Does God Matter? Essays on the Axiological Consequences of Theism 0415793513, 9780415793513

Citation preview

Does God Matter?

“The collection of essays in this volume make a major contribution to a much-neglected issue in philosophical theology—the issue of the value of God’s existence. It is of course a central question, but largely unexplored in the literature. The range of associated issues in the theory of value, metaethics, modal metaphysics, conceptions of God, and rational choice will interest a wide audience of scholars.” —Michael J. Almeida, University of Texas at San Antonio, USA

The question of whether God exists has long preoccupied philosophers. Many accounts of God have been proposed, and many arguments for and against God’s existence have been offered and discussed. But while philosophers have been busy trying to determine whether or not God exists, they have generally neglected to ask this question: “Does it matter whether God exists?” Does God Matter? features eleven original essays written by prominent philosophers of religion that address this very important, yet surprisingly neglected, question. One natural way to approach this question is to seek to understand what difference God’s existence would—or does—make to the value of the world and the well-being of its inhabitants. The first essay sets the stage for the discussion of this topic. The three essays in Section I defend versions of pro-theism: the view that God’s existence would—or does—make things better than they would otherwise be. The four essays in Section II defend anti-theism: the view that God’s existence would—or does—make things worse than they would otherwise be. The three essays in Section III consider the interplay between the existential and axiological debates concerning the existence of God. This book presents important research on a growing topic in philosophy of religion that will also be of keen interest to scholars working in other areas of philosophy (such as metaphysics, epistemology, and value theory), and in other disciplines (such as religious studies and analytic theology). Klaas J. Kraay is Professor of Philosophy at Ryerson University in Toronto, Canada. He is the editor of God and the Multiverse: Scientific, Philosophical, and Theological Perspectives (Routledge 2015).

Routledge Studies in the Philosophy of Religion For a full list of titles in this series, please visit www.routledge.com

10 God and the Multiverse Scientific, Philosophical, and Theological Perspectives Edited by Klaas J. Kraay 11 Christian Ethics and Commonsense Morality An Intuitionist Account Kevin Jung 12 Philosophical Approaches to the Devil Edited by Benjamin W. McCraw and Robert Arp 13 Galileo and the Conflict between Religion and Science Gregory W. Dawes 14 The Arguments of Aquinas A Philosophical View J. J. MacIntosh 15 Philosophical Approaches to Demonology Edited by Benjamin W. McCraw and Robert Arp 16 Eighteenth-Century Dissent and Cambridge Platonism Reconceiving the Philosophy of Religion Louise Hickman 17 Systematic Atheology Atheism’s Reasoning with Theology John R. Shook 18 Does God Matter? Essays on the Axiological Consequences of Theism Edited by Klaas J. Kraay

Does God Matter? Essays on the Axiological Consequences of Theism Edited by Klaas J. Kraay

First published 2018 by Routledge 711 Third Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2018 Taylor & Francis The right of the editor to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Names: Kraay, Klaas J., 1975– editor. Title: Does God matter? : essays on the axiological consequences of theism / edited by Klaas J. Kraay. Description: 1 [edition]. | New York : Routledge, 2017. | Series: Routledge studies in the philosophy of religion ; 18 | Includes bibliographical references and index. Identifiers: LCCN 2017044431 | ISBN 9780415793513 (hardback : alk. paper) Subjects: LCSH: Theism. | God. Classification: LCC BL200 .D55 2017 | DDC 211—dc23 LC record available at https://lccn.loc.gov/2017044431 ISBN: 978-0-415-79351-3 (hbk) ISBN: 978-1-315-21099-5 (ebk) Typeset in Sabon by Apex CoVantage, LLC

Contents

List of Contributorsvii   1 Invitation to the Axiology of Theism

1

KLAAS J. KRAAY

PART I

Arguments for Pro-Theism37   2 God and Intrinsic Value

39

SCOTT A. DAVISON

  3 Axiology: Theism Versus Widely Accepted Monotheisms

46

MICHAEL TOOLEY

  4 An Agreeable Answer to a Pro-Theism/Anti-Theism Question

70

T.J. MAWSON

PART II

Arguments for Anti-Theism93   5 If There Is a Hole, It Is Not God-Shaped

95

GUY KAHANE

  6 The Problem of Magic STEPHEN MAITZEN

132

vi  Contents   7 The Absurdity of Life in a Christian Universe as a Reason to Prefer that God Not Exist

147

ERIK J. WIELENBERG

  8 Recasting Anti-Theism

164

TOBY BETENSON

PART III

Connections between the Existential and Axiological Debates179   9 Triple Transcendence, the Value of God’s Existence, and a New Route to Atheism

181

J.L. SCHELLENBERG

10 Arguments from Evil and Evidence for Pro-Theism

192

MYRON A. PENNER AND BENJAMIN H. ARBOUR

11 Plantinga’s Defence and His Theodicy Are Incompatible

203

RICHARD B. DAVIS AND W. PAUL FRANKS

Index225

Contributors

Benjamin H. Arbour holds a doctorate in theology from the University of Bristol and currently serves as the executive director of the Institute for Philosophical and Theological Research. He has edited or coedited several books, and he has published journal articles in the International Journal for Philosophy of Religion and Bibliotheca Sacra. Toby Betenson is lecturer in philosophy of religion at Bangor University. Richard B. Davis is professor of philosophy at Tyndale University College. Scott A. Davison is professor of philosophy at Morehead State University. W. Paul Franks is associate professor of philosophy and chair of the Department of Philosophy at Tyndale University College. Guy Kahane is fellow and tutor in Philosophy, Pembroke College, University of Oxford. Klaas J. Kraay is professor of philosophy at Ryerson University. Stephen Maitzen is the W. G. Clark Professor of Philosophy and head of the Philosophy Department at Acadia University. T.J. Mawson is the Edgar Jones Fellow and Tutor in Philosophy at St. Peter’s College, University of Oxford. Myron A. Penner is professor of philosophy at Trinity Western University. J.L. Schellenberg is professor of philosophy at Mount St. Vincent University. Michael Tooley is professor emeritus and College Professor of Distinction at the University of Colorado, Boulder. Erik J. Wielenberg is professor of philosophy at DePauw University.

1 Invitation to the Axiology of Theism Klaas J. Kraay

“What difference would—or does—God’s existence make?” In recent years, philosophers of religion have begun to tackle this question with vigour and rigour.1 As with many philosophical questions, it is deceptively simple to pose, but enormously difficult to answer. In this chapter, I set the stage for the remainder of the volume by introducing the reader to some of the core issues at stake in this area. I hope to be an impartial guide to this unfolding discussion. The phrase “axiology of theism” can be misleading in two respects, and so some preliminary clarifications are needed. First, since “theism” is sometimes taken to mean “belief in God”, the phrase “axiology of theism” can prompt the thought that the chief task here is to evaluate the (dis)value or (dis)utility of belief in God, or perhaps of some individual or society engaging in religious practices oriented towards God. While these are important projects in their own right, this literature does not concern them. Instead, the central goal is to attempt to understand the axiological import of God’s existence, or non-existence, for the world and its inhabitants. Second, while most of the discussion has indeed been about this issue, an important subsidiary thread has considered what sorts of preferences can be rational with respect to God’s existence or non-existence. As we will see, one point of dispute has been whether preferences must track axiological judgements in order to be rational. In section 1, I attempt to clarify the central axiological question. In section 2, I distinguish a range of positions that might be held on this issue. In section 3, I set out some considerations favouring each of the main positions. In section 4, I consider some connections between the debate about the axiological consequences of God’s existence and the debate about whether God exists. In section 5, I turn to the debate about rational preferences concerning God’s existence or non-existence. Finally, in section 6, I summarize the contributions to this volume and draw key connections between them.

2  Klaas J. Kraay

1. Clarifying the Axiological Question In this section, I set out sixteen considerations relevant to how to understand the following question: “What difference would—or does—God’s existence make?” 1. This question is phrased carefully in order to be neutral between two perspectives: that of someone who believes that God exists and that of someone who does not believe that God exists. (The latter, of course, may be either an agnostic or an atheist. Moreover, the latter clearly needn’t be non-religious—there are many, many non-theistic religious systems and worldviews, after all.) The theist thinks that God exists, and so when she poses this question, she asks what difference God’s existence really does make. The non-theist, on the other hand, asks what difference God’s existence would make, were God to exist. 2. This question does not encode any account of who or what God is, and of course, many, many different models of God have been discussed by theologians, philosophers, and others throughout human history. Clearly, then, how one answers this question will depend enormously on who or what one takes God to be. Most contributors to this debate, including the authors featured in this volume, take God to be omnipotent, omniscient, perfectly good, and the creator and sustainer of all that contingently exists.2 This is an enormously influential and important model of God—and one much-beloved by analytic philosophers of religion—but of course, this question could just as well be tackled with respect to other models of God.3 More broadly still, similar axiological questions could be posed about nontheistic worldviews or belief systems, whether religious or otherwise. 3. Sometimes, contributors to this discussion focus on ‘bare theism’— that is, they concentrate on the characteristics of God set out above, and attempt to determine their axiological import. But sometimes other features of God (or more broadly, features of typical theistic worldviews) are considered. For example, some have discussed the axiological implications of the view that, given theism, all human persons possess eternal life. Additional features of God, or of theistic worldviews, can be called ‘expansions’ of bare theism. 4. The ‘difference’ referred to in the question is generally taken to be a difference in value. So, for example, someone might say that God’s existence would make, or does make, things better. There are, however, many ways of specifying value: intrinsic, instrumental, aesthetic, prudential, moral, etc. The philosophers involved in this discussion have generally focused on the intrinsic value of God’s existence, and on the intrinsic and moral effects of God’s (non)existence for the world, and for persons and their lives.

Invitation to the Axiology of Theism 3 5. This foregrounds the question of what the ‘things’ are to which God’s existence might make an axiological difference. One candidate is the actual world. On this way of understanding the matter, our question asks what difference God’s existence would or does make to the value of the actual world. Another candidate value-bearer, as suggested in point 4, is persons. On this way of framing the issue, our question asks what difference in value God’s existence would or does make for some or all persons (or perhaps to the lives they lead). In points 6–13 I consider the world-based construals of our question, and in points 14–16, I turn to person-based construals. World-Based Construals 6. Suppose, for the moment, that the relevant value-bearer is the actual world, and that the scope of the intended comparison encompasses all of it. Someone who tackles our axiological question in this fashion aims to determine what difference God’s existence would or does make to the overall value of the actual world.4 Alternatively, the question could be construed as asking what difference God’s existence would or does make to some feature(s) or proper part(s) of the actual world, without commitment to any global claims about the overall axiological status of any worlds. Answers that focus on the overall axiological status of a world, or on the overall status of some feature(s) or proper part(s) of a world, can be classified as wide. Answers that focus on the axiological status of a world, or on some of its features or proper parts, in some respect only, can be classified as narrow. 7. Whether the question is taken to concern the actual world as a whole, or just some of its features or proper parts, answering it seems to demand a comparison, and so it is important to consider exactly what is being compared. The broadest version would presumably involve comparing all worlds that include God with all worlds that lack God—either overall or with respect to some feature(s) or proper part(s). At the other extreme, the most restricted comparison would be between the actual world, as one takes it to be qua theist or nontheist, and the closest comparator world in which one’s own position about theism is false—again, either overall or with respect to some feature(s) or proper part(s). (The theist, for example, would aim to compare the value of the actual world—whether overall or in some respect—with how things would otherwise be; i.e., how things are in the nearest possible world in which theism is false.) In between these limit cases are other alternative construals of the comparison that could be explored. For example, one might think that the former construal is too broad, and the latter too narrow. Accordingly, one might wish to determine what difference God’s existence would

4  Klaas J. Kraay or does make, not just to the actual world, but to worlds that are relevantly and sufficiently similar to the actual world. Of course, attempts to construe the question in this fashion bear the burden of developing and defending some account of what the relevant similarities are, and of the desired degree of similarity, in order to fix the referents for comparison.5 Challenges for World-Based Construals 8. Suppose that someone aims to compare the value of all worlds that contain God with all worlds that lack God—either in toto or with respect to some feature(s) or proper part(s). An epistemic objection might be levelled: someone might say that such a task is simply beyond our intellectual abilities.6 Alternatively, a metaphysical objection might be levelled: someone might say that failures of commensurability or comparability render the desired task impossible. Suppose, instead, that someone aims to engage in the more restricted comparison sketched in point 7, by comparing the actual world, as one takes it to be qua theist or non-theist, with the closest comparator world in which one’s own position about theism is false. Here are three possible objections. First, there may be no such thing as a unique closest possible world in either scenario. Perhaps, for example, two or more worlds are equally—and hence unsurpassably—close to the actual world. Alternatively, perhaps worlds asymptotically approach ours in degree of similarity. Second, perhaps the term ‘closest’ is ambiguous between several different interpretations, none of which is the most suitable. And third, even if one clear sense of ‘closest’ can be agreed upon, perhaps it is still vague which world is closest. 9. Moreover, even if there is a unique closest world to the actual world, the resulting comparison may not be what the inquirer really intends. As several authors have pointed out, a theist might well think that the closest possible world in which theism is false nevertheless contains a being who is as similar to God as possible without being God.7 Perhaps, for example, this being fails to know just one true proposition (and accordingly, fails to be omniscient). On this construal, the comparative question for the theist would be posed as follows: “Which world is better (either overall or in some respect): the actual world, in which God exists, or the closest possible world in which God fails to exist, but an omnipotent, perfectly good being exists, who knows all true propositions save one?” The problem is that this may not be the comparison that fundamentally interests the inquirer. Indeed, most of the contributors to this debate have focused on comparing the actual world (on the assumption that God exists) to the nearest possible world in which naturalism is true—or alternatively, comparing the actual world (on the assumption that naturalism is true) with

Invitation to the Axiology of Theism 5 the nearest possible world in which theism is true.8 In the rest of this chapter, I will follow suit (notwithstanding the concerns noted in point 8). 10. A further difficulty looms for all attempts to compare worlds in which God exists with worlds in which God does not exist. Theists often hold that God not only exists in the actual world, but in all logically possible worlds. In other words, God’s existence is logically necessary. On this view, there just are no logically possible worlds lacking God available for comparison. (Likewise, suppose that God’s existence is impossible. If so, then there just are no logically possible worlds containing God available for comparison.) If a suitable account of counterpossible judgements could be found, this might permit the needed comparison. But on the standard Lewis-Stalnaker semantics, all counterpossible claims turn out trivially true. Accordingly (for example), the theist who takes God’s existence to be necessary would have to maintain that things would be better if, per impossible, God were not to exist, and also that things would be worse. Evidently, this is problematic.9 If no sense can be made of the relevant comparative judgements, conceptually or semantically, then this axiological project seems doomed. To date, various responses to this serious challenge have been explored, although no consensus has emerged. Here they are: (a) Kahane (2011) suggests that perhaps we can intelligibly evaluate impossibilities. In partial defence of this suggestion, he notes that certain other debates in the philosophy of religion appear to presuppose that we can do just this. Consider, for example, the debate about whether morality depends upon God. Kahane says: “when one of Dostoevsky’s characters asserts that ‘If God doesn’t exist, everything is permitted’, this is not meant to be an indifferent remark. It is supposed to be, and taken to be, a horrible and frightening implication of atheism” (677). Elsewhere, he rightly points out that discussions of Pascal’s Wager, and of the problem of evil, also appear to presuppose that we can sensibly compare what things would be like on both theism and naturalism, and make coherent axiological comparisons (2012, 36).10 Assuming that the participants in these discussions agree that God’s existence is either logically necessary or logically impossible, Kahane can be read as arguing for a legitimate presumption that the comparisons presently at issue should be deemed intelligible, since they are widely taken to be so in relevantly similar domains. (b) Mugg (2016) expresses some sympathy for this presumption, but seeks to go further, by offering a model of how one might consider and evaluate impossibilities. Drawing on the work of

6  Klaas J. Kraay Leslie (1987), Nichols and Stich (2003), and Stanovich (2011), Mugg describes a process called cognitive decoupling. This occurs when “subjects extract information from a representation and perform computations on that extracted information” (448). The remaining information is kept separate from the reasoning process: it is “screened out” or “cognitively quarantined” (448). Mugg offers the example of Bugs Bunny picking up a hole, throwing it against a wall, and then jumping through. We can understand the cartoon’s narrative only if we engage in cognitive decoupling, by screening out and quarantining certain beliefs that would threaten the narrative’s intelligibility—for example, the belief that it’s impossible to pick up a hole. By analogy, then, we can perhaps make progress on the axiology of theism by screening out beliefs that make trouble for our evaluation of impossibilities. Of course, this will require a principled way of quarantining beliefs that is invulnerable to the charge of being ad hoc, and to the charge of unduly favouring one’s own position in this debate. Moreover, reasons are needed for thinking that this method for assessing impossibilities is reliable.11 (c) Another strategy, proposed by Kahane (2012) and endorsed by Moser (2013), is to make axiological judgements about closely related states of affairs whose possibility is not disputed, and then extrapolate from these judgements to the target case. Here is an example of this method. S might confidently judge that her life is far better than it would have been, had she been born in the Middle Ages. But assuming the necessity of origins, there is no possible world in which S herself really was born in the Middle Ages. So perhaps S’s judgement is grounded in a comparison of her actual life with the medieval life of someone distinct from S, but sufficiently similar in the relevant ways. (The trick here, of course, is to identify what the ‘relevant’ ways are, and to clarify what ‘sufficiently’ means.) Likewise, if one takes God’s existence to be logically impossible, one might instead imagine a world in which a godlike being exists, and use this world as a proxy for the true object of one’s axiological assessment. Kahane refers to such a proxy as an “adjacent possibility” (2012, 38), but of course, one might wonder whether it is intelligible to think of logically impossible scenarios as being close to, or closest to, logically possible worlds.12 Moreover, one might wonder whether intuitions about such proxy worlds can plausibly transfer to the target.13 (d) Kahane (2012, 36) offers another suggestion: that progress can be made by considering epistemic possibilities. Consider, for example, someone who believes that God exists in all logically possible worlds, but who takes it that atheism is nevertheless

Invitation to the Axiology of Theism 7 epistemically possible—true for all she knows. Such a person might make headway by considering some epistemically possible (for her) world in which atheism is true, and comparing it to the actual world as she takes it to be. Likewise, someone who believes that God fails to exist in all possible worlds, but who nevertheless takes theism to be epistemically possible, can do likewise, mutatis mutandis. Of course, as Kahane notes, this strategy is only open to those who do not take themselves to know that theism (or atheism) is true. Those who take themselves to know that theism (or atheism) is true cannot, after all, deem the relevant alternative to be epistemically possible. Moreover, one might wonder whether assessments about epistemically possible worlds are indeed suitable proxies for impossibilities. (e) In his 2012 paper, and also in his contribution to this volume, T.J. Mawson recommends that the theist take God’s existence to be metaphysically necessary but not logically necessary, and that the atheist take God’s non-existence to be metaphysically necessary but not logically necessary.14 He says that both should picture modal space, geometrically, as follows. The actual world is at the centre, surrounded by a field of metaphysically possible worlds (which, of course, are also logically possible). Travelling outward through this field, one eventually crosses a boundary of sorts, and enters into a ‘doughnut’ of metaphysically impossible yet logically possible worlds. According to Mawson, the actual world can fruitfully be compared, from the perspective of both the theist and the atheist, with worlds in this doughnut. In his chapter, he proposes some candidate worlds and offers a comparative axiological assessment from the perspective of the theist and the perspective of the atheist. Of course, whatever its merits, this move is unavailable to those who think that God’s existence is either logically necessary or logically impossible. 11. Suppose that a way is found to handle the challenges raised in points 8 and 10. A further matter must be considered before a comparative judgement can be attempted. Let’s begin by considering the issue from the perspective of the theist who wishes to compare the overall axiological status of the actual world with the overall axiological status of the closest world in which naturalism is true. To make the issue vivid, picture what Alvin Plantinga calls the ‘book’ of each world—roughly, the exhaustive list of all true propositions describing that world.15 In order to fix the comparator world, one might think that the theist should just cross out ‘true’ beside the statement “God exists” in the book, and replace it with ‘false’, then do the opposite for the statement “naturalism is true”, and then compare the resulting book with the actual world’s book. The problem is that

8  Klaas J. Kraay doing so might well seriously understate the differences between the two worlds.16 Here are three examples to illustrate the point: (a) Suppose it’s a consequence of theism that without God, nothing contingent exists. If so, then the world to which the actual world is compared is sparsely furnished: it contains only whatever necessary existents there are apart from God. (b) Suppose that God is the unique source of all value. If so, then the proposed comparator world, whatever else it contains, is entirely devoid of value.17 (c) On many expansions of bare theism, God is thought to imbue the world with teleological features—for example, the world and its inhabitants are often thought to have divinely ordained purposes. No such feature, evidently, can be present on naturalism.18 Let’s consider the matter again, this time from the point of view of the naturalist. (d) Suppose that the naturalist takes cosmic history to have begun with the Big Bang, and takes the history of all living things on earth to be fundamentally shaped by random mutation and natural selection. As she considers the nearest theistic world, she must now imagine God playing a causal role in addition to these processes. After all, God is the creator and sustainer of all that contingently exists, and so at the very least, God must be causally responsible for the Big Bang, and must conserve its results in existence, including all the events described by evolution.19 (e) Suppose that the naturalist is convinced that the actual world contains gratuitous evils, and that, if God were to exist, there would be no instances of gratuitous evil whatsoever.20 If so, then when the naturalist attempts to hold before her mind the nearest world in which theism is true, she must either conceive of the gratuitous evils in the actual world as deleted from the theistic world altogether, or else must conceive of them as altered in some significant way. (For example, perhaps the suffering remains phenomenologically identical, but is now embedded in some larger axiological context that renders it non-gratuitous.)21 (f) Suppose that the naturalist is convinced that the actual world contains instances of non-resistant nonbelief, and that if God were to exist, there would be no such instances.22 Again, as she attempts to hold before her mind the nearest world in which theism is true, she must imagine these instances either being deleted altogether, or else being modified so as to render them consistent with theism. The moral of points (a)-(f) is just this: if the axiological inquirer aims to assess the difference that God’s existence would or does make to

Invitation to the Axiology of Theism 9 the value of the actual world as a whole, care must be taken to consider just how different the comparator world is.23 12. Perhaps, however, it is excessively hubristic to attempt axiological comparisons of the actual world as a whole with some comparator world(s). Perhaps, as mentioned in point 8, it is not possible in principle, due to failures of commensurability or comparability between the actual world and the intended alternative(s). Or perhaps it is possible in principle, but not in practice, due to our epistemic limitations. (Are our cognitive powers really up to the task of correctly tracking all these changes, and assessing their axiological import?) Such concerns might motivate narrowing the scope of inquiry to the axiological implications of theism for some feature(s) or proper part(s) of the actual world. But here, too, the cautionary lesson of point 11 must be borne in mind: care must be taken to track all the changes due to theism (or naturalism) for the relevant feature(s) or proper part(s) of worlds.24 Person-Based Construals 13. As mentioned in point 5, another possible locus of axiological comparison is persons. A personal construal of our question asks what difference God’s existence would or does make for persons. This could be a broad question about all persons, or it could be narrowed to pick out just some person(s). At its most autobiographical, the question could be this: would or does God’s existence make things better for me than they would otherwise be?25 14. Personal comparisons could again be wide or narrow. A wide personal comparison seeks to determine, for some or all persons, the overall axiological effects of theism. In contrast, a narrow personal comparison seeks to determine the axiological effects of theism for some or all persons, in some respect or other. 15. Once the question is construed in personalistic terms, the same issues about fixing the comparator worlds that were discussed in points 8–11 will arise. Suppose—perhaps rashly!—that they can be solved satisfactorily. Even so, further difficult questions about personal identity must be faced. On typical expansions of theism, God is taken to have created human beings in his own image, and has given them a certain teleological structure. Thus, for example, the Westminster Confession of Faith states that “Man’s chief end is to glorify God and enjoy him forever.” So consider the atheist who wishes to compare, with respect to persons, the actual world with the closest possible world in which theism is true. Both worlds feature persons, to be sure. But the persons in the theistic world are also creatures, divinely

10  Klaas J. Kraay created and supernaturally endowed with a characteristic telos—the “chief end”. If having such a telos is an essential characteristic of persons—as theists might well suppose—then the comparative autobiographical version of the personal question is no longer intelligible. Put plainly, the atheist in question simply would not exist in the relevant theistic world. That aside, other personal comparisons may yet be possible, even in the absence of trans-world identity. For example, one might simply judge that some or all of the persons in some world(s) are better off than the persons in the other(s), without assuming personal identity across the relevant worlds.26 16. Care should be taken to consider the relationships between impersonal and personal foci of axiological analysis. Consider the following example. Whether the actual world is theistic or naturalistic, there is wide agreement that it features enormous quantities of moral wrongdoing.27 But on certain expansions of bare theism, all this moral wrongdoing must be viewed against the cosmic backdrop of God’s nature, plans, and activities. These actions are not, theists might say, ‘just’ moral wrongdoing—they are also sins, offences against God.28 As a result, actions that are phenomenologically indistinguishable might be worse on theism, due to their being sinful, than they would be on naturalism.29 And plausibly, this would have personal consequences: a sinful action could well be personally worse for the sinner on theism than would the phenomenologically indistinguishable moral wrongdoing on naturalism. Aggregating many such actions together might have impersonal consequences as well: indeed, ceteris paribus, the world as a whole might just be worse with respect to moral action if theism is true, quite apart from being worse for persons. So it must be remembered that the very same feature may have both personal and impersonal axiological dimensions or consequences.

