior. Clearly, the adoring fan will more fully enjoy her place in the heavenly hierarchy than the resentful critic. The fan finds the presence of those who are recognizably superior pleasurable in a way the resentful critic cannot. Further, the fan’s greater dulia explains her greater enjoyment. Those who esteem their superiors are those who enjoy the presence of those they find superior. There is a danger of moving too far beyond the fan, past dulia, and into secular idolatry or sycophantic worship of the unworthy. The secular idolater would also be unable to enjoy the new creation. This is because the secular idolater will be disinclined to enjoy the fullness of the heavenly hierarchy. Just as a foolish fan, who cannot recognize the superiority of the Beatles to their favorite second-rate band, is unable to enter fully into the deepest musical enjoyments, so too the secular idolaters will be unable to fully enter into the deepest enjoyment of the hierarchy of heaven. They will miss out on certain goods due to a settled insensitivity. Gratitude Aquinas contrasts the virtue of gratitude with the vices of ingratitude and vengeance. Gratitude would enhance one’s ability to enjoy the heavenly hierarchy because only the grateful will be able to fully enjoy their own place in that hierarchy. All the many and varied goods we receive as part

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

25

of our status in the new creation are free gifts given to us by God, and the grateful enjoy the use of gifts more deeply than the ungracious. There is an enjoyment that comes when you see something you own as a gift given you by someone you love or respect. There is an additional enjoyment, even if just a sentimental fondness, in reading a book my mentor gave me that would not have been present had I purchased that same book on Amazon; parents often cherish gifts created by their children far more than anyone else would cherish that same object purchased at an estate sale; and lovers will treasure jewelry given by their beloveds more than that same jewelry given by expartners. Being able to recognize our place in the new creation as a gift given by our beloved savior imbues our enjoyment of that place in an alchemical way. The ungrateful will not be able to enter fully into that enjoyment. This tends to happen in one of three ways. First, the ungrateful might be unwilling to recognize their position as a gift. They deceive themselves into thinking they are where they are merely in virtue of merit. Thus, they lose out on the enjoyment of their position qua gift. Second, the ungrateful recognize their place is a gift, and resent that fact and so refuse to accept the gift on offer. Thus, they refuse to enter into all the goods of the new creation and instead hold back to avoid being indebted to another. Third, they embrace these gifts but retain that resentment of needing to accept these things as gifts. Thus, their ability to enjoy their position is polluted by the constant resentment they experience. Therefore, the vicious, in this case the ungrateful, are unable to fully enjoy the new creation, either because they do enter into the activity of the new creation (by refusing the divine gifts or by failing to enjoy them as gifts) or because the experience of that activity is polluted by resentment. Truthfulness Aquinas contrasts the virtue of truthfulness with the vices of lying, hypocrisy, boasting, and false modesty. In general, Aquinas’s treatment of truthfulness concerns honest self-disclosure. Boasting and false modesty, for instance, both explicitly involve false pronouncements about one’s own virtue. Aquinas explains that “it belongs to the virtue of truth

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

26   that a person openly show his true self in outward signs.”³³ Self-disclosure is necessary to fully enjoy one’s position in the new creation because enjoyment comes in finding approval of oneself following full and honest self-disclosure. One might feel a smug sense of superiority having received an A on a plagiarized paper, but she will miss out on a feeling of proper pride at being truthfully told of the excellence of her work. The untruthful then face our old dilemma of the vicious. Either the untruthful are able to hide who they really are, and thus hide their position in the heavenly hierarchy (for it seems that your place in the hierarchy is reflective of the life you have lived). In that case, the untruthful will be unable to receive that approval we all crave. Or they will be unable to hide who they are, leaving themselves persistently upset at having their own past failings and faults made known to others. The extent to which we are horrified about parts of us, things we would never want anyone else to know, is the extent to which we will be inclined to hypocrisy. Thus, the untruthful are those who cannot enjoy the prospect of being fully known by others. Friendliness Aquinas contrasts the virtue of friendliness with the vices of flattery and quarreling. Friendliness relates to the outward signs of congeniality and affability that we bear to those with whom we interact. In this way, the friendly person is one who is “affable to the congregation of the poor” even when she herself is not poor.³⁴ Friendliness describes the disposition with which we act towards everyone, not just those with whom we are similar and share interests. Plausibly, friendliness is important for enjoying the heavenly hierarchy because we will not be able to self-segregate into groups of the likeminded in the fashion we do here on earth. We will end up rubbing shoulders with a far greater variety of God’s kingdom than we tend to within our earthly hierarchies (where professors regularly talk to professors, students with students, members of the upper class with members of the upper class, etc.). The hierarchy of heaven will not stratify us, if it stratifies us at all, in the same way our current society is stratified. ³³ Summa Theologiæ, II-II q. 111 a1.

³⁴ II-II q. 114 a1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

27

In the new creation the friendly and affable will be able to enjoy this varied company far more than the flatterer or the querulous. If you find yourself in a group of strangers, it is easy to tell who in that group is particularly friendly: you look around to see who is enjoying the random company. The friendly are far more willing to enter into activity with those who are different from themselves, and they enjoy that activity far more deeply, than the querulous do. Either the vicious will refuse to enter into the activity of the new creation because they are unwilling to enter into activity with those who are different from them, or when entering in, they will be unable to fully enjoy that activity because they tend to only enjoy the company of those with whom they are similar. Liberality Aquinas contrasts the virtue of liberality with the vices of avarice and prodigality. The liberal are ones who value their own possession in accordance with the measure of their need. They are not wasteful with resources that really are necessary for their good (as the prodigal person is), nor do they selfishly store up for themselves resources beyond what conduces to their good. Aquinas treats this desire in accord with the measure of one’s need as central to the concept of liberality: Consequently the human good in them consists in a determinate measure, namely that a person seek to possess material wealth to the degree that it is necessary to a life suited to his station. The sin is to go beyond this measure, namely the will to acquire or to hoard material goods excessively. The meaning of avarice, defined as unchecked love to possess, involves this and so avarice clearly is sinful.³⁵

The avaricious are concerned with possessing more than conduces to their own good, the prodigal less. There are two ways that liberality will be necessary to enjoy the ideal conditions of the heavenly hierarchy. First, the vicious will desire either to possess more than conduces to their good, or less than conduces to their good. If the vicious are able to act on their desire, then they will end ³⁵ II-II q. 118 a1. Emphasis original.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

28   up with an amount of possession that is not perfectly suitable to their need, harming their well-being. If, on the other hand, the vicious are unable to act upon this desire, then they will end up with a persistent dissatisfaction with their inability to possess more, or their inability to show off by giving away more than everyone else. Second, and more fundamentally, both the prodigal and the avaricious are characterized by an excessive self-focus. Covetous people will focus on themselves as the possessor of things, while prodigal people will focus on themselves as the giver of things. In both cases, you gain enjoyment when you yourself take center stage. This inordinate self-focus makes it far more difficult to enjoy the goods of the heavenly hierarchy. When you are not concerned that all needs are met, but instead that you are the one who provides what others need, then it is impossible to be satisfied in a world where others’ needs are met without your assistance. Similarly, the avaricious are not equally joyed at others finally having what they need and deserve. As such, covetous people are limited in sources of happiness to the goods they themselves have received. Those people characterized by the virtue of liberality lack this inordinate self-focus. They are pleased at the whole right distribution of goods in the heavenly hierarchy, even those goods from which they do not directly benefit and which they had no hand in providing. The sheer scope of their enjoyment is thus opened up by living in a world where each individual person’s right place in the heavenly hierarchy is an independent source of joy.

Courage If the virtue of justice is necessary to enjoy the rightful equalities and hierarchies of the new creation, courage is necessary to enjoy the bodily presence of God. It is only the courageous who will both be sensible to the majesty and holiness of God and yet still be able to enjoy his presence. Courage falls between the vicious extremes of cowardice and fearlessness (or foolhardiness). This can make the argument dialectically complex. Many potential reasons to think that cowardice would be bad in the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

29

new creation, such as the fact that the fear of God would drive us from his presence, would seem to entail that the fearless would be best situated to enjoy the new creation. Because of this dialectical complexity, I will start by looking at both cowardice and fearlessness separately, and then try to weave together the lesson about why courage is necessary for enjoyment of God’s presence. Cowardice Cowards struggle when they face danger or evil. But there is no longer any danger of judgment in the new creation, and God is certainly not evil. So why then do we need courage to stand before God in the new creation? First, one thing courage does is ensure that your rational judgment correctly controls what you fear. So, one manifestation of cowardice is the inability of your reason to control your appetitive fear. Even if there is no rational basis for fear in the new creation, the coward could still fear irrationally. After all, many who fear flying are not comforted when told flying is safe, nor do I cease to be frightened while in a haunted house even though I know the whole thing is made up. The clichéd refrain that we “fear the unknown” gestures at a tendency of appetitive fear. We tend to fear that which we cannot understand, and God is well beyond our comprehension. As such, the person who lacks control over their appetitive fear will be unable to enjoy the presence of God in much the same way that a child who is afraid of the dark is unable to enjoy stargazing. There is also a second, and deeper, reason we need courage to enjoy standing before God. My first argument was that even if we had nothing to rationally fear in God after judgment, we would still require courage to keep our appetitive fears in check. However, we still will have a rational ground for fear after judgment. As Mr Beaver reminds us, you cannot move from a premise about God’s goodness to the conclusion that God is safe.³⁶ Aquinas and Psalm 19:9 agree with Mr Beaver—both maintain that fear of the lord endures forever.³⁷ Aquinas distinguishes servile fear of ³⁶ Lewis, The Lion, the Witch and the Wardrobe, 80. ³⁷ Summa Theologiæ, II-II q. 19 a11.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

30   God—a fear we have only while under judgment—from filial fear, the sort of fear a child rightfully feels towards a loving parent, and argues that only filial fear will remain in the new creation. Why would there be filial fear of God?³⁸ When we trust someone whom we cannot understand, as a child trusts her parents, courage is still necessary because the unknowability of the other person precludes us from setting limits on what we will be required to endure—even for our own good. Note the difference between my trust in a friend and a child’s trust in her parents. If I trust one of my friends to pursue my own good, I can generally predict what she will end up doing. After all, my friend and I are similar in how much we understand about the world. The child has no similar parity with a parent. She has no idea how her parents see the world, and thus has no way to limit rationally how much her parents might ask of her. The last time our imagined child went to the doctor’s office she was given a painful injection; who knows what will have to be faced next time? She may trust her parents, but that does not mean she does not still rightly fear the pain they might require her to endure. Nevertheless, if she is courageous, she will be able to face that fear, confident in her knowledge that what she endures ultimately tends to her good. That same filial fear should characterize our relation to God, even in the new creation. God’s ways are so far above our ways that we have no idea what God will decide we must endure for our own good. Given this, we should rightly fear God because even potential pain and difficulty are fearful things. Our knowledge of God’s goodness means we should endure that which we fear, but courage is necessary to enjoy doing so. If it is right to think that God is a terrifying and awful being to behold, and yet beholding God is also the greatest possible good for a human creature, then it follows that courage is necessary to enjoy the ultimate human good. Cowardice will either drive me away from my ultimate good by rendering me unable to worship before the throne of God or will infect my experience of that worship by allowing fear, rather ³⁸ While the conclusion that filial fear remains in the new creation seems well attested by church tradition, the explanations of why filial fear continues are often unclear. I find even Aquinas’s explanation of the phenomena uncompelling. As such, I am far more confident in the conclusion that filial fear continues than I am in any given explanation, including my own, of why filial fear continues.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

31

than love or reason, to direct my attention. Either way, vice renders me unable to enjoy good in an ideal state. Fearlessness We still need to explain why it is better to be courageous than fearless, since, presumably, the courageous and the fearless will both be able to stand before God. The problem for the fearless is that fearlessness is a manifestation of a harmful insensibility. In his treatment of fearlessness, Aquinas argues that if I do not fear my own death it must be explained in one of three ways: by insufficient love of my own life, by “a swelled head” in thinking myself too good to be threatened, or by my “sheer dim-wittedness” in being insensible to the threats around me.³⁹ Given that God is God, it is a similar failing to not fear him at all. We could be fearless before a fearsome God only if we are insensible to his love or majesty. Because the fearless are insensible to God’s glory and grace, they stand in God’s presence only because they cannot grasp the full significance of that act of standing. This insensibility means that the fearless are unable to enjoy the full depth of the presence of God. They will have at most a shallow understanding of what it is to be in God’s presence.⁴⁰ In eternity, then, courage is necessary to fully enjoy the presence of God. On the one hand, we must be sensible to God’s fearful nature to experience the full enjoyment of salvation, but on the other hand, only the courageous will be able to enjoy that fear to the fullest.

Temperance If the virtue of justice is necessary to enjoy the rightful equality and hierarchy of paradise, and courage is necessary to enjoy the bodily presence of the Lord, what feature of the new creation is temperance

³⁹ Summa Theologiæ, II-II q. 126 a1. ⁴⁰ If there is a natural human desire to be in the presence of a good and loving God, that might explain some people’s peculiar enjoyment of horror movies. It is a corrupted manifestation of that desire to experience fear in context of true and assured ultimate safety.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

32   necessary to enjoy? Ultimately, the eternality of immortal life makes temperance necessary. Temperance involves a moderation of appetite. In developed nations, we tend to worry about how intemperance harms us when we overconsume. Both obesity and alcoholism, for example, harm the health by intemperate consumption. However, temperance also plays an important role in helping us enjoy our consumption. Temperance helps us to enjoy what we are consuming without being constantly bothered by inordinate desires for more. We fast, not as a form of effective weight loss, but as a way to develop a character that is properly satisfied with what one has. Temperance is valuable, not just when we have too much to eat, but also when we have too little to eat. Hunger makes it challenging to live well. However, it is a far greater challenge to the intemperate than it is to the temperate. This is a problem because the tendency of intemperance is to grow. The danger of indulging in intemperate appetite is not just that it will have harmful health consequences, but also that once indulgence becomes the norm it will take ever-increasing indulgence to provide the same levels of satisfaction. Imagine one snobbish grumbler and one appreciative person attending a meal. The appreciative diner will be able to enjoy everything that is served, taking some pleasure in the mediocre food and great pleasure in exquisite food. The grumbler, in contrast, will be unable to take any pleasure in the mediocre food and will find the exquisite food merely passable. Intemperance grows over time. Aquinas, in his discussion of intemperance, warns that when unbridled, “lust gathers strength,” and quotes Augustine as saying that lust “becomes custom when given way to, and custom becomes compulsion when not resisted.”⁴¹ When you extend this process for an eternity the result is eternal dissatisfaction. This leaves us again with the dilemma of the vicious. Either indulgences will be limited and kept from increasing, in which case the intemperate will be dissatisfied eternally by always wanting a little more than they have, or else the intemperate will be able to indulge a little extra each time, and so will be stuck with an unsustainable eternal ⁴¹ Summa Theologiæ, II-II q. 142 a2.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

33

growth of appetite. The intemperate by nature grow dissatisfied with what has already been consumed. No matter how varied the gustatory pleasures of the new creation are, there simply cannot be new pleasures forever, at least not pleasures, like new foods, that would satisfy the intemperate.⁴² Temperance is the key to the eternal enjoyment of unchanging pleasures.

Prudence If justice is necessary to enjoy the equality and hierarchy of the heavenly kingdom, courage is necessary to enjoy the presence of God, and temperance is necessary for eternal enjoyment, then prudence is necessary for our enjoyment of the new creation because participation in the new creation will be a matter of activity. Earlier, we presented two reasons why prudence seems less valuable in the new creation. First, prudence is concerned with picking the right means to our ends; since the new creation is our ultimate end, once it is satisfied, prudence will no longer be needed. Second, prudence is valuable principally for the avoidance of bad consequences, but there are fewer (or no) bad consequences in the new creation. Both arguments, however, are only plausible if we see our final end as a state of passivity rather than activity. Some of the ends we have in life are passive states. For example, if you ask someone to teach you how to reattach a button, then once your button is reattached you no longer have use for that practical skill (supposing you do not expect to need to reattach a button again). However, many of our ends are active states. Thus, if you want to be good enough to play baseball for your school, then you won’t cease to care about your baseball skills once you qualify. You want those skills so

⁴² It is worth mentioning, just quickly, why the other extreme opposed to temperance, namely insensibility, would also undermine enjoyment of the new creation. The reason for this is straightforward. The insensible are unable to see what is good and pleasurable about good and pleasurable things. Because happiness consists, at least partly, in the proper enjoyment of good things, the insensible are thus unable to experience many of the primary joys of the new creation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

34   that you can play baseball well, and thus you value your baseball skills just as much once you are selected for the team and start playing. Similarly, if I am concerned with being a good potential spouse, I don’t give up that concern at my wedding. At that point, the characteristics of being a good spouse become even more important. I didn’t want to be in the passive state of “being married,” instead I wanted to excel at the activity of marriage. We can extend this analogy to talk of the new creation. We do not value prudential excellence and practical wisdom just so that we can get to the new creation. We chiefly value practical wisdom so that we can excel in and enjoy activity while there. If you want to be an active participant in the projects of God, it will be important that you be able to excel at those activities. Bach and I could not meaningfully coauthor a piece of music because I cannot contribute positively to any musical project with Bach. This stems from my lack of musical wisdom. Similarly, if we want to participate in God’s new creation, and not just passively receive goods, we will need practical wisdom.⁴³ Those who lack prudence will either be forced to stay out of many of the activities of salvation, or else will participate but be unable to see themselves as true partners (in the same way I could not see myself as a true partner of Bach in the production of music). If skill and excellence are required to conduct music for a concert at the Kennedy center, how much more will it be required to produce music in the throne room of God? If skill and excellence are required to run a restaurant here on earth, how much more will skill and excellence be required to produce a celebratory feast attended by our Savior? While mistakes in the new creation will no longer result in death, that does not mean our activities are thereby less important. Instead, the ideal conditions of the new creation should enlarge the importance and significance of our projects and activities. Imprudence, then, also exhibits the dilemma of the vicious. Either the imprudent will be mere spectators within the new creation, and so lose out on their chief good of participatory activity with God, or else they will

⁴³ This discussion is, of necessity, vague. While I expect many of the activities of the new creation will be the same as the activities here on earth (producing art, cooking food, worshiping God, engaging in competition, studying nature, etc.) I do not know that for certain.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

35

nominally participate, but be unable to make a real difference in the contribution. Either way, the imprudent lose out on the special enjoyment of a good world that they had a role in shaping.⁴⁴ God does not want us to be mere passive receptacles of happiness. Instead, God intends that we take pleasure in a good world that we had a hand in making. There is a special savor in eating good bread that you helped bake, and it is one of His great graces that God does not just give us good bread but gives us that special savor by teaching us to help in the kitchen. As such, the imprudent will lose out on a final enjoyment. They can enjoy the ideal state in which they find themselves, but they cannot enjoy it as an ideal state which they helped build.

Conclusion Nussbaum presents a serious challenge to Christian ethics. The standard action-mediated way of thinking about the value of the virtues implies a terminal conception of virtue. This forces the Christian ethicist to find an alternative way of understanding the value of the virtues. That alternative is provided by identifying virtues with those dispositions necessary to enjoy ideal conditions, an account which fits well with an endurentist conception. This account would be strengthened if it were shown to help solve other problems. Thus I will conclude by noting four places in ethics and philosophy of religion where I believe further explanatory work can be done by this account. ⁴⁴ Here I should note my tendency to see the new creation as a state of “activity” which requires practical wisdom. On some accounts of the new creation, including Aquinas’s, our activity in the new creation tends to look rather cerebral. If paradise is merely the eternal contemplation of God in the beatific vision, then it is difficult to understand exactly what role we play in changing the character of that experience. While such a view of the new creation would prove challenging to my account of all of the cardinal virtues, it would be especially threatening to prudence. If the new creation is mere contemplation, then perhaps the enduring view of the value of the virtues is true for justice, courage, and temperance, but is not true of prudence. It could be that the value of prudence alone terminates. The story here is that while justice determines what ends we should choose, and courage and temperance allow us to receive goods (by keeping our irascible and concupiscible parts in check), prudence is only needed for positive activity, and so, when that activity comes to an end, so does the value of prudence.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

36   First, if the virtues are necessary to enjoy ideal conditions, that could help establish the rationality of virtue. On a merely secular picture it seems likely that at least sometimes an individual life is best served by vice. In contrast, on many religious pictures it seems like virtue is always the right choice, but only because God is animated by a troubling retributivism (that seems to be in at least some tension with divine forgiveness and mercy). If, instead, all humans are destined for an ideal condition that is enjoyable only to the extent one is virtuous, it becomes easier to see how one is always best served by virtue. God ensures that enjoyment proportions to virtue, not via strict retributivism, but simply by establishing a perfect kingdom. Second, this account suggests a way to fill in a troubling lacuna in Christian Aristotelianism. Many Christian ethicists—myself included— are inclined towards a neo-Aristotelianism that defines “goodness” in terms of species-specific excellencies. The problem for the Christian Aristotelian is to integrate that view of goodness with the Goodness of God, a goodness which seems far more Platonic. If the virtues are ultimately necessary for flourishing not in just any conditions, but in ideal (or good) conditions, then it suggests a plausible way to integrate Christian Aristotelianism and Christian Platonism. Human good is defined by those dispositions necessary for human flourishing (of which I assume enjoyment is a part) in certain conditions, and goodness of those conditions is cashed out in a species-independent Platonic fashion. I don’t think it is coincidental that the cardinal virtues are necessary to enjoy four conditions so closely connected to Being. (The presence of God involves unity with being itself. The rightful hierarchies and equalities of the heavenly Kingdom involve an ordering in terms of the great chain of being. The eternality of immortal life involves the maximal temporal extension of our being. And activity opposes passivity as the being of life/agency.) Third, this account suggests a non-retributive way to cash out the doctrines of paradise, purgatory, and damnation. The new creation will be an immediate paradise to the virtuous, since they, and they alone, can fully enjoy the conditions in which they reside. By contrast, the new creation will be eternally unenjoyable to those trapped in vice. Finally, the new creation will be presently (but only temporarily) unenjoyable to

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

37

those still in the process of sanctification; at the culmination of sanctification, those in purgatory enter seamlessly into joy. Finally, this account suggests a way to explain the heavenly hierarchy of the saved. According to Christ, the greatest in the heavenly kingdom will be those who were servants of all on earth. One way to explain that hierarchy is by stipulating that each person will receive different goods in the new creation—God will lavish special gifts on his most virtuous sons and daughters. This would explain the hierarchy, but there is something unintuitive in picturing God as holding back any good thing from any member of the Church. God has no reason to be stingy. Perhaps the hierarchy is better understood because the more virtuous can more fully enjoy those gifts that God generously lavishes on all. A picture of this sort concludes C. S. Lewis’s Chronicles of Narnia. All the friends of Narnia have finally entered the new Narnia. All of them see and experience the same good world. Yet, in this very equality, Lucy, the most faithful and virtuous, experiences new Narnia differently: [Tirian] looked round again and could hardly believe his eyes. There was the blue sky overhead, and grassy country spreading as far as he could see in every direction, and his new friends all round him laughing. “It seems, then,” said Tirian, smiling himself, “that the stable seen from within and the stable seen from without are two different places.” “Yes,” said the Lord Digory. “Its inside is bigger than its outside.” “Yes,” said Queen Lucy. “In our world too, a stable once had something inside it that was bigger than our whole world.” It was the first time she had spoken, and from the thrill in her voice, Tirian now knew why. She was drinking everything in even more deeply than the others. She had been too happy to speak.⁴⁵

⁴⁵ This chapter has been significantly improved by the feedback of others. I received helpful content suggestions during two presentations of this view. I first developed the view in conversation with my Fall 2018 Introduction to Philosophy class. I then presented the core ideas to those of the Florida State philosophy department who attended my lunchtime talk. Michael Bukoski deserves a special shout-out for the detailed feedback he provided both

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

38  

Works Cited Adams, Robert Merrihew. A Theory of Virtue: Excellence in Being for the Good. Oxford: Clarendon Press, 2008. Adams, Robert Merrihew. Finite and Infinite Goods: A Framework for Ethics. Oxford: Oxford University Press, 1999. Alston, William. “Some Suggestions for Divine Command Theorists.” In Divine Nature and Human Language: Essays in Philosophical Theology. Cornell Paperbacks. Ithaca, NY: Cornell University Press, 1989. Alston, William. “What Euthyphro Should Have Said.” In Philosophy of Religion: A Reader and Guide, edited by William Lane Craig, 283–99. New Brunswick, NJ: Rutgers University Press, 2002. Anscombe, G. E. M. “On Promising and Its Justice, and Whether It Needs Be Respected In Foro Interno.” Crítica: Revista Hispanoamericana de Filosofía 3, no. 7/8 (1969): 61–83. Aquinas, Thomas. Summa Theologiæ: Latin Text and English Translation, Introductions, Notes, Appendices, and Glossaries. Translated by Blackfriars. 60 vols. Cambridge: Eyre & Spottiswoode, 1975. Butler, Joseph. The Analogy of Religion, Natural and Revealed, to the Constitution and Course of Nature. To Which Are Added Two Brief Dissertations: I. Of Personal Identity. II. Of the Nature of Virtue. 3rd ed. Eighteenth Century Collections Online. London: Gale, 1740. Butler, Joseph, and David McNaughton. Fifteen Sermons Preached at the Rolls Chapel: And Other Writings on Ethics. First edition. British Moral Philosophers. Oxford: Oxford University Press, 2017. DeYoung, Rebecca Konyndyk. Glittering Vices: A New Look at the Seven Deadly Sins and Their Remedies. Grand Rapids, Mich.: Brazos Press, 2009. Foot, Philippa. Natural Goodness. Oxford: Oxford University Press, 2001. . during and after my presentation. I also received helpful grammar and writing feedback from Dean Thompson, Jane Thompson, and Mitali Perkins—those three, in addition to catching a frightening array of typos, improved the readability of the chapter enormously. I also am extremely grateful to Grace Thompson, John Schwenkler, David McNaughton, Nick Cisneros, Paul Rezkalla, and Spencer Smith, who each provided feedback on multiple drafts of the chapter. Finally, my deepest thanks are due to Bob Bishop and Tucker Sigourney who read through at least five complete drafts of this chapter and edited, at a conservative estimate, over 75,000 words. There is no part of this chapter that has not been significantly improved by Bob and Tucker.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

39

Hursthouse, Rosalind. On Virtue Ethics. Oxford: Oxford University Press, 2001. . Korsgaard, Christine M. “Two Distinctions in Goodness.” The Philosophical Review 92, no. 2 (April 1983): 169. . Lewis, C. S. The Lion, the Witch and the Wardrobe, 2010. . Nussbaum, Martha. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Rev. ed. Cambridge, New York: Cambridge University Press, 2001. Nussbaum, Martha. “Transcending Humanity.” In Love’s Knowledge: Essays on Philosophy and Literature, 365–92. New York: Oxford University Press, 1992. Nussbaum, Martha. The Therapy of Desire: Theory and Practice in Hellenistic Ethics. Martin Classical Lectures, , 2. Princeton: Princeton University Press, 1996.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

2 Being Good and Loving God T. Ryan Byerly

Questions about the relationship between religion and morality have been of perennial interest to philosophers of religion. Perhaps the dominant questions have concerned whether salient features of the moral life, such as the existence of moral obligations (e.g., Evans 2013) or the appropriateness of moral motivation (e.g., Adams 1979), are best justified by the existence of a religious reality. If they are, or if it is reasonable for a person to think that they are, then this would provide for one way in which a person’s moral commitments could justifiably generate or sustain their religious commitments. Their justified commitments to the moral life having the relevant features, together with their justified commitment to a religious reality providing the best justification for the existence of these features, could justifiably lead them to cognitively commit to a religious reality. In this chapter, I will explore a different and less well-trodden way in which a person’s moral commitments might justifiably generate or sustain their cognitive religious commitments. My focus isn’t on cases in which a person has certain views about the nature of the moral life and is persuaded by an argument that these views evidentially support the existence of a religious entity. Rather, my focus is on cases in which a person’s commitment to trying to live a morally good life leads them to adopt cognitive religious commitments. More specifically, I am interested in cases in which a person embraces these religious commitments as a manifestation of their trying to be a good person—cases where taking on these religious commitments is just part of what being a good person is justifiably taken to demand. I am especially interested in cases where the explanatory role of moral commitment makes a determinative difference. Roughly, these are cases T. Ryan Byerly, Being Good and Loving God In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © T. Ryan Byerly 2022. DOI: 10.1093/oso/9780192862976.003.0002

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

41

in which the religious commitments either wouldn’t have arisen or would have been lost had it not been for the explanatory influence of the relevant moral commitment. Because I think it is most plausible that moral commitments of the relevant sort would justifiably make this kind of determinative difference in cases where a person’s evidence is otherwise roughly counterbalanced with respect to the religious commitments in question, I will concentrate on such cases where the person’s evidence is roughly counterbalanced. I wish to examine whether a person’s commitment to being a good person might move them off such an evidential fence, so to speak, to embrace cognitive religious commitments, and whether it might do so in a way that renders these commitments justified in a moral, epistemic, or all-things-considered sense. I will focus specifically on the case of embracing commitment to the existence of a God of the sort envisioned in the major monotheisms. So my focus is on cases in which a person’s evidence is roughly counterbalanced regarding the existence of such a God. I will argue that there is a suite of character traits that a person might reasonably take to be partially constitutive of being a good person, where living in accordance with these character traits requires that in such evidential situations one cognitively commits oneself to God’s existence. Given the existence of such character traits, a person might justifiably commit themselves to God’s existence in these evidential circumstances as a manifestation of their pursuit of living in accordance with a morally good character. Moreover, given the nature of the relevant commitments to God’s existence as spelled out below, this will be appropriately described as a case in which a person’s attempt to be good will have led them to love God. Indeed, it will be a case where loving God was just part of what it was to be good, or at least where this was a justifiable view. Section 1 introduces the suite of difference-making character traits that are at center stage in this story about the justification of religious commitment. It explains the nature of these traits, and illustrates how possessing or pursuing them can make a difference for a person’s cognitive commitments not just to God’s existence but to a broad range of claims, concentrating on cases involving rough evidential symmetry in which these traits can make a determinative difference. Section 2 defends the moral value of the relevant character traits.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

42 .   Section 3 turns to the question of the justificatory status of commitments that arise out of the pursuit or possession and exercise of these traits. I argue that such commitments, including commitment to God’s existence, are typically morally justified and all things considered justified, and they may even be epistemically justified.

1. Praisefulness, Thankfulness, and Contrition There are various ways that the character traits I have in mind might be conceptualized as following a unified pattern or contributing to a unified ideal. All of them can be thought of as tendencies to err in one way rather than another. And, as I will discuss in more detail in Section 2, the direction toward which they tend to err is a direction that tends to be conducive toward cultivating valuable interpersonal relationships. So, we might think of them as pro-relationship character traits. Moreover, while the examples of these pro-relationship character traits I will focus on in this section may at first strike some readers as rather narrow features of character, I will explain in Section 2 how they are related to broader features of character equally concerned with promoting valuable interpersonal relationships. As I will argue there, these narrow features of character have a fitting place within a broad, virtuously other-oriented character. While there are many candidates for such traits, I will focus my attention here on three that I call, respectively, “praisefulness,” “thankfulness,” and “contrition.” I use these labels stipulatively rather than in an effort to analyze some pre-theoretic phenomena already generally recognized using these terms. In speaking of “thankfulness,” for example, I should not be understood as offering a rival conception of the trait of gratitude which has received extensive attention from philosophers and psychologists, though thankfulness as I will conceive of it is closely related to gratitude as it is commonly conceived as I spell out further below. Praisefulness is stipulatively defined as a tendency to err on the side of giving credit to others for their accomplishments, rather than refraining from giving such credit. The praiseful person would rather give credit

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

43

when credit isn’t deserved than refrain from giving credit when credit is deserved. They are more tolerant of erring by offering credit when it isn’t due than they are of erring by failing to offer credit when it is due. The credit the praiseful person tends toward giving they tend toward giving sincerely. Theirs isn’t a tendency to feign giving others credit for their accomplishments, but a tendency to sincerely give credit. Nor is theirs a tendency to give others more praise than their accomplishments would merit, but a tendency to err on the side of giving others the praise their accomplishments would merit—if indeed they are accomplishments. The praiseful person therefore tends to err on the side of sincerely giving others credit commensurate with their accomplishments, rather than the side of refraining from giving others credit commensurate with their accomplishments. In the contemporary philosophical literature, the construct that is most closely related to praisefulness so conceived is appreciation. The kind of appreciation most commonly discussed by philosophers is aesthetic appreciation—the appreciation of beauty (e.g., Budd 2002). But aesthetic appreciation is just one kind of appreciation. Tony Manela (2016), for example, notes that there are also cognitive, ethical, and prudential kinds of appreciation. While noting that “there is no consensus philosophical account of what appreciation is,” Manela suggests that what is common to all these forms of appreciation is that they are each “a mode of valuing, that is, a certain kind of response to something good” (289). Accordingly, Manela proposes that appreciation includes both cognitive elements and affective elements. Focusing on the case of prudential appreciation—appreciation for the good things in one’s life—he proposes that “when I appreciate [such things], I do more than just get right certain facts about the value of those things; in addition I enjoy them as well” (289). Similarly, I propose that the praiseful person errs on the side of adopting a stance toward others’ achievements that includes both positive cognitive and positive affective elements. The stance includes both a positive cognitive stance toward the achievements as achievements and an appropriate positive valuing of those achievements as such. Without adopting such a stance their praise of others’ achievements would not be sincere, as it would not express the attitude of appreciation that sincere praise

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

44 .   expresses.¹ The praiseful person thus errs on the side of adopting a positive cognitive and affective orientation toward others’ achievements rather than refraining from adopting such an orientation. They err on the side of offering others sincere praise for their achievements as an expression of appreciation of those achievements. Thankfulness is structurally very similar to praisefulness. It is a tendency to err on the side of giving thanks to others for the valuable things others have done for one, rather than refraining from giving such thanks. The thankful person would rather give thanks when thanks isn’t deserved than refrain from giving thanks when thanks is deserved. They are more tolerant of erring by offering thanks when it isn’t due than they are of erring by failing to offer thanks when it is due. The thanks the thankful person tends toward giving they tend toward giving sincerely. Theirs isn’t a tendency to feign giving others’ thanks for their help, but a tendency to sincerely give thanks. Nor is theirs a tendency to give others more thanks than their help would merit, but a tendency to err on the side of giving others the thanks their help would merit—if indeed they have given such help. The thankful person therefore tends to err on the side of sincerely giving others thanks commensurate with the benefits they have given, rather than the side of refraining from sincerely giving others thanks commensurate with the benefits they have given. While thankfulness is structurally similar to praisefulness, it is not the same trait, nor is thankfulness a subordinate species of praisefulness. To recognize and value the excellence of someone’s performance in the way characteristic of praisefulness is a different thing from thanking them for the contribution this performance made to one’s own well-being. Giving thanks is not just what someone who has benefited from another’s excellent performance does when they recognize the excellence of that performance; it is its own distinctive sort of activity. Giving thanks involves adopting an orientation toward another as one’s benefactor and not just an orientation toward an achievement of another. ¹ Here I follow the “orthodox” philosophical approach to sincere speech acts according to which a speech act is sincere just in case it expresses the appropriately related mental state. On this conception, for example, sincere assertion expresses belief, while sincere thanksgiving expresses gratitude. For discussion and references regarding this orthodox conception of sincerity, see (Eriksson 2011).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

45

There is a large philosophical literature on gratitude that is directly relevant for informing our conception of thankfulness. Where the attitude expressed by sincere praise for others’ accomplishments is appreciation of those accomplishments, the attitude expressed by sincere thanks for the benefits others have given to one is gratitude. According to a growing consensus, such gratitude is understood by philosophers to have a “to-for” structure (see Manela 2015, section 1). The grateful person is grateful to their benefactor for the benefit they have received from them. As with appreciation, philosophers generally agree that gratefulness includes both positive cognitive (e.g., Berger 1975; Walker 1980) and positive affective (e.g., Fitzgerald 1998; Bruton 2003) elements. Likewise, I propose that the thankful person tends to err on the side of adopting a stance toward benefactors that includes positive cognitive and positive affective elements—chiefly, the positive cognitive recognition of benefits these benefactors have given them and positive valuing of their benefactors as sources of these benefits. The thankful person would rather offer sincere thanks that expresses such a stance when such a stance is not merited than fail to offer sincere thanks expressing such a stance when such a stance is merited. Whereas praisefulness and thankfulness so conceived govern how a person approaches certain positive behaviors of others, contrition governs how a person approaches their own negative behaviors. The contrite person errs on the side of apologizing for wrongs done to others. They would rather apologize when an apology is not warranted than fail to apologize when an apology is warranted. They are more tolerant of erring by offering an apology when it is not warranted than they are of erring by failing to offer an apology when it is warranted. The apologies they tend toward giving are sincere apologies, and apologies commensurate with the wrongs done. Thus, theirs is a tendency to err on the side of giving sincere apologies commensurate with the wrongs they’ve done rather than the side of refraining from giving sincere apologies commensurate with the wrongs they’ve done. The philosophical literature most relevant to contrition so conceived is the limited literature on apology. As with gratitude, apology has a “to-for” structure; a person apologizes to someone they have wronged for the wrong they’ve done. Sincere apologies are typically regarded as

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

46 .   including both cognitive and affective elements. Radzik and Murphy write that “a well-formed apology requires at least acknowledgement of both the fact of wrongdoing and responsibility by the wrongdoer, as well as an expression of regret or remorse” (2015, section 3.1). Here the cognitive element is positive while the affective element is negative. Accordingly, I propose that the contrite person errs on the side of adopting a stance toward their wrongdoing that involves a positive cognitive recognition of this wrongdoing as such and a negative affective evaluation of this wrongdoing as such. The contrite person would rather offer a sincere apology expressing such a stance when none is called for than fail to offer such an apology when it is called for. Let the foregoing suffice for a brief account of the nature of praisefulness, thankfulness, and contrition. I will concentrate my discussion on these three candidates for character traits that can fulfill the role of justifying commitment to God’s existence in cases of roughly counterbalanced evidence regarding God’s existence. But, as I have indicated above, there may be other good candidate traits for fulfilling this role as well. For example, we might imagine tendencies to err on the side of seeking the council of those who love one prior to acting, or tendencies to err on the side of sharing one’s joys and sorrows with those who love one, or tendencies to trust the promises of those who love one. While I will concentrate on the traits of praisefulness, thankfulness, and contrition, I invite readers to consider how such other traits might play a similar role in the story of the justification of religious commitment. My task in the remainder of this section is to illustrate how the possession or pursuit of praisefulness, thankfulness, and contrition can make a difference to the commitments a person holds, including commitment to God’s existence, especially in cases of rough evidential symmetry. Let us begin with a more mundane case. Imagine you are watching the final seconds of a basketball game with a tie score. An offensive player gets the inbounds pass and dribbles down the middle of the lane. They’re swarmed by the defense, so much so that you can hardly tell what’s happening. The crowd stands on its feet, further obscuring your view. What you are able to see clearly, though, is the ball popping up out of the crowded lane toward the basket, bouncing about, and falling in.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

47

Details about the case could be fleshed out in various ways. Let’s suppose, though, that the offensive players had spread the floor and so there was no other offensive player in the lane. And let’s stipulate that your ability to see the events was affected in just such a way that your evidence was exactly counterbalanced regarding whether the player who drove into the lane deserved credit for having made a winning shot. You have exactly as much reason for thinking the player does deserve credit for making a winning shot as for not thinking this. Our question is whether characterological features of the sort in view in this section might make a difference for your cognitive commitments in this case. Here it is the first character trait of praisefulness that is relevant. Suppose that you have a tendency to err on the side of giving others credit for their accomplishments. You’d prefer to give credit when it isn’t deserved than fail to give credit when it is deserved. Since you’re otherwise on the fence in this case regarding whether or not credit is deserved, this trait of praisefulness could indeed make a determinative difference for what you do. If you are praiseful, you will tend to offer sincere praise to the player for their having made the winning shot. Moreover, as we saw above, offering sincere praise in this case will require adopting a positive cognitive and affective stance toward the player’s achievement. It requires a positive cognitive stance toward the player’s deserving credit for having made the winning shot, and a positive evaluative stance toward this achievement as an achievement.² Now, this isn’t yet to have specified the exact nature of the cognitive commitment you must have. And here we may wish to take a liberal attitude—somewhat more liberal than is reflected in the literature on appreciation cited above. Specifically, we may not want to insist that any particular positive cognitive attitude is required for the praise to be sincere, as there may be multiple distinct positive cognitive attitudes that are sufficient. On this point, Daniel Howard-Snyder’s (2013; 2019a; 2019b) recent work on propositional faith is relevant and helpful. Howard-Snyder maintains that faith that p requires a positive cognitive

² Some readers may question whether adopting the cognitive commitment here is psychologically possible. I discuss this topic in Section 3 when considering the epistemic justification of such commitments.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

48 .   stance toward p, but he demurs from the suggestion that this positive cognitive stance must involve some particular positive cognitive attitude such as belief, instead proposing that any positive cognitive attitude can stand in for belief. Regarding the variety of positive cognitive attitudes that might prove serviceable, he writes: Notice the plethora of folk psychological terms for positive cognitive stances: “acceptance,” “acknowledgement,” “affirmation,” “assent,” “assumption,” “belief,” “confidence,” “conviction,” “credence,” etc. Although it would be hasty to suppose that each term stands for a different stance, it would be equally hasty to suppose that every term stands for the same stance. Interestingly, many philosophers think some of them stand for different stances. (2013: 361)

Indeed, preceding Howard-Snyder’s article, Alston (1996) and Audi (2008) had proposed to distinguish between belief and acceptance. Howard-Snyder himself goes on to distinguish assuming from each of these, and John Schellenberg (2005) has proposed to distinguish assent from all of the above. In accordance with this sort of openness to the plurality of positive cognitive attitudes, we might allow that adopting the positive cognitive stance required by sincere praise needn’t require adopting any particular positive cognitive attitude, but rather that it requires adopting some positive cognitive attitude or other. More specifically, it requires adopting a more positive cognitive stance toward the relevant claim than toward its negation. In the basketball case, sincerely praising the player requires adopting a more positive cognitive stance toward the player’s deserving credit for having made the winning shot than toward the player’s not deserving such credit. It requires, if you will, a cognitive leaning toward the player’s deserving credit. You might sincerely praise the player by assuming that they made the shot, or by accepting that they did, or by affirming or assenting to this, rather than by believing it.³ Acting in accordance with the character trait of ³ In partial defense of this idea, consider that describing the basketball case as one of beliefless assuming seems about as coherent as the descriptions of focal cases that HowardSnyder gives when illustrating and arguing for the phenomenon of beliefless assumption, such as his cases of the defensive captain and the army general (2019b: 9).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

49

praisefulness in the basketball case will push you toward adopting some such cognitive commitment toward the player’s achievement. It is notable that the praisefulness tendency needn’t be particularly strong to make this difference. We can contrast people who very strongly prefer offering sincere praise when it is not deserved to failing to offer sincere praise when it is deserved, on the one hand, with people who have much more slight preferences of this kind, on the other. The stronger preference is apt to make a difference in cases unlike the present one where the evidence is less friendly toward the player’s deserving credit for making the shot. A person who is strongly averse to failing to give praise when it is due may prefer taking on a greater risk in getting it wrong in offering sincere praise, for example praising the player for making the shot when their evidence is not so close to counterbalanced that the player made the shot. A weaker tendency toward praising would not push a person to offer praise in such a case, though it is sufficient to do so in cases approaching, including, or exceeding evidential symmetry. Structurally similar cases illustrate how even weak varieties of thankfulness and contrition can make a parallel difference for a person’s commitments in cases of rough evidential symmetry. A thankful person whose evidence is roughly counterbalanced regarding whether someone else has benefited them will tend toward sincerely thanking them for the benefit. Perhaps the supposed benefactor was aiming to give the benefit undetected, but couldn’t avoid leaving just enough evidence for the beneficiary’s evidence to be counterbalanced regarding whether they had given the benefit. Here the thankful person would rather err on the side of offering sincere thanks to the supposed benefactor. Yet, sincerely thanking them requires adopting a positive cognitive orientation toward this other’s having benefited them. So, their thankfulness can here make a difference to which cognitive commitments they take on. Similarly, the contrite person whose evidence is roughly counterbalanced regarding whether they have wronged another person will tend to sincerely apologize to this other, preferring to sincerely apologize when no apology is necessary than to fail to apologize when an apology is necessary. Here we might imagine that you’ve been hashing over the details of whether you have wronged your partner for some time, and you’ve reached the point that your evidence that you’ve wronged them is

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

50 .   roughly counterbalanced. Acting in accordance with contrition will push you toward offering a sincere apology. Yet in sincerely apologizing you will need to take on a positive cognitive attitude toward your having wronged your partner. Additionally, we might imagine that even those who possess only weak general preferences in favor of praising, thanking, or apologizing will more strongly prefer praising, thanking, or apologizing when that which is prompting the praise, thanks, or apology is of greater absolute value. For example, a person with a weak general preference for praising may have a stronger preference for praising greater accomplishments than lesser accomplishments. They might, for example, be more strongly disposed to give sincere praise for the game-winning shot than for a shot made in the second quarter. Likewise, a person might have a stronger preference in favor of apologizing if the wrong at issue is more egregious. In this way, tendencies of the kind in view here that are generically weak may alter a person’s commitments in cases further away from an evidential symmetry if these are cases where what prompts the praise, thanks, or apology is of sufficient absolute value. We’re now in a position to explain how a person’s possession or pursuit of character traits such as praisefulness, thankfulness, and contrition could manifest in their adopting or maintaining commitments to God’s existence in cases where their evidence is roughly counterbalanced regarding God’s existence. For, we must note that a person whose evidence is roughly counterbalanced regarding God’s existence is also a person whose evidence is roughly counterbalanced concerning whether God has achieved praiseworthy accomplishments, has benefited them in ways deserving thanks, and has been wronged by them. For a God of the sort envisioned in the major monotheisms is one who by definition has created the cosmos and ordered it to benefit human beings, all of whom this God loves. If God has created the cosmos and ordered it to benefit human beings, then God deserves praise and thanks from these human beings; and if God loves all human beings then any mature human being will have wronged God if in no other way than by wronging those God loves. Thus, for any mature human being, to have counterbalanced evidence that God exists is to have counterbalanced evidence that God deserves their sincere praise, thanks, and apology.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

51

For a person with roughly counterbalanced evidence that God deserves praise, thanks, and apology, possessing or pursuing the traits of praisefulness, thankfulness, and contrition can make a determinative difference for their commitments regarding God’s existence. For, if they are to act in accordance with these traits in this evidential situation, they will offer sincere praise, thanks, and apology to God. But to offer sincere praise, thanks, and apology to God requires taking a positive cognitive stance toward God’s having created and benevolently ordered the world, and toward one’s having wronged God. Taking such a positive cognitive stance, moreover, requires taking a positive cognitive stance toward God’s existence. Nor, as we’ve seen, need the traits of praisefulness, thankfulness, or contrition be particularly strong in order to generate this result. Weak preferences for praising, thanking, or offering apology can generate this result in cases of rough evidential symmetry. Yet, in this particular case we might expect the force of these traits to be even greater than in other cases in which they make a difference. For, first, the traits operate in concert here whereas they can operate independently in other cases. For example, whereas only praisefulness pushes you toward committing to the player’s having made the shot, all three of praisefulness, thankfulness, and contrition will push you toward committing to God’s existence if your evidence regarding God’s existence is roughly counterbalanced. Second, in this particular case, we might imagine that even those with weak general preferences for praising, thanking, or apologizing may have a stronger preference for praising, thanking, or apologizing to God, as that for which praise, thanks, and apology is being offered is arguably of immense absolute value. If God exists, then more or less all that we have and are is creditable to God insofar as it is creditable to anyone, and most any wrong we do also wrongs God. Thus, if a person is more strongly inclined in favor of offering praise, thanks, or apology for that with greater absolute value, they may be particularly inclined toward adopting a positive cognitive stance toward God’s existence in a case of rough evidential symmetry. The case of God is of course a somewhat unique case for the operation of praisefulness, thankfulness, and contrition, and this should be acknowledged. In this case, a person’s evidence for thinking that another

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

52 .   person has done something worthy of praise or thanks or has been wronged by them is about as strong as their evidence for thinking this person exists. Insofar as they have evidence for thinking this person, God, exists at all they also have evidence regarding God’s praiseworthiness, thankworthiness, and apologyworthiness. This is not typical of our evidential situation with fellow human beings. Typically with our fellow human beings our evidence that they exist is much stronger than our evidence that they are praiseworthy, thankworthy, or apologyworthy. Yet, this difference should not make a difference to the operation of the traits of praisefulness, thankfulness, and contrition as specified here. What matters for the operation of these traits is just their possessor’s evidence regarding whether another person is praiseworthy, thankworthy, or apologyworthy. If a possessor of one of these traits possesses roughly counterbalanced evidence that another person is praiseworthy, thankworthy, or apologyworthy, then these traits may make a difference for whether they offer praise, thanks, or apology. Thus, while the case of God differs from typical cases involving fellow human beings, it does not differ in a way that makes a difference for the operation of praisefulness, thankfulness, or contrition. Moreover, while the God case differs from typical cases involving our fellow human beings, it is worth noting that there are atypical cases involving our fellow human beings that are closer to the God case, which illustrate how praisefulness, thankfulness, and contrition may operate in this kind of case. For example, imagine a child attending to a parent on their deathbed. In some such cases, depending upon the child’s beliefs about an afterlife, the medical facts about the parent, and the parent’s treatment of the child during their life, the child’s evidence for thinking their parent exists may be about as good as their evidence for thinking this parent is praiseworthy, thankworthy, or apologyworthy. Insofar as they have evidence for thinking their parent is still with them, they likewise have evidence for thinking that someone is with them who is praiseworthy, thankworthy, or apologyworthy. If we imagine their evidence for thinking their parent is still with them is roughly counterbalanced, then this is a sort of case, a bit closer to the God case, where the operation of praisefulness, thankfulness, and contrition can make a determinative difference. Tending to err on the side of offering sincere

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

53

praise, thanks, or apology may lead the child to adopt a cognitive commitment to their parent’s continued existence as part of their offering sincere praise, thanks, or apology to them. Indeed, in some cases these mechanisms may operate long after the death of the parent. They may help to explain the extraordinary commonality of continued communication with the deceased relatives (Steffen and Class 2018).⁴ The aim of the foregoing discussion has been to illustrate how the character traits of praisefulness, thankfulness, and contrition can make a determinative difference for the commitments one holds, including one’s commitments to God’s existence, in cases of rough evidential symmetry. But it is important to note that these character traits needn’t make the sorts of differences cited in every such case. In particular, sometimes there will simply be something more important to do than offer the relevant praise, thanks, or apology. For example, we might imagine in the basketball case that at the moment the shot goes through you receive a phone call notifying you that your child has just been injured and is at the hospital. Given the relative significance of this turn of events and the time-sensitive nature of it, you may quickly exit the stadium without so much as a sincere cheer in order to quickly make your way to the hospital—and this remains the case even if you are a praiseful person. Your praisefulness leads you to err on the side of offering sincere praise rather than refraining from doing so, but only other things being equal; and here other things are not equal. Similarly, the praiseful person whose evidence regarding God’s existence is roughly counterbalanced may not always incessantly give sincere praise, thanks, and apology to God. Sometimes other cares of life may take precedence. Yet it is doubtful that a mature adult human being whose evidence regarding God’s existence remains roughly counterbalanced for much of an extended length of time and who is also either in possession or pursuit of praisefulness, thankfulness, or contrition during ⁴ We can imagine further cases even closer to the God case. For example, imagine that a late adolescent who has spent most of their life in the foster system has recently discovered ambiguous evidence of the existence of someone who showed them great love and care when they were very young. Their evidence for thinking this person exists may be about as good as their evidence for thinking this person is thankworthy. Thanks to two anonymous referees for pushing me to locate mundane cases more similar to the God case.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

54 .   that time will endure very long without these traits leading them to offer sincere praise, thanks, or apology to God. For, first, as we have just noted, the absolute value of that for which praise, thanks, and apology is here offered is very large. And, second, as has been emphasized in recent psychological literature on expressing gratitude and appreciation to other human beings, expressing sincere thanks, praise, and apology are relatively low-cost activities (see Kumar and Epley 2018c). Not only, then, can the possession or pursuit of praisefulness, thankfulness, and contrition manifest in a person’s adopting positive cognitive commitments to God’s existence in cases of rough evidential symmetry, but it is likely to yield this result in the lives of mature human beings whose evidence remains roughly symmetrical for any extended period of time. In this way, pursuing or possessing pro-relationship character traits such as praisefulness, thankfulness, and contrition can manifest in cognitive commitment to God’s existence, making a particularly determinative difference in cases of rough evidential symmetry. My task in the remainder of this chapter is to address the value of religious commitments held in this way. I begin this task in the next section by defending the moral value of praisefulness, thankfulness, and contrition.

2. The Moral Value of Praisefulness, Thankfulness, and Contrition Two central questions animate this chapter. One is the question of whether attempting to be a good person can manifest itself in making a determinative difference for whether a person cognitively commits to God’s existence. In the previous section we saw that pursuing or possessing the tendencies of praisefulness, thankfulness, and contrition can make a determinative difference for whether one cognitively commits to God’s existence. I now aim to defend the thesis that the tendencies of praisefulness, thankfulness, and contrition are morally valuable—that they are traits that a person might reasonably aim to cultivate and maintain as part of their pursuit of being good. If they are, then a positive answer to our first animating question follows: a person’s commitment

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

55

to being good may manifest in making a determinative difference for whether they cognitively commit to God’s existence. It is worth starting with the observation that for some readers little argument may be needed for the conclusion that the traits of praisefulness, thankfulness, and contrition are morally valuable. Some readers may simply, upon understanding the nature of the traits, be inclined to think that possessing them is morally preferable to not possessing them. They might be inclined to judge that these are traits they wished their colleagues had, or that their children will come to possess one day. Such intuitions regarding the value of these traits may be as persuasive as any rational argument could be for leading readers to the conclusion that these traits are morally valuable. A more demanding argument in favor of the value of these traits that remains in the neighborhood of such intuitions is one that appeals to the recently fashionable thesis of moral exemplarism. According to moral exemplarists such as Linda Zagzebski (2017), the emotion of admiration is a fallible guide to moral value. More specifically, the existence of widespread conscientious admiration for a character trait T is strong evidence that T is a virtue. Thus, if there is widespread conscientious admiration for praisefulness, thankfulness, or contrition and moral exemplarism is true, then these traits are likely virtues. I am in possession of no direct evidence regarding whether there is such widespread admiration, and so I will not concentrate on developing this line of argument further. A psychological study testing the existence of such admiration more directly would be philosophically informative. A different approach to defending the value of these traits, which will help us to uncover a deep account of their value, involves attending to other traits that have been regarded as virtues, and arguing that if these latter traits are morally valuable then so are praisefulness, thankfulness, and contrition. For example, several philosophers have been attracted to the idea that it is morally excellent to give others a certain kind of benefit of the doubt—to err, we might say, on the side of viewing others more positively or charitably. Susan Wolf, for example, in her classic essay on moral saints, proposes that a moral saint “should try to look for the best in people” and “give them the benefit of the doubt as long as possible” (1982: 422). Similarly, Ryan Preston-Roedder defends the value of a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

56 .   virtue he calls “faith in humanity” at length, where this virtue involves both a cognitive element and a volitional element. Of the cognitive element, he writes that “when someone who has faith in humanity morally evaluates other people’s actions, motives, or characters, she tends to give them the benefit of the doubt.” Moreover, she tends to “believe in people, trust in them, make presumptions in their favor, or see them in a favorable light, morally speaking” (2013: 666). This cognitive element is complemented by a volitional element which requires of the person with faith that it “be important for her, in itself, that [other] people act rightly” (667). Michael Pace likewise writes that “thinking charitably of others, may in fact be a prima facie moral obligation regarding evidential standards that one has to everyone . . . Other things being equal, adjusting one’s standards to give people the benefit of the doubt seems to be a moral good that flows from the good of treating others with respect” (2011: 258‒9; cf. Roberts and Wood 2007: 73‒5, Adams 1987: 155). In each of these cases, we find a proposal that a tendency to err on the side of viewing others favorably is a central component (or perhaps the whole) of some virtue or other—whether that virtue is the virtue of giving the benefit of the doubt, the virtue of faith, or the virtue of respect. What is important to note for our context is that if any of these traits is indeed a virtue and centrally includes the disposition to err on the side of viewing others favorably, then this provides reason for thinking that the traits of praisefulness and thankfulness are morally valuable. For these traits too are centrally constituted by dispositions to err on the side of viewing others favorably. It is just that these other-favoring dispositions are restricted to particular domains—namely, domains in which it is others’ achievements or acts of beneficence that are up for consideration—and they are accompanied by fitting favorable evaluative stances, where not all of the virtues identified above require such evaluative stances in addition to favorable cognitive stances. We might say, then, following Daniel Russell (2009: ch. 7), that praisefulness and thankfulness as defined here are “unique specifications,” or central components of unique specifications, of more cardinal virtues such as faith or giving the benefit of the doubt or respect, in much the way that magnificence is a unique specification of generosity.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

57

Moreover, we can locate a plausible deep story about the moral value of not only praisefulness and thankfulness but contrition as well by attending to why it is that these traits involving erring on the side of viewing others favorably are indeed valuable. A key idea appealed to by several authors who have written in favor of giving others the benefit of the doubt in one way or another has been the following. When we err on the side of giving someone the benefit of the doubt, we thereby enhance the expected value of personal relationship goods we will enjoy with this other over what we would enjoy if we instead erred on the side of refraining from giving them the benefit of the doubt. Which particular personal relationship goods are at issue may vary depending upon what sort of favorable views of which others one errs toward. But in any case in which one errs on the side of adopting certain more favorable views of certain others, one will thereby enhance the estimated value of some personal relationship goods with these others. These traits involving giving others the benefit of the doubt are in this way what we might call pro-relationship character traits. They are traits that, to borrow a metaphor from another recent author on faith, involve “leaning in” (Page 2017) toward relationship with others. Preston-Roedder implicitly relies upon this kind of point in his defense of the value of having faith in humanity. He writes that faith in humanity “partly constitutes a certain morally important relation, namely, a kind of harmony or solidarity, between the virtuous person and other members of the moral community” (2013: 676). “Having faith in people’s decency,” he continues, “despite reasons for doubt, is a way of standing by them.” By putting her faith in others, the faithful person “ties her own flourishing, in certain respects, to the quality of these people’s characters and actions” (683). This tying of one’s flourishing to others is a significant personal relationship good, and it is one that is promoted in greater measure when one errs on the side of adopting the positive cognitive and volitional stances toward others that Preston-Roedder has in view than when one errs on the side of not adopting such stances. Having faith in humanity, Preston-Roedder emphasizes, is one way in which “a morally virtuous person escapes her solitude and enters into [a valuable] form of community” (684).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

58 .   A similar instance of this pattern of argument can be found in the growing literature on epistemic partiality in friendship (see, e.g., Stroud 2006, Keller 2004). The driving thought behind the central problem in this literature is the thought that the value of a maintained friendship is better advanced by erring on the side of viewing one’s friends favorably than by not doing so. Writing about cases in which one friend, S, tells another, A, that p, thereby inviting A to trust S, Goldberg says that “A risks jeopardizing the friendship in any case in which it is true both that S is worthy of A’s trust and that A fails to trust S.” By contrast “the case in which S is not worthy of A’s trust but A trusts anyway is not one in which A damages the friendship” (forthcoming: 8). Other things being equal, it will follow from this observation that erring on the side of trusting rather than not trusting one’s friends better promotes a maintained friendship with them, and it is this fact which has seemed to some authors to provide moral reason for friends to err on the side of trusting one another—even if this involves epistemic irrationality. I propose here a similar deep defense of the moral value of praisefulness, thankfulness, and contrition. By erring on the side of giving others credit commensurate with their achievements, thanks commensurate with the benefits they have given us, and apologies commensurate with wrongs we have done to them, we enhance the expected value of personal relationship goods with these others over what we would achieve if we erred on the side of refraining from praising, thanking, or apologizing to them. For, just as in trusting a trustworthy friend one cultivates this friendship but in trusting an untrustworthy friend one does not comparably harm the friendship, likewise in praising a praiseworthy person and thanking a thankworthy person and apologizing to one’s victims one cultivates significant personal relationship goods but in praising an unpraiseworthy person and thanking an unthankworthy person or apologizing to someone one has not wronged one does not comparably damage these personal relationship goods. Similarly, just as one does greater harm to a friendship by failing to trust a trustworthy friend than by trusting an untrustworthy friend, likewise one does more harm to a relationship by failing to praise, thank, or apologize to one who deserves it than by offering praise, thanks, or apology to one who doesn’t deserve it. To put the point slightly differently, accurate praise, thanks,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

59

and apology is more likely to build up a relationship than is accurate refraining from these, and misplaced praise, thanks, and apology is less likely to damage a relationship than is their misplaced absence— particularly where evidence regarding whether these are warranted is roughly symmetrical. Erring on the side of praising, thanking, and apologizing therefore enhances the expected value of personal relationship goods. And as such there is available a powerful defense of the moral value of being a praiseful, thankful, contrite person that parallels existing defenses of various ways of giving others the benefit of the doubt. At this point it will be helpful to stress the significant value of the personal relationship goods promoted by praisefulness, thankfulness, and contrition. The importance of personal relationship goods for human well-being is stressed in growing bodies of both philosophical and psychological literature. In philosophy, several growing strands of research in recent decades have focused on personal relationship goods, including research in care ethics and research on associative duties. Where relationship goods in general are conceived as “those goods of constitutive (as well as, often, instrumental) value that accrue to individuals in virtue of them being in relationships with other people,” personal relationship goods are such goods that “accrue to individuals in virtue of them being in relationships that involve some kind of direct, personalized interaction” (Gheaus 2018: section 1). Examples include “companionship, affection, intimacy, attachment, love, friendship, empathy, social respect, solidarity, trust . . . attention, sympathy, encouragement, [and] acceptance” (section 1) among others. Philosophers have been in broad agreement that such goods “generate weighty reasons for action” (section 1) and “represent a significant and non-substitutable component of individuals’ well-being, are a significant kind of personal resource as well as a major determinant of individuals’ opportunities” (Gheaus 2018, Introduction). It is commonly affirmed that these personal relationship goods are non-instrumentally valuable (e.g., Seglow 2013) and are constitutive of good lives (e.g., Lynch et al. 2009). They are, in addition, indispensable to subjective life satisfaction (Vaillant 2012), and are instrumentally valuable as sources of self-confidence, self-respect, and self-esteem (Honneth 1995), as well as even mental, emotional, and physical health (Brownlee

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

60 .   2016). The value of having at least some minimal level of such goods is dramatically illustrated by the devastation which ensues when people are bereft of such goods for extended periods. Brownlee argues that there is a right against social deprivation on the basis that chronic lack of adequate social contact “generates the same threat response as pain, thirst, hunger, or fear by setting off a chain of anxiety-inducing physiological reactions known as the ‘fight or flight’ response” (2013: 211). She writes, further, that “when we are deprived of adequate social connections . . . we tend to break down mentally, emotionally, and physically” (2016: 55). The value of personal relationship goods is such that justice in the distribution of these resources has recently become a major topic of philosophical debate, with several philosophers defending the existence of various rights to personal relationship goods such as adequate social contact or even love (Liao 2015), and others defending the existence of duties to cultivate personal relationship goods such as friendship (Collins 2013). Recent psychological literature, some of which is relied upon in the philosophical research just surveyed, also provides confirmation of both the value of personal relationship goods and the way in which a concern for these goods unifies the dispositions of gratitude, appreciation, and apology. Even two decades ago at a time when it was relatively unfashionable for psychologists to argue in favor of the existence of basic psychological needs, Baumeister and Leary (1995) nonetheless found the evidence in favor of a basic need to belong so widespread and powerful that they published a seminal article on the topic. Now the need to belong, or for belongingness, is commonly recognized in psychological research. The need to belong is a “pervasive drive to form and maintain at least a minimum quantity of lasting, positive, and significant interpersonal relationships” in which “frequent, affectively pleasant interactions” take place in a context of a “temporally stable and enduring framework of affective concern for each other’s welfare” (497). Given the existence of such a need, it is hypothesized and confirmed that “real, potential, or imagined changes in one’s belongingness status will produce emotional responses, with positive affect linked to increases in belongingness and negative affect linked to decreases in it.” Moreover, as reflected above in recent philosophical literature, the absence of adequate relationships will be detrimental toward mental,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

61

emotional, and physical health while their presence will predict such health as well as life satisfaction. While not every person will be equally motivated to cultivate a positive personal relationship with every other person, “rejecting social attachment goes against some deeply rooted aspect of human nature” (520), and when one experiences such rejection, “as in unrequited love, the result is typically distress and disappointment” (505). Personal relationship goods are here seen to serve an indispensable role in fulfilling a basic psychological need. Moreover, there is evidence that tendencies to give thanks (i.e., gratitude), credit (i.e., appreciation), and apology (i.e., disposition to apologize) are closely interrelated aspects of a unified character distinctly sensitive to these personal relationship goods. These constructs are themselves significantly correlated; existing evidence suggests that they tend to come as a package deal. The psychological literature has in fact at some points blended research on gratitude and appreciation together (Wood et al. 2008), while gratitude has also been found to be positively correlated with humility (Uhder, Watkins, and Hammamoto 2010), to which the disposition to apologize is closely related. All three of appreciation (Martinez-Marti, Hernandez-Lloreda, and Avia 2016), gratitude (McComb, Watkins, and Kolts 2004), and the disposition to apologize (Howell et al. 2011) are correlated with the basic personality construct of agreeableness. Indeed, the turn of psychological attention to these facets of personality is likely representative of a broader development within positive psychology that confirms the unifying elements of these traits. Regarding the rapid growth of psychological research on humility, Worthington, Davis, and Hook (2017) write that “a number of researchers appear to have realized the limitation of focusing on individualistic virtues without also attending to the quality of social bonds that tie us together in relationships and communities” (16). Accordingly, there has been a dramatic “increase in [attention given to] other-oriented rather than self-focused virtues” (16) in recent psychological research. Traits such as dispositional appreciation, dispositional gratitude, and the disposition to apologize are plausibly united in being other-oriented in the sense that they each promote personal relationship goods that fulfill basic psychological needs. In a similar way, we may see praisefulness, thankfulness, and contrition as part of a unified character, distinctively

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

62 .   concerned with personal relationship goods, that may reasonably be viewed by aspirants as desirable. Let me conclude this section by briefly restating the basic case here offered in favor of the moral value of praisefulness, thankfulness, and contrition. These traits, I have proposed, are attractively viewed as unique specifications (or central components thereof) of more cardinal virtues such as faith, giving the benefit of the doubt, or respect (as in the case of praisefulness and thankfulness) or humility (as in the case of contrition). Insofar as these latter traits are morally valuable, so are the former. Moreover, a deep defense of their value is available that parallels the deep defense of the value of these latter traits: namely, their possession promotes extremely valuable personal relationship goods better than its absence, and thereby enhances human well-being. To err on the side of praising, thanking, and apologizing is to err on the side of promoting valuable personal relationship goods. These dispositions to err on the side of personal relationship goods are likely only part of the story of a virtuously other-oriented character that is increasingly attracting the attention of psychologists. As such, it may indeed be reasonable to seek to cultivate and maintain them as part of one’s project of aiming to be a good person. And since acting in accordance with these traits can, as argued in Section 1, manifest in making a determinative difference in favor of committing to God’s existence, we have a positive answer to our first animating question. Committing oneself to God’s existence can be a manifestation of one’s reasonable pursuit of being a good person. Aiming to be good can lead one to commit to God.