2. Axiological Positions We began with this deceptively simple question: “What difference would—or does—God’s existence make?” Speaking somewhat loosely for the moment, let’s say that the pro-theist holds that God’s existence would or does make things better than they would otherwise be, and that the anti-theist holds that God’s existence would or does make things worse than they would otherwise be.30 But these, of course, are not the only possible positions. The neutralist holds that God’s existence neither makes things better than nor worse than they would otherwise be.31 The quietist about this issue thinks that, in principle, the question is unanswerable.32 And, finally, the agnostic about this issue thinks that the question is answerable in principle, but nevertheless withholds judgement about it. This view could come in two forms. The positive agnostic

Invitation to the Axiology of Theism 11 has judged that, given the available arguments and evidence, suspending judgement is the most reasonable thing to do. The withholding agnostic, on the other hand, is someone who simply withholds judgement, even about the statement ‘agnosticism about the axiological import of God’s existence is the most reasonable position’.33 These positions are set out in the horizontal axis of the table below. On the vertical axis of the table are three familiar positions: theism, atheism, and agnosticism. (I call these ‘existential’ positions because they concern whether God exists.) The point of bringing these together in a table is to illustrate various combinations of existential and axiological positions. A familiar one is theistic pro-theism. A person who holds these views believes not only that God exists, but also that God’s existence makes things better than they would otherwise be. Another common combination is atheistic pro-theism. A person who holds these views believes that God does not exist, but nevertheless holds that God’s existence would make things better, if God were to exist. At first glance, it might seem that every cell in the above table represents a coherent combination of positions, but this has been contested. For example, in their contributions to this volume, Michael Tooley and J.L. Schellenberg both argue that anti-theism entails atheism. If they are correct, then ‘theistic anti-theism’ represents an incoherent combination of views.34 Table 1.1  Axiological Positions

Existential Positions

Pro-Theism Anti-Theism Neutralism Agnosticism Quietism Theism Atheism Agnosticism

Drawing on the conceptual resources set out in Section 1, we can begin to add more precision to this table by distinguishing personal and impersonal versions of these axiological judgements. The former, obviously, focus on the axiological implications of theism for persons. The latter, meanwhile, focus on the axiological implications of theism for worlds (or their features or proper parts), without reference to whether these implications are good (or otherwise) for persons. Equally, we can distinguish narrow and wide variants of these positions. The former concern the axiological consequences of theism in one respect only, while the latter focus on the overall axiological effects of theism. The table below illustrates these distinctions for one column only—pro-theism—thereby generating twelve different broad pro-theistic positions.

12  Klaas J. Kraay Table 1.2  Axiological Positions Pro-Theism Impersonal

Anti- Neu- Agnos- QuietTheism tralism ticism ism

Personal

Existential Positions

Narrow Wide Narrow Wide Theism Atheism Agnosticism

These distinctions can also be applied to each of the remaining four axiological positions (anti-theism, neutralism, agnosticism, and quietism), thereby generating twelve distinct variants for each, for a grand total of sixty unique combinations. Some of these sixty combinations of existential and axiological positions can consistently be held together with other pairs. For example, consider an atheist who thinks that God’s existence would make things better for her in certain respects, but who is unsure about the overall axiological import of theism. Such an atheist would be both a narrow personal pro-theist and wide impersonal agnostic. But clearly, not all combinations are compossible. Most obviously, for example, a quietist of any stripe cannot also be a pro-theist, anti-theist, neutralist, or agnostic of the same stripe. To date, most work on this topic has concerned protheism and anti-theism, and mostly from the perspectives of the theist and the atheist. But, given the vast array of views distinguished here, it is clear that this discussion could be broadened in many ways. For ease of expression, I have stated these axiological positions fairly loosely. However, as explained in Section 1, various precisifications of each are possible, and the main columns above could be further subdivided to distinguish these. A couple of examples will illustrate this point. A wide, personal pro-theist holds that God’s existence would or does make persons better off overall than they would otherwise be. But as noted in point (13), above, this could be a view about just some persons, or about all persons. And as noted in point (7), this view could be expressed with reference only to some possible worlds, or all possible worlds. Meanwhile, a wide, impersonal pro-theist holds that God’s existence would or does make ‘things’ overall better than they would otherwise be. But there are various ways to precisify ‘things’. As noted in point

Invitation to the Axiology of Theism 13 (6), this could be a view about one or more worlds as a whole, or it could be a view about some feature(s) or proper part(s) of worlds. The moral here is just this: it is very important to state these positions very clearly prior to evaluating them.

3. Some Considerations Supporting Each Axiological Position In this section, I set out some reasons that have been offered (or in some cases, merely might be offered) to support each of the five main axiological positions on the horizontal axis of the foregoing tables. I do not intend to conduct a comprehensive review of all the arguments, or a thorough analysis of the relevant disputes. Instead, my goal is simply to orient the reader to some considerations favouring each view. Quietism The quietist holds that some or all precisifications of the comparative question are in principle unintelligible, and hence impossible to answer sensibly. One reason for this view was mentioned in section 1, point 10: someone who holds that theism is logically necessary (or alternatively, that theism is logically impossible) might believe that this fact, in principle, defeats any attempt to engage in comparative axiological analysis.35 A different motivation for quietism could involve incommensurability or incomparability. For example, someone who believes that all theistic worlds are overall incommensurable and incomparable with all naturalistic worlds would presumably think that no ‘world-wide’ form of the comparative question can, in principle, be answered. It’s important to see that one might be a quietist about some forms of axiological comparison, but not others. For example, suppose one holds that worlds cannot sensibly be thought to bear overall axiological status. On this view, quietism about ‘world-wide’ axiological comparisons would be in order. But one might still think it possible to compare worlds in various respects, and so one might not be a quietist about more restricted axiological judgements. Agnosticism As I’ve defined this view, the agnostic believes that some relevant axiological comparison is possible in principle, but also thinks that, in practice, it’s best to withhold judgement about it. As mentioned above, this view could come in two forms: positive agnosticism and withholding agnosticism. Either version can be motivated by considering the difficulties involved in making the relevant comparisons. At the broadest level, can we really be confident that we have the modal abilities to

14  Klaas J. Kraay represent before our minds two or more worlds in their entirety, and then form an overall comparative axiological judgement about them? The agnostic might doubt this, perhaps on the general grounds that our computational power is just not up to the job of assessing all that needs to be assessed, or perhaps for the more specific reason that we lack the ability to properly grasp the import of, or connections between, some or all of the value-adding and value-diminishing features of worlds. At the narrow level, a different motivation for agnosticism could stem from a scepticism about our ability to adequately isolate some feature(s) or proper part(s) of a world for axiological analysis, or it could stem from scepticism about our ability to engage in comparative axiological evaluation of such proper part(s) or feature(s), even if we can isolate them adequately. Similar points might be made for personal construals of the axiological question. Neutralism The neutralist thinks that, with respect to some or all forms of comparison, God’s existence makes no axiological difference whatsoever. This view is distinct from and incompatible with quietism, agnosticism, protheism, and anti-theism. One way to be a neutralist is to take the view that God’s (non)existence has no axiological effects whatsoever. Another is to hold that the axiological downsides of God’s (non)existence are precisely counterbalanced by the axiological upsides, or perhaps by holding that the downsides and upsides are, in Ruth Chang’s sense, “on par” (Chang 1997). Neutralism has not been defended, or even discussed, in the literature to date. It might, however, inform a view that has been discussed: apatheism. The apatheist has an attitude of apathy or indifference towards questions pertaining to God’s (non)existence. If one thought that God’s (non)existence makes no axiological difference, one might thereby be led to apatheism.36 Pro-Theism The pro-theist thinks that God’s existence would or does make things better than they would otherwise be (for some precisification of ‘things’, ‘better’, and ‘would otherwise be’). Here are several motivations for this view: (a) One motivation stems from the familiar theistic idea that God is an unsurpassable being—and indeed the only such being.37 One might think that any world featuring such a being must contain superlatively more intrinsic value than any naturalistic world, merely in virtue of God’s presence. And one might further think that this increase in value outweighs any putative axiological downsides that

Invitation to the Axiology of Theism 15 the anti-theist might posit. Such a view amounts to a strong form of wide impersonal pro-theism.38 While the foregoing pro-theistic consideration appealed to God’s existence, the remaining ones set out here appeal to God’s nature and presumed activities, sometimes by invoking various expansions of theism. (b) Another impersonal motivation for pro-theism involves cosmic justice. For example, suppose it’s the case that if God exists, God brings it about that the wicked always ultimately receive their just deserts, and that the good always ultimately receive their just rewards. A protheist might say that this makes theistic worlds better in at least this respect, and perhaps even overall, than they would otherwise be. A pro-theist might even say that it is also personally better for the good—and even for the wicked—to receive their just deserts. (Of course, there is also a familiar sense in which the truth of such an expanded theism is personally worse for the wicked—they may not wish to receive their just deserts, after all!)39 (c) It has often been held that only God’s existence can anchor objective morality. Assuming that the presence of objective morality is a goodmaking feature, this claim can be harnessed to support pro-theism.40 (d) As noted earlier, it is often thought that God would prevent every occurrence of gratuitous evil, and that this fact would make things better than they would otherwise be. (e) Relatedly, some philosophers accept versions of a patient-centred restriction (PCR) on the divine permission of evil.41 Strong forms of this restriction hold that God will permit someone to suffer evil only if the suffering ultimately benefits that very individual. Since there is no such guarantee on naturalism, the pro-theist might argue, God’s existence makes things better in this respect for persons and for the world—and if this respect is significant enough, the pro-theist can argue that God’s existence makes person’s lives, or the world as a whole, better overall. (f) Another common pro-theistic consideration involves the meaning of life. It is sometimes argued that God’s existence is necessary for human life to be meaningful.42 If true, and if it is a good thing that life can be meaningful, this too could be harnessed in an argument for pro-theism. (g) Many expansions of theism include the idea that God ensures an eternal afterlife for all. This might be thought to favour pro-theism. In Chapter 4, T.J. Mawson discusses an important supplementary doctrine concerning the afterlife: universalism. On this view, not only does everyone receive an eternal afterlife on theism, all persons go to heaven. Given a suitably positive view of heaven, universalism could be harnessed to support pro-theism.

16  Klaas J. Kraay Davis (2014) is a dialogue involving a character, Jill, who is a pro-theist. Jill advances four additional pro-theistic considerations.43 (h) Better answers to our deepest human questions are possible on theism than on naturalism. (More generally, it is sometimes said that only on theism is the universe fully intelligible.) (i) On theism (and not on naturalism), our lives matter in a rich way because God created us and loves us. (j) On theism (and not on naturalism) it is possible to be in a relationship with God, and to experience the peace of God’s presence, and these are great goods that cannot be obtained on naturalism. (k) On theism, our guilt for moral wrongdoing can be forgiven in a richer way than it can be on naturalism. Anti-Theism The anti-theist thinks that God’s existence would or does make things worse than they would otherwise be (again, for some precisification of ‘things’, ‘worse’, and ‘would otherwise be’). The following anti-theistic considerations are most clearly presented in the work of Guy Kahane (see his 2011 and 2012 papers, and his contribution to this volume). Kahane urges that on theism, but not on naturalism, “we necessarily occupy a subordinate position in relation to a being that is vastly superior to us in every respect” (Chapter 5, page 110). This general point, in his view, has a number of untoward consequences: (l) On theism, human beings can never have complete privacy, given God’s omniscience—and this is an inherently bad-making feature of the relevant worlds and their inhabitants that does not arise on naturalism.44 (m) On some expansions of theism, God’s nature is taken to be incomprehensible, and so on these views, ultimate reality is more inscrutable than it would be on naturalism. (Clearly, this view is in tension with point (h) above.) (n) On theism, God is our creator and sustainer, and so we depend upon him for our existence. On naturalism, meanwhile, our existence does not depend upon any extra-mundane person—and this is a better state of affairs. (o) On certain expansions of theism, God is taken to have a plan for the cosmos, and purposes for each of us. If this is true, then our ability to live our lives according to our own plans and purposes is significantly impaired in ways that it would not be on naturalism. (p) On theism, God’s moral status vastly exceeds ours. This grounds various duties in human beings, such as the duty to obey and to worship. Since there is “something deeply undignified in occupying

Invitation to the Axiology of Theism 17 such a subservient position” (Chapter 5, page 111), God’s existence reduces the dignity we would otherwise have on naturalism. (q) There are individuals whose life plans and projects centrally involve achieving certain sorts of goods, like privacy, knowledge, autonomy, independence, and dignity. Since God’s existence would curtail or eliminate these individuals’ ability to achieve these life plans and projects, their lives, accordingly, are rendered meaningless on theism but not on naturalism, or are at least rendered significantly less meaningful.45 Stephen Maitzen has defended two anti-theistic considerations that can be read as inverting or undermining some of the pro-theistic considerations listed above. (r) As we saw in point (c) above, a pro-theist might argue that God’s existence anchors morality. But in a 2009 paper, Maitzen argues that theism actually undermines morality. Maitzen focusses on the expansion of theism, mentioned in point (e) above, according to which God ensures that those who experience involuntary undeserved suffering ultimately benefit from this experience. This view, Maitzen says, entails that it’s false that we sometimes have a basic obligation to prevent undeserved involuntary human suffering—a consequence he deems absurd and morality-undermining.46 (s) In Chapter 6 in this volume, Maitzen argues that theism—far from making reality intelligible, as mentioned in point (h)—in fact compromises our capacity to understand the universe. While Kahane focusses on the unknowability of God—see point (m) above— Maitzen urges that the universe is fundamentally mysterious on theism in a way that it is not on naturalism. Erik Wielenberg has also defended anti-theistic considerations that can also be read as inverting or undermining some of the pro-theistic considerations above. (t) Wielenberg (2005) imagines a mother who sacrifices her own life so that her child can live. If God guarantees ultimate cosmic justice, as theists sometimes suppose, then this mother’s actions will, ultimately, be rewarded. But on naturalism there is no such guarantee—and accordingly, the truth of naturalism makes possible acts of true self-sacrifice in a way that theism does not. While cosmic justice appeared in the list of pro-theistic considerations—point (b) above—Wielenberg’s argument here seeks to identify an anti-theistic drawback of this very feature. (u) In Chapter 7 of this volume, Wielenberg argues that the truth of Christian theism would render human life absurd in a very specific way: human beings who see what Christianity requires would

18  Klaas J. Kraay experience negative psychological consequences that would make it difficult or impossible for them to be happy. Wielenberg argues that on Christianity, we have a moral obligation to pursue the good of others. But the Christian God would also ensure that every instance of involuntary undeserved suffering ultimately makes the sufferer better off than she would otherwise have been. Accordingly, on Christianity, we have an obligation to inflict undeserved and involuntary suffering on others—and realizing this suffices to make life absurd in Wielenberg’s sense. This can be seen as inverting or undermining point (e), above, and as an extension of Maitzen’s argument mentioned in point (r) above.

4. Connections between the Existential and the Axiological Issues As visually depicted in the tables above, the axiological issues are orthogonal to the existential question of whether God exists. It might seem, accordingly, that they are entirely independent. But this has been contested in several intriguing ways. 1. In their contributions to this volume, Michael Tooley and J.L. Schellenberg both argue that the anti-theist must, in order to be consistent, be an atheist. Tooley argues that if (wide) anti-theism is true, then it is better to bring about a world in which God does not exist. An omnipotent, omniscient, perfectly good being would unerringly bring about the better in lieu of the worse. And so, if such a being were to exist at any time, that being would ensure that no such being exists at some future time. But since Tooley thinks that if God exists, then God exists at all times, it follows that if there is any time at which God does not exist, then God simply does not exist. (And alternatively, if there is no time at which God exists, God doesn’t exist either.) Accordingly, Tooley thinks, anti-theism entails atheism. For his part, Schellenberg thinks that any argument for the claim that there is a genuinely worse-making feature of theism simply amounts to an argument for atheism. This is because he holds that any being worthy of the name ‘God’ could not possibly bring about any worsemaking feature whatsoever. 2. Another connection between the existential and axiological issues is brought out in the chapter by Myron A. Penner and Ben Arbour in this volume. They argue that proponents of certain arguments for atheism that appeal to evil are rationally required to be protheists. Such arguments, they say, essentially involve the following claim: “If God were to exist in a world, some possible really bad feature would be precluded from that world”. This consideration

Invitation to the Axiology of Theism 19 constitutes one respect in which things would be better on theism, and so it is tantamount to a narrow form of pro-theism. They go on to argue that this consideration is weighty enough to justify wide pro-theism as well. 3. Travis Dumsday (2016) draws a connection between arguments for atheism that appeal to non-resistant non-belief and a certain form of anti-theism. On some expansions of theism, it is possible that persons will justly face harsh postmortem punishment for their misdeeds. This feature constitutes one respect in which people’s lives would be worse if such an expansion of theism is true, as noted above in section 3, point (b). Yet, while this thought supports narrow personal anti-theism, Dumsday argues, it also furnishes a response to arguments from non-resistant non-belief. It’s reasonable to suppose that God might ‘hide’—that is, might permit non-resistant non-belief in his existence—in order to mercifully spare at least some of us from awareness of our impending and just post-mortem fate. If this is right, Dumsday argues, then the occurrence of non-resistant nonbelief does not decisively count against the truth of theism, contrary to what philosophers like Schellenberg have held. 4. In Chapter 11, Richard Davis and Paul Franks draw a specific connection between the existential and axiological issues within the philosophy of Alvin Plantinga. Plantinga famously constructed a ‘defence’ against the logical problem of evil: a logically possible explanation of why God might permit evil to occur. In his more recent work, however, Plantinga has also offered a ‘theodicy’: a purported account of God’s actual reasons for permitting evil. This theodicy is unabashedly pro-theistic: it involves the claim that God’s existence ensures that things will be far better than they would otherwise be. Yet, by focusing on the case of Jesus, Davis and Franks argue that Plantinga’s defence is incompatible with his theodicy. 5. In a forthcoming paper, Daniel Linford and Jason Megill draw a very different connection between the existential and axiological issues. They describe several biases that plague our thinking (wishful thinking, valence effect, and the optimism bias), and point out that: “Common among these biases is the overestimation of the probability that a belief is true precisely because individuals associate positive outcomes with the truth of the belief” (13). If this is correct, then pro-theists are susceptible to over-valuing the probative force of the arguments and evidence for God’s existence—and likewise, anti-theists are susceptible to over-valuing the probative force of the arguments and evidence against God’s existence.47 Accordingly, those investigating the arguments and evidence for and against theism should carefully and conscientiously work to mitigate the effects of these biases.

20  Klaas J. Kraay

5. The Debate about Rational Preferences In an oft-quoted passage, Thomas Nagel remarks: “I want atheism to be true and am made uneasy by the fact that some of the most intelligent and well-informed people I know are religious believers. It isn’t just that I don’t believe in God and, naturally, hope that I’m right in my belief. It’s that I hope there is no God! I don’t want there to be a God; I don’t want the universe to be like that” (1997, 131). For ease of exposition, I will gather together sentiments like hopes, wishes, and desires under the heading ‘preferences’. So construed, Nagel here expresses a clear preference for God’s non-existence. Some authors have explored the issue of which preferences in this domain can plausibly be deemed rational. Suppose one holds that these preferences can be rational only if they track reasonable axiological judgements. On this view, for example, one can rationally prefer God’s non-existence only if one also takes it that things would be better without God than they would otherwise be (for some precisification of ‘things’, ‘better’, and ‘would otherwise be’). If this is right, then there is no separate issue about which preferences are rational in this domain—the discussion simply collapses into the debate about axiology discussed above. But some authors have cautioned that it can be rational, in certain cases, to prefer the worse to the better.48 If this is possible here, then the debate about rational preference is, after all, distinct from the axiological discussion. This will require some modification to the positions on the horizontal axis of tables 1.1 and 1.2. If the debate is construed to concern which preferences are rational, then the resulting positions (moving from right to left) are as follows. The quietist holds that, in principle, no plausible account can be given about which preferences are reasonable in this domain. The agnostic demurs, thinking that, at least in principle, some preferences could be deemed reasonable. (The positive agnostic holds that, at present, there is insufficient evidence and arguments to justify any preference. And the withholding agnostic simply withholds judgement, even about the statement ‘agnosticism about which preferences concerning God’s (non)existence are reasonable is the most defensible position’.) We saw earlier that the neutralist holds that God’s existence neither makes things better than, nor worse than, they would otherwise be. The corresponding position about rational preference is perhaps better termed indifferentism, and it is the view of one who thinks that, in practice, God’s existence can neither be rationally preferred nor dispreferred, and that, accordingly, the rational attitude to take is indifference. The anti-theist, of course, thinks that it is rational to prefer God’s non-existence, and the pro-theist thinks that it is rational to prefer God’s existence. The distinctions between wide and narrow, and between personal and impersonal, variants of the axiological positions can be applied here as

Invitation to the Axiology of Theism 21 well, thus generating sixty combinations of positions—combinations of a position on whether God exists with a view about rational preferences in this domain. Table 1.3  Preferences

Existential Positions

ProAnti- Indifferentism Agnosticism Quietism Theism Theism Theism Atheism Agnosticism

The clarificatory points about the axiological positions made in Section 1, above, can be applied, mutatis mutandis, to the preference-based construal of this debate. But there are some important differences to consider. For example, suppose one thinks that comparative axiological judgements are simply impossible, on the grounds that theism, if true, is necessarily true, and that, accordingly, no sense can be made of the comparative axiological question. Such a person would, of course, be a quietist about the axiological issue. But this needn’t commit such an individual to quietism about rational preferences. Stephen Davis offers a helpful example to illustrate this idea. While it is logically impossible for the laws of mathematics to be other than what they are, it is not clear that it is therefore irrational for the struggling math student to wish, per impossibile, that calculus was easier than it is.49

6. Chapter Summaries Chapters 2–4 develop and defend arguments for pro-theism. Chapters 5–8 develop and defend arguments for anti-theism. Finally, Chapters 9–11 explore connections between the existential and axiological issues.50 In Chapter 2, Scott A. Davison sets the stage by rehearsing his favoured account of intrinsic value. According to Davison, the intrinsic properties of a thing are those that it possesses by itself, quite apart from any relationships to other things. The intrinsic properties of a thing ground its powers, capacities, and dispositions, and these typically distinguish a thing from other things. According to Davison, for X to be intrinsically valuable, X must have an intrinsic structure that would provide a fully informed, properly functioning valuer with a reason to value X for its own sake. Armed with this conceptual framework, Davison offers an

22  Klaas J. Kraay argument for a limited form of pro-theism. He notes that if God were to exist, God would possess intrinsically a significant combination of great-making features (and perhaps even the best possible combination) to the highest possible degree. Moreover, God would be a fully informed, properly functioning valuer. Given his definition of intrinsic value, then, it follows that God would possess intrinsic value to a significant degree (perhaps even the highest possible degree). The result, says Davison, favours pro-theism, since worlds containing God would contain a significant amount of intrinsic value, whereas other worlds would contain nothing that possesses anything like this degree of intrinsic value. In Chapter 3, Michael Tooley offers three main arguments. First, he urges that the axiological and existential questions concerning God’s existence are not logically independent, since anti-theism entails atheism. Second, he defends impersonal pro-theism, and third, he indirectly defends personal pro-theism (by criticizing arguments for anti-theism).51 Tooley’s first argument begins by supposing that wide, impersonal anti-theism is true, and that God should be understood as an omnipotent, omniscient, and morally perfect good being who exists at every time. He argues that, given antitheism, it is better to bring about a world in which God does not exist than to refrain from doing so. An omniscient, omnipotent, and morally perfect being would, of necessity, choose the better action over the worse, when both are available.52 So, if such a being exists at any time, it would bring it about that God does not exist at some future time. And bringing this about, Tooley argues, would be within its power.53 If there is any time at which God does not exist, then God does not exist, given Tooley’s definition of God noted above. So, if an omnipotent, omniscient, and morally perfect being were to exist at any time, given wide anti-theism, atheism is true. And, equally, if an omnipotent, omniscient, and morally perfect being were to exist at no time whatsoever, atheism is true. Since these alternatives are exclusive and exhaustive, anti-theism entails atheism.54 Tooley’s second main argument is a defence of impersonal pro-theism. He offers two pro-theistic considerations: eternal life for human beings and the idea that justice will be achieved in the end. Both will plausibly occur on theism, Tooley thinks, but not on atheism—and both are very important good-making features of worlds. Tooley argues that the case for pro-theism is much stronger on atheism than it is on theism. One important reason is that the atheist, unlike the theist, can hold that all the suffering due to natural and moral evil that we find in the actual world either would not (or likely would not) obtain on theism.55 Finally, Tooley criticizes arguments for personal anti-theism, including those of Kahane (2011). The basic structure of Tooley’s criticism is this: for any putative personally worse-making consequence of God’s existence to which the anti-theist appeals, careful reflection on the nature of an unsurpassable being will reveal that it is not a consequence of God’s existence at all.56 For example: whereas Kahane argued that God’s omniscience would objectionably violate human being’s privacy, leading some people’s lives to

Invitation to the Axiology of Theism 23 be personally worse under theism than they would be on naturalism, Tooley counters that an omnipotent, perfectly good being would be extremely unlikely to be omniscient, on the grounds that it is very likely morally wrong, on at least some occasions, to access the thoughts and feelings of a person without his or her consent. Tooley thinks that similar moves could be made with respect to any other anti-theistic considerations. According to Tooley, the inclination to see personal downsides of God’s existence (as opposed to revising one’s concept of God such that these downsides do not obtain) is due to the inordinate conceptual influence of the major monotheistic religions, and correspondingly, inadequate attention to what should really be expected of an omnipotent, perfectly good being.57 In Chapter 4, T.J. Mawson seeks an answer to the axiological question on which theists and atheists can agree, even if the debate about whether God exists has not been settled. He proposes the following method, which he first advanced in an earlier paper (Mawson 2012). The theist and the atheist should each should begin with the actual world as they take it to be and should then search for the nearest comparator world in which the rival view is true. Since both take their own view to be metaphysically necessary, both will have to exercise their imaginations to travel ‘outward’ through modal space, past all the metaphysically possible worlds, and then into the merely logically possible worlds.58 Having arrived there, so to speak, each should inspect candidates for the closest world to the actual world (as they take it to be), and then engage in—admittedly speculative—comparative evaluation. Mawson first considers the matter from the theist’s perspective. As Mawson sees it, the theist has three defensible candidates for which world is the closest to (what she takes to be) the actual world. First, she might say that if God does not exist, nothing else does either, so the closest world may fairly be called Nothingness. This world, Mawson urges, is impersonally worse, since it lacks God. Moreover, he suggests, Nothingness cannot be personally worse for anyone, since it evidently contains no persons. Second, the theist might hold that the nearest logically possible world lacking God contains a simulacrum of the actual world, alike in all respects save those involving God’s existence and activity. This world, says Mawson, is impersonally worse—again since it lacks God—and personally worse too, since the theists in it mistakenly worship a non-existent deity.59 Third, the theist might hold that the nearest possible world lacking God contains not only a simulacrum of everything non-divine in the actual world, but also a being who is as similar as possible to God without being God, and who plays the same roles in that world that God does (according to theists) in the actual world. Mawson urges that such a world would be impersonally worse, since the deity it contains is inferior to God. Moreover, such a world would be personally worse for theists, since they would inadvertently be worshipping this inferior deity instead of God. Mawson then considers the matter from the atheist’s perspective, by exploring two candidates for the nearest logically possible world in

24  Klaas J. Kraay which theism is true. The first contains a simulacrum of the actual world as the atheist takes it to be, and in addition contains God. This world, Mawson says, is impersonally better in virtue of containing God. Moreover, Mawson says, atheists should agree that it is personally better for all, since theism logically entails universalism—the view that all persons ultimately enjoy a heavenly afterlife.60 The other candidate world that Mawson considers contains God but no creation at all. He urges that atheists should hold that it is impersonally better than the actual world as they take it to be, since God is such a superlative being. Moreover, Mawson notes that atheists should hold that such a world cannot be personally better for any creatures, since it evidently doesn’t contain any. Mawson’s official conclusion, then, is that until the debate about whether God exists is resolved, theists and atheists can agree on a view he calls weak pro-theism. The theist should hold that it would be somewhat worse (or at least no better) if atheism were true, and the atheist should hold that it would be somewhat better (or at least no worse) if theism were true. In Chapter 5, Guy Kahane develops and defends a novel argument for the anti-theistic claim that the worlds we should most prefer are ones in which God does not exist. Kahane concedes that God’s existence would make things better in various important respects. He has in mind the sort of consideration typically appealed to by pro-theists, such as the idea that theism ensures cosmic justice and eternal afterlives for everyone. But he cautions that these goods can be obtained without theism. Karma, for example, could secure cosmic justice, and afterlives (or at any rate, immortality) could likewise be guaranteed by various non-naturalistic nontheistic systems. Moreover, Kahane argues at length that theism would make things worse in other important respects. In so doing, he develops in greater detail some of the considerations initially set out in Kahane 2011: God’s existence, he says, would violate our privacy, compromise our autonomy, and make us objectionably dependent.61 Given these considerations, Kahane concludes that there are possible atheistic worlds that offer all (or at least most) of the benefits of God’s existence but without these serious costs. From this he infers that some atheistic worlds are either the best or at least among the best, and are at any rate superior to all theistic alternatives. And this, he says, justifies his overall conclusion about which worlds are preferable. Kahane takes pains to argue that, if his argument is successful, even theists should prefer God’s non-existence. Along the way, Kahane responds to criticisms of his earlier argument for anti-theism. In Chapter 6, Stephen Maitzen considers the relationship between, on the one hand, the truth of supernaturalism in general (and theism in particular) and, on the other hand, the intelligibility of the universe. The first half of his chapter consists in criticizing arguments for the claim that the intelligibility of the universe requires theism. One such argument invokes God to explain the laws of logic. Maitzen points out that this argument requires the claim that if God did not exist, the laws of logic would not hold—and he notes that the consequent of this conditional is senseless,

Invitation to the Axiology of Theism 25 since the laws of logic are necessary in the strongest sense. Another argument maintains that our cognitive equipment is reliable if and only if God exists. While Descartes famously held that God could not deceive, Maitzen demurs, and he thus objects that God’s existence would not suffice for the reliability of our cognitive equipment. A third argument holds that knowledge can occur only if God exists. Maitzen responds by saying that this argument illicitly requires that knowledge is infallible. In the second half of his chapter, Maitzen argues that supernaturalism in general (and theism in particular) seriously compromises humanity’s capacity to understand the universe.62 Maitzen defines naturalism to be the view that purposes aren’t fundamental: every being, action, or whatever that has a purpose (a goal, a telos) ultimately arises from things that have no purpose. In slogan form: “Purposes don’t go all the way down” (Chapter 6, page 138). He argues that this view is incompatible with many supernaturalistic views, including theism. While naturalism allows human discovery to be limitless in depth, theism holds that something magical stands at the foundation of our universe: a purposive being whose aims cannot be understood by our scientific method. Maitzen concedes that, given theism, we should still ask indefinitely many nonscientific questions about God, but he urges that the growth in our knowledge of the natural world over the past few centuries vastly outstrips the growth in our knowledge of God—and this is not the sort of asymmetry we should expect if inquiry into the nature of God were a promising line of research. Maitzen closes by briefly developing further related anti-theistic considerations: if God exists, he argues, the universe is perversely unpredictable, frustrating, and scary. In Chapter 7, Erik J. Wielenberg turns his attention to Christian theism. He argues that the existence of the Christian God would make life absurd in a very specific way.63 According to Wielenberg, the God of Christianity is essentially omnipotent, omniscient, morally perfect, triune, and has commanded us to love him with all our hearts and to love our neighbours as ourselves. The final clause is the main focus of his paper. Wielenberg offers the following definition of the expression claim C makes life absurd: “Claim C’s truth makes (or would make) true at least one claim C1 such that most (actual) human beings are such that if they were to accept C1 they would experience negative psychological consequences that would make it difficult or impossible for them to be happy (without also failing to accept at least one entailment of C)” (Chapter 6, page 148). Wielenberg takes for granted that a morally perfect God would not permit the occurrence of any gratuitous evil, and moreover, that such a being would ensure that every instance of involuntary and undeserved suffering ultimately makes the individual who experiences it better off overall then she would otherwise have been. A consequence of this view is that whenever A inflicts involuntary undeserved suffering on B, B is ultimately better off than she would otherwise have been. Moreover, since the Christian God requires us to love our neighbours as ourselves, and since this involves pursuing the good of others, it seems that, if Christianity is true,

26  Klaas J. Kraay we have a moral obligation to inflict involuntary and undeserved suffering on others. Realizing this appalling truth would cause most human beings to experience negative psychological consequences that would make it difficult or impossible for them to be happy, and so, by Wielenberg’s definition of absurdity, such individuals’ lives would be rendered absurd once they understood this. Wielenberg argues that these negative psychological consequences would occur whether or not people acted upon their moral obligation to inflict involuntary and undeserved suffering. After considering and rejecting some objections, Wielenberg connects his argument to the axiology of theism. He expresses doubt that either wide impersonal or wide personal anti-theism can be established. In the former case, this is because (a) God’s existence itself would add great value to the world, and because (b) God’s preventing all gratuitous evil would ensure that the absurdity appealed to in Wielenberg’s own argument could not possibly make the world worse overall than it would otherwise be. In the latter case, this is because God would ensure that the negative psychological consequences (which, after all, are involuntary and undeserved) ultimately benefit those who suffer them. Wielenberg instead defends the claim that, given his central argument, it is reasonable to prefer, desire, or hope that atheism is true—even if theism would make things better overall than they would otherwise be, both for the world as a whole and the individuals in it. This idea is explored in greater detail in the subsequent chapter. In Chapter 8, Toby Betenson develops and defends a new construal of anti-theism. He begins by noting that anti-theism is often taken to hold that one can reasonably prefer the non-existence of God only if God’s existence would make the world worse than it would otherwise be. Critics of anti-theism typically deny that this necessary condition is satisfied, and Betenson accepts that their arguments are entirely persuasive. (This is because Betenson, like Wielenberg, holds that if God were to exist, then God—a supremely powerful, knowledgeable, and loving being—would ensure that any ‘downsides’ of his existence would be used to bring about greater goods, and that this would ensure that the world is not worse than it would otherwise be.) But Betenson argues that this condition is not necessary at all: it can be reasonable to prefer God’s non-existence even while conceding that God’s existence fails to make things worse, and indeed even makes things better. He offers several analogies in support of this claim. One can rationally accept that the presence of security measures at airports makes the world better than it would otherwise be, and that the taking of painful medication makes one’s life better than it would otherwise be, while nevertheless rationally preferring that the world not be such that these measures or medicines are necessary. Likewise, he urges, one can accept that God’s existence makes things better, while rationally preferring that God not exist. In the final, more speculative, section of his paper, he argues that such preferences might even be appropriate and meaningful even if they are not, strictly speaking, rational.

Invitation to the Axiology of Theism 27 In Chapter 9, J.L. Schellenberg carefully considers the relationship between anti-theism and atheism. He begins by claiming that the property of God-ness must realize three forms of transcendence: metaphysical, axiological, and soteriological. When these forms of transcendence are maximally instantiated, the resulting divine reality must be deemed triply ultimate: it is the ultimate fact, it has the ultimate worth, and it makes for the ultimate good of creatures and the world. On this account of God-ness, Schellenberg argues, God’s existence simply could not make a world or this world or someone’s world—the world of someone’s personal experience— worse than it would be if naturalism were true. So pro-theism initially seems secure. If the anti-theist were to insist that some property of God would make the world worse in some way, Schellenberg replies that the very fact that such a property is worse-making is sufficient reason for thinking that it shouldn’t be attributed to God in the first place.64 But suppose that this move fails, and that the relevant property should indeed be ascribed to God. In this case, Schellenberg argues, defenders of theism will need to argue that it is not really worse-making after all. If this move also fails, then the result is that anti-theistic considerations collapse into arguments for atheism: any genuinely worse-making feature of God’s existence simply shows that, given the requirements of triple ultimacy, God does not exist.65 Schellenberg then considers whether such arguments would constitute a new kind of argument for atheism, or whether they would simply fall under the heading ‘problem of evil’. He suggests that they would properly be treated as distinct from arguments from evil—since they would concern ineluctable features of God’s existence that occur in every world in which God exists, whereas arguments from evil always involve contingent claims about events or circumstances in the actual world. But, Schellenberg notes, typical anti-theistic considerations (about God’s existence compromising privacy, meaning, independence, or autonomy in human life) won’t be relevant in all theistic worlds—since of course God needn’t create beings whose privacy, meaning, independence, or autonomy would be compromised by his existence. And so Schellenberg doubts that such arguments could, strictly speaking, constitute a new route to atheism, independent of the problem of evil. In closing, however, Schellenberg suggests that there is nevertheless a way to understand anti-theism as part of a broad new strategy for supporting atheism. He claims that recent insights about cultural evolution have suggested new ways to criticize theism. For instance, consider his own prominent ‘argument from non-resistant non-belief’. This argument centrally appeals to the idea that a perfectly loving God would seek ongoing intimate personal relationships with all creatures who are capable of it. Schellenberg thinks that this is a new, more culturally evolved way of understanding the divine character—one that blinkered adherence to tradition can easily suppress. Evidently, this argument turns on a pro-theistic

28  Klaas J. Kraay consideration: the goodness of intimate, personal relationships between creature and created that would be sought by God. But, he suggests, the considerations typically invoked by anti-theists also invoke values made more prominent by cultural evolution: privacy, meaning, autonomy, and independence. To the extent that these considerations ultimately support atheism, they are indeed part of this new approach, even if they can legitimately be classified as arguments from evil. In Chapter 10, Myron A. Penner and Benjamin H. Arbour argue that proponents of certain arguments from evil are rationally required to be pro-theists. They begin by noting that some prominent arguments from evil are committed to the following auxiliary assumption: “If God were to exist in a world, some possible really bad feature would be precluded from that world” (Chapter 10, page 194). For example, the logical argument from evil maintains that if God were to exist, there would be no evil at all to be observed, and the evidential argument from evil holds that if God were to exist, there would be no gratuitous evil to be observed. Since proponents of these arguments are committed to the idea that God’s existence would prevent certain really bad features from occurring, they are committed to the idea that God’s existence would make the world better than it would otherwise be in these respects. Thus far, then, such individuals are committed to narrow pro-theism. But must they be wide pro-theists? Perhaps not, one might say. After all, anti-theists have argued that God’s existence would entail the occurrence of other really bad features that might outweigh the positive effects of God’s preventing the really bad features already considered. Penner and Arbour think that the prospects for such a move are dim indeed. They consider the putative negative consequences of God’s existence that anti-theists have considered (loss of privacy, loss of autonomy, loss of understanding, and the inability to pursue a meaningful life), and they argue that it is deeply implausible to think that such features—even if they are uniformly badmaking—can outweigh the value added to a world by God’s precluding the really bad features appealed to in arguments from evil. So, they conclude, defenders of these arguments from evil are rationally required to be wide pro-theists.66 In Chapter 11, Richard B. Davis and W. Paul Franks consider the relationship between two different responses that Alvin Plantinga has offered to the problem of evil—and argue that they cannot both succeed. In response to the logical problem of evil, Plantinga speculated that all creaturely essences suffer from a condition called ‘transworld depravity’. This condition guarantees that any world that God actualizes, if it includes free creatures, will also include moral evil. The possibility that all creaturely essences suffer from this condition is essential for the success of Plantinga’s defence against the logical problem of evil. According to Plantinga, a defence is a logically possible explanation of why God might permit moral evil. God, Plantinga thinks, has morally

Invitation to the Axiology of Theism 29 justifying reasons for instantiating creaturely essences, and so, since it is logically possible that all creatures suffer from transworld depravity, it is logically possible that God has morally justifying reasons for permitting moral evil. Plantinga’s argument is not without its critics, but it has been widely accepted. Independently, Plantinga has also offered a theodicy— a purported account of God’s actual reasons for permitting evil. Here, Plantinga argues that God will choose one of the best possible worlds to actualize—and that any such world will include the Incarnation and the Atonement. (Plantinga is thus a pro-theist: he holds that God’s existence ensures that things will be far better than they would otherwise be.) Now, since the Atonement logically requires the occurrence of moral evil—otherwise there would be nothing to atone—we have an account of why God permits it to occur: moral evil is an ineluctable feature of the best possible worlds. Davis and Franks seek to show that if Plantinga’s defence succeeds, his theodicy fails. They do this by focusing on the essence of Jesus Christ. According to standard Christian doctrine, Jesus is the second person of the Trinity, God Incarnate, and the one who brings about our atonement for sin. Davis and Franks argue that if Plantinga is right that all creaturely essences suffer from transworld depravity, then Jesus’ essence must so suffer as well. But if so, it must be the case that in any world in which Jesus’ essence is instantiated, Jesus will, in fact, go morally wrong at least once. And this is a big problem for the doctrine of the Atonement, since Christian orthodoxy takes for granted that only a sinless agent can bring this about. Accordingly, Davis and Franks claim, if Plantinga’s celebrated defence succeeds, his pro-theistic theodicy fails.

Acknowledgements This volume is the culmination of a three-year research project entitled “Theism: An Axiological Investigation”, which was generously funded by the John Templeton Foundation. (The project’s website is www.ryer son.ca/~kraay/theism.html.) Ancestors of Chapters 2–3, and of Chapters 6–11, were presented at a capstone workshop held at Ryerson University on September 11–12, 2015. I am very grateful for the Templeton Foundation’s support, and for the outstanding research climate at Ryerson University. I am particularly grateful to the four research fellows whose work was supported by this grant: Myron A. Penner and Richard Davis (2013–2014), W. Paul Franks (2014–2015), and Toby Betenson (2015). Thanks are due to all the contributors to this volume, not only for producing their fine chapters, but also for generously giving valuable feedback on the work of other contributors, and on this introductory chapter. I am also thankful to Stephen T. Davis, Chris Dragos, Kirk Lougheed, Morgan Luck, Graeme McLean, and Josh Mugg for their feedback on this chapter, and for valuable discussions of these issues. Thanks are also due to the three anonymous referees who commented

30  Klaas J. Kraay on the initial anthology proposal for Routledge. Finally, and most importantly, I am profoundly grateful to my wife, Mary Beth, and our two children, Emma and Jacob, for their unconditional love and support.