3. The Justification of Committing to God’s Existence out of Praisefulness, Thankfulness, or Contrition The second animating question of this chapter focuses on the justificatory status of cognitive commitments to God’s existence that arise out of the pursuit or possession and exercise of morally valuable traits such as praisefulness, thankfulness, or contrition. The question is whether commitments that arise in this way are morally, epistemically, or all things considered justified. In this section, I discuss the moral justification,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

63

all-things-considered justification, and epistemic justification of such commitments in turn, identifying reasons for thinking that all three forms of justification may indeed be present in cases of rough evidential symmetry. That is, in those cases which, according to Section 1, are cases where the pursuit of being good is likely to make the most determinative difference in favor of committing to God, there is also reason to think that it will do so in a way the results in commitments to God that are morally, epistemically, and all-things-considered justified. The moral justification of the relevant commitments is the easiest case. Whether the relevant commitments are morally justified will co-vary with whether the balance of moral reason favors adopting them. We saw in the previous section that there are significant moral reasons for adopting the commitments. Adopting a commitment to God’s existence in a case of rough evidential symmetry regarding God’s existence is valuable as a manifestation of the pursuit or possession of praisefulness, thankfulness, and contrition. For the possession and exercise of these traits is itself morally valuable. So there is moral reason for a person whose evidence regarding God’s existence is roughly symmetric to adopt such commitments out of aiming to acquire or maintain such traits. The only way in which the balance of moral reason would not then favor such a person’s adopting these commitments is if the person in question also had comparably strong moral reasons to not adopt these commitments. Yet, such reasons are not easy to come by. We did note in Section 1 that even a praiseful, thankful, and contrite person whose evidence regarding God’s existence is roughly symmetric or better may at times fail to give praise, thanks, and apology to God, because they may be preoccupied with more important matters. Yet we noted there also that this state of affairs is unlikely to persist very long in the life of a mature adult whose evidence regarding God’s existence remains roughly symmetric, given the likely strength of the disposition in favor of giving praise, thanks, and apology in this particular case and given the relatively low cost of these activities. Similarly, it seems unlikely that a person with roughly symmetric evidence regarding God’s existence will possess moral reasons against giving praise, thanks, and apology to God that outweigh their reasons for doing so that persist for very long.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

64 .   The best hope for identifying outweighing moral reasons against offering praise, thanks, or apology to God in such a case rests with finding something inherently morally problematic with erring on the side of offering praise, thanks, or apology in general, or with doing so in this particular case. However the prospects for identifying such reasons are not very appealing. One somewhat tempting proposal for a reason against tending to err on the side of praising, thanking, or apologizing in general appeals to the idea of norms of assertion.⁵ Numerous authors have proposed that assertions are governed by norms such as the knowledge norm (assert only what you know), the belief norm (assert only what you believe), or the justification norm (assert only what you have justifying reason to believe). One might worry that the person who offers sincere praise, thanks, or apology under circumstances of rough evidential symmetry will violate these norms, thereby providing a moral reason against doing this. However, while this idea may be somewhat tempting initially, it is plausible that the kinds of cases we have in mind in this chapter are not ones in which the praising, thanking, or apologizing engaged in amounts to an act of assertion. Assertions have typically been identified by philosophers in terms of their communicative intentions, especially the intention of bringing it about that others believe what is asserted or the intention of supplying others with evidence for target propositions of inquiry (see Pagin 2014: section 3.1 for an overview). But the praising, thanking, and apologizing that is in view in this chapter is not engaged in out of an effort to guide collective deliberation about whether that for which praise, thanks, or apology is offered in fact occurred. It is instead offered as a response based on the assumption or assent or belief, etc., that this did occur. It is done after inquiry, so to speak, rather than as part of it. Or one might propose that there are particular moral reasons that favor erring on the side of skepticism regarding God’s existence. An especially interesting case of this kind, given the structure of this chapter, would be a proposal according to which erring on the side of skepticism regarding God’s existence is just a manifestation of the pursuit of good moral character. In this vein, one might argue that if one errs on the side ⁵ Thanks to an anonymous referee for suggesting this possibility.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

65

of thinking that no one else will act to help others, this may lead one to take more initiative in helping others. This sort of trait might attenuate the effects of the bystander effect, for instance. But, we might think it would also lead one toward skepticism regarding God’s existence. Erring on the side of thinking that no one else will act to help others will lead one to err on the side of thinking God will not act.⁶ In response, I would propose that assuming that God will not act to help others (in the relevant way) is not inconsistent with thinking that God exists. This sort of tendency might lead to a certain kind of (perhaps healthy) skepticism about special divine intervention, but it needn’t lead to a general skepticism about God’s existence. I of course can’t hope to rule out each and every possible countervalue that might tell against erring on the side of offering sincere praise, thanks, or apology to God. I must rest content with suggesting that finding a countervalue that plays the role needed here appears to be no easy task. In further support of this conclusion, it is illuminating to consider that in parallel discussions of the moral value of erring on the side of trusting or believing one’s friends, it is difficult to find philosophers arguing that there are moral reasons against such erring. Instead, it is argued at most that there may be epistemic reasons against such erring (see Faulkner 2018 for a recent overview). The same seems the most likely outcome in this case: if there is something problematic about erring on the side of giving praise, thanks, and apology to God it isn’t something morally problematic but something epistemically problematic. Thus, if we treat moral and epistemic reasons separately, there is reason to presume that commitments to God’s existence that arise out of the pursuit or possession and exercise of praisefulness, thankfulness, and contrition in cases of evidential symmetry will typically be morally justified. The possibility to which we have just alluded regarding a potential conflict between moral reasons and epistemic reasons leads us directly into the territory of considering the all-things-considered justification of the relevant commitments. If we suppose that the balance of moral reasons typically supports adopting such commitments but allow that ⁶ Thanks to another anonymous referee for suggesting this possibility.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

66 .   non-moral reasons may on balance favor not adopting such commitments, then we may wonder what—if anything—we are to conclude about whether adopting such commitments is all-things-considered justified. Here I will assume that there is such a thing as all-thingsconsidered justification,⁷ and I will identify reasons for thinking that if moral and non-moral reasons conflict in this case, with moral reasons favoring commitment to God’s existence, then committing to God’s existence will be all-things-considered justified. One prominent and appealing strategy available here is to appeal to the nature of moral reasons, qua moral, as trumping or outweighing non-moral reasons. Preston-Roedder appeals to this sort of strategy in his defense of the all-things-considered value of behaving in accordance with faith in humanity. He is willing to grant the possibility of some conflict between faith and epistemic rationality, writing that in some cases “someone who has faith can, without any failure of [this] virtue, form beliefs about people . . . that are to some degree irrational, given the evidence” (2013: 685‒6). Yet, he continues, “unless we assume that the moral importance of epistemic rationality is implausibly great, or the importance of [the moral aims of faith] implausibly slight, we should conclude that a virtuous person may sacrifice some degree of epistemic rationality, in certain respects and in certain cases, in her pursuit of these other aims” (687). Preston-Roedder concludes that having faith is constitutive of a “practical ideal, concerned with the sort of life one should live” (686). Epistemically disvaluable features of faith can only prevent faith from contributing to this all-things-considered ideal, according to Preston-Roedder, if they are of comparable moral importance to the moral values toward which such faith is conducive. Similarly, in the literature on epistemic partiality in friendship, it has been maintained that if there is conflict between the norms of friendship and the norms of rationality, it is the norms of rationality that must give way, as the epistemic values in play are not of comparable moral importance to friendship (cf. Stroud 2006).

⁷ For an influential account of the nature of all-things-considered justification, see (Chang 2004).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

67

This same strategy applies straightforwardly to our focal case. The moral values better secured by erring on the side of giving praise, thanks, and apology to God outweigh epistemic values that might be better secured by not so erring. The enhanced expected value of personal relationship goods better secured through such erring is all-things-considered more important than the enhanced epistemic value one might expect to secure through not so erring. As such, if we suppose that committing to God’s existence in cases of rough evidential symmetry out of an aim to acquire or exercise praisefulness, thankfulness, or contrition is morally justified, we likewise have reason to think doing so is all-things-considered justified. It is worth pausing to note here something else about praisefulness, thankfulness, and contrition that parallels Preston-Roedder’s remarks about faith. Having faith, he writes, “does not typically dispose one to make irrational judgments” (2013: 685). This is because in the great majority of evidential contexts, a slight disposition to err on the side of viewing others favorably will not make a determinative difference for which cognitive attitudes a person adopts; instead, it is only in particularly contexts of roughly symmetrical evidence that it will. Thus, “concerns about the irrationality of faith in humanity apply to a narrower range of cases than they initially appear to” (685). Likewise, praisefulness, thankfulness, and contrition will not typically lead one to adopt epistemically irrational attitudes. The kind of case in which they are apt to do so is the kind of case on which we are focusing in this chapter—a case in which a person’s evidence regarding the fittingness of praise, thanks, or apology is roughly symmetric. Here, and only here, are we to expect that these dispositions will typically lead one to adopt epistemically irrational commitments. This point should help us to see that any epistemic values that are sacrificed via the attempt to acquire or retain praisefulness, thankfulness, or contrition are relatively slight. While I think that the foregoing defense of the all-things-considered justification of the target commitments to God’s existence is forceful— just as forceful as parallel defenses of faith and epistemic partiality in friendship—it is worth noting that the defense makes an unnecessarily strong assumption. For, in order to maintain a defense of the all-thingsconsidered justification of giving praise, thanks, or apology in the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

68 .   relevant cases it isn’t necessary to insist as suggested above that moral values, qua moral values, trump just any non-moral values. It isn’t necessary to insist that non-moral values can never trump moral values. It is only necessary that the moral values at issue in this particular case trump the non-moral values at issue in this case. And it is indeed plausible that this is true regarding the epistemic values at issue in this case given what we have seen regarding the relative insignificance of the epistemic values that are sacrificed as compared with the relative significance of the moral value of personal relationship goods. So long as any epistemic values sacrificed by erring on the side of praising, thanking, or apologizing to God in these cases are not as all-things-considered important as the moral values better attained through this erring, the status of this erring as all-things-considered justified will not be impugned by these sacrificed epistemic values. A similar approach is appealing when we consider other non-moral values that might be sacrificed in this case. One example of such a nonmoral value is proposed by Susan Wolf (1982). Despite her affirmation that a moral saint will err on the side of giving others the benefit of the doubt, she suggests that this moral excellence may conflict with the nonmoral excellence of a certain kind of humorous personality. “A cynical or sarcastic wit,” she writes, “or a sense of humor that appreciates this kid of wit in others, requires that one take an attitude of resignation and pessimism toward the flaws and vices to be found in the world” (422)—and a tendency to adopt such an attitude is in “substantial tension” (421) with the disposition to give the benefit of the doubt. To put it slightly differently, a tendency to err on the side of viewing others positively, while perhaps morally virtuous, may prevent one from just as quickly identifying more negative interpretations of others in a way characteristic of non-morally excellent cynicism. Now, one might attempt to challenge Wolf ’s idea that excellence in cynicism or sarcasm requires a tendency to adopt attitudes of pessimism or resignation, thereby conflicting with giving the benefit of the doubt. As it happens, existing psychological evidence does not support the claim that people who exhibit the sort of humor Wolf seems to have in mind—called “aggressive humor” by psychologists—are more depressive or pessimistic than people who do not excel in this sort of humor (Martin

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

69

2006: ch. 7). Moreover we might, a priori, think that having a cynical or sarcastic wit needn’t require that one tend to err on the side of adopting more negative views of others (or oneself), but only that it requires an aptitude for identifying such views and an appreciation of them. We might imagine, for example, that a person who is simply very quick to see all angles on a performance might be prepared to engage in excellent sarcasm, even if they also err on the side of adopting more positive evaluations of these performances. Still, let us suppose that there is conflict between tending to give the benefit of the doubt and excellent cynicism/sarcasm. Much as in the case of apparently conflicting epistemic values, here too we might judge that the non-moral cynicism-relevant value lost by erring on the side of giving others the benefit of the doubt isn’t of comparable all-thingsconsidered importance to the moral value to be gained by erring in this way. It may be that there is little to say in defense of this judgment other than to appeal to readers’ intuition—a fact Wolf herself seems to acknowledge insofar as she opts for a kind of intuitionism regarding when moral or non-moral values win out. Still, recent psychological research on aggressive humor again may prove illuminating in this case. This research has found that “Greater use of aggressive humor is related to more frequent negative interactions with others, less giving and receiving of empathy, reduced ability to manage conflict and provide empathy in social relationships, and lower satisfaction with dating relationships and friendships, both for oneself and one’s partner” (Martin 2006: 303). In other words, just as we might suspect if it is in tension with orientations that tend to promote personal relationship goods, aggressive humor tends toward the destruction of personal relationship goods. Now this of course isn’t to say that such humor doesn’t have any (non-moral) value. But, given the importance of personal relationship goods for human well-being as highlighted above, it may indeed seem more reasonable to judge that this cynicism-relevant value is not worth its price all-things-considered. More generally, the allthings-considered justification of committing to God’s existence out of the aim to acquire or exemplify praisefulness, thankfulness, and contrition can be forcefully defended on the basis of the plausibly superior all-thingsconsidered importance of personal relationship goods over non-moral goods that might be sacrificed by adopting these commitments.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

70 .   I turn finally to the epistemic justification of commitments to God’s existence held as a manifestation of one’s pursuit or possession of praisefulness, thankfulness, and contrition. While in discussing the allthings-considered justification of these commitments, I allowed for the sake of argument that these commitments may be irrational and so epistemically unjustified, I now wish to bring this claim into question. There are two broad strategies that can be used for this purpose. The first strategy involves leaning heavily on the suggestion put forward in Section 1 that the kinds of cognitive commitments necessary for offering sincere praise, thanks, or apology may be quite varied and may be weaker than commitments such as outright belief that are typically taken to be subject to stronger standards for epistemic justification. If the suggestion offered there is correct, and sincere praise, thanks, and apology can be given when the subject only assumes or assents or accepts, etc., that something praiseworthy, thankworthy, or apologyworthy has occurred, then this provides one way for resisting the conclusion that offering such praise, thanks, or apology involves adopting epistemically unjustified cognitive attitudes. For in this case it is plausible that the cognitive attitudes such praise, thanks, or apology involve adopting are either not subject to epistemic standards at all or are subject to much weaker epistemic standards than outright belief or other stronger cognitive attitudes. Whereas it may be more tempting to judge that adopting the latter cognitive attitudes is epistemically unjustified in the target cases of praising, thanking, or apologizing to God, it is less tempting to judge that adopting the former cognitive attitudes would be.⁸ Still, I only offered these comments about the plurality of positive cognitive attitudes that may suffice for giving sincere praise, thanks, or apology as a suggestion and I do not want to insist on it. So, let us assume here that offering sincere praise, thanks, or apology in our target cases involves adopting cognitive attitudes of a sort that are subject to the kinds of epistemic standards typically thought to apply to belief. There remains a strategy that can be used to resist the conclusion that offering such praise involves adopting epistemically unjustified cognitive attitudes. For, there are three recent approaches to thinking about epistemic ⁸ Thanks to an anonymous referee for pushing me to highlight this strategy.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

71

justification that would each make room for allowing that the commitments in question would not be epistemically unjustified. The first of these approaches has been developed by Susanna Rinard (2017, forthcoming). Rinard defends a view she calls “Equal Treatment” for belief. On this view, there is no special guidance-giving normativity that applies only to beliefs; questions about what one should believe or is permitted to believe are to be answered in the same way as questions about what one should do or is permitted to do more generally. As a consequence, if having a belief is all-things-considered justified, it cannot be unjustified according to some standard of justification that applies only to beliefs. The implication for our focal case is straightforward: if believing that God exists in our focal case is all-things-considered justified, then it cannot be epistemically unjustified—it cannot be justified according to some standard of justification that applies only to beliefs. It is instructive to note the kind of case which plays a central role in Rinard’s defense of Equal Treatment. She asks us to imagine a case in which taking a pill will cause you to have a certain belief you would not otherwise have without taking the pill, and you know this. Moreover, as the case is described it is supposed to be one in which you all-thingsconsidered should take the pill, but your all-things-considered reasons favor this only slightly. Denying Equal Treatment and maintaining that there is a special guidance-giving normativity that applies only to beliefs allows it to be the case that you should not hold the belief that is caused by taking the pill, despite it being the case that you should take the pill. For there could be distinctively epistemic reasons against holding the belief that do not apply to taking the pill, and which are such that they are sufficient to shift the balance of all-things-considered reasons pertaining to holding the belief but not to taking the pill. Rinard finds this consequence implausible because it violates the principle of agglomeration: that if you should x and you should y then you should (x and y). Denying Equal Treatment violates this principle because it allows that you should take the pill and should not hold the belief, whereas surely it is not the case that you should (take the pill and not hold the belief), as the latter is not an option for you. Our own focal case significantly parallels this example of the beliefinducing pill. Let us suppose, as argued above, that offering praise,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

72 .   thanks, or apology to God is all-things-considered justified. If we suppose that there is a guidance-governing normativity distinctive to belief, then it could be that while offering praise, thanks, and apology is justified, believing that God has done something warranting praise or thanks or believing that one has wronged God is not all-thingsconsidered justified. This possibility runs afoul of agglomeration, if we suppose that you cannot offer praise, thanks, or apology to God without holding such beliefs. Thus, if we, with Rinard, are unwilling to give up agglomeration, we have a way of resisting the conclusion that offering sincere praise, thanks, or apology to God in our focal cases involves adopting epistemically unjustified beliefs. Two other approaches to epistemic justification maintain that there is a distinctive guidance-governing normativity that applies only to cognitive attitudes such as belief, but allow that cognitively committing to God’s existence in our focal case may be epistemically justified. The first of these is a moral encroachment approach to epistemic justification. On such a view, moral reasons for or against adopting a cognitive attitude can affect the level of evidential support needed in order for that attitude to be epistemically justified. Moral encroachment is in this way a species of the broader and better-known thesis of pragmatic encroachment, which allows that pragmatic facts about cognitive attitudes can make a difference for the epistemic status of those attitudes (see Kim and McGrath 2018). Not just any version of moral encroachment will allow us to maintain that in our focal case cognitively committing to God’s existence is epistemically justified. In fact, pragmatic encroachment views generally have tended to maintain that pragmatic factors can only raise the level of evidential support needed for a positive cognitive attitude to be justified. Yet what is needed in our focal case is a view according to which moral reasons for holding an attitude can lower the level of support needed for that attitude to be epistemically justified. For, presumably, absent such moral reasons, if a person’s evidence is roughly symmetric in that it supports a claim p at most just as much as it supports not-p, then adopting a strong positive cognitive attitude such as belief toward p will not be epistemically justified. What is needed, then, is a version of moral encroachment where moral reasons in favor of adopting a positive

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

73

cognitive commitment toward p can make adopting such a commitment epistemically justified despite evidence for p otherwise being roughly symmetric. Michael Pace (2011) has recently defended a version of moral encroachment that comes close to offering what is needed. Crucially, on Pace’s view, the fact that greater moral value is attainable via belief than via its absence can lower the evidential standards necessary for belief to be epistemically justified. As he puts it, “When there are significant positive benefits to be gained by having a true belief and relatively little practical cost of error, the evidential standards sufficient for justification dip below what they would be in contexts in which nothing much is at stake” (257). Yet, Pace stops just short of offering what is needed, because he stipulates that in order for belief that p to be justified, one’s evidence must make p at least more likely than not. It is only in cases where one’s evidence for p already makes p more likely than not that moral considerations can kick in to make a difference for how strongly this evidence must support p in order for believing p to be epistemically justified. Pace claims that there are principled reasons for imposing this “more likely than not” requirement. First, he worries that “when one recognizes that one’s evidence does not make a proposition more likely than not, to believe on the basis of pragmatic reasoning . . . may be psychologically impossible” (252). Second, he worries that even if not psychologically impossible, believing on the basis of such reasoning “may require deceiving oneself about the quality of one’s evidence” (252). These claims, however, are quite contentious,⁹ and they seem all the more contentious given Pace’s own approach to thinking about moral encroachment. On Pace’s approach, the best way to understand how moral encroachment works is that moral factors affect one’s “evidential standards” (254) for the extent to which evidence must support a claim p in order for belief that p to be epistemically justified. Pragmatic factors matter, in other words, for whether the extent to which a claim is supported by evidence is adequate for belief (cf. here Schoenfield 2014). Yet, there doesn’t seem to be anything self-deceptive involved in a person recognizing that their ⁹ For a review of recent debate which assesses it as largely standing at an impasse, see (Rinard 2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

74 .   evidence only supports a claim p just as much as its denial, and yet reflectively taking this degree of support as adequate for believing p given the moral gains to be had through such belief. Nor, indeed, does there appear to be anything psychologically impossible about a person who judges their evidence’s support for p to be adequate in this way adopting the belief that p (cf. Comesaña 2015). A moral encroachment view like Pace’s but without its “more likely than not” requirement therefore seems not much less well motivated than Pace’s own stated view. If such a moral encroachment view is correct, then it will support the judgment that our target commitments to God’s existence may be epistemically justified, rather than epistemically unjustified. Since the moral value to be gained via these commitments is significantly greater than that to be gained by their absence, the standards for the degree of evidential support necessary for these commitments to be adequately supported may be lowered—even lowered to the point that the evidence needn’t make God’s existence more likely than not. A person might selfreflectively assess their evidence regarding God’s existence as being roughly counterbalanced, and yet take this level of evidential support to be adequate for cognitively committing to God’s having achieved great feats, God’s having benefited them in important ways, and their owing God apology for their wrongdoing. On the present moral encroachment view, these commitments would be epistemically justified. One final view about epistemic justification differs from the previous moral encroachment view in that it will not straightforwardly support the view that committing to God’s existence in our focal cases is epistemically justified while not doing so is not. Instead, this view more straightforwardly allows that committing to God’s existence and not doing so may each be epistemically permissible options. The view in question is typically called “epistemic optionalism” or “epistemic permissivism.”¹⁰ In recent literature, epistemic permissivism is defined by reference to the thesis it must deny—namely, the uniqueness thesis. This thesis holds, roughly, that there is at most one unique justified attitude for a person to take toward any proposition given any body of evidence. Denials of this thesis can come in various forms and as such can be more or less ¹⁰ For a recent overview of permissivism in epistemology, see (Kopec and Titelbaum 2016).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

75

permissive. For example, one version of permissivism might hold that multiple distinct credences in a proposition p can each be justified given the same body of evidence as long as these credences fall within a certain narrow range. Another might hold that cognitive attitudes that are further apart, such as confident belief and confident disbelief, can each be justified given the same body of evidence. Just as we needed a certain version of moral encroachment in order to defend the justification of our target commitments to God’s existence, we will need a certain version of permissivism to do this. One important requirement is that the version allows for cases in which it is epistemically permissible (and so justified) for a person to take a positive cognitive attitude toward a proposition p despite the fact that her evidence regarding p is at most symmetric. Presumably, in such cases, the view in question would allow that both neutral and positive, or negative, neutral, and positive cognitive attitudes are all permissible. Not all versions of permissivism will meet this requirement, but some will. A common but not universal idea that motivates permissivist views is the Jamesian idea that there are two cognitive goals that are in tension with one another: namely, believing the truth (or, more broadly, adopting positive cognitive attitudes toward true claims) and avoiding false belief (or, more broadly, avoiding adopting positive cognitive attitudes toward untrue claims). Because the goals are in tension, how one weighs the goals can make a difference for which attitudes one adopts. As Kelly (2014) expresses the idea, “the more weight one gives to not believing something false, the more it makes sense to hold out until there is a great deal of evidence that p is true before taking up the belief that p. On the other hand, the more one values not missing out on believing the truth, the more it makes sense to take a somewhat more liberal attitude about how much evidence one expects before taking up the relevant belief” (104). Similar things are true for adopting more subtle positive or negative cognitive attitudes—a point Kelly in fact emphasizes. Wayne Riggs (2008) likewise notes that epistemic risk is ineliminable: “every single instance of belief or withholding represents an epistemic risk of one kind or another” (2)—either we risk failing to take positive cognitive attitudes toward true claims or we risk taking positive cognitive attitudes toward false claims. How these risks are managed, Riggs

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

76 .   suggests, is largely down to individual personality. “Our proclivities to act or believe in certain ways,” he writes, “tend to embody the values that we have, and the strengths with which we have them” (5). The more we value having positive cognitive attitudes toward true claims in a domain, the more we’ll risk adopting positive cognitive attitudes toward false claims in this domain. It coheres well with permissivist views of this sort to affirm that where a person’s evidence regarding p is roughly symmetric, it may be epistemically permissible both for them to adopt a positive cognitive attitude toward p, or a neutral cognitive attitude toward p, or a negative attitude toward p, depending in significant part on the extent to which they prefer to err on the side of gaining truth or to err on the side of avoiding error. Permissivist views of this sort apply straightforwardly to our focal case. In this case, we have a person whose character is such that they prefer erring on the side of giving praise, thanks, or apology; or at least they aim for their character to be this way. As such, they value adopting positive cognitive attitudes regarding the appropriateness of such praise, thanks, or apology when these attitudes are warranted more highly than they value not adopting these attitudes when they are not warranted. They would rather risk error in the matters of giving praise, thanks, or apology for the sake of gaining truth in these matters. As such, their personality is ripe to make a difference for which cognitive attitudes they adopt in the way suggested by Riggs. It will incline them to take positive cognitive attitudes that they might not otherwise have taken. Moreover, and importantly for our purposes here, the attitudes they in this way adopt will be epistemically permissible on the versions of permissivism here in view. We have now surveyed three views about epistemic justification that render defensible the judgment that the target cognitive commitments to God’s existence are not epistemically unjustified. To the extent that a disjunction of these views is preferable to its denial, we have reason to think that the target cognitive commitments in view in this chapter not only have moral and all-things-considered justification but also do not lack epistemic justification. Indeed, on two of the views surveyed, these commitments will possess epistemic justification. Moreover, it bears emphasizing here that the two strategies outlined above for resisting the epistemic irrationality of offering praise, thanks,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

77

or apology in our focal cases are not mutually exclusive. The views of epistemic justification surveyed toward the end of this section may be even more plausible when applied to cognitive attitudes weaker than belief, such as acceptance, assumption, or assent. In this way, the strategies for resisting the epistemic irrationality of offering sincere praise, thanks, or apology in our focal cases may work in tandem. It may be that neither of these two strategies, nor even their combination, convinces readers that it is more likely than not that the target cognitive commitments would not be epistemically unjustified. Still, it bears observing that whatever likelihood there is for thinking that these commitments would not be epistemically unjustified is relevant for computing the all-things-considered justification of our target cognitive commitments. For in computing this all-things-considered justification we must rely on an assessment of the expected epistemic disvalue of these commitments. Whatever positive likelihood there is that these commitments are not epistemically unjustified will lower this expected disvalue somewhat. In this way, even if the availability of the views surveyed here does not convince readers that our target cognitive commitments to God’s existence are not epistemically unjustified, it may bolster the case given above for concluding that these commitments are all-things-considered justified.

4. Conclusion I conclude by briefly connecting the preceding discussion with the topic of loving God. Praising, thanking, and apologizing to God are, I propose, all partly constitutive of what it is to engage with God in loving relationship. Indeed, more generally, sincere praising, thanking, and apologizing all require attitudes of appreciation or valuing of the one toward whom they are directed, where this appreciating or valuing is constitutive of relating lovingly to this other. As Roberts writes with respect to gratitude—indeed, gratitude to God specifically—“Gratitude is a kind of love . . . Thanksgiving is a practice of love” (2014: 68). Similar things could be said about appreciation and praise, as well as remorse or repudiation and apology. In this way, pursuing or possessing and exercising praisefulness, thankfulness, or contrition not only inclines one

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

78 .   more strongly toward adopting cognitive commitments to God’s existence but inclines one more strongly toward engaging with God in loving relationship. In this story of the justification of religious commitment, relationship takes center stage. Aiming to be good leads one not only to adopt cognitive commitment to God but to adopt such commitments as an expression of love.

References Adams, Robert, 1979. “Moral Arguments for Theistic Belief.” In Rationality and Religious Belief, ed. C. Delaney, 116–40. Notre Dame, IN: University of Notre Dame Press. Adams, Robert. 1987. “Moral Arguments for the Existence of God.” In The Virtue of Faith and Other Essays in Philosophical Theology, 144–63. Oxford: Oxford University Press. Alston, William. 1996. “Belief, Acceptance, and Religious Faith.” In Faith, Freedom, and Rationality, ed. Jeff Jordan and Daniel Howard-Snyder. Lanham, MD: Rowman & Littlefield. Audi, Robert. 2008. “Belief, Acceptance, and Faith.” IJPR 63, 1: 87–102. Baumeister, Roy, and Mark Leary. 1995. “The Need to Belong: Desire for Interpersonal Attachments as a Fundamental Human Motivation.” Psychological Bulletin 117, 3: 497–529. Berger, Fred. 1975. “Gratitude.” Ethics 85, 4: 298–309. Brownlee, Kimberly. 2013. “A Human Right Against Social Deprivation.” The Philosophical Quarterly 63, 251: 119–222. Brownlee, Kimberly. 2016. “Ethical Dilemmas of Sociability.” Utilitas 28, 1: 54‒72. Bruton, Samuel. 2003. “Duties of Gratitude.” Philosophy in the Contemporary World 10, 1: 1–5. Budd, Malcolm. 2002. The Aesthetic Appreciation of Nature. Oxford: Oxford University Press. Chang, Ruth. 2004. “All Things Considered.” Philosophical Perspectives 18, 1: 1–22. Collins, Stephanie. 2013. “Duties to Make Friends.” Ethical Theory and Moral Practice 16, 5: 907–21.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

79

Comesaña, Juan. 2015. “Can We Believe for Practical Reasons?” Philosophical Issues 25, 1: 189–207. Dunn, E. W., L. B. Aknin, and M. I. Norton. 2008. “Spending Money on Others Promotes Happiness.” Science 319: 1687–8. Epley, N., and J. Schroeder. 2014. “Mistakenly Seeking Solitude.” Journal of Experimental Psychology: General 143: 1980–99. Eriksson, John. 2011. “Straight Talk: Conceptions of Sincerity in Speech.” Philosophical Studies 153, 2: 213–34. Evans, C. Stephen. 2013. God and Moral Obligation. Oxford: Oxford University Press. Faulkner, Paul. 2018. “Giving the Benefit of the Doubt.” International Journal of Philosophical Studies 26, 2: 139–55. Fitzgerald, Patrick. 1998. “Gratitude and Justice.” Ethics 109, 1: 119–53. Gheaus, Anca. 2018. “Personal Relationship Goods.” In Stanford Encyclopedia of Philosophy, ed. Edward Zalta. Available at . Goldberg, Sanford. Forthcoming. “Against Epistemic Partiality in Friendship: Value-Reflecting Reasons.” Philosophical Studies. Honneth, Axel. 1995. The Struggle for Recognition: The Moral Grammar of Social Conflicts, trans. Joel Anderson. Cambridge: Polity Press. Howard-Snyder, Daniel. 2013. “Propositional Faith: What it is and What it is Not.” American Philosophical Quarterly 50, 4: 357–72. Howard-Snyder, Daniel. 2019a. “Can Fictionalists Have Faith? It All Depends.” Religious Studies 55: 1–22. Howard-Snyder, Daniel. 2019b. “Three Arguments to Think that Faith Does Not Entail Belief.” Pacific Philosophical Quarterly 100 (1): 114–28. Howell, Andrew, Raelyne Dopko, Jessica Turowski, and Karen Buro. 2011. “The Disposition to Apologize.” Personality and Individual Differences 51, 4: 509–14. Keller, Simon. 2004. “Friendship and Belief.” Philosophical Papers 33, 3: 329–51. Kelly, Thomas. 2014. “Evidence can be Permissive.” In Contemporary Debates in Epistemology, ed. Matthias Steup, John Turri, and Ernest Sosa, 298–311. Oxford: Wiley-Blackwell. Kim, Brian, and Matthew McGrath, eds. 2018. Pragmatic Encroachment in Epistemology. New York: Routledge Press.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

80 .   Kopec, Matthew, and Michael Titelbaum. 2016. “The Uniqueness Thesis.” Philosophy Compass 11, 4: 189–200. Kumar, Amit, and Nicholas Epley. 2018a. “It’s Surprisingly Nice to Hear You: Miscalibrated Expectations of Connection and Awkwardness Affect How People Choose to Connect with Others.” MS under review. Kumar, Amit, and Nicholas Epley. 2018b. “A Little Good Goes a Long(er) Way (than Expected): Differential Construals of Kindness by Performers and Recipients. MS. Kumar, Amit, and Nicholas Epley. 2018c. “Undervaluing Gratitude: Expressers Misunderstand the Consequences of Showing Appreciation.” Psychological Science 29, 9: 1423–35. Liao, Matthew. 2015. The Right to be Loved. Oxford: Oxford University Press. Lynch, Kathleen, John Baker, Maureen Lyons, Sarah Cantillon, Judy Walsh, Maggie Feeley, Niall Hanlon, and Maeve O’Brien. 2009. Affective Equality: Love, Care and Injustice. London: Palgrave. Manela, Tony. 2015. “Gratitude.” In Stanford Encyclopedia of Philosophy, ed. Edward Zalta. Available at . Manela, Tony. 2016. “Gratitude and Appreciation.” American Philosophical Quarterly 53, 3: 281–94. Martin, Rod. 2006. The Psychology of Humor: An Integrative Approach. Amsterdam: Elsevier. Martinez-Marti, Maria, Maria Hernandez-Lloreda, and Maria Avia. 2016. “Appreciation of Beauty and Excellence: Relationships with Personality, Prosociality and Well-Being.” Journal of Happiness Studies 17: 2613–34. McComb, D., P. Watkins, and R. Kolts. 2004, April. Personality and Happiness: The Importance of Gratitude. Presentation to the 84th Annual Convention of the Western Psychological Association, Phoenix, AZ. Pace, Michael. 2011. “The Epistemic Value of Moral Considerations: Justification, Moral Encroachment, and James’ ‘Will to Believe’.” Nous 45, 2: 239–68. Page, Meghan. 2017. “The Posture of Faith.” In Oxford Studies in Philosophy of Religion, vol. 8, ed. Jonathan Kvanvig, 227–44. Oxford: Oxford University Press. Pagin, Peter. 2014. “Assertion.” In Stanford Encyclopedia of Philosophy, ed. Edward N. Zalta. Available at .

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

    

81

Preston-Roedder, Ryan. 2013. “Faith in Humanity.” Philosophy and Phenomenological Research 87, 3: 664–87. Radzik, Linda, and Colleen Murphy. 2015. “Reconciliation.” In Stanford Encyclopedia of Philosophy, ed. Edward Zalta. Available at . Riggs, Wayne. 2008. “Epistemic Risk and Relativism.” Acta analytica 23, 1: 1–8. Rinard, Susanna. 2017. “No Exception for Belief.” Philosophy and Phenomenological Research 94, 1: 121–43. Rinard, Susanna. 2018. “Believing for Practical Reasons.” Nous 4: 763–84. Rinard, Susanna. Forthcoming. “Equal Treatment for Belief.” Philosophical Studies. Roberts, Robert, and Jay Wood. 2007. Intellectual Virtues: An Essay in Regulative Epistemology. Oxford: Oxford University Press. Roberts, Robert. 2014. “Cosmic Gratitude.” European Journal for Philosophy of Religion 6, 3: 65–83. Russell, Daniel. 2009. Practical Intelligence and the Virtues. New York: Oxford University Press. Schellenberg, John. 2005. Prolegomena to a Philosophy of Religion. Ithaca, NY: Cornell University Press. Schoenfield, Miriam. 2014. “Why Permissivism is True and What it Tells Us about Irrelevant Influences on Belief.” Nous 47, 1: 193–218. Seglow, Jonathan. 2013. Defending Associative Duties. New York: Routledge. Steffen, Edith, and Dennis Klass, eds. 2018. Continuing Bonds in Bereavement: New Directions for Research and Practice. London: Routledge. Stroud, Sarah. 2006. “Epistemic Partiality in Friendship.” Ethics 116, 3: 498–524. Uhder, J., P. C. Watkins, and D. Hammamoto. 2010, August. Would the Humble Please Stand: Can Self-Reported Humility Be Valid? Paper Presented to the Annual Convention of the American Psychological Association, San Diego, CA. Vaillant, George. 2012. Triumphs of Experience: The Men of the Harvard Grant Study. Cambridge, MA: Belknap Press. Walker, A. D. M. 1980. “Gratefulness and Gratitude.” Proceedings of the Aristotelian Society 81: 39–55.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

82 .   Wolf, Susan. 1982. “Moral Saints.” Journal of Philosophy 79, 8: 419–39. Wood, Alex, John Maltby, Neil Stewart, and Stephen Joseph. 2008. “Conceptualizing Gratitude and Appreciation as a Unitary Personality Trait.” Personality and Individual Differences 44, 3: 621–32. Worthington, Everett, Don Davis, and Joshua Hook. 2017. “Introduction: Context, Overview, and Guiding Questions.” In Handbook of Humility: Theory, Research, and Applications, ed. Everett Worthington, Don Davis, and Joshua Hook, 15–28. Routledge Press. Zagzebski, Linda. 2017. Exemplarist Moral Theory. Oxford: Oxford University Press.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

3 How to be a Mereological Anti-Realist Andrew Brenner

“It is a divine matter to possess sufficient knowledge, and at the same time sufficient competence, to mix a plurality into a oneness, and conversely to break a oneness up into a plurality; there is not now nor will there ever be any human being who is up to either of these tasks”—Plato, Timaeus (Plato 2008: 67)

1. Introduction Peter van Inwagen’s “special composition question” asks more or less the following question: what must some objects be like in order for them to compose another object (i.e., in order for there to be some object of which those former objects are the proper parts) (van Inwagen 1990)? There have been a number of proposed answers to this question. In this chapter I’m primarily interested in a type of answer to the special composition question which has not received a great deal of attention, namely an answer to the special composition question according to which the circumstances under which composition occurs have something to do with one or more (actual or counterfactual) agent’s mental states. Call views of this sort “mereological anti-realism.” Mereological anti-realism has been defended by Uriah Kriegel (2008, 2012) and, under the label “mereological idealism,” Kenneth Pearce (2017).¹ ¹ Cowling (2014) defends a view which he calls “mereological anti-realism,” but his view is very different from the sort of view I discuss in this chapter. According to Cowling’s “mereological anti-realism” the world lacks “mereological structure,” so that, for example, there are no composite objects, and there are no mereologically simple objects. By contrast, according to the Andrew Brenner, How to be a Mereological Anti-Realist In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Andrew Brenner 2022. DOI: 10.1093/oso/9780192862976.003.0003

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

84   In this chapter I have two main tasks. First, I describe the variants of mereological anti-realism defended by Kriegel and Pearce, and present some objections. Second, I introduce and defend a theistic variant of mereological anti-realism. Theistic mereological anti-realism comes in a weaker and a stronger variety. According to the weaker thesis: there is a y such that the xs compose y if and only if God wills that there is a y which those xs compose. According to the stronger thesis: there is a y such that the xs compose y if and only if God wills that there is some specific y which those xs compose. To some extent my two projects in this chapter are independent of one another: (1) my objections to non-theistic variants of mereological anti realism (specifically, the versions of mereological anti-realism defended by Kriegel and Pearce) do not rely on the truth of theistic mereological anti realism; (2) there is some motivation to accept theistic mereological anti-realism, even if my objections to non-theistic variants of mereological anti-realism fail. But it is also worth noting that the objections I present to non-theistic variants of mereological anti-realism fail to undermine theistic mereological anti-realism. So, to the extent that one is attracted to mereological anti-realism in general, the failure of nontheistic versions of mereological anti-realism lends some support to the theistic variety of mereological anti realism. One way to construe the chapter, then, is as a sustained argument for the conclusion that the best way to develop mereological anti-realism is in theistic terms.² In the end I do not endorse any variant of mereological anti-realism. This chapter’s arguments are nevertheless worth developing, for at least two reasons. First, extant varieties of mereological anti-realism have not to my knowledge received sustained critical attention. Second, while there has been some discussion of the relationship between God and composition (see §3 below), the theistic variant of mereological antirealism has received little attention. To my knowledge, the only previous sort of mereological anti-realism I discuss in this chapter, the world has “mereological structure,” but in certain important respects that structure is determined by the cognitive activities of one or more agents. ² In this respect the present chapter is similar to Alvin Plantinga’s well-known “How to be an Anti-Realist” (Plantinga 1982). In that paper Plantinga argues that proponents of a certain sort of global anti-realism should be theists. My chapter’s thesis also resembles Berkeley’s contention that subjective idealism should be developed in theistic terms.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

85

defense of theistic mereological anti-realism is given by Eric Yang and Stephen Davis (2017) (although under the name “the Will of God Theory,” rather than under the name “theistic mereological antirealism”).³ I think that theistic mereological anti-realism should receive more attention, and that it should be taken seriously by those interested in the metaphysics of composition, as well as those interested in the metaphysical import of theism. Here’s the plan for the remainder of this chapter. In §2 I discuss the non-theistic variants of mereological anti-realism defended by Kriegel and Pearce. I argue that both views face formidable objections, objections which do not undermine theistic mereological anti-realism. In §3 I more fully introduce and develop theistic mereological anti-realism. In §4 I develop some motivations for theistic mereological anti-realism. §5 concludes the chapter.

2. Non-Theistic Mereological Anti-Realism Non-theistic versions of mereological anti-realism are “non-theistic,” not because they are incompatible with theism, but rather because God plays no special role in the formulation of these versions of mereological antirealism. Non-theistic mereological anti-realists think that the circumstances under which composition occurs have something to do with the thoughts or practices of non-divine beings. (It is open to the non-theistic mereological anti-realist to think that the circumstances under which composition occurs might also have something to do with the thoughts or practices of a divine being. For simplicity, I will generally ignore this possibility.) In practice, non-theistic mereological anti-realists (such as Kriegel and Pearce) emphasize the connection between the circumstances under which composition occurs and human thoughts or

³ The epigraph at the beginning of this chapter might give the impression that Plato, or at any rate a character in one of Plato’s dialogues, endorsed theistic mereological anti-realism. But the epigraph is taken out of context, and in fact I doubt that Plato, or the character in Timaeus who spoke the words in the epigraph, intended to endorse theistic mereological anti-realism. My impression is that one can take these sorts of liberties with epigraphs. It is a pretty good epigraph, right?

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

86   practices. Sometimes this sort of anti-realism is restricted to particular sorts of composite objects, rather than composite objects in general. Lynne Baker (2000: 29‒30, 34‒5), for example, contends that some composite objects (e.g., works of art) only come into existence as a result of human cognitive practices. The sort of mereological anti-realism I’d like to focus my attention on here is broader than this. Non-theistic mereological anti-realism, as I conceive of that thesis, contends that all composite objects exist at least partially as a result of the actual or counterfactual cognitive activities of non-divine beings. Of course, composite objects may also depend on their parts for their existence. But according to non-theistic mereological anti-realism that some parts compose a composite object is a result of the actual or counterfactual cognitive activities of non-divine beings. Mereological anti-realism should be distinguished from quantifier variance (although mereological anti-realism is compatible with quantifier variance). Proponents of quantifier variance such as Hilary Putnam and Eli Hirsch think that there are multiple possible quantifiers, and that which sentences talking about the existence of composite objects are true will depend on how we interpret the quantifiers used in those sentences (Putnam 1987: 18‒19, 32‒6; 2004: 33‒40; Hirsch 2011: 144‒6). On this sort of view there is no sense in which our cognitive practices create composite objects. Rather, proponents of quantifier variance think that, in some cases, our cognitive practices can pick out possible quantifier meanings according to which sentences involving composite objects are true, even if some such sentences turn out to be false when the quantifier expressions involved in those sentences are given other permissible meanings. Mereological anti-realism, as I think of it, says that, keeping our quantifier meanings fixed, the circumstances under which composition occurs are connected to our cognitive practices, in the sense described above. A number of philosophers have endorsed views which might be described as anti-realism with respect to composite objects, nonfundamental objects, or even physical objects in general (see, e.g., Heller 2008; Chalmers 2009; Remhof 2018). In this chapter I confine my attention to two philosophers who most unambiguously defend the sort of view which I aim to refer to with the label “non-theistic

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

87

mereological anti-realism,” namely Kriegel (2008, 2012) and Pearce (2017). As we’ll see, the varieties of mereological anti-realism defended by Kriegel and Pearce both fall prey to serious objections, objections which do not undermine theistic mereological anti-realism. I’ll begin by discussing Pearce’s view.

2.1. Pearce Pearce (2017) defends a thesis which he calls “mereological idealism.” According to mereological idealism, composite objects come into existence as a result of their proper parts being “unified in thought under a concept” by some mind (Pearce 2017: 200). Some objects compose a further object if and only if they are unified in thought in this manner. For example, my desk exists as a result of my mentally unifying the proper parts of the desk under the concept of a “desk.” I see the parts of the desk, I conceptualize those parts as uniting to form an object which falls under the concept of a “desk,” and so, as a result of my mental activities, the parts of the desk do compose a desk. For Pearce, it is very easy for this sort of mental “unifying” of parts to occur. For example, a natural worry to have about mereological idealism is that it fails to account for the existence of composite objects which, it seems, nobody ever thinks about. What about all of those boulders on Mars, for example, which nobody has ever thought about? If those boulders exist, then, according to mereological idealism, someone must have mentally unified their parts under a concept. But it is difficult to see who could have done that. Of course, the theist who also happens to be a mereological idealist will have an easy response to this concern: God thinks about all of the boulders on Mars, even if we do not. Pearce’s own response to the concern is that, simply by entertaining the question “who mentally unified the parts of all those boulders on Mars?” we mentally unify all of those boulder parts (Pearce 2017: 206‒7). So, according to Pearce it is extremely easy to engage in this sort of mental unification of parts, since in fact we accidentally unify parts on another planet, simply by entertaining the question “who mentally unified the parts of all those boulders on Mars?”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

88   Notably, these observations undermine one of the motivations Pearce gives for mereological idealism. One motivation for mereological idealism is its purported ability to allow us to secure the existence of those composite objects which we tend to think exist (e.g., desks), without having to posit the existence of the sorts of exotic composite objects posited by mereological universalism⁴ (e.g., a composite object composed of my phone and the Eiffel tower). Plausibly, however, mereological idealism does lead to universalism, if it is as easy as Pearce thinks it is to mentally unify the parts of composite objects. Many philosophers think that universalism is true. These philosophers might very well unify all of the composite objects posited by universalism into existence. If I can (accidentally) unify all of the proper parts of boulders on Mars, and so bring all of those boulders into existence, it seems as if the universalist should similarly be able to unify all objects (or all non-overlapping objects) in thought, and thereby bring into existence all of the composite objects which exist according to universalism. If we must specify some concept under which all of the parts are unified, there seems to me to be nothing stopping us from unifying all of the parts under the “composite object” concept. By contrast, theistic mereological anti-realism does not lead to universalism—in fact, theistic mereological anti-realism might lead us to reject universalism. I discuss this point further in §4 below. An additional concern for Pearce’s mereological idealism is that it seems to attribute to us spooky mental powers: on this view we have the ability to create large heavy physical objects with our minds, simply by mentally “unifying” some other objects. Pearce offers the following response to this concern: “Our ability to influence what there is by means of our thought is not some sort of spooky, supernatural power we have; it stems from the fact that the co-apprehension of the constituent parts is a metaphysically necessary condition for the existence of a composite object” (Pearce 2017: 203). Later he reiterates this point: the mind-dependence in question “does not involve attributing any spooky powers to minds; it merely rests on a particular account of the existence ⁴ According to mereological universalism, any non-overlapping objects compose a further object.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

89

conditions for composite objects” (Pearce 2017: 206). But the fact that the existence conditions for composite objects make reference to our mental activities is precisely one of the things which makes mereological idealism “spooky”! In fact, this downplays the spookiness. According to mereological idealism, the existence conditions for composite objects don’t merely make reference to our mental states. In other words, the existence of composite objects does not merely supervene on our mental states. Rather, according to mereological idealism, composite objects partially depend for their existence on our mental states. Think, for example, of an object composed of Mount Everest and whatever book is nearest to your left hand. According to mereological idealism, this big physical object (bigger than Mount Everest) exists, in part, because we think about it. In fact, it partially owes its existence to the fact that I just told you to think of an object composed of Mount Everest and whatever book is nearest to your left hand. By reading this chapter you’ve helped create a physical object which is larger than Earth’s tallest mountain. It just seems very implausible to me that human beings can change the furniture of the physical world so easily.⁵ By contrast, there is nothing “spooky” about the idea that composite objects owe their existence to God, as theistic mereological anti-realists maintain. God, given God’s omnipotence, can bring composite objects into existence in all sorts of ways which are not available to us (cf. §4.1 below). For example, it may be spooky to suppose that we human beings could bring big heavy physical objects into existence simply by “unifying” their parts in thought. By contrast, however, there is nothing spooky about the supposition that an omnipotent God could bring physical objects into existence in a similar manner, by, say, “‘unifying” some objects in the sense of willing that those objects compose a further object.

⁵ Presumably, nobody before you has “unified” Mount Everest and whatever book is nearest to your left hand. (This is, in fact, why I used this example.) Earlier I suggested that mereological universalists might mentally “unify” every non-overlapping object. If that’s right, then the object composed of Mount Everest and whatever book is nearest to your left hand owes its existence to whoever first endorsed mereological universalism (or, perhaps, to the first person to endorse universalism after the book nearest to your left hand was printed). But it is just as implausible that that big physical object would owe its existence to the first universalist as it is that it would owe its existence to you.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

90  

2.2. Kriegel I now turn to a discussion of Kriegel’s variant of mereological anti-realism (Kriegel 2008, 2012). According to Kriegel, composition occurs whenever a subject of a certain sort would think that it occurs. This basic idea could be construed in a few different ways, and Kriegel’s several formulations of mereological anti-realism are rather complicated. Here is one example: Metaphysically necessarily, for any (non-overlapping) objects O₁, . . . , On, there is an object O, such that O₁, . . . , On compose O iff O₁, . . . , On are disposed to elicit the automatic and non inferential but intellectual judgment that there is an O in most subjects who are capable of producing automatic and non-inferential but intellectual judgments, and whose faculties dedicated to the production of such judgments are well-functioning, under forced choice conditions that do not perturb the exercise of the relevant faculties. (Kriegel 2008: 367)

Kriegel goes on to suggest that facts regarding composition may not track intuitions regarding composition, but rather may track whether some proper parts instantiate properties or relations which would serve as the categorical basis for the disposition, in normal intuiters, to form the intuition that those proper parts compose a whole. In a later article Kriegel suggests that his thesis should be formulated in terms of ideal intuiters, rather than “normal” intuiters (Kriegel 2012: 29). There’s no need at this point to get bogged down in the details of Kriegel’s proposal, although I’ve alluded to some of those details in order to give the reader a sense for some of the complications which crop up when Kriegel tries to give a precise formulation of mereological antirealism. The basic idea, again, is that the circumstances under which composition occurs have something to do with the circumstances under which subjects of a certain sort (e.g., normal intuiters, ideal intuiters) would think that composition occurs. If, for example, some objects compose a table, this has something to do with the fact that normal intuiters judge (or would judge) that those objects compose a table, or it has something to do with the fact that (counterfactual) ideal intuiters would judge that those objects compose a table.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

91

I think it is reasonable to make the assumption that, for Kriegel, facts regarding the circumstances under which composition occurs track the judgments of normal or ideal intuiters because the judgments of those intuiters ground or otherwise explain the fact that composition occurs under those circumstances. Kriegel himself is somewhat ambivalent on this issue (cf. Kriegel 2008: 376‒8), but I don’t think that he should be. Kriegel should not be committed only to the weaker thesis that the judgments of normal or ideal intuiters are merely correlated with, but do not ground or explain, the facts regarding when composition occurs. If the relevant facts regarding composition neither explain, nor are explained by, the judgments of the relevant intuiters, then the correlation between the compositional facts and the judgments would be objectionably brute. If the correlation results merely from the fact that the relevant judgments are explained by the relevant facts regarding composition (e.g., an ideal intuiter would intuit that such-and-such objects compose something because in fact such-and-such objects do or would compose something), then Kriegel’s so-called “mereological anti-realism” is neither novel, nor a form of anti-realism, nor a view according to which, in Kriegel’s terms, composition is a “secondary quality” (and, in fact, Kriegel admits as much—see Kriegel 2008: 378). So, it is best to interpret Kriegel’s mereological anti-realism as being committed to the idea that our judgments, or the judgments of ideal intuiters, track the compositional facts because which compositional facts obtain is grounded in or otherwise explained by the actual or counterfactual occurrence of the relevant judgments.⁶ The primary motivation for Kriegel’s version of mereological antirealism is that it allows us to preserve our intuitive judgments regarding the circumstances under which composition occurs. According to Kriegel, there is a very strong presumption that our account of composition should adhere to our pre-philosophical intuitions regarding the circumstances

⁶ Another option is that the facts regarding composition and the facts regarding our judgments regarding composition are reliably correlated because they have a common explanation (e.g., a common cause or ground). Kriegel just doesn’t seem to be endorsing a view of this sort. The theistic mereological anti-realist, however, might endorse a view of this sort: God’s decrees/ intentions/whatever explain both why such-and-such xs compose something, and why we form the judgment that such-and-such xs compose something.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

92   under which composition does and does not occur—that is, it should adhere to those of our intuitions which do not lie “downstream of philosophical theorizing” (Kriegel 2008: 363). This is because these sorts of pre-philosophical intuitions are the only data with which our theories of composition should be in accord (Kriegel 2008: 362‒3).⁷ It is not difficult to understand how our mereological intuitions track the mereological facts if those mereological facts are in part made true by our mereological intuitions or judgments. I’m now ready to present my objection to Kriegel’s variant of mereological anti-realism. Kriegel’s view faces a dilemma. For Kriegel, facts regarding the circumstances under which composition occurs either track our judgments concerning composition (or subjunctive conditionals regarding our judgments), or they track subjunctive conditionals regarding the judgments of (counterfactual) ideal intuiters. First horn of the dilemma: if Kriegel’s thesis is that facts regarding composition track our judgments, or subjunctive conditionals regarding our judgments, then the thesis faces the “spookiness” objection which I presented to Pearce’s mereological idealism. It is very spooky to suppose that composite physical objects come into existence as a result of our judging that their parts compose something, or as a result of the fact that we would judge that those parts compose something. And this is so despite the fact that on one variant of Kriegel’s view composite objects will only come into existence as a result of our being inclined to make an “automatic and non-inferential” judgment that its proper parts compose something. The supposition that big heavy physical objects should come into existence as a result of our cognitive activities is not made less spooky by supposing that the cognitive activities in question are automatic or non-inferential. One might as well say that my having telekinetic abilities is made less spooky if those telekinetic abilities are only employed automatically or without deliberation. As we saw earlier, according to one variant of Kriegel’s view, some parts compose a whole in response to their instantiation of those ⁷ This methodological constraint is in line with a broader methodological inclination: “On the conception I have in mind, a central—perhaps the central—function of philosophy is to vindicate (most of) our pre-philosophical worldview, by providing some sort of rational reconstruction of our everyday world-model” (Kriegel 2008: 364).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

93

properties or relations which would serve as the categorical basis for the disposition, in normal intuiters, to form the judgment that those proper parts compose a whole. In this case we would not directly create composite objects with our judgments. So, is this variant of Kriegel’s mereological antirealism less spooky than the alternative variant according to which objects compose a whole in response to our judgments that those objects compose a whole? No. As I argued above, if some parts compose a whole in response to their instantiation of those properties or relations which would serve as the categorical basis for the disposition, in normal intuiters, to form the intuition that those proper parts compose a whole, then the parts care about those properties and relations only because they are connected with our disposition to form the intuition that those parts compose a whole. In other words, our judgments or intuitions regarding composition still explain why the parts in question compose a whole. If they didn’t—if, that is, the parts in question compose a whole as a result of their instantiating the relevant properties or relations in question, and not in response to the fact that those properties and relations are connected with our judgments regarding composition—then Kriegel’s “mereological anti-realism” wouldn’t be a novel thesis, and wouldn’t really be a form of anti-realism. Earlier I said that according to one variant of Kriegel’s view what matters are the (counterfactual) judgments of ideal intuiters. Now I need to say a bit more about what Kriegel means when he writes of “ideal” intuiters.⁸ Kriegel says that an ideal intuiter is “an otherwise normal human subject who (i) performs or undergoes all and only epistemically justified/warranted cognitive processes and (ii) knows all the non-decompositional facts regarding the world as a whole” (Kriegel 2012: 31). What does Kriegel mean by “otherwise normal human subject”? He means that the human subject in question “is set up like actual subjects, in particular in terms of the biological function of its reasoning faculties; namely, the function of promoting its prospects for survival and reproduction” (Kriegel 2012: 30). I think it’s rather odd that Kriegel stipulates that ideal intuiters must be human, and must have the sort of cognitive architecture actual ⁸ Kriegel also suggests that composition tracks the judgment of ideal intuiters in ideal conditions, and he gives an account of what it means for some condition to be “ideal.” Luckily, however, we can ignore this complication, since my objections to Kriegel don’t depend on this detail of his view.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

94   humans have (in any case with respect to its promotion of survival and reproduction). In fact, it seems objectionably spooky and anthropocentric to suppose that compositional facts would care specifically about human ideal intuiters. Human mentality is an eminently contingent, and recent, feature of the world, and human history is extremely short in comparison with the history of the universe. So, why ideal human intuiters, rather than, say ideal dog intuiters, or ideal martian intuiters? I suggest, then, that Kriegel drop the stipulation that ideal intuiters be human, and, for similar reasons, suggest that he drop the stipulation that ideal intuiters have cognitive architecture which is similar to human cognitive architecture with respect to its promotion of survival and reproduction. But now we come to the second horn of the dilemma: if facts regarding composition track the (counterfactual) judgments of ideal intuiters, then Kriegel’s version of mereological anti-realism is unmotivated. As we saw earlier, the primary motivation for Kriegel’s version of mereological anti-realism is that it allows us to preserve our intuitive judgments regarding the circumstances under which composition occurs. Again, it is not difficult to understand how our mereological judgments track the mereological facts if those mereological facts are in part made true by our judgments. The problem now is that if facts regarding the circumstances under which composition occurs are tied to the judgments of non-actual ideal intuiters, then there is little reason to think that our pre-philosophical intuitions regarding the circumstances under which composition occurs are correct. Perhaps our intuitions will resemble, to some extent, the intuitions of ideal intuiters. But given the vast gulf between their epistemological credentials and our own, our intuitions regarding composition would presumably diverge significantly from the intuitions of ideal intuiters. The fact that we have the mereological intuitions that we have, and more generally the fact that we have mereological concepts at all, is likely an eminently contingent feature of our cognitive lives, a result of contingent forces in our evolutionary history, and contingent limitations on our cognitive resources, which would presumably not shape the intuitions of ideal intuiters. It seems plausible, for example, that our tendency to “chunk” objects together (such as our tendency to “chunk” birds together into a “flock” of birds) is a result of our brains making do with limited cognitive

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

95

resources—it’s much easier for our visual representational systems to keep track of a big composite object (like a flock of birds), rather than a bunch of individual objects (like each of the individual birds in the purported flock) (cf. Gobet et al. 2001; Alvarez 2011; Osborne 2016: §3.4; Brenner 2018: 662). We fail to prereflexively “chunk” arbitrary parts together for presumably much the same reason, namely because it would impose a needless burden on our cognitive capacities. It’s also important to remember that, given mereological anti-realism, we cannot suppose that the judgments of ideal intuiters would match our judgments because we are all predisposed to track the mereological facts. This is because, given mereological anti-realism, the mereological facts (and in particular the circumstances under which composition occurs) are made true by our judgments, or by the judgments of ideal intuiters. So, there are no pre-existing mereological facts which might constrain our (or ideal intuiters’) judgments. I reiterate, then, my conclusion that there is little reason to think that the mereological judgments of ideal intuiters would match our own mereological judgments. So, to recap the dilemma: we either formulate Kriegel’s version of mereological anti-realism in terms of actual intuiters, and are left with an objectionably spooky view, or we switch to ideal intuiters, but lose our motivations for adopting Kriegel’s version of mereological anti-realism. Theistic mereological anti-realism faces neither of these problems. As I noted above, there is nothing “spooky” about the idea that composite objects owe their existence to God. What’s more, theistic mereological anti-realism has a number of motivations, which I discuss in §4 below. As we’ve seen, extant non-theistic varieties of mereological antirealism face severe difficulties, difficulties which theistic mereological anti-realism does not face. I turn now to a more detailed discussion of theistic mereological anti-realism.

3. Theistic Mereological Anti-Realism Theistic mereological anti-realism comes in two varieties, one weak and one strong. The weaker thesis is that there is a y such that the xs compose y if and only if God wills that there is a y which those xs compose. The

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

96   stronger thesis is that there is a y such that the xs compose y if and only if God wills that there is some specific y which those xs compose.⁹ The general idea is that, just as the divine command theorist thinks that moral facts are within God’s control, the theistic mereological anti-realist thinks that mereological facts are within God’s control, in two respects: (1) God decides when composition occurs (this is the weaker version of theistic mereological anti-realism); (2) God decides which composite objects are associated with which proper parts (this is a consequence of the stronger version of theistic mereological anti-realism). The distinction between the weaker and stronger versions of theistic mereological anti-realism is not very important for much of what I write in this chapter. So, I often write of “theistic mereological anti-realism,” without bothering to distinguish between the weaker and stronger versions of that thesis. Where the distinction between the two versions of the thesis is important, I will distinguish between them. The sort of theism I have in mind in this chapter is of the standard Western monotheistic variety. Below I will appeal to the notion that God is omnipotent, and I will assume that the problem of evil is something the theist will feel motivated to address. The latter assumption comports with a conception of theism according to which God is perfectly good, or at any rate so good that the existence of evils, or certain sorts of evils, is prima facie in tension with the existence of God. Other than that I don’t think I need to put much stress on one particular conception of God as opposed to another. Now, admittedly, there is a sense in which, construed as an answer to the special composition question, theistic mereological anti-realism (in either variant) is not very informative. Even if composition occurs when and only when God wills that it occurs, we might still wonder when, in fact, God wills that composition occurs.¹⁰ More generally, we might ⁹ Yang and Davis (2017) also defend theistic mereological anti-realism, and it’s worth stating how they understand the thesis: “The xs compose some y at t if and only if (i) the xs exist at t, (ii) God wills that there be a fusion of the xs at t” (Yang and Davis 2017: 220). This is basically equivalent to what I call the “weaker” form of theistic mereological anti-realism. ¹⁰ A similar point can be made regarding divine command theories. Divine command theories say that actions are right or wrong only insofar as they are commanded or forbidden by God. But divine command theory by itself doesn’t tell us which actions God commands or forbids.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

97

think that theistic mereological anti-realism is uninformative because it is strictly compatible with all extant answers to the special composition question. All of those answers say that composition occurs under (and only under) such and such circumstances. There is nothing incoherent in supposing that composition happens in those circumstances because God wills that those are the circumstances under which composition occurs. Even if theistic mereological anti-realism is relatively uninformative for these reasons, it might still be a thesis worth defending. First, it will at the very least give us an answer to a higher-order corollary of the special composition question, one which asks, of the correct answer to the special composition question, “why is this the correct answer to the special composition question?” The theistic mereological anti-realist answer to this question is informative and interesting, even if it tells us relatively little about the circumstances under which composition in fact occurs. But second, if we know that God decides the circumstances under which composition occurs, this will open up new potential lines of evidence regarding the circumstances under which composition occurs: (1) we may in principle have theological evidence regarding the circumstances under which composition occurs (e.g., from revelation); (2) we may have normative evidence regarding the circumstances under which composition occurs (e.g., we may have reasons to suppose that a good God would or would not allow composition to occur under certain circumstances). To be honest, I’m not sure that theistic mereological anti-realism is correct. But, I think there’s a lot to be said for it, and so in the remainder of this chapter I’ll motivate the thesis as best I can. If I’m successful, theistic mereological anti-realism will be taken seriously as a live option in debates over the metaphysics of composition, and debates over the special composition question in particular. Some of the motivations for theistic mereological anti-realism which I discuss will appeal only to theists, or only to particular sorts of theists (e.g., Christian theists). Some of the motivations, however, should appeal even to non-theists. Before I turn to discuss motivations for theistic mereological antirealism, I would like to note that, while theistic mereological anti-realism has, to my knowledge, only been discussed once before (Yang and Davis

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

98   2017), several other philosophers have suggested that God might have something to do with the circumstances under which composition occurs. For example, Jonathan Edwards is sometimes interpreted as endorsing a view according to which God can decide when diachronic composition occurs. God can decide, for example, whether stages at different times compose some temporally extended object.¹¹ Alvin Plantinga (2011: 289‒90) and Christopher Menzel (2018) defend the view that God brings about the existence of sets. Part of the motivation here is that sets are brought about by “collecting” objects together in some sense, but there are sets which could plausibly only result from the collecting activity of a divine being. On some views sets involve mereological relations (cf. Lewis 1991). The classical Indian philosopher Udayana defends an argument for theism from the existence of certain sorts of composite objects (for a discussion, see Kronen and Menssen 2013). The idea seems to be that only intelligent agents can create composite objects, but there are some composite objects whose existence cannot plausibly be attributed to any non-divine agent. By contrast, Edward Feser (2017: ch. 2), taking inspiration from Plotinus (Ennead V.4; see also Gerson 1994: ch. 1), argues that any sort of composition requires theism. The basic idea is that something must bring parts together and make them compose something, but the only thing which can do something like that must be absolutely simple (since otherwise it would be composite, and it would need something else to unite its parts). Feser identifies this absolutely simple thing with God. Aquinas also endorses the idea that God, being absolutely simple, unites parts in order to make them form composite objects (Summa Theologica I, Q3, A7), although Aquinas, unlike Feser, does not make this point as part of an argument for theism. Finally, Ross Inman and Alexander Pruss (2019) also argue that theism has an important connection with the circumstances under which composition occurs. They argue that Christian theists should not be mereological universalists.