Notes 1. Although some themes from this discussion are present in Rescher (1990), the contemporary literature on this topic began with Guy Kahane’s very important and influential 2011 paper, entitled “Should We Want God to Exist?” 2. Some theists have also held that God is responsible for the existence of noncontingent entities as well. For more on this issue, see Gould (2014). 3. For discussions of other models of God that have been discussed by analytic philosophers, see Diller and Kasher (2013) and Nagasawa and Buckareff (2016). 4. This way of framing the question evidently presumes that possible worlds can sensibly be thought to have an overall axiological status. While this has been contested (e.g., Thompson 2008), it is widely granted in contemporary analytic philosophy of religion, and indeed by the contributors to this debate. For more on the standard assumptions about possible worlds and their value in analytic philosophy of religion, see Kraay (2008). 5. See Kraay and Dragos (2013, 162). 6. Peter van Inwagen’s (1998) modal skepticism offers some important reasons for thinking that we are just not up to the job. 7. This issue is discussed in Kahane (2011), and in Kahane’s and Tooley’s chapters in this volume, and in Mawson (2017). 8. In chapter 5, Kahane points out there are many different variants of naturalism to consider as well. 9. Davis and Franks (2015) propose a novel semantics informed by the null world hypothesis—the idea that if God doesn’t exist, nothing else does either. On this view, all counterpossibles involving the non-existence of God turn out to be false. As Mugg (2016) points out, this move also threatens the intelligibility of many comparative judgements at issue in the debate about the axiology of theism, and so, whatever its merits, it will not suffice if the goal is to avoid quietism about the axiology of theism. For more on this issue, see Kahane (2012, 36–37). 10. Kahane (2012) and Mugg (2016) also discuss philosophical disputes outside the philosophy of religion in which each side takes the other’s view to be impossible. Kahane discusses libertarianism versus compatibilism in the debate about free will, and Mugg discusses presentism versus eternalism in the philosophy of time. 11. Guy Kahane raised this point in personal correspondence. 12. On this point, see T.J. Mawson’s chapter in this volume. 13. On this point, see Mugg (2016). 14. Moser (2013) agrees, and so does Tooley, in his chapter in this volume. 15. More precisely: “. . . for any possible world W, the book on W is the set S of propositions such that p is a member of S if W entails p. Like worlds, books too have a maximality property; if B is a book, then for any proposition p, either p is a member of B or else not-p is” (Plantinga 1974, 46). 16. And as a result, the revised book might not represent a genuine possibility. 17. For discussions of this issue, see Kahane (2017), and the chapters by Mawson and Kahane in this volume. 18. While divinely ordained purposes cannot be present on naturalism, it might nevertheless be the case that the world or its inhabitants have been created

Invitation to the Axiology of Theism 31 with non-divinely-ordained purposes on naturalism. (Consider, for example, the idea that we are living in a simulation created by super-intelligent aliens.) I owe this point to Guy Kahane. 19. T.J. Mawson discusses this issue in chapter 4. 20. This is the dominant view in contemporary analytic philosophy of religion, but there have been some important dissenters. For details, see Kraay (2016a; 2016b). 21. For discussions of this point, see Kahane (2011, note 11) and Michael Tooley’s chapter in this volume. 22. Here I have in mind the argument, championed by J.L. Schellenberg, that has come to be known as the argument from divine hiddenness. See, for example, Schellenberg (1993). 23. It’s not the case that all such differences are relevant to this axiological inquiry: it is the changes due to God’s existence (or non-existence) that are germane. On this point, see Kraay and Dragos (2013, 161–162). 24. Sticking with Plantinga’s metaphor, we might think of features or proper parts of worlds as ‘chapters’ of books. And my point is just removing God from a chapter (or adding God to a chapter) may have many effects on that chapter. 25. To bring out the difference between personal and impersonal approaches, Mawson (2012) offers the vivid example of a meteorite striking and killing a relatively unimportant person. While this event may have only negligible impersonal significance in the grand scheme of things, or for the world as a whole, it clearly has enormous personal significance for the victim. 26. Scott Davison touches on this issue in chapter 2, section 3. 27. Moral nihilists, of course, demur. 28. For more on sin in this context, see Davis (2014, 152–154). 29. Contrariwise, someone might hold that otherwise indistinguishable morally good actions are better on naturalism than they are on theism. See section 3, point (t), below. 30. I should note that Kraay and Dragos (2013) followed Kahane (2011) in construing anti-theism as the view that God’s existence makes things far worse (and in construing pro-theism as the view that God’s existence makes things far better). Under pressure from Moser (2013), I here omit the modifier ‘far’. 31. This term is due to Mawson (see chapter 4 in this volume). Kraay and Dragos (2013) use the term ‘indifferentism’ instead, but on reflection, it is best avoided here, since it connotes a lack of preferences rather than an axiological judgement. 32. The application of this term in this context is due to Mugg (2016). 33. One might of course make the same distinction with respect to agnosticism about the existential question. I thank Nathan Ballantyne for clarifying this distinction for me. 34. I will return to this in Section 4, below. 35. A robust defence of this view would presumably provide reasons why none of the proposed solutions to this problem are viable. 36. For a discussion of apatheism, see Hedberg and Huzarevich (2017). 37. Whether this counts as an expansion of bare theism depends on whether it is a logical consequence of the attributes listed under ‘bare theism’. I concentrate on the attribute ‘unsurpassable’ rather than the more familiar ‘greatest conceivable’ in order to circumvent worries about whether conceivability tracks possibility. 38. Arguments in this vein are advanced by Davison (in chapter 2) and by Penner and Lougheed (2015). 39. This line of thought evidently presumes that no purely naturalistic mechanism could guarantee cosmic justice. But this might be contested: suppose,

32  Klaas J. Kraay for example, that our universe was created and sustained by very advanced, powerful, and benevolent aliens who take it upon themselves to ensure that we humans all eventually get our just deserts. One might think that, relative to this alternative, and in this respect, God’s existence cannot make things better than they would otherwise be. Here and in what follows, I bracket this type of consideration, by stipulating that the naturalistic alternatives to theism at issue do not contain mechanisms that play the same role as God is thought to play on theism. But for more on this issue, see Kahane’s chapter in this volume. 40. Of course, as T.J. Mawson has pointed out in correspondence, if one thinks that without God, there would be no value whatsoever, then one cannot sensibly be a pro-theist. One would have to hold that value is independent of God, but that morality depends upon God. 41. See, for example, Stump (1990); Tooley (1991); Rowe (1996). 42. A wonderful entry point to this literature is Seachris (2012). For a detailed treatment, see Mawson (2016). 43. They are expressed as preferences, not as axiological judgements, but Davis appears to presume that the former track the latter. 44. Of course—and here I echo a point made in note 39—there could be versions of naturalism on which human beings cannot have complete privacy either. Suppose, for example, that we are really just characters is a simulation designed by intelligent aliens, and that our thoughts are perfectly transparent to them. One might think that, relative to this alternative, and in this respect, God’s existence cannot make things worse than they would otherwise be. Here and in what follows, I bracket this type of consideration, by stipulating that the naturalistic alternatives to theism at issue do not contain mechanisms that play the same role as God is thought to play on theism. 45. Penner (2015) criticizes this argument. Lougheed (2017) responds. 46. Related moves are made in Hasker (1992; 2004; 2008) and Jordan (2004). But Hasker and Jordan aim to convince theists to give up the expansion of theism at issue, whereas Maitzen argues that theism, so construed, undermines ordinary morality. 47. Nagel agrees: “[I]t is just as irrational to be influenced in one’s beliefs by the hope that God does not exist as by the hope that God does exist” (1997, 131). See also Kahane (2012, 31–32). 48. Several versions of this view are sympathetically discussed in Kahane (2011), Luck and Ellerby (2012), and in Erik Wielenberg’s and Toby Betenson’s chapters in this volume. McLean (2015) responds to Luck and Ellerby (2012). 49. That wish could be construed as a wish for enhanced cognitive power, but I take it that Davis’ point is that it needn’t be. 50. Michael Tooley’s chapter not only defends pro-theism, but it also explores the connections between the axiological and existential issues, and so it might have been placed in the final section of this anthology. 51. Strictly speaking, Tooley only argues that there is no respect in which the world would be better if God did not exist. This is, of course, compatible with views other than pro-theism, including the view that God’s existence makes no difference to the overall value of a world. 52. Tooley makes an exception to this principle for cases in which the alternatives are situated in an infinite hierarchy of increasingly better actions—but he thinks that this is not such a case. 53. Tooley recognizes that this claim is controversial, given that God is often taken to be a necessary being. But he argues that God should not be understood to exist of logical necessity.

Invitation to the Axiology of Theism 33 4. Schellenberg also argues that anti-theism entails atheism in Chapter 9. 5 55. Penner and Arbour endorse a similar argument in Chapter 10. 56. Schellenberg endorses a similar argument in Chapter 9. 57. Schellenberg exhibits a similar sentiment in Chapter 9, and Kahane replies in Chapter 5. 58. Thus Mawson agrees with Tooley that God needn’t be regarded as a logically necessary being. 59. While Mawson thinks of worship as an appropriate response to God, in Chapter 3, Tooley suggests that no being should be deemed worthy of worship. 60. He argues for this entailment in Mawson (2005). 61. For dissenting views, see the chapters by Tooley and Schellenberg, both of which argue that a truly appropriate conception of God could not plausibly be thought to have these axiological downsides. Kahane’s chapter responds to these concerns. 62. Kahane (2011) gestures at such an anti-theistic consideration, but Maitzen develops it in greater detail. 63. Kahane (2011) also argues that on theism, certain people’s lives are absurd, but for different reasons: Kahane focusses on those whose life plans and projects are rendered unachievable by God’s existence, arguing that such people’s lives are meaningless, and hence absurd. 64. Michael Tooley endorses a related argument in Chapter 3. 65. Michael Tooley endorses a similar argument in Chapter 3. 66. It’s important to see the difference between Penner and Arbour’s argument here and the argument that Betenson endorses in Section 1 of Chapter 8: while Penner and Arbour maintain that the putative ‘downsides’ of God’s existence cannot outweigh the upsides of God’s existence, Betenson argues that God would ensure that any downsides of his existence would themselves be used to bring about greater goods.

References Buckareff, Andrei, and Nagasawa, Yujin, eds. 2016. Alternative Concepts of God: Essays on the Metaphysics of the Divine. Oxford: Oxford University Press. Chang, Ruth, ed. 1997. Incommensurability, Incomparability, and Practical Reason. Cambridge: Harvard University Press. Davis, Richard, and Franks, W. Paul. 2015. “Counter Possibles and the ‘Terrible’ Divine Command Deity.” Religious Studies 51: 1–19. Davis, Stephen T. 2014. “On Preferring that God Not Exist (or that God Exist): A Dialogue.” Faith and Philosophy 31: 143–159. Diller, Jeanine, and Kasher, Asa, eds. 2013. Models of God and Alternative Ultimate Realities. Dordrecht: Springer. Dumsday, Travis. 2016. “Anti-Theism and the Problem of Divine Hiddenness.” Sophia 55: 179–195. Gould, Paul. ed. 2014. Beyond the Control of God? Six Views on The Problem of God and Abstract Objects. New York, NY: Bloomsbury. Hasker, William. 1992. “The Necessity of Gratuitous Evil.” Faith and Philosophy 9: 23–44. Hasker, William. 2004. “Can God Permit ‘Just Enough’ Evil?” Chapter 5 in Providence, Evil, and the Openness of God, 81–94. New York, NY: Routledge. Hasker, William. 2008. The Triumph of God over Evil: Theodicy for a World of Suffering. Downer’s Grove: InterVarsity Press.

34  Klaas J. Kraay Hedberg, Trevor, and Huzarevich, Jordan. 2017. “Appraising Objections to Practical Apatheism.” Philosophia 45: 257–276. Jordan, Jeff. 2004. “Divine Love and Human Suffering.” International Journal for Philosophy of Religion 56: 169–178. Kahane, Guy. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696. Kahane, Guy. 2012. “The Value Question in Metaphysics.” Philosophy and Phenomenological Research 85: 27–55. Kahane, Guy. 2017. “If Nothing Matters.” Noûs 51: 327–353. Kraay, Klaas. 2008. “Creation, World-Actualization, and God’s Choice Among Possible Worlds.” Philosophy Compass 3: 854–872. Kraay, Klaas. 2016a. “God and Gratuitous Evil (Part I).” Philosophy Compass 11: 905–912. Kraay, Klaas. 2016b. “God and Gratuitous Evil (Part II).” Philosophy Compass 11: 913–922. Kraay, Klaas, and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 153–178. Leslie, Alan. 1987. “Pretense and Representation: The Origins of ‘Theory of Mind’.” Psychological Review 94: 412–426. Linford, Daniel, and Megill, Jason. Forthcoming. “Cognitive Bias, the Axiological Question, and the Epistemic Probability of Theistic Belief.” In Ontology of Theistic Beliefs: Meta-Ontological Perspectives, edited by Miroslaw Szatkowski. Berlin: de Gruyter. Lougheed, Kirk. 2017. “Anti-Theism and the Objective Meaningful Life Argument.” Dialogue 56: 337–355. Luck, Morgan, and Ellerby, Nicholas. 2012. “Should We Want God Not to Exist?” Philo 15: 193–199. Maitzen, Stephen. 2009. “Ordinary Morality Implies Atheism.” European Journal for Philosophy of Religion 1: 107–126. Mawson, T. J. 2005. Belief in God. Oxford: Oxford University Press. Mawson, T. J. 2012. “On Determining How Important It Is Whether or Not There Is a God.” European Journal for Philosophy of Religion 4: 95–105. Mawson, T. J. 2016. God and the Meanings of Life. New York, NY: Bloomsbury. Mawson, T. J. 2017. “Doing Natural Theology Consistently with Theism and Why One Might Stop Trying.” Religious Studies 53: 339–351. McLean, Graeme. 2015. “Antipathy to God.” Sophia 54: 13–24. Moser, Paul. 2013. “On the Axiology of Theism: Reply to Klaas J. Kraay.” Toronto Journal of Theology 29: 271–276. Mugg, Joshua. 2016. “The Quietist Challenge to the Axiology of God: A Cognitive Approach to Counterpossibles.” Faith and Philosophy 33: 441–460. Nagel, Thomas. 1997. The Last Word. Oxford: Oxford University Press. Nichols, Shaun, and Stich, Stephen. 2003. Mind Reading: An Integrated Account of Pretence, Self-Awareness, and Understanding Other Minds. Oxford: Clarendon Press. Penner, Myron A. 2015. “Personal Anti-Theism and the Meaningful Life Argument.” Faith and Philosophy 32: 325–337. Penner, Myron A. and Lougheed, Kirk. 2015. “Pro-Theism and the Added Value of Morally Good Agents.” Philosophia Christi 17: 53–69. Plantinga, Alvin. 1974. The Nature of Necessity. Oxford: Clarendon Press.

Invitation to the Axiology of Theism 35 Rescher, Nicholas. 1990. “On Faith and Belief.” In Human Interests, 166–178. Stanford: Stanford University Press. Rowe, William. 1996. “William Alston on the Problem of Evil.” In The Rationality of Belief and the Plurality of Faiths, edited by Thomas Senor, 71–93. Cornell: Cornell University Press. Schellenberg, J. L. 1993. Divine Hiddenness and Human Reason. Ithaca: Cornell University Press. Seachris, Joshua, ed. 2012. Exploring the Meaning of Life: An Anthology and Guide. Madden, MA: Wiley-Blackwell. Stanovich, Keith. 2011. Rationality and the Reflective Mind. Oxford: Oxford University Press. Stump, Eleonore. 1990. “Providence and the Problem of Evil.” In Christian Philosophy, edited by Thomas Flint, 51–91. Notre Dame: University of Notre Dame Press. Thomson, Judith Jarvis. 2008. Normativity. Chicago: Open Court. Tooley, Michael. 1991. “The Argument from Evil.” Philosophical Perspectives 5: 88–134. van Inwagen, Peter. 1998. “Modal Epistemology.” Philosophical Studies 92: 67–84. Wielenberg, Erik. 2005. Value and Virtue in a Godless Universe. Cambridge: Cambridge University Press.

Part I

Arguments for Pro-Theism

2 God and Intrinsic Value Scott A. Davison

1. The Question There are many ways in which people value things, and corresponding to these, there are different concepts of value: intrinsic, instrumental, prudential, aesthetic, sentimental, epistemic, economic, etc. In this paper, I will focus only on intrinsic value. With respect to the intrinsic value in the world, would it make any difference if God were to exist? Yes, it would, or so I shall argue here. Consider the possible positions represented by the chart found on p. 11 of Chapter 1. I will defend a very limited version of a position that would be found in the vertical protheism column (one of the families of views identified as “axiological positions”); here I shall take no stand with regard to what are called the existential positions with respect to God’s existence.

2. The Structure of Intrinsic Value There have been many different views concerning the possible bearers of intrinsic value. Some authors hold that the bearers of intrinsic value are states of affairs (e.g., Chisholm 1986; Lemos 1994; Bradley 2006, and many others); some hold that the bearers are states of concrete objects (e.g., Zimmerman 2001; Tännsjö 1999); and still others hold that the bearers of intrinsic value can occur in both of these categories (e.g., Davison 2012). I will not rehearse here my arguments, developed elsewhere, for the conclusion that concrete particular things can be bearers of intrinsic value (see Davison 2012, Chapter 3); instead, I will simply assume here that this is so. The intrinsic properties of a thing are those that it possesses by itself, apart from any relationships it has to other things. If we were talking about a concrete particular thing, for instance, and we possessed a machine that could create an atom-for-atom duplicate of it, then the duplicate would have the same intrinsic properties as the original at the moment of creation. (Here I am restricting myself to so-called qualitative properties, and ignoring those individual essences or haecceities posited

40  Scott A. Davison by some medieval thinkers, should they exist.) With regard to Rodin’s (original) statue known as the Thinker, for instance, we should say that being made of bronze, having a certain shape, being solid, and being inflexible are intrinsic properties of the statue. By contrast, the extrinsic properties of a thing are those that it possesses only because it is related to other things. If we could use a machine to create an atom-for-atom duplicate of some concrete particular thing, then the duplicate would not have the same extrinsic properties as the original, right from the moment of creation—for one thing, the duplicate would be the duplicate, and the original would be the original. With regard to Rodin’s (original) Thinker statue, we should say that being heavy, having been designed by Rodin, being located in France, and being both rare and important are among the extrinsic properties that it possesses. Although people sometimes conflate these categories, not all intrinsic properties are essential. The property of being made of bronze seems to be both intrinsic and essential to the Thinker statue designed by Rodin, but the property of having a tiny little piece on the side that is part of the thinker’s hair is an intrinsic property of the statue that is not essential to it. (If that tiny piece were broken off by mistake during a tour of the United States, and lost forever, France would be utterly unsympathetic to the claim that the original statue could not be returned, since it no longer existed, having lost an essential part.) So the intrinsic value of something might not be essential to it—in fact, it might change over time. For example, the dying mosquito on my window sill has less intrinsic value today than it had yesterday, when it was healthy and vibrant; the statue of Winged Victory (Nike) had more intrinsic value when it still possessed its head; and the creature that was the fertilized egg that became my son Andrew has more intrinsic value now than it did a couple of days after conception. (I will say more about degrees of intrinsic value below.) The intrinsic properties of a thing are the basis or grounds of its powers, capacities, and dispositions, which typically serve to distinguish it from other kinds of things. For X to be intrinsically valuable, then, is (roughly) for X to have an intrinsic structure that would provide a fully informed, properly functioning valuer with a reason to value X for its own sake. (Here I am inclined to follow Thomas Scanlon’s (1998) “buck passing” account of value, as opposed to G. E. Moore’s concept of a nonnatural property as the ground of intrinsic value, but I will not discuss this question here. For more on it, see Davison 2012, Chapter 1.) When I mention a fully informed valuer in this formulation, I have in mind someone who knows all of the relevant facts. This is an epistemic ideal that actual human valuers only approximate. When I speak of a properly functioning valuer, I have in mind someone who reliably judges value by having a properly disinterested stance, using all relevant

God and Intrinsic Value 41 information, etc. This is a normative ideal that actual human valuers only approximate. Different starting points have been suggested with regard to characterizing the notion of intrinsic value, of course; for example, G. E. Moore prefers the duty to choose one thing over another (Moore 1912, 66), C. D. Broad prefers to speak of the fitting object of desire (Broad 1930, 283), and Roderick Chisholm prefers to speak of what is required in contemplation (Chisholm 1986, 52–53). But all of these starting points presuppose some normative notion or other, and for reasons that I explain elsewhere, I prefer my formulation (see Davison 2012, Chapter 1). We make judgements concerning comparative value in general rather easily and naturally, and this seems true also with respect to intrinsic value. Following Aristotle’s hierarchy, for instance, it seems appropriate to say a human being has more intrinsic value than a rabbit, and a rabbit has more intrinsic value than a plant, and a plant has more intrinsic value than a rock. What determines degrees of intrinsic value? There seem to be several possibilities. First, it could be the case that certain combinations of intrinsic properties are intrinsically better than others, and this is just a brute, unexplained fact. (This seems to be Nozick’s view: see Nozick 1981, Chapter 5. For an extended discussion, see Davison 2012, Chapter 5.) Second, we could appeal to a notion of nesting to explain degrees of intrinsic value. Suppose that intrinsic property P includes intrinsic property Q if and only if (a) necessarily, if something has P then it has Q, and (b) possibly, something has Q but lacks P. Now we can say that if intrinsic properties P and Q confer intrinsic value on things, and P includes Q, then all other things being equal, things that possess P and Q are better than those that possess P but not Q. To illustrate this nesting account, consider again the Aristotelian hierarchy. A plant is better intrinsically than a rock because in addition to having all of the properties that make the rock intrinsically valuable, it also has intrinsic properties that serve as the basis or grounds of its distinctive powers, capacities, and dispositions, properties that the rock lacks. (Aristotle would describe the basis of these things as a “vegetative soul.”) A rabbit is better intrinsically than a plant because in addition to having all of the properties that make the plant intrinsically valuable, it also has intrinsic properties that serve as the basis or grounds of its distinctive powers, capacities, and dispositions, properties that the plant lacks. (Aristotle would describe the basis of these things as an “animal soul.”) And a human being is better intrinsically than a rabbit because in addition to having all of the properties that make the rabbit intrinsically valuable, it also has intrinsic properties that serve as the basis or grounds of his or her distinctive powers, capacities, and dispositions, properties that the rabbit lacks. (Aristotle would describe the basis of these things as a “rational soul.”)

42  Scott A. Davison So far, one could agree with everything I have said and still hold that actually, there is no intrinsic value in the world at all—so far, I have tried only to explicate the concept, without assuming (or arguing) that anything at all actually possesses intrinsic value. With regard to the actual distribution of intrinsic value in the world, there seem to be several possibilities. Elsewhere, I have argued that the most likely one is that everything that exists has some degree of intrinsic value, however small. I do not claim to have a powerful argument for this conclusion, but after comparing all of the possibilities with regard to other things we seem to know, seeking reflective equilibrium, this view emerges as the most likely, in my humble opinion. Although not widely accepted, this view provides a satisfying answer to what I have called the Cutoff Question (namely, what explains why some things have intrinsic value and others do not?). It is also supported by an argument that I have called the “Anything Is Better than Nothing” argument. (For extended discussion of these arguments, see Davison 2012). But the reader could reject my view that everything has intrinsic value and still embrace my main conclusion in this paper, as I will now explain.

3. God and Intrinsic Value Some have held that we can explain the existence of something like intrinsic value by appealing to what God values (e.g., Adams 1999). Of course, if intrinsic value did depend upon God’s valuing activity, then a world in which God were to exist would clearly be better than a world in which God did not exist, at least with respect to intrinsic value, since the former would presumably contain some and the latter would contain none. (I owe this argument to Klaas Kraay.) But this approach to intrinsic value seems obviously mistaken to me. Intrinsic value is independent of a thing’s relations to anything else, so it cannot depend upon God’s valuing activity. If my son has a certain degree of intrinsic value in a world in which God exists, for instance, then he must have that very same degree of intrinsic value in a world in which God does not exist. Of course, here I am assuming certain things about which properties are essential to my son. I am also assuming that we can compare meaningfully a world in which God and my son exist, and a world in which God does not exist and my son does; some traditional theists would hesitate here because they would hold that God is a necessary being (so that in fact, there are no possible worlds in which God does not exist and my son does). Still I would argue that the claim in question (namely, that my son must have the same degree of intrinsic value in a world without God as he possesses in the actual world) is more plausible than its denial. (For more on this set of issues, see the discussion in Davison 2012, Chapter 7.) To return to the larger question of the relationship between God’s valuing activity and the existence of intrinsic value: if God

God and Intrinsic Value 43 exists, then it is up to God whether or not to create anything, so it is up to God whether intrinsically valuable things exist—except for God, that is, as I will explain now. To return now to my main question: what difference, if any, would the existence of God make to the amount of intrinsic value in the world? Consider the following argument: P1. Something has intrinsic value to a certain degree if a fully informed, properly functioning valuer would value it for its own sake to that degree. P2. If God were to exist, God would be a fully informed, properly functioning valuer. P3. If God were to exist, God would possess intrinsically a significant combination of great-making features (perhaps even the best possible combination) to the highest possible degree. P4. If God were to exist, God would value God for God’s own sake to a significant degree. C. If God were to exist, God would possess intrinsic value to a significant degree (perhaps even the highest possible degree). P1 follows from the account of intrinsic value I have defended elsewhere (see Davison 2012), which was described above in section 2. Since being a properly functioning valuer involves having a combination of cognitive and evaluative skills, P2 follows from the traditional theistic picture, according to which God is omniscient and morally perfect. Together, these attributes guarantee that God would be a fully informed, properly functioning valuer in the sense described above in section 2. As a fully informed, properly functioning valuer, God would judge correctly and infallibly the degree of intrinsic value of everything (if any). P3 also follows from the traditional theistic picture, according to which God possesses the best combination of great-making features such as power, knowledge, goodness, etc. Of course, there are questions about how all of these attributes fit together into the greatest possible package, but according to traditional theists, however exactly those questions should be resolved, that’s exactly what God is like. (Here I am also ignoring questions about the doctrine of divine simplicity, which would require formulating the argument differently, but would not weaken the substance of the argument.) If P3 is true, and God is a fully informed, properly functioning valuer (P2), then God would value God, for God’s own sake, to a significant degree, which is what P4 says. Since the argument is clearly valid, and each premise is at least plausible, I take it that it provides a strong reason for thinking that the existence of God would make a big difference with regard to the intrinsic value in the world. For if God did not exist, there would be nothing that possessed anything like this degree of intrinsic value.