¹¹ For this interpretation of Edwards, see Chisholm 1976: 139; Johnston 2010: 123. I don’t know much about Edwards, but this interpretation has come under what seems to me to be convincing criticism. See Rea 2007: §2; LoLordo 2014: §§4–5.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

99

4. Motivations for Theistic Mereological Anti-Realism One motivation for theistic mereological anti-realism stems from the fact that, as we’ve seen, extant non-theistic variants of mereological antirealism are objectionable. To the extent that one thinks that we should endorse some version of mereological anti-realism, then, one should think that we should endorse theistic mereological anti-realism. I turn now to discuss some additional motivations for theistic mereological anti-realism.

4.1. Motivation 1: God is omnipotent If God is omnipotent, then if God wills that some objects compose a further object, then they will. Similarly, if God wills that some objects fail to compose a further object, then they will not. That doesn’t strictly get us all the way to theistic mereological anti-realism, since God might just be indifferent with respect to whether some xs compose a y, and so God might neither will that they do nor will that they don’t compose the y. (In other words, it may be the case that, for some objects, those objects either compose something or do not compose something, but whether or not they compose something is not something which God decides.)¹² All of this does get us pretty close to (the weaker variety of) theistic mereological anti-realism, however: there is a y such that the xs compose y if and only if God wills or allows that there is a y which those xs compose. Is this thesis correct? The main concern here is that philosophers often assume that it is necessarily true that composition occurs or doesn’t occur under any particular circumstances. Omnipotence is often restricted so that God cannot contravene necessary truths. So, the thought goes,

¹² Or perhaps in cases where God is indifferent with respect to whether or not composition occurs it would be indeterminate whether composition occurs. If that’s right, then this sort of case may be compatible with theistic mereological anti-realism, construed as the thesis that there is (determinately) a y such that the xs (determinately) compose y if and only if God wills that there is a y which those xs compose. But indeterminate composition does not mesh well with the second motivation for theistic mereological anti-realism, which I’ll discuss below.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

100   whether or not composition occurs under any particular circumstances has nothing to do with what God wills: whether or not composition occurs under some particular circumstances is a necessary truth, and God doesn’t have the power to decide which necessary truths obtain. This is an important objection, in part because it would undermine the “God is omnipotent” motivation for theistic mereological anti-realism, but also because it would directly undermine theistic mereological antirealism itself. But there are a few responses which might be made to this objection. First, it is open to debate whether truths regarding the circumstances under which composition does or does not occur are necessary, rather than contingent. Ross Cameron (2007) and Kristie Miller (2010) argue that composition is contingent. Similarly, Alexander Skiles (2015) argues that grounds need not necessitate that which they ground. If that’s true of some grounds, it seems to me to lend support to the idea that parts arranged in such and such a manner need not necessitate their composing something (assuming, as seems plausible, that parts generally ground the existence of the thing(s) they compose). We can reach the same conclusion by way of concerns regarding necessary connections between distinct existences (cf. Bohn 2014). Assuming that composite objects are not simply identical with their proper parts, to suppose that those proper parts (or those proper parts configured in a particular manner—e.g., arranged in a particular manner) necessitate the existence of this other thing, the thing which they compose, strikes me as mysterious, and potentially objectionable for precisely the same reason that philosophers have thought many other necessary connections between distinct existences are objectionable. Here are these objects, and here is this other object, and why should the former objects force or necessitate the other object to exist? Second, some theists have argued that God decides, of at least some necessary truths, whether or not they are true (cf. Leftow 2012). Third, simply because some truth is necessary, and so could not have been otherwise, it does not follow that the necessary truth is brute. We might suppose that, even if God is not free to decide which necessary truths regarding composition obtain, God’s decrees nevertheless ground or otherwise explain why those necessary truths regarding composition

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

101

obtain. In this case we might be forced to think that God makes the same such decrees in all possible worlds, since the compositional facts which those decrees ground obtain in all possible worlds. But that’s compatible with theistic mereological anti-realism. Fourth, we might think that the motivations for theistic mereological anti-realism, if sufficiently compelling, give us some reason to endorse the disjunction of the previous three responses.

4.2. Motivation 2: Theistic mereological anti-realism potentially allows us to endorse a moderate answer to the special composition question Let’s say that an answer to the special composition question is “moderate” if it is neither mereological nihilism, according to which composition never occurs, nor mereological universalism, according to which composition always occurs (or, more carefully, according to which any two or more non-overlapping objects compose a further object). Many philosophers have wanted to endorse a moderate answer to the special composition question, but have felt compelled to instead endorse either nihilism or universalism. But theistic mereological anti-realism allows us to avoid one of the chief motivations which is generally offered in favor of nihilism and universalism, that moderate answers to the special composition question will require that we accept either objectionable ontic vagueness or a sharp, but arbitrary, boundary between circumstances in which composition occurs and circumstances in which composition does not occur.¹³ The vagueness concern stems from the apparent fact that any plausible moderate answer to the special composition question will involve vague criteria. For example, van Inwagen’s moderate answer to the special composition question is organicism, according to which some objects compose something if and only if the activities of those objects constitute

¹³ See Lewis 1986: 212–13; Horgan and Potrč 2008: ch. 2. For a response on behalf of moderate answers to the special composition question, see Korman 2015: chs 8–9.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

102   a life (van Inwagen 1990). But van Inwagen is very upfront about the fact that it is sometimes a vague matter whether or not the activities of some objects constitute a life. It follows then that, on van Inwagen’s view, it will sometimes be vague whether composition occurs, and so it will sometimes be vague whether a composite object exists. Most philosophers think this sort of ontic vagueness is unacceptable. By contrast, neither nihilism nor universalism will give us conditions in which composition occurs which could ever be vague: nihilism says that composition never happens, and universalism says that composition always happens, and neither of those criteria is vague. Here’s where theism comes into the picture. If God decides when composition does and does not occur, then God can ensure that it is never a vague matter whether the criteria for composition are met. In other words, God can ensure that those conditions which are such that composition occurs in, and only in, those conditions, are not vague. Now you might think that the criteria God sets for when composition occurs could be vague. But in that case you should, at any rate, accept that they need not be vague. But if ontic vagueness is objectionable because it is impossible, as its opponents generally argue, then God might ensure that it is never vague whether composition occurs because God must ensure that it is never vague whether composition occurs. A similar point can be made about the concern regarding arbitrariness. The concern here is that we will have trouble finding a plausible moderate answer to the special composition question which distinguishes in a non-arbitrary manner between cases where composition occurs and cases where composition does not occur. But as Yang and Davis (2017: 220) note, given theistic mereological anti-realism, there is a non-arbitrary distinction between cases where composition occurs and cases where composition does not occur: God wills that composition occurs in the former circumstances, but not in the latter circumstances. What’s more, some of God’s decisions regarding the circumstances in which composition occurs may themselves be non-arbitrary, insofar as they may be tied up with the distribution of goods and evils. For example, God might want to ensure that there are composite persons, but God might also want to ensure that there are not any composite persons experiencing gratuitous suffering. Whether there are composite persons,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

103

and whether there are composite persons who experience gratuitous suffering, will partially depend on where and when composition occurs.¹⁴ I should mention at this point that there is some pressure toward thinking, not only that theistic mereological anti-realism gives us the resources to defuse otherwise powerful objections to moderate answers to the special composition question but also that it may push us toward actually endorsing one of those moderate answers. Again, theistic mereological anti-realism is strictly compatible with any extant answer to the special composition question, as far as I can tell. But there are some considerations which might reasonably lead some proponents of theistic mereological anti-realism to reject both mereological nihilism and mereological universalism. Start with mereological nihilism. Nihilism is often thought to require that we reject the existence of human persons, assuming that human persons are composite objects (cf. Rosen and Dorr 2002: 159‒60; Sider 2013: 268‒9). While theism is compatible with the nonexistence of human persons, some prominent theistic religious traditions arguably are not. For example, it is hard to make sense of Christian soteriological doctrines absent a belief in the existence of human persons. If, then, mereological nihilism requires that we reject the existence of human persons, then theistic mereological anti-realism, conceived in, say, Christian terms, will entail that mereological nihilism is false.¹⁵ Why might theistic mereological anti-realism lead us to reject mereological universalism? My arguments here must be very tentative, since they rely on very controversial assumptions. First, universalism may make the problem of evil more difficult for the theist to resolve. For anyone who experiences a severe evil (e.g., severe hardship or suffering), universalism gives us innumerable other individuals who we might not initially think about who also experience that evil—e.g., the object composed of that person and the Eiffel Tower, the object composed of that person and my cell phone, etc. You might doubt

¹⁴ I discuss these points more below, especially in my discussions of motivations 4 and 5. ¹⁵ This is assuming, of course, that mereological nihilism really does entail that there are no human persons. That’s debatable. If human persons are mereologically simple, for example, then mereological nihilism is compatible with the existence of human persons.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

104   that such gerrymandered individuals would have any experiences. But assume that I am a composite object. In virtue of what do I have the conscious experiences that I have? Everyone is going to say that the answer to that question has something to do with the fact that some of my parts are in certain physical states—e.g., that my brain is in such and such a physical or functional state. It’s natural, then, to think that any composite object with those sorts of parts will also have those conscious experiences.¹⁶ Or, at any rate, this seems to me to be epistemically possible, which is enough for us to worry that the existence of the gerrymandered individuals in question might pose a difficulty for theism. The presence of so many more suffering individuals in the world will presumably strengthen the problem of evil for theism. And since it is universalism which leads us to posit the existence of all these strange gerrymandered individuals, and the existence of all those gerrymandered individuals would strengthen the problem of evil for theism, we should lower our credence in the conjunction of theism and universalism.¹⁷ Earlier I wrote that Inman and Pruss (2019) argue that Christian theists should not be mereological universalists. Their arguments might provide a bit of additional support for the idea that theistic mereological anti-realism might lead us away from universalism (and it’s also worth noting that while their arguments specifically concern Christian theism, it seems to me that many other theists might also endorse their arguments). Here I will describe just one of their arguments. God is the most valuable being. But mereological universalism is committed to there being a composite object composed of God and some other object, and this object will plausibly be at least as valuable as God. More strongly, God is the most valuable possible being. But universalism is committed to there possibly being some composite object composed of God and some other object, a composite object which is at least as valuable as God. ¹⁶ Johnston (2017: 626–8) makes a similar point regarding the very short-lived individuals normally posited by four-dimensionalists. ¹⁷ In a similar vein, R. T. Mullins (2014: 130–1) argues that the conjunction of mereological universalism and four-dimensionalist accounts of persistence over time will make the problem of evil harder for the theist to resolve.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

105

4.3. Motivation 3: Theistic mereological anti-realism allows us to explain mereological pairing relations Recall that theistic mereological anti-realism comes in two varieties. According to the weaker variety, there is a y such that the xs compose y if and only if God wills that there is a y which those xs compose. According to the stronger variety, there is a y such that the xs compose y if and only if God wills that there is some specific y which those xs compose. The weaker version of theistic mereological anti-realism tells us that God decides when composition occurs, but the stronger version of theistic mereological anti-realism goes a step further and also tells us that God decides how the relata of mereological relations are paired. God decides not only that some xs will compose something or other but also that the xs will compose this particular thing, y. Why is this significant? Well, one objection to composition, which I have pressed elsewhere (Brenner 2015: 328‒9; 2017: 471‒4), is the “mereological pairing problem.” The mereological pairing problem is to some extent modeled after Jaegwon Kim’s pairing problem for substance dualism (see, e.g., Kim 2005: ch. 3).¹⁸ Kim thinks the substance dualist cannot tell a satisfactory story regarding the manner in which immaterial souls are paired with material bodies. For example, why is this soul paired with this body, rather than that one? Kim contends that the substance dualist cannot provide a satisfactory answer to this question. Similarly, I contend that those who believe in composition cannot tell a satisfactory story regarding the manner in which composite objects are paired with their proper parts. Why do these proper parts compose this composite object? Why is this composite object composed of these proper parts? I contend that those who believe in composition cannot provide satisfactory answers to these questions. Rather, the mereological pairing relations will be brute. This is problematic, I contend, because brute relations add to the complexity of our total theory.

¹⁸ The pairing problem is most often associated with Kim, but he says that the pairing problem originates with Foster 1968, 1991. In fact, while Foster does develop the pairing problem independently of earlier philosophers, the earliest discussion of the pairing problem which I am aware of occurs in Vasubandhu’s Abhidharmakośa (Vasubandhu 2009: 301).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

106   But, as we’ve seen, the strong variant of theistic mereological anti-realism tells us that mereological pairing relations are not brute, but are rather decided by God. The strong variant of theistic mereological anti-realism therefore resolves the mereological pairing problem. This theistic resolution to the mereological pairing problem resembles Richard Swinburne’s (1986: 198‒9) and John Foster’s (2001: 29) suggestion that God decides which souls are paired with which bodies. There are several possible variants of this theistic response to the mereological pairing problem.¹⁹ Perhaps we need not endorse a particularly strong variant of theistic mereological anti-realism, according to which God can decree, for any things arranged, say, human-wise, that those things compose a different human every five seconds. Perhaps we need only endorse a variant of theistic mereological anti-realism according to which God decides, of some things arranged human-wise, which human will be associated with those parts over the course of an entire human lifetime. The theistic response to the mereological pairing problem is preferable to many alternative responses we might think of. By way of illustration, consider one response to the pairing problem for dualism which Foster endorsed prior to his endorsement of the theistic response to the pairing problem. Foster suggests that a resolution to the pairing problem might be found in brute laws regarding the interaction of bodies and souls, where each law of this sort is such that it only governs the interaction of one particular body and one particular soul (Foster 1991: 167‒9). This response to the pairing problem requires an immense number of brute laws: one law for each soul (or, perhaps, for each body). We should avoid postulating so many fundamental laws, since they complicate our total theory immensely. For the same reason we should also refrain from endorsing a similar solution to the mereological pairing problem, involving mereological laws each of which governs the pairing of a single composite object with its parts—or, even worse, since we will require more laws, mereological laws each of which governs the pairing of a single part with any composite objects it composes. It is much simpler to suppose that a single being, God, pairs composite objects with their parts, ¹⁹ Thanks to an anonymous referee for making this point.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

107

rather than be forced to posit an extremely large number of brute mereological laws governing the pairing of composite objects with their parts.

4.4. Motivation 4: Theistic mereological anti-realism provides novel resources for making sense of the general resurrection of the dead Many theists (especially many Jewish, Christian, and Muslim theists) believe that many or all human beings will be resurrected, along with their physical bodies, at some point in the future. It is often supposed that this sort of resurrection will be particularly problematic if we are each identical with some composite physical object (e.g., we are each identical with an organism, or a brain, or whatever).²⁰ After all, when we die the physical objects with which we are identical decay, and our proper parts are eventually scattered. If we do not cease to exist immediately at death, then at any rate many of us cease to exist relatively soon after our deaths, well before any general resurrection of the dead is supposed to occur. How can God bring us back if we are destroyed? Even dualists have often faced this sort of problem, since dualist theists who believe in a general resurrection of the dead have traditionally held to the thesis that our bodies will be resurrected as well. How can those bodies be resurrected if they have long since gone out of existence? These problems have been addressed before.²¹ I don’t claim that proponents of the doctrine of the general resurrection of the dead must appeal to theistic mereological anti-realism in order to make sense of that doctrine. But it is still worth mentioning that theistic mereological antirealism gives us an easy way to make sense of the general resurrection in ²⁰ See, for example, Eric Olson (1997: 71): “The Biological Approach . . . bears on some religious doctrines. On that view, you are an animal, and an animal ceases to exist when it dies—when its vital functions cease and its tissues decay beyond the point where they can be reanimated. So existence after death seems to be ruled out. Once biological death has occurred, not even God can call you back into being, at least if I am right about what it takes for an animal to persist through time . . . ” ²¹ Among analytic philosophers see, for example, van Inwagen 1978, Zimmerman 1999, Baker 2001.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

108   a manner which has mostly been neglected. If we are physical objects, how are we resurrected? Well, if we are physical objects then we are presumably composite objects. As we saw above, the stronger variety of theistic mereological anti-realism ensures that God decides which composite objects are paired with which proper parts: God decides not only that some xs will compose something or other but also that the xs will compose this particular thing, y. When it’s time for the general resurrection to take place, God can simply will that some proper parts compose some specific person who has been dead for many years. God can make the same sort of move with our bodies, assuming that we are not identical with our bodies. Theistic mereological anti-realism allows us to avoid an additional difficulty which traditionally plagues attempts to make sense of the general resurrection. It is often supposed that our bodies at the general resurrection must have some or all of the proper parts we had at the moment of our deaths (cf. Hershenov 2002). This requirement becomes particularly difficult to satisfy in scenarios involving cannibals, cannibals who eat only other cannibals, and so on. Theistic mereological antirealism allows us to avoid all of these difficulties, since, given theistic mereological anti-realism, we need not accept any requirement to the effect that, at the general resurrection, one (or one’s body) must have some or all of the proper parts one had at one’s death.²² Yang and Davis (2017: §11.3) also note that theistic mereological antirealism can help us make sense of the general resurrection of the dead, and much of what they say about this subject is similar to what I say about it. But they assume that the resurrected body must have “a sufficient number of particles that once composed the pre-mortem body” (Yang and Davis 2017: 223). As I note above, this requirement seems to me to be unnecessary. My theistic mereological anti-realist account of the general resurrection also resembles Lynne Baker’s contention that God can freely decide at the general resurrection which bodies constitute which people (Baker 2011: 56). But my account does

²² This point is consonant with the Qur’ān (36: 77–82), which emphasizes God’s ability to create people (either initially, or at the general resurrection of the dead) from whatever materials God wants.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

109

not require that we accept a view in personal ontology similar to Baker’s, according to which people are constituted by, but not identical with, their bodies.²³ We can also note that theistic mereological anti-realism can help us make sense of reincarnation, under the assumption that we are physical objects. God can ensure that the composite object associated with some parts arranged embryo-wise, or fetus-wise (or whatever) is the same composite object as some deceased person.²⁴

4.5. Motivation 5: Theistic mereological anti-realism provides novel resources for responding to the problem of evil A large part of the problem of evil involves reconciling the existence of God with the existence of apparently gratuitous suffering. We have some idea of what sorts of physical brain states some phenomenal states are supervenient upon, and it is undeniable that the physical brain states associated with severe suffering are often instantiated. However, we cannot tell, on the basis of empirical observation, whether in these cases there are composite persons who correspond to the brain states in question. Perhaps, sometimes, when parts arranged brain-wise (or person-wise, or animal-wise, or whatever) are just about to bring about horrible phenomenal states, God ensures that those parts cease to ²³ My account of the general resurrection superficially resembles the idea that, in reassembling some physical parts, God can ensure that they compose one past person rather than another by ensuring that the reassembly of parts is causally related to one of those people rather than the other one. This suggestion was made by Dean Zimmerman (2013: 138–9) in order to account for presumably extremely unlikely cases where two people who died at different times died with all of the same proper parts. God ensures that God’s reassembly of those proper parts is causally related to one of those people’s past lives rather than the other in order to break the tie between them, since otherwise they would have equal claim to being the person who results when God reassembles the proper parts. My suggestion is, by contrast, much more general: God can always decide which people are composed by which proper parts. God need not do this merely in order to “break ties.” What’s more, God does not decide who is composed of which proper parts merely indirectly, by ensuring that the reassembly of proper parts is appropriately causally related to some particular past person. ²⁴ Western cultures rarely associate reincarnation with theism. But reincarnation is endorsed by some major theistic traditions, including some strands of Hinduism and Judaism (cf. Goldschmidt and Seacord 2013).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

110   compose anything, so that there is temporarily no composite object associated with those parts, and so no composite person²⁵ to experience the horrible phenomenal state.²⁶ I recognize that this proposal will be rejected by many readers out of hand. It can initially seem ridiculous, or even offensive. I share this reaction, and perhaps it’s worth reminding the reader at this point that I don’t actually endorse the proposal, since I don’t endorse theistic mereological anti-realism. On reflection, however, the proposal is more defensible than it initially seems. This is surprising, but it is true nevertheless. So before you reject the proposal, you should see whether the objections you have stand up under closer scrutiny. Upon reflection, I’m not confident that they will, and so I think that the proposal is worth serious consideration. So, why might the proposal seem to be so implausible? First, it can seem wildly ad hoc. But on reflection the proposal is not ad hoc. For, first, the proposal involves theistic mereological anti-realism, so that God has the ability to decide when composition does and does not occur. As we’ve seen, theistic mereological anti-realism is motivated on other grounds as well—we do not appeal to theistic mereological anti-realism here merely in order to help resolve the problem of evil. A second reason the proposal is not ad hoc is that if phenomenal states are identical with micro-physical or functional states, then if God wants to prevent some instance of suffering, God may have no better option than to ensure that, for as long as the suffering would persist absent God’s intervention, the proper parts instantiating the micro-physical or functional state associated with that suffering fail to compose anything.²⁷ God might, of course, simply ensure that the micro-physical or functional states fail to obtain. But this would presumably require that God

²⁵ For simplicity I’ll write of “persons” in this context. But this account of how God might avoid the occurrence of otherwise gratuitous suffering could also be extended to non-person animals. ²⁶ Where do the composite persons go? There are a few options here. Perhaps they briefly cease to exist. Or perhaps they briefly exist as simples. Or perhaps they are briefly composed of some parts other than the parts which normally compose them. ²⁷ By contrast, if some sort of dualism were true then perhaps God could simply suspend the psycho-physical laws, and ensure that the physical or functional states normally associated with some objectionable mental state fail to bring about that mental state.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

111

intervene in the causal order, to curtail the causal antecedents of the micro-physical or functional states in question. It would probably require, then, that God suspend whatever physical laws govern the transition from those antecedent states to the micro-physical or functional states God wants to avoid. Either way, then, God must suspend some law of nature in order to prevent the suffering from occurring: either the physical laws governing the causal interaction between microphysical or functional brain states and the causal antecedents of those states, or the mereological laws which ensure that, normally, those micro-physical or functional brain states are associated with a composite object which experiences intense suffering. I don’t see why God should obviously be more inclined to modify or suspend the former causal laws rather than the latter mereological laws. Non-theists who advance the problem of evil often contend that if God existed God would regularly suspend the causal laws (i.e., God would regularly bring about miracles) in order to ensure that various instances of gratuitous suffering do not occur. If such non-theists are so confident that God would regularly suspend the causal laws, why should they scoff at the suggestion that God might, for the same reason, sometimes suspend the mereological laws? The problem of evil is often motivated by the empirically observable fact that causal laws are often not suspended: Fawns obviously do burn in fires, cancer cells obviously do spread, and so on.²⁸ By contrast, the operation of the mereological laws which God might instead suspend cannot be observed, so that if on occasion God does suspend those laws, we wouldn’t be able to tell on the basis of observation. To recap: Some instances of suffering seem to be particularly hard to reconcile with the existence of an all powerful and all good God. The critic of theism might reasonably contend that God should (or would be expected to) miraculously intervene to ensure that these cases of suffering do not occur. Given theistic mereological anti-realism, we have a new way of showing how God might very well intervene in these cases, without suspending the causal order, and in such a manner that God’s

²⁸ Or, more carefully: things arranged fawn-wise obviously do burn in things arranged firewise, and things arranged cancer cell-wise obviously do spread among things arranged bodywise.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

112   actually intervening to prevent the suffering in question is compatible with our empirical evidence. So, the suggestion that God does in fact occasionally suspend the mereological laws does not seem to me to be at all an ad hoc response to the problem of evil—in fact, it seems to be what we might reasonably expect to happen, if God can in fact decide when composition occurs. An additional concern regarding the idea that God might occasionally prevent otherwise gratuitous suffering by temporarily preventing the parts of the suffering person from composing that person is that on this proposal God engages in massive or widespread deception. It is often assumed, in discussions of the problem of evil, that we should avoid proposed solutions to the problem which appeal to God’s engaging in massive deception (cf. Swinburne 1998: 139; van Inwagen 2006: 120‒1). God might be engaged in massive deception in this case for at least two reasons. First, it will seem to be the case that people are suffering when they are not. Second, people will seem to remember painful experiences which they didn’t actually experience, since their parts failed to compose anything at those times at which they think that the suffering occurred. I agree that there is a great deal of prima facie plausibility in the assumption that God would not engage in massive deception, at any rate, as Swinburne emphasizes, regarding important matters. But God need not engage in massive deception in order to prevent suffering by temporarily modifying or suspending the mereological laws. I never meant to suggest that God invariably suspends those laws when God might thereby prevent a great deal of suffering. I want to make the much more modest suggestion that God might sometimes take advantage of this possibility, just as God might occasionally suspend other sorts of laws of nature in order to prevent certain sorts of evils. It is acknowledged by many people that God (if God exists) might occasionally be expected to suspend laws of nature in order to prevent certain evils—for example, God might occasionally be expected to suspend whatever laws might otherwise result in the spread of a malignant cancer. Similarly, I claim, God might occasionally suspend mereological laws in order to avoid otherwise gratuitous cases of suffering. If we are deceived in some manner insofar as God occasionally suspends those laws, we are not at any rate massively deceived, if God’s suspension of the mereological laws

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

113

is not too frequent. And to the extent that we are deceived, the badness of this deception might be overridden by the good which results from God’s temporary suspension of the mereological laws. I think we would say more or less the same thing regarding God’s occasional suspension of causal laws. Perhaps there is a sense in which God’s suspension of causal laws deceives us regarding the world’s causal order. But this is a mild sort of deception, the badness of which can presumably often be overridden by the good which results from God’s temporary suspension of the causal laws. Importantly, if we thought that God did invariably intervene to prevent great suffering, then we might think that we are never morally obligated to act to prevent suffering, since we can expect God to unilaterally prevent that suffering. And it might be thought to be objectionable if theistic mereological anti-realism undermines our motivation to act to prevent suffering. This problem too is avoided, since my proposal is only that God might be expected to occasionally suspend composition in order to prevent cases of great suffering, just as God might be expected to occasionally intervene in other ways to prevent cases of great suffering. A final concern for my proposal regarding the problem of evil is that, even if the composite person normally associated with some parts does not experience the severe suffering which they would normally experience, their parts would continue to (collectively) suffer. It’s great that the composite person would not suffer, but it’s also objectionable if their parts collectively suffer.²⁹ My response to this concern requires that I take on a controversial assumption, namely that objects can only jointly instantiate phenomenal states (such as the phenomenal states associated with suffering) by composing something which instantiates those states.³⁰ So, for example, an organism might suffer, but the proper parts of that organism do not also (collectively) suffer—or at any rate the

²⁹ This is a concern which a typical Buddhist proponent of the non-self doctrine might have. According to proponents of the non-self doctrine selves (including composite persons) do not ultimately exist. Nevertheless, Buddhist proponents of this doctrine typically still think that simples can collectively suffer, and we should work to eliminate this suffering, even if there are no persons or selves which suffer. See Siderits 2007: ch. 4. ³⁰ For some philosophers who think that this assumption is plausible, see van Inwagen 1990: §12; Olson 2007: 188–93; Bailey 2016; Dowland 2016. For philosophers who do not think that the assumption is plausible, see Rosen and Dorr 2002: 159–60; Sider 2013: 268–9.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

114   only sense in which they do collectively suffer is insofar as they compose an organism which suffers. By contrast, if the organism weighs 100 kg, its proper parts will also collectively weigh 100 kg.³¹

5. Conclusion In this chapter I’ve discussed mereological anti-realism, the idea, roughly, that the circumstances under which composition occurs have something to do with one or more (actual or counterfactual) agent’s mental states. I described two non-theistic variants of mereological antirealism, due to Kriegel and Pearce, and found that they both suffer from significant objections. I went on to describe and motivate a theistic version of mereological anti-realism, which itself came in weaker and stronger variants. The weaker variant: there is a y such that the xs compose y if and only if God wills that there is a y which those xs compose. The stronger variant: there is a y such that the xs compose y if and only if God wills that there is some specific y which those xs compose. At the end of the day I’m not sure whether theistic mereological anti-realism is true. Nevertheless, I think it has a lot to be said for it, and it does not fall prey to the objections which plague extant nontheistic versions of mereological anti-realism. Where this leaves us, I think, is with a distinctively theistic perspective on the metaphysics of composition, one which merits further scrutiny.³²

³¹ Molto (forthcoming) argues that in cases where we have indirect evidence of animal suffering (e.g., we find an animal carcass), for all we know God may have intervened in order to ensure that the animal in question did not suffer. This thesis coheres very well with the point I’ve made in this section, although I suggest that God might intervene in order to prevent any sort of suffering, rather than unseen animal suffering in particular. DeRose (2020) also suggests that God might surreptitiously shield us from some otherwise gratuitous evils, in the sense that the evils do not occur, despite the fact that they seem to occur. This idea also coheres very nicely with what I’ve written in this section, and DeRose provides additional reasons to think that this response to the problem of evil is not as objectionable as we might initially think it is. DeRose does not, however, develop his proposal in term of the mereological mechanism I’ve described in this section. ³² Thanks to Spencer Case, Rebecca Chan, Dustin Crummett, David Efird, Peter Finocchiaro, Jack Himelright, Michael Longenecker, Tim O’Connor, Timothy Perrine, Callie Phillips, Michael Rea, Dean Zimmerman, several referees for Oxford Studies in Philosophy of Religion, and the audience at my presentation of this chapter at Wuhan University.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

115

References George A. Alvarez. Representing multiple objects as an ensemble enhances visual cognition. Trends in Cognitive Science, 15(3): 122–31, March 2011. Thomas Aquinas. Summa Theologica. Benziger Bros., New York, 1947. Andrew M. Bailey. Composition and the cases. Inquiry, 59(5): 453–70, 2016. Lynne Rudder Baker. Persons and Bodies: A Constitution View. Cambridge University Press, Cambridge, 2000. Lynne Rudder Baker. Material persons and the doctrine of resurrection. Faith and Philosophy, 18(2): 151–67, April 2001. Lynne Rudder Baker. Christian materialism in a scientific age. International Journal for Philosophy of Religion, 70(1): 47–59, August 2011. Einar Duenger Bohn. From Hume’s dictum via submergence to composition as identity or mereological nihilism. Pacific Philosophical Quarterly, 95(3): 336–55, September 2014. Andrew Brenner. Mereological nihilism and theoretical unification. Analytic Philosophy, 56(4): 318–37, December 2015. Andrew Brenner. Mereological nihilism and personal ontology. The Philosophical Quarterly, 67(268): 464–85, July 2017. Andrew Brenner. Science and the special composition question. Synthese, 195(2): 657–78, February 2018. Ross P. Cameron. The contingency of composition. Philosophical Studies, 136: 99–121, 2007. David J. Chalmers. Ontological anti-realism. In Ryan Wasserman, David J. Chalmers, David Manley, editors, Metametaphysics: New Essays on the Foundations of Ontology, pages 77–129. Oxford University Press, Oxford, 2009. Roderick Chisholm. Person and Object: A Metaphysical Study. Open Court Publishing Co., La Salle, IL, 1976. Sam Cowling. No simples, no gunk, no nothing. Pacific Philosophical Quarterly, 95(2): 246–60, June 2014. Todd DeRose. Empirically skeptical theism. Faith and Philosophy, 37(3): 323–35, July 2020. S. Clint Dowland. Embodied mind sparsism. Philosophical Studies, 173(7): 1853–72, July 2016.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

116   Edward Feser. Five Proofs for the Existence of God. Ignatius Press, San Francisco, 2017. John Foster. Psychophysical causal relations. American Philosophical Quarterly, 5: 64–70, 1968. John Foster. The Immaterial Self. Routledge, London, 1991. John Foster. A brief defense of the cartesian view. In Soul, Body, and Survival: Essays on the Metaphysics of Human Persons, pages 15–29. Cornell University Press, Ithaca, NY, 2001. Lloyd P. Gerson. Plotinus. Routledge, London, 1994. Fernand Gobet, Peter C. R. Lane, Steve Croker, Peter C.-H. Cheng, Gary Jones, Iain Oliver, and Julian M. Pine. Chunking mechanisms in human learning. Trends in Cognitive Science, 5(6): 236–43, June 2001. Tyron Goldschmidt and Beth Seacord. Judaism, reincarnation, and theodicy. Faith and Philosophy, 30(4): 393–417, October 2013. Mark Heller. The donkey problem. Philosophical Studies, 140(1): 83–101, July 2008. David B. Hershenov. Van inwagen, zimmerman, and the materialist conception of resurrection. Religious Studies, 38(4): 451–69, December 2002. Eli Hirsch. Quantifier Variance and Realism: Essays in Metaontology. Oxford University Press, New York, 2011. Terence Horgan and Matjaž Potrč. Austere Realism: Contextual Semantics Meets Minimal Ontology. The MIT Press, Cambridge, MA, 2008. Ross Inman and Alexander Pruss. On Christian theism and unrestricted composition. American Philosophical Quarterly, 56(4): 345–60, 2019. Mark Johnston. Surviving Death. Princeton University Press, Princeton, 2010. Mark Johnston. The personite problem: Should practical reason be tabled? Noûs, 51(3): 617–44, September 2017. Jaegwon Kim. Physicalism, or Something Near Enough. Princeton University Press, Princeton, 2005. Daniel Z. Korman. Objects: Nothing out of the Ordinary. Oxford University Press, Oxford, 2015. Uriah Kriegel. Composition as a secondary quality. Pacific Philosophical Quarterly, 89: 359–83, 2008. Uriah Kriegel. Kantian monism. Philosophical Papers, 41(1): 23–56, March 2012.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

117

John Kronen and Sandra Menssen. The argument from wholes: A classical Hindu design argument for the existence of God. Faith and Philosophy, 30(2): 138–58, April 2013. Brian Leftow. God and Necessity. Oxford University Press, Oxford, 2012. David Lewis. On the Plurality of Worlds. Blackwell, Oxford, 1986. David Lewis. Parts of Classes. Blackwell, Oxford, 1991. Antonia LoLordo. Jonathan Edwards’s argument concerning persistence. Philosophers’ Imprint, 14(24): 1–16, July 2014. Christopher Menzel. The argument from collections. In T. Doughertyand J. Walls, editors, Two Dozen (or so) Arguments for God: The Plantinga Project. Oxford University Press, Oxford, 2018. Kristie Miller. The existential quantifier, composition and contingency. Erkenntnis, 73(2): 211–35, September 2010. Daniel Molto. The problem of evil: Unseen animal suffering. Religious Studies, forthcoming. R. T. Mullins. Four-dimensionalism, evil, and Christian belief. Philosophia Christi, 16(1): 117–37, 2014. Eric T. Olson. The Human Animal: Personal Identity without Psychology. Oxford University Press, Oxford, 1997. Eric T. Olson. What Are We? A Study in Personal Ontology. Oxford University Press, Oxford, 2007. Robert Carry Osborne. Debunking rationalist defenses of common-sense ontology: An empirical approach. Review of Philosophy and Psychology, 7 (1): 197–221, March 2016. Kenneth L. Pearce. Mereological idealism. In Kenneth L. Pearce and Tyron Goldschmidt, editors, Idealism: New Essays in Metaphysics, pages 200–16. Oxford University Press, Oxford, 2017. Alvin Plantinga. How to be an anti-realist. Proceedings and Addresses of the American Philosophical Association, 56(1): 47–70, September 1982. Alvin Plantinga. Where the Conflict Really Lies: Science, Religion, and Naturalism. Oxford University Press, Oxford, 2011. Plato. Timaeus and Critias. Oxford University Press, Oxford, 2008. Plotinus. Ennead. Hilary Putnam. The Many Faces of Realism. Open Court, La Salle, IL, 1987.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

118   Hilary Putnam. Ethics Without Ontology. Harvard University Press, Cambridge, MA, 2004. Michael C. Rea. The metaphysics of original sin. In Dean Zimmerman and Peter van Inwagen, editors, Persons: Human and Divine, pages 319–56. Oxford University Press, Oxford, 2007. Justin Remhof. Nietzsche’s Constructivism: A Metaphysics of Material Objects. Routledge, New York, 2018. Gideon Rosen and Cian Dorr. Composition as a fiction. In Richard Gale, editor, The Blackwell Guide to Metaphysics, pages 151–74. Blackwell, Oxford, 2002. Theodore Sider. Against parthood. In Karen Bennett and Dean W. Zimmerman, editors, Oxford Studies in Metaphysics, volume 8. Oxford University Press, Oxford, 2013. Mark Siderits. Buddhism as Philosophy: An Introduction. Ashgate, Aldershot, 2007. Alexander Skiles. Against grounding necessitarianism. Erkenntnis, 80(4): 717–51, August 2015. Richard Swinburne. The Evolution of the Soul. Oxford University Press, Oxford, 1986. Richard Swinburne. Providence and the Problem of Evil. Clarendon Press, Oxford, 1998. Peter van Inwagen. The possibility of resurrection. International Journal for Philosophy of Religion, 9(2): 114–21, June 1978. Peter van Inwagen. Material Beings. Cornell University Press, Ithaca, NY, 1990. Peter van Inwagen. The Problem of Evil: The Gifford Lectures Delivered in the University of St Andrews in 2003. Clarendon Press, Oxford, 2006. Vasubandhu. Vasubandhu’s abhidharmakośa: The critique of the soul. In Jay L. Garfield and William Edelglass, editors, Buddhist Philosophy: Essential Readings, pages 297–308. Oxford University Press, Oxford, 2009. Translated by Charles Goodman. Eric T. Yang and Stephen T. Davis. Composition and the will of god: Reconsidering resurrection by reassembly. In Paradise Understood: New Philosophical Essays About Heaven, pages 213–27. Oxford University Press, Oxford, 2017.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

     -

119

Dean W. Zimmerman. The compatibility of materialism and survival: The “falling elevator” model. Faith and Philosophy, 16(2): 194–212, April 1999. Dean W. Zimmerman. Personal identity and the survival of death. In Jens Johansson, Ben Bradley, and Fred Feldman, editors, The Oxford Handbook of Philosophy of Death, chapter 4, pages 97–154. Oxford University Press, Oxford, 2013.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

4 Creativity in Creation Meghan D. Page

1. Introduction My household contains five food allergies and two working parents; deciding what to make for dinner poses no small challenge. Our current response to this conundrum is a binder that contains various allergyfriendly recipes which can be prepared in twenty minutes or less. When it’s time to make dinner, or to meal-plan for the week, my spouse and I sort through the binder and decide which recipes we will bring into existence. Occasionally, I lament the days when I could exercise creativity in the kitchen—times I would wander through the farmer’s market looking for inspiration among the eggplants, plotting how I might build a meal around the perfect bulb of fennel. Although I am no expert, I’ve always found joy in cooking “off-book,” fusing different techniques and flavor profiles to create novel dishes. Even for those who don’t cook, the distinction between recipe-based cooking and creative cooking is an intuitive one. It is an example of a more general distinction among artists—those who make and those who create. While a novice sculptor aims to imitate the forms of authorities in her field, the master sculptor develops her own unique style and artistry. Creative artistry requires originality. Although I currently make something for dinner several nights a week, it’s increasingly rare that I create something for dinner. Fair enough. Many nights just getting something on the table feels like a heroic act. But is this lower form of craftship also adequate for describing the creative choices of an all-powerful God forming the world? According to the primary philosophical picture of creation, it is. As Klaas Kraay writes, Meghan D. Page, Creativity in Creation In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Meghan D. Page 2022. DOI: 10.1093/oso/9780192862976.003.0004

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

121

In contemporary analytic philosophy of religion, this [God’s choice in creation] is construed as follows: God surveys the set of possible worlds, and then freely chooses exactly one for actualization on the basis of its axiological properties. (Kraay, 2008, 855)

If we translate this model into culinary terms, God surveys the set of possible recipes, and then chooses exactly one to make for dinner. On this picture, even if God made the world, they didn’t really create it. Although making requires skill—obtaining a recipe for Bérnaise sauce from Thomas Kellar won’t prevent you from accidentally scrambling the eggs—the creator of an artifact bears a unique relationship to her work that is unparalleled by the maker. We understand an artist’s creation to reveal something about her creative intentions and intellect. As an example, Fallingwater is taken to display the style and artistry of its designer, Frank Lloyd Wright, rather than the contractor who oversaw its construction. Theists often assume God’s relationship to the world is analogous to an artist’s relationship to her art; the natural world reveals the mind of God in the same way Fallingwater exemplifies the mind of Wright. This assumption is implied by both common religious practices—for example our tendency to praise God for the glory revealed in creation—and natural theological arguments, like the argument from design, which take the structure of the world to suggest the existence of a designer. But if God creates the world by means of actualization, it’s unclear whether the artist‒artifact analogy holds. In this chapter, I highlight several ways in which the worldactualization model of creation (hereafter WAM) is in tension with divine creativity. First, I introduce three types of creative thought processes described by Margaret Boden and argue the God of WAM fails to exhibit any of them. Next I explore ways to ease the tension between divine creativity and the world-actualization model. I then consider potential transformations of WAM that might be friendlier to divine creativity, but conclude that it is difficult to maintain the artist‒artifact analogy in conjunction with WAM. Finally, I offer three desiderata to keep in mind in the search for a creative model of creation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

122  . 

2. Combination In order for the artist‒artifact analogy to hold, it is crucial that God acts creatively in creating the world. But what precisely does creative action entail? Cognitive scientist Margaret A. Boden offers a particularly clear and rigorous taxonomy of creative thought processes in her book Creativity & Art. Boden’s primary interest is whether machines can exemplify creativity. In order to answer this question, she first defines the various ways agents display intellectual creativity, and evaluates whether machines are capable of such processes. Although I will adhere to Boden’s taxonomy of creative processes, my aim in this chapter is not to evaluate whether God is capable of creative thought but instead to evaluate whether the God of WAM invokes these creative thought processes in constructing the world. I understand Boden’s types of intellectual creativity to be ways an artist makes creative contributions to her work. In order for God to be related to the world in a way that justifies the artist‒artifact analogy, it is important that God makes creative contributions to the world. In this section, I describe Boden’s three types of creative processes. I then turn to the work of Alvin Plantinga to flesh out the key features of the world-actualization model of creation. Finally, I argue that actualizing the world does not involve making a creative contribution.

2.1 Boden on Creativity According to Margaret Boden, there are three different kinds of creativity: combinational, exploratory, and transformational. Creativity in general is the generation of novel, surprising, and valuable ideas. “Ideas”, here, is a catch-all term covering not only concepts and theories but also (for example) music and literature, and artefacts such as architecture, sculpture, and paintings. The three types of creativity . . . are defined by the different kinds of psychological processes that generate the new structures. (Boden, 2010, 1)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

123

Boden’s three categories of creativity describe ways agents engage with conceptual spaces, which are “structured styles of thought” built by a community and taken on by an individual (Boden, 2010, 32). I understand these different styles of thought as the ways an artist makes a creative contribution to an artifact. In the first sort of creativity, combinatorial, an artist combines images or models from different conceptual spaces in a novel way, “making unfamiliar combinations of familiar ideas.” Puns exemplify this sort of creativity. Suppose upon returning from a bank in the UK I make the comment, “Do I look heavier to you? I just picked up 100 pounds.”¹ This pun recombines two familiar uses of ‘pound’—a measurement of weight and a monetary unit—in a humorous way. This sort of creative recombination is not random; the concepts are recombined in just the right way to produce some kind of “value” (in this case, attempted comedic relief).² Boden’s second category, exploratory creativity, involves working inside of a particular conceptual space and exploring all of the possibilities within it. Boden uses the example of chess. The rules of chess define a (rather large) number of possible moves—some of which have never been played. A chess player displays exploratory creativity when she discovers a move or sequence of moves that she was previously unaware of. Here, the agent sheds light on new areas of the conceptual space that, though allowed by the “rules,” so to speak, have not been illuminated or clarified. Similarly, to construct a new painting in a familiar style displays exploratory creativity. An artist explores the bounds of a particular tradition, for example cubism, as she paints various objects within the framework.

¹ Since the topic of this chapter is creativity, I find it important to point out that my spouse, not I, was the author of this pun. I merely recalled the last pun I could remember. ² Of course, while this may be the first time I came up with the pun, these two ideas have been combined in a similar way many times before. For this reason, Boden distinguishes between psychological (P) and historical (H) types of creativity. H-creativity is a special case of P-creativity, because the creative act is surprising to the artist as well as the world (as it’s never been done before). While the pun may exhibit P-creativity, it fails to exhibit H-creativity. However, in the case of the divine, I will assume that any case of P-creativity also implies H-creativity. Nevertheless, my primary interest here is an account of how a creator is connected to her work rather than how that work connects to a community. For that reason, following Boden, I will leave issues about H-creativity to the side and instead focus on P-creativity (Boden, 2010, 30).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

124  .  Transformative creativity, Boden’s third category, is particularly radical. The transformative artist changes the conceptual space in which she works; she creates possibilities which did not previously exist. As Boden describes, The deepest cases of creativity involve thinking something which, with the respect to the conceptual spaces in their minds, they couldn’t have thought before. The supposedly impossible idea comes about only if the creator changes the pre-existing style in some way. It must be tweaked, or even radically transformed, so that thoughts are now possible which previously (within the untransformed space) were literally inconceivable . . . (Boden, 2010, 34‒5)

A paradigmatic case of a transformative artist is Jackson Pollock. Although Pollock was trained in representational art, he transformed his style by moving from the use of a brush to the pouring (and ultimately dripping) of paint. Introducing this alternative method allowed him to create paintings that were inconceivable within his previous style. We can think of similar movements within science—the introduction of wave-particle duality introduces a host of physical possibilities which were ruled out according to classical mechanics. One thing to note, which will become salient later, is that on Boden’s picture the transformative artist creates new possibilities with respect to a particular possibility space. While there may be some sense in which dripstyle paintings were possible prior to Pollock, the conceptual spaces of Pollock and his contemporaries ruled them out.

2.2 Creation in the World-Actualization Model Now that we have an account of creativity, let’s turn our attention to WAM. Although it’s hard to trace WAM to a single origin, Leibniz certainly popularized the picture in his argument that this world is the best of all possible worlds. On Leibniz’s view, God’s perfect knowledge entails knowledge of every possible world, God’s goodness entails the desire to make the best world possible, and God’s power entails the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

125

ability to make any possible world. Given that we are in the world God made, we can be sure it is the best world possible (Leibniz, 1989, 48). Although many think of Leibniz as the father of WAM, contemporary discussions of the model are fairly disconnected from the Leibnizian tradition and instead fueled by the resurgence of possible-world talk that occurred after Kripke’s work on formal semantics for modal logics in the late 1950s. Renewed philosophical interest in modality influenced much philosophy of religion that occurred in the latter half of the twentieth century. One prime example of this sociological phenomenon is the work of Alvin Plantinga, particularly his book God, Freedom and Evil, which surveys several traditional philosophical problems in light of contemporary modal logic. Of particular interest to the discussion at hand is Plantinga’s response to J. L. Mackie’s logical problem of evil. Mackie follows Leibniz in thinking that a perfect God would surely create the best of all worlds. However, unlike Leibniz, Mackie is certain this world is not the best possible. As a result, Mackie argues that there is not a perfectly good God, because such a God would not create a world that contains evil. To those who suggest the evil in this world is a consequence of free will, Mackie insists that a perfect God would create beings that freely choose good rather than evil. In response to this, Plantinga denies that there is necessarily a best possible world and further argues that not every possible world can be actualized. The Free Will Defender disagrees with both Leibniz and Mackie. In the first place, he might say, what is the reason for supposing that there is such a thing as the best of all possible worlds? . . . But what is really characteristic and central to the free will defense is the claim that God, though omnipotent, could not have actualized just any possible world he pleased. (Plantinga, 1977, 34)

Before discussing why Plantinga thinks God could not have actualized “just any possible world he pleased”, I want to say a bit more concerning the process of world actualization. On Plantinga’s view, “a possible world is any possible state of affairs that is complete” (Plantinga, 1977, 36). He

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

126  .  further suggests that each possible world can be described by a distinct set of propositions, or world book. For Plantinga, possible states of affairs are coeternal with God. Every possible world (and corresponding worldstory) just is. This allows Plantinga to make an important distinction between actualization and creation. . . . we must note that God does not, strictly speaking, create any possible worlds or states of affairs at all . . . What He did was to perform actions of a certain sort—creating the Heavens and Earth, for example—which resulted in the actuality of certain states of affairs. God actualizes states of affairs. He actualizes the possible world that does in fact obtain; He does not create it. And while He created Socrates, He did not create the state of affairs consisting in Socrates’ existence. (Plantinga, 1977, 38‒9)

God creates by making the physical world we inhabit. God makes the physical heavens and the physical earth. But the state of affairs that describes heaven and earth is not created by God. God creates physical persons, but those persons instantiate already existing states of affairs. To return to our earlier analogy, God executes the recipe for the world, but God does not write it. On Plantinga’s view, the free choices of creatures partly determine which possible world is actualized. God strongly actualizes a state of affairs, and this state evolves according to the choices of free creatures (perhaps together with other indeterministic forces if some kind of physical indeterminism is true) into a complete state of affairs. This latter possible world is weakly actualized by God, given that they strongly actualized its initial conditions (so to speak). Those who have a robust conception of divine foreknowledge, such as Molinists, will claim that God chooses to strongly actualize a state of affairs based on the world it will weakly actualize. But Plantinga’s view also leaves room for open theists who deny that there are facts about what world will be weakly actualized by the world God strongly actualizes. If the actual world is only weakly actualized by God, and partly determined by the free choices of creatures (and/or physical indeterminism), then it’s false that God could create any possible world, because

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

127

God couldn’t create a world where free creatures make choices other than the ones they in fact make. Therefore, Mackie’s argument fails. However, what’s most relevant to the project at hand is Plantinga’s sense of what creation looks like on the WAM. Regardless of whether God strongly actualizes an entire world or a set of initial conditions, for God to create is merely to instantiate (or materialize, or bring into existence) a pre-existing model. According to Plantinga, states of affairs just exist. As an omniscient knower, God is aware of every possible state of affairs, and God chooses to materialize one.³ If God creates by instantiating a pre-existing state of affairs, does God exemplify creativity? Let us consider Boden’s three categories, beginning with combinatorial creativity. If God merely selects a pre-existing state of affairs, God is not recombining anything. Every state of affairs is complete—the world book which describes this state of affairs contains every proposition needed to describe the world. Therefore, we can’t think of God as making a creative contribution to the world through combination. Exploratory creativity requires conceiving of a possibility that lies in the hidden corners of a conceptual framework, an option that exists but is not yet illuminated. But for God, all of the possible states of affairs are equally illuminated. It’s not the case that in actualizing a world God illuminates a hidden region of the divine conceptual space. Therefore, the God of WAM does not make a creative contribution by exploration. This leaves transformative creativity. To transform a conceptual space is to create a new possibility space. In humans, this happens when we find a way out of our limited conceptual space. However, God’s infinite conceptual space already contains all of the possible possibilities, as all possible states of affairs “just are.” For God to “transform” this conceptual space would imply that God was previously working inside a limited space. This would generate serious problems for divine omniscience, as it suggests that there are possibilities which God does not know (but then comes to know). The God of WAM does not creatively contribute to the world through transformation. ³ Elsewhere, Plantinga suggests these possible worlds depend on God, existing as ideas in God’s mind (see Plantinga, 1980). I will consider whether this resolves the problem in §4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

128  .  It seems there is no point at which the God of WAM makes a creative contribution to the world. Perhaps the God of WAM is a maker, but not a creator.

3. Exploration 3.1 Appreciating the Tension In the previous section, I argued that a tension exists between the worldactualization model of creation and traditional understanding of divine creativity. Before exploring whether this tension can be resolved, I will consider the significance of divine creativity. After all, the simplest way to relax a tension is to give up a tenet, and in this case it may be tempting to give up our view of God as “creative.” This resolves the tension, but I think it’s a mistake to dismiss divine creativity too quickly. Divine creativity plays a central role in religious practice. Religious traditions are filled with scriptures, hymns, and practices that represent God as a creator. For example, suppose a worshiper were to offer God praise for the exquisite design of a pomegranate. Surely the intent of such a worshiper is to marvel at the complexity of the fruit, the workmanship of the mind that created bright, juicy spheres of sweetness encapsulated within a greater sphere for protection, and not “thank you, God, for actualizing the world where pomegranates are juicy instead of the world where they are dry.” Such praises inherently represent God as making creative contributions to the pomegranate. One might object that I am placing too much emphasis on intellectual creativity. Even if God doesn’t create the blueprint of the world, God still provides it with substance. Many of the Psalmist’s praises, for example, extol the handiwork of God, or seem to emphasize God’s artisanship. Why not think that God’s creativity is found in the materialization of a world-book rather than its writing? On this picture, God’s majesty is reflected in creation by means of how well God executes a chosen design. In Boden’s general description of creativity she points out the categories are intended to apply to both concepts and artifacts. God may not create ideas but God does create the physical world. At times it sounds as if

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

129

Plantinga has this sort of creative process in mind when he suggests that God “creates the Heavens and the Earth” and in doing so actualizes a possible world. While I’m sympathetic to the idea of conceiving of creator-God as an artisan, I think most cases of artisanship still invoke intellectual creativity. In the human realm, “making” and “creating” are the ends of a continuum. A student might “make” a term paper by copying and pasting an article from the internet, but such a paper lacks any creative contributions from the student. However, in most cases where a human agent creates a new artifact, such as a painting or sculpture, she makes some sort of creative contribution even when following a model. A potter may have a particular design in mind when she sits down to throw at her wheel, but as she works she continuously reconceives of her intended artifact. Or consider, for example, a class where students are instructed to paint a copy of a particular painting. While the students will all follow the same design, their paintings will differ, and these differences will reveal unique creative contributions made by each student to her work. In the mortal realm, the creation of artifacts is never as simple as having a rigorous idea and physically actualizing it; even if a cook is following a recipe, she will likely have to make adjustments along the way. These adjustments constitute creative contributions. A baker may intend to make focaccia, and follow a standard recipe for doing so, but precisely how much flour she adds in at the first and second provings will depend on how the dough is taking shape. If the dough seems particularly porous, or if it’s having a difficult time rising, she may add less oil when she moves to the final bake in order to prevent it from sagging— and this requires intellectual creativity. But note that, according to WAM, there is no space for flexibility between recipe and dish. The world-recipes are so precise that God makes no creative contributions in bringing one of them about. God functions as a printer for pre-existing books. The minimal role that WAM assigns to God not only clashes with religious practice, it also raises worries for two common theological views. First, if God’s role in creation is mere actualization, this threatens to undermine traditional arguments from design. According to design arguments, the intricacies of nature imply the world was designed by an

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

130  .  intelligent creator. However, on WAM, this “creator” doesn’t really make creative contributions to the design of the world. Perhaps the force of such design arguments can be recovered within WAM, but there is at least an initial tension between the method of creation invoked in these two models. Moreover, some theists think this world not only points to a designer but reveals the nature of that designer in the same way an artist’s artwork reveals something about her. In order to view the Book of Nature as an additional source of divine revelation, it is critical that we understand God to have made specific creative contributions to its design. But on WAM, it appears this sort of contribution is eradicated. Perhaps we learn about God’s nature from which world God chose to actualize. But if God makes the choice to actualize based on overall aggregate goods, it will be difficult, from inside of the world, to extract what those goods are. Moreover, it will be unclear whether any particular part of the world was a reason why God chose to actualize or just an unintended but foreseen consequence of actualization. It seems, then, that divine creativity plays an important role in both religious practice and natural theology. Is there a way it can be reconciled with WAM?

3.2 Possible Dissolutions One way to salvage divine creativity is to redefine it. I have shown that WAM is inconsistent with Boden’s creative categories, but it’s possible these categories fail to capture the type of creativity God exemplifies. Perhaps there is another sort of creative contribution that Boden and I have left out. Although this is possible, I don’t know of any actual definitions that will do the trick. All definitions of creativity appeal to originality, and that aspect of creativity is precisely what WAM seems to exclude. Therefore, if we want to maintain that God is creative, our best bet is to modify WAM. But how can we massage WAM in order to make space for divine creativity? One option is to argue that while God does not combine any worldcomponents to make a world, God does choose a best world and in this

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

131

sense combines states of affairs with value. However, to say that God bestows worlds with this sort of value suggests that the value of worlds is subjective—which is contrary to the central tenets of WAM. As Klaas Klaay writes, Discussions of God’s choice of a world typically assume that worlds have axiological status, and that they can (at least in principle) be evaluated: some are good, others are bad; some are better, others are worse . . . it seems plausible to suppose that, if actual, they can properly be said to bear world-good-making properties (hereafter WGMPs). These are properties which, ceteris paribus, tend to make worlds good. Similarly, it is reasonable to suppose that there are world-badmaking properties (hereafter WBMPs). On this account, the axiological status of a world can be understood to depend upon which WGMPs and WBMPs are instantiated in the world, and the degree to which they are exemplified. (Kraay, 2008, 855)

According to WAM, the content of the worlds fixes an objective ranking. Of course, there have been a variety of discussions over whether these rankings are well ordered, etc., but the basic assumption is that the worlds themselves bear good and bad properties, thereby grounding an objective understanding of goodness. Perhaps we could affirm that God surveys the set of possible worlds but deny that there is a ranking or ordering of worlds (or, if such a ranking does exist, it is superfluous to God’s choice).⁴ When God selects a world to actualize, they combine a state of affairs with their being in order to make something of value. Here, the artistry of God is in the “eyes,” so to speak. God is like a composer who breathes into a sequence ⁴ Alex Pruss argues for something like this in his paper “Divine Creative Freedom” (Pruss, 2016). He suggests that various possible worlds may be incommensurable, and God can choose between them for artistic reasons. However, Pruss does not deny that the worlds have goodmaking properties, he merely denies that there is an objective measure of “goodness” which selects a particular world. Instead he allows for multiple goods which may be in tension with one another. Additionally, while Pruss appeals to “artistic choice” in his argument, his sense of what that means is quite different than what I’ve defined as creativity. Pruss suggests that we can have aesthetic values that are not based in moral values that can ground contrastive choices between worlds. I am making the much stronger argument that modeling creation as a selection between possible worlds eviscerates genuine creativity.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

132  .  of notes and makes them a melody. This change happens when God bestows a state of affairs with meaning—that state of affairs particularly reflects their nature and essence, it reveals to creatures who they are. This type of value doesn’t depend merely on objective good making properties but on what the creator sees in a state of affairs. The tricky part of this response is it depicts God as outside of the set of possible worlds but, as Kraay points out, this is a mistake. “Since the possible worlds there are exhaust the way things could be, there simply is no vantage point, divine or otherwise, outside this set” (Kraay, 2008, 856). Possible worlds include facts about meaning, as they are complete descriptions of the world. Therefore, God couldn’t somehow project meaning and purpose onto a world-story because these are contained within the worlds themselves. However, another alternative is to defend that combining states of affairs with God’s being is itself creative on Boden’s account. Here’s an analogy: we might claim that God’s knowledge of every possible world is similar to knowing every potential chess move. But there is a distinction between knowing a potential move and playing it, and when God actualizes a particular world they play a chess move. Shawn Bawulski and James Watkins defend a version of this view (Bawulski and Watkins, 2012). They argue that WAM provides a way of reconciling classical theism with divine creativity. Although they appeal to Boden’s work on creativity, they focus on her claim that creativity involves bringing about something “new” without further exploring her three categories. Because the actual world is something new, they argue it constitutes a case of genuine creativity. While this satisfies part of Boden’s notion of creativity, it misses her more general point. The reason we ascribe a certain kind of “creativity” to players who use new chess moves, or use moves they have never seen used, is that they uncover or recognize some novel way to exploit the space they are working in. That sort of discovery is absent in this case. We can reintroduce an element of discovery by requiring that the world God actualizes is indeterministic, and denying that God has foreknowledge of how it will evolve. If this is the case, God’s creative act displays a kind of exploratory creativity because it illuminates a previously unknown possibility. However, while this allows God to make a creative contribution to the world, it’s unclear whether that

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

133

contribution is substantial enough to reconcile WAM with nature as designed or revelatory. If the only creative choice God makes is to add a bit of randomness, it’s still unclear how this depicts God as the intelligent designer presumed in teleological arguments for the existence of God or nature as revelatory. One last option is to drastically modify WAM in a way that introduces divine transformative creativity. In the previous section, I pointed out that transformative creativity, as it is instantiated in humans, is a relativized concept. Artists transform conceptual spaces by identifying possibilities in an alternative conceptual space, and moving from their current space to a novel one. I further argued that, given this sort of relativization, it’s impossible to ascribe this sort of creativity to the God of WAM whose conceptual space is unlimited. But that argument requires maintaining that transformative creativity is relativized. Another way of thinking about divine transformative creativity is that it’s like human transformative creativity insofar as it involves restructuring a conceptual space but unlike human creativity in that the novelty generated is not relativized. Divine transformative creativity involves the introduction of new possibilities full stop. This is similar to the conceptual move made in developing an account of creation ex nihilo. When humans make artifacts, they always make them from something else. A painter makes a painting from paints, a brush, and canvas. A sculptor turns bronze into a statue. We know what it means to create by altering previously existing physical substances. So what does it mean to create ex nihilo? Well, it’s like what humans do when they make an artifact, except it doesn’t require any previous physical material. Divine transformative creativity extends this doctrine from matter to concepts—not only did God create matter where there was no matter, God invented the concept of matter. God creates genuinely new states of affairs.⁵ ⁵ McFarland also claims the theological doctrine of ex nihilo demands that there be no conceptual or physical predecessor to creation, and, coincidentally, draws similar culinary connections. “The creation of the world from nothing is not a matter of God actualizing one possibility out of many, in the way that I might choose to make chocolate rather than carrot or lemon cake . . . conceiving creation from nothing as choosing among possible worlds constitutes a logical mistake. God does not actualize one possibility among others; rather, God makes it that there can be such a thing as the actualization of possibilities” (McFarland, 2014, 187).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

134  .  While any picture that suggests God creates states of affairs will significantly diverge from Plantinga’s description of WAM, we may be able to reconcile the two by supposing that God first “writes” all of the world-books, so to speak, and then selects one of them.⁶ In the next section, I’ll turn my attention to this suggestion. After all, Boden’s third form of creativity suggests the creation of possibilities that did not previously exist. Perhaps the solution to this puzzle is that God designed the framework of possibility.

4. Transformation In §2, I combined Margaret Boden’s account of creative processes with WAM to reveal a tension between divine creativity and the preferred philosophical model of creation. In §3 I explored the space, concluding that WAM is problematic because it rules out the possibility of God making original creative contributions to the structure of the world. However, I suggested this problem might be resolved if God is the origin of possibility space; perhaps on a view where God exhibits transformative creativity, creating the possibility space from which they select a world to actualize, the problems divine creativity raises for WAM disappear. In this section, I examine one view according to which God creates possibility space, and argue it fails to resolve the tension. I then sketch out three desiderata for any model of creation that maintains a robust picture of divine creativity.

4.1 Creating Possibilities Transformative creativity, on Boden’s view, requires the creation of possibilities that did not previously exist. In the current context, this suggests God must somehow create possibility space itself—the recipe

⁶ I speak of these things as if they occurred in time, but we may be able to massage time out of the picture by appealing only to logical orderings.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

135

book, so to speak. But what does it mean to say that God creates possible worlds?⁷ In their paper “Absolute Creation,” Chris Menzel and Tom Morris argue that God is creator not only of the actual world but of the entire platonic realm. For Menzel and Morris, the worry is not that an actualizing God fails to make creative contributions, but that the doctrine of creatio ex nihilo implies “that the realm of necessity as well as that of contingency is within the province of divine creation” (Morris and Menzel, 1986, 353). If God truly creates from nothing, then all things, including maximal and complete states of affairs, must depend on God for their existence. Menzel and Morris flesh out their view by claiming that all abstract objects are the product of divine intellectual activity. Unlike human concepts, then, which are graspings of properties that exist ontologically distinct from and independent of those graspings, divine concepts are those very properties themselves; and unlike what is assumed in standard platonism, those properties are not ontologically independent, but rather depend on certain divine activities. We think this view can be extended to the rest of the traditional platonic domain. All necessarily existent propositions, for example, can be thought of as “built up” out of properties. Thus, in the way in which we characterize properties as God’s concepts, we can characterize propositions as God’s thoughts. So, for example, the proposition that red is a color can be construed as the content of God’s thinking: Red exemplifies being a color. So the existence of propositions as well derives from an efficacious divine conceiving. (Morris and Menzel, 1986, 355)

According to Menzel and Morris, in the same way that we, by an intellectual process, come to clearly grasp the concept of a triangle, ⁷ Whether or not God creates possibility space, and how that might happen, has received a good bit of discussion. See, for example, Leftow (2012), Craig (2016), or Plantinga (1980). It’s possible that some other description of how God generates possibility space may ultimately square better with divine creativity than the theistic activism I describe here. To explore all of these views on created necessity is outside the scope of this chapter. However, I found the picture given by Menzel and Morris to be a useful heuristic for drawing out what must be included in a model of creation in order for it to exemplify divine creativity.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

136  .  God, by an intellectual process, generates the concept itself. They go on to equate such concepts with properties, which populate God’s thoughts to form propositions. On this view, which they call theistic activism, the truth values of all propositions are also determined by processes in the divine intellect. Propositions that are contingently true (e.g. “some peppers are red”) are a part of God’s “beliefs” while contingently false propositions are not. Necessary truths are true in virtue of the divine process by which they are conceived; they could not be disbelieved by God. Possible worlds are sets of these propositions; the set of possible worlds models the modal propositions produced by the divine intellect. Menzel and Morris don’t discuss actualization, or how God “creates” the world. However, I will use their view as a springboard for what I call WAM+. On WAM+, God creates the set of possible worlds as described in theistic activism, and then chooses one of those worlds to actualize on the basis of the world-good-making and world-bad-making properties described by Kraay. Leaving aside the details of WAM+ and whether or not it is palatable, the view suggests God exhibits Bodenesque creativity. God displays transformative creativity by generating entirely new concepts, uses combination to make properties into propositions, and explores the space by populating it with every possible state of affairs. It appears the God of WAM+ is a creator rather than a maker, and the force of the creator‒ creation analogy often supposed in religious practice and theology is recovered. Or so we might hope. Unfortunately, the picture of creation given by WAM+ remains difficult to reconcile with a traditional understanding of creation.⁸ Although God generates and combines concepts, these processes occur independently of the intentions typically associated with creating. Consider, for example, the way intention is ascribed to God in the first creation myth of Genesis.

⁸ It’s worth pointing out that by “traditional understanding” of creation, I don’t intend a literal understanding of Genesis. As I explain in the discussion of the following passage, the real issue is that WAM+ does not allow for God to act with intentionality as is suggested in Genesis. This is problematic even without a literal reading of Genesis.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

137

And God said, “Let there be lights in the expanse of the heavens to separate the day from the night. And let them be for signs and for seasons, and for days and years, and let them be lights in the expanse of the heavens to give light upon the earth.” And it was so. And God made the two great lights—the greater light to rule the day and the lesser light to rule the night—and the stars. And God set them in the expanse of the heavens to give light on the earth, to rule over the day and over the night, and to separate the light from the darkness. And God saw that it was good. (Genesis 1:14‒18, ESV)

Genesis suggests a God who creates the world with purpose. The lights in the heavens are created to be signs, to mark the seasons, and to give light. That the world was created with intention has often served as the basis for teleological pictures of nature. But if WAM+ is correct, God does not create parts of the world (e.g. the light) for these sorts of purposes. The worlds are collections of consistent propositions; the lights in the heavens mark the seasons in this world for the same reason that other worlds fail to have seasons altogether—because each of those worlds exemplifies some part of possibility space. In other words, WAM+ rules out God exhibiting the sort of purposiveness in the creation of this world posited by most theistic traditions. This purposiveness connects to what Boden calls “value.” As you may recall, her definition of creativity involves “the generation of novel, surprising, and valuable ideas” (emphasis mine). Creativity isn’t just about making something new, but making something that is both novel and valuable. We get a glimpse into what this means with respect to creative combination in the case of the pun. A pun isn’t creative just because it combines two different meanings of a word—that might happen in a misunderstanding. To create a pun one combines two different meanings of a word for the purpose of comedic value. Even in the case of transformative creativity, where “new possibilities” are created, these possibilities serve some purpose. Wave-particle duality suggests new ways of understanding confusing physical phenomena. Pollock created a new way of interacting with paint and canvas in order to express meaning. These new possibility spaces are created as a way of accomplishing a particular goal or advancing a general purpose.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

138  .  But WAM+ makes no mention of this sort of value. Although theistic activism suggests God “combines” properties to make propositions, it fails to posit any reasons why God engages in this activity. Unlike the pun case, where two meanings are combined for the purpose of making a joke, all WAM+ tells us is that properties are combined. The same is true of the other types of creativity the God of WAM+ allegedly displays; while God creates a new possibility space through the generation of concepts and explores this space by populating it with possible worlds, theistic activism fails to connect these activities to any creative purpose. This is not intended as a criticism of theistic activism—which is a view intended to solve a different problem—but reveals that in order for God to fully satisfy Boden’s definition of creativity, God must act with intention. Perhaps this problem can be solved simply enough if we include values in the creative process described in WAM+. Suppose God creates possibility space in order to make a truly excellent world. God generates properties, combines them into propositions, and then explores different possibilities by recombining the propositions to represent every logically consistent world. God then generates a ranking of these worlds according to the objective good and bad properties they bear, and chooses the best for actualization. On this view, I’ll call it WAM++, God fully displays the purposiveness required for Bodenesque creativity. Although the God of WAM++ is creative in Boden’s sense—God creates possibility space for the purpose of making a world—it’s still unclear whether this purposiveness translates to world-specific intentions like those described in Genesis. If WAM++ is true, and we are living in the world God actualized, it’s still true that the lights weren’t placed in the heavens in order to be signs or order the seasons or shine on the Earth: the lights are in the heavens because that was one logically consistent possibility that needed to be explored. All worlds—including this one—obtain their structure because they exemplify one of many possible orderings of propositions. We might try to recover the relevant kind of purpose by appealing to God’s choice in actualizing. Even if the world wasn’t ordered according to world-specific intentions, perhaps it was chosen for them. When surveying the worlds, God was pleased that in this one the lights in the expanse of the heavens separate the day from the night, so they decided to actualize it.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

139

But this just returns us to the problem discussed in §3.2. Recall that, according to the decision procedure outlined in WAM, worlds are ranked according to their aggregate goodness (and badness), which is determined by objective properties, and not ordered by God’s preferences. We might try to collapse this distinction and say that the “objective-good-making-properties” Kraay describes are identical to God’s preferences; if God is pleased by some property x (i.e. the lights separating the day from the night), then x is a good-making property. On this view, the best world will also be the world that includes the greatest number of things that God prefers. But even if the best world is the world that contains the greatest number of God’s preferences, God still does not choose on the basis of any specific design feature(s), but according to the aggregate good of the world. This implies there may be specific features of other worlds that God prefers to specific features of this one; perhaps when it comes to skies, God prefers green to blue. However, if it turns out that the overall best world has a blue sky, God would choose to actualize a blue sky over a pink one. The worry, then, is that even if WAM++ recovers world-specific creative intentions, which I think is unclear, it generates an epistemic dilemma for those who claim that God reveals divine nature in creation. On WAM++, some parts of creation reveal God’s preferences, and other parts of the world are only here as a consequence of consistency. However, there is no principled way for perceivers to determine which elements of creation God considers good and which are merely accidental. From our point of view, it will be perpetually unclear whether any actual feature of the world is one that reflects divine preference and nature or is just an unintended but foreseen consequence of the best overall world. We might spend our whole life praising God for the beauty of sunsets only to have it later revealed that they strike the divine eye as a bit gaudy. This makes learning about God from the created order particularly problematic.⁹

⁹ There is a temptation to say we know which parts of this world are of God—those that reveal God’s glory. However, to say this is just to presuppose that we already know what God’s nature is like from some other epistemic ability. If that’s the case, it’s still true that we don’t learn about God from the natural world. I suspect some will argue that we know what God is like

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

140  .  An additional worry for WAM++ remains for proponents of design arguments. Classical design arguments, like Paley’s watchmaker, claim the complexities of nature suggest this world was designed to be just so, rather than accidentally arranged by natural forces. But on WAM++, the structure of the world is generated by a consistency algorithm. Of course, you might think that this kind of algorithm requires a complex mind— only such a mind could arrange a consistent world. But if each possible world is composed only in virtue of consistency, as WAM++ suggests, then the operations of the divine mind do not seem altogether different than natural forces. The algorithm by which the worlds are generated is, at the very least, describable by something structurally similar to “natural laws.” Interestingly, the force and degree of both of these worries may be related to questions about what sorts of things count as basic properties. Following theistic activism, I have said little about how God generates properties and what sorts of properties are generated. But what properties are basic will give us some insight into what sort of algorithm might be used to generate possible worlds. For example, if the basic properties of worlds are particles, and various possibilities represent different arrangements of those particles, then the “arranging” of these particles by some kind of algorithm is very similar to any picture on which the world evolves by natural forces. On the other hand, if we take particular sunsets to be basic properties, we might recover interesting world-level intention (e.g. God put together a particular sunset for creative reasons in the process of generating the property), but it seems unlikely that anyone would be committed to the basicness of a sunset. In any event, I think how one fleshes out the manner by which God generates properties plays an important role in thinking about God’s creative intentions.

already, and so what we learn from creation is that God in fact exists and created the world. The short response to this is that it’s just a different view, and therefore may not be subject to the problems I raise for divine revelation. Additionally, this alternate view is vulnerable to another problem: from the standpoint of the perceiver, we might constantly receive evidence that nature is contrary to God, as things will happen that are contrary to God’s nature. Of course, for philosophers like Plantinga, this is precisely our situation and why the problem of evil occurs. But if God’s intentions in creating the world are to confirm divine nature rather than reveal it, and the world that is created isn’t one that obviously confirms that nature for the sorts of agents God creates, this starts to look like a serious design flaw.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

141

To summarize, then, the assumption that God creates possibility space is not sufficient for recovering divine creativity. Additionally, even if we recover some level of divine creativity in an account, that alone may not be sufficient support for the type of creativity assumed in natural theological arguments.

4.2 Morals I began with the question of whether the world-actualization model of creation describes a God that is genuinely creative. I argued that if God creates by actualizing, then God fails to make intellectual contributions to the design of the world required for a creative creator. I then attempted to recover divine creativity by assuming God not only actualizes a world but generates the entire space of possible worlds from which the actual world is chosen. Interestingly, this assumption alone is not enough to recover the type of divine creativity central to several common philosophical positions. Of course, this by no means proves that WAM and divine creativity cannot be reconciled; I have only considered one rather sketchy view of how God might create possibility space. Nevertheless, it does suggest that maintaining a robust sense of divine creativity within models where God creates by actualization is no simple task, and one to which philosophers should pay more attention. Moreover, I think this particular exploration of the tension between WAM and divine creativity suggests at least a few desiderata that a model of creation need satisfy in order to maintain a robust picture of divine creativity. The first desideratum for creativity in creation is originality; at least some part of the world’s design must originate in God. Using Boden’s taxonomy of creative thought processes, I sketched several ways in which this might occur. However, if the world-books are, as Plantinga states, not something God writes, it is unclear how mere actualization leaves God the type of creative space required for originality. The straightforward solution to this is to say that all possible worlds originate in God. However, that the worlds originate in God is not enough to recover creativity—we also need to know something about how they originate. Boden’s account shows they must be generated for a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

142  .  purpose. A purely random generation of possible worlds does not exemplify creativity, as creativity is the generation of something new and valuable. Let’s call this second desideratum purpose. The final desideratum is that God maintains world-specific intentions. Religious practice, arguments from design, and the belief that the divine nature is revealed in creation all suggest the God made the world as it is for reasons specific to it. These creative intentions are precisely what natural theology takes to be revealed in creation. Because WAM suggests that God chooses this world for aggregate reasons, which are based on comparisons across different possible worlds rather than specific features of the actual world, it is hard to reconcile actualization with worldspecific intentions. An example may be helpful to highlight the importance of worldspecific intentions. As I mentioned in the beginning of the chapter, theists often claim the world is related to God in the way an artist is related to an artifact she creates. This undergirds idea that we can learn about God from creation in the same way we learn about an artist from her work. For example, just as we take the Heydar Aliyev Center to display the brilliant mind of Zaha Hadid, creation displays the mind of God. Ordinary artifact to artist inferences—for example every element of the Heydar Aliyev Center reflects Hadid’s love of fluidity—imply that Hadid specifically intended (and constructed) the fluid elements of the building. Therefore, if we want to make these same sort of inferences about God from nature— for example the harmonic planetary motions reflect the divine love of order—it is crucial that God intended these design elements.¹⁰

¹⁰ Of course, one could move in the opposite direction and give up world-specific intentionality altogether. This approach does help with the problem of evil. If any adjustment to this world turns out to make it a worse world overall, then we don’t need explanations for particular cases of evil. In other words, the answer to any “why” question about a particular evil turns out to be “because it was a necessary part of the best overall world.” I suspect this powerful response to the problem of evil is one reason WAM has become so popular. In fact, this forceful strategy is found in Leibniz, and Plantinga introduces God’s limited choice in actualizing as part of his response to Mackie’s concerns about the problem of evil. But what this chapter points out is that constraining God to this sort of aggregate choice about the features of the world is at tension with the creator-God that underlies parts pf natural theology like the argument from design. Moreover, if we lose world-specific intentionality altogether, it will cause serious issues for many specific tenets of specific religious traditions like Christianity. What does it mean to say that Jesus died for humanity because of God’s love for the world if God lacks world-specific intentions?

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

143

One very important and related issue that I have left to the side in this chapter is the role of human agents in the structuring of the world. Of course the question of God’s creative role in the world will become increasingly complex if we take humans to also make creative contributions to why things are the way they are. In order to screen some of this off, I have primarily focused on God’s creative role in the world’s origins. I think any theist—open, classical, etc.—who is committed to creatio ex nihilo will agree that God creates at least the starting point for the actual world. Any account of creation that includes a creative God will need to show that what God created was an original design made for a purpose. To maintain our traditional understanding of a creator God, those purposes will also need to be specific to this world, rather than generic (e.g. to make the best world possible).

5. Conclusion In this chapter, I have considered whether we can reconcile divine intellectual creativity with the world-actualization model. I have relied on Boden’s three categories of creativity to frame this discussion. I combined Boden’s definition of creativity with WAM to generate a puzzle. I explored the conceptual space surrounding the puzzle to see how we might escape it. I transformed WAM into a model on which God creates possibility space, to explore whether this assumption is sufficient for recovering divine creativity, and concluded it is not. Finally, I sketched three desiderata to help guide the construction of a model of creation that leaves space for a creative God. The tension between WAM and divine creativity appears to be quite deep. In order to find the best possible way of modeling creation, we might need to start from scratch.¹¹

¹¹ This chapter would not have been actualized without the unending advice and support of Allison Krile Thornton, my sous-chef in philosophy and life. Additional thanks to Molly Brown, Kyla Ebels-Duggan, Elizabeth Jackson, Chris Menzel, and Fr Philip Neri Reese for their creative contributions. I am also grateful to audiences at the Notre Dame Center for Philosophy of Religion reading group, the Ryerson analytic philosophy of religion work-in-progress group, the Practical Philosophy Workshop at University of Chicago, two anonymous referees, and the editors for helpful feedback. This chapter is dedicated to the memory of Brian Siken, who always encouraged me to believe in a creative God.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

144  . 

References Bawulski, S. and Watkins, J. (2012). Possible worlds and god’s creative process: How a classical doctrine of divine creation can understand divine creativity. Scottish Journal of Theology, 65(2): 174–91. Boden, M. A. (2010). Creativity and art: Three roads to surprise. Oxford University Press. Craig, W. L. (2016). God over all: Divine aseity and the challenge of Platonism. Oxford University Press. Kraay, K. J. (2008). Creation, actualization and god’s choice among possible worlds. Philosophy Compass, 3(4): 854–72. Leftow, B. (2012). God and necessity. Oxford University Press. Leibniz, G. W. (1989). The monadology. In Philosophical papers and letters, pages 643–53. Springer. McFarland, I. A. (2014). From nothing: A theology of creation. Westminster John Knox Press. Morris, T. V. and Menzel, C. (1986). Absolute creation. American Philosophical Quarterly, 23(4): 353–62. Plantinga, A. (1977). God, freedom, and evil. Wm. B. Eerdmans Publishing. Plantinga, A. (1980). Does god have a nature. Marquette University Press. Pruss, A. R. (2016). Divine creative freedom. Oxford Studies in Philosophy of Religion, 7: 213–38.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

5 Defining Religion Samuel Lebens

1. The Project: Where Angels Fear to Tread The history of religious studies is littered with failed definitions of ‘religion’. Past attempts have proven to be too narrow, too broad, too vague, or otherwise problematic. Accordingly, most religious studies scholars make do either with somewhat ad hoc and narrow definitions, for the technical purposes of specific empirical inquiries, or with familyresemblance definitions. Indeed, in the last serious attempt to offer a definition of religion in the history of analytic philosophy, William Alston (1967) opted for a familyresemblance account. The best he thought we could do was to list ‘religion-making characteristics’. None of these characteristics were deemed necessary nor sufficient for making something a religion; nor were the listed characteristics supposed to be jointly sufficient. This was, he thought, the best we could do. By contrast, I shall offer a definition of religion which, I suggest, provides us with the necessary and sufficient conditions for a set of ideas and/or practices to constitute a religion. Moreover, I suggest that the definition is extensionally and intensionally adequate, and that it captures what ordinary speakers mean when they talk about religion. Despite the proven ability of religious studies scholars to operate without an agreed upon definition of ‘religion’, it would still be desirable to find one. Empirical studies about religion can hope to be much less qualified and hedged, if only a precise, substantive, and accurate definition could be found for ‘religion’. It is just such a definition that I shall propose in this chapter. In the wake of such a definition, philosophical Samuel Lebens, Defining Religion In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Samuel Lebens 2022. DOI: 10.1093/oso/9780192862976.003.0005

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

146   discussions about the distinctive virtues and vices of religion can aim to be much more confident and general. Moreover, philosophy of religion—in recent years—has focused much more on theological problems, and discussion of evidence and warrant for theological beliefs, than it has focused on the nature, role, and value of religion itself. Perhaps having a definition of religion would help to remedy that state of affairs. Moreover, beyond the confines of the philosophy of religion narrowly conceived, a number of thorny issues in political philosophy might be rendered less thorny if we had a clear definition of what religion actually is. For example, scholars debate the extent to which it is acceptable, in a pluralistic society, for citizens and politicians to appeal to religious reasons in public discourse.¹ That is to say: is it appropriate to justify public policy in a pluralistic society on the basis of religious reasons? Similar debates take place regarding religious exemptions.² Can it be moral or appropriate for people to demand exemptions from legal requirements that bind other citizens, purely on religious grounds? Those who argue for and against such exemptions, and those who argue for and against the acceptability of religious reasons in public discourse, would surely stand to benefit from a clearer conception of what religion actually is. I am aware that the project of this chapter manifests a sort of audacity that constructive philosophy is known for, and that religious studies might sneer at. Surely only fools rush in where angels fear to tread. And yet, perhaps there’s a surprisingly simple reason why so many past attempts have failed. Indeed, I shall argue that in order to arrive at our hoped-for definition, we simply need to reverse the order of explanation commonly posited between religion and religiosity. Once we recognize that religiosity is explanatorily prior to religion, rather than the other way around, it becomes surprisingly easy to formulate a definition of ‘religion’ that seems to be immune to any obvious counter-example. In the end, criticisms of the definition I propose will be moot until critics find a clear counter-example, either in the form of a religion that

¹ See, for example: Greenawalt, 1988; Audi, 1990; Raz, 1990; Perry, 1991; Kymlicka, 1992; Quinn, 1995; and Habermas, 1995. ² See, for example: Leiter, 2013; and Vallier, 2016.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

147

isn’t covered by the definition, or in the form of something that clearly isn’t a religion that is covered by the definition. Ultimately, I leave that ball in the court of would-be critics, and hope that the definition I propose will prove itself fit for purpose.

2. Religiosity First Counter-examples seem to lie in wait for every candidate definition of ‘religion’; showing that the definition is either too narrow, or too broad. The existence of atheistic religions, for example, will rule out a definition of religion in terms of theology (such as James Martineau’s suggestion that (1888, p. 15), ‘Religion is the belief in an ever living God . . . ’).³ Some have tried to define religion in terms of morality and duty (Kant, 1999, p. 153). But these attempts are not sufficiently discerning. We should be careful not to confuse systems of ethics with religions. Others have tried to define religion in terms of the experience of the holy (Otto, 2010). But does every religion, and do only religions, have such a concept? Secular atheists report moments of awe and transcendence, and they use the language of sanctity to describe them (Wettstein, 2012, p. 33, documents such uses). Émile Durkheim defined religion as a ‘unified system of beliefs and practices relative to sacred things . . . which unite into one single moral community called a Church all those who adhere to them’ (Durkheim, 2008, p. 47). Durkheim therefore added the notions of community and practice to Kant’s morality, and to Otto’s sanctity. One wonders whether these additions are enough. One could imagine a club for art enthusiasts. They have an organized community. They have a regular practice, which is to visit museums and galleries together. They share certain values. They take art to be, in some sense, sacred. And they seek experiences of awe in the ‘chapel’ of the art museum. Is this club a religion, or yet another counter-example to yet another failed definition? ³ Having recognized that there can be atheistic religion, Alston (1967) goes on to offer what he seems to regard as an exhaustive taxonomy of religion in terms of where they locate divinity (in various sorts of objects, in certain words or literature, or in mystical experience). This is surprising since Alston surely recognizes that not every religion has a conception of divinity at all.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

148   Notice that all of these definitions seek to define religion independently of religiosity. Presumably, all of these definitions would relate to religion as explanatorily prior to religiosity. First they would tell us what religion is, and then they could define religiosity in terms of some sort of allegiance to religion. John Dewey (2013, p. 25) and Paul Tillich (1963, p. 4) focus upon religiosity instead of religion. They define ‘religiosity’ in terms of its affective qualities, and how it plays a central role in organizing one’s activities and concerns. But this approach doesn’t give us an obvious route back to a definition of religion. A commune of Marxists might say that their political ideology has affective qualities, and that it plays a central role in organizing their activities and concerns. We do recognize that people can dedicate themselves with a religious zeal to things that are not religions. Religiosity and religion can come apart.⁴ But if you can be a religious Marxist as easily as you can be a religious Hindu, then what is it that makes Marxism an ideology, and Hinduism a religion? Dewey’s (2013, p. 1) suggestion is that ‘religion’ in its classic sense has something to do with belief in the supernatural. But not every belief in the supernatural is religious, and it’s not clear that every religion would accept that they do believe in the supernatural, nor is it obvious how, exactly, to draw the distinction between the natural and the supernatural (see Lebens, 2020, pp. 226‒8). And yet, I shall argue that what’s distinctive about religions is that they call for religious devotion. Marxism and football fandom can give rise to religiosity, but, I shall argue, they don’t call for it. Religion does. If that’s right, then once we have a clear conception of what religiosity is, the following definition of ‘religion’ should suffice, without admitting of counter-examples: Religion: A religion is a system of thought and/or practice that calls for religiosity from its adherents.

⁴ Indeed, this was a central contention of Dewey’s definition of religiosity (Dewey, 2013).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

149

Now all we need is a definition of religiosity that’s precise enough to power this definition of ‘religion’, and to close the door on possible counter-examples. That is the task of this chapter. First, a disclaimer: although I deny that ‘religion’ is a family-resemblance term, since I’m in the business of providing a definition, I do however recognize that religiosity can come in degrees. It’s not always a black and white issue as to whether a person is religious or not. Accordingly, the criteria I lay out for living a religious lifestyle are only intended to define a ‘norm-kind’ for religiosity. Religious lifestyles are only religious to the extent and degree that they approximate the norm-kind.⁵ A religion, I shall argue, is a system of thought and practice that calls upon its adherents to do their best to approximate that norm-kind.

3. Religiosity In this section, I present a three-part definition of ‘religiosity’.⁶

3.1. Community There is a cognitive component to religiosity, and we shall get to it, in time. But it’s important to note that religiosity has a communal or social aspect. Indeed, religiosity often has more to do with community than it has to do with belief. Even when a religion formulates a binding catechism, the socio-historical act of forming that catechism often has more to do with belonging than it has to do with content. For some religions, what I’ve said about belonging and belief in the previous paragraph is uncontroversial. Judaism is a good example. Witness the classical rabbinic text that comes closest to approximating a catechism; the Mishna in tractate Sanhedrin 10:1: ⁵ Thanks to Terence Cuneo for this suggestion. ⁶ I shall be using ‘religiosity’ and ‘religious’ almost interchangeably. Religiosity is a property instantiated by lifestyles (among other things). A person is religious to the extent that her lifestyle exhibits religiosity. It is in terms of these personal properties that I hope to define what a religion is.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

150   These [Jews] have no share in the World to Come: One who says that the resurrection of the dead is not from the Torah, or that the Torah is not from Heaven, and an Apikoros . . .

The text isn’t explicitly concerned with what people think but only with what they say and do. Denial of the doctrine of resurrection, and denial of the divinity of the Oral Torah, were the rallying cries of the Sadducees—a sectarian rival to Rabbinic Judaism. The Apikoros is understood, by the Talmudic gloss on this Mishna, to be a person who disrespects the rabbis.⁷ In other words: it might not matter what you think, so long as you don’t say that the Oral Torah isn’t from heaven. If you go about saying that, then you’ll be adding to sectarian strife—you’ll be lending support to the Sadducees, or you’ll be undermining the rabbis. Indeed, the notion that these people have no share in the World to Come is a shorthand for saying that they’re not really Jews, since the Mishna begins with the assertion that ‘All of Israel have a share in the world to come.’ Sectarians do not belong. They must be excluded from the community. On these grounds, Menachem Kellner (2006) concludes that this Mishna is more concerned with sectarianism than catechism. But Judaism isn’t alone in this regard. Christian conciliar statements spend a great deal of time anathematizing certain views—a practice that isn’t merely about labelling a view as false; it’s also about shutting certain sorts of believers out of the community; saying to them that they don’t belong (unless they change their views). You might accuse me of trying to fit a Christian peg into a Jewish shaped hole, but Aquinas himself argues that holding a false belief can never be sufficient grounds for accusing a person of heresy. Rather, heresy occurs when a person publicly defies the authority of the Church (Summa Theologiæ, II-II, Q11, art. 2).⁸ So far so Catholic. Protestantism, by contrast, was built upon defiance in the face of the authority of ‘the Church’. Some Evangelicals claim to

⁷ It doesn’t matter whether this gloss is anachronistic. One way or another, it reveals a sociopolitical agenda behind Talmudic (if not Mishnaic) talk of apostasy. ⁸ Thanks to Simon Hewitt for pointing this out to me.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

151

abhor established religion. Is community belonging truly a part of their religiosity? I would argue that it is. Many Protestants do feel a close connection and loyalty to their particular denomination. Moreover, those devout Christians who see themselves as falling outside of any organized religion will nevertheless see themselves as part of a community of fellow travellers, fellow followers, or disciples of Jesus, etc. Even the founder of a new religion, who has no community to belong to, hopes to be the first link in a community that will extend on in time.⁹ Similarly, the last surviving member of a religious community still views herself as loyal to her communal forebears, even if she has no contemporaries in her community. Religiosity, it seems, has a lot to do with belonging to a community. Accordingly, the first criterion for living a religious life is: 1. A religious life is a life that is meant to be lived as a part of a community that defines itself around a system of ideas and/or practices. Remember that we’re trying to articulate a norm-kind. Accordingly, it’s not inappropriate to talk about how the life is meant to be lived. The last practitioner of a religion may lose hope of rebuilding a vibrant communal life. But she’ll recognize that her lifestyle wasn’t meant to be lived apart from a community.¹⁰ This first criterion follows directly from my contention that religion is, at root, a sociological phenomenon. To subscribe to a set of theological doctrines, but not to see yourself as a part of any community at all— not to see your fate as somehow bound up with the fate of your co-religionists—simply falls short of living what most of us would call a distinctively religious life. And, indeed, a person with no sense of communal belonging can have a theology and a set of rituals, but not a religion in any sociological, and therefore in any standard, sense of the word. A wise and spiritual person living on top of a mountain only becomes a Guru in the context of a community.

⁹ I will introduce (and rebut) a further possible counter-example to this claim later on, with a Robinson Crusoe type castaway. ¹⁰ Thanks to Daniel Howard-Snyder for discussion of this point.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

152  

3.2. Fundamentals of Faith Our first criterion entails that beliefs cannot be sufficient for fully-fledged religiosity. Now the question is, to what extent are beliefs a necessary component of a religious life? It’s clear and uncontroversial that religiosity and faith have an intimate relation. We sometimes describe religious people simply as people of faith. But then the question becomes: what is the relationship between faith and belief? A debate currently rages among philosophers about the nature of propositional faith (i.e., faith that some proposition is true). Some argue that it doesn’t require belief (McKaughan, 2013; Howard-Snyder, 2013). And some have argued that it does (Malcolm & Scott, 2017). I have argued that the whole debate misfires (Lebens, Forthcoming). To believe that p is to have a certain degree of confidence that p is true, but that degree of confidence is context-sensitive. Sometimes belief requires more confidence; sometimes less. Given the contextsensitivity of ‘belief ’, the question as to whether faith requires belief seems to be ill formed. It’s like asking whether somebody is tall, without providing a comparison class. What counts as tall in one context doesn’t count as tall in another. Similarly, the degree of confidence that counts as belief in one context doesn’t count as belief in another. Accordingly, we’d be wise not to define other attitudes in terms of belief and disbelief where possible. One thing seems clear: there are contexts in which it can be appropriate to say that a person has faith that p, without it being appropriate, in the same context, to say that that person believes that p. That is to say, there are some contexts in which belief that p demands more confidence than does faith that p. But all in all, when defining ‘faith’, we’re better off leaving belief talk behind. With this qualification in place, I think that Daniel Howard-Snyder (2013) has a compelling analysis of propositional faith. According to him, faith that p has four ingredients: (i) A positive evaluation of p (ii) A positive conative orientation towards p (iii) A positive cognitive attitude towards p

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

153

(iv) Resilience in the face of counter-evidence for p, and in the face of non-evidential challenges.¹¹ There’s something somehow inappropriate in saying that you have faith that p, when you realize that p isn’t the sort of thing that you should want to be true. This leads to Howard Snyder’s first ingredient. To afford p a positive evaluation is to think that p is the sort of thing that people should want to be true.¹² In his second ingredient of faith, Howard-Snyder gestures towards the difference between wanting something intrinsically and wanting something instrumentally. The mother in the midst of an agonizing cancer treatment may no longer care, in and of herself, whether she lives or dies, but she must at least have some relevant desire, perhaps the desire to be there for her children as they grow up, if we’re to make sense of the claim that she has faith that she’ll survive (Howard-Snyder, 2013). And thus, what Howard-Snyder means, when he requires ‘a positive conative orientation towards p’, is that faith that p demands that we either want p to be true intrinsically, or, as in the case of the suffering mother, we want its truth indirectly, or instrumentally. To have faith that p doesn’t merely require the thought that people should want p to be true. It also requires the person of faith to want it to be true themselves.¹³ When Howard-Snyder talks about a ‘positive cognitive attitude’ towards a proposition, he simply means that you have to have some confidence that p is true. How much confidence? I don’t think that’s a fair question, since, as with belief, the confidence threshold for faith will be context-sensitive. However, we can say that faith, in certain contexts, can survive less confidence than belief can normally survive. Sometimes you can have faith that p even if you only think that p is the least unlikely of the relevant options. Apparently, when T. S. Eliot was questioned about his Christianity, he replied that it was the least false-seeming of the

¹¹ The notion of resilience to non-evidential challenges was developed later than HowardSnyder 2013. See Howard-Snyder & McKaughan, forthcoming—especially pp. 14–15. ¹² In Lebens 2020, pp. 276–7, I discuss an objection to this ingredient of faith. I refer readers to the discussion there. ¹³ In Lebens 2020, p. 277, I discuss a potential counter-example to this ingredient of faith. I refer readers to the discussion there.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

154   options that was open to him (Howard-Snyder, 2013, p. 365). That degree of confidence was sufficient to give rise to faith. Finally, Howard-Snyder insists that faith has to be resilient in the face of both counter-evidence and in the face of non-evidential challenges.¹⁴ Resilience to counter-evidence, I would suggest, follows from the fact that, in many contexts, it will be true to attribute ‘faith that p’ to a person even when, in the very same context, it will be inappropriate to attribute ‘belief that p’ to the same person. And this is because, in a given context, faith will often demand a lower degree of confidence than belief. This alone will make faith more resilient, in general, to counter-evidence than is belief.¹⁵ Moreover, the positive evaluation and the positive conative attitude can probably explain why faith will be resilient in the face of non-evidential changes too.

¹⁴ See footnote 11. ¹⁵ In correspondence, Simon Hewitt raises two putative counter-examples to HowardSnyder’s account of faith: (a) Hewitt’s own belief that Fermat’s last theorem is true; (b) the devil’s belief that God exists. In neither case would we want to say that there’s faith, but Howard-Snyder’s definition, according to Hewitt’s objection, can exclude neither of them. The devil wants it to be true that God exists since, if God didn’t exist, neither would the devil! Accordingly, the devil has a positive conative attitude towards the proposition, via an instrumental desire, and thus can be said, counter-intuitively (and contra the New Testament, James 2:19) to have faith. To respond: I think it’s possible for people to have pro-attitudes (i.e. favourable attitudes) towards necessarily false propositions. And thus, even though it’s necessarily false that the devil could exist without God, the devil may still wish, per impossibile, that he could exist without God, and therefore, he might have an overall positive desire for God’s non-existence. Accordingly: he believes that God exists, but he doesn’t have faith. Regarding the first counter-example, Hewitt thinks that true mathematical propositions are the sort of propositions that people should believe. He therefore has a positive evaluation of Fermat’s last theorem. His belief in the theorem is also very resilient to counter-evidence, since the theorem has been firmly established by the mathematical community. But a positive evaluation of the theorem, in combination with resilience to counter-evidence, isn’t sufficient to generate faith. You also have to have a positive conative attitude towards the proposition in question. If it turns out that Andrew Wiles’s proof of the theorem is faulty, and if it turns out that the theorem itself is actually false, then what will Hewitt have lost—apart from a mistaken belief? It therefore makes sense to say that Wiles has faith that the theorem is true, but not that Hewitt has faith in its truth, unless he’d also be terribly disappointed by its disproof. In further correspondence, Hewitt said that, although the disproof of Fermat’s last theorem wouldn’t be much of a blow to him personally, he could think of other mathematical propositions in which he was more invested. For example, he would be disappointed if it turned out that there was actually a highest prime number, because the loss of the infinite structure of the sequence of prime numbers would be a loss of an elegance that Hewitt treasures. If that’s the case, then I’m very willing to say that Hewitt doesn’t just believe that the sequence is infinite, he hopes that it is; he has faith that it is. I don’t consider this a damaging counter-example to Howard-Snyder’s account. I think it’s simply an example of faith in a mathematical proposition.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

155

To commit to a religion will require faith in some set of propositions. Some religions have book length catechisms. Some have just a handful of fundamental principles. Sometimes a religion has no principles at all, and defines itself, instead, around a set of practices. Even so, to commit oneself to such practices will require faith in some set of propositions or other, even if two practitioners of the same religion can justify their identical practices via faith in different sets of propositions. So, whether a religion is orthodox or orthoprax, religiosity will still require propositional faith. A Catholic might think that their religiosity has been misrepresented. According to Catholicism, faith is an infused virtue, given by God. Faith might dispose you towards believing the content of revelation, but it is belief, and disbelief, rather than faith, that are going to be the difference between a heretic and a non-heretic. So how can I not write ‘belief ’ into my analysis of religiosity? And yet, I will stick to my guns. Saint Teresa of Calcutta (2007), in her darker, most anguished moments, prayed to God, confessed that she lacked belief, and asked for his help to be saved from doubts. Will the Catholic maintain that she was a heretic? All the while she wanted God to exist. She thought it plausible that he did. She certainly had what I’m calling ‘faith’. Perhaps she had belief too, under some contextually determined interpretation of the word ‘belief ’, but her confidence sometimes slipped below the levels that we’d generally pick out as relevant for belief. What matters is that she had what Howard-Snyder would recognize as propositional faith. In a religious context, you can call that belief, if you like. Or you can call it faith. But it’s what she had! We can now state our second essential ingredient of a religious life: 2. To live a religious life requires propositional faith directed towards the fundamental principles of the system of thought referred to in criterion 1 (or, at least to their conjunction¹⁶), and/or towards some set of propositions such that faith in them can warrant commitment to the practices referred to in criterion 1.

¹⁶ I add the parenthesis because you might desire that P&Q be true without desiring that one of the conjuncts be true without the other. Accordingly, faith in a conjunction won’t always distribute over the conjuncts.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

156   Marxism does call for its adherents to unite into something like a community. After all, the Communist Manifesto ends with its rallying call: ‘Workers of the world, unite!’ Moreover, Marxism seems to call upon its adherents to want certain things, and to hope for them, to believe that certain things are possible, to have faith in certain principles. If we define religiosity in terms of our first two criteria, and if we define a religion as a system of thought that calls for religiosity, then Marxism will be defined—erroneously—as a religion, even as the Marxist would insist that religion is the opiate of the people. But in fact, religiosity includes more than just community membership and faith. It requires that one engage one’s imagination. This is the third criterion.

3.3. Imaginative Engagement Religions tend to demand that their followers imaginatively engage with a particular set of narratives; a narrative canon. To engage with a narrative is, first and foremost, to engage one’s imagination. Whether we’re dealing with a fictional narrative, or a non-fictional narrative, if it’s written as a narrative, then we engage our mind’s eye. We imagine the scenes described unfolding, as if we’re watching them. Neurological research suggests that we use the same regions of our brain in witnessing an event of type X, as we do when we process a mere narrative about an event of type X (Oatley, 2008; Marr, 2011; Young & Saver, 2001). To read or listen to a narrative is to engage in a sort of offline mental simulation of witnessing the events described. I can’t argue that religiosity per se requires narrative engagement, since not all religions have a narrative canon. Admittedly, some forms of Buddhism revolve around stories about the Buddhas, but Zen Buddhists, despite their own body of legends and stories, seem to think that such stories are something of a distraction from the endeavour of enlightenment.¹⁷ Quakerism, despite its roots in Christianity, today

¹⁷ Hence the phrase, attributed to the Zen Master Linji Yixuan, ‘If you meet the Bhudda on the road, kill him.’ On the road to your own personal enlightenment, engaging with stories about the Bhudda will only be a distraction. We each have our own path to take.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

157

eschews any particular canon of narratives. What I can say, is that every religion demands some form of imaginative exercise. In order to develop my claim, we need to distinguish between a number of forms of imaginative engagement. As you imagine the events of a narrative unfolding, you’re not necessarily projecting yourself into them. Imagining events in this somewhat detached fashion, as unfolding without you, is what Peter Alward (2006) calls de dicto imaginative engagement. Sometimes a narrative calls for de re imaginative engagement when, for example, you’re invited to imagine, concerning an actual location in London, namely Baker Street, that Sherlock Holmes once lived there. Sometimes, de se imaginative engagement is what’s called for. When we play games of role-play or make-believe, for example, we thrust ourselves into the imaginative action. We imagine that we are soldiers, or superheroes, or butterflies, or what have you. Zen Buddhism, despite eschewing narrative, certainly seems to place a great weight upon acts of de se imagination. Certain elements of its meditative practice, known as zazen, could be characterized as a very minimalistic, and intentionally sparse, form of de se imaginative engagement: you are your breath.¹⁸

¹⁸ Hewitt, in correspondence, objects. Zazen cannot be de se imaginative engagement because there is nothing that it is like, for a human animal, to be breath. There are a number of different concerns that could be lurking under the surface here. You might think that there can be no such thing as a nonsensical proposition. And thus, you might think that the sentence ‘I am my breath’, as uttered by a human being, simply fails to express anything. Wittgenstein, for example, thought it impossible to have nonsensical, or illogical, thoughts. This was part of his critique of Russell’s theory of assertion (Wittgenstein, 1961, 5.422). On this view, there can be no propositional attitudes towards the proposition that a human being is a breath, for there can be no such proposition. ‘Read sympathetically,’ Hewitt suggests, ‘we should see this kind of locution as a poetic way of saying “focus on your breath to the exclusion of all else.” ’ I don’t feel the pressure of this objection because I think that certain sorts of ‘nonsensical’ sentences do express propositions, and thus, it is possible to have propositional attitudes towards certain sorts of ‘nonsense’—especially towards category mistakes, like the one at the heart of zazen. For arguments to bolster this claim, see Ofra Magidor (2013). In fact, Hewitt accepts that category mistakes can still express propositions. So perhaps his worry is this: even though my uttering, ‘I am my breath’, may express a proposition, that proposition isn’t one that I can imagine to be true, since I can’t visualize its truth, since there is nothing it would be like for it to be true! In response to this objection, I might point towards the work of Tamar Gendler (2000), who explores obviously and explicitly impossible stories that still serve as perfectly good stories, for which narrative uptake occurs. Surely we engage our imagination with such stories in some way or other.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

158   When thinking about de re and de se imaginative engagement, we can make some further distinctions. Sometimes we are invited to imagine ourselves, or something around us, in what can only be called, a literally true light. According to Rabbi Sampson Raphael Hirsch, for example, we are not simply commanded to believe that God exists, we have to view ourselves as living in a world in which God exists, and to see ourselves as his creations (see his commentary to the first of the Ten Commandments, in Exodus (Hirsch, 2009)). According to Judaism, we do live in a world in which God exists. We are God’s creatures. But that doesn’t mean that we automatically view ourselves as living in such a world, as his creatures. When you’re being asked to imagine yourself, or something around you, in a true light, I would call it ‘attentive-seeing-as’. This can be de se, as when the Jew tries to see herself as a creature of God, or de re, as when the Quakers ‘endeavor to see “that of God” in every person’ (Clarke et al., 2011). I call it attentive-seeing-as because you don’t believe that you’re making something up—instead you’re trying to attend to something that’s all too easily ignored. It’s as if you’re engaging your imagination in order to see the world more accurately, in accordance with what you believe, or in accordance with your faith. I call engagement in an act of attentive-seeing-as, adopting a perspective. Besides regular make-believe (which can be de dicto, de se, or de re), and attentive-seeing-as (which can only be de se or de re), there’s another type of seeing-as that is relevant to the religious life. I call it metaphoricalseeing-as. Elisabeth Camp (2009) helps us to distinguish between regular de se make-believe, and metaphorical-seeing-as. In order to make believe that she is Anna Karenina, she has to forget all about Elisabeth Camp, or at least to ignore everything that’s distinctive about herself. She ignores that she is a philosopher, and a professional woman in the twenty-first century, and tries, instead, to pretend, even to herself, that she’s a Russian aristocrat in the nineteenth century. She tries, in some sense, to lose

Alternatively, and perhaps relatedly, I could simply concede: yes, it’s true that nobody can successfully imagine being their own breath. This doesn’t mean that they can’t try! A child might try to jump to the moon. It isn’t possible for her to jump that high, but it won’t stop her from trying. More generally, it is possible for an agent to try to Φ, even when Φ-ing is impossible for them. Perhaps the Zen master merely tries to imagine that he is his breath—and this attempt, though it’s doomed to fail, might be thought to bring certain positive effects in its wake.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

159

herself. We’re talking here about a process akin to method acting. Compare this process with the metaphor that Elisabeth Camp is Anna Karenina. Processing this metaphor, she has to view herself through the prism of Anna Karenina. She says (Camp, 2009, pp. 112‒13): I might decide that Anna’s conflict between her love for her son and her love for Vronsky mirrors my own struggle to reconcile parental devotion and professional ambition . . . The overall result of this [metaphorical] matching process is a restructured understanding of myself, one which highlights, connects, and colors my Anna-like features while downplaying the rest.

Throughout this process, Camp cannot lose sight of herself, and her own life story, which she matches up with details of Anna Karenina’s story. Attentive-seeing-as harnesses the power of your imagination to help you see its object for what you already believe it to be. Metaphorical-seeing-as, by contrast, ‘helps you to reconstruct your understanding’ of its object. Similar to the distinction between attentive-seeing-as and metaphoricalseeing-as is Terence Cuneo’s distinction between playing a role and playing a target role. To play a role is consistent with losing yourself in an act of de se imagination. To play a target role, on the other hand, is to play a part ‘for the purpose of being that way, becoming like or identifying with that which one imitates’ (Cuneo, 2016, p. 78). To play Anna Karenina as a target role would be to act like her in order to become more like her or to identify with her. Cuneo argues that the adoption of target roles is commonplace in Christian (and especially Eastern Orthodox) liturgy. Target role-play is related to, but importantly distinct from, another notion that Cueno (2017) develops; that of aligning yourself with another person. Sometimes the religious life exhibits an attitude, say towards the saints, that is something more than veneration, but which can’t be called emulation or identification either, because emulation and identification might be, in certain cases, inappropriate. The attitude in question, Cuneo calls ‘alignment’. To align yourself with a person is in some sense to stand with them, and to share in their ideals and their goals. To stand with a person has nothing to with spatial proximity of course; the proximity is generated instead, symbolically, by an act of the imagination. Indeed,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

160   religiosity often seems to demand aligning yourself with a personal God, or with the saints, or with some person or other, even if those persons are beyond our emulation or identification. Howard Wettstein (2002; 2012) talks about signing on to an image. Take the image of God judging us on Rosh Hashona. What it means to sign on to that image, I take it, is to agree to structure your life through its prism, to engage your emotions with it, to make it your own, to choreograph your life with this image as part of your personal symbolic landscape. What religious people do, characteristically, is to engage in a very powerful and intimate way, with certain images at certain times; to sign on to them. This is where religiosity goes further than propositional forms of faith and manifests what is sometimes called ‘global faith’ (Audi, 2011, pp. 53, 57‒8), and faith as a venture; models of faith that require a person to organize and orient their lives in a certain way (Kvanvig, 2013; 2015). And thus we can now formulate our third and final essential ingredient of a religious life: 3. To live a religious life requires imaginative engagement (either via a species of make-belief, attentive or metaphorical seeing-as, target role playing, alignment, or in terms of signing on, depending on the context) with the canonical narratives, metaphors, prescribed games of makebelief, persons, and/or perspectives of the system of ideas and practices in question.¹⁹ As a shorthand, I will refer to all of these imaginative acts under the umbrella term of ‘signing on’. Religious life, at least as a norm-kind, is supposed to be absorbing. It is the imaginative component of religion that gives rise, most centrally, to this quality. It is one thing to believe in a religion; it is another thing to sign on. Signing on is what it means, to live one’s life in service of an ideal (to echo Kvanvig, 2013; 2015). Signing on engages the imagination. There is something defective about a religiosity

¹⁹ For extensive and fascinating discussion of the use of imagination in the religious life of Evangelical Americans, see Tanya Luhrmann (2012).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

161

that believes in a creed but fails to engage the imagination; that would be a faith without a full-blooded religious psychology. Some people might point to Yeshayahu Leibowitz as a counterexample to this third criterion, and perhaps to the second as well. He was a religious Jew. And yet his theology was remarkably austere. He thought that it wasn’t our place to investigate the reasons for the commandments, and that we should simply obey them (Leibowitz, 1992, pp. 14, 16). Moreover, his negative theology led him to think that substantive theology quickly descends into idolatry (Leibowitz, 1979, p. 129). Does this leave room in his religiosity for any cognitive content? And yet, Leibowitz must at least have had faith in some set of principles sufficient to justify, if only to himself, continued Jewish practice. And so, his religiosity must have satisfied our second criterion. And, despite the austerity of his religious vision, he was a man who engaged deeply with the narratives of the Bible, about which he wrote compelling homilies (Leibowitz, 2002). Indeed, he wrote (Leibowitz, 1992, p. 140): From the standpoint of religious faith, the Torah and the entirety of Holy Scripture must be conceived as a demand which transcends the range of human cognition—the demand to know God and serve Him—a demand conveyed in various forms of human expression: prescriptions, vision, poetry, prayer, thought, and narrative.

Poetry and narrative are among the vehicles for the message of the Torah. Accordingly, I wouldn’t accept that Leibowitz is a counterexample to my definition of religiosity. And if he were, then I would be willing to entertain the thought that, despite his intensity, his profundity, and his moral vision, perhaps he wasn’t a paradigm case of religiosity. Perhaps there is, indeed, something sufficiently austere about his religiosity to render it something of an outlier, at least in relation to the normkind. But that’s precisely why my definition of religiosity only seeks to delineate a norm-kind; a paradigm towards which real-world tokens of religiosity approximate in varying degrees. But this needn’t affect the precision of the forthcoming definition of ‘religion.’

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

162  

4. Religion Defined In the previous section, I argued that the norm-kind for religiosity can be defined in terms of the following three criteria for living a religious life: 1. A religious life is a life that is meant to be lived as a part of a community that defines itself around a system of ideas and/or practices. 2. To live a religious life requires propositional faith directed towards the fundamental principles of the system of thought referred to in criterion 1 (or, at least to their conjunction), and/or towards some set of propositions such that faith in them can warrant commitment to the practices referred to in criterion 1. 3. To live a religious life requires imaginative engagement (either via a species of make-belief, attentive or metaphorical seeing-as, target role playing, alignment or in terms of signing on, depending on the context) with the canonical narratives, metaphors, prescribed games of make-belief, persons, and/or perspectives of that system of ideas and/or practices. The notion of a person living a life that satisfies these three criteria provides us with a norm-kind for religiosity. A person is religious to the extent that her lifestyle approximates this norm-kind. And of course, this allows, as we should allow, that a person can commit oneself to something that isn’t a religion, such as Marxism, or even football fandom, in a distinctively religious way. Where past definitions of ‘religion’ have gone wrong, I claim, is that they haven’t recognized that religiosity is explanatorily prior to religion. With an understanding of religiosity already in place, however, we can define ‘religion’ as follows: Definition: For any system of ideas and/or practices R, R is a religion iff: i. R calls upon its adherents to live their lives as part of a community that defines itself around R, and ii. R calls upon its adherents to have propositional faith in the conjunction of some set (or other) of propositions, and

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

163

iii. R calls upon its adherents to engage their imagination with a set of canonical narratives, metaphors, persons, prescribed games of makebelief, and/or perspectives. A person can exhibit all of the elements of religiosity even in their football fandom. But to be a fan of a given team doesn’t require, or call for, all three elements in the way that a religion does. What does it mean for religion to call for religiosity? Since a religion is a set of ideas and/or practices, it can contain prescriptions. A religion prescribes religiosity of its adherents (and sometimes, in the case of proselytizing religions, it prescribes adherence of all people). That’s all it means for a religion to call for religiosity. A person can be a Marxist with truly religious zeal. Not only might they see themselves as members of the global proletariat, not only might they have faith in the fundamental principles of Marxist ideology, but they might even see the world, imaginatively, through the prism of Marxist symbols and metaphors. But Marxism isn’t a religion. Our definition can explain why. Marxism doesn’t call upon its adherents to engage their imagination in this distinctively religious way. You’re no less good a Marxist for not being interested in Marxist art, narratives, metaphors, or poetry. It is because Marxism doesn’t call upon Marxists to instantiate this third element of religiosity that Marxism isn’t a religion. Indeed, a Marxist might rightly point out that it is this imaginative element of religion that can make religion such a dangerous opiate to begin with. The cult of personality that has developed around the Kim dynasty in North Korea might be classified, by my definition, as a religion. It really does seem to call for every element of my definition of religiosity from the citizens of North Korea. But this definitional scheme seems to get things right. The official political ideology of North Korea has, to all intents and purposes, become a state religion. It’s certainly a cult-like and pernicious religion. But it’s a religion nonetheless. If accepting that fact is a consequence of my definition of ‘religion’, then it seems to be a strength of my definition rather than a weakness. But what about a desert island castaway—a Robinson Crusoe; can’t he found a religion, even once he’s lost hope of founding a community? Or

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

164   what about the last surviving human after a nuclear holocaust, can she not found a religion, even without hope of founding a community? Doesn’t the reliance of my definition upon community erroneously rule out such possibilities? No. I think that a Crusoe figure can found a religion, but only if he harbours the desire to found a community around the system of ideas and/or practices in question; only if that system calls for community. Only if the lifestyle in question is meant to be lived as part of a community. The fact that he knows that this desire and this calling is unlikely, or even impossible to satisfy, doesn’t, on my definition, undermine the fact that he founded a religion. But, what if Robinson Crusoe founded a system of beliefs and/or practices according to which other people are supposed to find their own paths? What if his system of beliefs and/or practices renounces any call for community building around it? Dean Zimmerman put it to me (in correspondence) that, if the tenets of Crusoe’s faith were rich, and if there was, for example, lots of prayer, and ritual, we’d still want to call it a religion, despite its renunciation of community; even if we could only call it a private religion. My definition, and its reliance upon the call to community, seems erroneously to rule out Crusoe’s communityrenouncing religion. The community-renouncing lifestyle that our Crusoe adopts certainly contains elements of religiosity. A person is religious to the extent to which his lifestyle approximates the norm-kind laid out in §3. This Crusoe’s lifestyle approximates two elements of that norm-kind to a very high degree, high enough—perhaps—for us to call him a religious person. You might even think that he’s founded a community of people who shun conformity. But if Crusoe’s lifestyle shuns community, and not merely conformity, then it’s going to be very hard to maintain that his lifestyle satisfies the first characteristic of a religious life. If to shun communal belonging is automatically to belong to the community of community shunners, then we’ve probably made communal belonging far too easy to achieve. However religious this Crusoe may be, my definition can’t allow that he’s founded a religion. But I think my definition is right to deny Crusoe’s communityshunning lifestyle the title of ‘religion’. I can demonstrate the point via a reductio.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

165

Assume that our community-renouncing Crusoe is not just religious, but that he has founded a religion. On the basis of this assumption, we’re going to find it hard to avoid the conclusion that there are, in the world, at least as many religions as there are idiosyncratic spiritual people. But, surely we don’t want a definition of ‘religion’ to have that consequence. Accordingly, I would suggest that a ‘private religion’ is no more of a religion than ‘false teeth’ are teeth. Private religions are not religions. Critics will be eager to find more putative counter-examples. Most of my argument was drawn from Judaism and various Christian confessions, although I did make reference to the Zen practice of zazen. Will my definition of religion extend naturally to Islam, Buddhism, Hinduism, and more? I’ll leave it to critics to find counter-examples. I don’t know of any. Indeed, I claim that our definition includes atheistic, orthodox, orthoprax, western, eastern, new, and old religions, and that it doesn’t include anything that isn’t a religion. Accordingly, we have a definition of ‘religion’ that seems fit for purpose. A religion is a system of ideas and/ or practices that calls for religiosity from its adherents.²⁰

References Alston, W. P., 1967. Religion. In: P. Edwards, ed. Encyclopedia of Philosophy. New York: Macmillan. Alward, P., 2006. Leave me out of it: De re but not de se imaginative engagement with fiction. The Journal of Aesthetics and Art Criticism, 64 (4), pp. 451–9.

²⁰ Thanks to Dean Zimmerman, Lara Buchak, and to two anonymous reviewers. Thanks also to Simon Hewitt, and Dan Howard-Snyder for comments on an earlier draft. I was fortunate to attend two summer workshops on the value and nature of faith, supported by the Templeton Religion Trust. The workshops gave shape to this chapter. I would like to thank the organizers and participants of both workshops. An ancestor of this chapter was presented at a workshop at the University of Haifa on rational belief and normative commitments, organized by Arnon Keren. I’m grateful to Arnon and to the participants of the workshop for their insightful comments. Finally, this publication was made possible by a grant from the Templeton World Charity Foundation. The opinions expressed in this publication are those of the author and do not necessarily reflect the views of the Templeton Religion Trust or Templeton World Charity Foundation Inc.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

166   Audi, R., 1990. Religions and the ethics of political participation. Ethics, 100, pp. 386–97. Audi, R., 2011. Rationality and Religious Commitment. Oxford: Clarendon Press. Camp, E., 2009. Two varieties of literary imagination: metaphor, fiction, and thought experiments. Midwest Studies in Philosophy, 33, pp. 107–30. Clarke, E. et al., 2011. An Introduction to Quaker Testimonies. Philadelphia: American Friends Service Committee. Cuneo, T., 2016. Ritualized Faith: Essays on the Philosophy of Liturgy. Oxford: Oxford University Press. Cuneo, T., 2017. Aligning with lives of faith. Religious Studies, 81, pp. 83–97. Dewey, J., 2013. A Common Faith (The Terry Lectures). 2nd edn,. New Haven: Yale University Press. Durkheim, E., 2008. The Elementary Forms of the Religious Life. New York: Dover Publications Inc. Gendler, T. S., 2000. The puzzle of imaginative resistance. Journal of Philosophy, 97, pp. 55–81. Greenawalt, K., 1988. Religious Convictions and Political Choice. New York: Oxford University Press. Habermas, J., 1995. Reconciliation through the public use of reason: remarks on John Rawls’ political liberalism. The Journal of Philosophy, 92, pp. 109–31. Hirsch, S., 2009. The Hirsch Chumash. New York: Philipp Feldheim. Howard-Snyder, D., 2013. Propositional faith: what it is and what it is not. American Philosophical Quarterly, 50(4), pp. 357–72. Howard-Snyder, D. & McKaughan, D. J., Forthcoming. Theorizing about faith with Lara Buchak. Religious Studies. doi:10.1017/S0034412520000359 Kant, I., 1999. Kant: Religion Within the Boundaries of Mere Reason and Other Writings. Cambridge: Cambridge University Press. Kellner, M., 2006. Must a Jew Believe Anything? 2nd edn, Oxford: Littman Library Of Jewish Civilization. Kvanvig, J., 2013. Affective theism and people of faith. Midwest Studies in Philosophy, 37, pp. 109–28. Kvanvig, J., 2015. The idea of faith as trust: lessons in noncognitivist approaches to faith. In: M. Bergmann & J. Brower, eds, Essays on Faith and Reason. Oxford: Oxford University Press.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 

167

Kymlicka, W., 1992. Two models of pluralism and tolerance. Analyse und Kritik, 14, pp. 33–55. Lebens, S., 2020. The Principles of Judaism. Oxford: Oxford University Press. Lebens, S., Forthcoming. Will I get a job? Contextualism, belief, and faith. Synthese. https://doi.org/10.1007/s11229-021-03045-3 Leibowitz, Y., 1979. Talks on the Ethics of the Fathers and on Maimonides [Hebrew]. Jerusalem: Schocken. Leibowitz, Y., 1992. Judaism, Human Values, and the Jewish State. Cambridge, MA and London: Harvard University Press. Leibowitz, Y., 2002. Accepting the Yoke of Heaven: Commentary on the Weekly Torah Portion. New York: Urim Publications. Leiter, B., 2013. Why Tolerate Religion?. Princeton and Oxford: Princeton University Press. Luhrmann, T. M., 2012. When God Talks Back: Understanding the American Evangelical Relationship with God. New York: Vintage. McKaughan, D. J., 2013. Authentic faith and acknowledged risk: dissolving the problem of faith and reason. Religious Studies, 49(1), pp. 101–24. Magidor, O., 2013. Category Mistakes. Oxford: Oxford University Press. Malcolm, F. & Scott, M., 2017. Faith, belief and fictionalism. Pacific Philosophical Quarterly, 98(S1), pp. 257–74. Marr, R., 2011. The neural bases of social cognition and story comprehension. Annual Review of Psychology, 62, pp. 103–34. Martineau, J., 1888. A Study of Religion: Its Sources and Contents. Oxford: Clarendon Press. Oatley, K., 2008. The mind’s flight simulator. The Psychologist, 21, pp. 1030–2. Otto, R., 2010. The Idea of the Holy. Martino Fine Books. Perry, M. J., 1991. Love and Power: The Role of Religion and Morality in American Politics. New York: Oxford University Press. Quinn, P. L., 1995. Political liberalisms and their exclusions of the religious. Proceedings and Addresses of the American Philosophical Association, November, pp. 35–56. Raz, J., 1990. Facing diversity: the case of epistemic abstinence. Philosophy & Public Affairs, 19(1), pp. 3–46. Teresa of Calcutta, 2007. Mother Teresa: Come Be My Light: The Private Writings of the Saint of Calcutta. New York: Doubleday.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

168   Tillich, P., 1963. Christianity and the Encounter with World Religions. New York: Columbia University Press. Vallier, K., 2016. The moral basis of religious exemptions. Law and Philosophy, 1–28, p. 35. Wettstein, H., 2002. Poetic imagery and religious belief. In: D. Shatz, ed., Philosophy and Faith: A Philosophy of Religion Reader. Boston: McGrawHill, pp. 107–14. Wettstein, H., 2012. The Significance of Religious Experience. Oxford: Oxford University Press. Wittgenstein, L., 1961. Tractatus Logico-Philosophicus. London: Routledge. Young, K. & Saver, J. L., 2001. The neurology of narrative. Interdisciplinary Perspectives, 30, pp. 72–84.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

6 Time and the Nature of the Atonement Amy Seymour

1. Introduction Standard practice in philosophy of religion is to evaluate certain theological positions with respect to how well they comport with desirable philosophical positions and vice versa. Objections to skeptical theism generally posit that the position requires an unacceptable level of skepticism. Compatibility with the doctrine of the resurrection is used in an attempt to break ties in the debate between dualists and materialists about human persons. And so forth.¹ In this chapter, I undertake such a discussion regarding theories of time and the nature of the atonement. One’s views about the metaphysics of time can be used to help clarify one’s views about the nature of atonement, and vice versa. There has been surprisingly little written about the relationship between time and atonement, and I think it time to start the conversation in earnest.² To this end, I discuss the current advantages and disadvantages presentism and eternalism have with respect to atonement theory and provide direction for future conversation. The atonement has received quite a bit of discussion regarding its ethics—for instance, whether notable theories are indeed ethically

¹ For an introduction to the debate regarding skeptical theism, see Almeida & Oppy (2003) and Bergmann & Rea (2005). For introduction to debate about the resurrection, see Baker (2007). ² Notable exceptions are O. Crisp (2009), Lebens & Goldschmidt (2017), and Stump (2012b). For a discussion about the intersection of theories of time and the problem of evil, see Mullins (2014). Amy Seymour, Time and the Nature of the Atonement In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Amy Seymour 2022. DOI: 10.1093/oso/9780192862976.003.0006

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

170   permissible.³ But along with questions about whether and how atonement is permitted, we must also inquire about the mechanics— when and how atonement might work, and for whom. Asking these questions ushers us into metaphysics. The viability of a particular account of atonement depends on what exists—both according to the account and according to our ontology—along with what atonement is to remove or change. Asking questions about when and how this change occurs invites investigation into the nature of time, identity, and persistence. At its core, debate about the metaphysical nature of the atonement focuses on existence puzzles, along with related questions about the nature of identity and change. Someone was not reconciled, and now they are. Must a person exist for this to occur, and if so, when? Since these questions are part and parcel of philosophy of time and persistence, examination of the atonement in light of these debates—and vice versa— promises to be a fruitful endeavor. In this chapter, I will argue that theories of time can be used to help settle or clarify one’s views about the nature of the atonement, and vice versa. In order to establish this thesis, I will focus on what I call transfer and solidarity views of atonement and compare the theoretical costs and benefits of conjoining these views with presentism, on the one hand, and eternalism on the other. I will show that presentism and eternalism face different sorts of difficulties regarding both what exists and cross-time relations. These puzzles can be answered by adherents of both views; whether the solutions are regarded as theologically acceptable is another matter. In short, I will show that which theory of time one accepts affects the specifics of what one can say about the atonement and opens oneself to different objections regarding one’s views on the correct theory. The following discussion will be largely limited to presentist and eternalist theories of time. Presentism is the thesis that it is always true that there are no non-present objects or events. When quantifying over everything that exists, the only objects there are turn out to be present objects—while dinosaurs used to exist, they do not any longer. If an

³ See O. Crisp (2007 & 2008), Fee (2006), Lewis (1997), Murphy (2009), Strabbing (2016), Stump (2012b), Swinburne (1989), and Wright (2006) for a small sample of the literature.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

171

atoning event occurred in the past, then it no longer exists. But so too, events of past sin. Eternalism is the thesis that the objects and events from all times— past, present, and future—exist. The unrestricted quantifier ranges over objects at every time. Dinosaurs and future great-great-grandchildren exist; they are simply not present. Past and future atoning events exist, full stop. But so too, past and future events of sinning. Throughout most of the discussion, I will consider the B-theoretic version of eternalism, wherein all times are equally ontologically privileged, and there is no objective past, present, or future.⁴ I focus on presentism and eternalism because other theories of time add additional ontological complications to the discussion without offering clear ontological benefits over and above what either the presentist or eternalist can offer regarding atonement theorizing.⁵ I argue that neither presentism nor eternalism is clearly to be preferred with respect to atonement theorizing, since each view has advantages and disadvantages related to the questions and desiderata put forth below. Which theory is to be preferred depends, in large part, on what one thinks “at-one-ness” requires. Likewise, particular theories of atonement may require certain temporal ontologies. One goal of this chapter is to help the reader reach reflective equilibrium with respect to these theory choices. Decision points will be laid bare.