44  Scott A. Davison Of course, this does not mean that every possible world in which God exists is overall better than any world in which God does not exist. (Here I am assuming that God is not a necessary being for the sake of the argument; as noted above, most traditional theists will regard any world in which God fails to exist as an impossible world, but that is a question I will not attempt to resolve here.) If God existed in a world that was otherwise absolutely terrible (per impossibile), or God’s existence brought with it a degree of organic disunity that rendered a world awful on the whole, then some possible world without God might be overall better, on the whole, than some possible world in which God exists. Typically, though, the axiological debate about God’s existence concerns not every possible world, but rather the actual world or other possible worlds that are very much like our world. Setting aside the disputed question of the possibility of organic unities or disunities (see Chisholm 1986; Zimmerman 2001; Davison, ms.), if we adopt a simple additive view of the intrinsic value of a possible world, then there are good reasons to think that a world just like ours in which God exists is intrinsically better than a world just like ours in which God does not exist. This is the limited version of pro-theism that I wanted to defend in this paper. Finally, the argument from this section also has the following interesting consequence: traditional theists have wavered concerning whether there exists anything that is intrinsically valuable in the world. If this argument is sound, though, there is no room for doubt—given the way I have characterized the notion of intrinsic value, traditional theists are committed to the existence of intrinsic value, at least in the person of God. So those traditional theists who wish to deny the existence of intrinsic value altogether owe us an alternative characterization of the notion of intrinsic value, at the very least. In addition, if my account of intrinsic value is correct, and I am also correct in thinking that everything that exists is intrinsically valuable to some degree, it follows that our world contains a great deal of intrinsic value, because it is very big and very old. In fact, it may contain so much intrinsic value that the evil things in our world are overcome in some sense by the value of the whole world, and that even suffering and death are overcome by the intrinsic value of the creatures that experience them. (For more on this, see the discussion in Davison 2012, Chapter 7, and Davison, ms.)

Acknowledgements Thanks to participants at the Axiology of Theism Conference (held at Ryerson University, Toronto, Canada, on September 11–12, 2015) for helpful discussion concerning the questions raised in this paper, and to Klaas Kraay for many helpful comments concerning earlier drafts of this paper.

God and Intrinsic Value 45

References Adams, Robert. 1999. Finite and Infinite Goods. Oxford: Oxford University Press. Bradley, Ben. 2006. “Two Concepts of Intrinsic Value.” Ethical Theory and Moral Practice 9: 111–130. Broad, C. D. 1930. Five Types of Ethical Theory. New York, NY: Harcourt Brace. Chisholm, Roderick. 1986. Brentano and Intrinsic Value. New York, NY: Cambridge University Press. Davison, Scott A. 2012. On the Intrinsic Value of Everything. New York, NY: Continuum Press, 2012. Davison, Scott A. “A Naturalistic Intrinsic Value Theodicy.” Unpublished manuscript presented at the Rutgers Conference on Value Theory and Philosophy of Religion, Rutgers University, New Jersey, August 10, 2015. Kraay, Klaas J. and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 57–178. Lemos, Noah. 1994. Intrinsic Value. Cambridge: Cambridge University Press. Moore, G. E. 1912. Ethics. New York, NY: Henry Holt and Company. Nozick, Robert. 1981. Philosophical Explanations. Cambridge: Harvard University Press. Scanlon, Thomas 1998. What We Owe to Each Other. Cambridge, MA: Harvard University Press. Tännsjö, Törbjorn. 1999. “A Concrete View of Intrinsic Value.” Journal of Value Inquiry 33: 531–536. Zimmerman, Michael J. 2001. The Nature of Intrinsic Value. Lanham, MA: Rowman Littlefield Publishers.

3 Axiology: Theism Versus Widely Accepted Monotheisms Michael Tooley

1. Overview The structure of this paper is as follows. First, I start off by briefly explaining the concepts of pro-theism and anti-theism, and by distinguishing both between personal and impersonal versions of those views, and also between a more modest and a less modest claim connected with the impersonal version of pro-theism. I then introduce a distinction that is itself quite trivial, namely, that between pro-theism (and anti-theism), on the one hand, and pro-monotheism (and anti-monotheism), on the other, where the latter are defined in terms of the deity of some particular religion. The triviality notwithstanding, I shall argue at various points in the subsequent discussion that even the best discussions of the pro-theism versus anti-theism issue are vitiated to some extent by a failure to keep that distinction clearly in mind. Next, I consider how the pro-theism versus anti-theism question is related to the theism versus atheism question. These questions are often thought to be quite independent of each other, but I shall argue that that is not so. In particular, I shall argue that these two issues are neither logically nor evidentially independent of one another. The rest of the essay is then devoted to discussing the strength of the case for pro-theism, and there my basic claims are as follows. First, the case for the impersonal pro-theistic thesis—the thesis that, all things considered, the existence of God is better than the non-existence of God—is very strong relative to facts about which there is no dispute. Second, and contrary to the views of many philosophers, there is no respect in which the world is better if God does not exist than if God exists. Third, the case for the personal pro-theistic view that it is better for every person if God exists than it is if God does not exist is also very strong. Fourth, however, it is not clear that these pro-theistic conclusions remain justified if one thinks that one can take one’s evidence as including not just facts about which there is no dispute, but also the proposition that God exists, and I shall argue that if one does assume that theism is true, the support

Axiology: Theism vs. Accepted Monotheisms 47 that can otherwise be offered for some pro-theistic claims is seriously weakened, perhaps even to the point where those claims can no longer be sustained. My general conclusion, accordingly, might be put as follows: while there is no significant anti-theistic thesis that can be sustained if one is an atheist, some pro-theistic theses may become problematic if one is a theist.

2. Pro-Theism and Anti-Theism 2.1. Pro-Theism and Anti-Theism: Impersonal Versus Personal Versions Pro-theism and anti-theism are answers to evaluative questions, one of which is as follows: Are things as a whole better if God exists than if God does not exist? So on one interpretation of the relevant terms, one can take pro-theism to be the view that things as a whole are better if God exists, and anti-theism as the view that things as a whole are better if God does not exist. It might be suggested, however, that there is another important evaluative question in this general area, namely this: If things as a whole are better if God exists than if God does not exist, is it also true that a world in which God exists is a good world, that is, a better world than, say, sheer nothingness? One might, then, define pro-theism as the view that the right answers to both questions are affirmative, and anti-theism as the denial of protheism. Moreover, it seems to me that the case for pro-theism remains very strong under this more demanding interpretation. In the discussion that follows, however, I shall not be addressing the second question, so I shall only be defending the more modest version of pro-theism. In any case, pro-theism and anti-theism, understood in either of these ways, are theses about how the value of reality as a whole depends upon the existence or non-existence of God. It is therefore natural to label these views as impersonal pro-theism and impersonal anti-theism respectively. In addition to questions about the value of reality as a whole, however, one can also ask about how good a given life is for the individual whose life it is. One can then speak of personal pro-theism and personal antitheism, where the former is the view that it is better for every person if God exists than if God does not exist, while the latter is the view that, for at least some people, their lives are better if God does not exist. Thus defined, I think that many people will initially view impersonal pro-theism as immensely plausible, but may very well be strongly inclined to reject personal pro-theism, the thought being, on the one hand, that it is hard to see how things on the whole could possibly be worse if there were an omnipotent, omniscient, and morally perfectly good deity than

48  Michael Tooley if there were no such being, whereas, on the other hand, given that the world contains some serious evildoers indeed, their lives, surely, will not be as good for them if God exists, since then their evildoing will not go unpunished, and thus that it is not true that everyone will enjoy a better life if God exists. Nevertheless, this rather natural line of thought notwithstanding, my goal, as I have indicated, is to defend both impersonal pro-theism and personal pro-theism. 2.2. Pro-Theism and Anti-Theism Versus Theism and Atheism The issue of pro-theism versus anti-theism is thus rather different from that of theism versus atheism, and virtually everyone who has written in this area appears to hold that the different answers that can be given to the question of whether impersonal pro-theism, or impersonal antitheism, is true are not only logically independent but also evidentially independent of the answers to the question of whether theism is true, or atheism, so that a theist can be either an impersonal pro-theist or an impersonal anti-theist—although almost all theists are, I think, impersonal pro-theists—and similarly, an atheist can be either an impersonal pro-theist or an impersonal anti-theist. Shortly, however, I shall offer an argument for the thesis that if impersonal anti-theism is true, then atheism is true, so that the issues of pro-theism versus anti-theism and of theism versus atheism are not logically independent. Then, in the main section of my paper, I shall offer further reasons for also holding that they are not evidentially independent. 2.3. The Question of the Comparison I formulated one relevant evaluative question as follows: Are things as a whole better if God exists than if God does not exist? In answering that question, then, one is comparing some world in which God exists to some world in which God does not exist. But what are the relevant worlds? On some ethical views, an omnipotent and morally good person is morally obligated to create one of the best of all possible worlds, if such there be. In that case, how good a world will be with God in it is fixed and settled. (It may be, of course, that for every world there is a better one, or it may be that a morally good God need only create a world that is sufficiently good. If so, things are more complicated.) But what is the relevant Godless world? That this is a tricky question is apparent when one notices that among the Godless worlds are not only purely physical worlds, but also worlds of the sort that the Buddha believed in, where there are immaterial souls, along with reincarnation and karma, so that good deeds and evil deeds affect how good one’s state is in the next life.

Axiology: Theism vs. Accepted Monotheisms 49 One way of handling this question, if one is a theist, is to ask how this world would be if God did not exist, and to ask whether that world would be better than the actual world. But if one is an atheist, that way of putting things will not be satisfactory: the question will instead be, first, how the world would be if God existed, and, second, whether that world would be better than the actual world. Another complication is that some theists believe that the existence of God is necessary, in which case there are no possible worlds where God does not exist, which would mean that one is comparing the actual world with an impossible world. I shall ignore this complication, however, in part because I think that it can be demonstrated that the existence of a necessarily existent being that is capable of entering into causal relations is logically impossible. I am inclined to think that the question that is most worth pursuing here starts from the assumption that God does not exist, and then asks whether, if God did exist, the world would be better than the actual world. This is in part because I think there is an inductively sound version of the argument from evil (Tooley 2008, 70–150), and thus that it is reasonable to believe that the actual world is a Godless world. In addition, however, and as I mentioned earlier, I think that impersonal anti-theism entails that atheism is true, so that if God does exist, then impersonal anti-theism is false. My main focus, accordingly, will be on the question of whether a world containing God would be better than the actual world. But I shall also consider how things stand if one is a theist. 2.4. A World that Is Bad in Some Respects? Impersonal pro-theism is the proposition that the world as a whole is better if God exists than if God does not exist. But one might agree with that proposition while also affirming the following proposition: The ‘Worse in Some Respect’ Thesis: There is at least one good-making property, P, such that a Godless world is better with respect to instances of property P than a world containing God. Indeed, defenders of impersonal pro-theism, such as Guy Kahane (2011), appear to be inclined to accept this thesis. I shall be arguing, however, that this thesis is unsound. If one can defend the view that things as a whole are better if God exists than if he does not, does it matter whether there are some respects in which things are worse? The answer is that one way in which it may matter arises when one comes to consider personal pro-theism—the thesis that every individual has a better life if God exists than if God does not—for if there are some respects in which the world may very well be

50  Michael Tooley worse if God exists than if God does not exist, that may enable one to develop a convincing argument against personal pro-theism, as Kahane’s discussion (2011, 681ff.) illustrates. So there seems to me good reason to think very carefully about the thesis that the world is better in some respects if God does not exist, and I shall argue that the “worse in some respect” thesis is false.

3. The Crucial Distinction There is a distinction that I think it is absolutely crucial to keep in the forefront of one’s thinking: it is that between pro-theism and pro-monotheismof-type-X, where, for reasons that will emerge later, I take theism as the view that there is a morally perfectly good and omnipotent—or at least very powerful—creator of the physical universe who exists at all times, and where pro-monotheism-of-type-X is the view that the world is better if a deity of the sort involved in religion X exists than if such a deity does not exist. Why is it crucial to draw that distinction? First of all, for almost any monotheistic religion that comes to mind, a person who is not religious, and who actually focuses on religion X, and thinks about that religion, is very likely to accept pro-theism, but to reject pro-monotheism-of-type-X. Second, I believe that the vast majority of existing discussions of protheism and anti-theism suffer from a failure to make this distinction. This failure is most likely to occur if one is oneself some sort of monotheist, broadly understood, since then the temptation to identify the god that one believes in with the god of theism, as defined above, is likely to be virtually irresistible. But even highly intelligent and very good philosophers who are atheists, in thinking about the pro-theism versus anti-theism issue, seem prone to make the same mistake, for although an atheist is very likely to think, for example, that both the god of the Old Testament, and also the god of the New Testament, would be, if they existed, not only less than morally perfect deities, but profoundly evil ones, atheists can easily make more subtle errors, attributing properties to a morally perfect deity that there is, upon reflection, good reason not to attribute. (One of the more subtle errors here, as I shall argue later, is thinking that a deity than which no greater can even be conceived would be omniscient.)

4. Are the Pro-Theism Versus Anti-Theism Issue and the Theism Versus Atheism Issue Logically Unrelated? 4.1.  Overview: A Common Assumption The view that seems almost universally accepted by philosophers writing in this area is that there are no logical connections between, on the one

Axiology: Theism vs. Accepted Monotheisms 51 hand, answers to the pro-theism versus anti-theism question and, on the other, answers to the question of whether God exists. Thus Guy Kahane, for example, in his very thoughtful and interesting article, says: But anti-theism makes no claim about religion or about the value of belief. More importantly, anti-theism and atheism are independent claims. Nagel is not a theist. But a theist could be an anti-theist, although this combination of views is obviously not common. Antitheism is silent about the question of God’s existence. (2011, 679) Now it is true both that anti-theism does not explicitly say anything about the existence of God and also that a theist could be an anti-theist. But it is not true that anti-theism and atheism are logically independent claims. Why not? Because anti-theism logically entails atheism, as can be seen as follows. 4.2. Mini-Theorem 1: Anti-Theism Entails Atheism One way of setting out my first mini-theorem would involve the following premise: (*) An omnipotent, omniscient, and morally good being must always perform the morally best action. But this proposition is open to at least two rather plausible objections. One is that a being can be good without always performing the morally best action: it is enough if the being performs an action that is morally sufficiently good. The other is that it is possible that in some situations there is no morally best action: there are just actions that are morally better and better without limit. To deal with the first objection, I shall use the expression “morally perfectly good” so that only a being that always performs the morally best action whenever such an action exists qualifies as morally perfectly good. Since a being that is not morally perfectly good is not a being than which no greater can be conceived, it is surely reasonable to say that God must be a morally perfectly good being. The second objection can be addressed by revising (*) as follows. Some possible actions are such that they belong to a series of possible actions that differ only with respect to some quantity. Suppose, for example, that creating a million very happy people is a morally good action. Then that action belongs to an infinite and unbounded series of actions, all of which have the form: creating N very happy people. When both an action A and some better alternative B belong to an infinite and unbounded series of possible actions that differ only quantitatively in the way just described,

52  Michael Tooley performing action A instead of B does not entail that one was not morally perfectly good. In contrast, if A and some better alternative B do not belong to such an infinite and unbounded series, then performing action A instead of B does entail that one was not morally perfectly good. So we have, in short, the following principle: (**)  If a being performs action A rather than some morally better action B, when A and B do not belong to an infinite and unbounded series of actions that differ only quantitatively along some dimension, that that being is not morally perfectly good. Given that principle, the mini-theorem that anti-theism entails atheism can be proved as follows: 1. God is by definition an omnipotent, omniscient, and morally perfectly good being who exists at every time. 2. It is logically impossible for there to be logically necessary connections between temporally distinct states of affairs. Therefore, from 1 and 2: 3. If an omnipotent, omniscient, and morally perfectly good being existed at any time earlier than now, that being had the power to bring it about that no omnipotent, omniscient, and morally perfectly good being exists now, and that being knew that it had that power. From 1: 4. If no omnipotent, omniscient, and morally perfectly good being exists now, then God does not exist. Therefore, from 3 and 4: 5. If an omnipotent, omniscient, and morally perfectly good being existed at any time earlier than now, it had the power to bring it about that God does not exist, and knew that it had that power. Assume, as a premise in a conditional proof: 6. Anti-theism is true: The world would be better if God did not exist. Therefore, from 6: 7. It is better to bring about a world where God does not exist than to refrain from doing so.

Axiology: Theism vs. Accepted Monotheisms 53   8. If A is the action of bringing about a world where God does exist at all future times, and B is the action of bringing it about that God does not exist at all future times, then actions A and B do not belong to an infinite and unbounded series of actions that differ only quantitatively. Next, let us introduce (**) as a premise:   9. If a being performs action A rather than some morally better action B, when A and B do not belong to an infinite and unbounded series of actions that differ only quantitatively along some dimension, then that being is not morally perfectly good. Therefore, from 7, 8, and 9: 10. If a being brings about a world where God exists at all later times, rather than bringing it about that God does not exist at some future time, then that being is not morally perfectly good. Therefore, from 5 and 10: 11. If an omnipotent, omniscient, and morally perfectly good being exists at any time, that being will bring it about that at some future time God does not exist. In view of 1: 12. If there is some time at which God does not exist, then God does not exist. Therefore, from 11 and 12: 13. If an omnipotent, omniscient, and morally perfectly good being exists at any time, then God does not exist. In view of 1: 14. If an omnipotent, omniscient, and morally perfectly good being does not exist at any time, then God does not exist. Therefore, from 13 and 14: 15. God does not exist. Therefore, from 6 and 15, via conditional proof: 16. If anti-theism is true, then God does not exist.

54  Michael Tooley 4.3. An Objection: The Existence of God Is Logically Necessary There is a familiar idea that, if sound, would be a devastating objection to the preceding argument. It is that the idea of a being greater than which no being can even be conceived entails that such a being would be a logically necessary being, and a being all of whose basic powers would also be logically necessary. My response to this objection is that both of these supposed attributes of a being than which no greater can even be conceived are logically impossible attributes. First, as regards the idea that such also a being would be logically necessary, I would argue as follows: 1. A being than which no greater can even be conceived would be omnipotent. 2. An omnipotent being can enter into causal relationships. Therefore, from 1 and 2: 3. A being than which no greater can even be conceived would be able to enter into causal relationships. 4. No contradiction can be derived from the proposition that the only things that exist that can enter into causal relationships are a number of neutrinos in an infinite Euclidean spacetime. 5. If no contradiction can be derived from a proposition, then it is logically possible for the proposition to be true. Hence, from 4 and 5: 6. It is logically possible that there does not exist anything that can enter into causal relationships beyond a number of neutrinos in an infinite Euclidean spacetime. 7. A world that contains nothing that can enter into causal relationships beyond a number of neutrinos in an infinite Euclidean spacetime does not contain any omnipotent being. Therefore, from 6 and 7: 8. It is logically possible that there is no omnipotent being. Therefore, from 1 and 8: 9. It is logically possible that there does not exist a being than which no greater can even be conceived.

Axiology: Theism vs. Accepted Monotheisms 55 What about the claim that the properties of a being than which no greater can even be conceived would all be logically necessary properties, so that if something existed with the relevant properties at any time, it would have to have those properties at any later time? Here my basic argument would be as follows: 1. A being than which no greater can even be conceived would be omnipotent. 2. Omnipotence is an intrinsic property of an entity at a time. 3. When a property is an intrinsic property of an object at a time, the object in question can possess that property regardless of what things exist at other times. Therefore, from 2 and 3: 4. An object’s possessing the property of omnipotence at one time cannot entail that it possesses that property at any other time. Therefore, from 4: 5. Omnipotence cannot be a logically necessary property of any entity. 6. If something is not a logically necessary property of any entity, an omnipotent being acting at time t could bring it about that no entity possessed that property at any time later than t. Therefore, from 5 and 6: 7. An omnipotent being could bring it about that it was not itself omnipotent at any later time.

5. The Case for Impersonal Pro-Theism 5.1.  An Overview How should one develop a case for pro-theism? The usual way in which philosophers have proceeded, I think, is as follows. First, one cites various properties that are plausibly viewed as good-making properties and argues that the world will have those properties if God exists, but not if God does not exist. Second, one considers arguments that might be offered for anti-theism, in which other properties are cited that are also claimed to be good-making, and which it is argued will be present if and only God does not exist. At that point, there are two ways in which defenders of pro-theism typically proceed. One is to argue that the properties in question are not in fact good-making properties, so that

56  Michael Tooley their absence does not make the world worse than it would be if those properties were present. The other way is to grant that the properties in question are indeed good-making, but to argue that they are less significant than those good-making properties that, arguably, will be present if God exists, and not otherwise. So while, on this latter view, there will be respects in which the world will be worse if God exists, all things considered the world as a whole will be better if God exists. As regards these two pro-theistic ways of responding to considerations that are advanced in support of anti-theism, my view is that the first is unsound, while the second is plausible. But there is a third response, which seems not to be made, and which I think is clearly stronger: it involves arguing that the good-making properties that are cited in support of anti-theism will be present in a world where God exists, and that the failure to see that that is so arises out of a failure to keep in the forefront of one’s thinking the crucial distinction that I mentioned earlier. The case that I shall set out for pro-theism differs from standard cases in three other important, and related respects. First, I shall make a point of asking whether the case for pro-theism is equally strong if one adds to what one takes to be the relevant evidence either the assumption that God exists or the assumption that God does not exist. Once this question is asked, the answer turns out to be clear: the case for pro-theism is not only stronger, but much stronger if one assumes that God does not exist. Second, if the case for pro-theism is weaker if one assumes that theism is true, how much weaker is it? Does one still have a good case for protheism? I shall offer a reason for thinking that this may not be so. Finally, how do things stand if one confines oneself to facts about which there is no dispute in making out one’s case? In view of the preceding two points, the answer depends on the strength of arguments for and against the existence of an omnipotent and morally perfectly good deity. If, as some have claimed, there is a sound version of the evidential argument from evil, then the case for pro-theism is very strong, based simply upon facts about which there is no dispute. 5.2. Bad Arguments for Pro-Theism Before proceeding to set out the case for impersonal pro-theism, I want simply to mention some arguments that are sometimes offered for protheism that I think are bad arguments. The ones that I have in mind here involve the following claims: (1) If God didn’t exist, there would be no objective values. (2) If God didn’t exist, there would be no moral obligations. (3) If God didn’t exist, life would have no objective meaning.

Axiology: Theism vs. Accepted Monotheisms 57 My reasons for rejecting the first two claims are a matter of familiar arguments in meta-ethics, going back to Plato’s famous argument in the Euthyphro, along with the following argument: 1. Basic normative statements are necessarily true. 2. It is not necessarily true that God exists. Therefore, 3. Basic normative statements would be true even if God did not exist. In the case of the third argument, I shall return to it, in effect, when I defend personal pro-theism. 5.3. Religiously Neutral Arguments for Impersonal Pro-Theism Let me begin by briefly describing some arguments in support of impersonal pro-theism that are religiously neutral in the following way: the features of the world that it is claimed would make the world a better place would generally be viewed as doing so by both theists and atheists, and it would also be generally agreed that those properties are much more likely to be present if God exists than if God does not exist. Consideration 1: Persons Living Forever Is a Good-Making Feature of a World First of all, if God does not exist, it is unlikely that human persons will live forever, since it is unlikely that they will survive death. It is of course perfectly possible for humans to live forever even if there is no omnipotent, omniscient, and morally perfectly good person. There could, for example, be less powerful beings that bring it about that human persons survive death. Or it could be that humans have immaterial minds or souls that survive the destruction of their bodies, perhaps to be reincarnated, as in Hinduism or Buddhism. But I would claim, first of all, that we have no evidence for the existence of any such powerful beings who might have an interest in our surviving death, and second, that we have very strong evidence—as I have argued elsewhere (Tooley et al. 2009, 17–19)—that humans do not have immaterial minds or souls, so that such alternative scenarios are very unlikely. Would it be a good thing if humans survived death? The answer to that clearly depends on what life would be like after one’s earthly death. But clearly, if life after death were no worse than life before death, most people would want to survive, especially if one could maintain the

58  Michael Tooley relationships that one has with those whom one cares about, and if one could either continue on with one’s present activities and projects, or else move on to other ones that were at least as satisfying. In addition, on standard views concerning how good a life is for a person, a longer life is, other things being equal, a better life, and if survival of death would thus be better for individuals, it would be surprising if the world as a whole, accordingly, were not better. Finally, although the idea is rejected by some religious believers, many philosophers believe that persons have a right to continued existence, in which case it would surely be good if that right were not violated by earthly death. Consideration 2: Justice Will Be Achieved, at Least in the End Second, if God does not exist, then, as regards life here on this earth, there will be many people who will have suffered injustices that will never be rectified—and in some cases absolutely horrendous injustices, such as the suffering and deaths of millions of people in the Holocaust. In addition, some morally very bad people have had very happy lives, often because they profited by inflicting undeserved sufferings upon others. A world with God, however, will be a world where justice is ultimately done, and a world where, at least in the end, good triumphs over evil. Surely this is an immensely good feature of a world. In short, then, there are at least the following two features that it is reasonable to expect will be present in a world in which God exists, that it is also reasonable to believe would contribute greatly to the goodness of a world, and that there is no good reason to suppose would be present in a world in which God does not exist. First, personal existence, rather than ending with death, will potentially continue forever; second, the world will be a just world, and one where goodness triumphs over evil, at least in the end. 5.4. Atheism as Strengthening the Case for Impersonal Pro-Theism Suppose that God does not exist. Why should that make a difference to the strength of the argument in support of impersonal pro-theism? The answer is that the actual world contains a number of states of affairs whose known bad-making properties far outweigh any known goodmaking properties of those states of affairs. As I have argued elsewhere (Tooley 2008), an inductive argument enables one to show, of each such state of affairs, both that it is more likely that that state of affairs is, given all of its morally significant properties, both known and unknown, an undesirable state of affairs, and also that its expected value, given all of its morally significant properties, both known and unknown, is equal to

Axiology: Theism vs. Accepted Monotheisms 59 the value that it has given only its known morally significant properties. There is good reason for believing, then, that the world that would exist if there were an omnipotent, omniscient, and morally perfectly good person would at the very least lack much of the suffering that one finds in this world due to diseases, natural disasters, the existence of carnivorous animals, and so on. But it is not just natural evils that it is reasonable to expect would be dramatically reduced: the same is true surely of moral evils. Suppose that one could have prevented the sufferings and deaths due to people such as Stalin and Hitler. Would one have refrained from doing so, and would it have been reasonable for one to have done that—on the grounds, say, that refraining from intervening might have some great, unknown rightmaking property that made it best, all things considered, not to intervene? Surely the answer is, first of all, that one would have acted to prevent Stalin and Hitler from performing the acts in question. Similarly, if one could prevent the suffering that children undergo at the hands of adults, would not one choose to do that? If so, then one must believe that it is unlikely that there are any unknown right-making properties that, if they were known to one, would justify one in not intervening to prevent those sufferings and deaths. But if it is reasonable for you to believe that there are no such counterbalancing right-making properties—and if the type of inductive argument from evil just mentioned is sound, this is indeed reasonable—then it must also be reasonable to believe that if there were an omnipotent, omniscient, and morally perfectly good person, that person would also act to prevent such moral evils. In short, an atheist can, unlike a theist, point to the enormous suffering and deaths involved in either natural evils or moral evils, or both, and argue that, at the very least, it is likely that most such evils would not be present in a world in which God existed. In addition, consider again the religiously neutral considerations in support of impersonal pro-theism mentioned earlier, and notice that if one asks about the details of those two considerations, an atheist can offer a picture that is arguably better than what the theist can offer. The first consideration was that it is a very good thing if persons, rather than ceasing to exist at some point, potentially live forever. A theist is forced here to think in terms of a person’s undergoing bodily death, and then either surviving because they have immaterial minds or souls, or else being resurrected. An atheist, on the other hand, can refer to the possibility of persons being supermen and superwomen who, being made of indestructible material in a world without kryptonite, can never be killed when they want to go on living. On the latter conception, unlike the former, one is never separated, even temporarily, from those that one loves. Surely this is better than what one has on the theist scenario.

60  Michael Tooley 5.5. A Religious Consideration: Atheism as Again Strengthening the Case for Pro-Theism If theism is true, the existence of God is, to put it mildly, rather less evident than it could be. As J.L. Schellenberg (1993; 2015) has eloquently, and in my view cogently, argued, this hiddenness of God is a very strong argument against the existence of God. In response, theists generally attempt to argue, contrary to what Schellenberg and others have claimed, either that this hiddenness is not in a fact a bad feature of the world, or else that God is not really hidden, since his existence is clearly evident to all who have not perversely closed their minds and hearts to his existence. What is true, in any case, is that it is easy to imagine a world where God’s existence would be completely evident, and the atheist can proceed to argue that it would be a good thing if people could know of God’s existence, and thus interact with him, just as they can in the case of other persons. One has, then, another good-making feature that the atheist can argue would be present in the world if God existed, and otherwise absent, and thus one has additional support for pro-theism. 5.6. Theism as Weakening the Case for Impersonal Pro-Theism We have just seen that if atheism is true, there are more states of affairs that one can appeal to in support of pro-theism, since one can appeal to any apparent evil that the world contains, and argue that it is likely that, if there were an omnipotent, omniscient, and morally perfect person, that apparent evil would not exist. So the case supporting pro-theism is weaker given the assumption that theism is true rather than given the assumption that atheism is true. But the assumption that theism is true weakens the case in another, albeit related way, as follows. If one is a theist, confronted with states of affairs that are such that when one considers the known right-making and wrong-making properties possessed by an action of not preventing the existence of those states of affairs, the known wrong-making properties outweigh the known right-making properties, one is forced to maintain that those states of affairs not only possess some unknown right-making properties, but right-making properties of such a magnitude that the total weight of the right-making properties, known and unknown, is greater than the total weight of the wrong-making properties, known and unknown. But if one claims that an extraordinary number of apparent evils are actually not evils, all things considered, how can one be justified in claiming that things that it appears would be good actually would be good, rather than bad? If an omnipotent being’s intervening to eliminate diseases, to stop natural disasters, and to prevent the terribly evil actions of

Axiology: Theism vs. Accepted Monotheisms 61 people such as Stalin and Hitler would have been morally wrong, how can one be justified in holding, for example, that an omnipotent being should not allow death to be the end of human existence? In short, once one contends, as a theist, that an enormous number of apparent evils are actually not evils, all things considered, how can one answer the antitheist who parallels the sceptical theist’s contention in order to support the conclusion that we have no way of knowing that such things as survival of death are good all things considered, and thus no way of knowing that such things will exist if God exists? Can a theist’s case for pro-theism survive this objection? If one is prepared to offer a theodicy, perhaps, but not, I think, if one is a sceptical theist. However, regardless of whether that is so, the basic point here is the theist’s case for pro-theism is clearly weakened in a way that an atheist’s case for pro-theism is not. 5.7. How One’s Religious Views Affect the Pro-Theistic Arguments that One Can Offer It is now clear that one’s religious beliefs can affect how one argues for pro-theism. On the one hand, if one is an atheist, a natural argument for pro-theism is that if God existed, the present and past lives of people here on this earth would contain or have contained much less suffering than they do, and many more pleasurable states. A theist, on the other hand, cannot argue in that way. The goods that a theist can cite must either be goods that lie in the future, or else goods that supposedly exist now. The latter include such things as libertarian freewill, or assistance in achieving spiritual goals, where, while there may well be no evidence for the existence of those goods, perhaps at least there is not any clear-cut and decisive evidence against their existence.

6. Objections to Pro-Theism? How could a world be worse if God existed? It would have to be the case that there was some state of affairs S with the following characteristics: (1) S is a bad state of affairs; (2) The badness of S outweighs the goodness of those states of affairs that exist because of God’s existence; (3) S is a state of affairs whose existence is entailed by the existence of God. Accordingly, when a consideration is offered in support of anti-theism, the pro-theist has three ways of replying. The first is to challenge the claim that the purportedly bad state of affairs S is in fact bad. The second is to argue that God’s existence does not entail the existence of S. The third is to argue that although S is bad, and its existence is entailed by God’s existence, the badness of S is not sufficient to outweigh the goods that result from God’s existence, goods such as immortality, justice being done, and good triumphing over evil.

62  Michael Tooley Consider, for example, the claim that immortality is undesirable. If a person advancing this claim has in mind the Christian conception of the afterlife, one might argue that it is not bad that sinners are justly punished by eternal torment in hell. Alternatively, one could argue that while it is undesirable that some people suffer eternally in hell, the undesirability of that is outweighed by the goods that God’s existence makes possible, such as people enjoying eternal happiness in heaven. Or third, if neither of those lines of thought seems appealing, one can argue that the existence of God does not entail the existence of that kind of afterlife. But some people object to immortality on quite different grounds. One of the most striking is Walter Kaufmann, who in a chapter entitled “Death” in his book The Faith of a Heretic, quotes a famous passage from Shakespeare’s Tempest that concludes We are such stuff As dreams are made on, and our little life Is rounded with a sleep. He then goes on to say: It is possible that this is wrong. There may be surprises in store for us, however improbable it seems and however little evidence suggests it. But I do not hope for that. Let people who do not know what to do with themselves in this life, but fritter away their time reading magazines and watching television, hope for eternal life. If one lives intensely, the time comes when sleep seems bliss. If one loves intensely, the time comes when death seems bliss. . . . The life I want is a life I could not endure in eternity. It is a life of love and intensity, suffering and creation, that makes life worthwhile and death welcome. There is no other life I should prefer. Neither should I like not to die. . . . As one deserves a good night’s sleep, one also deserves to die. Why should I hope to wake again? To do what I have not done in the time I’ve had? All of us have so much more time than we use well. How many hours in a life are spent in a way of which one might be proud, looking back? For most of us death does not come soon enough. Lives are spoiled and made rotten by the sense that death is distant and irrelevant. One lives better when one expects to die, say, at forty, when one says to oneself long before one is twenty: whatever I may be able to accomplish, I should be able to do by then; and what I have not done by then, I am not likely to do ever. One cannot count on living

Axiology: Theism vs. Accepted Monotheisms 63 until one is forty—or thirty—but it makes for a better life if one has a rendezvous with death. Not only love can be deepened and made more intense and impassioned by the expectation of impending death; all of life is enriched by it. (1963, 372–373) Once again, one might either challenge the evaluative claims that Kaufmann is making, or argue instead that the goods that God’s existence makes possible far outweigh any disvalue associated with immortality. But both of those responses have features that are somewhat undesirable. As regards the second, it means that one is conceding the more modest anti-theist thesis, namely, that there are respects in which God’s existence is undesirable. As regards the first, it leaves it open for one to argue against personal pro-theism, since one can argue that the existence of God is not good for everyone, since immortality means that someone like Kaufmann would be unable to live the type of life that he deems best, so that the existence of God would not be in his best interests. For this reason, the third response mentioned above is clearly the most desirable, and it is the response that Guy Kahane (2011) makes: If immortality is indeed undesirable, then the possible world described by some theists might also be undesirable. But if it is constitutive of God’s goodness that he is omnibenevolent and supremely good, then this could not be a genuine Godly world. So this is the wrong way to argue for anti-theism. (681) Since immortality seems so obviously desirable to most people, this response may seem unacceptable. But notice that this may be because one is thinking not in terms of theism, but in terms of the views of some monotheistic religion that one accepts, and here I am not thinking of the idea of heaven and hell. What I am thinking of is that immortality in familiar monotheistic religions is not an option: it is compulsory. But if one keeps in mind that God is defined as perfectly good, why should that being not allow people to opt out of immortality if, having considered arguments on the other side, their preferred life, as in the case of Walter Kaufmann, is a life where one decides for a time limit on one’s life, where one has “a rendezvous with death”? But while this type of response is open to one in the case of immortality, is it always available when one is confronted with an argument in support of anti-theism? May there not be things that are intrinsically valuable—such as privacy, for example—that are incompatible with the

64  Michael Tooley existence of God? If so, then one is faced with the options of arguing either that the thing in question is not intrinsically desirable, or else that it is, but that it is outweighed by the intrinsically desirable things that will exist if and only if God exists. The former response seems not at all plausible in some cases, while the latter response concedes the modest anti-theist claim, and opens the door for an argument against personal pro-theism. It is preferable, then, to argue, if one can, that there is nothing intrinsically desirable that is incompatible with the existence of God, and it seems to me that this can be done.

7. In Defence of Personal Pro-Theism? Kahane develops an argument in support of personal anti-theism. I believe that Kahane’s argument is unsound. The starting point for Kahane’s argument is the idea that there may be a sound argument for the modest impersonal anti-theist view that if God exists then the world will contain some features that are intrinsically undesirable: I don’t know why Nagel is an anti-theist—to what he is referring when he writes that he doesn’t want the universe to be like that. But I suspect Nagel has in mind something like the following. A world in which God exists is a world where human beings stand in a distinctive and inescapable relation to another person. It is a world where we are the subordinates of a moral superior, a superior that deserves our allegiance and worship, and where we have been created to play a part in some divine cosmic plan. It is a world where everything about us is known and fully understood by another, a world where even our innermost thoughts and feelings are not entirely private. It is a world in which we are never truly alone, away from the presence and attention of another. And if the true nature of God is beyond human comprehension, it would also be a world that we can never hope to fully understand. The idea is that God’s existence is logically incompatible with the full realization of certain values. Thus a world in which God exists is a world where we would not be the moral equals of all other rational beings—equal members of a kingdom of ends that has no ruler. Such a world seems incompatible with complete independence, or with complete privacy and genuine solitude. And it might also be a world where it would be pointless for us to strive for a complete and unqualified understanding of the universe. (2011, 682) But why should one accept the claim that if God exists, the world will contain such apparently undesirable features? Consider, for example, the claim that if God exists, it will be a world where “even our innermost

Axiology: Theism vs. Accepted Monotheisms 65 thoughts and feelings are not entirely private.” Kahane, I believe, thinks that that would be an undesirable feature of a world, and that seems to me right. For suppose that a machine was developed that enabled one to know what brain states another human was in and that could then tell one what that person’s thoughts and feelings were. Would it be morally acceptable for one to use that machine, without getting the other person’s permission? The answer, it seems to me, is that that would be prima facie wrong. But then it would also be prima facie wrong for God to read the minds of other people without their consent. Accordingly, should we not conclude that, in general, God would not do so? In reply, it will probably be said that God cannot avoid reading the minds of others, since omniscience is part of the very definition of the term “God”. My reply is that, if so, the definition is ill-conceived. Our starting point should be to use the term “God” to refer to a being than which no greater can even be conceived, and then we need to ask what properties such a being will have, and, specifically, whether such a being will be omniscient. The answer to this question is given by a second mini-theorem. Mini-Theorem 2: If God Exists, It Is Extremely Likely that God Is Not Omniscient 1. God is by definition a being than which no greater can even be conceived. 2. A being than which no greater can even be conceived will be morally perfectly good. 3. A being than which no greater can even be conceived will be omnipotent. 4. An omnipotent being can avoid performing actions that have any wrong-making properties. 5. A morally perfectly good being will avoid performing an action that has any wrong-making property if it can do so, unless the action also has one or more right-making properties that outweigh the wrongmaking property in question. Therefore, from 4 and 5: 6. A being that is both morally perfectly good and omnipotent will not perform any action that has a wrong-making property, unless the action also has one or more right-making properties that outweigh the wrong-making property in question. 7. All persons have a right to the privacy of their thoughts and feelings. Therefore, from 7: 8. Accessing the thoughts and feelings of another person without his or her consent is prima facie wrong.

66  Michael Tooley Therefore, from 6 and 8:  9. A being that is both morally perfectly good and omnipotent will never access the thoughts and feelings of another person without his or her consent, unless doing so has one or more right-making properties that outweigh the wrong-making property in question. Therefore, from 9: 10. If there is a being that is both morally perfectly good and omnipotent, and if there are also other persons who want to enjoy privacy of their thoughts and feelings, and who do not consent to those thoughts and feelings being accessed, then the morally perfectly good and omnipotent person will not access the thoughts and feelings of those other persons, unless doing so has one or more right-making properties that outweigh the wrong-making property in question. 11. Given the enormous number of people who exist at some time or other, and the extremely large number of thoughts and feelings that almost every person has, it is unlikely in the extreme that there is never a single thought or feeling where there is no right-making property, or combination of right-making properties, that outweigh the wrongness of accessing that thought or feeling without the consent of the person in question. 12. An omniscient person would have knowledge of all of the thoughts and feelings of another person. Therefore, from 11 and 12: 13. If there were an omniscient person, it is extremely likely that such a being would have acquired knowledge of someone’s thoughts or feelings without that person’s consent, and where the wrongness of thus violating that person’s privacy was not outweighed by any combination of right-making properties. 14. An omnipotent person could always avoid violating a person’s right to privacy when the wrongness of doing so was not outweighed by any combination of right-making properties. Therefore, from 13 and 14: 15. If there is an omnipotent and omniscient being, it is extremely likely that that being has done something morally wrong. Therefore, from 15: 16. If there is a being that is both omnipotent and morally perfectly good, it is extremely likely that that being is not omniscient.

Axiology: Theism vs. Accepted Monotheisms 67 Therefore, from 1, 2, 3 and 16: 17. If God exists, it is extremely likely that he is not omniscient. Kahane, in a footnote, considers, in effect, the sort of response that I have just offered, and says: It might be replied that, if these value claims are true, God would have to respond by restricting Himself in some way. But how, for example, could God be omniscient yet fail to know our inner thoughts and feelings? And if God is inscrutable to human understanding, could He really make himself simpler so that we would be able to understand Him? I take it that these suggestions are absurd. If they can be shown to be coherent, then perhaps God’s existence would be compatible with at least some of the values I listed. This, however, would require a truly radical revision of the traditional theist conception of God. (2011, 682, fn. 18) I do not think that there is anything very radical about the idea that God might limit his knowledge in certain ways. But regardless of whether it a radical revision, the point is that the concept of the God of theism should surely not diverge from that of a being a greater than which cannot even be conceived, and this means that unless one is prepared to deny that persons have a right to privacy of thoughts and feelings, one must conclude that an omnipotent and morally perfect good person, though he would have the capacity for omniscience, would not, in the absence of countervailing right-making characteristics, exercise that capacity when doing so would violate the rights of persons. This seems to me to be another case where it is easy to allow one’s thinking to be shaped by assumptions about what God would be like that are found in familiar monotheistic religions. Moreover, I think that the same thing occurs with regard to other features that Kahane assumes would be present in the world if God existed. For consider the following sentences from the passage I quoted earlier: A world in which God exists is a world where human beings stand in a distinctive and inescapable relation to another person. It is a world where we are the subordinates of a moral superior, a superior that deserves our allegiance and worship, and where we have been created to play a part in some divine cosmic plan. Think of it his way. Imagine a person whose power and moral character were comparable to our own. Now imagine that person gradually becoming morally better. Is there some point at which that being deserves to be worshipped, rather than merely deeply admired? I cannot see why

68  Michael Tooley this should be so. Nor does it seem to me that there is any more ground for thinking that worship is appropriate if we also imagine the being in question becoming more and more powerful, until he is eventually omnipotent. Or consider the idea that we ought to subordinate ourselves to such a being. Do not persons have a right to self-determination, a right to shape their own lives as they see fit, provided that they do not violate the rights of others? But if so, there would be no moral obligation to fit into some “divine cosmic plan.” Once again, then, it seems to me that many philosophers working in this area fail to think critically about the conception of God that they have, and are assuming, in effect, that the properties that are attributed to God in familiar monotheistic religions are properties that should be attributed to a being than which no greater can even be conceived. So they think of God as a being that wants to be worshipped, and that should be worshipped, and that has some divine plan that other persons should accept and fit into. What I am suggesting, then, is that the considerations that are advanced either in support of the modest anti-theist claim, or in support of personal anti-theism—such as that a world containing God will be a world where one’s inner life is no longer private, a world where one must bow down and worship some being, and a world where, rather than one’s right to self-determination being respected, one is expected to abandon one’s own goals in order to fit into some divine plan—are all based upon the mistaken view that various properties that characterise the gods of familiar monotheistic religions will be properties of God, of a being than which no greater can even be conceived. Jettison that mistaken idea, and it seems to me that all of the anti-theistic arguments disappear. The underlying point, in short, is this. If some way of acting is in some respect prima facie wrong, then that provides one with a good reason for concluding that an omnipotent and morally perfect being would not, in the absence of countervailing right-making properties, act in that way. Consequently, one should not attribute to a being than which no greater can even be conceived any property, such as omniscience, that entails that there is some type of action that is prima facie wrong and which is such that the being in question would never refrain from performing an action of that type. Summing Up The main theses for which I have argued in this paper are as follows: 1. The argument for impersonal pro-theism is very strong. 2. The argument is strongest if atheism is true. 3. In contrast, the argument is less strong if theism is true, since, first of all, the types of goods to which the theist can appeal are much more limited.

Axiology: Theism vs. Accepted Monotheisms 69 4. Second, the fact that the theist has to argue that many apparent evils are only apparent exposes the theist to the objection that the goods that it is claimed would exist if God exists may be only apparent goods, and thus might not exist. 5. Contrary to what many believe, there is also a strong case for the still stronger pro-theist thesis that there is no respect in which the world would be better if God did not exist. 6. Because of this, there is also a strong case for personal pro-theism. 7. In thinking about all of these issues, one has to avoid incorporating into one’s conception of God properties that are often part of the conception of the deity in familiar monotheistic religions, but that, upon reflection, would not be properties of a being than which no greater can even be conceived.

References Kahane, Guy. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696. Kaufmann, Walter. 1961. The Faith of a Heretic. Garden City, New York, NY: Doubleday and Company. [Published by Doubleday and Company as an Anchor Books paperback in 1963, from which the quotes are taken.] Schellenberg, J. L. 1993. Divine Hiddenness and Human Reason. Ithaca, New York, NY: Cornell University Press. Schellenberg, J. L. 2015. The Hiddenness Argument: Philosophy’s New Challenge to Belief in God. New York, NY: Oxford University Press. Tooley, Michael, and Plantinga, Alvin. 2008. Knowledge of God. Oxford: Blackwell. Tooley, Michael, Jaggar, Alison, Devine, Philip E., and Wolf-Devine, Celia. 2009. Abortion: Three Perspectives, edited by James Sterba. Oxford: Oxford University Press.

4 An Agreeable Answer to a Pro-Theism/Anti-Theism Question T.J. Mawson

In this chapter, I address some difficulties that stand in the way of reaching an answer that will be agreeable across the theist/atheist divide to a particular pro-theism/anti-theism question, the question I call ‘the’ comparative pro-theism/anti-theism question. Really, one might say, ‘the’ comparative pro-theism/anti-theism question is two questions, one for the theist and one for the atheist. For the theist: would God’s not existing have been better or worse than is His existing? For the atheist: would God’s existing have been better or worse than is His not existing? Assuming that theist and atheist alike should agree that the issue of whether or not there’s a God concerns a metaphysical necessity and thus that each should think of themselves as involved in counterpossible reasoning when addressing their variant of ‘the’ comparative pro-theism/anti-theism question, I shall argue that the difficulties they face in reaching an agreement on its answer, whilst not insignificant, can be overcome, at least to an extent. I shall tentatively suggest that an answer that I shall call ‘weak pro-theism’ emerges as one that is agreeable across the theist/atheist divide, agreeable as what I shall call the ‘lower epistemic bound’ to ‘the’ correct answer to ‘the’ comparative question. More specifically, the theist should answer his variant of the question by saying that it would have been somewhat worse (or at least no better) were God not to have existed. The atheist should answer his variant by saying that it would have been somewhat better (or at least no worse) were God to have existed. I shall argue that arguments for a stronger pro-theistic conclusion to this question largely wait on a resolution to the theism/atheism question.