⁴ I will consider A-theoretic eternalism toward the end of the chapter, in the section “Eternalist A-theory to the Rescue?” ⁵ Notable theories that I ignore are the growing and shrinking block theories. According to growing block theory, only past and present objects and events exist—but given the progression of time, times that were present become past and new presents come into existence, bringing along with them new objects and events. The temporal extent of reality thus grows over time. Shrinking block is the reverse of growing block—only present and future objects and events exist, and fewer and fewer objects and events exist as future times become present and then pass out of existence. These two views appear to inherit problems from both presentist and eternalist approaches. Strategies that rely on the existence of objects and events of all times cannot be used by the growing and shrinking block theories. And these views inherit existence problems— according to growing block your sins will always exist and according to shrinking block the crucifixion no longer exists. The notable exception that may allow a growing block theorist to escape problems on both sides while retaining all advantages—hypertime—is addressed below. According to some hypertime theorists, God could retain the good objects and events (including atonement) while completely eliminating the bad. See Hudson (2010) for the beginnings of the hypertime view at work and Lebens & Goldschmidt (2017) for its application to the atonement.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

172  

2. Key Theological Assumptions Whether the proper account of the atonement favors or makes trouble for either presentism or eternalism—or whether a theory of time rules out a particular account of atonement—will depend on what it takes to have an acceptable account of atonement itself. We must know some particulars about atonement, other than general statements such as “We are made at one with God and each other.” There appear to be many ways in which such oneness could be achieved. In what follows, I will discuss atonement in terms of the Christian tradition. One reason for this focus, and that of the preceding literature, is due to the unique features of the Christian atonement story. The Christian account appears to require temporally exacting conditions— Christ’s suffering and death was a one-time, sufficient offering for the sins of the world.⁶ Not only is some sort of divine action for sinremission involved, but there are explicitly temporal stakes which could be theoretically difficult. Better to engage with what appears most difficult in a ground-clearing project. Additionally, the atonement is notable for its centrality in the Christian faith, playing a distinctive role. Though I focus on the death of Jesus, my general approach can be used by anyone who thinks that (a) there is some special moment or moments in time where forgiveness of sins happens (e.g., the Day of Atonement) and (b) special divine action is required to remit sin—that is, atoning acts themselves are not an everyday, commonplace occurrence.⁷ For those concerned with these two conditions, much of the material and concerns carry over, mutatis mutandis. However, to the extent one thinks the relevant special divine action can happen at multiple times, the less one may be concerned with some of the particular problems related to existence and forgiveness discussed below. I focus on a narrower treatment of the existence, forgiveness, and removal of sin or effects of sin, all of which is of fundamental importance

⁶ See Hebrews 7:27, Hebrews 10:10, and 1 Peter 3:18. ⁷ Though the effects of said atoning acts, such as the forgiveness of sins, might be more commonplace.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

173

to the atonement. A complete treatment would give proper attention to the scriptural idea that atonement is required for much more than just moral transgression.⁸ But moral transgression is a good place to start, especially given the historical focus on ethics, and is a way to keep the discussion manageable. For similar reasons I focus specifically on what was accomplished through Christ’s suffering and death. Accounting for other aspects of the atonement magnifies the problems discussed below—if there are additional problems of purity that need to be dealt with (e.g., in the case of non-living objects), they will be subject to similar discussions regarding the existence and removal of the problem. I assume that both God and sin exist and that, somehow, humans are made right with God despite their sin. While sin and its effects may be broad-reaching, I will further limit discussion to individual sin and its effects. For present purposes, let sin be either an event or act that assigns some property to an individual (e.g., having sinned) or that has continuing effects (or both). Whatever else the atonement may encompass, it (a) relevantly removes or reverses the evil of our sin⁹ (b) at least in part by Christ’s sacrifice on the cross—for “Christ suffered for our sins once for all time, the just for the unjust.”¹⁰

3. Potential Roadblocks to the Debate The widespread disagreement regarding how atonement is to be understood accounts for much of the reason why the metaphysical side of the atonement has been underexplored. Three diagnoses of the nature of the disagreement and ensuing oversight will be useful for the following discussion. First, there is disagreement about what is required for being made “at one” with another human person after more ordinary

⁸ For instance, one must make sense of the cultic practices of Israel that are repeated in the book of Hebrews regarding impurity and atonement made for places and inanimate objects. (Thanks to Carl Mosser for this point.) But what it takes to make something clean may map onto a taxonomy not dissimilar to the one I present below. ⁹ 1 John 3:5: “You know that he was revealed to take away sins, and there is no sin in him.”. Biblical passages are from the NRSV translation unless otherwise specified. ¹⁰ 1 Peter 3:18a, NASB.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

174   transgressions.¹¹ Some transgressions may be smoothed over after a simple apology. For others, it is perhaps not clear how the parties could possibly be fully reconciled at all.¹² Religious discussions of atonement add an additional layer—the relationship between humans and God has been broken, and some sort of special divine action is needed in order to make things right again. One reason for this needed divine action is that the break in the relationship is due to human sinful action, which brought about, among other things, consequences which sinners are unable themselves to remove.¹³ Given the widespread disagreement about the nature of human-tohuman forgiveness, it is no surprise that disputes increase when discussing divine action. In other doctrinal disputes, one way to find common ground is to start with what is required according to a central creedal understanding of the faith. For this reason, Christian philosophers commonly return to the Apostle’s or Nicene creeds. While dualists and materialists argue over whether the doctrine of the resurrection is consistent with certain metaphysical views, it is generally agreed that a proper account of the resurrection must affirm the raising of physical bodies to life. “I believe in the resurrection of the dead” cannot simply mean, “When I think about God on a bad day, my mood is improved.” This is in stark contrast to debates about the atonement. Despite its centrality, there is no universal, ecumenically accepted account of atonement or even what is required for such forgiveness of sins. I take this to be the second main reason for the metaphysical oversight—it is unclear where there is enough common ground for metaphysical theorizing to occur. Too much theoretical disagreement can make for unwieldy metaphysical discussions.

¹¹ This is one reason it will behoove us to look at several differing sorts of approaches to the atonement. In so doing, we will see important similarity in the temporal questions asked: we cannot escape them. ¹² See Rea (2019) and Strabbing (forthcoming) for discussion of the difficulties of full reconciliation. ¹³ Importantly, according to the Jewish and Christian traditions, the relationship between human beings and the rest of creation has also been broken. The breaking of the divine‒human and human‒creation relationships has traditionally been thought to have occurred as a result of the Fall. I will abstain from theorizing about the nature of the Fall itself for the purposes of this chapter. It has also been argued that the continued break in this relationship is not entirely due to sinful human action. See, for instance, Adams (2006) and Rea (2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

175

For the Christian, the ecumenical creeds do provide some common ground: The Nicene Creed asserts, “for our salvation, [the Son] came down from heaven, and was incarnate by the Holy Spirit of the Virgin Mary, and was made man, and was crucified also for us under Pontius Pilate,” and the Apostles’ Creed affirms that there is “the forgiveness of sins.” But little is said about how exactly (and non-metaphorically) the work of Christ accomplishes salvation and forgiveness, both in terms of what is done and when it is done. Given this, there are more theoretical questions to be answered about the atonement than in many other areas of philosophical theology. One large point of agreement, though not universal: The crucifixion of Jesus is somehow both causally efficacious and necessary for atonement between God and human beings.¹⁴ I call this the necessity requirement: Something about Christ’s death on the cross itself was necessary for atonement between God and humans.¹⁵ Eleonore Stump (2018) puts the point nicely: If Christ’s passion and death are to be efficacious as the remedy for human sin, it seems there ought to be something about the passion and death of Christ itself that is directly involved in the remedy; that is, Christ’s passion and death ought to be directly or else somehow crucially instrumental to the giving of grace in humans. (31)

There is debate regarding the necessity requirement, but Stump assumes that Christ’s passion and death are somehow necessary to atonement, and that this assumption is widely shared throughout church history.¹⁶ I join her in this assumption—especially because it is better for us to

¹⁴ Indeed, it is mentioned in both of the above creeds and explicitly linked to our salvation in the Nicene Creed, though not in terms of the necessity requirement per se. One of the reasons in support of the necessity requirement, in connection with the crucifixion, is that the Father did not “take this cup” from the Son (see Matthew 26:39). Perhaps it was not possible to do so. ¹⁵ There are varying degrees of strength one can give to the necessity requirement. One need not think this requires something like metaphysical necessity. Perhaps accidental necessity is enough. Debate about this point will take us beyond the scope of this chapter. ¹⁶ This is one of the problems with the Thomistic interpretations of the atonement, according to Stump: According to those accounts “it is not the case that a perfectly good God is unable to give grace to human beings without the passion and death of Christ” (31).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

176   assume more stringent requirements at the outset of our discussion, and loosen them later if need be. Better to start with the harder project first. The phrases “for our salvation” and “the forgiveness of sins” give no concrete ontological guidance. What we do know is that our sins are no longer held against us, but there could be many ways in which forgiveness is accomplished. How does atonement work, exactly? For example, if it requires payment, what are the particulars of that payment? If a person’s bad deed is paid for, does that person need to exist? The major models of atonement in the Christian tradition offer some answers, but said answers involve much metaphor. Widespread use of metaphor is the third reason for the historical oversight. Ransom theory models maintain that Christ’s death freed us from the power of sin and death by serving as a ransom on our behalf. However, it appears this model cannot be literally true, since if it were it would make the Devil powerful enough that God would need to pay some sort of ransom to him to make things right. Christus Victor models maintain that Christ’s death and resurrection is a victory over evil that broke the power that sin had over humanity—but it is not metaphysically clear how this was a victory that broke said power. Satisfaction models maintain that Christ’s death satisfied the wrath of God that we deserved; penal substitution models that Christ, in his death, bore the punishment for our sins in our place. Both satisfaction and substitution models appear to engage, at crucial points, in “forensic fiction” (O. Crisp, 2009, 436ff.). And even if it is literally true that Christ took away either our sins or punishment for our sins, satisfaction and substitution models themselves offer no metaphysical guidance for how this was accomplished.¹⁷ Finally, moral exemplar theories maintain that Christ, in his life and death, served as a moral teacher and exemplar that we must emulate in order to become right with God. But it is not clear how Christ’s example makes a sinner right with God—especially if the sinner died well before the Incarnation. The mere existence of an exemplary person gives no obvious benefit. Nor is it clear why Christ’s death was a necessary act. ¹⁷ How one might go about this sort of account is discussed below in the section “Consequence Removal Questions.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

177

These models are not all mutually exclusive, but there is widespread disagreement about which (if any) gives a proper account. These models are thin on metaphysical particulars. While metaphor is not without some interpretive guidance—after all, none viewing a performance of Romeo and Juliet would take Juliet to be a gaseous giant upon hearing “Juliet is the sun”—I take there to be either anxiety or complacency regarding the above atonement metaphors and others used in Scripture, which has prevented robust metaphysical discussions. Here metaphysics can be of some use. One of the tasks of atonement theorizing is to distinguish metaphorical language from literal use. Is it literally the case that Jesus was “crushed for our iniquities . . . and by his bruises we are healed” (Isaiah 53:5)? Is Christ literally a “ransom for all” (1 Timothy 2:6)? These are hard questions; but if we discover, say, that presentism is incompatible with a literal construal of one or more of these characterizations and we are presentists, then we have been helped a bit in our interpretive work. Introducing theoretical considerations can often help illuminate what is ultimately at stake and whether our beliefs are consistent with the introduced theory. This is not to say that we cannot affirm that certain key ingredients of atonement theorizing, such as salvation (whether in terms of the forgiveness of sins or in the reconciliation of God with humanity), happen on any given theory of time. If the primary function of the atonement is salvation, this function can be fulfilled on either presentism or eternalism. But notably, what salvation looks like—when and how it comes to be—will vary. What does it mean to say that our sin is relevantly removed? The available metaphors in Scripture also give us little guidance. Isaiah 44:22 says, “I have swept away your transgressions like a cloud, and your sins like mist; return to me, for I have redeemed you”, while Acts 3:19 declares, “Repent therefore, and turn to God so that your sins may be wiped out.” Regarding these metaphors, both presentists and eternalists may attempt to insist that their view is superior, in that it is the one that properly accounts for how sins are “swept away” or “wiped out.” Indeed, even the scriptural passages appear to be in temporal tension, as sins appear to have already been removed in one but have yet to be eliminated on the other.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

178   Once the advantages and disadvantages of each theory have been laid bare, we can move toward reflective equilibrium between our theory of time and interpretation of atonement metaphors. If any reasonable understanding or interpretation of a metaphor includes p and a theory says not-p, then we are forced to decide between the two. Here we should not be content to sit with mystery. Highly abstract metaphysical debates are relevant to philosophical theology in ways that are deeply important for even pastoral discussions and understanding— consider, for example, the importance of understanding what it means for Christ to be fully human and fully divine. What we can truthfully assert, even in pastoral contexts, will depend in part on our metaphysics. Pastoral utterances about the atonement include statements like “Jesus died specifically for your particular sin.” Is this a statement about Jesus’s general salvific work or about payment for a particular action that a specific person performed? The latter will require certain metaphysical machinery. Another pastoral comment, “One day there will be no evil,” similarly has different implications depending on one’s metaphysics. Is it that evil will cease to be performed? Or will evil be removed from reality altogether?

4. Existence Assumptions and Approaches to the Problem At the heart of these questions are existence puzzles: Who needs to exist (or not) in order to remove the consequences of sin and evil generally, along with those of a particular sinner? In what follows, I assume NeoQuineanism: There are no objects which do not exist. Thus, if an object or event has a property or stands in a relation, that object or event must exist. There is a question not just about the existence of particular objects or events, but the nature of the properties they bear—depending on what kind of properties they are, different sorts of solutions will be available (or not). Here I will assume that there is something about the nature of personal sin and evil which has to do with the person themselves and is not merely an extrinsic property. I assume this due to a widespread

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

179

intuition about personal responsibility, which I call the substantivity intuition: If a person is to be properly held responsible for an action or have responsibility for an action remitted, such responsibility or lack thereof cannot be due to mere Cambridge change. A mere Cambridge change is a change in purely relational or extrinsic properties, which has nothing to do with the actual objects or events in and of themselves.¹⁸ For instance, suppose Alice is shorter than Bob at a time t₁ and taller than Bob at t₅ and Bob remains the same height. Bob thus has the property being taller than Alice at t₁ and changes to having the property being shorter than Alice at t₅. In contrast, an intrinsic property that Bob has is being 50 1100 (or also consider the changes in intrinsic properties that he can undergo, such as being 50 in 2000 and being 50 1100 in 2020). The latter sorts of properties have something to do with Bob in an important way that the former properties do not. This is not to say that atonement has nothing to do with changes in relational properties. Indeed, such changes are to be expected if we are reconciled and made “at one” with another! But the point of this assumption is that atonement requires something more substantive than mere Cambridge changes. If I have wronged you, the problem cannot be solved simply by some sort of change that does not properly affect you. This idea generalizes to the substantivity intuition. One can certainly resist the substantivity intuition, or perhaps reject Neo-Quineanism in favor of multiple quantifiers.¹⁹ However, one task of this chapter is to clear ground for future discussions, along with determining which ontological commitments constrain our theological

¹⁸ The term “Cambridge change” is due to Peter Geach (71–2). Geach uses it to argue against the view of change put forward by Cambridge professors, such as Bertrand Russell. According to Russell and many others, property change just is having a property at one time and lacking it at another (or vice versa). A poker changes from cold to hot simply by having the property being cold at t₁ and being hot at, say, t₅. While there are debates about whether this account of change is substantial enough, I will not engage them here. I shall assume this account of change— popular with eternalists—works. The “mere” in the above classification of mere Cambridge change is thus of utmost importance. The example of the poker appears to involve either intrinsic property change or a robust relational change in the poker—something about the poker itself at least partially explains or accounts for the change. This is in contrast to mere Cambridge changes, which occur without reference to things like the nature of the poker itself. ¹⁹ Though I ultimately have my doubts about the prospects of the latter approach, sharing concerns related to those in van Inwagen (2009) and (2014).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

180   commitments and vice versa. Thus, it behooves us to explore more constrictive requirements first and loosen those later if need be. Together, Neo-Quineanism and the substantivity intuition generate some existence puzzles: It appears things need to exist in order to be atoned for. Thus, one might think all objects and events from all times need to exist in order for sin and evil to be removed. On the other hand, complete removal of sin and evil appears to require the elimination of sin and evil from the domain of things that exist. Evil is eliminated, full stop. Here, then, is a choice point about the nature of the atonement itself. When we are made “at one,” how much evil and suffering is removed, and in what way? We must decide between: (1)

Evil Elimination: Atonement makes it possible for all sin and evil to be removed from reality entirely; or

(2)

Consequence Removal: Atonement makes it possible for the consequences of sin and evil to be removed from reality.²⁰

Further, the Christian will have the following condition, related to but weaker than the necessity requirement: (3)

The means of the atonement—the work of removing sin, evil, or its consequences—is primarily Christ’s sacrifice.

The fundamental choice between Evil Elimination and Consequence Removal is whether what has to be removed from reality is evil itself— including sin and all its negative consequences—or only the consequences of sin and evil. On this point, there are a few simple answers proponents of different temporal ontologies can give, in the hopes of claiming theological advantage. Here it may seem this choice point offers a quick advantage to the presentist: She can affirm Evil Elimination, while the eternalist may only affirm Consequence Removal. Given eternalism, the past and the future also exist, which means that our past sinful acts still exist, even if we are

²⁰ Or, at minimum, it allows for the consequences to be transformed or overcome.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

181

forgiven them. A robust view of evil elimination may seem to require a presentist metaphysics. The eternalist’s ontological commitment to the continued existence of my sin may appear problematic, given what many Christians want to say about Christ’s salvific work. Due to the cross, the sins of those who believe have been “wiped out,” swept away like a cloud. But, given eternalism, there is a very real sense in which the sins of the repentant haven’t been swept away—they still exist. Thus, it looks like sin has an unacceptable foothold in reality, given that Christ’s sacrifice and death is to have defeated sin and evil. Given eternalism, this sin remains. Presentism thus appears to be the superior view according to the Evil Elimination approach, since according to it the sins for which you have been forgiven no longer exist. Here the eternalist will object that this is moving much too quickly. According to presentism, it is not the death or sacrifice of Christ that causes sin to cease to exist, but the mere passage of time. This is a violation of (3), the necessity requirement, and cannot be the correct view of the atonement, since it would trivialize the death of Christ—if all that is required is for sin to cease to exist, Christ’s death is not needed at all! Sins and evil disappear, no special action needed. Additionally, the positive events associated with the atonement have, along with past sin, also been eradicated. According to presentism, the atoning death of Christ no longer exists—surely, the eternalist says, that is a worse consequence!²¹ So perhaps eternalism should be favored, since it affirms not just the continual existence of the death of Christ, but the eternal existence of his salvific death. It’s always true that Christ’s salvific act exists, full stop. While our past and future sinful acts will always exist, the act of

²¹ Whether this follows depends on one’s theory of events. While not a presentist, Szabo (2006) thinks presentists should say events always exist, but do not always occur. However, it is unclear whether this view will fulfill either the Neo-Quinean condition or the substantivity intuition. The presentist also faces fatalist difficulties regarding future contingents, which will likely cause difficulty for parts of this strategy which involve future contingents (see Rea (2006)). The presentist can escape such fatalist concerns, but plausible ways of doing so involve rejecting the need for truths to be grounded in or supervene on reality (see Merricks (2007)) or denying that there are true future contingents (see Seymour (2016)). The viability of this general strategy remains one for future research.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

182   redemption also always exists. Thus, the continued existence of our sinful acts need not bother us—they are eternally covered by the blood of the Lamb.²² The presentist, on the other hand, must admit an existence gap between a large number of sins and the act that covers and atones for sins. Those that lived and died pre-Christ were without an act central to their forgiveness. If Christ’s sacrifice on the cross is what “wipes out” sin, then even the pre-Incarnation faithful died without their sins properly dealt with. But this appears to be in contrast to passages of scripture which tell of forgiveness of sin before Christ.²³ The presentist can say, however, that the forgiveness given before Christ’s death was fundamentally forward-looking—sins could be forgiven precisely because God knew of and planned the coming sacrifice of Christ.²⁴ However, the forgiveness was not complete, since a central piece required for it was missing until Christ’s death. This incompleteness of forgiveness is not necessarily a drawback for the presentist, since this seems to be precisely what is taught in the book of Hebrews 11:39‒40: “Yet all these, though they were commended for their faith, did not receive what was promised, since God had provided something better so that they would not, apart from us, be made perfect.”²⁵ The presentist need not deny (3): She can say that Christ’s death is what is responsible for complete consequence removal. Presentists and eternalists can both affirm that sin is “wiped out” if this is understood as Consequence Removal. But eternalists appear unable to ²² Since eternalists affirm that all times and the events therein exist, they can say that there is never a time that the atonement doesn’t reach, and thus affirm the declaration in Revelation 13:8 that Jesus is “the lamb that was slain from the foundation of the world.” The presentist is not forced to deny this, but she must explain the passage merely in terms of what God planned or intended and cannot take the passage literally. ²³ Isaiah 44:2; 2 Samuel 12:13. Note that this concern holds for any theory which affirms (3). As long as the primary means of atonement is Christ’s sacrifice, something necessary for atonement was missing before said sacrifice, regardless of the particular mechanics of salvation itself. ²⁴ Isaiah 44:2; 2 Samuel 12:13: “David said to Nathan, ‘I have sinned against the Lord.’ Nathan said to David, ‘Now the Lord has put away your sin; you shall not die.’ ” ²⁵ Additional support for this view is found in Hebrews 11:13: “All of these died in faith without having received the promises, but from a distance they saw and greeted them.” There is a view, however, that can affirm both that the sacrifice of Christ must exist in order for there to be forgiveness and that the pre-Incarnation faithful were forward-looking, having not received the promises: an eternalist A-series, wherein all objects and times exist, but time is also dynamic so that objects and times are objectively future, become present, and then pass into the past. This potential will be discussed in depth in a later section.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

183

affirm Evil Elimination—to the extent that the continuing and eternal existence of sin and evil is a problem, her theory is worse off. This is not necessarily a disadvantage according to much of the tradition: Most Christians affirm the doctrine of hell. Given hell, there will be some sort of unrepentant sin that continues to exist after the second coming of Christ and Judgment Day. As long as we admit that some sort of sin or evil will exist post-Judgment Day, why should it bother us if more sinful things still exist in time, even given redemption? The relevant difference lies in the ability to affirm Evil Elimination: Presentism makes such a view at least theoretically possible, but eternalism does not. According to the presentist who believes in hell, the only continued existing sin or suffering from sin comes from those who rejected Christ. On eternalism, the sins of those at one with God and fully redeemed forever exist. The presentist also need not adopt the doctrine of hell and the continually existing suffering from sin that accompanies it. The presentist can maintain that God could prevent creatures from continuing to resist God (either by annihilating them or, if universalism is possible, by saving everyone); whereas, according to the eternalist, God cannot go back and erase an existing time where someone sinned. Presentism allows for the option of the removal of all sin and evil from existence (via annihilationism or universalism), whereas eternalism cannot. While atonement itself does not remove all sin and evil, a combination of presentism, atonement, and some other special divine action makes Evil Elimination possible.

5. Hypertime to the Rescue? However, recent work by Samuel Lebens & Tyron Goldschmidt (2017) argues we could eat our cake and have it, too. They argue that theories of hypertime would allow God to change the past—atonement would thus remove all sin and evil. The good would remain, including the atoning acts, but the evil goes.²⁶ Accordingly, they present a picture where Evil

²⁶ They apply this to the Jewish tradition, though their temporal framework could be used by other religious traditions.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

184   Elimination is accomplished by the atonement on what they call the Divine Proofreader Theory. To do this, they utilize hypertime theories, according to which the domain of what exists keeps changing, as spacetime grows, shrinks, or develops. Growing block theories posit that past and present objects and events exist. As time moves on and new times become present, reality itself grows. But if reality is growing, “there must be a ‘second sort of time’ in which it does its growing” (van Inwagen, 2010, 14). This “second sort of time” is hypertime, and it is what makes sense of accidental necessity—there is a record of what that block used to be like, even after or despite radical changes.²⁷ Peter van Inwagen utilizes hypertime to argue for a non-paradoxical way of changing the past. One cannot, it was assumed, travel back in time, successfully kill one’s own grandparent, and have them remain dead. For the grandparent’s existence is causally necessary for the time traveler’s existing, traveling back in time in the first place, et cetera. Permanently removing the grandparent from the spacetime block is impossible—at least, according to traditional eternalist conceptions of time travel. Combining hypertime with a spacetime block that can change changes everything. On a growing block theory, the time traveler could travel back to an earlier point in the block—and in so doing obliterate all that exists between her former present and her destination. Reality would move from being larger to smaller and then grow again. The traveler could kill her grandparent on this story, and the block continue to grow from there. But what the traveler could not do is make it so her grandparent never hyper-existed—for hypertime records or bears witness to the traveler’s being born, traveling backward in and through times which no longer exist according to the block, et cetera. While the traveler can eliminate her grandparent from the spacetime block, she cannot completely erase them from reality altogether. That is, she cannot eliminate them completely from the domain of existence, full stop. It remains true that there was such a person in the hyper-past. ²⁷ Accidental necessity: If it was the case that p, then it always will be the case that it was the case that p.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

185

Hud Hudson (2014) develops this framework for theological use, according to which pain and suffering will continue to exist according to hyperhistory, but not in the past (193‒4). Lebens and Goldschmidt go farther and attempt to completely eliminate sin and evil from hypertime itself—thus removing sin and evil from the domain of existence altogether. Utilizing an infinite series of hypertimes, they argue that God could always hyper-make it so that evil and sin did not exist according to an earlier hyperhistory. Consider an event of evil, E. Lebens and Goldschmidt . . . propose that God can complete an infinite number of tasks, deleting all traces of E from time, hypertime, hyper-hypertime and so on. God can’t undertake this supertask in time, nor in any level of the hypertime hierarchy, if he wants completely to eradicate E from every level. (2017, 7)

God, being atemporal, performs this supertask outside of time. Thus, God could eliminate E from infinitely many timelines with no paradox (7‒8). They think, “The supertask would atemporally exist, but the deleted evil wouldn’t. There is no reason, besides incredulity, to deny that God could perform the supertask” (8). While the theory is interesting, I believe it moves far too quickly. It is debatable whether such supertasks are possible, especially if the supertask is performed by a being who relates and responds to the free actions of temporal beings (regardless of whether the being itself is atemporal). Further, if such a supertask were accomplished, it appears we should never know about it—a temporal supertask should be manifest to us wherever we are in time. So, if God will eventually eliminate all evil from the world in this way, we should see no evil now.²⁸ Otherwise, the supertask would not be accomplished. Second, accidental necessity appears to be conceptually, as well as metaphysically, necessary. Lebens and Goldschmidt label their complete elimination of evil as “Ultimate Forgiveness” (1). But it is difficult to see ²⁸ Here I agree with an anonymous reviewer cited in Lebens & Goldschmidt (2017) in footnote 7 on page 7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

186   how the view is actually about forgiveness when, arguably, the forgiveness itself will be erased, and not even God will see or know otherwise. For if there is a complete removal from reality of what was forgiven, it is difficult to see how the act of forgiveness could remain or even be intelligible—forgiveness requires something which was forgiven.²⁹ Consider: It seems reasonable to think inhabitants of heaven could wonder “What was I forgiven for?” But Lebens and Goldschmidt’s account seems in tension with any narrative the redeemed could tell in answer to such a question. Either God’s atoning act(s) remain in the domain of existence or they do not. If God’s atoning acts are eliminated by the supertask, then the atonement drops out of the story of union with God altogether, even to God. I wasn’t forgiven. No atonement occurred. The story simply is that we had a perfect life and now we are enjoying great union with God. On this option, their theory isn’t actually a theory about the atonement at all and can be ignored for now.³⁰ Lebens and Goldschmidt, however, endorse the other option, the view that sin and evil is gone, but God’s forgiveness and atoning acts remain (8). Herein lies the conceptual difficulty: You cannot be forgiven if you haven’t done anything wrong! And if there is no accidental necessity and my sin has been completely eliminated from reality, it simply is true that I have never done anything wrong, nor will I ever. My being forgiven, then, appears incoherent. Surely God would want for our experience to make sense—for us to have internal self-awareness about what has happened to us and why we find ourselves in the position we are in. On Lebens and Goldschmidt’s view, if I ask, “What was I forgiven for?,” even God doesn’t know! We need to hold onto the principle of accidental necessity in the case of sin and forgiveness, because our reality in heaven won’t make sense without it—not just to us, but to God. Initially, one might think Lebens and Goldschmidt can make use of the same strategies that are available to the presentist when trying to

²⁹ This is not just an issue about what must exist in order for certain relations to hold, though that is relevant. The stronger point is that the concept of forgiveness requires that there be wrongdoing that is no longer held against someone. ³⁰ Thanks to Jonathan Jacobs for this point.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

187

answer questions about the nature of the past and future. Suppose the presentist agrees that truth supervenes on being: If a proposition about the past is true, then there must be an appropriate object or event for that truth to supervene on. Objects and events that are merely past, however, lack anything for a truth to supervene on—or so the objection goes.³¹ Thus, the presentist cannot adequately account for the difference between a world in which it is true that a brachiosaur walked on this spot 150 million years ago and a world in which it is true that the world came into existence five minutes ago. The same would hold true for atoning events, such as the crucifixion. The presentist has available a method to solve the puzzle that Lebens and Goldschmidt do not—appeal to something which exists on which the truth supervenes (or perhaps more strongly: grounds or makes true the proposition in question). Bigelow (1996) argues that the world itself instantiates properties like there were previously brachiosaurs; T. Crisp (2007) appeals to presently existing, abstract times. These and other presentist strategies are used in order to account for truths, whether in terms of supervenience, truth-making, or grounding: Given this, Lebens and Goldschmidt cannot adopt any of them without it being true that, for instance, I sinned.³² And a Christian has some reason to think evil is not eliminated from the narrative entirely but somehow redeemed—for the resurrected Christ’s body still bears signs of crucifixion (see John 20:24‒9).³³ A main motivation of Lebens and Goldschmidt is to make sense of scriptural passages about forgiveness and atonement by examining the imagery to see what is to be taken literally, and what is metaphysically required to do so. I share this motivation. One of their key passages is Isaiah 43:25: “I, I am He who blots out your transgressions for my own

³¹ See Sider (2001), 35ff. ³² Another presentist strategy is to deny that truth does supervene on being (See Merricks (2007)). This may be a promising route for Lebens and Goldschmidt, though they will still face the conceptual difficulty about forgiveness. ³³ Other methods Lebens and Goldschmidt use (e.g., the moving spotlight and hyperpresentism) appear to give theoretical advantages only on par with the presentist’s ability to have evil eliminated as discussed in the previous section. While Lebens and Goldschmidt do not specifically reference the above debate and my use, they acknowledge the shared benefits and drawbacks I’m noting when discussing “Ostrich Temporalism” (10).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

188   sake and I will not remember your sins.” They take this lack of remembrance seriously. If God does not remember and God is omniscient, then the sins must not exist. Or at least, we have good reason to think so. However, while I find the metaphysical theorizing rich, such literal non-remembrances do not appear necessary or merited given a strict reading of the textual language alone. The passages regarding the word ‘remember’ do not appear to promise or indicate complete removal of evil and sin from God’s thought. The root Hebrew word that is here translated as ‘remember’ is ‘‫‘[ ’זכר‬zakar’], which has an active component.³⁴ The remembering is not simply intellectual—it is meant to move one to action based on what one is remembering and one’s relationship or duties to the events or persons involved. It is also used in contexts of announcing, praising, holding against, and proclaiming. We can see this in other passages with the same word, where it seems clear God had not eliminated either the thought or existence of the relevant person: “Then God remembered Rachel, and heeded and opened her womb” (Genesis 30:22).³⁵ If I ‘zakar’ my mother’s birthday, I will not just know that it is today—I will give her a call. Remembering that it is trash day, on this conception, is not just intellectually knowing trash is collected—I will go outside and move the cans to the curb. If I do not remember, in this way, an action someone did against me, I will not go about proclaiming it to others and acting in accordance with it. Given this, I set their theory aside for now and will assume accidental necessity: If it has been the case that p, then it always will be that it has been the case that p. Evil Elimination needs not just a dynamic theory of time, but one in which past (sinful or evil) times go out of existence. Accidental necessity mandates that sin and evil cannot be eliminated from reality entirely on a non-presentist view—and perhaps on a presentist ontology as well, depending on what accounts for the grounding of past and future truth-claims. So rather than focusing simply on the advantages of mere existence (or non-existence), it would behoove presentists and eternalists to examine

³⁴ See Logos Bible Database, ‘zakar’ (2020). ³⁵ Among other passages that use this conception of memory, see Genesis 8:1, Genesis 19:29, Esther 9:28, Psalm 106:45, Isaiah 12:4, Isaiah 43:26.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

189

more closely what is meant by the Christian metaphors of “wiping away” sin and being “covered” by the blood of Christ. Let us turn, then, to Consequence Removal theories.

6. Consequence Removal Questions The first question to ask is: How does the consequence removal of atonement work? First, we must figure out what the consequences are. Then, presentism and eternalism will have importantly different implications. We first start with the general: (4)

Mechanics Question: How does atonement happen?

Answering (4) will tell us important things about the metaphysics underpinning the act(s) of atonement, and different atonement theories will have different implications. An informative answer to (4) will involve both a description of the relevant consequences removed by atonement, as well as an explanation of how they are removed.³⁶ Any such explanation, by its nature, will involve answers to the: (5)

Temporal Question: When does atonement happen?

Answering (5) will tell us important things about cross-time relations, should there be any. Importantly, an answer to the temporal question will at least attempt to explain in what sense Christ’s death is “once for all.” Hebrews 9:28a tells us that Christ was “offered once to bear the sins of many.”³⁷ Does this mean that Christ’s redemptive work was completely accomplished on the cross?³⁸ ³⁶ The term “consequences” applies broadly, to cover the effects of sin that must be dealt with in order to have restoration—whether those effects amount to something like debts that need to be paid for or shame and guilt that needs to be healed depends on the particular atonement theory in question. ³⁷ This requirement is also found in 1 Peter 3:18a: “For Christ also suffered once for all, the righteous for the unrighteous, in order to bring you to God.” ³⁸ I do not mean to indicate that the Father and Spirit play no role in atonement—only that being an offering for the forgiveness of sins is the work of the Son. The full work, timing, and effects of the atonement involve more than one agent (indeed, for the Christian, it must involve at least three!).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

190   There is reason to think that while something was completely accomplished during the crucifixion³⁹—perhaps forgiveness—atonement itself cannot be accomplished all at one time. For atonement involves multiple parties being made “at one.” To truly be made at one with another, forgiveness is not enough. One must be reconciled to another. The former can occur without the latter. For instance, one can forgive a debt without being reconciled, as one may continue to be estranged from the former debtor. Richard Swinburne argues that atonement requires four features: repentance, apology, reparations, and penance.⁴⁰ The initial work of atonement, reparations, is the feature that plausibly happens at one time on the cross, fulfilling the “offered once for many” scriptural desideratum.⁴¹ But the other three features do not need to happen all at one time. If they had to be all at once, then Christians would be committed to, at minimum, a view on which every individual eventually made at one with God exist at the moment(s) of the crucifixion. This is highly implausible, as I will argue below. For now, then, we will limit consideration of “Christ’s redemptive work” to what Christ himself accomplished. But even if the “at one time” requirement is limited to reparations (or something like it), the temporal question admits further fineness of grain, such as: (5a) Is the initial act of Christ’s atoning work different from the effects of this atoning work, and if so, how? That is, is the reparative work of the atonement (both the action and effects thereof) accomplished all at one time (or within a small period of time)?⁴² Or does the initial divine saving act occur all at one time, which ³⁹ See John 19:30. ⁴⁰ Swinburne (1989), 81. ⁴¹ One may respond that apology, repentance, and penance are creaturely actions, and thus do not count as Christ’s redemptive work. This sort of response appears to have too narrow of a conception of effects. I am able to make proper penance only if the offended party will accept what I do, and the penance I give to God for my sin is accepted because of the atoning work of Christ. However, a discussion focusing more particularly on Christ’s specific work is below. ⁴² Some may reject any interpretation that implies that Christ’s death was necessary for atonement. Instead, his death accomplished atoning work that could have been accomplished some other way. As noted earlier, this is set aside for our purposes. There is an additional question of whether Christ’s atoning work at the crucifixion was, by itself, the initial act, or

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

191

then has its salvific effects spread throughout history? To answer these questions, we must determine what sort of cross-time relations are involved (if any). There are further puzzles, too, about the temporal boundaries of events and when the full effects of events are settled. One quick response is to insist that there are no cross-time relations involved: Even if atonement requires acts (or acceptance of an act) from at least two agents, these happen all at one time. While the initial act of Christ’s salvific work is conceptually distinct from its effects, the act and effects are synchronic (or near enough to not cause great difficulty). Thus, the full, complete effects of Christ’s reparative atoning work for individuals occurred all at one time. This response is implausible, as it requires an extreme form of mind‒ body dualism. If Christ’s full and complete atoning work for individuals had to happen all at one time—no cross-time relations allowed—it plausibly requires not only the existence of every individual atoned for at the time of the crucifixion but their conscious awareness and assent or acceptance of the atoning act.⁴³ If that is the case, then one would need an extreme form of dualism, wherein each atoned person not only existed but was conscious and able to make decisions at the moment(s) of the cross. It would not be enough for those individuals to simply exist, quantifiers unrestricted, as part of the atoning work for individuals existing in the past or future relative to the cross would then take place at those other times when those individuals exist. Thus, unless either party wants to be an extreme dualist, presentists and eternalists must agree that the full scope of atonement takes place, in some sense, at different times. The disagreement will then center on whether the individuals existing at times other than those of the cross

rather one among many. That is, did the acts and events at that time encompass all of the reparative work needed for atonement or all of the divine atoning act? Perhaps Christ’s earthly ministry somehow helpfully contributes, or the like. One might also think that Christ’s death was both necessary for atonement and for any further atoning work to be accomplished. Disambiguating acts and their effects will get more complicated if one thinks that atonement is not fully captured by the forgiveness of the sins of individuals. ⁴³ Unlike ordinary Cartesian dualism, which can allow for souls to come into existence after the time of the crucifixion, this dualism requires that all people exist at the time of the crucifixion. And unlike a Williamsonian view of necessary existents, all people must have robust enough existence so as to have certain sorts of conscious experience and ability to, for example, assent.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

192   must exist simpliciter when the crucifixion exists and the nature of cross-time relations—that is, over the truth of eternalism, among other things.⁴⁴ If we (reasonably) think that the full or initial divine act(s) of atonement were not instantaneous, but happened across time, how much does that theoretically constrain us?⁴⁵ Unfortunately, it is even unclear what it takes for some events to happen “all at one time.”⁴⁶ Judith Jarvis Thomson (1971) famously raises puzzles for what time an event like a killing takes place. With some events, it seems clear how to make this determination. Suppose Jill and Jane jump at t₁. Jill and Jane’s jumping then happens all at one time—t₁. But some actions do not seem to admit this sort of easy analysis. Suppose instead that Jill shoots Jane at t₁ and Jane succumbs to her wounds and dies at t₅. When did Jill kill Jane? It’s hard to say—one attempted answer is that Jill’s killing of Jane happened “all at one time”, since what was necessary for the death (and thus killing) of Jane all occurred at t₁.⁴⁷ A verdict of the killing’s occurring “all at one time” due to the necessitation of the future death seems unsatisfactory. It is true that the important act of bringing about Jane’s death occurred at t₁. It is even true that Jill intended to kill Jane at t₁. But another important element in the killing of Jane—Jane’s death—does not occur until t₅. Though Jill shot Jane at t₁, Jill’s killing Jane depends on whether Jane dies.⁴⁸ Given this,

⁴⁴ This is not quite right, since even if one thinks that all individuals that need to be atoned for must exist, one is not forced to be an eternalist. One can posit, for instance, a modified shrinking block which stops shrinking once humans begin to exist. But this is an odd view, and one I cannot see anyone wanting. Hence it is easier, as noted above, to simply limit the discussion to presentism and eternalism. ⁴⁵ For sake of simplicity, I had assumed Christ’s death had occurred at one moment. But this does not seem to be the case—deaths can be long, drawn-out affairs (and one might expect a particularly painful death that is to do atoning work for all humans for all time to take at least a little while—as it turns out, around three hours). One might think that this instantly favors nonpresentist views, since presentism appears to have a difficult time accounting for events that last over a period of time. However, this again is simply another manifestation of the problem of cross-time relations and I refer the interested reader to T. Crisp (2005) for a presentist account of how to deal with cross-time relations. ⁴⁶ Thanks to Michael Rea for prompting this discussion. ⁴⁷ This example is taken, with names changed, from Thomson (1971). ⁴⁸ Securing this death turns out to be tricky business. Suppose the laws of nature ensure that Jill’s shooting Jane at t₁ would result in Jane’s death at t₅. We still cannot say that Jill killed Jane

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

193

one of the most plausible answers to the question, “When did Jill kill Jane?” is that “there is no true answer to the question of when A killed B that gives a time-span smaller than the minimal one that includes both A’s shooting of B and the time of B’s death”—at least, this is the answer Thomson herself gives.⁴⁹ Jill’s killing of Jane seems to occur across time, not all at one time. The killing depends on something to do with Jane as well as Jill. The question, then, is what this dependence requires. For Christ’s atoning work, though occurring at the moment(s) of the crucifixion and death, does not seem relegated to those times. Unless one thinks that Christ’s work on the cross necessitates the atonement of the elect,⁵⁰ the completion of atonement somehow depends on us.⁵¹ Those who hold views according to which Christ’s atoning work at the crucifixion somehow necessitates salvation—certain Calvinists, determinists, fatalists, or universalists (as the last group believes all will be saved)—are arguably able to treat Christ’s “all at one time” atoning work in the same manner as Jill and Jane jumping together. While perhaps linguistically awkward, it does not seem incoherent to say that Jill killed Jane at t₁ even though Jane only died at t₅ if the shooting event at t₁ necessitates Jane’s death. After all, Jane can even cry out at t₁, “You’ve killed me!” The question is what happens if we take the necessitation view of an event’s being accomplished: Jill killed Jane at t₁ because it was settled at that time that Jane would die, even though it took Jane until t₅

at t₁—God could decide to obliterate the universe at t₄, and thus Jill could fail to kill Jane. Those who wish to identify the killing of Jane with the shooting that occurs at t₁ must either deny that such miracles can happen or be determinists. ⁴⁹ Thomson (1971), 122. This is, however, by no means the only reasonable answer. Another compelling answer is that terms like ‘killing’ are inexact, and are used, at various times, as terms indicating the success of an intentional action, to demonstrate an agent’s responsibility (e.g., Jane died because of something Jill did), or to draw one’s attention to a certain state of affairs. Each of these other answers opens up further avenues of exploration but examining each one is beyond the scope of this chapter. ⁵⁰ See, for example, the Calvinist doctrine of irresistible grace. ⁵¹ Affirming this dependence need not make one Pelagian nor semi-Pelagian (heretical views according to which, to varying degrees, humans have the ability to see that they need God and accept Christ’s sacrifice on their behalf without divine help). This dependence need not be robust. Consider again the case of Jill’s killing Jane. The killing depends on Jane, but Jane did nothing to actively bring it about (in fact, one may imagine she tried to avoid it).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

194   to die.⁵² So, the killing happened all at one time.⁵³ Mutatis mutandis for Christ’s salvific work: If the event(s) of the crucifixion in some way necessitated the atonement of the elect (whoever they may be), then Christ’s atoning work could have happened all at one time at the moment(s) of the crucifixion.⁵⁴ There is, however, a general theological reason along with the metaphysical and linguistic reasons to think that ultimate atonement is temporally distinct from forgiveness. The doctrine of unlimited atonement is widely accepted (though not universally), and states that the forgiveness of sins that happens on the cross is not just for those who respond favorably and are ultimately reconciled to God, “but also for the sins of the whole world.”⁵⁵ Christ atones without limit for all, regardless of individuals’ responses.⁵⁶ On this view, the reparative work of Christ can take place at one time: The relevant act thus appears to be the forgiveness of sin, or the defeat of principalities and powers generally, et cetera. And if this is the case, then it appears that the act of sins being forgiven happened all at one time, when Christ was on the cross. While forgiveness can happen at one time, the effects of such forgiveness (e.g., individuals’ acceptance of the gift of the cross; the struggle with and elimination of a sinful nature) happen at different points in time. If one distinguishes between the initial act of atonement (reparations of some sort) and the effects of the atonement (e.g., individuals’ acceptance of the gift of the cross), atonement seems to operate in a manner akin to

⁵² There is dispute about what this sort of settling requires—see footnote 49 for some of the difficulty. But whatever it requires, the sort of necessitation which follows from fatalism, determinism, and the like—views which guarantee the outcome in question—will count as settling in this case, at t1, Jane’s future death. After t1, it is unavoidable. ⁵³ Thomson rejects this sort of account (120), in large part because she does not think there is this sort of necessitation—though she thinks the proposal sounds odd even with necessitation. She says the above example is rather the “Hollywood” use of language. There are important questions, too, about what is necessary in order to have this event be settled. ⁵⁴ Though necessitation views may have other sorts of concerns related to the problem of evil. ⁵⁵ 1 John 2:2. The full verse is “and he is the atoning sacrifice for our sins, and not only for ours but also for the sins of the whole world.” Due to verses which reference the Son dying for the whole world (cf. John 3:16‒17), I will not delve into the Calvinist view of limited atonement, according to which Christ only died for the elect. ⁵⁶ Although a universalist can say that there is atonement for all and universal response.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

195

Jill’s killing Jane—and our answers to the above questions should reflect our general solutions to the “time of a killing” problem. At this point, another quick response emerges: Isn’t the presentist who avoids such a necessitating response, as discussed above, done for? The view under consideration is now committed not only to cross-time relations but quantifying over time-spans. If any proper account of Jill’s killing Jane must include the time-span of Jill’s shooting of Jane to the time of Jane’s death, it may appear that we are forced to quantify over the things that exist during these times. And in the case of the atonement, this would mean from Christ’s death to the full atoning effects in the lives of individuals across time. Thus, it appears we are forced to be eternalists. Given our Neo-Quineanism, we either need to ontologically commit to such time-spans or give a proper paraphrase. Here presentists are not without reply. They have several strategies for solving the problem of cross-time relations. Even if one thinks that proper accounts of the temporal location of the event must “include the time-span,” one is not forced to quantify over a set of times, fusion of time slices, or objects from multiple times in so doing. And the problem raised here is not particular to theorizing about the atonement—the above examples are instances of a more general objection to presentism that charges that the view cannot properly account for events, by assuming that a proper account of events must refer to concrete times, time slices, or objects existing at multiple times. One way for the presentist to give such an account is to reduce—the presentist can say that “time-spans” and events properly reduce to a presentist-friendly view of causal interaction of objects. She could escape the difficulty by maintaining that the world itself contains many irreducible past-tensed properties such as being such that Jill shot Jane, being such that Jane suffered for a bit, and being such that Jane died from her gunshot wound from Jill at t₅.⁵⁷ Or she could maintain that times are abstract objects akin to possible worlds, and thus make reference to times and time-spans without committing herself to non-present things.⁵⁸ How this works depends on the particulars of one’s atonement theory, which will be delved into more deeply below. If one finds the above sorts ⁵⁷ See Bigelow (1996).

⁵⁸ See T. Crisp (2007).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

196   of presentist responses unsatisfying, then one appears to simply find presentism itself unsatisfying. In that case, one is not rejecting presentism on any grounds particular to the atonement but is instead making a theory choice due to much more general reasons (discussion of which is beyond the scope of this chapter). The failure, then, is not due to atonement theorizing but to general theoretical difficulties with crosstime relations. In fact, the distinction between the initial act of atonement and the effects of said initial act helps the presentist answer a particular existence problem raised earlier about the existence of the crucifixion and its causal efficacy over time. The passage of time is not all that is required in order for things to be made right in the world and making things right in the world need not require that the death of Christ always exist. Suppose I burn down my neighbor’s house. According to the presentist, while my act of burning down the house no longer exists, there is still a scar on reality that needs to be taken care of that the mere passage of time cannot make better. There are ashes where a house once stood. While my act of sinning no longer exists, the consequences of my sin do. The house must be rebuilt. Similarly, good effects can last over time. The presentist can say that while Christ’s death no longer exists, the effects of his death last forever. And the effects of Christ’s death are needed to make things ultimately right with the world and to ensure that the least amount of evil exists, post-Judgment Day. Thus, again, it is available to the presentist to say that post-Judgment Day, sin and evil cease to exist entirely—the events themselves are erased by the passage of time, while the effects are wiped away by Christ’s work and our repentance.

7. Explicating the Metaphors: ‘Transfer’ Views Given this, it is worth exploring how the effects of sin and evil are “wiped away.” One way of understanding this interaction is in terms of some sort of property transfer between Christ and the relevant person(s). Christ appears to gain a relevant property, while the sinner loses it—or at least, the conditions are provided for the sinner to lose the property in

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

197

question at a later point in time. This is a way of understanding 2 Corinthians 5:21: “For our sake he made him to be sin that knew no sin, so that in him we might become the righteousness of God.” There are many ways this could be understood, so here I must speak in some generality. The property or properties in question may be understood as guilt, unrighteousness, being counted as a sinner, needing to pay a debt, or the like, which are transferred to Christ—the particular properties will depend on the theory in question. Other accounts involve a transfer from Christ to the sinner, perhaps of moral uprightness, perfect obedience, or not being under the control of sin (though Christ will also retain these properties). Making sense of this involves not only a question about cross-time relations but the nature of the properties in question. The nature of the properties themselves will determine what sort of cross-time relation is necessary, and thus our ontological commitments generally. A family of views appear to require a property transfer.⁵⁹ I call these transfer views: Something is transferred between Christ and another party due to (and possibly during) the atoning act. Among the transfer views are satisfaction theory, penal substitution, vicarious punishment, and ransom theory. On satisfaction, penal substitution, and vicarious punishment theories, some property or properties are transferred between Christ and the sinner(s); this property transfer of, for example, guilt resolves the need for punishment or payment. God the Father may also be involved in the transfer as well, depending on the theory. Some sort of payment by Christ is necessary for the removal of the consequences of sin for the individuals who sin. On a ransom view, some relevant property is transferred between the devil and Christ (though humans may be involved in the transfer as well).

⁵⁹ Christus Victor accounts, according to which atonement requires defeat of certain principalities and powers, may also count as property transfer views, depending on how said defeat occurs. Perhaps one event that happens at atonement is that the property having dominion or authority over person p is transferred from the principality or power to Christ. However, both this understanding of Christus Victor (and ransom theory) must answer concerns about Christ’s power and preeminence: It appears that, as God, Christ must already have the properties of having dominion and authority over everything, which includes person p. And as noted before, a common complaint against ransom theory is that it appears to give the devil entirely too much power.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

198   In what follows, I will assume some sort of penal substitution theory for ease of explication: that is, that on the cross, Christ had to pay a penalty for us in order to remove our sin or the consequences of it. How did his payment work? Here again, there are overly simplistic answers which fail and more nuanced (and promising) responses. On one simple conception of property-transfer, we might insist that the property transfer in question itself happens during the atoning act. This sort of property transfer appears to require non-presentism, since it requires the existence of all the objects and properties related to such a property transfer: Both the sinner and the relevant properties must exist in order for those properties (of guilt, unrighteousness, et cetera) to be transferred to Christ on the cross (and to have any relevant properties transferred from Christ to the sinner).⁶⁰ It cannot be the case that Christ only later has the relevant properties transferred to himself in the future when new sinners come into existence and sin, since this violates the condition that Christ be offered only once for the sins of many. Similarly, if eternalism is false, Christ cannot have transferred to himself the sins of those who have already died, since they no longer exist and thus cannot have their properties transferred to Christ. While some properties related to past sinners might exist, they cannot be properly thought to be the properties of non-existent objects. Again, given Neo-Quineanism, there are no objects which do not exist. However, this notion of transfer “during” the event turns out to be unwieldy. Even if the property transfer requires that all parties and properties exist, the mere existence of said parties and properties is not enough to secure said transfer during the moment(s) of Christ’s atoning act(s). One may wonder how, exactly, the properties are to be transferred away from the sinner, given the eternalist picture (where everything always exists). Here, any acceptable account of property transfer must offer an account of change over time. I have thus far been speaking of eternalists as if they all universally endorsed the same account of the past, present,

⁶⁰ Unless one is committed to the extreme mind‒body dualism discussed earlier.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

199

and future and change over time. I will continue, for now, to speak of eternalists in terms of static eternalism, also known as the B-theoretic eternalists.⁶¹ Static eternalists reject a dynamic theory and argue there cannot be any change that is not simply a difference between what properties objects have at different times. For instance, a poker’s having the property of being cold at t₁ and being hot at t₁₀ just is what it takes for there to be property change in the poker from cold to hot. But an overly simplistic view of property transfer appears also to require a dynamic understanding of said transfer, which begs the question against such an eternalist. On a dynamic conception of change, it is possible for an object to simply not have a property anymore, full stop. A particularly strong reading of transfer appears to require noneternalism. A view one might take of property transfer is that of a “property swap,” somehow occurring at one time. But this view appears to be in a double bind. If the properties are always transferred in such a way that the existence of all parties is necessary, surely the transferred properties of redemption must always be there and have always been transferred. There is always redemption—and if that transfer as understood in a strong sense has always happened, the view appears to be contradictory. For, somehow, it is both true that: It is always the case that I am a sinner at t, and it is always not the case that I am a sinner at t; I’m always redeemed! Such an understanding of a property swap requires both a dynamic theory of time and non-presentism, since all relevant parties and properties need to exist at the moment(s) of the initial atoning act. Those who want to endorse a view like this are led back to Lebens and Goldschmidt, and the previous difficulties discussed.⁶² The problem thus appears not

⁶¹ I address the alternative in the later section “Eternalist A-theory to the Rescue?” ⁶² Notably, Lebens and Goldschmidt’s proposed “Ultimate Forgiveness—Amputation” (13) appears to be a transfer view (though, importantly, without the Christian presuppositions). Again, some of the discussion can apply to multiple traditions’ views of atonement. Lebens and Goldschmidt’s view faces fewer problems than the above response and discussion, since identifying Christ’s actions as (an initial component of) the divine atoning supertask poses many problems due to being in time. Lebens and Goldschmidt do not identify the divine supertask with this event and thus avoid the particular problem.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

200   to be with presentism or eternalism, but with this conception of property transfer.⁶³ Given the immense difficulty of making sense of a simplistic “property swap” property-transfer view, we have good reason to think the imagery supporting the view, that is, “God made him [Christ] who had no sin to be sin for us, so that in him we might become the righteousness of God,”⁶⁴ non-literal. If they are to be viable, transfer views should not entail that Christ became a sinner.⁶⁵ After all, Christ cannot literally be made to be sin—first, because persons cannot be changed into events or properties and, second, because divinity and sin are incompatible. Conversely, we do not literally become the righteousness of God. Our understanding of Christ becoming a sinner in our place has certain ontological boundaries—the focus must be on the notion of “in our place” or something (either a property, event, or experience) relevantly close to being a sinner or in our place, rather than a simplistic, literal understanding.⁶⁶ Here we must choose what to prioritize in our theorizing: What is more important, the existence of objects, other properties, or events, or something dynamic? Eternalists are able to say that the transfer happened during the initial atoning event(s)—or, at least, enough for the property to count as transferred over time. Cross-time relations, and thus relations of property transfer, simply hold between objects or events happening at different times. The cross-time event, then, is simply the event of such a relation’s coming to hold. However, there are still some puzzles to be dealt with, regarding identifying that event of a relation’s coming to hold, along with what it ⁶³ The eternalist can sidestep problems related to the property swap view—that is, it’s always being true that I am a sinner and it’s always being true that I’m not seen by God as a sinner—by making a distinction between logical ordering of propositions and properties, even among atemporal truths. But interesting questions remain regarding when to identify our status of being sinners, being forgiven, et cetera. ⁶⁴ 2 Corinthians 5:21. ⁶⁵ The deeply problematic nature of Christ’s becoming a sinner—or having an experience near enough—is discussed in the next section. ⁶⁶ In general, the transfer views seem to focus more on understanding what it means for Christ to be “in our place” or to take our place (in terms of punishment, et cetera). For a focus on being relevantly close to the sinner in the needed (atoning) respect, see the solidarity views discussed in the next section.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

201

is for properties to be transferred during an act.⁶⁷ Suppose my greatgreat-grandmother and I are exactly the same height when each of us is 4 years old.⁶⁸ It is unclear when this event of our “being the same height” takes place.⁶⁹ At both times? Only later when I reach the same height? Or possibly across all the times between our being the same height? Since there is a temporal gap in which nothing relevant seems to hold between my great-great-grandmother’s height and my own, intuitions which tempt one toward a response affirming the event happens across times may expand to include all times whatsoever: Perhaps it is always the case that my great-great-grandmother is this height in 1836 and I that height in 1989. And it is a logical truth that these heights are identical. Or we might deny that there is such an event altogether. At the very least, the puzzle regarding when the relation comes to hold may be dismissed for now for one of two reasons: It is either a particular case of the more general “time of a killing” puzzle discussed in the previous section (and thus we should give an answer consistent with our previous one here) or it is irrelevant, as the relation described involves mere Cambridge changes (or something near enough). Which it is depends on the nature of the properties or relations in question. Thus, it can be set aside for now. One might insist, however, that at least the eternalist has options. Certain plausible, or possibly necessary views, appear unavailable to the presentist. For the presentist is restricted to just one time in which the relation can hold. The presentist cannot immediately appeal to

⁶⁷ There are puzzles about whether there can be a moment or instant of change—that is, a moment during which change actually happens (see Priest (2006)). ⁶⁸ Thanks to Dean Zimmerman for this example. ⁶⁹ This is a motivating concern for Szabo (2006). He cites Sextus Empiricus’ example: “Helen [of Troy] had three husbands” (399). But she didn’t have them all at the same time; she had them successively. So at what time is it true that Helen had three husbands? Szabo gives a view on which events always presently exist but only sometimes occur. The presentist, ironically, may regard this as a Trojan Horse: such a view will likely force the presentist to surrender her NeoQuineanism or her non-fatalist intuitions. One way to escape these problems is to highlight the presentist’s commitment to a dynamic theory of time—but such a non-fatalist commitment to dynamic theories likely involves serious tensing of the sort which would allow her to escape this kind of counting problem altogether. For the presentist who wants atemporal propositions, which may appear to generate the problem, T. Crisp’s (2007) offers a promising route forward.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

202   backward-looking properties, should those properties have causal power.⁷⁰ But this all depends, again, on the nature of the properties transferred. There are many cases of property transfer which are not obviously a problem for the presentist. Certain sorts of property transfers, especially in a legal setting, do not appear overly mysterious: A will, for example, dictates the conditions for transferring certain possessions of mine to someone else upon my death (and perhaps upon other conditions contained within the will), and there are consequences that follow for the recipient. A common complaint about penal substitution theories, however, is that they are morally inappropriate in a way that a will is not.⁷¹ There is nothing prima facie morally wrong about gifting someone my money; there does seem to be—at least in ordinary cases—something morally wrong with taking an earned severe punishment on behalf of someone else. Consider again the will example: While someone will end up with my money, this change seems extrinsic to them—or at least, too close to a mere Cambridge change to rest satisfied that this is the correct account of how the atoning property transfer works. One way to escape the moral concern is to assume that some sort of relevant property connected to being a sinner or deserving punishment is

⁷⁰ Further, doing so will result in fatalist problems, see Rea (2006) and Seymour (2016), especially if one argues that an event that occurs entirely at t can depend for its occurrence on much later events occurring. A presentist might say that the event of a cross-temporal relation coming to hold is restricted entirely to one of the two times. This may be the case, but it does not immediately solve the problem since the relevant properties must be transferred. Dean Zimmerman has suggested to me the following: Suppose a particular painting p was worth one million dollars in 1990. In 2020, I bid on another painting, causing that second painting to be worth twice as much as the first painting p was in 1990. However, this appears to be a mere Cambridge change. But suppose instead that I bid on the first painting in 2020, raising the price of p from one million to two million dollars. I thus cause the painting to be worth twice as much in 2020 as it was in 1990. While my bidding changes a property of p (how much it is currently worth), it is important to note that this need not change anything about the properties of painting p in 1990—to say otherwise is to run afoul of the in/at distinction with worlds and times. Consider a related example with possible worlds: It is possible that I have a child, but that does not mean there is an existing possible child out there in, say, world w with properties. So, the second painting case does not appear to be a true case of property transfer after all. There are further interesting cases of when properties come to be and how they do so, but these will take us to questions addressed in the “Consequence Removal Questions” section’s discussion of Judith Jarvis Thomson. ⁷¹ For a defense of penal substitution, however, see Strabbing (2016).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

203

somehow transferred between the relevant parties. At the very least, Christ represents the sinner in some way as being a bearer of a relevant property. And here, existence questions will arise again: Must the person in question exist in order for this property to be transferred? Here we must remember the Neo-Quinean commitment. We might think this concern can be quickly answered in the way the presentist does in other sorts of cases. Consider other cases of property transfer: Coaches of football teams can gain and lose properties such as being the coach with the most Superbowl wins or even being the first coach of the football franchise team with the most total Superbowl wins. These sorts of property transfers can take place regardless of whether the coaches in question are living or dead.⁷² One might think that Christ’s atoning act(s) and the relevant property transfers can work in this way: If Christ’s representing me (a sinner) is important, and if Christ’s representing me can be determined later (as per the earlier discussion of Judith Jarvis Thomson), then why not think that some later act of mine, such as repentance, would transfer the property backward in the relevant way? My repenting, then, is not unlike how my coaching of a football team could transfer a property to a long-dead person or how my accepting the conditions of a will could transfer riches to me from someone now dead. However, the mechanics of this suggestion are unclear and of the utmost importance. Gaining and losing properties like being the first coach of the football franchise team with the most total Superbowl wins appear to be mere Cambridge changes, or near enough for our purposes. For there is nothing about my winning or a currently existing franchise that looks like it directly relates in the right sort of way to a particular past coach. While a past coach’s efforts and wins might contribute to the current state of affairs and wins of a team, the current state of affairs does not appear to be about the past coach in the right sort of way, since the property involves games won by the team by later coaches. (In fact, there are Neo-Quinean concerns regarding the above way of even framing the case, which will be addressed below.) Here’s one way of seeing the difficulty: Even if we limit discussion to relational properties, some are more substantive or robust than others. ⁷² Thanks to Dean Zimmerman for pressing these examples.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

204   Utilizing less robust properties in certain circumstances sounds, at best, odd. Consider the property of coach with the most lifetime wins. This is a property that some coach has and can lose at a later time—even after death, should another coach surpass one’s total lifetime wins posthumously. But a substantive posthumous transfer of that property (if such a property is indeed substantive) can only go one way: from the deceased football coach to some currently living coach. A substantive property cannot be transferred backward from a living football coach to one who is long dead, since it appears incoherent to say that someone in the past can change in a non-mere-Cambridge way in virtue of something’s happening at a later time. Given the substantivity intuition about the atonement—that it, at the very least, involves something more robust than mere Cambridge change—we have reason to think that such minimal, backward-looking accounts of property transfer are insufficient. The event of the crucifixion is in the past. Transferring a property from a presently existing (or future) sinner back to the crucifixion appears more akin to transferring the property of coach with the most lifetime wins backward to someone.⁷³ For Christ is, on these views, supposed to take something substantive of ours upon himself. At the very least, more must be said about how such backward-looking property transfer is substantive enough.⁷⁴ This sort of change appears akin to mere Cambridge change rather than something substantial in and for the agent herself. And atonement requires, at some point, something substantial for the agent herself. Finally, there are futureoriented difficulties for attempts to apply such accounts of change to

⁷³ Here one might quibble: Suppose the coach currently said to have the most lifetime wins is revealed to have cheated. In that case, couldn’t the property of coach with the most lifetime wins be transferred back to someone else posthumously? I say no: The intuitions which would lead me to revoke the title of “most lifetime wins” from the current seeming bearer also lead me to think that the impostor coach never actually held the property coach with the most lifetime wins in the first place. So, there was no property transfer to begin with. And if such a property could be backwardly transferred to someone long dead, it appears to be a mere Cambridge change. And in which case we must find a better example to illustrate the above. ⁷⁴ More, too, will need to be said about how it avoids fatalist problems. For this understanding of atonement appears to land one immediately into problems addressed in Rea (2006) if one is to avoid the Neo-Quinean concern.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

205

possible future persons, who at the time of the crucifixion have no relevant properties to transfer, which are addressed below. In addition, the presentist has a further, Neo-Quinean difficulty with the above sort of understanding of property transfer. There cannot be things which do not exist. In order to have a property transferred to someone, then, that someone must exist. Most plausible presentist accounts offer something like haecceities or other properties to perform the relevant property-transfer work of the above examples.⁷⁵ Presentists who are committed to a transfer view must have an appropriate ontological stand-in. The above applies more generally to views that require the existence of the relevant acts in order for payment to be made for them. If the negative consequences (penal or otherwise) of all of one’s specific sins and all the specific sins of everyone who has existed or will exist were taken care of by Christ on the cross, presentism appears to falter. If presentism is true, it looks like Christ cannot have specifically taken on any particular sin of mine at the moment of the crucifixion, since neither I nor that sin existed yet. Transfer views thus must address both important questions about how robust this transfer must be and what is ontologically required for a transfer of such robustness. According to transfer views, Christ takes something of ours upon himself (and perhaps vice versa). Is this transfer specific or general? The answer given here will constrain our temporal ontology, and vice versa, and leads us to a general taxonomy of transfer views. Consider, again, a penal substitution view and the transfer therein: Christ pays for my sins. This payment could be specific, applying a particular payment for each of my particular sins, or it could be general. Is there a specific, say, “dollar amount” accounting for each individual sin, with separate accounts? Or is there a spiritual “blank check”? The transfer of payment could thus work in two main ways: a tokentoken view and a type-token view.⁷⁶ On token-token views, for every ⁷⁵ See T. Crisp (2007). ⁷⁶ While the focus here is on moral transgression, the following discussion is also useful for Christus Victor accounts, according to which atonement requires defeat of certain principalities and powers. A similar type-token versus token-token discussion regarding the metaphysics of

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

206   individual sin that someone commits, Christ paid for that particular sin on the cross.⁷⁷ This view is captured nicely by the sentiment “Jesus died for your specific sins” or after an act of sin when a friend says, “Jesus paid particularly for that.” Type-token views, on the other hand, do not require Christ to pay for each and every particular sin token that was committed—rather, on this understanding it is enough for Christ to have paid for each sin type (e.g., gluttony, adultery, lying, or perhaps even something much more general . . . ).⁷⁸ Both presentists and eternalists can easily accommodate a type-token view of sin payment. But presentists appear to have a problem with token-token views that eternalists do not. Presentists are unable to accept a token-token view according to which Christ’s payment of the penalty (or substitution) once on the cross for every single exact sin past, present, and future requires the existence of those acts. However, while the presentist cannot endorse token-token views which require existence of the individual and sins in question, the presentist can endorse a sort of token-token view. Token-token views, as it turns out, are of a broad family. It all depends on the nature of the property token in question. If God has simple foreknowledge or middle knowledge of each of the individual sins each creature will commit, then God could transfer the penalty of each of those sins (or the relevant property that did or will attach to the creature—or perhaps just properties similar enough to ones that did or will exist) to Christ on the cross. If God knew what you would do, nothing would seem to prevent him from assigning to Christ the atonement will also arise. For instance, does Christ’s victory through his death and resurrection require a specific defeat of particular principalities and powers? Are the defeated powers and principalities such that they all existed at the key moments of the atonement and were defeated at that time? If not, was a type-token victory, against any possible kind of principality or power enough? If the proper account requires a token-token view, did Christ have to defeat each token specifically, or did what he accomplished wipe out any potentially existing principality or power that could ever arise? How does this defeat fit with Christ’s sacrifice as given once and efficacious for all? Similar discussions regarding Christ’s sufferings and their completion (see Colossians 1:24) can also be applied to this framework. ⁷⁷ The atoning event on the cross is, of course, a token. Transfer views take this token event of Christ’s sacrifice to involve something additional in terms of properties—the question now is whether this additional work in terms of properties transferred is to be understood in terms of types or tokens. ⁷⁸ On this alternative, tokens are not directly paid for, but only indirectly paid for. Thanks to Dean Zimmerman for this point.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 27/1/2022, SPi

      

207

penalty for your sin, along with the penalties of the sins of all who do, did, or will exist. The existence of sin x thus does not seem required for the proper application of, say, penal consequences for x. Another option available to presentists is to say that instead of accounting for each individual sin that does end up happening, Christ suffered the negative effects of every sin that might be committed (albeit in the form of experience of simulacra, the property of guilt, or punishment for sin). This particular type of answer also allows the presentist to deny that nonpresent objects and events need to exist in order for atonement to work. If Christ dies for every possible token of sin, then surely the ones you commit have been atoned for and we can also truly say that “Jesus died for your sin.” Understanding Christ’s payment in terms of such broad payment may, in fact, appear beneficial according to those who hold the view of unlimited atonement, wherein Christ atones for all sinners. Given the above, the following sin payment taxonomy can be established:⁷⁹ (6)

Type-Token Views

(7)

Token-Token Views (7a) Existence Requirement Views: All of our sins exist, and Christ pays for each and every one of them—no more and no less. (7b)

Pay It Backward and Forward View: Though they may not have existed at the moment of payment (being either past or future), God knew all of the sins that were being, had been, or would be committed—and so Christ’s death pays for each one of them exactly.

(7c) Saturation Views: God knows all of the possible ways that any creature that could ever exist could ever sin and pays the price for all of those on the cross, just in case. (7c*) Future-Oriented Saturation: Exact payment was given for past sins and those current to the time of the cross (since those occurred and were known), but saturation payment was given for any possible future sins. ⁷⁹ The taxonomies are stated in terms of payment for sin, but they can also be stated according to other conceptions of guilt and reparations (such as satisfaction or vicarious punishment theories). Language is limited to penal substitution only for ease of explication.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

208   With the exception of (7a), presentists are able to adopt token-token views. But there are two problems with the presentist-friendly tokentoken accounts. First, it might be objectionably unjust to have such “overkill” accounts of the atonement as (7c) and (7c*), where Christ suffers unnecessarily—after all, on this picture, not all of Christ’s suffering was efficacious, since every sin token he died for does not match up with an actual sin that needs forgiveness. However, it seems that there is some over-payment built into most commonly accepted accounts of the atonement—there is not a one-to-one correspondence between sins that Christ died for and sins that are forgiven in a way that results in ultimate reconciliation. Unless one is a universalist (and thinks that all will be saved) or a Calvinist who believes in limited atonement, one must affirm that it is not the case that all sins paid for result in ultimate reconciliation. However, there seems to be a difference between paying for something that ultimately did not need to be paid for (e.g., what happens on the saturation views) and paying for something that needed to be paid for, but where the gift was rejected. Second, one might question whether the presentist modifications of token-token accounts are acceptable on the grounds that they do not appear to properly take away my sin. Why should I care if a duplicate of my sin is removed or paid for? The presentist looks like she has a grounding problem—if the sinner no longer exists, how can their sin be paid for? Like other grounding problems and presentism, it is not clear that this is insurmountable, but it requires more discussion.⁸⁰ Here there is possibly tension between the two commitments of NeoQuineanism and the substantivity intuition. There is thus more fruitful work to be done. However, there is enough to generate a useful taxonomy for how such transfer views can work and said taxonomy can be fruitfully applied to other accounts of atonement, as will be shown below. And here we can note that the eternalist has the advantage concerning the wider range of moves she can make regarding how atonement happens over time.

⁸⁰ Additionally, the presentist must provide an explanation of how the work accomplished via the token-token view substitution does not pass out of existence.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

209

8. Solidarity Views and Stump’s Account Similar temporal questions and puzzles also arise for accounts which appear quite disparate. When we speak of the removal of sin and evil, there is the ontological removal of sin or sin events as such and then there is the removal of the consequences of that sin or evil: the weight, guilt, or psychic stain of evil and sin. It is a further question what is ontologically and theologically required for the removal of such a weight, guilt, or stain. Transfer views explore this in terms of property transfer. But another general approach to atonement theorizing—what I call solidarity views—faces similar sorts of temporal puzzles as those addressed previously. Given this, the earlier proffered taxonomy can, with some minor adjustments, be utilized in our understanding and debating these theories as well. The framework offered is thus broadly useful. Solidarity views aim to provide a robust understanding of how Christ is able to sympathize with our weakness—Christ has seen and experienced life with us to the full extent that a sinless being can. By undergoing the crucifixion, death, and horrors generally, Christ removes an ontological obstacle preventing our union with God. The ontological obstacle is not simply due to our sins—there is something about our suffering and our response to suffering that is in need of deep repair that we cannot do ourselves. Christ has important de se knowledge of our guilt and shame—in some theologically appropriate way, Christ knows what it is like to have such guilt and shame—and suffers in solidarity with us. While views such as penal substitution are concerned with reparations for sin, solidarity accounts provide an alternate way to make sense of the Hebrews 9:28a reference to Christ’s being “offered once to bear the sins of many,” focusing on his bearing of the sins, rather than the offering. What is it to bear sin? Certain transfer views assume that bearing sin requires some sort of payment, punishment, or reparation for sin. But this “bearing” might instead be some sort of mental sharing of the psychic weight of sin, which in turn allows sinners to turn toward God and away from their guilt and shame. Of course, certain sin payment and solidarity accounts need not be mutually exclusive. Due to the wide variety of metaphors and images used to describe what Christ’s

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

210   crucifixion and death accomplished, one might think more than one account is needed. According to solidarity accounts, the ontological gulf preventing our union with God is due in part to something previously lacking; God’s solidarity with us provided us something we needed. Perhaps, as Eleonore Stump argues (2018), the issue lies in our inability to overcome our own guilt, shame, and the psychic stains of our sin. Christ’s empathetic mental engagement with us offers us a way for this psychic weight of the stain of sin to be overcome. Or perhaps Christ defeats, redeems, and overcomes horrendous evils, as Marilyn McCord Adams argued (in her 1999 book and elsewhere). On this account, Adams famously suggested that we—sinful humans—might have a complaint or lament for having undergone suffering which needs to be addressed (2006). Through his suffering, Christ identifies and has solidarity with victims of evil and addresses such complaint and lament.⁸¹ Solidarity views thus offer more than mere moral exemplar views, although moral exemplar views do not entirely escape the previous temporal puzzles discussed.⁸² (For example: How was Christ’s moral example to be followed or of benefit for those who lived and died preIncarnation?) An important part of a solidarity view is our recognizing, interacting with, or somehow participating in something with Christ. What is metaphysically required in order to have such recognition, interaction, and the like? Here again we encounter temporal puzzles,

⁸¹ I do not mean to suggest that Adams’s and Stump’s approaches are strictly opposed. Adams and Stump agree that part of our being at one with God involves redemption where suffering is ultimately made into something for the benefit of the individual undergoing the particular suffering. See Stump (2010, especially 393ff.). Adams goes further, particularly due to her focus on horrendous evils: “evils the participation in which (that is, the doing or suffering of which) constitutes prima facie reason to doubt whether the participant’s life could (given their inclusion in it) be a great good to him/her on the whole” (1999, 26). For Adams, Christ’s defeat of horrors is not merely due to the fact that someone’s life turns out to be better on the whole; the evil is overcome in such a way that it becomes an integral part of an organic unity of overall goodness. The evil is organically part of a good whole and cannot be eliminated from that good whole (28‒9). For both Adams and Stump, each individual sufferer and sinner will think that their existence is, on the whole, good. I am not the first to note the connection with solidarity; Rea discusses Adams’s view in these terms in his (2019). Julian of Norwich likely also held some sort of solidarity view. ⁸² Indeed, the proponents of solidarity views mentioned take pains to avoid the unorthodox implications of a mere moral exemplar theory, as in Stump (2018, 167): “A position often attributed to Abelard and commonly considered heretical is the view that the point of Christ’s

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

211

which are heightened to the extent we think such interaction happens during Christ’s passion. For this reason, I will focus on Stump in what follows. (Adams’s view will face questions and puzzles regarding when our lives are made right, et cetera: that is, questions akin to those asked in the “Consequence Removal Questions” section.) Stump has offered a rich, appealing account of solidarity that provides the most metaphysical machinery, focusing particularly on what happens during Christ’s passion. Thus, it raises the most immediate temporal puzzles, which can be addressed within the current framework. Stump presents a “mind-reading” account.⁸³ According to Stump, in Christ’s suffering on the cross, Christ’s human psyche is somehow joined with the psyches of every human being, then at one and the same time Christ will mind-read the mental states found in all the terribly evil human acts humans have ever committed . . . In this condition, Christ will have in his psyche a simulacrum of the stains of all the evil ever thought or done, without having any evil acts of his own.⁸⁴

Part of the work of Christ on the cross is that he has second-personal shared mental attention, that is, shared focus between individuals, with every creature that has or will exist, in order to be unified with us. This attention is robust and requires the existence of all human minds and Christ’s attention to them during his passion: “ . . . the psyches of all human beings pour into the human mind of Christ, which is open to them in his suffering and dying. This openness on Christ’s part during

passion and death is to teach by example what real love is. If this supposedly Abelardian position marks one end of the spectrum for interpretations of the doctrine of the atonement, the account I am delineating here marks the other end of the spectrum. On the Abelardian view, the point of Christ’s passion is to set an example for human beings. On the account argued for here, the point of Christ’s passion is to provide for human beings a metaphysical analogue of the union of the persons of the Trinity, in which each person is in the other.” ⁸³ The reader should note that Stump’s account offers far more than just the shared attention material mentioned here. There is a deep richness in her account concerning the guilt and shame of sinners that needs to be removed. But if one finds even this tidbit of Stump’s view attractive, one will have a difficult time being a presentist and adopting her view wholesale. ⁸⁴ Stump (2012a), formulated exactly in Rea (2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

212   his crucifixion to allow all human psyches is his contribution of what is needed for mutual indwelling between God and human persons” (173‒4). Thus, Stump’s view appears to rule out almost all versions of presentism, requiring either eternalism or an extreme form of mind‒body dualism, wherein all sinners not only exist at the time of the cross but are conscious to a robust enough level that they can be joined with Christ’s human psyche.⁸⁵ In order for Christ to join with the psyche of every human being, every human being must exist and have a psyche to which Christ can be joined. However, a presentist who wants to avoid commitment to extreme mind‒body dualism is not left without resources to try to account for something like Stump’s view. Examining the available options requires a brief exploration of the second-personal experiences and shared attention which are central to Stump’s account. Second-personal experiences are experiences whereby one has direct experience of another person as a person. This often involves sharing attention—for example, two people mutually gazing into each other’s eyes. Such experiences show how persons can share attention with each other and be unified. Stump identifies the central obstacles to atonement with God as guilt (when we think that others do not desire our good) and shame (when we think that others reject not our good, but our very selves), which we experience because of our sin.⁸⁶ Christ’s secondpersonal experience of us and our psyches helps overcome the guilt and shame we experience due to sin: Christ’s joining his psyche with ours demonstrates that he does desire our good and wants to be united with us. Stump’s conditions for second-personal experience might seem to rule out any second-personal experience of one who no longer or does not yet exist, since “One person Paula has a second-person experience of another person Jerome only if (1) Paula is aware of Jerome as a person (call the relation Paula has to Jerome in this condition ‘personal interaction’), (2) Paula’s personal interaction with Jerome is of a direct and ⁸⁵ Stump herself thinks her view is consistent with presentism due to her position on divine eternity. This will be discussed in the following section. For now, it is sufficient to note that there is at least a prima facie difficulty. ⁸⁶ Stump (2012b), 130 & (2010), 354.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

213

immediate sort, and (3) Jerome is conscious.”⁸⁷ But Stump admits that what counts as being aware of someone “as a person” is not clear cut.⁸⁸ Stump says that thinking about another person is not enough to be aware of someone as a person, but interaction via email is. So, it is possible to be temporally removed in certain respects from someone with whom you have personal interaction of a direct and immediate sort. One can also have a second-person experience via a letter read after the author’s death. Stump admits that her consciousness requirement (i.e., (3) above), which seemed incredibly strict, does not require that the two people sharing personal interaction be conscious at the same time. What matters is that the author was conscious at the time they wrote the letter. If reading a letter after someone’s death can count as a proper secondpersonal experience, then our lack of conscious awareness or nonexistence at a particular time should not be an insurmountable obstacle to Christ’s having personal interaction with us on the cross. It depends on what is needed in order to have a personal interaction that is “of a direct and immediate sort.” Clearly, not every case of mediation of mental experience is unacceptable—direct and immediate awareness of another does not preclude communication mediated through written and spoken word. The question, then, is whether the mediation in question allows for the sort of awareness needed for second-personal experience (and is thus similar to email or a letter), or whether it is too far removed, like cases of third-party communication regarding a particular person. The presentist is thus able to account for second-personal awareness over time generally, without any need for extreme mind‒body dualism. However, second-personal awareness is only one of the conditions necessary for Stump’s account: shared attention is also required. It seems plausible to think that sharing attention requires the existence of all relevant parties—given Neo-Quineanism, if there is a relation, all relata must exist. Here, too, though, there is some aid from Stump’s offered metaphysics. While Christ shares attention with us and our experiences, such sharing is not always of a direct and immediate sort. For there are certain ⁸⁷ Stump (2010), 75‒6.

⁸⁸ Stump (2010), ch. 4, footnotes 72‒4 for the following.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

214   elements of our experience, such as knowing directly what it is like both to sin and to want to be the person who commits such a sin, that Christ cannot know directly. For Christ is without sin and cannot know this directly. But this is no obstacle to Christ’s having the right sort of awareness of and attention to those particular pains. Christ is able to have the necessary solidarity with and attention to the stain of sin and related desires via mental simulacra. Christ, being and remaining sinless, experiences simulacra of the stains of all evil and this is enough to do the work.⁸⁹ Christ cannot experience the stain of sin itself, but Christ can experience it in a manner similar to experiences of empathetic pain.⁹⁰ If simulacra are sometimes sufficient for certain kinds of sharing, then perhaps a presentist can make broader use of simulacra generally in order to have an account in the spirit of Stump’s. This could be done in the form of simulacra of either persons themselves or types or tokens of mental states. Here is one such presentist gloss: Experiences of mere simulacra take the place of mind-reading. Instead of sharing joint attention, Christ experiences simulations of all of the mental states of all human persons who did, do, or will exist. In that way, Christ knows what it is like to be in the position of each individual, even though joint, shared attention at the time of the crucifixion would be impossible.⁹¹ Perhaps in this way Christ mentally “joins” us in our suffering. And perhaps this sort of experience would allow for a later “meeting of minds” between Christ and other humans, as Christ would have at one time paid mental attention to the same kind of things each human experienced. One might rightly question whether this is enough.⁹² Simulacra of persons or general experiential content are certainly not as appealingly ⁸⁹ Though one must note that these simulacra comes about, for Stump, as a direct result of Christ’s mind-reading the mental states involved in all evil human acts. ⁹⁰ Stump (2018), 161. ⁹¹ This can be further nuanced to account for all the possible mental states of those who might exist. ⁹² Kenny Boyce suggested in conversation that perhaps the presentist can say that the second person of the Trinity, throughout time, shares mental attention with those beings that exist, as they exist, so that by the end of time the Son will have experienced what it is like to be each human being. However, this violates the Hebrews 9:28 requirement that Christ’s work on the cross be an offering that occurs only once.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 27/1/2022, SPi

      

215

robust as sharing attention with persons themselves. But it is hard to see how such mediation would be immediately unacceptable, given the earlier account of Christ’s solidarity vis-à-vis the stain of sin. And the creation of simulacra of the mental states, such as of those who died precrucifixion, would not be the result of their own efforts but would instead be the work of God. The particular nature of such simulacrum would depend on one’s particular account. But perhaps one thinks the simulacra of the events Christ attends to must have already happened—otherwise, our conscious experience is not suitably responsible for the simulacra to which Christ attends secondpersonally. After all, letters must first be authored before one is able to have a direct second-personal experience via reading them. However, if God has simple foreknowledge or middle knowledge, God knows what we will do and experience. Given this, why not think that a simulacrum of a mental experience to come could function in the same way as reading a letter from the past? If mediation via letter is acceptable, it is hard to see why such simulacra of future mental states would be unacceptable. Both sorts of simulacra would come from the same source—God—and in each case God knows exactly what the relevant experiences were or will be due to what the agent herself is like and chooses. Thus, robust shared attention between Christ on the cross and all persons throughout time is not necessary in order for Christ to gain a deep understanding of each person and demonstrate that he desires their best and wants to be united to them—the heart of Stump’s account. If such simulacra accounts are acceptable, a taxonomy of secondpersonal experience and interaction parallel to the transfer view’s taxonomy is available (with the same basic moves available to the presentist, mutatis mutandis): (8)

Type-Token Views of Personal Experience: Christ experiences, on the cross, certain general kinds of conscious experience—either via those who currently existed at that time or simulacra—to appropriately understand, for every particular token experience, what it is like to have undergone that token experience.

(9)

Token-Token Views of Personal Experience

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 27/1/2022, SPi

216   (9a) Existence Requirement Token Views: All of our conscious experiences exist, and, on the cross, Christ’s human psyche is joined with the psyche of every human, with Christ mindreading the mental states of every human and experiencing a simulacrum of all the evil that exists—no more, no less. (9b)

“Read It” Backward and Forward Views: Though they may not have existed at the moment of mind-reading (being either wholly past or future), God knew all of the mental states that existed at the moment(s) of the crucifixion, had existed, or would exist—and so Christ mind-reads the mental states concurrent with the cross and experiences a simulacrum of all other mental states and of all the evil that existed at the cross, had existed, or would exist.

(9c) Saturation Views: God knows all of the possible experiences any creature that could ever exist could ever have, and so Christ experiences simulacra of those and all possible evils on the cross, just in case. (9c*)

Future-Oriented Saturation: Christ experiences simulacra of all mental states that had existed, mind-reads mental states concurrent with the cross, and experiences simulacra of all possible future mental states (along with experiencing simulacra of past evil, evil present at the time of the cross, and all possible future evil).

One issue appears to be whether Christ can be properly considered to be participating in personal interaction on (9c) and (9c*), since Christ experiences simulacra that end up having no human person who ever identifies with them and claims them as simulacra of her own experience. By itself, this doesn’t appear to be a problem: Personal interactions require another and sometimes attempts fail. Some letters, for example, never reach a recipient. Additionally, the presentist inherits a grounding problem similar to the grounding problem regarding certain “overkill” transfer views—it appears that Christ does not properly share my experience, but instead experiences a qualitative duplicate.⁹³ And it is important to Stump that ⁹³ Importantly, in Stump’s account, the simulacrum of the stain of evil that Christ has comes about as a direct result of Christ’s mind-reading. Thus, the simulacrum Christ has does not detach or float free from actual human experience.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

217

the human psyche of Christ is, on the cross, joined to the psyche of every human.⁹⁴ This is what makes it so that Christ is truly sharing in our experience (though not completely, as Christ can only experience a simulacrum of the stain of evil). While one can have second-personal experience of someone after they have ceased to exist, certain shared experiences require robust consciousness and awareness of the other person.⁹⁵ There is a way of putting pressure on this requirement, however, utilizing the framework Stump has offered. Suppose one has a remote conversation, perhaps by video, with someone who continually goes into and out of existence.⁹⁶ On a strict reading of shared attention, this will not count as a case of shared, joint attention between two parties. But it seems we have something near enough for a good, shared conversation and attention therein. (One can even heighten the case by having the parties both pop into and out of existence, never existing at the same time.) Cases of such temporally staggered attention may be enough to allow us to reap the robust benefits of Stump’s view. For while Christ is aware of all persons and their suffering during the crucifixion, all existing persons are not so aware of him—there is at least not mutual awareness of the shared attention. And said awareness of Christ’s experiencing with us is what psychologically allows us to overcome our guilt and shame. This psychological awareness and overcoming (of course, by God’s grace) can happen at different moments in time. All that is needed for our intrinsic change, including conscious awareness, can happen at later moments in time.⁹⁷ Here Stump may complain that I am pressing distinctions she avoids on her account. Stump thinks she can have both presentism and true shared attention at the moments of the crucifixion, thus escaping the above grounding problems. Additionally, she thinks that there is the right sort of mutual awareness of the shared attention, even if my ⁹⁴ Stump (2012a) & Rea (2019). ⁹⁵ Stump (2010), ch. 6. ⁹⁶ Thanks to Dean Zimmerman for this example. ⁹⁷ Notably, even the most robust notions of intrinsic change due to atonement, such as being joined to another object such as redeemed humanity (O. Crisp, 2009) do not require my being immediately aware of such intrinsic change or joining.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

218   awareness of the sharing occurs millennia after Christ’s passion. My speaking of “later moments” is thus inapt. Stump thinks she can accomplish all of this by utilizing her account of divine eternity, according to which earlier and later moments in our timeline are simultaneous to God. If this is the case, the above taxonomy, while interesting, is not necessary for making her account consistent with a variety of temporal views. I now turn to this account.

9. Divine Eternity Stump & Kretzmann present an account of divine and human temporality wherein God, as eternal, has “the complete possession all at once of illimitable life.”⁹⁸ This illimitable life is a timeless mode of duration, such that everything (including all events in human history) are present to God. In addition to this limitless duration, God does not experience temporal succession, wherein events are either past, present, or future or even occur temporally earlier or later than each other.⁹⁹ From God’s perspective, my writing this chapter, your reading this sentence, and Washington’s crossing the Delaware are all simultaneous. Stump and Kretzmann label this simultaneity for God “‘ET-simultaneity’, for ‘simultaneity between what is eternal and what is temporal’.”¹⁰⁰ Humans, on the other hand, experience temporal succession—events are such that they are future, become present, and then move into the past. From the non-eternal, human perspective, you are reading this sentence in the present, and my writing it and Washington’s crossing the Delaware are in the past. According to Stump, “The logic of the doctrine of eternity has the result that every moment of time, as that moment is now in time, is ET simultaneous with the eternal life of God.”¹⁰¹ The difference between ⁹⁸ Stump & Kretzmann (1981), 431. ⁹⁹ Stump (2018), 119. She writes, “Because an eternal God cannot have succession in his life, neither of the series (the so-called ‘A series’ or ‘B series’) characteristic of time can apply to God’s life or to God’s relations with other things. That is, nothing in God’s life can be past or future with respect to anything else, either in God’s life or in time; and, similarly, nothing in God’s life can be earlier or later than anything else either” (119). ¹⁰⁰ Stump (2018), 119. ¹⁰¹ Stump (2018), 119, emphasis Stump.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

219

divine and human temporal experience is analogous to the difference in experience of the spatial dimension that would exist between someone living in only two dimensions (a “flatlander”) and someone who lives in three.¹⁰² The nature of reality does not change, though the perspectives are different. A flatlander moving along a line would regard some of the line as behind her and some of the line as before her. But the threedimensional person would see the whole line before her all at once. Given divine eternity, God’s present awareness of the mental states of all those who exist, existed, or will exist need not be mediated via one point on the timeline. God is thus able to have interactions with multiple humans existing at different times. According to the divine eternity account, what is present depends on whether one is divine or merely human. There is no single, objective answer to the question “What is present?” since what is present depends on one’s mode of duration.¹⁰³ And there are two modes of duration— timeless and temporal. Admitting that “present” is a relative matter, however, does not make it so that one has introduced a theory of time over and above presentism and eternalism.¹⁰⁴ Divine eternity appears to require an eternalist spacetime block universe, albeit in terms which describe God’s illimitable life in relation to it. For God sees everything: from our perspective, everything that is, was, or will be. While our sins may be in our past, they are always present to God. Everything that ever happened or will happen is present to God. And if something is present to God, it must exist, full stop—especially given our Neo-Quinean commitments. Someone who wants to resist the move from divine eternity to eternalism must surrender a commitment to Neo-Quineanism and the univocal use of the universal and existential quantifiers. One could maintain that if an object O is present to God, it must exist in eternity. But it may nevertheless not exist—because “exist” in our mouths, in normal

¹⁰² Niven (1967) & Stump (2018), 120. ¹⁰³ Stump explicitly appeals to “modes of duration” in describing the theory (119). ¹⁰⁴ Other (eternalist) approaches that argue for multiple, differing true tense claims according to which there is not a single objective standpoint by which the question “What is present?” can be answered include Fine’s fragmentalism (2006).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

220   contexts, is restricted in scope to what presently exists. To use this to resist eternalism, one must say that there are two equally good quantifier meanings which we can shift between, and there is nothing more real about the things in a particular domain. But this position pays an ontological cost in addition to those which accompany a surrender of Neo-Quineanism. For this position maintains that God, in some important sense, is not the final word or arbiter of reality or what exists. Many theists will want to maintain that, if push comes to shove, God’s perspective or quantifier should take ontological privilege. And given the ontological gulf required to make this position work, it is also unclear how we could refer to those things in the domain of God’s quantifier and vice versa. The other way of understanding the difference in the use of “exists,” between normal contexts and those of the ontology room, is simply in terms of restricting the domain of the quantifier in ordinary contexts. All of these things exist, but only some of them are present. And we normally only refer to what is present. But to say this, as Sider (2001) points out,¹⁰⁵ is simply to hold an eternalist view. We can speak of the sum total of things in the domain—which includes what is present to God, that is, all objects, events, and times—but restrict in other contexts. By my lights, the divine eternity account thus turns out to be a particular version of eternalism, even though Stump does not think it is. Thus, any account of atonement which utilizes divine eternity must surrender either presentism or Neo-Quineanism.¹⁰⁶

¹⁰⁵ pp. 15ff. Stump (2018) resists this move, because she conflates a non-dynamic fourdimensional block (which is frequently labeled ‘B-theoretic eternalism’) with succession. But succession and temporal experience are different than temporal ontology simpliciter, for example, see McTaggart (1998). An example Stump uses involving a yellow point versus an illuminated line of “now” (120) might require quantifier variance or fragmentalism, if there’s a commitment to there being two objective, differing “nows.” ¹⁰⁶ There remain interesting questions about the nature of the Son’s experience of time as an incarnate being on earth. The divine eternity account might be particularly suited to helping address these issues, along with exploring how temporal experience may or may not be a fundamental aspect of human experience. Unpacking Christ’s relationship to time will help clarify our temporal commitments. Those who do not utilize divine eternity—such as presentists—might have other advantages. Perhaps this is one aspect of the Incarnation that is less mysterious, as the Son’s experience of time is the same as ours: There is simply no other way for it to be, post-creation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

221

10. Eternalist A-Theory to the Rescue? In raising the question of what counts as present, the divine eternity account helps bring to the fore a crucial, unargued-for assumption that has been lurking throughout the discussion. Thus far I have assumed (for ease of explication) that since eternalism maintains that the objects and events from every time exist, there was no privileged time or set of times. While there exists a timeline in which events are ordered from earlier to later (called a B-series), talk about the past, present, or future is not objective. The term “present” is as much an indexical as the term “here.” Thus, I assumed that, according to the eternalist, all objects and events enjoy the same ontological status. But not all eternalists are B-theorists. While the above assumption is adopted by the majority of eternalists, some think that the existence of all objects and events from all times does not mean that these objects and events exist just as we do. For we have a property that grants us an objective, ontological privilege that objects at other times do not have— being present. These eternalists posit an A-series: “a series of positions that runs from the far past to the near past to the present, and then from the present to the near future to the far future, or conversely.”¹⁰⁷ On an eternalist A-theory, all objects, events, and times exist—but time is also dynamic, so that objects and times are objectively future, become objectively present, and then pass into the objective past. While all objects exist, only some are objectively present. The present can be envisioned as a sort of “spotlight” that moves across spacetime, “lighting” things up as it goes along. Objects, events, and times that were “lit up” are past, while those yet to be “lit” are future.¹⁰⁸ The A-theory allows for the ontological privileging of some objects and events over others, while eternalism guarantees the existence of all objects that are, were, and ever will be. Thus, according to an eternalist A-theory, while past sins exist, they do not have the same ontological status as what is present. Non-present sins exist, and thus can serve as a

¹⁰⁷ McTaggart (1998), 68. Broad (1923) articulates, but rejects, this view. Contemporary supporters of eternalist A-theory include Cameron (2015) and Sullivan (2012). ¹⁰⁸ Zimmerman (2008), 213.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

222   ground for, say, an existence requirement view of sin payment or an existence requirement solidarity view. Exact sins can be forgiven and paid for; exact conscious states can be utilized. But the addition of the A-theory makes it so that the continued existence of those sins need not be worrisome—sins in the past are not as real as the present.¹⁰⁹ Unlike B-theoretic eternalism, according to which past moments of sin and suffering have the same ontological status as the joy of the saints in heaven, eternalist A-theorists say that while past sin and suffering exist, they have lesser ontological status than the joys experienced in heaven will have, come the eschaton. The continued existence of past sins is not as great a blight on reality as the continually existing sins of a traditional B-theoretic eternalism. Eternalist A-theory thus attempts to get the best of both worlds— something intuitively near enough to an Evil Elimination approach (since the ontological status of sins and evil is somehow lessened, though they are not completely eradicated) and the ability to adopt existence requirement views of atonement (since, for example, each sin exists and thus is able to be paid for). What an eternalist A-theory is able to accomplish, however, depends on the ontological specifics of the privileged present. What is the benefit, or what follows from, being objectively present? One approach is simply to assume that nothing more follows than simply being objectively present. Time is fundamentally dynamic, and the only ontological difference between times (and the objects and events therein) is whether they are objectively past, present, or future.¹¹⁰ However, if this is the case, the benefit(s) of being objectively present are at best unclear. If there is nothing more to dynamic time, than Atheory eternalism appears to have no advantage over the B-theoretic approach—especially regarding the Evil Elimination approach.

¹⁰⁹ Thanks to Meghan Sullivan for suggesting this view. Eternalist A-theorists differ with respect to exactly how past and future objects and events are not as real as those that are present. The language of “not as real” is borrowed and need not suggest degrees of being. ¹¹⁰ This is the view of the growing block that Broad (1923) endorsed. The past is just as real as the present; it is simply not present (66). This view of dynamic time can also be adopted by the eternalist.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

223

This view also inherits a problem that David Braddon-Mitchell (2004) and Trenton Merricks (2006) raise for the growing block theory of time: How do I know that I am objectively present? On the above A-theory, it subjectively seems to me that I am in the objective present, but it would seem that way to me regardless of whether I am. I think to myself, “I am sitting here at the present time.” Or pick an intensely felt moment of my suffering or sinning. I think, “I am suffering now.” There is no ontological difference between these conscious states and those in the objective present, except that the earlier ones lack the primitive property being present.¹¹¹ When these moments have objectively passed, I will still be thinking these things to myself—at these past times, it will seem to me that I am in the present. Even more problematic for our purposes, I will still be vividly and intensely feeling the experience of my suffering or sin. Thus, this view cannot allow for Evil Elimination, and the concerns that motivate the desire for Evil Elimination remain. The growing block theorist is not left without a reply to Merricks, nor, mutatis mutandis, is the eternalist A-theorist. Peter Forrest (2004) proposes the Dead Past Hypothesis: Conscious experiences and certain kinds of events (like an explosion) are activities and require objective presentness (359). Forrest uses this to defend the growing block theory and says that being on the present edge of reality is what allows for activity, but the Dead Past Hypothesis can be expanded to the Dead Past and Future Hypothesis (DPFH) for the eternalist A-theory.¹¹² Objectively past and future people cannot think, sin, or suffer. The eternalist A-theorist endorsing the DPFH then must explain the differences between the past and future, as opposed to the present. And therein lie difficulties, regardless of whether the past and future are more or less ontologically robust. The more robust the existence of non-

¹¹¹ And the other relevant primitive properties that also follow, such as being past. ¹¹² Cameron (2015) agrees with Forrest that only present beings are conscious (23‒4). However, Cameron does not think this solution solves the epistemic problem (28‒9). He also thinks that a merely past person, such as Caesar, is dramatically different than present beings: “Caesar is now completely different from how he used to be: not only is he not conscious, but he no longer has any of the ordinary properties we think relevant to something’s being conscious . . . his being a certain height, mass, and having certain neural firings in his brain, etc., are merely ways things were, and they are no part of how reality is” (50). As will be discussed below, a view like this will be functionally similar to the presentist approaches.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

224   present things is, the more the existence of past sin and evil appears threatening. One problem that both the presentist and eternalist A-theorist share is that Christ’s salvific death is not as ontologically robust as what is present (though both can affirm the continuing salvific effects, and for the eternalist this is only a matter of degree). Both can attempt to explain why this is not a problem, due to the continuing causal effects of the cross versus the non-continuing effects of past sin. The existence of Christ’s sacrifice is enough to do some salvific work. The eternalist A-theorist who accepts the DPFH also shares with the presentist the inability to adopt certain versions of Stump’s account: Those which require that the conscious states of persons atoned for be in the domain of existence at the time of the crucifixion. More robust versions of joint attention will preclude this sort of eternalist A-theorist along with the presentist. Eternalist A-theory, though, can attempt to affirm both that the sacrifice of Christ needed to exist, simpliciter, in order for there to be any forgiveness and that the pre-Incarnation faithful were forwardlooking, having not received the full benefit of promises to come. For when those pre-Incarnation were present, the sacrifice of Christ existed but was in the future. Their sins could be forgiven, since Christ’s sacrifice existed, but the full promise—Christ’s being with us—was not yet present. However, this view has the consequence that the more efficacious the work of Christ is in this particular manner (when non-present), the more past and future sin and evil has its mark on reality.¹¹³ Eternalist A-theory, then, does not provide obvious help regarding the forgiveness of persons who existed pre-crucifixion over and above solutions available to either the presentist or B-theoretic eternalist.¹¹⁴ ¹¹³ To say otherwise, and ontologically privilege positive events at a time but not negative events at that same time, will land us back in the territory of Lebens and Goldschmidt. For this would require that positive events remain in a way that negative events do not and must then make use of something like their hypertime strategy of eliminating evil from existence. ¹¹⁴ Note the eternalist A-theorist’s ability to use a token-token existence requirement view of sin payment is not affected. For all an existence requirement view needs is that the exact sins, et cetera, existed. In the case of Christ’s sacrificial work, noting that the sins in question will be paid for is not enough if Christ’s work on the cross is supposed to do something for the sinner. But on this picture, Christ’s work, though it exists, appears just as efficacious as it is for the presentist— the sins will have reparations made for them, but those reparations are not present. There may

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

225

So, the eternalist A-theorist faces a dilemma: If the present is ontologically privileged to the extent that the existence of past suffering and sin is ontologically so non-robust as to approach the advantages of Evil Elimination, then the account will rule out conscious experiences in the past and future and cannot utilize theories which require them. If conscious experience is not reserved only for those who are present, then there will be conscious experience of sin and suffering at other times even as the saints rejoice in heaven post-eschaton. And this is precisely one of the problems of B-theoretic eternalism that the A-theory tries to avoid. So, while the eternalist A-theorist is able to combine some of the benefits of the other two theories, she is unable to claim the full benefits of either. The eternalist A-theory was introduced in an effort to provide a middle way between presentists and eternalists. However, the view appears to suffer from weaknesses of both accounts. Presentists will not be pleased that sin and evil still exist according to the view. But if the eternalist A-theorist emphasizes how little ontological significance past sins and sufferings have in the present, then she is in danger of losing advantages provided by ontologically robust existing things, such as the moment of atonement or conscious states. The eternalist A-theorist can still retain the effects of Christ’s sacrifice, but so, too, can the presentist. At minimum, positing an eternalist A-theory does not allow one to bypass the issues. However, one notable advantage of the eternalist A-theory is the ability to minimize the effects of past sin while adopting an existence token-token view of sin payment. It is unclear whether this (or the theory) ultimately helps matters, but it is another option worthy of exploration.

11. A Continuing Eucharist? However, the presentist and eternalist A-theorist may attempt to retain benefits of a robustly existing atoning event by questioning the earlier be advantages here, which merits further exploration, but they are not readily apparent: for instance, the mere existence of a non-robust atoning event does not appear to secure the fulfillment of promises of forgiveness in a way that allows for an obvious advantage over presentism. For presentists can also say that the cross was going to happen (even determinately at a particular point in time, if they wanted to), and can thereby secure the promises of forgiveness as well.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

226   notion of the atonement happening “all at once.” To do so, they will need a strong view of the sacraments: A potential way out is to maintain that Christ’s redemptive work on the cross, unlike sins which are here today and gone tomorrow, is continually carried out via the Church’s continued participation in the Eucharist. Christ’s sacrificial work never becomes past, as the Church has carried the salvific work continually in the present. At first glance, this appears to violate the condition that Christ is offered once and only once. In the continual practice of the Eucharist, is not Christ continually sacrificed? No: One can hold that the Eucharist is the continuing of a single act of Christ’s ongoing sacrifice that began on the cross. There are several things to note here. First, this requires an unusual understanding of what is considered to be a single act, particularly if the act is to be continuous. Second, Christ’s ongoing sacrifice, even with a robust understanding of the Eucharist, does not appear to be continuous: It seems highly implausible that the Eucharist would have been practiced in the interim period between Christ’s death and resurrection, let alone continuously in the early Church. If the Eucharist had not been offered in the interim, it would appear that Christ was offered at least twice—once on the cross and once at the first Eucharist after the Resurrection (three times, if you count the Last Supper). Perhaps, instead, the act simply needs to be continuing, rather than continuous. However, this will face concerns about what counts as a single act.¹¹⁵ Third, this understanding may not do justice to the last words of Christ: “It is finished.”¹¹⁶ One can interpret Christ to be speaking solely about the end of his human life, but that ignores the surrounding texts in which Christ says things to fulfill the scriptures. Perhaps, though, Christ’s last words on the cross were correct—the salvific work of reparations was finished, but something sacrificial related to the event, such as Christ’s sacrificial blood, need continually exist (and it does so via the Eucharist).

¹¹⁵ This could occur if events can be temporally gappy or scattered. (Thanks to Alicia Finch for this point.) However, if events can be temporally gappy, the gappy event need not involve the Eucharist: perhaps it is something else that most every theorist can utilize. This is worth further exploration. ¹¹⁶ John 19:30.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

227

The above account of the Eucharist may not be popular, but it does demonstrate that B-theoretic eternalism need not be the only way to say that the act of atonement (and not just its effects) continually exists. It also raises a further possibility for presentism: If the event of Christ’s sacrifice or some element of it did not stop, it solves some of the earlier difficulties regarding cross-time relations and the coexistence of Christ’s act and the effects on us. As of now, though, the particulars of the account are not clear, and so it is not apparent whether or not this sort of move works. But if one is attracted to presentism or an eternalist A-theory but thinks that the act of atonement still needs to exist, this is a route worthy of exploration.

12. Taking Stock There is much more to be said about these topics, and certain seemingly simple or easy answers are unavailable. Answers which appeared to give simplistic solutions turned out to offer anything but ontological ease: for they involved commitment to things like extreme mind‒body dualism. And certain overly simplistic views of atonement, used to object to either presentism or eternalism, failed in large part due to issues with those views themselves. For example, in what sense is Christ’s death “once for all”? A wooden reading of this condition, in terms of a property swap at a single time, for example, might initially appear to rule out presentism— but such a reading appears to be incoherent. There are at least several ways of interpreting such conditions, leading to different ontological options: either in terms of distinctions made, available views in a taxonomy, or even, ultimately, temporal ontologies themselves. Importantly, the temporal questions, distinctions, and taxonomies introduced above must be dealt with despite differing accounts of the atonement, such as transfer views and solidarity theories. One cannot remain neutral with respect to the ontological options. While the answers one can give will vary depending on one’s view of time or one’s preferred atonement theory, there is no way to avoid these temporal questions entirely—at least, not if one is to offer a complete theory.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

228   Neither presentists nor eternalists can, at this point, claim to better fit the proper account of the atonement, largely due to disagreement regarding what is required for a proper account of the atonement. However, presentism and eternalism have been used to highlight and aid in choice points regarding what an atonement account can and should look like. Presentism may have advantages if one wants an Evil Elimination view, or theories which focus on the elimination of evil and sin; Eternalism, on the other hand, allows for broad token-token views (in terms of both property transfer and mind-reading) and many other theoretical options.¹¹⁷ One final complication, however, must be addressed. There appears to be tension in the scriptural metaphors regarding the forgiveness and removal of sin with which we began this conversation. Isaiah 44:22 says, “I have swept away your transgressions like a cloud, and your sins like mist; return to me, for I have redeemed you,” while Acts 3:19 declares, “Repent therefore, and turn to God so that your sins may be wiped out” (emphasis mine). Isaiah indicates that sin has already been taken care of, while Acts indicates that something more is needed in order to wipe out sin. This calls into question some of the fundamental assumptions of this chapter. First, I had assumed that the major work of forgiveness and reparation for sin were accomplished on the cross regardless of whether people are repentant. Second, it was tacitly assumed that the mechanics of the atonement did not vary by people group—if there is a required order to the atonement (e.g., forgiveness, then repentance), that order is the same for everyone.¹¹⁸ ¹¹⁷ Though the latter is a limited sort of advantage—it offers more options, and thus allows for someone to choose among a group of potentially attractive views. But at the end of the day, only one of the competing options (e.g., type-token versus token-token views) will be true. And it may very well be that the true theory is consistent with presentism. ¹¹⁸ One might think that while the mechanics of the atonement do not vary by people group or person, messages communicated to humans about atonement might vary due to the unique and particular point of the cross in eschatological history. Before Christ, things regarding the atonement were by their very nature “forward looking,” whereas now we are “backward looking.” But the above verses are not what one might expect on this sort of picture, since the message of currently existing forgiveness was to a pre-Christ Israel (though one might think that such forgiveness could be given because of what Christ would do), and the need for repentance in order to have forgiveness was addressed to an audience post-resurrection, while making use of prophetic passages from pre-Christ Israel.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

229

One can account for the consistency of these scriptural metaphors by saying the imagery applies to different steps of atonement: say, reparations and repentance. If one finds the previous discussion in this chapter regarding the Temporal Question of atonement convincing, such seeming tension is solved with the appropriate distinctions and framework. Simile and metaphor are malleable; an image used in one context need not communicate the same content as the same image used in another context. However, one might read these two statements as indicating that the major salvific act can occur at more than one time, thus dropping the constraint that the atoning work of Christ happen only once, on the cross. If so, then our theoretical options widen considerably. Presentists, for example, will be able to utilize certain previously unavailable existence requirement token-token views. The presentist may say that individual sins are paid for as they happen, that Christ shares relevant mental attention with each person when they exist, and that Christ’s atoning sacrifice is present and real in a way that sin is not, since Christ’s sacrifice is always happening. One should be careful, though, when dealing with imagery and metaphor, to interpret the more poetic by means of the less poetic. One cannot read 1 Peter 3:18 (“Christ suffered for our sins once for all, the just for the unjust”) as rejecting the one-time constraint without doing violence to the text, whereas “sweeping away” and “blotting out” can be interpreted more loosely. While there are many ontological options available for consideration, our theorizing can be helpfully constrained by the texts themselves, which also deserve further study. Where should the conversation go from here? I recommend we aim for reflective equilibrium between what we take to be the proper theory of time and how we understand the atonement. The previous discussion offers tools, by way of providing distinctions and taxonomies and also in highlighting areas in need of further exploration. I thus end with the hope that there will be such exploration: further detailed work is needed in order to make sense of these issues.¹¹⁹

¹¹⁹ Thanks to the 2012 Notre Dame Center for Philosophy of Religion reading group, the 2012 Baylor-Georgetown-Notre Dame Philosophy of Religion conference, Kenny Boyce, Jonathan Jacobs, Erik Johnson, Jon Kvanvig, Carl Mosser, Faith Pawl, Tim Pawl, Luke Potter,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

230  

References Adams, M. M. (2006). Christ and Horrors. Cambridge University Press: Cambridge. Adams, M. M. (1999). Horrendous Evils and the Goodness of God. Cornell University Press: Ithaca, NY. Aletti, J. (2006). “ ‘God made Christ to be Sin’ (2 Corinthians 5:21): Reflections on a Pauline Paradox,” in The Redemption: An Interdisciplinary Symposium on Christ as Redeemer. Davis, S. T., Kendall, D., & O’Collins, G. (eds). Oxford University Press: Oxford. Almeida, M. J. & Oppy, G. (2003). “Sceptical Theism and Evidential Arguments from Evil,” Australasian Journal of Philosophy, 81 (4): 496–516. Baker, L. R. (2007). “Persons and the Metaphysics of the Resurrection,” Religious Studies, 43: 333–48. Bergmann, M. & Rea, M. (2005). “In Defence of Sceptical Theism: A Reply to Almeida and Oppy,” Australasian Journal of Philosophy, 83 (2): 241–51. Bigelow, J. (1996). “Presentism and Properties,” Philosophical Perspectives, 10: 35–52. Braddon-Mitchell, D. (2004). “How Do We Know it is Now Now?,” Analysis, 64: 199–203. Broad, C. D. (1923). Scientific Thought. Routledge & Kegan Paul: London. Cameron, R. (2015). The Moving Spotlight. Oxford University Press: Oxford. Chisholm, R. (1976). Person and Object: A Metaphysical Study. Open Court Publishing: LaSalle, IL. Crisp, O. D. (2009). “Original Sin and Atonement,” in The Oxford Handbook of Philosophical Theology. Flint, T. P. & Rea, M. C. (eds). Oxford University Press: Oxford. Crisp, O. D. (2008). “Penal Non-Substitution,” The Journal of Theological Studies, 59 (1): 140–68. Crisp, O. D. (2007). “Non-penal Substitution,” International Journal of Systematic Theology, 9 (4): 415–33.

Alan Rhoda, Meghan Sullivan, Joshua Thurow, Peter van Elswyk, and Dean Zimmerman for helpful comments and discussion. Special thanks to Alicia Finch, Michael Rea, and Warren Rogers for formative discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

231

Crisp, T. (2007). “Presentism and the Grounding Objection,” Nous, 41 (1): 90–109. Crisp, T. (2005). “Presentism and Cross-Time Relations,” American Philosophical Quarterly, 42 (1): 5–17. Crisp, T. (2003). “Presentism,” in The Oxford Handbook of Metaphysics. Loux, M. & Zimmerman, D. (eds). Oxford University Press: Oxford. Fee, G. (2006). “Paul and the Metaphors for Salvation: Some Reflections on Pauline Soteriology,” in The Redemption: An Interdisciplinary Symposium on Christ as Redeemer. Davis, S. T., Kendall, D., & O’Collins, G. (eds). Oxford University Press: Oxford. Fine, K. (2006). “The Reality of Tense,” Synthese, 150 (3): 399–414. Forrest, P. (2004). “The Real but Dead Past: A Reply to Braddon-Mitchell,” Analysis, 64 (4): 358–62. Geach, P. (1969). God and the Soul. Routledge and Kegan Paul: London. Hinchliff, M. (1996). “The Puzzle of Change,” Philosophical Perspectives, 10: 119–36. Hudson, H. (2014). The Fall and Hypertime. Oxford University Press: Oxford. Hudson, H. (2010). “An Essay on Eden,” Faith and Philosophy, 27 (3): 273–86. Lebens, S. & Goldschmidt, T. (2017). “The Promise of a New Past,” Philosophers’ Imprint, 17 (8): 1–25. Lewis, D. (1997). “Do We Believe in Penal Substitution?,” Philosophical Papers, 26 (3): 203–9. McTaggart, J. M. E. (1998). “Time: An Excerpt from The Nature of Existence,” in Metaphysics: The Big Questions. Van Inwagen, P. & Zimmerman, D. (eds). Blackwell Publishing: Malden. Mellor, D. H. (1998). Real Time II. Routledge: London. Merricks, T. (2007). Truth and Ontology. Oxford University Press: New York. Merricks, T. (2006). “Goodbye, Growing Block,” Oxford Studies in Metaphysics, 2: 103–10. Mullins, R. (2014). “Four-Dimensionalism, Evil, and Christian Belief,” Philosophia Christie, 16 (1): 117–37. Murphy, M. C. (2009). “Not Penal Substitution but Vicarious Punishment,” Faith and Philosophy, 26 (3): 253–73.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

232   Niven, Larry (1967). “Flatlander,” Worlds of If, 17 (3): 78–131. Priest, Graham (2006). In Contradiction, 2nd edn, Oxford University Press: Oxford. Rea, M. (2019). “The Ill-Made Knight and the Stain on the Soul,” European Journal for Philosophy of Religion, 11 (1): 117–34. Rea, M. (2006). “Presentism and Fatalism,” Australasian Journal of Philosophy, 84: 511–24. Rea, M. (1998). “Temporal Parts Unmotivated,” The Philosophical Review, 107: 225–60. Seymour, A. (2016). “Presentism, Propositions, and Persons: A Systematic Case for All- Falsism,” (PhD Thesis), University of Notre Dame. Sider, T. (2001). Four Dimensionalism: An Ontology of Persistence and Time. Oxford University Press: Oxford. Strabbing, J. (2016). “The Permissibility of the Atonement as Penal Substitution,” in Oxford Studies in Philosophy of Religion. Kvanvig, J. L. (ed.), vol. 7, 239–70. Strabbing, J. (forthcoming). “Forgiveness and Reconciliation,” Australasian Journal of Philosophy. Stump, E. (2018). Atonement. Oxford University Press: Oxford. Stump, E. (2012a). “Atonement and the Cry of Dereliction from the Cross,” European Journal for Philosophy of Religion, 4 (1): 1–17. Stump, E. (2012b). “The Nature of the Atonement,” in Reasons, Metaphysics, and Mind: New Essays on the Philosophy of Alvin Plantinga. Clark, K. & Rea, M. C. (eds). Oxford University Press: Oxford. Stump, E. (2010). Wandering in Darkness: Narrative and the Problem of Suffering. Oxford University Press: Oxford. Stump, E. & Kretzmann, N. (1981). “Eternity,” Journal of Philosophy, 78 (8): 429–58. Sullivan, M. (2012). “The Minimal A-Theory,” Philosophical Studies, 158 (2): 149–74. Swinburne, R. (1989). “Guilt, Atonement, and Forgiveness,” in Responsibility and Atonement. Oxford University Press: Oxford. Szabo, Z. (2006). “Counting Across Times,” Philosophical Perspectives, 20 (1): 399–426. Thomson, J. J. (1971). “The Time of a Killing,” Journal of Philosophy, 68 (5): 115–32.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 25/1/2022, SPi

      

233

Van Inwagen, P. (2014). “Modes of Being and Quantification,” Disputatio, 38 (5): 1–24. Van Inwagen, P. (2010). “Changing the Past,” Oxford Studies in Metaphysics, 5: 3–28. Van Inwagen, P. (2009). “Being, Existence, and Ontological Commitment,” in Metametaphysics. Chalmers, D., Manley, D., & Wasserman, R. (eds), 472–506. Van Inwagen, P. (1990a). “Four-Dimensional Objects,” Nous, 24 (2): 245‒55. Van Inwagen, P. (1990b). Material Beings. Cornell University: Ithaca, NY. Williamson, T. (2002). “Necessary Existents,” in Logic, Thought, and Language. O’Hear, A. (ed.). Cambridge University Press: Cambridge. Wright, N. T. (2006). “Redemption from the New Perspective? Towards a Multi-Layered Pauline Theology of the Cross,” in The Redemption: An Interdisciplinary Symposium on Christ as Redeemer. Davis, S. T., Kendall, D., & O’Collins, G. (eds). Oxford University Press: Oxford. Zimmerman, D. (2008). “The Privileged Present: Defending an ‘A-Theory’ of Time,” in Contemporary Debates in Metaphysics. Sider, T., Hawthorne, J. & Zimmerman, D. (eds). Blackwell Publishing: Malden. Zimmerman, D. (2006). “Can One ‘Take Tense Seriously’ and Be a B-Theorist?,” in Persistence. Haslanger, S. & Kurtz, R. (eds) MIT Press: Cambridge, MA.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

7 Why the One Did Not Remain Within Itself Timothy O’Connor

The central metaphysical question for Plotinus was: Why did the One not remain within itself? Transposed from Plotinus’s Neoplatonist to a classical theistic framework, the question becomes: Why did the omnipotent, omniscient, unsurpassably and perfectly good being who is necessary in himself, and having a supremely rational will, contingently create ex nihilo? What motivation could account for such freely undertaken activity, displaying it as neither necessary nor less than fully rational? My point of departure is a richly detailed and carefully argued article by Mark Johnston (2019) addressing this ancient question. Johnston makes several astute critical points and correctly identifies some of the central pieces needed to answer the question satisfactorily, but his proposed answer is at best significantly incomplete. I will argue that a more complete answer will plausibly posit an ordered complexity in God’s motivations, with surprising implications, possibly severe, for the scope of contingency.

I. The Problem Given God’s supremely rational will, his creative activity must have completely adequate reasons: there must be considerations that motivate his so creating that are not outweighed by the reasons for creating differently, or for not creating at all. Given that God is of himself

Timothy O’Connor, Why the One Did Not Remain Within Itself In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Timothy O’Connor 2022. DOI: 10.1093/oso/9780192862976.003.0007

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

235

unsurpassably good—indeed, is the Good itself—his activity cannot arise from need, or more generally from any desire or goal that, if unrealized, would render him somehow incomplete, less good than he would be were he to realize it. This suggests to Johnston and many others that while God’s reasons for acting as he does must be completely adequate, they cannot be coercive: they cannot outweigh reasons for acting in any other way, or for not acting at all. Putting these two constraints together, the challenge is to offer an intelligible account of God’s creating what he does such that his reasons for so doing neither outweigh nor are outweighed by reasons he had for not creating, or for creating differently in any number of ways (Johnston, 112‒15). Some would object to this way of framing what we should be after in an account of the reasons God might have for acting as he does. They will say that if the totality of reasons God has for acting as he does satisfies the second of these two constraints—it is not “coercive”—then his action will fail to satisfy the Principle of Sufficient Reason (PSR), as traditionally and stringently understood by rationalists. Let us table this objection until the end and, for now, focus on whether there is an intelligible way of thinking about the divine motivation such that both constraints, complete adequacy and non-coercion, are satisfied.

II. Two Rejected Answers and Johnston’s Proposed Third Answer A. Meliorism One reason for creation that Johnston considers and rejects is melioristic: the aim of improving (or preventing the diminishment of) reality overall (151‒3). This reason is rejected on the grounds that improvement to reality is impossible for an unsurpassably good being. Whether we think of the value of such a being as measured by an infinite cardinality greater than that measuring any other possible reality or as beyond (even infinite) measure altogether (as Cantor himself supposed), its act of creating a lesser reality would be like adding a point to a line (151): it is no addition at all.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

236   ’ 

B. Generosity Consider a second possible reason God might have to create: generosity. It is not that God in creating is moved to be generous to us, or to any possible creatures in particular, as he cannot have concern for or attachment to beings who do not yet exist, merely possible beings. Instead, the suggestion must be that God simply is disposed to be generous, he has the virtue of generosity, and he creates so that there might be beings (as it happens, us) who receive his generosity in coming to be. Johnston contends that this proposal, too, must be rejected. He gives two reasons. First, precisely because this reason does not suffice for creating us (along with all the other denizens of contingent reality) in particular, it cannot be completely adequate. It is incomplete, being a reason only for creating someone(s) or other. In consequence, there must be a further reason that points to our created reality in particular. And whatever that further reason might be, it will do all the explanatory work, making idle the appeal to God’s generosity. A second reason to reject the generosity explanation is that it succumbs to the threat of necessary creation. It would be a reason that, if it were completely adequate, would have no counterpart reason not to create, since (Johnston contends) “there cannot be some positive value which he would have instantiated had he not created, but did not instantiate because he created.” So if instantiating the value of generosity were the reason God created, it would be rationally coercive. Since this is contrary to our aim of finding a non-coercive, though adequate reason, we must look elsewhere (148‒50).

C. Manifestationism Following Aquinas, Johnston suggests that the needed insight is that God necessarily wills or affirms his own goodness. Any reasons for particular forms of creative activity or inactivity are ordered to (are contingent modifications of) this fundamental and necessary aim. They are “adverbial”: different ways of willing his own goodness. Johnston further suggests that Cantor’s account of the transfinite enables us to see God’s

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

237

reasons for acting as he contingently does as “extra” reasons, unmatched by reasons for not so acting, without thereby being weightier in the final analysis, and so making Creation compulsory for a fully rational will. Whether God creates or not, he fundamentally wills an unsurpassable good, his own goodness. Johnston then proposes that the reason for creating (as an optional way of willing his goodness) is to manifest, to show forth, his own glory in what he creates. This provides an extra reason, one that would not have a counterpart had he willed his own goodness in creating nothing at all, but in either case, God’s overall reasons for choosing as he does would be equally weighty, as both target an unsurpassable good (155–57).

III. Going Deeper Notice that the specific modification of God’s central, necessary willing of himself that Johnston settles upon as a reason for creating is highly unspecific and does not point to any particular creative activity. But (plausibly) manifesting the glory of perfect goodness does exclude certain choices. For example, a perfectly good God could not will a created reality that is fundamentally and ultimately unjust—notwithstanding the fact that the aggregate value of such a world would be unsurpassable. Furthermore, important facets of God’s perfection center on his personhood, and in particular encompass his generosity; a creation lacking creatures who bear his image in this most profound sense and who can receive his generosity seems deficient in the way it manifests his glory. And since God’s perfection is multifaceted, perhaps infinitely faceted, a created reality that contains widely diverse kinds of good creatures would seem to better manifest his glory than one that is more uniform. The general point is that there are more or less good ways that God might manifest his glory, and it seems that a perfectly rational will would tend towards the better ones. Thus, it is inevitable that there will be further structure to God’s willing, rooted in the general, fixed aim of willing goodness that Johnston (following classical thinkers) has identified. God’s reasons for

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

238   ’  creating this reality, for example, might have something like the form: willing his own goodness by manifesting his glory in a Creation of great value which includes creatures to whom he may express his generosity. (Note that this provides the basis of a response to Johnston’s first criticism of the “generosity” answer: while generosity may not be wholly adequate, it may be a necessary component of a wholly adequate, complex reason for God to create what he does. But more below on whether the expression of generosity should, in the final analysis, be thought of as a partial reason for God’s creation.) Since unsurpassable aggregate value—the sum of individual values—is a given no matter what, God’s willing must be sensitive to something other than simple aggregate value. The natural alternative is that it is the organic value of Creation as a whole. It is a delicate matter how to understand this. It is initially tempting to say that this value attaches to possible creations considered in abstraction from God, their necessary source and ground: their value qua divine manifestation. But reflection suggests otherwise. On theism as opposed to deism, God is the one in whom we live and move and have our being. Every created thing is by very nature ontologically dependent moment to moment on God’s sustaining activity. Which is to say, the created order cannot be properly and fully understood in abstraction from God. Furthermore, theists typically and plausibly suppose that union with God is our ultimate end (and perhaps that of other sentient creatures as well). For such creatures, at least, there can be no describing their place in the created order apart from God. Thus, the organic value of every possible creation is partly constituted by creaturely relations to God on which such possible creations are centered. Think of God as like a master craftsman. He would want to fashion only those created orders that have great value as organized totalities. Within the created order, he would also take into account the value of organisms and perhaps things such as ecosystems—components of the universe that are interesting unities and so whose own organic value can sensibly be gauged. Should we suppose that God would be further motivated to multiply copies of such valuable things, without end? Not obviously so. Certainly, the master craftsman analogy doesn’t suggest it: it is not merely for lack of time that human craftsmen aren’t highly

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

239

motivated to duplicate their valued outputs—apart from considerations of instrumental utility to them (e.g., money), which have no analogue in God’s case. There will be only a qualified meliorism of desiring to make things better for those sentient creatures there are, whose reason for existing lies elsewhere. If we take the above considerations to provide decisive reasons to favor some creative options over others, then we will have narrowed the scope of contingency. (There is no possibility that the necessarily existing God would select the rejected options, and his creative power sets the bounds on what concrete realities can exist. If classical theism is true, fundamentally unjust worlds are impossible.) Will this lead to some options being the best possible manifestation, leading to complete modal collapse? I think not. There may well be kinds of goods that are noncompossible (at least within a single universe) and incomparable. If so, the choice of any option among such pairs or n-tuples will not be rationally coerced. Incomparable values no less than equal values can serve to block there being coercive reasons for one particular course of action. Each option will be weightier than the others in some respect, but none will be weightier simpliciter. (At least, that will be so if there is no generic scale of value with respect to which they have a positive value relation—that is, if the values are not just incomparable but are incommensurate.)¹ A further hedge against modal collapse will be present if there are variable options for attaining the maximal value along any of the incomparable value dimensions—that is, significantly different ways of expressing each maximal compossible mix of values. But what if, for each of the incomparable value dimensions, there is no maximal value, and possible creations instead either approach aleph-null as an unrealizable limit or, crossing that threshold, fail to top out at any higher aleph? (I find the former of these scenarios more plausible for systems of finitely valuable fundamental created entities, even when they are oriented on an unsurpassably valuable source and telos. But it is hard to identify a firm basis for such a conjecture. It is not obviously false that fractal-like nested and fathomless systems with infinitely many elements ¹ On this distinction, see Chang (2013).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

240   ’  at each level, arranged in aesthetically and otherwise valuable ways, might have transfinite organic value.) It is a little jarring to contemplate an infinitely perfect creator having to choose a creation of some particular, limited value in the face of attainable alternatives of arbitrarily greater value. If the structure of possibility space rendered it inevitable, we should accept it with the piety of the modal investigator;² but if there were a way around such submaximalization, it is plausible that a supremely rational and unsurpassably good chooser would take it. And indeed, there appear to be two natural and possible alternatives: 1. Ratherthan create a single, organically unified totality, a limitless creator might opt for a multiverse of infinitely many, wholly discrete universes. This would circumvent submaximalization provided the individual evaluable universes were formally orderable by ascending value along some value axis with no upper bound. Submaximalization of universe creation is avoided by “covering all the bases.” If (contrary to my own instinct) one maintains that there is a kind of non-additive value attaching to multiverses, despite their spatiotemporal and causal isolation, then the original question of whether there might not be a maximal value seems answerable in the negative: an appropriately diverse, infinitely numbered multiverse plausibly has maximal such value. 2. A second,single-universe option depends on a view about time, on which there is real becoming in the created order, rather than time being, like space, a fully realized dimension. Given such a metaphysics of time, it appears there could be an ever improving created order that has merely potential infinity of value which it approaches as a limit.³ Selecting it does not require settling for a fixed value less than that of other options. This option seems most attractive on the assumption that time necessarily involves real becoming. If it is not, then a proponent must confront ² William Rowe (2004) sees, unconvincingly, the makings of an argument for atheism here: if a creator of limitless power and knowledge were to choose any option under such constraints, of value x, then we might imagine a morally better being, one who places a floor to his preferences at the value of 2x. As Chris Tucker (2016) observes, Rowe’s argument rests on a principle (that the goodness of one’s character is necessarily expressed by the value of one’s choice) that is widely rejected in ordinary human contexts. Dean Zimmerman (2018) extends Tucker’s point to analogous “expression” principles concerning certainty and desire. ³ This option has been explored recently by Nevin Climenhaga (2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

241

the question of the value of such a universe relative to the wholly realized value of a “static” (B-theoretic), temporally infinite universe with increasingly valuable stages.⁴ For each of these stratagems, the threat of modal constriction re-emerges (rationality would seem to compel the selection of one of these options), and there is an analogous response to the original: there appear to be infinitely many distinct ways to realize the stratagem’s central feature. Let us press on by returning to the most fundamental choice point faced by God: whether or not to manifest his glory in Creation. Is it plausible that this might be contingent, such that there was a real possibility that God might not have manifested himself at all? Johnston thinks that we should agree that it is, once we see the option of manifesting himself as “adverbial” expression—as one way of willing his own unsurpassable goodness, alongside another, equally weighty way that involves no creating at all. But we have already suggested that some determinate ways of willing God’s own goodness might be better than others—such that God’s selecting the others would not be backed, in Johnston’s terminology, by “fully adequate” reasons. I am inclined to follow pseudo-Dionysius, when he said that goodness naturally “diffuses itself”—goodness has a creative impetus built into it. And when it comes to God’s perfect goodness, there is opportunity, no limits on resources, time, or energy, and no conflicting obligations or needs to adjudicate. In short, there is nothing to restrain or override such an impetus. The natural conclusion to draw is that God inevitably creates something or other. (Thomas Aquinas plainly was mightily tempted to agree in a number of places. But finding the conclusion theologically unpalatable, he draws back and allows only that it is “fitting” that God, who is perfect goodness, should create. But what does that mean, if not that, other things being equal, it is better for him to do so than not?)⁵ The thesis that perfect goodness naturally diffuses itself outward in Creation does not entail that generosity is a fundamental creative

⁴ I thank Dean Zimmerman for pressing this modal question. ⁵ For trenchant discussion of assorted texts in Aquinas discussing this matter, see Norman Kretzmann (1986).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

242   ’  motivation for God. God has a natural impetus to create diverse good kinds of things, and these include self-aware beings. So it was perhaps inevitable that he would create some self-aware beings or others, and this would rightly be perceived by them as generous. However, generosity is best seen not as a fundamental attribute, but only as a consequence of the more fundamental attribute of goodness, which is creatively disposed. To those creatures capable of receiving his generosity, there is nothing stinting or qualified about it. Even so, if (as I am supposing) God is not an aggregate value maximizer, we will avoid the suggestion that God is as generous as he can be in an extensive sense.⁶ Alex Pruss (2016) has recently pushed back on this necessity-ofcreation conclusion by arguing that there can be positive reasons not to create that counterbalance reasons to create and hence reasons that contribute to grounding real contingency in whether, and not just how, God creates. By not creating (and merely contemplating the possibilities), God would realize three world values not obtainable otherwise: maximal simplicity, uniformly maximal excellence, and all beings perfectly achieving their natural telos (and so a world without any tincture of metaphysical ‘evil’ or defect). Pruss does not argue that a world manifesting such values would be on balance better than ones that lack them (that would be an argument for atheism!). Rather, he sees them simply as among the many incommensurable world values, any of which a perfect being might opt to realize, foregoing the others. In reply, I observe that each of the values Pruss cites are simply and uniquely characteristics of God, as classically conceived, here considered in relation to God-centered worlds. They apply to a world in the unique case where God is the sole reality. Thought of as world values, they are disguised conjunctive values: there being the unique possible being that is maximally simple, excellent, and fully realizing the “potential” or “end” of its nature, and there being nothing else. Whether God creates or not, he necessarily realizes the first of the conjuncts in himself. The question is whether the second, negative conjunct—the absence of anything else, which necessarily would fall short of these characteristics—contributes ⁶ Thanks to Mark Johnston for helpful correspondence on the issued discussed in this paragraph.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

243

to a reason for not acting that would be just as adequate as the positive reasons for acting in specific ways. I am skeptical. In a world where God creates lesser realities, he is not just one thing among others in the world, nor even is he merely the “biggest” reality, ontologically speaking. His being swamps all such lesser beings and permeates them. When the world is seen aright, the divine attributes are no less lustrous in the dependent presence of other realities, each of which “speaks” of the divine nature. As Augustine put it, “ . . . all these things have the same message to tell, if only we can hear it, and their message is this: we did not make ourselves, but he who abides forever made us” (Confessions, Bk XI). I don’t see that there is a kind of additional value had by actuality in the case where the unsurpassable God is all there is. Pruss is part of a long tradition in seeking a way to reject the necessity of divine creative activity. But what is the problem, exactly? One worry is that it renders God unfree. We could spend a fair bit of space scrutinizing the distinctive character of divine freedom. Here I will say only that its being open to God to creatively express or manifest himself in infinitely many ways, circumscribed only by his own selfsubsisting nature that is unmarred by compulsions and guided only by rationality and goodness, is to be mighty free indeed. For those who think this not freedom enough, the natural alternative is to suppose that it is open to God—really open to God, something he might actually have done—to do anything that is possible in itself (including nothing). But reflection on that maximal-scope understanding of perfect freedom reveals it to be a chimera. It could be true only of something that had no preferences, no character at all. We are here supposing that God naturally expresses himself somehow or other, much as he naturally thinks the eternal truths. This is not a constraint on freedom, but a reflection of the perfect divine nature that provides the context in which choice among alternatives makes any sense at all. A second worry might be this: If it is necessary that God would manifest goodness in some manner or other, it follows that there is some incompleteness of God’s perfection “within himself,” so to speak, such that he is completed only in the activity of creation. But while this worry has a little more force to it, I think it can be comfortably resisted. If God’s necessarily thinking the eternal truths is not a “completion” of his

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

244   ’  nature, neither need this be the case when it comes to the necessity of his expressing his goodness in some creative fashion. It simply reflects his perfection; there is a congruence between this necessary impulse to selfmanifestation and what his perfect rationality judges as fitting. If one worries that creatures praise God’s generosity but one ought not to praise (in the moral sense) what is necessary, the answer of course is that we can praise his freely given generosity in creating us, non-coercively.