Introduction Many philosophers have argued for theism, and many others have argued for atheism. Yet, whilst many have addressed this, which we may call the ‘theism/atheism’ question, few philosophers have addressed any of the family of closely related (but logically distinct) questions which we may call the ‘pro-theism/anti-theism’ questions.1 I wish to address one of

Agreeable Answer to Pro-/Anti-Theism Question 71 these questions now, the one that I shall call the comparative pro-theism/ anti-theism question. The comparative pro-theism/anti-theism question is difficult to state in a way that will be acceptable to both theist and atheist alike, as it is about the difference in value between the actual world and the nearest world in which ‘God exists’ has the opposite truth value to that which it has in the actual world; and theists and atheists obviously disagree about what truth value ‘God exists’ has in the actual world, and only slightly less obviously, they disagree about the nature of the right comparator world, that is to say the world a comparison with which is needed in order to answer the question. Really, one might thus say, ‘the’ comparative pro-theism/anti-theism question is two questions, one for the theist and one for the atheist. For the theist: would God’s not existing have been better or worse than is His existing? For the atheist: would God’s existing have been better or worse than is His not existing? Kraay and Dragos define pro-theism as the view that it would be far better if God exists than if he does not; anti-theism, as the view that it would be far worse if God exists than if he does not; and they distinguish various sub-views within each (2013). Taking my terminological lead from them, in this paper, I want to argue for the acceptability of a particular answer to the comparative pro-theism/anti-theism question, an answer that I think (just) deserves the name ‘pro-theism’ even though it doesn’t fit Kraay and Dragos’s definition. This is the view that, as one will think of it if one is a theist, it is somewhat better overall or at least no worse that God exists than it would have been if He had not existed or, as one will think of it is one is an atheist, that it would have been somewhat better overall or at least no worse had God existed than it is, given that He does not exist. We may call this view ‘weak pro-theism’. It will be seen immediately that weak pro-theism is the weakest view that could possibly count as nevertheless on the pro-theistic side of this debate although if the way it comes out as true is by its being no better or worse whether or not God exists, weak pro-theism is really marking the border between pro-theism and anti-theism and thus a name such as ‘neutralism’ would better specify it. Why then do I set my ambitions so low? The short answer is that—as I shall, I hope, establish—only weak pro-theism can commend itself to theists and atheists alike as the answer to this pro-theism/anti-theism question. In that sense then, I am arguing for weak pro-theism as the epistemic lower-bound on the correct answer to the comparative pro-theism/anti-theism question pending resolution of the theism/atheism question. Thus weak pro-theism as defined is the agreeable answer spoken of in my title. In my conclusion, I will push the boat out just a little bit further, arguing that if my—only hesitatingly endorsed—arguments against certain contenders for being comparator worlds are right, weak pro-theism will come out true in a non-neutralist way. That is, as the theist will put it, it is not just no worse that God exists than it would have been had He not existed, but better (albeit only mildly

72  T.J. Mawson better). As the atheist will put it, it is not just no better that God doesn’t exist than it would have been if He had existed, but worse (indeed quite a bit worse). Before making this case, I need to explain why I am not taking either of two paths, each of which appear easier and each of which appears to lead to a much grander destination. First, one might suggest the following as a way to a stronger pro-theistic conclusion. Question for the theist: Given that we exist and theism is true, how much value is there in the world? Answer from the theist: The universe and our ante-mortem lives contain a certain amount, but—in addition to that—a large amount of value is added impersonally by the mere existence of God and, for us personally, by the sort of post-mortem lives that God will give us. Question for the atheist: Given that we exist and atheism is true, how much value is there in the world? Answer from the atheist: The same certain amount that the universe and our ante-mortem lives contain according to the theist and nothing else.2 Comparative question for both: So, which of you is evaluating reality as higher than the other? Answer from both: theism, and by quite a bit. The crucial point to make here is that this isn’t a way to answer the comparative pro-theism/anti-theism question. The question I’m addressing is not, ‘How much better does theism say the world is than atheism says it is?’ Although that is in itself controversial, I presume that the answer to that question may be addressed in this way and that it does come out as ‘Quite a bit’. And that is, it seems to me, speaking to one of the family of pro-theism/anti-theism questions. But it’s not speaking to the comparative question as I have defined it. The issue I am addressing is rather, for the theist, how much better or worse would it have been had theism been false? And for the atheist, how much better or worse would it have been had theism been true? And as already indicated, I’m interested in exploring whether or not we can reach agreement on the answer to this question across the theist/atheist divide. Second, there is what I shall call ‘the easy way’ to travel to the answer to my question that Kraay and Dragos call wide impersonal pro-theism, and it is noted by them. Roughly, it goes like this: God is by definition a being of supreme value, and thus worlds with Him in them thereby have a value-adding entity that swamps out any disvalues; therefore, every Godly world is better than every ungodly one just in virtue of every Godly world containing God and every ungodly world not doing so. I do not take the easy way for the following reasons, a defence—rather than articulation of which—would take me beyond the scope of this paper. I hold to something akin to what Parfit (1984, 488) calls the Full Comparative Requirement for assessing things as being good or bad for people, viz. that if something’s going to be good for a given person, it has to be the case that he or she would have been worse off had it not obtained (and for something to be bad, that he or she would have been better off

Agreeable Answer to Pro-/Anti-Theism Question 73 had it not obtained). That makes any person’s actually existing neither good nor bad for him or her. And that being so, I don’t count God’s existence as a good for Him—His existing doesn’t make Him better off than He’d have been had He not existed. (Nor would His non-existence be bad for Him.) I am somewhat diffident with respect to a person-affecting requirement on value per se, viz. the requirement that things can only be good or bad if there is actually someone for whom they are good or bad. On balance, I reject such a requirement, but I’m hesitant about accepting arguments that need it to be false, as does the easy way if one accepts my variant of the Full Comparative Requirement. As I nevertheless do— albeit hesitatingly—reject a person-affecting requirement on value, so I do—hesitatingly—think of it as impersonally good that God exists— even though (given my variant of the Full Comparative Requirement) it is not good for Him that He do so. I do—hesitatingly—think that it would have been good that God existed even if He had chosen not to create any other people, and thus His existence would have been good for no one. There is then, I would say, this much truth in the easy way. But, whilst I do on balance reject a person-affecting requirement on value, I do think that impersonal values (i.e., non-person-affecting ones) just aren’t that significant when personal values (i.e., person-affecting ones) are in play, that personal values are trumping over impersonal ones; and that bars the easy way to me. So, consider two worlds, A and B, that are almost exactly alike and each of which is much like the actual world. In each, there is an Earth, with people like us on it. The only difference between the two is as follows. In world A, an object with considerable impersonal value—perhaps a supremely beautiful crystal formation—exists on a planet beyond the light cone of any inhabited planet (e.g., its Earth), and on its Earth, a certain person suffers without respite from a migraine headache for an hour, a not-inconsiderable personal disvalue. In world B, that object fails to exist, and the person fails to have the headache. I evaluate World B as lexically better than world A. Similarly, it seems to me, the mere fact that God exists may be said to add great—even supreme, in its own terms— impersonal value to a world, but if it can’t be said—in addition—to add personal value, then that’s not saying much; if the fact that God exists makes it worse for people generally (this ‘if’ is, I of course think [defending as I am weak pro-theism], a per impossibile ‘if’), it must thereby make it worse overall. That being so, I’m going to focus on what personal difference God makes to the value of the world for us—individually and collectively. I’ll mention the impersonal difference as I pass, but not put much weight on it. I shall be taking theism to be the thesis that there exists a being with the properties ascribed to God by traditional Perfect Being Theology. Someone who believes that God exists then, I shall be taking it, simply believes that there is an eternal, omnipotent, omniscient, perfectly good

74  T.J. Mawson creator of everything other than Himself; and so on.3 And I shall be assuming throughout my presentation that this God, if He does exist, exists of metaphysical necessity (and, if He does not exist, then He does not exist of metaphysical necessity).4 In Mawson (2012), I argued that the only way to ground judgements that it would be better or worse were situations which obtain of metaphysical necessity to have been different is to portray metaphysically possible worlds as a proper subset of logically possible worlds.5 On this picture, we have the actual world as our centre of focus, naturally enough. Around it in logical space, we have metaphysically possible worlds (which are also, of course, logically possible). And then we cross a significant boundary, into a ‘doughnut’ of worlds that are logically possible yet metaphysically impossible. Comparative value judgements of a merely counterfactual sort—Would it have been better had Hitler died in his crib?—involve comparisons between worlds, all of which are in the centre of this doughnut. Comparative value judgements of a counterpossible sort (where the sort of possibility ‘countered’ is metaphysical in nature)—Would it have been better had I used a time machine in order to kill Hitler in his crib?6—involve comparisons between worlds that are in the centre of the doughnut and worlds which are in the doughnut itself. Only by thinking of the modal landscape in this way can we make both metaphysically possible and metaphysically impossible worlds available to us in logical space for evaluative comparison, and thus only by thinking in this way can we say of some metaphysical necessities that it would be better, not worse, were they otherwise and say of others that it would be worse, not better. (Thinking this way means that we do not need to say of every metaphysical necessity that it would be both better and worse were it otherwise—we avoid the counterintuitive consequences of the Lewis/ Stalnaker semantics for counterpossibles, viz. that they’re all true.)7 Rather than repeat the general arguments of my earlier paper in favour of the method, I shall start by giving an illustration of it at work on another—it seems to me, far easier—issue. Then I shall go on to apply the method more fully to the God issue. Personally, as already indicated, I think that time-travel of the sort one usually sees depicted in science-fiction stories is a metaphysical impossibility (because backwards causation is metaphysically impossible), but I believe that careful science-fiction writers can nevertheless write logically consistent stories which involve it. One day it occurs to me to reflect on whether it would be better for me personally were I to have access to a time machine of the sort I regard as metaphysically impossible—to keep things simple, were I to have access to such a time machine but nobody else to do so. I am reminded as I reflect of a scene that occurs in the film Bill and Ted’s Excellent Adventure. Our heroes arrive at a police station, where various historical figures who they have brought forward in time to assist them in passing a history

Agreeable Answer to Pro-/Anti-Theism Question 75 exam (‘report’) have been incarcerated by Ted’s father, the local policeman. If only they could use Ted’s father’s keys to release them, muses Bill. Ted reports that, sadly, his father lost his keys a couple of days ago. (In an earlier scene in the film, the father had even accused Ted of stealing them.) The following exchange then takes place: Bill: “If only we could go back in time to when he had them and steal ’em then.” Ted: “Well, why can’t we?” Bill: “’Cause we don’t have time!” Ted: “We could do it after the report.” Bill: “Good thinking, dude! After the report we’ll travel back to two days ago; steal your dad’s keys; and leave them here.” Ted: “Where?” Bill: “Don’t know. How about behind that sign? [Indicates sign] That way when we get here now, they’ll be waiting for us. [Bends down; pokes behind sign; hand emerges with keys] See!” Ted: “Whoa! So after the report, we can’t forget to do this otherwise it won’t happen. . . . But it did happen! So it was me who stole my dad’s keys.” Bill: “That’s exactly it, Ted!” It seems to me obvious that it would be great for me if I were able to pull off tricks like this (at least presuming nobody else is able to do so). As I look into the nearest world in logical space in which I have a time machine and nobody else does, I find that I’m clearly better off there than I am here; I never lose anything important (in the sense that if I do ‘lose’ something I care about enough, I go back in time to when I had it and safely secure it somewhere for my later self to find; I ‘do a Bill and Ted’). That being so, I can reach a well-grounded judgement that it’s relatively bad for me, personally, that I don’t have a time machine. I can also consider the more general issue of how things would work out for persons other than myself were I to have access to a time machine and nobody else to do so. And only slightly more hesitatingly, I conclude that it would be better for them too—my own good nature and sense ensure that, as well as purely self-interested ‘Bill and Teds’, I’d perform quite a few generally beneficent ‘Bill and Teds’ too. At my most bold, I find that I can even assess with tolerable confidence the value of such a world entirely impersonally: aesthetically, a world with causal loops has certain elegant aetiological structures that add value, independent of their effects on people; there is something about the original Bill and Ted story that pleases aesthetically. Similarly then, the suggestion goes, if one is a theist being asked whether it would be better or worse personally were God not to exist, one may look into what one judges to be the nearest world in logical

76  T.J. Mawson space in which there is no God and seek to see of oneself and others in that world whether or not one is better off and these others are better off. If one finds that one is better off, one should conclude that it would be better for one were there not to be a God; if one finds that one is not better off, that it would not be better for one were there not to be a God. The same goes for the others. And the same goes, mutatis mutandis, for an atheist who is asked to assess how much better or worse it would be for him or her and for others were there to be a God. And indeed, the same can be made to go for more impersonal judgements of value too: theist and atheist can assess how much better or worse the world would be, impersonally, were whichever of theism and atheism they believe to be actually true to have been false. So, anyway, that’s the methodological suggestion. Assuming it’s basically right, let’s turn to see how it plays out. First, I shall consider the comparative pro-theism/anti-theism question from the theistic point of view—my own, as it happens. Then I shall turn to consider it from the atheistic point of view. We shall see that what theists and atheists are justified from their respective starting points in thinking of as the relevant comparator worlds differ (and within each camp there is scope for division about what the relevant comparator world is); some candidates for comparator worlds support stronger pro-theistic conclusions than weak pro-theism. (Only very implausible candidates support anti-theistic conclusions.) To that extent then, the precise answer to the comparative pro-theism/anti-theism question waits on resolving the theism/atheism question and some other issues, issues I shall draw to our attention as we pass them. In that sense then, as I indicated earlier, what I want to be seen to be doing in this paper is establishing weak protheism as an epistemic lower-bound to the correct answer to the comparative pro-theism/anti-theism question, pending resolution of the theism/ atheism debate and these other issues. *** Considering the God issue as a theist, one thing I might be tempted to say is that in the closest world in which God doesn’t exist, nothing else exists either because it’s a metaphysical necessity that there’s a God; it’s a metaphysical necessity that every substantial thing that’s not God depends on God for its existence; and these two metaphysical necessities are logically discrete. If all that is right, as I journey from the centre outwards along the theism/atheism axis8 and into the metaphysically impossible doughnut, I ‘void’ the metaphysical necessity that is God, but having done so, I am left with the metaphysical necessity that everything substantial that’s non-God depends on God for its existence, and so the nearest world in which there’s no God has nothing else in it either; it is complete nothingness. (Perhaps abstracta still exist—I shall ignore them.) Let us call this world ‘Nothingness’.

Agreeable Answer to Pro-/Anti-Theism Question 77 If, as a theist, one were attracted to the view that standards of value logically depend for their existence on God, then one would say that there wouldn’t be any value at all in Nothingness—for good or bad. So, Nothingness would then have to be judged an un-evaluable world. This then indicates the first (at least somewhat controversial) assumption that one must make as a theist to use the method. One mustn’t hold, for example, to a Divine Command Theory as a logical necessity. One must hold that one of the ‘right’ value theories is true—‘right’ by reference to the criterion of keeping a standard of evaluation intact as one crosses the boundary into the doughnut whilst travelling on this axis or one must allow at least that some theory of value (different from the one that is actually right but that nevertheless broadly mirrors the ‘outputs’ of the theory that is actually right) is true in the relevant closest world. This adds a complication worthy of further consideration, but I should move on as it doesn’t affect my own line of argument. If one held to a personaffecting requirement on value as a logical necessity, such that nothing could, of logical necessity, be good or bad unless there was someone for whom it was good or bad, then again one would have to say that Nothingness would have no value in it, for good or bad. Again, Nothingness would have to be judged to be an un-evaluable world.9 I don’t hold to a Divine Command Theory (let alone to one as a logical necessity; in fact, I think that Platonism is true of logical necessity) and, as already indicated, I—hesitatingly—reject the problematic person-affecting view; it certainly doesn’t seem to me to be a logical necessity. So, I can in principle proceed further. I can in principle proceed further, but I might in practice get stalled at this point. If, even with all this in place, I just failed to form a judgement of the value of this metaphysically impossible world as compared to the actual world, that would be the end of the matter for me. But in fact, I am able to form the necessary evaluative judgement of Nothingness. As I look into Nothingness, I judge it to be worse impersonally than the actual world for the easy reason: it doesn’t have God—a being of great impersonal value—in it, whereas the actual world does. As I have already said, I don’t think that saying that alone is saying much, but it is saying something. So, if I think of the first world within the doughnut as Nothingness, my hesitating conclusion is that because Nothingness is worse impersonally than the something that we have, we should say it’d be worse were God not to exist, even though it wouldn’t be worse personally— for me or anyone else—if He were not to exist. As I’m not in fact as confident about the necessary judgements in the God case as I am in the case of considering myself alone having access to a time machine, so, via this route, I myself would hold only more-hesitatingly that it would be worse impersonally, but not personally, were atheism true. So far then, getting to a pro-theistic conclusion has required a number of controversial assumptions and then, finally, an unusual value

78  T.J. Mawson judgement—Nothingness is worse than the something we actually have; and not everyone shares these assumptions or makes this judgement. One of these assumptions—that standards of value don’t logically depend on God such that they drop out simultaneously with Him dropping out as one crosses into the doughnut (or that some other similar standard would take over of logical necessity)—is essential if the method is to work. But another of these assumptions is only required due to the need to evaluate Nothingness given my variant of the Full Comparative Requirement—the falsity of a person-affecting view of value. It would be nice then if we could show that a route that requires one to evaluate Nothingness (rather than more-easily-evaluable somethings) is not one that the theist would be well advised to take, as we could thereby bypass these issues and perhaps justify a stronger pro-theistic conclusion. One way in which one might hope to do this is by showing that it is a logical necessity that if there’s a God, then everything non-God depends on God. Then ‘both’ of these necessities will drop out at once as one crosses the border. I think that this can in fact be shown, through reflection on God’s essential properties of omnipotence and being the creator of everything other than Himself. As I reflect on what the classical theistic notion of God entails, it seems to me that it entails of logical necessity (through two of God’s essential properties, His omnipotence and His being the creator of everything other than Himself) that anything substantial that’s non-God depends on God. It depends on God both in the sense of depending on Him for its initial existence and character—He is the ultimate creator of everything other than Himself—and in the sense of depending on Him for its continued existence and character—even if, in creating certain creatures, He gives them some ‘existential inertia’, as we may put it, it must nevertheless be the case that, in virtue of His omnipotence, He retains the capacity to annihilate or change them thereafter; He cannot have made any creatures metaphysically necessary. These properties—ultimate creatordom, as I have elsewhere called it (Mawson 2005, 71), and omnipotence— taken together then make it a logical necessity that in any world in which there’s a God, anything non-God depends on God in the fullest sense of depends.10 The ‘two’ necessities are in this way logically linked.11 If then we do suppose these ‘two’ metaphysical necessities to be logically linked in this way, in travelling into the doughnut along the theism/ atheism axis, the first world one comes to is going to be one in which ‘both’ the metaphysical necessity that there’s a God ‘and’ the metaphysical necessity that anything that’s non-God depends on God for its existence are voided—there isn’t even a logically possible world where the first is voided and the second is not. On such a presumption, what should one then think about the nature of the first world? The world must be as close as is logically possible to the actual world given the voiding(s) mentioned. I find it no easy task to say with confidence what that world is like, but I do have some intuitions.

Agreeable Answer to Pro-/Anti-Theism Question 79 First, it seems to me that we should think of this world as containing a simulacrum of our physical universe. To assert that one would do better by saying that nothing that actually needs God happens in this closest world as there’s ex hypothesi no God there to make it happen—i.e., to assert that the nearest world is Nothingness once more—seems to assert as closer a world that is far more different from the actual than is a world with a simulacrum. Nothingness just is vastly different from any something, a fortiori from the actual something. The first view of this simulacrum I wish to consider is one in which the Big Bang; the sustaining of the laws of nature; and anything that’s happened in the actual universe as a result of miraculous intervention, have happened as a result of chance. So, in all properties (other than aetiological ones, obviously), this world is exactly the same as the actual; its history is exactly the same (or as close as it can be in logical consistency with there being no God), as is its present and future; and so we are there, thinking the same thoughts as we are thinking in the actual world. Given that in this world chance does all the things God does in the actual world, as well as the simulacrum of our earthly lives, there is also a simulacrum of our afterlives. Personally, I incline to think that Universalism, the thesis that everyone gets to Heaven, follows of logical necessity from theism and the fact that God’s created beings like us,12 but many theists would disagree with me about that and so we’ll need to consider how that would affect things as and when we come to it. Let’s call this—somewhat underspecified then— world ‘Simulacrum minus God’. That there are persons in Simulacrum minus God means that, unlike when evaluating Nothingness, we don’t need to rely on the assumption of the falsity of a person-affecting requirement on value in order to evaluate it and that these persons are us means that concerns about the Full Comparative Requirement become moot. And then, finally, that all these persons (minus God again, obviously) are doing exactly the same things and enjoying or enduring the same experiences and fates as we do in the actual world means that one might suggest that we can easily judge that things are no better or worse for anyone in Simulacrum minus God than they are in the actual world (or if they are worse for anyone, they are only worse for God, and that would require bringing in a denial of my variant of the Full Comparative Requirement). However, to conclude that things are no worse for us in Simulacrum minus God would be too quick. In Simulacrum minus God—most acutely in the afterlife section of it— our worship is everlastingly directed towards a non-existent God. This fact about Simulacrum minus God is sufficient, I take it, to make its heaven—let’s call it ‘ersatz heaven’—worse for us than real Heaven. We are deceived about what it is we are worshipping. Of course, in Simulacrum minus God we don’t realise that it’s worse—the ersatz heaven in Simulacrum minus God is phenomenologically identical to the real Heaven. But that ignorance of our everlastingly worshiping a God who

80  T.J. Mawson doesn’t exist doesn’t make ersatz heaven anything other than worse for us than real Heaven. Simulacrum minus God then is worse for each of us personally. (It is also worse impersonally, of course, through the nonexistence of God; an item of great impersonal value has been lost in crossing the border. But that is the easy result.) As I have said, I think that Universalism follows of logical necessity from theism and the fact that we exist; some theists differ; they think that not everyone who actually exists will be even as well off as are those in ersatz heaven; according to them, some go to Hell and thus in Simulacrum minus God some go to ersatz hell. It seems to me that ersatz hell would be worse than real Hell. At least in real Hell, ex hypothesi, one has been sent there after having been judged to be worthy of it by God; in ersatz Hell, one would have been sent there as a result of chance. This difference again seems to me to make it worse. And the same goes for purgatory and so forth. All in all, then, it seems to me pretty clear that Simulacrum minus God is at least somewhat worse for each of us personally (and, for the easy reason, worse impersonally). It’s personally worse, but—I would say—only mildly worse. But is Simulacrum minus God the right comparator? The qualitative similarity at the superficial level between Simulacrum minus God and the actual world is what I have so far suggested could be argued to make it closest to the actual world. But there are other contenders for closest world, even if we suppose theism and the voidings of it and the necessity that everything non-God depends on God to be simultaneous in travelling into the doughnut. The reason for calling Simulacrum minus God as closest is that, given that one’s knocked out God in travelling into the doughnut along this axis, one (arguably) does best to secure similarity by having anything God is in fact causally responsible for—plausibly the Big Bang; the sustaining of the laws of nature; miracles and our fates in the afterlife—happen anyway (otherwise there’d be needless difference) but happen without Him; hence the appeal to chance (i.e., uncausedness, i.e., nothing) to fill God’s role. But there is great plausibility in the assertion that one would do best, not by using chance but by using another substantial entity—the best candidate would have to be one that was as close to God in properties as is conceptually possible whilst remaining distinct, ‘God minus’ we might call him13—to play God’s causal role in this nearest world. After all, ‘chance’ doing something is just another way of saying that nothing is causing that thing and it is hard not to have sympathy for the view that a world in which the Big Bang had no cause; the laws of nature continue to operate without anything sustaining them; various happenings that in the actual world were miracles—Jesus raising from the dead, say—happen without causes; and then, all by chance, we have an everlasting afterlife of just the phenomenological sort God arranges for us in the actual world, whilst superficially identical to the actual world, is in fact vastly different below