IV. The Principle of Sufficient Reason While many theists are concerned to maximize the contingency of creation, some others will equate any contingency in the divine choice with the breakdown of rational explanation. If God does A when he might just as well have done B, then whatever allegedly explains A cannot explain why God did A rather than doing B. So this contrastive fact, at least, will be unexplained. In recent philosophy of science, there is some debate about what contrastive causal explanation requires, with two main accounts on offer, one of which is actually consistent with the availability of true contrastive explanations for indeterministic events in many kinds of scenarios. Peter Lipton (1990) contends that cause C explains why P rather than the non-occurring Q if and only if C causes P and there is no “corresponding” event to C in the history of ~Q—no relevantly similar event C* that would have been a cause of Q had Q occurred instead. Setting aside complications concerning how we might understand God’s agency in causal terms, any analysis along these lines will plausibly rule out contrastive explanations for undetermined divine choices of the sort that theists generally envision, as they are ones in which certain divine motivational states explain what actually occurs and other motivational states would have explained alternative possible outcomes. Let us call this the “strong account” of contrastive explanation. Christopher Hitchcock (2012) argues for a significantly weaker analysis. As he sees it, C explains why P rather than Q just in case C raises the objective probability of P more than it, C, raises the probability of Q. This “weak account” will easily accommodate contrastive psychological explanations of divine

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

245

choices for all scenarios save “Buridan’s ass” cases where the very same motivation would explain either or any of the arbitrary choices made. One might well wonder whether one or the other (or some third) analysis must be the correct analysis. The question, “Why P rather than Q?” is in general a way of requesting a particular kind of explanation of P against the background assumption that it was not possible in the circumstance that P and Q both obtain. That much is clear. But it is possible to hear the question as imposing more or less strong requirements on a satisfactory answer, as is brought out by placing verbal emphasis on “rather than,” or by repeating the question with emphasis on the fact that the explanation doesn’t show why Q could not have occurred. No matter. The most fundamental point to see is that contrastive explanation strongly construed is not some kind of explanatory “gold standard,” let alone the only standard. We may accept the more stringent requirement on contrastive explanation while rejecting the claim that the absence of a contrastive explanation entails a deficiency in explanation simpliciter. On the envisaged scenario, God and his creative motivations are the source of contingent reality, and thereby explain whence it comes. (Some will think of this as noncontrastive “causal source” explanation, others as tacitly pointing to a weaker, Hitchcockian variety of contrastive explanation. In what follows I will assume that it is noncontrastive, as I wish to show that conceding the absence of contrastive explanations does not have the damaging consequence that some suppose.) We can go on to ask more fine-grained explanatory questions about aspects of this reality, including (strongly) contrastive questions. But we should observe that contrastive truths correspond to nothing real—there are no “contrastive facts” in the sense of concrete structures or constituents in reality. The concrete totality that contingently obtains has a “completely adequate” explanation in the productive activity of God, guided by particular reasons. But (an objector may continue) even if we grant that there is a noncontrastive explanation of why this in terms of God’s motivated generative activity, allowing that there would be no explanation why this as opposed to that (where “that” is any possibility not precluded by God’s perfect goodness and rationality) is to accept an explanatory

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

246   ’  surd, something incompatible with the philosophical theist’s pursuit of ultimate explanation of reality. But this, too, I deny. Notice that our conception of a necessarily existing, purposive, and free source of contingent reality provides an explanation of why there can be no explanation of certain contingent contrastive truths. Explanatory completeness, we might say, is restored at a “higher level” of explanation. There are the first-order concreta, explained by God’s reason-governed activity. And there are various second-order truths about them, including those contrastive truths. We can have a robust and principled PSR that is also open to all epistemic options concerning the “causal character” of what lies at the necessary foundation of reality. It can allow that there are some second-order truths lacking an explanation, provided that there is an explanation of this lack of explanation, a third-order truth. Such a formulation of PSR will require only a kind of completeness theorem that, at some level, complete explanation is gained in the sense that every lower-order truth falls under the umbrella of a true explanation, and this persists up the hierarchy. If this makes sense, then it could be a necessary, grounded truth that there are (contrastive) truths without explanation. This general truth, notice, will not be a result of what God does. It would obtain were God not to create anything at all; it is grounded in his nature, in the fact that he has the all-things-considered potential for a plurality of outcomes. It also does not reflect something distinctive about rational explanation of the actions of a free agent. Were the One (contra theism) a “blind,” impersonal source of contingent reality, a nonintentional but non-deterministic cause, the explanatory situation would be precisely parallel. Given that indeterministic activity is consistent with a kind of explanatory completeness, on which nothing is objectionably brute (i.e., lacking any basis whatsoever), we should ask ourselves: by what epistemic right do we formulate the PSR in the manner of Leibniz and Spinoza so as to preclude these epistemic possibilities from the outset?⁷ ⁷ I presented material resulting in this chapter at colloquia at the Center for Philosophy of Religion, Rutgers University, Central European University, the University of Wuhan, and Taiwan National University. I thank the audiences on these occasions for helpful feedback. I especially thank Mark Johnston and Alex Pruss for private conversation that has (I hope) led to numerous improvements, as well as Dean Zimmerman for queries and objections on the penultimate draft.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

       

247

References Chang, Ruth (2013). “Incommensurability (and Incomparability),” in Hugh LaFollette, ed., The International Encyclopedia of Ethics. Oxford: Blackwell, 2591–604. Climenhaga, Nevin (2018). “Infinite Value and the Best of All Possible Worlds,” Philosophy and Phenomenological Research 97(2), 367–92. Hitchcock, Christopher (2012). “Contrastive Explanation,” in Martijn Blaauw, ed., Contrastivism in Philosophy: New Perspectives. London: Routledge, 11–34. Johnston, Mark (2019). “Why Did the One Not Remain Within Itself?,” Oxford Studies in Philosophy of Religion 9. Oxford: Oxford University Press, 106–64. Kretzmann, Norman (1986). “A General Problem of Creation: Why Would God Create Anything At All?,” in Scott MacDonald, ed., Being and Goodness. Ithaca, NY: Cornell University Press, 208–28. Lipton, Peter (1990). “Contrastive Explanation,” Royal Institute of Philosophy Supplement 27, 247–66. Pruss, Alex (2016). “Divine Creative Freedom,” in Jonathan Kvanvig, ed., Oxford Studies in Philosophy of Religion 7. Oxford: Oxford University Press, 213–38. Rowe, William (2004). Can God Be Free? Oxford: Clarendon Press. Tucker, Chris (2016). “Satisficing and Motivated Submaximization (in the Philosophy of Religion),” Philosophy and Phenomenological Research 93, 127–43. Zimmerman, Dean (2018). “Ever Better Situations and the Failure of Expression Principles,” Faith and Philosophy 35(4), 408–16.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

8 God’s Perfect Will Remarks on Johnston and O’Connor Kenneth L. Pearce

. . . that you may discern what is the good, pleasing, and perfect will of God. Romans 12:2, HCSB

According to classical theism, contingent reality exists because of the free and rational choice of a necessary, perfect being. On the one hand, this appears to be a very promising explanation of contingent reality, since free choice can explain outcomes without taking away contingency. On the other hand, given that a perfect being would not lack or want or need anything of value, it is hard to understand how such a being could have a reason to act (as a being who acts rationally must). It is even harder to understand how or why such a being would create a world like this one with all of its evils. This aporia at the heart of classical theism has recently been addressed by Mark Johnston (2019) and Timothy O’Connor (this volume). Central to Johnston’s treatment is the classical suggestion, developed perhaps most clearly by Aquinas (Summa Contra Gentiles 1.74–88), that God’s will consists in the affirmation of God’s own goodness. According to Johnston, following Aquinas, any creative act God could have performed would have been a different way of affirming God’s own goodness. Furthermore, since God’s goodness is infinite, the addition of a finite created world does not increase the total amount of goodness in existence. As a result, the total quantity of goodness in the created world is, in a sense, irrelevant to God’s decision. Kenneth L. Pearce, God’s Perfect Will: Remarks on Johnston and O’Connor In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Kenneth L. Pearce 2022. DOI: 10.1093/oso/9780192862976.003.0008

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

 ’   

249

According to Johnston, this approach secures a very wide scope for divine freedom. O’Connor disagrees, arguing that even on this approach divine perfection will significantly restrict the scope of contingency. For instance, “fundamentally and ultimately unjust” worlds will be impossible (O’Connor, this volume, § III). I argue that O’Connor does not go far enough. Although Johnston’s strategy can explain why God would create rather than “remaining within Godself,” it makes no progress toward explaining how it is possible that God should not create the best of all possible worlds. The difficulties faced by Johnston’s approach are internal to the Neoplatonic framework he adopts. One of the most famous and influential doctrines of Plato is that all willing is directed at the good (Plato, Gorgias 466d–468e). Neoplatonists identify God with the good. Thus, the fact that God wills (affirms) Godself, and this is the motive of God’s action, does not differentiate God from us. Even our sinful actions are, in a sense, ways of affirming God’s goodness: what we pursue, even in sinful actions, is the good which (perhaps unbeknownst to us) is God. This same idea is at the root of the Thomistic doctrine that what is chosen is always chosen “under the guise of the good” (see Aquinas, Summa Theologiae I-II q1 a6). It is not enough, then, to say that God’s willing is an affirmation of God’s own goodness. We must say that God’s willing is a perfect affirmation of God’s own goodness, in contrast to our deficient willings. We thus reach by a different route one of the central points of O’Connor’s reply to Johnston (O’Connor, this volume, § III): there are better and worse ways of affirming the divine goodness. Not only is God’s willing an affirmation of God’s own goodness, but it is a perfect affirmation. Although Johnston recognizes this point in passing (Johnston 2019, 158), he supposes that, since the result of God’s willing makes no difference to the total quantity of value that exists, the necessary perfection of God’s willing places few or no constraints on its results. This assumption is far from evident. According to Plato, perfection in willing the good resides in the intellect. When we act wrongly, our action is deficient because we do what we think to be best. Since such opinions are often mistaken, doing what we think best does not allow us to obtain the true object of our will, the good (Plato, Gorgias 466e–467a).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

250  .  This approach is applied to God by ibn Sina (Metaphysica ch. 33). According to ibn Sina, the very fact that our will is distinct from our intellect is a defect. Because our will is distinct from our intellect and our intellect is limited and fallible, we often will what is not in fact the greatest good. (Compare Descartes’s doctrine, in the Fourth Meditation, that we err because the scope of our intellect is more limited than the scope of our will.) Since God is perfectly simple, God’s will and God’s intellect are one and the same. For a perfect being, there is no distinction between knowing a thing to be best and willing that thing. Thus, ibn Sina concludes, “[God’s] will is no other than [God’s] knowledge of the best order for all things” (ibn Sina, Metaphysica 68). On ibn Sina’s view, the perfection of God’s willing implies that, necessarily, God’s willing results in the best of all possible worlds. Although the Neoplatonic framework is less explicit in Leibniz, his argument is essentially the same. Leibniz agrees that the fundamental principle here is that “God . . . acts in the most perfect manner” (Leibniz [1686] 1989, § 1, emphasis added). He claims, however, that the perfection of God’s act implies the perfection of God’s product (Leibniz [1686] 1989, § 3). If God’s product were improvable, God would be dissatisfied with God’s own creative activity (Leibniz [1710] 1985, § 201). Hence, the Neoplatonic thesis that God’s willing is the affirmation of God’s own goodness does not, by itself, solve the problem of divine freedom in creation or the problem of evil. God must affirm God’s own goodness perfectly. This would seem to limit God’s possibilities for willing, just as O’Connor says. The central point of Johnston’s argument is that, since God’s own value is unsurpassably infinite, God’s creative activity does not—and cannot—change the total quantity of value that exists. What follows from this observation is that the perfection of God’s creative activity cannot be judged in a consequentialist fashion by the total amount of goodness that results. Rather, as our discussion of the nature of God’s willing also suggests, it must be the willing itself that is perfect (Johnston 2019, 158). Ibn Sina and Leibniz suppose that perfect willing must have a perfect product. Note that this brand of optimism does not fall prey to Johnston’s worries about “meliorism.” We do not suppose that God is

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

 ’   

251

motivated to increase the total amount of goodness in existence (an impossible task). Rather, God perfectly affirms God’s own goodness, and the perfection of this affirmation is seen in the perfection of the world that is thereby created. This yields a principled reason for attending only to the goodness of the created world, excluding the infinite divine goodness. (This may also provide a reply to Johnston’s objections to meliorism in the human case, since we may rationally desire to affirm the divine goodness by participating in the perfection of God’s creative product.) At the same time, the focus on the perfection of God’s activity shows that this supposition made by ibn Sina and Leibniz—that perfect activity must have a perfect product—is contestable. Indeed, Leibniz was forced to defend this supposition explicitly because it had been challenged by Malebranche ([1680] 1992, §§ 1.13–14). Theists who endorse the broadly Neoplatonic framework, whether they accept or reject optimism, are committed to the claim that God’s willing, God’s creative activity, is perfect. A defense of contingency in divine creation, then, must show that any of a wide variety of worlds—or perhaps none at all—might have resulted from God’s perfect willing. Similarly, a solution to the problem of evil would make it plausible that this world could have been produced by God’s perfect willing. To see whether such solutions are available, we would need to understand in what the perfection of God’s activity consists. One answer might be that God’s activity is perfect precisely insofar as it is a wholehearted affirmation of the highest good (i.e., Godself), rooted in the most complete knowledge of that good (i.e., Godself). This, however, still leaves us with what I take to be O’Connor’s deepest questions: what are the possible results of such activity? Do they include the existence of anything distinct from God? Do they include worlds like this one that contain enormous quantities of evil? This is closely connected to the disagreement between Leibniz and Malebranche: must perfect creative activity have a perfect product? It is worth noting that certain common analogies for divine creation tend to focus our attention on God’s product rather than God’s activity, favoring Leibniz over Malebranche. For instance, it is very natural to suppose that the excellence of an act of carpentry or painting is judged

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

252  .  primarily, if not exclusively, by the excellence of its product.¹ If it is the perfection of God’s activity with which we are concerned, then God might be better analogized to a storyteller (Lebens 2015, 2017; Goldschmidt and Lebens 2020) or a dancer (Pearce 2017b, 250). I have argued elsewhere that these kinds of analogies also provide a better model for the metaphysics of divine creation: as the dance is nothing over and above the activity of the dancers, so also God’s creation is nothing over and above God’s creative activity (Pearce 2017a, 2017b). The storyteller analogy, however, suggests a rather shocking answer to our question about the product of God’s creative activity. The suffering of the characters in a story in no way detracts from the excellence of the act of storytelling. A storyteller is under no obligation to ensure that the characters “live happily ever after,” since there are distinctive aesthetic values to be found in the genre of tragedy. Indeed, some neo-Thomists have explicitly held that, as a storyteller has no moral obligations to the characters in her story, so God has no moral obligations to us and the problem of evil therefore dissolves (see Ross 1969, ch. 6). This conception of God is clearly religiously inadequate. The Abrahamic religions (at least on most interpretations) are committed to the idea that God cares for us and wills our good. This religious commitment is independent of the controversial philosophical question of whether God has a moral obligation to care for us (and, indeed, whether God has any moral obligations at all). But is there philosophical reason to reject this conception? Progress can be made by combining the storyteller analogy with the observation that God’s willing must be an affirmation of God’s own goodness. This gives us a distinctive way of understanding an idea discussed by both Johnston and O’Connor, that God creates to “show forth God’s glory.” The story God tells should be a story that expresses God’s own character and values, and not only God’s excellence as a storyteller. Such a performance would be an appropriate way of expressing God’s affirmation of God’s own goodness, as the Neoplatonic

¹ Malebranche thus does himself no favors by comparing God to a craftsman in this very context! (Malebranche [1680] 1992, § 1.13).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

 ’   

253

tradition has it, and such a story would involve God’s care for the story’s characters. This is, admittedly, more of a picture than a theory. Nevertheless, it does seem to provide an explanation of God’s motivation for creating and God’s motivation for caring about the well-being of creatures. One worries, however, that we may have arrived back at the optimism of ibn Sina and Leibniz. If there is a unique best creative act for expressing God’s goodness, then it seems that, necessarily, God chooses this act. On the other hand, if there is not a unique best act, then we must explain how God’s choice is not merely random or capricious. In other words, shifting our attention to the goodness of God’s act makes little or no difference in our struggle to understand how God’s choice can simultaneously exhibit perfect freedom, perfect goodness, and perfect rationality. Furthermore, by recovering God’s care for creatures we also revive the problem of evil. I conclude that, while the Neoplatonic framework discussed by Johnston and O’Connor provides a promising answer to the question of why God would create at all, it makes very little difference to the problem of divine freedom or the problem of evil. We cannot avoid acknowledging that some acts available to God are better than others and, once we have acknowledged this point, it is difficult to see how we avoid the (deeply implausible) conclusion of ibn Sina and Leibniz that, necessarily, God creates the best of all possible worlds.²

References Aquinas, Thomas. 2018. Summa Theologiae. Edited by Laurence Shapcote. Revised by The Aquinas Institute. Steubenville, OH: Emmaus Academic. . Aquinas, Thomas. 2019. Summa Contra Gentiles. Edited by Laurence Shapcote. Revised by The Aquinas Institute. Steubenville, OH: Emmaus Academic. .

² I thank Timothy O’Connor for helpful comments on a previous draft.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

254  .  Descartes, René. (1641) 1984. “Meditations on First Philosophy: In Which Are Demonstrated the Existence of God and the Distinction Between the Human Soul and the Body.” In The Philosophical Writings of Descartes, translated by John Cottingham, Robert Stoothoff, and Dugald Murdoch. Vol. 2. Cambridge: Cambridge University Press. Goldschmidt, Tyron, and Samuel Lebens. 2020. “Divine Contractions: Theism Gives Birth to Idealism.” Religious Studies 56 (4): 509–24. ibn Sina, Abu ‘Ali al-Husayn. 1973. The Metaphysica of Avicenna (Ibn Sīnā), translated by Parviz Morewedge. Persian Heritage Series. New York: Columbia University Press. Johnston, Mark. 2019. “Why Did the One Not Remain Within Itself?” In Oxford Studies in Philosophy of Religion. Vol. 9. Oxford: Oxford University Press, 106–64. Lebens, Samuel. 2015. “God and his Imaginary Friends: A Hassidic Metaphysics.” Religious Studies 51 (2): 183–204. Lebens, Samuel. 2017. “Hassidic Idealism: Kurt Vonnegut and the Creator of the Universe.” In Idealism: New Essays in Metaphysics, edited by Tyron Goldschmidt and Kenneth L. Pearce. Oxford: Oxford University Press. Leibniz, G. W. (1710) 1985. Theodicy: Essays on the Goodness of God, the Freedom of Man, and the Origin of Evil. Edited by Austin Farrer. Translated by E. M. Huggard. La Salle, Ill.: Open Court. Leibniz, G. W. (1686) 1989. “A Discourse on Metaphysics.” In Philosophical Essays, edited and translated by Roger Ariew and Daniel Garber. Indianapolis: Hackett. Malebranche, Nicolas. (1680) 1992. “Treatise on Nature and Grace. Selections.” In Philosophical Selections, edited by Steven Nadler. Indianapolis: Hackett. O’Connor, Timothy. This volume. “Why the One Did Not Remain Within Itself.” Pearce, Kenneth L. 2017a. “Counterpossible Dependence and the Efficacy of the Divine Will.” Faith and Philosophy 34 (1): 3–16. Pearce, Kenneth L. 2017b. “Foundational Grounding and the Argument from Contingency.” In Oxford Studies in Philosophy of Religion. Vol. 8. Oxford: Oxford University Press, 245–68. Plato. 1987. Gorgias. Translated by Donald J. Zeyl. Indianapolis: Hackett. Ross, James F. 1969. Philosophical Theology. Indianapolis: Bobbs Merrill.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

9 Divine Self-Manifestation Reply to O’Connor and Pearce Mark Johnston

Let me begin by thanking Timothy O’Connor and Kenneth Pearce for their insightful comments on “Why Did the One Not Remain Within Itself?”¹ O’Connor suggests that my account of creation as a rationally uncoerced, and so an utterly free and contingent act of Divine self-manifestation nonetheless restricts the scope of contingency as it is ordinarily understood. In what follows I indicate just how that might be so, in a way that should be unsurprising given the overall account. He also suggests that God might have created out of generosity. I offer an alternative view to the effect that conditional upon creating us, God is generously disposed to us. And I indicate why I do not follow O’Connor in thinking that the Neoplatonist thesis of Pseudo-Dionysius—that Goodness automatically diffuses itself (like a gas?)—is helpful in examining the lineaments of creation. Pearce avers that the account of creation as a rationally uncoerced, and so an utterly free and contingent act of Divine self-manifestation leaves us with the familiar problem of evil in the form of the observation that that this is not the best of all possible creations. My own view, developed briefly below, is that God’s indefective self-manifestation involves the creation of free rational beings as perfect as they can be compatible with their being creatures. That raises the question of the etiology of evil; in particular how any such free rational will could rationally, i.e. clearheadedly and prudentially, valorize its own good over the Good. As I have ¹ Oxford Studies in Philosophy of Religion Vol. 9 (Oxford: Oxford University Press, 2019), pp. 106–67.

Mark Johnston, Divine Self-Manifestation: Reply to O’Connor and Pearce In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Mark Johnston 2022. DOI: 10.1093/oso/9780192862976.003.0009

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

256   argued elsewhere, the whole problem of evil looks very different—indeed considerably more chilling—if we start from God’s indefective selfmanifestation, and enquire after the source of corruption within the manifestation that is the first, i.e. direct and immediate, creation.²

Real Contingency Some things are really possible, i.e. compatible with the essential natures of existing things, and not simply conceptually consistent suppositions, from which no contradiction can be derived. But, contra Spinoza, not everything that is really possible is necessary; not every real possibility follows from the essential nature of necessary beings, if any there be. There are real contingencies. Actuality could have varied in myriad ways, though perhaps not as widely as many theorists of possibility have supposed. The recognition of real contingency prompts the intellect to look for contingency’s ground or ontological explanation. That was Immanuel Kant’s insight in his pre-critical work The Only Possible Ground of Proof [Beweisgrund] for a Demonstration of the Existence of God. In a recent incisive investigation of the structure of Kant’s “demonstration” Nicholas Stang offers this initial sketch: Facts about what is possible cannot be “brute”; they require grounds. [Kant] identifies the grounds of possibility with the powers of existing substances; an object or state of affairs is possible in virtue of the power of some substance to produce that object or cause that state of affairs to obtain. He goes on to argue that all possibilities must be grounded in a single necessarily existing substance, God. If God did not exist, nothing would be possible.³

The same thought about real contingency requiring a ground can be found in a very different intellectual setting, namely David Lewis’s On the ² The 2019 Gifford Lectures (St Andrews), Lecture 5: “On the Etiology of Evil.” ³ “Kant’s Possibility Proof,” History of Philosophy Quarterly 27 (2010), p. 275.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

257

Plurality of Worlds. Lewis is no theist in any ordinary sense. But he insistently complains against “magical ersatz” theorists of modality— theorists of what is merely possible and what is necessary—who help themselves to those notions without offering an account of their basis in reality, a basis that Lewis attempts to provide by way of his realm of (at least) Beth 3 existing concrete “worlds.” Even those who believe that the structure of Lewis’s proposed ground for modal truth itself reeks of real contingency—why Beth 3, etc., etc.?—and so is not in the end a satisfactory ground for real contingency, should recognize the significance of Lewis’s attempt to find a ground for contingency. Lewis at least faced a question avoided by those who merely provided abstract items— “possible worlds”—in a model theory that facilitates proofs of the consistency and completeness for this or that system of modal logic. Lewis’s massive plurality of insolated concrete “places” is an attempt to articulate a ground for what he took to be the truths concerning necessity, possibility, and contingency. Commentators have noted Kant’s later diffidence concerning his precritical proof, explaining it in part by the gap between an ens realissimum—a “most real being” that somehow grounds its own existence and that of real contingency—and the God who emerges in Kant’s Religion Within the Limits of Reason Alone—the author and upholder of an ethical economy in which happiness is ultimately made proportionate to virtue, thereby serving the demands of justice and realizing the hope for the Summum Bonum, a hope which Kant understood to be internal to the ethical enterprise. One of the things going on in “Why Did the One Not Remain Within Itself?” was the attempt to show how we might begin to understand how the ground of contingency might also be the guarantor of ethical life.⁴ Offered there is an account of the ground of real contingency that leads us to understand the “ens realissimum” as a person, indeed the eminent exemplar of Personhood or Rational Willing—a personal creator ⁴ In particular, see the discussion of how God’s authoritative commands might fill the practical void opened up by the recognition that reality as a whole is utterly unimprovable; see pp. 158–64 of “Why Did the One Not Remain?” See also my “Is Hope for Another Life Rational?” in Paul Draper, ed., Current Controversies in Philosophy of Religion (New York and London: Routledge, 2019), pp. 47–68.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

258   creating for a reason, a reason which was completely adequate but non-coercive, thereby allowing for the grounding of contingent creation as contingent. Completely adequate reasons for an act are justifying reasons that are not outweighed by the reasons for any alternative to the act; whereas coercive reasons for an act are justifying reasons that are weightier than the reasons for any alternative act. If in creating, God had a libertarian free choice, a choice not even coerced by the force of his reason to create, then we would have a fundamental ontological explanation of the contingency of creation. The proposal was that what is respectable in the intellect’s search for a ground of real contingency can captured by the following variant on the Principle of Sufficient Reason. Where an autonomous fact is a fact that arises wholly from the essences of the items involved, we have: THE BASIC IDEA: All non-autonomous facts are explicable, in the sense that there must be completely adequate reasons for them. THE CRUCIAL GLOSS: However, these reasons can take different forms. Sometimes such reasons are found in a necessitating ontological explanation, which would show how derivative things must be as they are, given more fundamental things; but alternatively there can be completely adequate but non-coercive practical reasons for choice and intentional action, reasons which provide a non-necessitating explanation of the choice and corresponding action via those antecedent reasons. So we can ground contingency as such if we see it as the outcome of a free rationally uncoerced creative act of a being which is itself an autonomous existent, or better is Existence itself—participation in which via creation is the ground of the existence of all else that exists. After developing this argument “Why Did the One Not Remain?” enquires as to the type of reason God had for creating—to be distinguished from God’s determinate plan, as it may be a loving or generous salvific plan, conditional on his creating. God was no meliorist. In creating he was not trying to make things better, for they were already unsurpassably good. In creating, God was not perfecting his nature. Nor was he advancing his own self-interest; i.e.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

259

fulfilling some need of his which otherwise would have been unmet. Nor was he under an authoritative command (explicit or implicit) to create, or if he was it can only have come from himself, which simply pushes the question back to the reason for that command. Nor was he morally obliged to create. To whom? And if there are moral obligations that are not obligations to anyone, then they still can be ruled out. Moral reasons to create would be coercive reasons, i.e. more weighty than any reason to refrain from creating. A perfect rational will would by its nature act in accord with them, so that creation then would be necessary. But that dissipates the distinctive nimbus of explanatory grounding which surrounds the ontology of traditional theism, i.e. the grounding of the contingency of creation in the genuinely free choice of a necessary being acting on a completely adequate, but non-coercive, reason for creating. Contingency implies real possibility; the contingent could be or not be. But what is the ground of real possibility? It cannot be a contingent ground, since that simply reproduces the question at the lower level. It must be a necessarily existing ground having its nature necessarily because essentially, and it must be a necessarily existing ground whose grounding of contingency is not the necessitation of what then would emerge as only apparently contingent things, as in the metaphysics of Spinoza. It is God and his free—even rationally free— choice to create which grounds the contingent as contingent. Here “Der einzig mögliche Beweisgrund zu einer Demonstration,” as Kant might have put it, goes beyond the postulation of a mere ens realissimum. God must be a free rational will in order to ground the contingent as contingent, and so ground those real possibilities that are not also necessary. Nor did God create us because he “already” loved us. For a necessary part of the ground of particular loving or generous relations with creatures was established by creating them, and so the existence of such relations could not have provided the reason for creating, since the existence of such relations presupposes creation. Though God’s plan conditional on his creating us may be a loving plan for us, there is what we might call a proleptic fallacy in supposing that his reason for creating was that he loved us, or was generously disposed to us. Nor did God create from a reason of generic generosity toward whatever creatures there might be. As “Why Did the One Not

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

260   Remain?” makes explicit, generic generosity is for us an imperfect duty, we are not required to be as generous as we can be in every circumstance because of the resultant life-diminishing costs of always and everywhere being so generous. There are no such life-diminishing costs for God, and so if he created from reasons of generosity those reasons would comprise a perfect duty, a duty which he, as the Eminent Exemplar of Rational Willing, would find coercive. He would then have had coercive ethical reasons to create, and so again the distinctive nimbus of explanatory grounding that surrounds the ontology of traditional theism would be dissipated.⁵ Indeed it would also follow that every being whose life God could foreknow—or “overknow” from a perspective outside of time—as meeting some appropriate standard, as it might be that the life was overall worthwhile, or involved final turning toward the Good, would then be a necessary being. But now God’s free generosity, his gratuitous favor to each such being in creating them disappears from the picture. He was rationally coerced to do it. By proceeding in this way through an inventory of the various types of reasons for action, and their requirements and entailments, we arrive by rational reflection at the traditional Catholic view, announced as de fide by Vatican 1 and reaffirmed in the Catechism.⁶ We should understand creation as an expressive act. God’s reason for creating was to manifest his glorious nature: to manifest his Existence, his Goodness, his Power, and his Free Rational Will—the very features that the grounding account of real contingency invoke. “Why Did the One Not Remain?” outlined three ways in which this reason, though completely adequate, might not amount to a coercive reason to create. The first two ways involve “offsetting” reasons. There might have been an equally weighty reason in favor of remaining, arising ⁵ O’Connor does not engage with these considerations concerning generosity as they are set out on pages 149–50 of “Why Did the One Not Remain?” ⁶ “If anyone . . . denies that the world was created for the glory of God, let him be cut off [from the community of the faithful and the saving grace of the sacraments].” Session 3 of Vatican 1 “Dogmatic Constitution on the Catholic Faith: 1. On God the Creator of All Things” Canon 5. And the Baltimore Catechism presents the following dialog: Q. “Why did God make us?” A. “God made us to show forth His goodness and to share with us His everlasting happiness in heaven” (Baltimore Catechism No. 2 at Project Gutenberg, Question 3.)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

261

for example from God’s foreknowledge that creation would be badly tarnished by free powerful created beings choosing evil, i.e. the valorization of their own good over the Good. Alternatively, there might have been an incomparable or incommensurable reason on the side of remaining, for example the avoidance of polluting reality by the pointlessly gargantuan and entropic character of the material universe—as it has now been revealed by astrophysics. Since such offsetters are wholly speculative as to their nature and weight as reasons, a third way is explored and then endorsed: Given the unsurpassable character of God’s goodness there is conceptual space for the idea that God’s extra, and perhaps not offset, reason for creating might not have given him more of a reason to create than to remain. In this way he could have had a completely adequate reason to create which, thought not offset, was nonetheless not a coercive reason to create. From this sheer freedom contingency arises. O’Connor is unsure about my attitude to incomparability. It is not that I reject the idea. It is just that I do not think that it should be our only model of how God could have had a completely adequate but noncoercive reason to create. The third model is to be favored because it derives from God’s nature as unsurpassable Goodness. And unlike the appeal to supposed balanced reasons for creating and for remaining, it avoids the implication that God just opted to create. Instead he had a justifying and explanatory reason to create that was not offset by the force of any reason to remain. Contrast this account of freely chosen Divine self-manifestation with Neoplatonic emanationism. Pseudo-Dionysius, a fifth and sixth-century emanationist who reads back his Neoplatonism into early Christian teaching, using the literary tactic of presenting his works as authored by St Dionysius, a first-century Athenian who was converted by Paul, says that Goodness diffuses itself. At a crucial point in his discussion of creation, Aquinas, unaware of the tactic, quotes Pseudo-Dionysius as having almost apostolic authority. As Norman Kretzmann and other commentators have noted this seems to reinforce Aquinas’s attraction to—an attraction short of outright endorsement of—the thesis that it was necessary that there be some creation or other. (Vatican I condemns this as a view that makes the holder subject to excommunication latae

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

262   sententiae, i.e. by the very holding of the view, which is a somewhat awkward position for a great and worthy saint to be in.) The diffusion thesis concerning creation does not invoke a reasonbased account. There is then no impetus to think of Goodness itself as a person whose rational will is the source of creation, let alone as the eminent exemplar of personhood. Since rational willing is not in play in creation, we cannot measure the creation we observe against God’s reason for creating and so discern just how evil could have arisen within it. Consequently, the obvious defects in the creation that we do observe remain surds, as evidenced in Middle-Platonic, Christian “Sethian” literature such as The Secret Revelation of John.⁷ The myths describe the weakening of the One’s originally pure emanation as it proceeds through its successive stages, as if there were an independent resistant force of anti-axiological friction opposing the original manifestation of Goodness. But whence the resistant force? Absent the resistant force we might worry how the emanationist can account for notoriously catastrophic progressions in the emanation of the One, such as the mythic progression from the marvel Barbelo to the disaster Yaldabaoth, the alleged creator of the material universe? The appeal to the only slightly corrupt will of Sophia, the mother of Yaldabaoth, itself testifies to the antecedent effects of an independent force resistant to the emanation of Goodness itself. Given the myth a natural thing to say is that these more remotely emanated archons or ruling agents, Sophia and Yaldabaoth, chose freely and hence blamably to valorize their own good, the enjoyment of the exercise of their power, in a way at odds with the Good. But how did that capacity for free choice arise, if it was not prefigured in the One from which these emanations arose? This is the obverse side of the Imago Dei doctrine: if God created free creative wills in his imago—his image and

⁷ See Karen Leigh King, The Secret Revelation of John (Cambridge, Mass.: Harvard University Press, 2006). On the connection with Platonism, see Philip Tite & John Turner, “Sethian Gnosticism and the Platonic Tradition,” Journal of Biblical Literature 123 (2005). See also Jay Bregman & Richard T. Wallis, Neoplatonism and Gnosticism (SUNY Albany Press, 1992) and Karen Leigh King, What is Gnosticism? (Cambridge, Mass.: Harvard University Press, 2003), pp. 156–62.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

263

likeness—should we not expect God to be a free creative will, not a mere impersonal emanating source?

Alethic Modality Refigured It is not a reasonable default to ask an account of the ground of real possibility to coincide with what is possible merely in the sense of containing no internal conceptual incoherence. There is no internal conceptual incoherence in supposing that it could have been that all that contingently existed was a single common or garden chicken. Lewis’s account of the ground of ontological possibility counts this as a real possibility! On the present account of the ground of contingency this is not a real possibility. A creation simply consisting of a single chicken would not be an adequate manifestation of God’s nature as the eminent exemplar of Perfect, Powerful, Free Rational Willing, unless there is something very special about chickens that is entirely lost on us. Nor is there reason to believe that there is only one possible adequate manifestation of God’s nature as the eminent exemplar of GoodAffirming,⁸ Powerful, Knowing, and Free Rational Willing. The adequacy-making features of one manifestation may be balanced by or alternatively incomparable with those of another manifestation. But of course, if the ground of real possibility is God’s freely chosen selfmanifestation then the scope of real possibility will be massively restricted relative to anything like the Lewis view. There is the really possible world consisting of God and no creation at all—the world where God remains within himself—along with the really possible worlds involving distinct adequate manifestations of God’s nature as the eminent exemplar of Good-Affirming, Powerful, Knowing, and Free Rational Willing. It is very natural to suppose that in each adequate manifestation created free rational wills appear and make genuinely free choices, where the available options for choice make for a dense garden

⁸ Here I take the traditional view that God necessarily and completely affirms his own Goodness. Since he is the Good, he does not face the fundamental choice confronting his rational creatures: to valorize the Good or one’s own good.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

264   of contingent forking paths. So there are still a myriad of real possibilities on this account. But in every non-empty world there is a first creation, and the present account of freely chosen self-manifestation helps us get some grip on what forms that might take.

The First Creation As “Why Did the One Not Remain Within Itself?” emphasizes (p. 158) there can be no defect in God’s act of self-manifestation. As an act, it is perfect. So it is natural to think that the expressive manifestation of God’s Good-Affirming, Powerful, Knowing, and Free Rational Willing would involve the immediate creation of powerful free rational wills. Since there could be no defect in God’s act of self-manifestation the immediate realm comprised of the first created free rational wills could be expected to be as perfect a manifestation as it could be compatible with those wills being creatures. That naturally suggests a plenitude of kinds of first created wills, with each created will being perfect of its kind, i.e. showing no defects in its kind-specific intellect, will, and power. The plenitude would be full in the sense of being a fully adequate manifestation of God’s nature. Given God’s nature, there is no reason to think that the creation of either matter or its cousins that inhabit the extensive fields of other “Multiverses” would be part of the self-expressive manifestation of God’s nature found in his first creation. Matter, extension, and their ilk are no part of God’s nature. Of course, when it comes to animal willing, the considerations which present to any animal’s will essentially come by way of its sensing, emoting, imagining, and remembering, and these are all capacities which require a material substrate for their operation. But in speaking of the full plenitude of kinds of first created wills—each created will being perfect of its kind, i.e. showing no defects in its kindspecific intellect, will, and power—we are not speaking of the animals, human or otherwise. In so denying that the first created wills would need or have any material embodiments, we are in a position to see just how a variety of no-best creation arguments stemming from William Rowe’s original

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

265

version go wrong.⁹ One strategy for imagining a creation “better” than any supposed creation is to increase the number of created loci of the targeted value of the supposed creator. Another is to increase the targeted value at the loci. There is no largest number, so the loci of the targeted value can be increased without limit. And if the targeted value is thought of merely as created goodness, it may appear to have no upper limit. So given any proposed creation we are able to conceive coherently a better creation, by one or other of the two strategies. But then it may seem that if God could have generated a better created realm and did not, there is some defect in his act of creation, and so in him. So the creator is not the highest one, and so not God. As against this adaptation of a now familiar worry, recall that God, in creating, is not a meliorist. He is not aiming to make things better in the sense of incorporating more good. For God is the Good, and so constitutively and unsurpassably good. Things are “already” as good as they can be. Creation, no matter what goods or absences of good it contained, could not make things better or worse. The dimension along which to evaluate creation is the dimension of its fit with the Divine intention to give expression to the Divine nature. If there is a lack of fit there, then the creator is not God, for his act of creation is defective. God’s reason for creating is to manifest his nature. His nature is to be Goodness and Power and Knowledge and Free Rational Willing itself. So the way in which his Goodness is manifested in creation is itself constrained by the co-manifestation of these other features of his nature. The full plenitude of kinds of first created free wills, each created will being perfect of its kind, i.e. showing no defects in its kind-specific knowledge, will, and power, is a non-defective fit with God’s creative intention. No kind of free created will that is capable of willing without reliance on any material embodiment is omitted, and each created will is perfectly good in the ontological sense of lacking no perfection which is owed to it as the kind of thing it is. So there is no scope to add more good at each of the loci of the targeted value.

⁹ William L. Rowe, “The Problem of No Best World,” Faith and Philosophy 11 (1994), 269–71.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

266   Still, there may seem to be scope for the creation of more loci, and in two ways: first by directly creating materially embodied wills like us, and second by multiplying ad infinitum the number of the free created beings within each kind that is to be found within the full plurality of the first creation. Even if the kinds of free rational wills in question are infima species, i.e. utterly determinate kinds of free rational wills, why could there not be numerically distinct twins, or triplets or quadruplets or . . . of each such utterly determinate kind? Against the first way, although it is within God’s power to cause matter to be, God has no reason to do this if his intention in creating is to give expression to his nature. For materiality, spatiality, temporality, and their ontic cousins are no part of his nature. Matter must have a different, presumably derivative, source. Astrophysics tells us that matter is not by its nature favorable to complexity, let alone life and mentality. The parameter “knobs” of the basic laws of matter have to be “tuned very finely” to allow for anything more complex than a helium atom. Even then complexity is a temporary aberration in Spacetime. On the now preferred “endless future” model almost all of Spacetime consists of thin gas, getting thinner and thinner in saecula saeculorum. If there was finetuning it was not aiming at the manifestation of God’s nature as GoodAffirming, Powerful, Knowing, and Free Rational Willing. Or if it was, it was ineptly done, and so either way it was not part of God’s direct selfmanifestation, the first creation. Here is a new empirical fact which theology must now assimilate, if it is to have a serious future. Astrophysics shows that the heavens, if understood as the immensely vast regions of Spacetime, do not proclaim the glory of God. Dwell briefly on a few of the more detailed disappointments Spacetime offers to friends of complexity, life, and mentation. It took more than five billion years for even the most primitive first planets, stable clumps of boiling elements, to form. Then there is the fact that great preponderance of the element that is best suited to combine complexly with other atoms, namely carbon, is locked up in red dwarf stars during the planetary period, when it could otherwise combine to create complexity. Then there is the unhelpful, continuous, and often catastrophic asteroid bombardment of planets in the so-called “goldilocks” zones, i.e. orbits where things resembling complex organisms as

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

267

we know them have at least a minimal chance of arising. To take one case, the Late Heavy Bombardment of our solar system by asteroids and failed planets lasted between 20 million and 200 million years. The larger-scale vista is even less encouraging; for the total entropy or disorder of the universe—already 100 times greater than was previously estimated¹⁰—is continually increasing. Complexity appears only for what turns out to be a cosmic instant and only in eddies swirling against the cosmic tide. In the large scale, the complexity of the planetary period is a temporary anomaly, a brief prelude to unendingly thinning gas. None of this looks anything like the outcome of Power itself giving expression to its nature as Good-Affirming, Knowing, and Free Rational Willing. Thus, the endless improvability of Spacetime-like realms is not a relevant consideration given the present account of Divine self-manifestation. What then of the idea that the plurality of first created rational wills, each perfect of its kind, could be improved by multiplying the individual cases or examples of the utterly specific kinds in question, without limit? Against this, there is the hylomorphic thought deployed by Aquinas in his discussion of the angels; namely that the ground of difference solo numero, the ground of there being many duplicates of an utterly specific kind, is difference in their material embodiments.¹¹ Aquinas famously concludes that, since they can only be individuated by their forms, each angel is of a different species. Then there is the Kripkean thought that the ground of difference solo numero would lie in the distinct material origins of the duplicates. But neither of these grounds applies to the free rational wills which make up the first creation. They are not embodied, and so they have no material origins. And since God creates the pleroma ex nihilo, “all at once” and not successively in time, there is no room for difference solo numero within a kind deriving from different moments of coming into being in some successive creation.

¹⁰ Chas A. Egan and Charles H. Lineweaver, “A Larger Estimate of the Entropy of the Universe,” ArXiv/Astrophysics (2010). ¹¹ It is important to remember that the winged angels may be no more than florid stand-ins for the first created rational wills. As Gregory the Great emphasized, “angel”—i.e. messenger or representative—is the name of an office or function, not the name of a nature. And “archon” and “aeon” have acquired a cultic connotation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

268   Moreover, and independently of these genuinely engaging subtleties in the theory of individuation and numerical distinctness, there remains the fact that God is not a meliorist. He is not aiming to maximize value. It is not a defect in Leonardo’s expression of his creative talent that he did not paint the Mona Lisa twice or thrice or . . . True, there is a thin conception of artistic value on which Leonardo failed to maximize artistic value in not so doing. But he was not aiming to maximize artistic value. That is what hacks and gallery entrepreneurs do. Likewise God was not aiming to maximize the number of good beings. He was aiming to give expression to his nature as Goodness, Power, Knowledge, and Free Rational Willing Itself. In first creating the “pleroma”—the full plenitude of kinds of created wills that require no material embodiment, with each created will being perfect of its kind, i.e. showing no defects in its kind-specific intellect, will, and power, and with none of the kinds being defective in themselves—God achieved that end. The first creation was thus indefective relative to God’s reason for creating. There was no defect in God’s creative act. Hence there is no good Rowe-like a priori argument against God being the first creator from the endless improvability of any creation. The first creation, the pleroma, was unimprovable along the relevant dimension of giving expression to God’s nature as the eminent exemplar of GoodAffirming, Powerful, Knowing, and Free Rational Willing. Obviously, this raises many crucial questions, some of which are addressed in my recent Gifford Lectures. Where and how did things go wrong with the first creaturely wills? How did the pointlessly gargantuan material universe come to be? Why does God abjure when it comes to the subsequent depredations of those among the first creation who fell by valorizing their own good over the Good? Is it because the first creaturely wills only face an uncoerced free choice as to whether to valorize the Good over their own good if they know, and know that they know, that God will abdure, i.e. not undermine the self-regarding projects of those first creaturely wills who choose wrongly, even when those projects involve tempting more vulnerable others, such as the human beings, to join them in their fall? And more generally, what would be involved in recovering a theology of creation that avoided the absurd anthropocentrism of making

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

 -

269

it all about H. sapiens, or the Earth, or Spacetime, or—of late—a supposed Multiverse of Spacetime-like universes? Tellingly, this sequence consists in expanding the horizon of reality outwards from our point of view, rather than beginning with God’s own reason for creating.¹² Perhaps the creation of human wills by God was merely an exploitation of a foreseen but adventitious anomaly of the appearance in Spacetime of neural complexity sufficient to provide expressive embodiments reflecting the character of those wills. However, given the present limitations of space, in responding to O’Connor and Pearce, I have here only attempted to indicate what the conception of freely chosen Divine self-manifestation implies concerning the scope of real possibility, and how that conception might distinctively reframe the question of the origin of evil as, in effect, the question of the coherence of that fall into evil traditionally regarded as antecedent to any human fall.¹³

¹² The best excuse for thinking of ourselves and our extended perspective being at the center of creation lies in a natural misconstrual of the doctrine of the Incarnation: surely if God became a member of the species H. sapiens, then H. sapiens is somehow central. Behind the excuse is a very poor inference. For it narcissistically supposes that the salvation history of human beings is salvation history period. But worse, it elides the fact that the Incarnation was an act of egregious—even, as Kierkegaard put it, “absurd”—generosity directed at the utterly underserving. To construe it as a response to our prior importance in creation is deeply mistaken. ¹³ See Augustine, On the Free Choice of the Will, On Grace and Free Choice, and Other Writings, trans. Peter King (Cambridge: Cambridge, 2010), Anselm, Three Philosophical Dialogues: On Truth, On Freedom of Choice, On the Fall of the Devil, trans. Thomas Williams (Indianapolis: Hackett, 2002), Thomas Aquinas, The De Malo of Thomas Aquinas, trans. Richard J. Regan SJ (New York: Oxford, 2001). For a brilliant overview of the tradition and a discussion of the distinctive position of Scotus, see Giorgio Pini, “What Lucifer Wanted: Anselm, Aquinas, and Scotus on the Object of the First Evil Choice,” in Oxford Studies in Medieval Philosophy Vol. 1 (Oxford: Oxford University Press, 2013), pp. 61–82.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

10 To What Extent Must Creatures Return to the One? Caleb Cohoe

If there must be an ultimate being, incomparable in goodness and entirely unsurpassable, why are there any other entities? In his article, Timothy O’Connor responds to a recent piece by Mark Johnston addressing this issue, which arguably goes back at least to Parmenides. To put it in its ancient form: if Being (in Parmenides’ terms) or the Good Itself (in the terms of the Platonic tradition) is so good that nothing can add to it, what reason can there be for anything non-identical to it to exist? I agree with Johnston that Plotinus, together with many thinkers from the Abrahamic traditions, rejects the meliorist answer on which creation increases the amount of goodness in the world. Both Johnston and O’Connor explore answers on which creation is a manifestation of God’s goodness to the glory of God, not a project to add more value to reality, as if that were possible. Before turning to one point of agreement with O’Connor and one problem for his approach, I wish to draw attention to a lacuna in both pieces. Both focus on how a contingent creation can be compatible with both a principle of non-coercion (God has reasons to create, but not ones that impel or reduce God’s freedom) and a principle of sufficient reason, in order to avoid a Spinozistic universe in which everything is necessitated. Johnston and O’Connor both move directly from the necessary being to considering reasons for the existence of contingent reality. Yet historically, many of the most influential thinkers in the perfect being tradition have addressed the question of why the One does not remain within itself by including both necessary and contingent elements within Caleb Cohoe, To What Extent Must Creatures Return to the One? In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Caleb Cohoe 2022. DOI: 10.1093/oso/9780192862976.003.0010

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

         ? 271 the set-up of reality. The metaphysical alternatives are not just “a single necessary being plus an entirely contingent reality” or “an entirely necessitated cosmos.” Instead, thinkers such as Plotinus, Ibn Sīnā, and Thomas Aquinas recognize, in addition to a contingent creation, a necessary emanation from the first principle, albeit in quite different ways. For Plotinus, even if the order of material reality involves contingency, Nous (Intellect) and the Forms must flow forth from the One, as water from a fountain. For Aquinas, the Son is the internal procession of the Father, the eternal intellectual emanation that is the “the word of the Heart.”¹ We should note a real divergence here between versions of perfect being theology that involve some sort of necessary emanation from the ultimate being in addition to the contingent creation of the material universe and those that deny any such emanation. Those who allow for some sort of necessary emanation can appeal to it to accommodate the thought that goodness and being are diffusive. The One is already expressing itself, on Plotinus’ view, by emanating Nous, the perfectly self-aware Being that contains all Forms. On the classical Christian view, God the Father is always emanating an internal Word that is “light from light, true God from true God.”² This means that the Good is already manifested in the most important and perfect way (albeit a way much different than contingent material reality). This can accommodate the insight that goodness is diffusive by internal necessary emanations, without requiring or involving anything contingent or external. The question of creation is then a question about an external and contingent emanation in addition to the perfect internal emanation, for classical Christians, or the initial generation of Being and the Forms, for Platonists. On the Christian view, creation is the gratuitous and contingent add-on to the perfect and internal divine life. This arguably lessens any coercive pressure to create. I agree with O’Connor that we should not require a contrastive explanation for why God created this world rather than another. Artistic production provides a better analogy here than explanations

¹ Summa Theologiae (ST) Ia 27.1 corpus, see articles 2–3 for the generation of the Son and article 4 for the procession of Love or the Holy Spirit. ² Nicene Creed.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

272   modeled on the physical sciences and contrastive hypotheses. If we ask why Johann Sebastian Bach’s Mass in B Minor exists, we are asking about how Bach made it and served as its source. We need to appeal to Bach and his mastery of the musical art to explain his Mass, but we do not need a further explanation for why Bach didn’t write a different mass. Bach could have written many complete masses, as he wrote many cantatas, each manifesting something of his musical genius in a different way. The explanation for why this piece of music rather than another depends on contingent facts, often external to the production or artistry itself. For example, the presence of a five-part chorus is explained by the style and structure of masses popular at the Dresden court while the orchestral talents Bach had to work with explain which instruments get the most prominent solos.³ In these artistic cases, the causal explanation of the resulting product appeals to both the art itself and the limitations of the artist and the matter. Together, this determines a particular manifestation. But this is different from the divine case where there are no external factors and where God creates both the forms of created things and their matter and so has no such limitations.⁴ In the divine case, the creative power is unlimited and there are no external factors, so we have a supremely sufficient source with no contrastive reasons that are relevant for determining one creation over another. I agree with O’Connor that such an approach fits with the PSR where it can be necessarily true (and also explained) that there are some contrastive truths without explanation (where the truth of that statement is itself explained). This does not mean that there are no constraints on God’s creative action. God’s free creative activity (when it occurs) will plausibly satisfy the Best Ordering Principle, as I have called it, that Thomas Aquinas articulates.⁵ Any world created by God will be ordered in the best way possible, given that it is being directed by God’s infinite wisdom and justice. Nevertheless, there are no contrastive reasons for why these

³ Markey Rathey (2003). “Johann Sebastian Bach’s Mass in B Minor: The Greatest Artwork of All Times and All People,” Tangeman Lecture, New Haven: Yale. ⁴ Cf. ST Ia 44. ⁵ Caleb Murray Cohoe (2014). “God, Causality, and Petitionary Prayer.” Faith and Philosophy 31 (1), 29.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

         ? 273 creatures are created rather than those, because there is no best creature or set of creatures. There is a problem lurking for O’Connor, however, resting not in the initial act of creation, but on what happens to creatures once created. O’Connor’s piece, like Johnston’s, focuses on the procession of creation from the first principle, but there is a corresponding puzzle about the return of creatures to the first principle. Does a world in which there is only a limited return of creation to God manifest God’s own goodness similarly to a world in which some or even all creatures fully unite with God and even share in the infinite divine nature, giving them infinite value? O’Connor rightly resists the idea that God would be required to multiple copies of valuable things as much as possible, so that we would end up with a world containing an infinite number of things. Yet at the same time O’Connor accepts constraints on the kinds of creatures that must feature in whatever God creates, insisting that “a creation lacking creatures who bear [God’s] image . . . and who can receive his generosity seems deficient in the way it manifests his glory.” This suggests that God needs to create a world that features creatures who are able to and do, in fact, achieve “union with God” in a maximal way. A universe with millions of mollusks but no other living things may not be a good enough manifestation. O’Connor lays out several significantly different modes in which God could create worlds. Even allowing that there are an infinite number of ways in which God could realize certain sorts of created orders, there are still theological and philosophical worries about coercion.⁶ O’Connor is willing to accept the “necessity of divine creative activity,” but his claims have stronger implications. O’Connor accepts “a qualified meliorism of desiring to make things better for those sentient creatures there are” and also is confident, as we saw, that God must create a world with sentient creatures capable of meaningful union with the divine. But for such creatures perfect and infinite union with God is better than lacking such union, so if God’s goal is to make things better for those creatures then it seems that God must do everything possible to allow

⁶ O’Connor mentions “a single, organically unified totality,” “infinitely many, wholly discrete universes,” and “an ever improving created order that has merely potential infinity of value which it approaches as a limit.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

274   these creatures to achieve full communion with God and infinite happiness. If this is the case, O’Connor’s meliorism is much stronger than it initially seems. It is not just, as O’Connor puts it, that “God inevitably creates something or other,” but that God must create a world with creatures who can achieve union with him and do everything possible to ensure that they all achieve full communion. This not only suggests that O’Connor is underplaying the constraints he thinks are present, it calls into question whether O’Connor’s position can respect the non-coercion condition. Non-coercion initially required that the reasons for God’s actions “cannot outweigh reasons for acting in any other way, or for not acting at all.” O’Connor explicitly gives up on this last clause but he is also in danger of modifying the first clause in ways that undermine it. O’Connor’s position does avoid complete modal collapse. There may be an infinite number of ways of getting to the destination, so any of the steps before the final state can be seen as contingent. Nevertheless, his view is in danger of violating divine coercion theologically and substantively. If God must create creatures who perfectly unite with him in an infinitely good universe (or series of universes) in order to properly manifest his goodness, then God’s salvific acts might seem forced, rather than gratuitous or freely given. If the ultimate state must feature one sort of thing (the infinitely valuable union of image bearers with God) for whichever image bearers God creates, then the scope of contingency for divine action might have narrowed too far. Having “infinitely many distinct ways to realize the stratagem’s central feature” does not ensure contingency in a significant way. If I need to tip my server, having infinitely many methods of payment for doing so does not add significant contingency or choice to the matter. We can see the issue by contrasting O’Connor’s view with the constraints put forward by Thomas Aquinas. Aquinas agrees that God’s own goodness and justice guide how God creates, but the requirements Aquinas posits are relative to the natural powers of the creature and are thus much weaker. For any creature God creates, God will give to that creature “what is owed to it according to the account of its nature and status [secundum rationem suae naturae et conditionis].”⁷ For Aquinas, ⁷ ST Ia 21.1 ad 3; note that this is an internal requirement of God’s justice that involves giving “what is owed to Himself,” not a constraint coming from creatures externally.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

         ? 275 this means that God creates creatures with the tools they need for their activities and places them in an environment which enables them to exercise their causal powers. If God creates a mollusk, God’s justice requires that the mollusk be placed in a world that allows it to do the work proper to its nature. This requires, in general, that the creature can causally interact with other things and, in the specific case of the mollusk, that it be able to nourish itself, grow, and reproduce in its own characteristic ways. In acting and being acted upon in ways that express their nature, each creature manifests God’s goodness in its own limited way. This expressive requirement is much weaker than the position to which O’Connor is moving. God could create a universe full of creatures that satisfy this condition without featuring beings that can achieve perfect divine union (though Aquinas recognizes other reasons for creation that may be more constraining, as we will see soon). O’Connor’s views, by contrast, seem to imply that God must create creatures who can fully enjoy him and then give them everything required to make that enjoyment as perfect and infinite as it can be. Some, such as Nevin Climenhega, wholeheartedly embrace this approach: God will continue creating creatures who can know and love God and will ensure that each and every one achieves moral perfection and “full communion with God,” guaranteeing that the world will have an infinite number of creatures with entirely fulfilled lives and lives as perfect as they can be.⁸ This view has antecedents in the insistence of Origen and Gregory of Nyssa that the goodness of God will ultimately bring all rational creatures to their best possible end. As Gregory puts it, “[God’s] end is one: when the complete whole of our nature shall have been perfected from the first human to the last . . . to hand over to all the sharing in the goods in Him . . . this is nothing other than to come to be in God himself, for the Good which is above hearing and eye and heart must be that Good which transcends the universe.”⁹ But obviously affirming the universal restoration of all things and their perfect union with God is much stronger than conceding, as O’Connor does, that “God inevitably creates something or other.” The issue here is not primarily ⁸ Nevin Climenhaga (2018). “Infinite Value and the Best of All Possible Worlds.” Philosophy and Phenomenological Research 97 (2), 372. ⁹ Trans. Moore and Wilson with modifications; Patrologia Graeca 46, Peri Psuchēs Kai Anastaseōs Ho Logos (Migne, 152); cf. Origen, Peri Archōn (De Principiis), III 6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

276   universalism: you could take a position on the nature of free will and creaturely freedom that would prevent universalism from being necessarily true. The issue is rather how much generosity God is rationally required to show to creatures. Given God’s goodness and power, does God need to give any willing creature infinite happiness? O’Connor’s commitments seem to imply not only that God must create but that God must create worlds with image bearing creatures in it who all (at least to the extent that they become open to this possibility) achieve lives of infinite value and union with God. This is not to say that this issue arises only for O’Connor or Origen. It confronts any view of reality on which created things are in some way capable of either attaining or approaching the infinite. The tension shows up even in Aquinas. He argues that there are strong reasons for the universe not only to contain a wide variety of creatures (Summa Contra Gentiles [SCG] 2.45) but to specifically contain intellectual creatures, since they better show forth the perfection of God’s mind in their intellectual and volitional activities and thus can more fully imitate and return to God than non-intellectual creatures (SCG 2.46). Aquinas rejects the idea that there is some best world God must create by emphasizing creatures’ limited ability to express God’s goodness (ST 1a 21.1, corpus). At the same time, he also recognizes the infinite goodness creatures can participate in through their union with God. Aquinas concedes that “from the fact that . . . created happiness is the enjoyment of God . . . it follows that [it has] a certain infinite dignity that stems from the infinite goodness which is God. And in this respect nothing can be made that is better than [it is], just as there cannot be anything that is better than God.”¹⁰ So Aquinas thinks there are strong reasons for God to create intellectual creatures and Aquinas thinks that such creatures are capable of sharing in God’s infinite value. But this may undermine Aquinas’s insistence that every created order will be a similarly limited manifestation of divine goodness. If there are universes which contain created things than which “nothing can be made that is better,” then it seems that ¹⁰ Trans. Freddoso, Ad quartum dicendum quod humanitas Christi ex hoc quod est unita Deo, et beatitudo creata ex hoc quod est fruitio Dei, et beata virgo ex hoc quod est mater Dei, habent quandam dignitatem infinitam, ex bono infinito quod est Deus. Et ex hac parte non potest aliquid fieri melius eis, sicut non potest aliquid melius esse Deo.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

         ? 277 God should create such universes, where creation itself shares in infinite goodness, over any universe which features only finite and surpassable creatures and activity. Indeed, since Aquinas thinks humans can only achieve perfect bliss if God supernaturally grant us union with the divine, there is an ongoing debate in the Thomist tradition over whether God needs to give us this beatific vision in order to fulfill our human nature.¹¹ While there would be many ways for God to do this, there is once again a risk of divine coercion. So perhaps O’Connor, who already takes Aquinas to be tempted by the idea that God necessarily creates, will also read Aquinas as joining him in thinking that God must create a universe in which created things share in infinite goodness. We can express the problem more fully by returning to the art analogy. There is a difference in kind between all limited manifestations which only express aspects of the art (even if the artist produces them with perfect wisdom and justice) and a manifestation that, through return to and union with the artist, came to express the whole art itself, not some part of it. This would be not just Pygmalion’s statue come to life, but a living statue that, through its ongoing connection to the artist, fully expressed all the shapes of which sculpture is capable. If such a thing were possible, it would be a better manifestation, even if the more limited statue was created and ordered with the same perfect skill. If some particular pieces of art have the capacity to somehow, with the artist’s help, become such manifestations that they express not just their own being but the whole of the artist’s art, then would it not be better for the artist to create and bring to completion such manifestations? Even an infinite number of partial manifestations would not have the same value as such a perfect image. Thus there seems to be pressure to make some or all of these partial manifestations of God into manifestations that universally and perfectly manifest the art itself. Now there are places where this train of reasoning can be resisted. There are questions about whether even an endless relationship of union to a being of infinite goodness is sufficient for giving the limited creature

¹¹ See, inter alia, Henri de Lubac, The Mystery of the Supernatural (New York: Crossroad Publishing, 1998) and Jacob W. Wood, To Stir a Restless Heart: Thomas Aquinas and Henri de Lubac on Nature, Grace, and the Desire for God (Washington, DC: CUA Press, 2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 20/1/2022, SPi

278   an axiologically infinite value that would be of a higher order than other creatures with limited connections to God. Yet to some extent this tension is unavoidable for any thinker who thinks that perfect union with God is best for us and what our natures want, even though we are not capable of achieving it on our own. Dante’s Aristotle, living forever in limbo, enjoying the full fruits of the natural knowledge he achieved, is living a good life, but not an infinitely good one, and there is something deeply sad about that fact (Inferno, Canto IV). The question going forward is whether O’Connor and others working on divine freedom and creation can avoid divine coercion when it comes to the return of all things to God. Can you affirm that union with the Infinitely Good God is the best thing for creation, while still seeing such union as a gracious gift that is not guaranteed and is, in fact, one that God is free to withhold? Or is there inevitable slippage towards the radical position of Origen and Gregory where God must ultimately be “all in all” by ensuring that each and every image of God becomes as divine and fully manifesting as it can be?¹²

¹² Thanks to Thérèse Cory for sharing her insights on how to read Aquinas on these issues, though she should not be held liable for the reading I put forward.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

11 God Does the Best A Neoplatonic Meditation on O’Connor Katherin A. Rogers

Timothy O’Connor (this volume) has written a thoughtful response to Mark Johnston’s “Why did the One not Remain Within Itself ” (2019). Johnston takes it that God must have been free to create other worlds, or not to create at all. This is St Thomas Aquinas’s understanding.¹ Johnston explains that were God to have reasons for creating, and creating this world, that were stronger than His reasons for doing otherwise, divine creation would be “coerced” (Johnston’s word) and hence not free. The puzzle, then, is to explain how creation is rational, if, so far as the divine reasoning goes, not creating is an equally good option. O’Connor argues to the contrary that it is a positive attribute of the divine nature that God be self-diffusive. God could not fail to create. However he agrees that divine freedom requires that God have the option to create other worlds than the world He has made. So the puzzle is, is there any explanation for why God made our world rather than any of the others He might have made? Both Johnston and O’Connor hold that they are assuming classical theism when they begin with the claim that God freely—in a libertarian sense—creates a contingent world. But classical theism does not begin in the thirteenth century. Earlier, more Neoplatonic philosophers, like ¹ For some of us a strong point in favor of this view is that the First Vatican Council (1869–70) delivered a statement which, at least prima facie, looks to say this. The statement holds that God creates freely and not by any necessity. But the statements of councils are not works of metaphysics. The philosopher is within her rights to wonder exactly how the Council limits what are appropriate ways to understand difficult and debated terms like “freedom” and “necessity.” Katherin A. Rogers, God Does the Best: A Neoplatonic Meditation on O’Connor In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Katherin A. Rogers 2022. DOI: 10.1093/oso/9780192862976.003.0011

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

280  .  Augustine and Anselm, are usually considered “classical theists.”² And certainly both would say that God creates freely and that the created world is contingent. Both take divine freedom to mean that God does whatever it is that He wills to do, and His will is not subservient to anything. It is not even correct to say that His will follows upon His wisdom or His nature since He is simple and His will and wisdom just are His nature. But nonetheless (at least arguably) both hold that God inevitably produces the best creation.³ As O’Connor says, it is a divine perfection that God should be self-diffusive but, contrary to O’Connor’s view, that entails that God does the best and our world is it.⁴ The created world is contingent in the sense that it is dependent on God’s creative will; contingent per se. But it could not have failed to exist. Given the perfect goodness of the divine nature our world is necessary per aliud. The earlier, Neoplatonic, thinkers would be puzzled by the thought that God’s freedom must be what St Thomas apparently held it to be, libertarian free will.⁵ They would be puzzled at the insistence that God, although perfectly good and perfectly wise, should be thought to have had the real possibility of doing other than He has done. That is, they would be puzzled by the same issues that Johnston and O’Connor raise. In the present chapter I attempt a brief response to Johnston and O’Connor from the more Neoplatonic perspective which questions their initial assumption. Perhaps it is just a mistake to ascribe libertarian free will to “that than which no greater can be conceived.” The first point is, perhaps, merely one of terminology. Johnston writes, and O’Connor repeats, that if God has reasons that entail that He inevitably creates our world, then these reasons are “coercive” and ² O’Connor cites the Pseudo-Dionysius with approval, and Pseudo-Dionysius’ thought represents an even closer reflection of Plotinus’ world-view than does that of Augustine and Anselm. ³ See chapter 10 of Rogers (2008). ⁴ On the assumption that human agents have libertarian freedom, it is up to us, to some extent, how the world goes. Thus God does the best, taking human free choices into account. ⁵ Though St Thomas seems to insist upon libertarian freedom for God, there is good reason to think that he was a compatibilist on two counts when it comes to human free will: God causes all our choices and we choose what our preceding practical reason judges to be the better option. I argue (Rogers 2008) that St Anselm supports libertarian free will for created agents, since that is the only way we can imitate divine independence and choose on our own, it being the case that everything we have and are comes from God. But God exists in complete independence and does not need open options to choose from Himself.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

   

281

“controlling” and so God is not free. And ordinarily if we said that some agent’s decision or action was “coerced” or “controlled” that would suggest that the decision or the action was not free. But “coercion” or “control” point to something outside the agent himself that is the source of his decision or action.⁶ That God inevitably wills the best creation because He is perfectly good and rational does not fit our ordinary understanding of coercion or control. Perhaps I can make the case vivid by looking at the divine attribute of omniscience. Imagine a disagreement between someone who accepts Perfect Being Theology (Ms PBT) and someone who is not versed in philosophy and who takes most biblical statements in what appears to him to be their immediate, plain sense (Mr IPS). Mr IPS has read in the Bible that we should ask God not to remember our sins and that God has indeed forgotten them. He takes this to mean that there are facts about human behavior that God just does not know because He deliberately chose to forget them. Ms PBT insists that these biblical passages should be interpreted to mean that God, though He knows everything, can choose to forgive those sins. Divine Omniscience is a necessary divine attribute, so God cannot literally forget—in the sense of no longer knowing—our sins. Mr IPS might fasten on that claim that God cannot forget. Does this mean that God is not free to forget, even if He wants to? Is He coerced and controlled by His omniscience, and the facts of our sins, such that He must keep those sins in mind? Ms PBT has to allow that God cannot do other than keep our sins in mind. This is an entailment of divine omniscience. And He cannot want to forget. But it seems rhetorically abusive to say that God is “coerced” or “controlled” by His perfect knowledge, such that He “remembers” our sins by a necessity that undermines divine freedom. And it seems a similar misuse of the terms to hold that God would be “coerced” by His perfect goodness and wisdom, were He to inevitably produce our world, the best world. Johnston, in an effort to find a non-coercive reason for God’s creating our world, argues against a “meliorist” approach that would hold that the

⁶ A created agent might be “coerced” by his motivations since, in both a theist and a naturalist universe, his desires and character come ultimately from outside of himself. God is different.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

282  .  universe is better overall if God creates. God is infinite good and so no addition to the universe could make it any better. O’Connor does not dispute this argument. One might suppose that this point would undermine the Neoplatonic thesis that God inevitably creates the best world. If creation can’t make the universe better, then there seems to be no decisive motivation for God to create. But from the perspective of the Neoplatonic tradition, this argument misses the mark. If it were being used to try to show that God might fail to create, or fail to create the best creation, then it would simply beg the question. The Neoplatonic claim is that a transcendent being who might choose to remain within itself and not create is not “that than which nothing greater can be conceived.” It is not infinitely good in that it lacks the important attribute of being necessarily self-diffusive. Moreover, the Highest Good must inevitably produce the best creation. The Neoplatonic thesis is that supposing that God must choose between equally good options, or must be free to choose less than the best, is to posit an imperfect God.⁷ But suppose, as St Thomas, and Johnston, and O’Connor do, that there is no best creation. Then God cannot create the best world, but that is not a limitation, since even God cannot do the impossible. There are at least two responses to make here. First, the claim that there is no best creation is open to doubt. The earlier, Neoplatonic approach accepted the Principle of Plenitude—God would not leave any good kind of thing unactualized. And if one argues that it is intrinsically impossible to fit all the good kinds of things into our universe, one might appeal to a multiverse in which all (perhaps an infinity) of the good kinds of things can be actualized.⁸ (O’Connor mentions this possibility (pp. 000‒0) but does not embrace it.) O’Connor writes (p. 000) that there is no generic scale of value, so there may be goods that are not compossible, that are incomparable, and that are incommensurate. But the core of the Neoplatonist doctrine is that there is one standard for all values, and

⁷ O’Connor (this volume, p. 000) notes Alexander Pruss’s argument that a universe in which God chooses not to create could realize certain goods such as, “uniformly maximal excellence, and all beings perfectly achieving their natural telos.” But if this argument is aimed against the Neoplatonist claim that the Highest Good inevitably produces a best creation, then it, too, begs the question. ⁸ See Rogers (2020b: 23–39).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

   

283

that is God. However one compares, or fails to compare, the values of created things to one another, all creatures reflect divine goodness, and— at least in a multiverse where God does not need to choose what creatures can all fit together in a single causal system—the proposal of a best reflection seems coherent. A second response is perhaps even more fundamental. As O’Connor spells out the apparent impossibility of a best universe—and many people understand the argument the same way—the claim seems to be that if there were a best possible world God could make, He would do so, but since there is not, He can’t. The advantage of this view is taken to be that God, even if He is necessarily self-diffusive, has libertarian freedom regarding which world He chooses to create. But the disadvantage, from the Neoplatonic perspective, is that God’s creative activity would be limited by truths about what is possible in logical space—truths apparently independent of God’s will. This would constitute a limitation on God’s will. But don’t we have to say that God’s will is circumscribed by what is possible? St Anselm argues that we do not. What is possible and what is necessary follow from God’s will.⁹ Does God, in His nature, transcend the laws of logic, such that we cannot think or talk meaningfully about Him in Himself? No. God is simple, so His nature and His will and the act by which He wills the possible and the necessary are identical to one another. But if God wills the laws of logic, could they have been otherwise? That seems incoherent. On the Neoplatonic view I am supporting God inevitably, eternally, and immutably wills the best. Trusting that it is best that logic be what it is, we do not need to locate logical space outside of the will of God in order to ensure its stability. Arguing that God cannot create the best world, due to necessary truths independent of God, defends divine libertarian freedom at the expense of undermining divine supremacy. Johnston’s concern is to show that God might create our world for reasons, even though His reasons do not outweigh His reasons for not creating or for not creating our world. As noted, he holds that such outweighing reasons would be “coercive.” O’Connor, though allowing ⁹ Cur Deus Homo Book 2, chapter 17.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

284  .  that God is necessarily self-diffusive, tries to answer a somewhat similar problem with the view that God was libertarianly free to create a different world from ours. Prima facie we would like to suppose that there is an explanation for why God made our world rather than some other. The Neoplatonist holds that the explanation lies in the necessary selfdiffusive goodness of God that inevitably does the best. But if God chooses with libertarian freedom, then there can be no such contrastive explanation in terms of God’s nature or reasons. O’Connor proposes (pp. 000‒0) that we might invoke a hierarchically ordered Principle of Sufficient Reason where “ . . . it could be a necessary, grounded truth that there are (contrastive) truths without explanation.” This epistemically acceptable PSR “ . . . can allow that there are some second-order truths lacking an explanation, provided that there is an explanation of this lack of explanation, a third-order truth.” And it is the assumption of libertarian freedom on God’s part that explains the lack of explanation. “Notice that our conception of a necessarily-existing, purposive, and free source of contingent reality provides an explanation of why there can be no explanation of certain contingent contrastive truths.” If this proposal is taken to be a response to the critic who is asking for a contrastive explanation for divine creation, then the argument seems to be that if we assume there cannot be a contrastive explanation for divine creation, then we should be satisfied with the lack of a contrastive explanation for divine creation. If the critic finds this unsatisfactory then she has another reason to consider the Neoplatonic approach, which does offer a contrastive explanation. Why did God create our world? Because He is perfectly good and wise, and our world is the best. A further worry with assuming libertarian freedom on the part of God is that it is hard to square with another basic assumption held by all of the classical theists of late antiquity and the Middle Ages and that is that God is simple and (an entailment of simplicity) immutable. Neither Johnston nor O’Connor discusses the doctrine of divine simplicity, and their spelling out of their views suggests that they may reject it. For example, Johnston does not qualify the thought that God might have reasons, in the plural, for doing what He does. And O’Connor speaks of God having various motivational states and “potentialities” for various outcomes. The defender of divine simplicity might use these locutions as

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

   

285

well, but neither Johnston nor O’Connor explain that when they seem to ascribe multiplicity to God they are speaking quoad nos, and neither mentions the issue of simplicity. The problem is that if one accepts the classical doctrine of divine simplicity, then one takes all of the divine attributes, including all of God’s knowing and doing, to be absolutely one and to be identical with God’s nature. A difficult doctrine on any account. Now add that God has libertarian freedom. God might not have created, or creation might be other than it is. Prima facie this looks to entail that God’s act would have been different had He failed to create or had He created a different world. But then there seems to be at least the multiplicity of a fundamental, immutable divine nature and then the act of divine creation which might have been other than it is. (On at least some readings) St Thomas “solves” the problem by holding that, contrary to the prima facie appearance, God, including the act which is identical to Himself, is exactly the same whether He creates our world or a different world.¹⁰ Indeed, He is exactly the same, even had He not created anything at all. To speak correctly of a relationship between God and creation is really to speak about creation, and not about God at all. For the Neoplatonist this problem of how to square divine libertarian freedom and divine simplicity does not arise.¹¹ It is not the case that God might be doing other than He is doing. (The problem of divine immutability can be resolved by embracing the thought that divine eternity “contains” everything at all times.) Johnston and O’Connor may well reject the doctrine of divine simplicity, but if so, it should be noted that their “classical theism” is very different from that of St Augustine, St Anselm, and St Thomas. Why, then, insist that God have libertarian free will? One reason might be that one takes “free will” to be defined as libertarians define it, to include genuine alternative possibilities. But why define free will this way? St Anselm argues that created agents must have libertarian free will—they must confront genuinely open options. This is because everything about the human being comes from God. Were the agent simply to ¹⁰ Matthews Grant (2012; 2019) has tried to explain and defend this difficult Thomist position. ¹¹ I try to spell out the (or “a”) Neoplatonic approach to divine simplicity in Simplicity (Rogers 2020a).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

286  .  pursue his strongest, god-given desire, his “choice” would be necessitated by God. Only by choosing between morally significant alternatives can the human agent contribute anything at all to his own orientation towards, or away from, the good. Anselm recognizes that no contrastive explanation will be available for why the created agent chose one way over the other. In a morally significant choice, there is nothing about the created agent’s preceding reasons or character that fully explains his choice for this over that. But, in the case of created agents, asking for such an explanation gets the dynamic of free choice backwards. Choices are not determined by character, character is determined by choices.¹² But the reasons for attributing libertarian freedom to created agents do not apply to God. Freedom had best be defined as “the ability to keep justice for its own sake.” That way we can say that both God and created agents are free. But we need not assume that that entails that the processes of divine and human willing work the same way. Thus Anselm preserves the value of human libertarian freedom without incurring the pitfalls of supposing that God must be libertarianly free. But doesn’t this Neoplatonic approach entail a “modal collapse”? Let us allow the claim that there are no unactualized possibilities. Is that a dire consequence? Note first that this approach does not shut down human libertarian free will. It can be entirely up to the created agent which option is chosen. One way to understand the thesis that God is eternal (another plank of the classical theist platform) is to hold that all times are equally existent. This means there is a truth to propositions about free choices in the future. (The “future” relative to some temporal agent at some particular time.) But it is the actual future choice of the agent himself that grounds the truth of the proposition, just as it is the agent’s free choice that grounds the truth of propositions about free choices in the past and present. God does the best given the free choices of created agents. This does mean that all of the things and events in all of the dimensions of space-time (perhaps in all of the universes of the multiverse) just are as they are. But this does not foreclose human libertarian free will.

¹² I argue this point in chapter 8 of Rogers (2015).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

   

287

And O’Connor himself, in discussing the lack of contrastive explanations regarding divine creation, suggests a way of thinking about the “modal collapse”—or perhaps “modal simplicity” would be a less worrisome term—such that it might inspire less fear. He writes that, “ . . . we should observe that contrastive truths correspond to nothing real—there are no ‘contrastive facts’ in the sense of concrete structures or constituents in reality” (p. 000). So with all unactualized possibilia. In positing modal simplicity we lose nothing of what there is in the world. We may have to allow that what appear to be propositions about unactualized possibilia have a truth value only if they can be translated into propositions about what is actually the case, or that they state only what is epistemically possible quoad nos, or that they state what is “possible” on some picture in which we have bracketed the existence of (the Neoplatonic) God. But none of these exercises is terribly onerous. A final criticism of the Neoplatonic approach would point out that, looking at our actual world, the thesis that it is the best world God could actualize is wildly implausible. Our world is full of wickedness and pain and suffering. Surely God can do better! But this argument is not one to be employed with much force by St Thomas, or Johnston, or O’Connor. St Thomas, though insisting that God could have done otherwise, including that He could have chosen not to create a world at all, holds that if God does create, then whatever world He makes must be well ordered and good overall. Johnston concludes his search for a “noncoercive” motive for divine creation with the traditional view that the reason God creates is to manifest His glory. And O’Connor, holding that God is necessarily self-diffusive, proposes various limits on what worlds God could choose to make, in accordance with His goodness. These defenders of libertarian divine freedom do not allow that God might freely choose to make a bad world, so they, as much as the Neoplatonist, face the problem of evil. The difference between God’s making the best world and God’s making one of a number of possible, exceedingly good worlds is minimal when the problem is to reconcile divine power and goodness with what we observe in our actual world. And down the ages classical theists of all stripes have proposed solutions to the problem of evil, so this worry does not count conclusively against the thesis that God does the best. The puzzles raised by Johnston and O’Connor, on the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

288  .  assumption that God has libertarian free will, do not arise on the Neoplatonic view that God, as perfect Good, inevitably does the best. The Neoplatonic approach, then, is worth considering.

References Grant, W. Matthews (2012). “Divine Simplicity, Contingent Truths, and Extrinsic Models of Divine Knowing,” Faith and Philosophy 29: 254–74. Grant, W. Matthews (2019). Free Will and God’s Universal Causality. London: Bloomsbury Academic. Johnston, Mark (2019). “Why Did the One Not Remain Within Itself,” Oxford Studies in Philosophy of Religion 9. Oxford: Oxford University Press, 106–64. Rogers, Katherin (2008). Anselm on Freedom. Oxford: Oxford University Press. Rogers, Katherin (2015). Freedom and Self-Creation: Libertarianism. Oxford: Oxford University Press.

Anselmian

Rogers, Katherin (2020a). “An Anselmian Approach to Divine Simplicity,” Faith and Philosophy 37: 308–22. Rogers, Katherin (2020b). “Classical Theism and the Multiverse,” International Journal for Philosophy of Religion 88: 23–39.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

12 The Contingency of Creation and Divine Choice Fatema Amijee

Is the contingency of creation compatible with the rationalist demand that all facts, including those about God’s actions, be governed by the Principle of Sufficient Reason (henceforth “PSR”)? According to the PSR, every fact has an explanation—a sufficient reason—for why it obtains and not otherwise.¹ Thus, given that God created this world, there must be an explanation for why God brought this world into existence rather than not. But if facts about God’s actions have sufficient reasons, it is difficult to see how God’s actions leave any room for contingency: if God’s choice to create this world as opposed to any other is backed by a sufficient reason that necessitates God’s choice, then it seems—at least on the face of it—that God could not have done otherwise.² I will argue, however, that despite appearances, it is possible to both preserve the contingency of divine choice and endorse the PSR in an unrestricted form so that it also applies to all facts about God’s actions and his preferences. In §1, I discuss in more detail how the tension between the PSR and the contingency of creation arises. In §2, I argue

¹ This contrastive formulation of the PSR traces back to Leibniz. In §31–2 of The Monadology, Leibniz writes: “Our reasonings are based on two great principles, that of contradiction . . . And that of sufficient reason, by virtue of which we consider that we can find no true or existent fact, no true assertion, without there being a sufficient reason why it is thus and not otherwise, although most of the time these reasons cannot be known to us.” (Leibniz, 1989, p. 217) ² In his chapter “Why the One Did Not Remain Within Itself ”, Timothy O’Conner writes: “[w]hile many theists are concerned to maximize the contingency of creation, some others will equate any contingency in the divine choice with the breakdown of rational explanation” (this volume, p. 000). Fatema Amijee, The Contingency of Creation and Divine Choice In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Fatema Amijee 2022. DOI: 10.1093/oso/9780192862976.003.0012

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

290   that, contra Timothy O’Connor’s suggestion that the rationalist should adopt a restricted PSR that allows for brute contrastive facts in order to secure the contingency of divine choice, there is a sufficient reason available for God’s choice of one world over another even if God could have chosen otherwise. But my proposal in §2 seems to generate a brute fact further downstream. I resolve this worry in §3, and thereby complete my case for the compatibility of an unrestricted PSR and the contingency of creation.

1. The Problem Why is the contingency of creation apparently incompatible with the PSR? Suppose that God’s actions are governed by the PSR and that God can bring only one world into existence. Then there has to be an explanation for why God creates our world as opposed to any other world (or no world). A “world,” as I am using the term, is not a maximally complete way that things simpliciter could have been, but rather a maximally complete way that created things could have been. Thus, the term “world,” as I am using it, picks out a created world. This sense of the term is specific to a theistic or deistic framework on which God, as creator, is not part of his creation, but rather stands outside of it. As Leibniz puts it, God “is above the world, and, so to speak, extramundane.”³ By contrast, I will use the term “possibility” to pick out a maximally complete way that things (created or otherwise) could have been. On a theistic or deistic framework committed to the existence of God as a necessary being and creator, God does not occupy a vantage point that is outside of any possibility. Rather, God exists in every possibility. A possibility thus includes both God and his creation (if he creates anything in that possibility). Now let us suppose for the sake of illustration that God is deciding between two worlds, w₁ and w₂ in a given possibility. Then, by the PSR, God must have a sufficient reason for bringing one of the worlds into existence over the other. This does not mean merely that God has a ³ Leibniz, 1989, p. 149.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

      

291

reason to bring w₁ into existence, for God might likewise have a reason to bring w₂ into existence. Suppose for example, that w₁ is more symmetric than w₂ and God values symmetry, but w₂ contains more happy people than w₁, and God also values happiness. Then it would seem that God has a reason to bring w₁ into existence, but also a reason to bring w₂ into existence. But if facts about God’s actions are governed by the PSR, then God’s choices are backed by the weight of reasons: they are the choices that God has most reason for making. God’s choice—or so the argument goes—cannot be arbitrary in the sense that God chooses to actualize w₁, but he might well have chosen to actualize w₂. But if the PSR requires that God’s choice not be arbitrary, then God’s choice is seemingly necessitated: if the weight of reasons supports God’s bringing about w₁, then God could not possibly bring about w₂. Why is that? Given the truth of the (unrestricted) PSR, two assumptions generate the threat to contingency. First, that, necessarily, God chooses the world he has most reason to bring into existence. To assume otherwise would be to suppose that God is irrational, because he could choose to bring about w₂ even if he had most reason to bring about w₁. Second, that, necessarily, w₁ (i.e., our world) is the world that God has most reason to bring about. If it were a contingent matter that w₁ is the world that God has most reason to bring about, then the necessitarian conclusion would not follow, for even if God necessarily chooses whatever world he has most reason to bring about, that world could have been w₂, not w₁, even if it is in fact w₁. And so, it would not follow that w₁’s existence is necessary. The first assumption follows from the claim that God is necessarily rational. To suppose otherwise would be to admit a defect in God, which goes against the conception of God as a being who is perfect in all respects. If the first assumption cannot be given up, the only way to save the contingency of creation is to deny the second assumption. Within a framework on which God must choose whatever world God has most reason to bring about, denying the second assumption requires insisting that a world other than ours could have been a world that God has most reason to bring about. And at least on the face of it, this claim does not look tenable. Suppose that God’s reason for bringing about a particular world is that it is the best of all possible worlds. The standards for what count as “best” cannot vary between possibilities (otherwise, it

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

292   would be true for us that God is possibly evil). And the set of worlds from which God must choose remains invariant across possibilities. Thus, it would seem that if God’s choices are governed by the PSR, the existence of our world is not contingent, but necessary.⁴ Interestingly, in order to derive the tension between the claim that God’s actions are governed by the PSR and the claim that our world exists contingently, we need not assume that God is omniscient, omnipotent, and omnibenevolent. Suppose, for example, that God lacks access to some of the facts about the worlds that he might bring into existence. Nevertheless, if God’s actions are governed by the PSR, then God must have a sufficient reason for choosing one world over another based on the facts that he has access to. Again, suppose that there was a world that God could not possibly actualize because it was beyond his power to do so, and he thus lacked omnipotence. Still, God can nevertheless have a sufficient reason for actualizing the world he brings about. It might not be the world he has most reason to actualize, but of the options that are within God’s power, it can nevertheless be the one with the weight of reason behind it. Finally, suppose that God is not omnibenevolent. The PSR does not place any kind of constraints on the kinds of reasons God might have for choosing to actualize one world over another. It might not be in God’s nature to create the world that is morally speaking the best world. The PSR requires only that whatever the reasons might be for God’s choice, they constitute a sufficient reason, such that if God’s reasons support actualizing w₁, he could not then actualize w₂ while remaining rational.

2. Non-moral Divine Preferences I will argue that despite appearances, the contingency of creation is not inconsistent with the claim that facts about God’s actions are governed ⁴ According to Leibniz’s correspondent Samuel Clarke, God’s reason for choosing a particular world over others can consist merely in God’s will (cf. Clarke 2000, p. 11). If God creates w₁ over w₂ simply because that is what his will dictates, then God could just as easily have chosen w₂. However, Clarke’s version of the PSR is not unrestricted, for it entails that there is no sufficient reason for why God wills what he does.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

      

293

by the PSR. Thus, one can accept both the contingency of creation, and endorse an unrestricted PSR that permits no brute facts, even if the facts in question are about God’s choices, preferences, and reasons for acting. For the sake of argument, let us suppose that God is omnibenevolent, and thus seeks to create the best of all possible worlds, where “best” picks out what is morally best.⁵ Suppose also that worlds w₁ and w₂ are equivalent in terms of bestness or, equivalently, moral optimality. Would not God’s choice of one world over the other then violate the PSR, for God would have no reason to choose w₁ over w₂, or vice versa? I will show that such a choice need not violate the PSR. Let us suppose that the moral optimality of a world is not one of its primitive properties, but is rather instantiated because the world instantiates some other properties. Perhaps it is a just and egalitarian world that is free from poverty. However, it need not be the case that every property the world instantiates plays a role in grounding the fact that the world is morally superior. For example, properties that pertain to shape—i.e., whether the world is cubical or spherical—might play no role in grounding the world’s moral optimality.⁶ Let us call such properties “extraneous.”⁷ On the assumption that God is omnibenevolent, God has a reason to choose the world that is morally the best. But unless we suppose that God has no other preferences except to bring about what is morally the best, it does not follow that God’s choice of world w₁ over world w₂ has no sufficient reason. And it is very implausible to think that God’s nature precludes his having any preferences apart from the preference for what is good. Finite creatures like us are not limited in this way: my friend Sam is perfectly good, and capable of no evil, yet that does not mean he thereby has no non-moral preferences. He still, for example, prefers chocolate over vanilla ice-cream. If finite creatures like us can have both moral and non-moral preferences, there’s no reason to think that ⁵ My argument does not require this assumption. Presuming omnibenevolence merely permits us to focus on a particular kind of reason—namely, a world’s moral optimality—as a reason that motivates God’s choice about which world to create. ⁶ I’m assuming for convenience that worlds can be cubical or spherical. The argument goes through for any pair of incompatible properties which are morally irrelevant. ⁷ One might claim—as Leibniz arguably does—that no properties of a world can be extraneous to the world’s overall goodness. But this view about the properties instantiated by worlds is not forced upon us.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

294   God—a being who is infinitely greater and more powerful than us— cannot. Moreover, to suppose that God can only have moral preferences would be to presuppose that God is limited in his abilities: that he is not able to have non-moral preferences. Yet, one might object—correctly—that not just any limitation in God’s abilities qualifies as a weakness. For example, God does not have the ability to act immorally, but that God lacks this ability does not imply that God is deficient in his power. However, God’s inability to have nonmoral preferences does, I contend, suggest a deficiency in God’s power. The difference between God’s inability to be immoral and his putative inability to have non-moral preferences lies in this: God lacks the former ability because he is perfectly moral, but there is no obvious explanation that stems from God’s nature for why God should lack the ability to have non-moral preferences. In other words, God’s lacking an ability does not imply that God is deficient, as long as that lack is explained by a feature of God that belongs to God’s nature. By contrast, God’s lacking an ability does imply that God is deficient if the lack is not explained by any feature that belongs to God’s nature. Thus, if God lacked the ability to have nonmoral preferences, God would be limited in his abilities in an objectionable way. Now suppose that w₁ and w₂ have some extraneous properties, i.e., properties that do not play a role in grounding the moral bestness of those worlds. For concreteness, let us suppose that w₁ has the extraneous property of being cubical, whereas w₂ has the extraneous property of being spherical. Now because God may have both moral and non-moral preferences, God can choose to actualize w₁ because there is no world that is better than w₁, morally speaking, and because it is cubical. Thus, even in a situation where there is more than one morally optimal world, God’s choosing to actualize one morally optimal world over another can be backed by a sufficient reason. God’s preference for a world that is morally optimal but also cubical provides the sufficient reason for God’s choosing to actualize that world. Moreover, it also serves as a sufficient reason for the contrastive fact that God chooses to actualize w₁ rather than w₂. Still, one might worry that by allowing a non-moral reason—such as the preference for a cubical world—to be part of the sufficient reason for

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

      

295

choosing to create w₁, we risk either losing the contingency of creation or violating the PSR. In what follows, I discuss each risk in turn. Why would the claim that God has a sufficient reason—one that has both moral and non-moral components—for choosing to create w₁ over w₂ threaten the contingency of world w₁? If the PSR is true, then all facts—including facts about God’s preferences—have an explanation. God’s preference for a world that is morally optimal is easily explained: it is part of God’s nature to be omnibenevolent, and this explains why the moral optimality of a world can figure as a reason for God’s decision to actualize w₁. However, what could possibly explain God’s preference for a cubical world? At least on the face of it, it seems that we have two options. We can either say that it is part of God’s nature that he prefers a cubical world, or we can say that there’s no further explanation for why God prefers a cubical world—he just does. The second option clearly violates the PSR. By the PSR, there must be an explanation for why God has the preferences he does. One might adopt a restricted PSR that does not apply to God’s preferences, even if it applies to his choices and actions. But from a purely rationalist perspective, such a restriction would be ad hoc. Let’s then explore the first option. According to this option, it is part of God’s nature that he prefers a cubical world, just as it is part of God’s nature that he is omnibenevolent and thus prefers a morally optimal world. But if God’s preference for a cubical world were part of his nature, then it is difficult to see how God could have chosen to actualize a morally optimal world other than the one that is cubical. In particular, he could not have actualized w₂, which is also morally optimal, but spherical (at least not on the assumption that God cannot have both a preference for a spherical world and an equally strong preference for a cubical world at the same time).⁸ Thus, it seems that if God’s preference for a cubical world were part of his nature, then God could not have done otherwise.

⁸ Of course, if God has both a preference for a cubical world and an equally strong preference for a spherical world, then the fact that a world is morally optimal and cubical cannot explain God’s decision to actualize that world over one that is also morally optimal but spherical.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

296  

3. Contingency Regained Is there a way to preserve the contingency of creation without introducing a brute fact about God’s preferences? I will argue that there is: an unexplored third option remains. On this option, God’s preference for a cubical world (say) is neither completely explained by his nature nor unexplained. Instead, it is part of God’s nature to have a disjunctive set of preferences: a preference for cubicity, or a preference for sphericality, and so on. Moreover, God’s actual preference need not be a necessary preference, where a necessary preference is a preference that God has in all possibilities. The appearance of a brute fact about God’s preferences results from an overly simplistic conception of how God’s preferences could relate to his nature. Once we have a more complex picture of God’s preferences, we can resolve the apparent tension between the PSR and God’s ability to contingently choose to create a particular world. Consider a pair of incompatible preferences: the preference for cubicity over sphericality, and for sphericality over cubicity. No coherent being could have both of these preferences at once. If any creature has coherent preferences, it is surely God, so God cannot have both of these preferences. It is possible, of course, that God is agnostic about whether he prefers cubicity over sphericality or vice versa. But while God may be agnostic with respect to some non-moral preferences, he cannot be agnostic with respect to every set of incompatible non-moral preferences: that would be an unacceptable limitation on his capacities. So, let us assume God does have a preference between cubicity and sphericality. Must this preference be necessary? Surely not. Nothing about God’s nature suggests that his non-moral preference for cubicity (or sphericality) must be necessary. In fact, it seems that with respect to a nonmoral choice like that between cubicity and sphericality, God should have the ability to have contingent preferences: in some possibilities he prefers cubicity, in other possibilities he prefers sphericality. Generalizing, it seems plausible that for any set of pairwise incompatible preferences, God should be able to have each of the preferences in the set. Hence, for each of these preferences, there is a possibility in which God has that preference.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

      

297

What then explains why God has a certain preference in a given possibility? I suggest that the question is ill-formed in this case. We cannot pick out a possibility and then ask: why does God have such-andsuch preference in this possibility? God’s having a particular preference in a possibility is simply essential to that possibility. My suggestion is that it is part of God’s nature that he has a disjunctive set of non-moral preferences, and that for each preference in the disjunction, there is a possibility in which God has that preference. Thus, God does not have the same non-moral preference in each possibility. Moreover, that God has a particular preference in some possibility is not unexplained: it is explained by the fact that God’s nature requires that he have some preference of the relevant kind in each possibility, and that he could have had any one of the preferences. What is ill-formed is the demand to explain why, in a given possibility, God has a certain preference. It is simply essential to a possibility that God has a certain preference in that possibility. But there seems to remain a recalcitrant brute fact. One might ask: why is the possibility in which God has a preference for sphericality over cubicity actual? To ask this question is to just ask why the possibility we are in is actual—a question that is notoriously difficult to answer. Yet proponents of an unrestricted PSR must answer it. Fortunately, a way out is available: one can adopt a modal realism according to which all and only those possibilities obtain, or exist, in which God chooses a morally optimal world. To make this picture vivid, suppose that there are only two possibilities, P1 and P2, in which God chooses a morally optimal world. On the restricted modal realism I am proposing, both possibilities obtain. It is an essential feature of P1 that God prefers cubicity in P1 and an essential feature of P2 that God prefers sphericality in P2. Then in P1 God chooses to create a morally optimal world that is cubical, and in P2, he chooses to create a morally optimal world that is spherical. But there is no possibility in which God creates a morally sub-optimal world. So, unlike a standard Lewisian modal realism (cf. Lewis 1986), the modal realist view I have just sketched does not have the result that an unrestricted plenitude of worlds exists—some of which are evil, or otherwise morally sub-optimal. In this respect, the view I am proposing comes apart from “theistic modal realism” as discussed and defended by

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

298   Michael Almeida.⁹ Like standard Lewisian modal realism, theistic modal realism takes possible worlds—both actual and non-actual—to be existing, concrete entities. On such a view, “actual” is in indexical, and simply picks out the possibility that we inhabit. Theistic modal realism extends this standard modal realism to allow for God’s necessary existence.¹⁰ But while theistic modal realism, like standard modal realism, is unrestricted, and is thus committed to the existence of every possible world, the restricted modal realism I am proposing does not entail the existence of every possible world. It entails only the existence of those possible worlds that are morally optimal. Put differently, on theistic modal realism, for every possible world, there is an existing possibility in which God creates it. By contrast, on my restricted modal realism, only for every possible morally optimal world is there an existing possibility in which God creates it. Thus, on my restricted modal realism, there are fewer existing worlds than on (unrestricted) theistic modal realism. At least one reason to prefer my restricted modal realism to the theistic modal realist’s unrestricted modal realism is that it does not run into the problem of having to reconcile the existence of morally suboptimal or evil worlds with God’s benevolence.¹¹ My restricted modal realism cleanly captures the contingency of creation. A world exists contingently just if it exists in some but not all possibilities. My restricted modal realism preserves the contingency of divine choice by allowing that there are non-actual (though existing) possibilities in which God chooses otherwise (i.e., chooses a morally optimal world that is cubical rather than spherical). God chooses otherwise in these possibilities because God’s non-moral preferences in those possibilities are distinct from God’s non-moral preferences in this possibility, and so the other possibilities include created worlds that differ from our own (despite being equally worthy of creation from a purely moral point of view). For those that complain that this is not genuine

⁹ See Almeida (2008, ch. 8; 2011; 2017). ¹⁰ See Cameron (2009) for a helpful and compelling discussion of how the theistic modal realist can accommodate the necessity of God’s existence. ¹¹ Almeida argues that theistic modal realism is compatible with God’s omnibenevolence, because, on that view, God couldn’t possibly have prevented the existence of possible worlds with evil in them, and even an omnipotent being can’t do that which is impossible.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

      

299

contingency, since a modal realist holds that all the relevant possibilities exist, I can appeal to the standard modal realist reply. For modal realists, “actual” is an indexical, and so while each possibility (and thus created world) counts as actual relative to itself, it is false (relative to each possibility) that every possibility is actual. The way in which my restricted modal realism captures the contingency of creation also highlights how the view substantially differs from the multiverse hypothesis. According to the multiverse hypothesis, as discussed by Kraay (2010), a single possibility includes multiple worlds (in my sense of “world”). These worlds may differ from one another as long as they are logically compossible, i.e., they do not contradict one another. Unlike my restricted modal realism, the multiverse hypothesis does not permit the different worlds to be created in virtue of incompatible preferences that form God’s disjunctive non-moral preference (since God could not actually have incompatible preferences). And the actual creation of distinct worlds, which the multiverse hypothesis allows, is not sufficient to secure the result that God could have done otherwise. God could have done otherwise only if there is some possibility in which he does otherwise. Hence, it is not sufficient to multiply created worlds; these worlds must be distributed amongst distinct possibilities. Finally, the compatibility of the unrestricted PSR and the contingency of divine choice does not require the truth of the restricted modal realist view that I have sketched: it requires its mere coherence. And there is no reason to think that the view is incoherent. What I have provided is a model on which we can have both the contingency of divine choice and the truth of an unrestricted PSR. A skeptic might argue that even if I have shown that the contingency of divine choice is compatible with the unrestricted PSR, it is too costly to endorse them both at once, since any form of modal realism is too bizarre or objectionable (even if coherent). I cannot engage with this skepticism fully here, except to say that insofar as we have independent reasons to endorse both the unrestricted PSR and the contingency of divine choice, we arguably have prima facie reason to adopt the modal realist view I have sketched. Objections to that modal realist view must therefore overcome that prima facie consideration in favor of it.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

300  

4. Concluding Remarks My goal has been to show that the contingency of creation and divine choice is compatible with an unrestricted PSR. I have put aside the questions of why we might want to preserve the contingency of God’s choice, or endorse an unrestricted PSR that governs all facts, including facts about God’s actions and preferences. Those inclined to accept both will see my chapter as a welcome salvation from the threat of inconsistency. And those who may have hoped to use the one against the other— who argue either from the unrestricted PSR to the necessity of God’s choice or from the contingency of God’s choice to a restricted PSR—face a more difficult task than they may have assumed: they must provide a direct argument against the model I have proposed, a model that renders the contingency of God’s choice compatible with an unrestricted PSR.¹²

References Almeida, Michael. (2008) The Metaphysics of Perfect Beings (New York: Routledge). Almeida, Michael. (2011) “Theistic Modal Realism?,” in J. L. Kvanvig (ed.) Oxford Studies in Philosophy of Religion, 3 (Oxford: Oxford University Press), 1–15. Almeida, Michael. (2017) “Theistic Modal Realism II: Theoretical Benefits,” Philosophy Compass 12. Cameron, Ross. (2009) “God Exists at Every (Modal Realist) World: Response to Sheehy,” Religious Studies 45: 95–100. Clarke, Samuel. (2000) “Clarke’s Second Reply,” in Roger Ariew (ed.) G. W. Leibniz and Samuel Clarke: Correspondence. Indianapolis: Hackett. Kraay, Klaas. (2010) “Theism, Possible Worlds, and the Multiverse,” Philosophical Studies 147 (3): 355–68. Leibniz, Gottfried. (1989) Philosophical Essays. Roger Ariew and Daniel Garber (trans.). Indianapolis: Hackett. Lewis, David. (1986) On the Plurality of Worlds. Oxford: Blackwell.

¹² Many thanks to Dominic Alford-Duguid, Michael Della Rocca, and Dean Zimmerman for written comments and detailed discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

13 Ineffability and Creation Alexander R. Pruss

This chapter offers a unified speculative solution to two seemingly different theological problems: What it means to say that God is ineffable and how it is that a being with the omni-properties of the God of theism could fail to create. St John Chrysostom (2010) once preached a series of sermons against a heretical sect who apparently argued against the doctrine of the Trinity on the grounds that if God was a Trinity, then God’s essence was incomprehensible, but on the contrary God’s essence was entirely comprehensible, and hence God couldn’t be a Trinity. Chrysostom argued, among other things, that if God’s ways are incomprehensible, as scripture insists, then a fortiori so is God’s nature. That sect was a rare occurrence in the history of Judaism, Christianity, and Islam. Most monotheists have at least paid lip service to God’s incomprehensibility and its linguistic analogue, God’s ineffability. But doing so is very tricky. It is easy to contradict oneself by saying that nothing can be truly said of God. A simplistic approach to ineffability is this: we can never say everything that is true of God. But that probably doesn’t distinguish God from many other things that we would not say are ineffable. There are infinitely many irreducible truths about the integers, and we have access only to finitely many of them, but that doesn’t make the integers ineffable.¹ Historically, besides heretics (maybe including Hegel, among the philosophers) who think they’ve comprehended God, and people who ¹ A referee suggests a stronger mediating position: God has some attributes that are wholly incomprehensible and others that are partly comprehensible and univocally linguistically expressible. On this view, it still seems, God will only be partially ineffable, whereas it seems plausible that the doctrine of divine ineffability is meant to say that God is fully ineffable. Alexander R. Pruss, Ineffability and Creation In: Oxford Studies in Philosophy of Religion Volume 10. Edited by: Lara Buchak and Dean W. Zimmerman, Oxford University Press. © Alexander R. Pruss 2022. DOI: 10.1093/oso/9780192862976.003.0013

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

302  .  pay mere lip service to the doctrine, there have been two main historical approaches to ineffability: apophatic theology, paradigmatically found in Eastern Christianity and in Moses Maimonides, and analogy, as in strands of Western Christianity influenced by Aquinas. Apophatic theology says that we cannot correctly say anything positive about God, but can only say what God is not. This neatly avoids the charge of self-contradiction as it is a negative, rather than positive, statement. Unfortunately, it is difficult to reconstruct the content of monotheistic devotion in apophatic terms without cheating. The believer wants to thank God for being good. The apophaticist has to reword that as thanks to God for not being non-good. That God is not non-good is true, but it is a cheat: being non-good is clearly a negative (without that observation, the popular privation theory of evil couldn’t get off the ground), and to deny a negative of God is, alas, to affirm a positive. Defenders of analogy, the most famous being St Thomas Aquinas (1920, I.12), add a second way of speaking to the apophaticist’s willingness to make denials. We can use words that apply to God and to humans, but do so in different yet analogous ways. Thus, God is good and Mother Teresa is good. But what makes God good is radically different from what makes Mother Teresa good, since God is goodness itself while Mother Teresa is good by participation in God. There are difficulties here, too. A cheeky question to ask is whether analogously good applies analogously or literally to God: if analogously, then we are off to an endless regress but if literally, then something non-negative is said literally of God. Similarly, we may wonder whether the disjunctive predicate “has creaturely goodness or has non-creaturely goodness” doesn’t apply univocally to God and creatures, and yet that predicate is not negative (though it does contain a “non”). More deeply, one may worry with Duns Scotus whether we can really know what we are saying if we are speaking analogically through and through. The above theories of ineffability, as well as other recent proposals (e.g., Jacobs, 2015), just as their criticisms, all have one thing in common: they are focused on semantics. But if we focus on pragmatics instead, we can find a promising theory of ineffability that does not fall prey to the difficulties. You ask me about a student I had in my logic class and how he did, and I say: “He passed.” And I say no more. But in fact, he did brilliantly,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

303

getting a grade far above 100%, given all the bonuses. What I said was true, but I misled you by saying a truth that falls far short of reality. I said something from which, given natural background assumptions, you would draw a false conclusion. Or suppose I say “Socrates was Plato’s fellow countryman” to someone who thinks Plato was a Roman. I have misled him, perhaps unwittingly: he will draw the false conclusion that Socrates was a Roman. In these examples, I can avoid misleading by being more explicit. I can say about my student: “In fact, it was the highest pass I have ever seen.” Or about Plato and Socrates I can add: “ . . . being both Greeks.” Here then is a theory of ineffability: everything sayable about God is misleading. Of course, everything false we say about God is automatically misleading. But according to the theory, the true things are all misleading, too. Think of how a simple believer may hear all sorts of orthodox sermons and still come to think that God is a bearded man in the sky. But unlike in my earlier ordinary examples of misleading language, I cannot add anything to my misleading statement to get rid of all the misleadingness. In fact, if my theory is correct, my adding a statement to what was previously misleading will mislead in a new way—though it might remove one misleading aspect of the original statement. Let’s say I tell the simple believer that God is not a bearded man in the sky. But perhaps now the believer concludes that God is not a person. Here is a deeper example from Christian theology. It is true that God is perfectly one and simple. But this statement, while true, is apt to mislead one into thinking that God isn’t a Trinity of persons. And, conversely, saying that God is a Trinity of persons is apt to mislead one into thinking that God isn’t one. And saying both is apt to mislead into thinking that God is self-contradictory. We might then add that there is no contradiction in God. Now we are apt to mislead the listener into theories like the shamrock theory of the Trinity that too easily eliminate the contradiction. So we add that the compatibility of the claims is incomprehensible to us.² But the misleadingness continues: perhaps now the listener is led to think that God is not really three persons, but is three persons in some Pickwickian way of speaking. That implication can be clarified, but the problem remains. Or so the misleadingness theory holds. ² I am grateful to the referee for the no-contradiction and incomprehensibility qualifications.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

304  .  Furthermore, it is not unlikely that at some point in such a series of clarifications, we will have shifted the misleading bump under the rug to a point where we will no longer be able to point it out and clarify it. Obviously, fleshing out an example here is not going to be possible. If I affirm the theory, have I contradicted myself in the way that literally affirming analogical predication of God does? No. Granted, if the theory that everything we say about God is misleading is true, then this theory is itself misleading. But it can be true and yet misleading.³ And pragmatically speaking, even when our concern is for truth, we can have good reason to speak misleadingly. Pretty much anything we say about physics that small children—and probably ordinary adults as well—can understand will be misleading (we can avoid saying the false by hedging with “roughly speaking,” but that doesn’t avoid misleading), but that does not mean we should give up on science education. Judaism, Christianity, and Islam all have a deep reverence for scriptures. On the misleadingness theory of ineffability, even if a scriptural text about God is true, it will be misleading. We might see this as helping to explain both the division between the religions and within each religion. Thus, a Jew or Muslim could see the Christian doctrine of the Incarnation as a response to a misleading implication of God’s aloofness in the doctrine of absolute divine transcendence, while a Christian could see Jewish and Islamic rejection of the doctrine of the Trinity as a response to a misleading implication of divine complexity in the doctrine. And seeing scripture as misleading is quite compatible with a high view of the inspiration of that scripture: that a text is misleading in at least one way is compatible with its being illuminating in a great many other ways and with its being simply true. The misleadingness theory of ineffability could stand alone, but it could also be a complement to one of the semantically oriented theories. There are at least two ways this could be worked out. First, it could be a

³ In what way? I could point something out—perhaps it could lead to undue pessimism about theology—but if I pointed it out, that would merely shift the bump under the rug. Alternately, one could follow a suggestion from a referee and distinguish “substantive” and “merely formal” claims about God—much as one might in defense of Aquinas from the question whether analogicity is merely analogous when applied to God—and hold that the misleadingness theory is merely formal.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

305

semantic theory of ineffability could explain why our pragmatic theory is true. On that view, the pragmatic theory would be a corollary rather than a stand-alone theory. Second, the pragmatic theory could add to the semantic theory. Thus, all the semantic theories agree that we have true literal negative predications of God: say, that God is not an airplane. However, our pragmatic theory will apply even to this seemingly innocent claim. Perhaps the claim is misleading by implicating that God lacks some perfection of an airplane, or that there is a negativity in God’s being. Given this sketch, let us move on to the seemingly unrelated problem of God’s freedom in creation. God knows everything and is perfectly rational. So if God has better reason to do A than to do an incompatible act B, then God will know about that better reason and being perfectly rational will act in accordance with it, and hence won’t do B. But it seems that God has better reason to create than not to create: as Johnston (2019) and O’Connor (this volume) suggest, by creating God manifests his glory. There may be multiple incommensurable⁴ or equally good ways for God to manifest his glory, and hence there may be some scope for contingency as to what God creates, but it seems that God has better reason to create than not to create, and hence God will not fail to create. This line of reasoning, however, will strike many theists as problematic. God’s creation is often seen by theists as something gratuitous, and the idea that God has to create at least appears in tension with divine selfsufficiency. And for Catholics, the First Vatican Council even defined it as dogma that God did not have to create. Now, ineffability was a thesis about God’s relationship to language. But it would be good if we could ground our pragmatic ineffability in something deeper and even more general than the shortcomings of linguistic communication. To that end, I want to suggest that the pragmatic theory of ineffability can be grounded in a deeper hypothesis: (NMM) Necessarily, any manifestation of God is misleading. When someone talks about God, they either speak truly or falsely. If they speak falsely, that misleads. And if they speak truly, God is manifested in ⁴ See Pruss (2016) for an account of creation focused on incommensurability.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

306  .  the utterance—and hence the utterance is misleading by NMM. Thus, NMM together with the observation that false utterances mislead explains why all that is said about God is misleading. We can make NMM plausible given classical theism’s doctrine of divine simplicity. Aquinas argues that only God is perfectly simple, since perfect simplicity implies a unity between an entity’s nature and its act of being, and it is only in a being that is the fundamental and necessary ground of reality that the form and the act of being can be one. Thus, on Aquinas’s view, all creatures necessarily mislead as to God’s simplicity. But we can speculatively say something more. God is perfectly simple and yet his nature is infinitely rich, whereas among creatures, a richer nature, like that of a human, is linked to a greater complexity, while a simpler nature, like that of an electron, is impoverished. Insofar as God is manifested in more diverse things, that manifestation misleads with respect to God’s simplicity. But insofar as God is manifested in less diverse things, this manifestation misleads with respect to the infinite richness of God’s simple perfection. It is only in God that the seeming opposites of richness and simplicity can meet, and they do so mysteriously. On the Oppy‒Pruss‒Brower account of divine simplicity (Oppy 2003, Pruss 2008, Brower 2009), divine simplicity means that God is the sole truthmaker for every true intrinsic claim about God. This ontological simplicity, however, is compatible with what one might call infinite richness, the idea that there are infinitely many correct intrinsic diverse (and perhaps radically diverse) attributions of a perfection to God. In creatures, on the other hand, it seems plausible that an infinite multiplicity of diverse perfections cannot meet with a single truthmaker. There would be something quite rationally intelligible about someone who accepted the misleadingness theory of ineffability therefore resolving never to speak of God, so as never to mislead about God. This might not be the best course of action, however, in light of the needs of others, which may require that we not be so fastidious about avoiding misleading that we fail to inform them of important truths. But in any case, we can see that there is good reason to be silent about God, even if this good reason is not conclusive.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

307

Similarly, God has good reason to avoid misleading manifestations of God. Once God has decided to create people, the reason to avoid misleading manifestations may sometimes be overridden by the needs of these people. But explanatorily prior to the decision to create people, the reason to avoid misleading manifestations of God seems like a compelling reason, one that can stand incommensurably against the reason to manifest. There are two ways of exhibiting respect for a value: by promoting the value and by fighting off what is contrary to that value. Sometimes the two ways will generate opposed reasons. The reason to speak of God, in order that truth be communicated, and the reason to be silent about God, lest one mislead, are such opposed reasons, and unavoidably so if the misleadingness account of ineffability is correct. Similarly, God finds himself with two opposed reasons: to manifest and to avoid misleading manifestation. And it is plausible that in the case of the fundamental question whether to create, those two reasons are incommensurable, and God can rationally choose to manifest or not to manifest himself. This is a brief sketch that raises multiple questions. There is only room for one of these in this chapter. We can ask whether in NMM, “manifestation” and “misleadingness” should be understood in such a way that that there is always someone (other than God) to whom God is made manifest and who is misled. If yes, then it appears that God would have two ways of avoiding misleading manifestation: by not creating at all or by creating a reality that does not contain persons. On this reading, we may have given a rationally compelling, though not conclusive, reason for God not to create persons, but it is not a reason for God not to create at all. However, when thinking of God as artist, we need not think of God’s manifestation as always manifestation to someone. We are discovering that our galaxy is full of planets. It is likely that many of these planets are empty of life, and they are likely to remain unobserved, except perhaps as the tiniest blurred spots in massive telescopes. And yet it seems that the adherents of the monotheistic religions will insist that the unseen-bycreatures beauties and wonders of nature of these planets manifest the glory of God. And if there can be unseen manifestations, there can be

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

308  .  unseen misleading manifestations. This seems paradoxical, but the following story may help. You are shipwrecked on a desert island with no hope of rescue. You could make a sketch of someone you love on bark with a piece of charcoal. If your artistic skills are at the level of mine, your sketch will be a stick figure, and such a manifestation of your beloved will be so misleading—even if no one is actually misled by it—that artistic integrity and love for the subject might well lead you to refrain from the production. If your sketching skills are at the Da Vinci level, you will have good reason to sketch your beloved, but you will still recognize that your sketch is so far short of your beloved, so misleading, that you also have an incommensurable good reason not to. The last story suggests that God’s reasons to avoid a misleading artistic production, even absent an audience, would be grounded in love of the subject, namely God. We may be troubled by the thought that God’s love of God is what keeps God from creating a world: that seems egoistic. Note, however, that the worlds we are now considering are ones where there are no non-divine persons, since there was no difficulty as to victims of the misleadingness in worlds with non-divine persons. Furthermore, God is the ultimately lovable being, and so a perfect being’s response of love to God must be the ultimate love—even if the perfect being is identical with God. Those who are not convinced by this response to the egoism worry, on the other hand, may instead find themselves drawn to see a richer Trinitarian structure in creation, whereby the love of God for God is an interpersonal Trinitarian love.⁵

References Aquinas, Thomas (1920). Summa Theologiae. Trans. Fathers of the English Dominican Province. New York: Benziger Brothers. Brower, Jeffrey R. (2009). “Simplicity and Aseity.” In: T. P. Flint and M. C. Rea (eds), Oxford Handbook of Philosophical Theology. Oxford: Oxford University Press. Pp. 105–28. ⁵ I am grateful to Dean Zimmerman and an anonymous referee for a number of very useful comments that have greatly improved this chapter. I am grateful to Michael Willenborg and my audience at St Benet’s Hall for discussions of the ineffability part of this chapter.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

  

309

Chrysostom, John (2010). On the Incomprehensible Nature of God. Trans. Paul W. Harkins. Washington: Catholic University Press. Jacobs, Jonathan (2015). “The Ineffable, Inconceivable, and Incomprehensible God: Fundamentality and Apophatic Theology,” Oxford Studies in Philosophy of Religion 6: 158–76. Johnston, Mark (2019). “Why Did the One Not Remain Within Itself,” Oxford Studies in Philosophy of Religion 9: 106–64. O’Connor, Timothy (this volume). “Why the One Did Not Remain Within Itself ”, Oxford Studies in Philosophy of Religion 10: 000–00. Oppy, Graham (2003). “The Devilish Complexity of Divine Simplicity,” Philo 6: 10–22. Pruss, Alexander R. (2008). “On Two Problems of Divine Simplicity,” Oxford Studies in Philosophy of Religion 1: 150–67. Pruss, Alexander R. (2016). “Divine Creative Freedom,” Oxford Studies in Philosophy of Religion 7: 213–38.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

Index Adams, Marilyn McCord 173, 209 Adams, Robert 14, 15–16, 20, 38, 40, 56, 78 Aknin, L. B. 79 Aletti, Jean-Noel 229 Alighieri, Dante 277 Almeida, Michael 168, 229, 297, 299 Alston, William P. 13, 38, 48, 78, 144, 146, 164 Alvarez, George A. 94, 114 Alward, Peter 156, 164 Anscombe, Elizabeth 6, 38 Anselm 268, 282 Aquinas, Thomas 5, 23, 24, 25, 26, 27, 30, 31, 32, 35, 38, 97–8, 114, 149, 235, 247, 248, 252, 260, 266, 268, 270, 271, 272, 274, 275–6, 279, 281, 284, 301, 305, 307 Aristotle 16 Audi, Robert 48, 78, 145, 159, 165 Augustine 32, 242, 268, 284 Avia, Maria 61 Bailey, Andrew M. 112, 114 Baker, Lynne Rudder 85, 106, 108, 114, 229 Baumeister, Roy 60–1, 78 Bawulski, Shawn 131, 143 Berger, Fred 45, 78 Bergmann, Michael 168, 229 Berkeley, George 83 Bigelow, John 186, 194 Braddon-Mitchell, David 221, 229 Bregman, Jay 261 Brenner, Andrew 94, 104, 114 Brower, Jeffrey R. 305, 308 Boden, Margaret A. 120–3, 131, 133, 140, 143

Bohn, Einar Duenger 99, 114 Broad, C. D. 220, 221, 229 Brownlee, Kimberly 59–60, 78 Bruton, Samuel 45, 79 Budd, Malcolm 43, 79 Butler, Joseph 1, 7, 17, 38, 39 Cameron, Ross P. 99, 114, 220, 225, 229, 297, 299 Camp, Elisabeth 157–8, 165 Chang, Ruth 66, 79, 238, 246 Chalmers, David J. 86, 114 Cheng, Peter C. H. 115 Chisholm, Roderick 97, 114, 229 Chrysostom, John 300, 308 Clarke, Emily 157, 165 Clarke, Samuel 291, 299 Climenhaga, Nevin 239, 246, 274 Cohoe, Caleb Murray 271 Collins, Stephanie 60, 79 Comesana, Juan 74, 79 Cowling, Sam 82, 114 Craig, William Lane 134, 143 Crisp, Oliver 168, 175, 203, 216, 229 Crisp, Thomas M. 191, 194, 200, 229 Croker, Steve 115 Cuneo, Terence 158, 165 Davis, Don E. 61, 101 Davis, Stephen T. 84, 95, 107, 117 De Lubac, Henri 276 DeRose, Todd 113, 114 Descartes, Rene 249, 253 Dewey, John 147, 165 DeYoung, Rebecca Konyndyk 20, 39 Dorr, Cian 102, 112, 117

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

312  Dowland, S. Clint 112, 113 Dunn, E. W. 79 Durkheim, Emile 146, 165 Edwards, Jonathan 97 Egan, Chas A. 266 Epley, Nicholas 54, 79, 80 Eriksson, John 44, 79 Evans, C. Stephen. 40, 79 Faulkner, Paul 65, 79 Fee, Gordon 168, 229 Feser, Edward 97–8 Fine, Kit 218, 230 Fitzgerald, Patrick 45, 79 Foot, Phillipa 6, 13, 39 Forrest, Peter 222, 230 Foster, John 104, 105, 115 Freddoso, Alfred J. 275 Geach, Peter 178, 230 Gendler, Tamar 157, 165 Gerson, Lloyd P. 97, 115 Gheaus, Anca 59, 79 Gobet, Fernand 94, 115 Goldberg, Sanford 58, 79 Goldschmidt, Tyron 108, 115, 168, 170, 182, 183–7, 198, 223, 230, 251, 253 Grant, Matthews 284, 287 Greenawalt, Kent 145, 165 Gregory of Nyssa 274 Habermas, Jurgen 145, 165 Hammamoto, D. 61 Heller, Mark 115 Hernandez-Lloreda, Maria 61 Hershenov, David B. 107, 115 Hinchliff, Mark 230 Hirsch, Eli 85, 115 Hirsch, Samson 157, 165 Hitchcock, Christopher 243 Hook, Joshua 61 Honneth, Axel 59, 79 Horgan, Terrence 100, 115

Howard-Snyder, Daniel 47–8, 79, 151, 152–3 Hudson, Hud 170, 183, 230 Howell, Andrew 61 Hursthouse, Rosalind 13, 16, 39 ibn Sina, Abu ‘Ali al-Husayn 249–50, 252, 253, 270 Inman, Ross 98, 103, 115 Jacobs, Jonathan 301, 308 Johnston, Mark 103, 115, 233–6, 246, 247–8, 249, 251–2, 253, 256, 278–87, 304, 308 Jones, Gary 115 Kant, Immanuel 146, 255–6, 258 Keller, Simon 58, 79 Kellner, Menachem 149 Kelly, Thomas 75, 80 Kim, Brian 72, 80 Kim, Jaegwon 104, 115 King, Karen Leigh 261 King, Peter 268 Klass, Dennis 53, 81 Kolts, R. 61 Kopec, Matthew 74 Korman, Daniel Z. 100, 115 Korsgaard, Christine 8, 39 Kraay, Klaas 119–20, 130, 131, 143, 298, 299 Kretzmann, Norman 217, 231, 240, 246, 260 Kriegel, Uriah 82, 83, 84, 86, 89–94, 113, 115 Kronen, John 97, 116 Kvanvig, Jonathan 159, 165 Kumar, Amit 54, 80 Kymlicka, Will 145, 166 Lane, Peter C. R. 115 Leary, Mark 60–1 Lebens, Samuel 147, 151, 152, 166, 168, 170, 182, 183–7, 198, 223, 230, 251, 253 Leftow, Brian 100, 116, 134, 143

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

 Leibniz, G.W. 123–4, 141, 143, 249–51, 252, 253, 288, 289, 291, 299 Leibowitz, Yeshayahu 160 Leiter, Brian 145, 166 Lewis, C. S. 30, 37–8, 39 Lewis, David 97, 100, 116, 168, 230, 255–6, 262, 296, 299 Liao, Matthew 60 Lineweaver, Charles H. 266 Lipton, Peter 243 LoLordo, Antonia 97, 116 Lurhmann, Tanya 159, 166 Lynch 59 Mackie, J. L. 124 Magidor, Ofra 151, 166 Maimonides, Moses 301 Malcolm, Finlay 151, 166 Malebranche, Nicolas 250–1, 253 Manela, Tony 43, 45 Marr, Raymond A. 155, 166 Martin, Rod 68, 69 Martineau, James 146, 166 Martinez-Marti 61 McComb, D. 61 McKaughan, Daniel J. 151, 152, 166 McFarland, Ian A. 132, 143 McGrath, Matthew 72, 80 McTaggart, J. M. E. 219, 220, 230 Mellor, David Hugh 230 Menssen, Sandra 97, 116 Menzel, Christopher 97, 116, 134, 143 Merricks, Trenton 180, 186, 221, 222, 230 Miller, Kristie 99, 116 Molto, Daniel 113, 116 Moore, Edward 274 Morris, Thomas V. 134, 143 Mullins, Ryan T. 103, 116, 168, 230 Murphy, Colleen 46, 81 Murphy, Mark 168, 230 Niven, Larry 217, 230 Norton, M. I. 79 Nussbaum, Martha 1–5, 6, 7, 11, 12–15, 16, 17, 18, 19, 35, 39

313

Oatley, Keith 155, 166 O’Connor, Timothy 247–8, 250, 251–2, 253, 254, 256–60, 263, 268, 269, 269–76, 278–87, 288, 304, 308 Olson, Eric 106, 112, 116 Oliver, Iain 115 Oppy, Graham 168, 229, 305, 308 Origen 274 Osborne 94, 116 Otto, Rudolf 146, 166 Pace, Michael 56, 73 Page, Meghan 57 Pagin, Peter 64 Parmenides 269 Pearce, Kenneth 82, 83, 84, 86–8, 113, 116, 251, 253, 254, 268 Perry, Michael J. 145, 166 Pine, Julian M. 115 Pini, Giorgio 268 Plato 19, 82, 84, 116, 248–9, 253 Plantinga, Alvin 83, 97, 116, 124–6, 134, 140, 141, 143 Plotinus 97, 116, 232, 270, 279 Potrč, Matjaz 100, 115 Preston-Roedder, Ryan 55–6, 57, 66 Priest, Graham 230 Pruss, Alexander 98, 103, 115, 130, 143, 241, 242, 246, 304, 305, 308 Pseudo-Dionysius 260, 279 Putnam, Hilary 85, 116 Quinn, Philip L. 145, 166 Radzik, Linda 46, 81 Rathey, Markey 271 Raz, Joseph 145, 166, 172 Rea, Michael C. 97, 117, 168, 173, 180, 200, 209, 210, 215, 229, 230, 231 Regan, Richard J. 268 Remhof, Justin 85, 117 Riggs, Wayne 76, 81 Rinard, Susanna 71, 72, 73, 81 Roberts, Robert 56, 78, 81 Rogers, Katherin 279, 281, 284, 285, 287

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 22/1/2022, SPi

314  Rosen, Gideon 102, 112, 117 Ross, James F. 251, 253 Rowe, William 239, 246, 264 Russell, Bertrand 156, 178 Russell, Daniel 56, 81 Saver, Jeffrey L. 155, 166 Schellenberg, John 48, 81 Schoenfield, Miriam 74, 81 Schroeder, J. 79 Scott, Michael 151, 166 Scotus, Duns 268, 301 Seacord, Beth 108, 115 Seglow, Jonathan 59, 81 Seymour, Amy 180, 200, 230 Sider, Theodore 102, 112, 117, 185, 219, 230 Siderits, Mark 112, 117 Skiles, Alexander 99, 117 Spinoza, Baruch 255, 258 Stang, Nicolas 255 Steffen, Edith 53, 81 Strabbing, J. 168, 172, 201, 231 Stroud, Sarah 58, 66, 67, 81 Sullivan, Meghan 220, 231 Stump, Eleonore 168, 174, 207–19, 231 Swinburne, Richard 105, 111, 117, 168, 189, 231 Szabo, Zoltan 180, 199, 231 Teresa of Calcutta 154, 166 Titelbaum, Michael 74 Tillich, Paul 147, 167 Tite, Philip 261 Thomson, Judith Jarvis 191–2, 231

Trigg, Joseph Wilson 274 Tucker, Chris 239, 246 Turner, John 261 Uhder, J. 61 Udayana 97 Valliant, George 59, 81 Vallier, Kevin 145, 167 van Inwagen, Peter 82, 101, 106, 111, 112, 117, 178, 183, 231, 232 Vasubandhu 104, 117 Walker, A. D. M. 45, 81 Wallis, Richard 261 Watkins, James 131, 143 Watkins, P. C. 61 Wettstein, Howard 159 Williams, Thomas 268 Williamson, Timothy 232 Wittgenstein, Ludwig 156, 167 Wolf, Susan 55, 68, 81 Worthington, Everett 61 Wood, Alex 61 Wood, Jacob W. 276 Wood, Jay 56, 81 Wright, N. T. 168, 232 Yang, Eric T. 84, 95, 107, 117 Young, Kay 155, 166 Zagzebski, Linda 55, 81 Zimmerman, Dean 106, 108, 118, 163, 200–1, 220, 232, 239, 246