Agreeable Answer to Pro-/Anti-Theism Question 81 the surface—aetiologically—and these below-the-surface aetiological differences should count for more when it comes for determining closeness. In travelling to Simulacrum minus God, one has voided God and a host of causal relationships; in travelling to Simulacrum minus God plus God minus, one has voided only one property of one entity. So, one might argue, it’s Simulacrum minus God plus God minus that one should call as closest. Let me explore then this alternative line of thinking a bit more. A hope—as always—is that we might find we can get to the same weak pro-theistic conclusion via different routes. The line of thinking that we’re now exploring then is that things that don’t happen by chance in the actual world won’t happen by chance in the nearest world within the doughnut along the theism/atheism axis. This line of thinking might take one in either of two directions: thinking that in the nearest world they won’t happen at all or thinking that in the nearest world they’ll still happen but be caused by the thing that’s as close to God as can be in properties whilst not being God—God minus, as we have dubbed him. First then, one might be led to think that perhaps it would, after all, be better to think of the first world one comes to as Nothingness, because even if in the first world one comes to when entering the doughnut along this axis it’s not a metaphysical necessity that anything that’s non-God depends on God, it’s still the case that things like universes coming into existence and laws continuing to operate don’t happen by chance; it’s not there a necessity that they don’t happen by chance, but it’s still true there that they don’t happen by chance. Nothingness then is the first world after all. If travelling by this route, then we can revert to the earlier discussion; I’ve already evaluated Nothingness and—hesitatingly—found it to be worse impersonally or at least—less hesitatingly—no better personally. Sadly, then, travelling by this route we will not have avoided having to evaluate Nothingness and the reasons for hesitation that having to do that brings (or at least brings me, given my only-hesitating rejection of a person-affecting requirement on value and belief in my variant of the Full Comparative Requirement). A stronger form of pro-theistic conclusion than weak pro-theism cannot thereby be justified. Second, one might be taken by this line of thinking to conclude that perhaps it would, after all, be better to think of the first world one comes to within the doughnut as having in it an entity that’s as close to God as can be in its properties without being God—God minus—playing the same causal role as God plays in the actual world. If so, one will think of the nearest world as Simulacrum minus God plus God minus. This world is very similar, but—I think—clearly (if only mildly) worse impersonally than the actual world in that the supreme being in it is God minus, who then is by definition at least somewhat worse than God14—a mild variant of the ‘easy’ conclusion. In addition, although God minus has been introduced under the description of causing everything God

82  T.J. Mawson actually causes (insofar as is logically consistent with God not causing it), in the ersatz heaven that he arranges for us, we’d be inadvertently worshiping God minus, not—as we suppose (and as things will be in the actual world)—God and that make things personally worse for each of us too. But in Simulacrum minus God plus God minus, we’d at least be worshiping something—God minus—and worshipping something that is at least somewhat worship-worthy. That seems to me to make Simulacrum minus God plus God minus better for us than Simulacrum minus God. And the same goes for the variants of theism which don’t embrace Universalism—the ersatz Hell that some are consigned to by chance in Simulacrum minus God only has people in it who are consigned there by God minus and being consigned there by God minus seems to me clearly better than being consigned there by chance. That then, it seems to me, is the way it would play out if one gave in to any feeling of sympathy for thinking that a world where chance fills the role vacated by God would be more distant than a world in which either it remains unfilled or God minus fills it. I hope to have shown then that, however these things play out, my weak pro-theistic conclusion is warranted from a theistic starting point. It would be worse were atheism true impersonally, for the easy reason that God, a being of great impersonal value who actually exists, wouldn’t exist were atheism true; but, for reasons given at the outset, that doesn’t count for much. What counts for more is the fact that it would ‘probably’— in the sense that one can get to this conclusion on more combinations of assumptions than not (but not on all combinations)—be worse were atheism true for each of us personally. I have not been able to eliminate all routes by which the theist will find himself or herself called upon to evaluate Nothingness, but—nevertheless—I have done something to indicate how, on certain presumptions, Nothingness is not the right comparator— the right comparator is either Simulacrum minus God or Simulacrum minus God plus God minus, each of which are much-less-problematically evaluable and much-less-problematically evaluable as at least somewhat (even if only mildly) worse for us personally. But obviously, even if Nothingness is the right comparator and is only problematically evaluable as worse, it is non-problematically evaluable as no better than the actual world. Therefore, weak pro-theism is justified.15 Let’s turn now to consider the issue from an atheistic perspective. *** Things are of course going to appear somewhat different to the atheist. But let’s consider the sort of atheist for whom they’re going to appear as little different as possible. Let us assume then that the atheist we are considering accepts the general account of the metaphysically possible as a proper subset of the logically possible and accepts one of the ‘right’

Agreeable Answer to Pro-/Anti-Theism Question 83 value theories, right by reference to the criterion of keeping a common standard of evaluation (or at least one that is sufficiently close) in place as we first journey into the doughnut along the theism/atheism axis or, as he or she may prefer to call it, the atheism/theism axis. What should the atheist say about the first Godly world he or she comes to, after crossing the border into the metaphysically impossible doughnut along the atheism/theism axis, as he or she might put it? I previously discussed different attitudes that one might take towards the logical discreteness of the (putative, as the atheist will insist) metaphysical necessities that there’s a God and that everything that’s non-God depends on God for its existence. For the atheist, as for the theist, this makes a difference as we shall see. The first issue I shall address is whether or not the atheist should think of the first world he or she comes to within the doughnut as containing a simulacrum of our universe. Obviously, a case can be made for thinking that the first world contains a simulacrum of our universe as its doing so keeps that world, superficially at least, as close as possible to the actual. Given what I’ve already said, it will come as no surprise that I am convinced by this case. If the atheist keeps discrete the necessities of God’s existence and everything that’s non-God depending on God, then the first logically possible Godly world that he or she comes to will be one in which a simulacrum of our physical universe (all that there is, I take it, on atheism) and also God exists and yet, as there’s no metaphysical necessity that the physical simulacrum that’s in it depend on God and as the actual physical universe doesn’t depend on God, the physical simulacrum that exists in it doesn’t depend on God either. In the first world then, God stands oddly devoid of work to do when it comes to the starting of universes and the sustaining of laws of nature. Alternatively, an atheist who considers the relevant ‘two’ metaphysical necessities to be logically connected in that if there’s a God, then, of logical necessity, anything non-God depends on God, and who does not regard under-the-surface aetiological differences as making that much difference, might think a simulacrum of our universe exists in the nearest world, but that it is linked to God in the conventional theistic way; in this world, God does the sorts of things theists characteristically say He does with respect to the physical world—e.g., He starts it off and sustains it. Again then, there’s a physical simulacrum and a God. The lack of dependence or dependence of a physical simulacrum on God doesn’t seem to me to matter, value-wise, or at least matter much.16 So I think that—sweeping some minor issues aside—we can actually lump both these worlds together for purposes of evaluation. That’s what I’ll do. But simulacra-plus-God worlds, as we may call them, are not the only contenders for comparator worlds. As earlier, one might think that it’s not superficial similarities, but rather the under-the-surface aetiological properties that are more cogent

84  T.J. Mawson in determining which world is closest. The atheist thinks of the actual world as having God-independent physical stuff in it. In the nearest world in which there is a God, then, if there is a logical necessity linking God’s existence to the necessity that everything that’s non-God depends on God, any stuff that existed would have to be God-dependent physical stuff. But a world with God-dependent physical stuff and God would be different from the actual world in two significant ways—there’d be God-dependent physical stuff and God—not just one—there’d be God (but still no God-dependent physical stuff). If so, then perhaps the atheist should think of the first world within the doughnut as God and nothing else, no simulacrum of our universe then. We could call this world ‘Just God’. How then to evaluate simulacra-plus-God worlds and Just God relative to the actual world as it is seen by the atheist? Turning to the simulacraplus-God worlds first, it seems to me that the atheist may use the easy reason to judge any simulacrum-plus-God world as better impersonally, even if I’m right that that alone will not be much. In addition, and more weightily, it seems to me that the atheist should judge of these worlds that they are somewhat worse for them personally in one respect (not, n.b., overall) than the actual in that in them atheists are mistaken about a matter of metaphysics which they get right in the actual world. That seems to me to do something to make simulacra-plus-God worlds somewhat worse for atheists personally.17 But it also seems clearly to make these worlds not a great deal worse and the badness of mere wrongness about the theism/atheism question is swamped out by another factor—Heaven. As already indicated, I think it is a logical necessity that if God and we exist, God gives us all an afterlife in Heaven. If that’s right, then—of logical necessity—as soon as we have God and a physical simulacrum in a world, we have Heaven for all of us in that world too. And the fact that the atheist should thus say that we have a heavenly afterlife in the first Godly world we come to makes it easy for him or her to evaluate that world as significantly better for each of us than the actual world as it is pictured by the atheist (where, I take it, we don’t—any of us—get to Heaven). And it seems to me obvious that Heaven for the atheist swamps out the badness of his or her having picked the wrong side on the theism/ atheism question in his or her earthly life. So simulacra-plus-God worlds are to be judged a lot better than the actual world by the atheist. We might even imagine that the atheist will want to rename them now that he or she has drawn the implication that each contains a Heaven, ‘Simulacra-plus-God-and-therefore-plus-Heaven worlds’. It is hard to say what the atheist should in consistency conclude if he or she thinks that I’m wrong in logically tying a heavenly afterlife for all of us in physical simulacra to God’s existence in this way. It depends, obviously, on what, if anything, he or she thinks does follow from our having ante-mortem lives in physical simulacra plus the existence of

Agreeable Answer to Pro-/Anti-Theism Question 85 God—follow of logical necessity—about God’s giving us post-mortem lives (and of what sort) in those simulacra. If nothing follows, then it seems as if the atheist should conclude that the first world we come to within the doughnut is one where we have just a physical simulacrum plus God—no afterlife at all (as we don’t have an afterlife in the actual world, he or she will presumably assert). If so, then there’s not the sort of swamping that I suppose. It would hence be mildly worse for atheists personally if theism were true (in virtue of it being mildly worse to be wrong rather than right about an issue in metaphysics).18 If the atheist thinks that it follows of logical necessity from there being a God that He’ll send everyone to Hell, then he or she in consistency will think that there’s a personal swamping of a small bad thing for him or her (that he or she backed the wrong answer on the theism/atheism debate) with a huge personal bad thing for him or her and everyone else. A strong antitheistic conclusion would be warranted for such atheists. But that that would follow of logical necessity seems to me very implausible. What’s much more plausible is that it’s entailed by the concept of God that God will give each person the sort of afterlife that is best for him or her. I infer ‘Heaven for all’ from that; but an atheist who thinks that leap too great to be called logically valid can still conclude that it would be personally good for everyone were theism true (as long as some sort of afterlife is best for everyone and getting the sort of afterlife that’s best for one is always better for one than having been wrong about an issue in metaphysics is bad for one). It seems then that weak pro-theism, even if nothing stronger, should commend itself to any atheist who considers a simulacrum-plus-God world to be the closest within the doughnut along the relevant axis. How to evaluate Just God? Similar issues arise as arise for the theist in evaluating Nothingness—none of us are in Just God to benefit or be harmed (if we assume my variant of the Full Comparative Requirement) by not existing. So (if we do assume this), Just God is neither better, nor worse, for any of us personally. Is it better or worse impersonally? Well, on the up side, it has God in it, which adds, I take it, great impersonal value—the easy conclusion. But, on the down side, as with Nothingness, it doesn’t have any of the sorts of impersonal values that we create in the actual world—objects of beauty, philosophy, and so forth. I would hesitatingly say that the atheist should say of Just God that it would be better than the actual world impersonally (because God adds more value than is taken away by the absence of the sorts of impersonal values that we create), but I am hesitant about that. (As already stated, I only hesitatingly reject a person-affecting requirement on value.) I can of course be more confident that Just God would not be worse for any of us, as we wouldn’t be there and I am, as always, assuming my variant of The Full Comparative Requirement. So, again weak pro-theism is justified; anything stronger I find myself unable to support without some hesitation.

86  T.J. Mawson

Conclusion I’ve explored the routes by which one might travel to a particular answer to the comparative pro-theism/anti-theism question as it will be configured by the theist and as it will be configured by the atheist, and I have shown that all these routes leave one justified in endorsing weak protheism; it is somewhat better or at least no worse that there is a God than it would have been had there not been a God/it would be somewhat better or at least no worse were there to have been a God than it is, given that there’s not. Some of these routes lead to stronger pro-theistic conclusions than that. My point is that none lead to weaker (not that there is a weaker form of pro-theism than my weak pro-theism; the ‘weaker’ form would be some form of anti-theism). In summary, depending on which route he or she takes, the theist should judge the nearest world within the doughnut along the relevant axis as either (a) Nothingness, (b) Simulacrum minus God, or (c) Simulacrum minus God plus God minus. All of these candidates for comparator world, I have suggested, can be seen to be worse than the world theism commits one to thinking is actual. Nothingness—(a)—is worse impersonally than the actual world (for the easy reason) but not worse— or, significantly, better—for anyone personally (given my variant of the Full Comparative Requirement). Simulacrum minus God—(b)—is worse impersonally (the easy reason, again) than the actual world and mildly worse for all of us personally; it’s worse even than Simulacrum minus God plus God minus—(c)—in that the people in Simulacrum minus God are worshipping nothing at all. Simulacrum minus God plus God minus—(c)—is worse than the actual world impersonally (the easy reason, again) and even-more-mildly worse for all of us personally; at least the people in this world are worshiping something that is at least somewhat worship-worthy. Depending on which route he or she takes, the atheist should judge the relevant comparators to be either (a) A Simulacrum-plus-God world or (b) Just God. As much as the theist, the atheist may use the easy reason to evaluate simulacra-plus-God worlds—(a)—as better impersonally. Simulacraplus-God worlds don’t differ much in value from one another; each are worlds in which there’s a physical simulacrum of the actual world, which may or may not depend on God, but, in any case, after each of us has lived in that physical simulacrum, we enjoy a heavenly afterlife (presuming that I am right in thinking that this is logically entailed by there being a simulacrum of our earthly lives and God). Simulacra-plus-God worlds are thus clearly considerably better personally than the world that the atheist thinks is actual. If I’m not right in thinking that a simulacrum of our earthly lives and God entail Heaven for all, the route the atheist will need to tread in order to evaluate them as better for us than the actual becomes more treacherous. But still, I have suggested, it will be navigable

Agreeable Answer to Pro-/Anti-Theism Question 87 to my weak pro-theistic conclusion (or even something stronger). Just God—(b)—is hard to evaluate for similar reasons to Nothingness being hard to evaluate for the theist, though it seems to me—hesitatingly—that the right thing for the atheist to say is that it would be better impersonally than the actual world (for the easy reason). It seems to me—far less hesitatingly—that the right thing for him to say is that Just God is not better or worse personally than the actual world, in that we don’t exist in it. Just God then is—hesitatingly—to be said to be impersonally better; unhesitatingly, to be said to be not personally worse. Weak pro-theism again is supported. If that though is all that can be said in going along this route, it reveals once more quite what a weak pro-theistic view weak pro-theism really is. Still, for the reasons discussed in this paper, I do not think that anything stronger than weak pro-theism can rationally commend itself to theists and atheists alike as the answer to the comparative version of the pro-theism/anti-theism question. One might hope that progress could be achieved on the comparative pro-theism/anti-theism question prior to progress being made on the theism/atheism question merely by knocking out some contenders for comparator worlds, but a quick look at the various contenders and what would follow from knocking out various combinations of them reveals that doing merely that would not, in fact, advance the case that agreement across the theist/atheist divide can coalesce on something stronger than weak pro-theism. Nor would merely knocking out my variant of the Full Comparative Requirement. Nothingness was worse only impersonally than the world the theist thinks actual due to my variant of the Full Comparative Requirement; without it, it might be said to be worse, indeed—why not?—a lot worse for each of us personally that we don’t exist in it. That then might justify (assuming the necessities of God’s existence and everything non-God depending on God can be kept separate) from a theistic starting point, a strong pro-theistic conclusion. However, knocking out my variant of the Full Comparative Requirement would, by the same token, also leave Just God worse for each of us personally than the world the atheist is committed to thinking actual, justifying for the atheist an anti-theistic conclusion. So, we’d lose agreement. Remember, I am looking for an agreeable-across-the-theist/atheist divide answer. Of course, if one could knock out Just God in addition—leaving the atheist with a simulacrum-plus-God comparator, that might help somewhat. In fact though, I have argued, Nothingness should not be taken to be the comparator world for the theist; the theist should take Simulacrum minus God (or Simulacrum minus God Plus God Minus) to be the comparator world and this is only plausibly mildly worse, personally, than the world the theist thinks actual; similarly, I have argued, the atheist should not take Just God to be the comparator, but rather take a simulacrumplus-God world to be the comparator, which he or she should think is quite a bit better, personally (in that we all get to Heaven on it [assuming

88  T.J. Mawson my argument to Universalism is right]), than the world that the atheist thinks actual. That being the case, again agreement across the theist/atheist divide on anything stronger than weak pro-theism can’t be reached if considering personal value alone. But perhaps it could be reached on weak pro-theism coming out true in a way that means it doesn’t deserve the name ‘Neutralism’, as I introduced it earlier. Regardless of my variant of the Full Comparative Requirement being right (it is just not needed to evaluate any simulacra-minus-God worlds or any simulacra-plus-God worlds), the theist must rate the alternative to there being a God as only mildly worse personally than reality; the atheist must rate the alternative to there not being a God as considerably better personally than reality. Both then can agree on weak pro-theism coming out true in the nonneutralist way on grounds of personal value. But, even with Nothingness and Just God knocked out, nothing more strongly pro-theistic than that can be agreed with respect to personal value. Of course, one may bring in the easy reasoning for something stronger on the grounds of impersonal value, but then I have my hesitations about that (generated by my only hesitatingly rejecting a person-affecting requirement on value anyway and less hesitatingly, thinking that impersonal values don’t count for much when personal ones are in play). If we want to get more agreement than that, we’ll need to resolve the theism/atheism question first so we can work out which variant of the comparative question we should be asking. Or we’ll need to switch to other questions. As I started by arguing, the following is an issue which is different from the comparative pro-theism/anti-theism issue as I defined it, but which admits of easier agreement across the theism/atheism divide. The question as it will strike the theist: How much worse would the world be were I to be wrong in my theism?19 Answer: Quite a bit. The question as it will strike the atheist: How much better would the world be, were I to be wrong in my atheism? Answer: Quite a bit. So, on these grounds, both theist and atheist can join in reasonably hoping that theism is right and atheism is wrong. But—to repeat the point made at the start—that issue is just not the same as the comparative pro-theism/anti-theism one that I’ve been addressing. The comparative pro-theism/anti-theism question as it may strike the theist is this: How much worse would the world have been had theism been wrong? Answer (I have argued): Impersonally (and thus to be said only hesitatingly), quite a bit; personally, only mildly. The pro-theism/anti-theism question as it may strike the atheist is this: How much better would the world have been had atheism been wrong? Answer (I have argued): Impersonally (and thus to be said only hesitatingly), quite a bit; personally, quite a bit. Given my reasons for hesitation about impersonal value, I myself hence conclude that weak pro-theism as defined is the answer that must be given to the comparative pro-theism/ anti-theism question pending a resolution of the theism/atheism question or other issues, e.g., those leading to my hesitation. But at least—due to

Agreeable Answer to Pro-/Anti-Theism Question 89 the issue on which they can reach easier agreement—theists and atheists can join in reasonably hoping that they’ll discover that the answer to the theism/atheism question is theism. We, theists and atheists alike, can all agree that if theism’s right, things are a lot better than if atheism’s right20 even though we can’t agree that if theism’s right, things are much better than they would have been had it been wrong.21

Notes 1. Some philosophers who have engaged with pro-theism/anti-theism questions are Kahane (2011; 2012) and Kraay and Dragos (2013) 2. I am of course ignoring in all this those atheists who think things such as that some trans-dimensional natural being will give us each a pleasant immortality. 3. See, for example, the first part of Mawson (2005) for a description of these standard properties. 4. These definitional moves actually have more implications than they may seem to have at first sight. My way of understanding theism brackets off what would otherwise be possibilities in the following way. (Allow me to assume that God exists for ease of presentation.) Later on, I shall argue that the nearest thing in conceptual space to God is a being that has properties that are as similar as possible (without being identical) to those that God actually has; I call such a being ‘God minus’. (I go into this in more detail in due course, but the rough idea is that God minus would fail by the narrowest of margins to be perfect; perhaps he might be ignorant of one insignificant truth, say.) Had I thought of ‘God’ as a logically proper name, picking out—assuming as I am here that He exists—a being that has these properties, then it would have been logically possible that there be another being who was qualitatively identical to God differing simply in having a differing haecceity (assuming haecceities are logically possible). A consideration of worlds containing that being doing the actual God’s ‘job’ then would have had to take the place of my discussion of worlds containing God minus closely approximating in his doings God’s job; and the argument would be affected accordingly. I am grateful to Brian Leftow and Martin Pickup for pointing this out to me. Fortunately, how the overall argument would be affected were one to understand theism and ‘God’ in this alternative way can be easily seen—it would make the case for something stronger than weak pro-theism even harder to make than I contend that it is in the main text, as it would threaten to come out true that if there hadn’t been this God, there’d have been another one, equally as good (on the assumption that a difference in haecceity is of logical necessity the smallest difference possible between two beings).   The issue of God’s existence (or non-existence) being metaphysically necessary is also important, as we’ll see; though it seems to me that similar issues would play out in a similar fashion were one to think that the issue of whether or not there’s a God is a contingent one, thinking of it as non-contingent is standard and throws the issues into sharper relief. 5. Mawson (2012) was a development of an argument I presented first in a paper that appeared in print later (Mawson 2013). In both of these papers I use the terms ‘personally’ and ‘impersonally’ with different meanings from those they have in this chapter. The argument of Mawson (2013) was summarised and challenged by Rota (2013). With a few uninteresting caveats, I now accept Rota’s arguments.

90  T.J. Mawson 6. This is an example, of course, only on the assumption that time travel is metaphysically impossible but not logically impossible. If you don’t happen to so grade it, you could consider arriving at a hotel late at night and being told that all the rooms are occupied, the last one being taken only ten minutes before you arrived. Counterfactual question: would it have been better for you had you arrived fifteen minutes earlier? Counterpossible question: would it have been better for you had this been Hilbert’s Hotel? 7. Well, I say this, but it’s far from clear that we will avoid all problems. For example, I would incline to think that some counterpossibles with logically impossible antecedents are informatively true and some false, and the method won’t allow us to see how that could be the case. (If pi had been 3, then my math homework would have been a lot easier.) 8. That is to say, whilst travelling from the actual world at the centre along the axis which, if theism is true, makes the first world one comes to within the doughnut the one where theism is false but everything else is as close as it can be—in logical consistency, theism having been knocked out—to the actual world. If atheism is true, then in travelling along the theism/atheism axis (or—as atheists might prefer to call it—the atheism/theism axis), the first world one comes to within the metaphysically impossible doughnut is one where atheism is false but everything else is as close as it can be—in logical consistency—to the actual world. In this sense of ‘axis’, there are innumerable axes along which one may travel out from the actual world and into the metaphysically impossible doughnut (innumerable presuming we cannot enumerate metaphysical necessities). 9. Or more precisely, it would have to be so judged if one added in my variant of the Full Comparative Requirement on assessing harms and benefits accruing to persons; without that addition, then, for example, one’s not existing in Nothingness could potentially be counted as a harm (or indeed a benefit) for oneself. Brian Leftow has put pressure on the Full Comparative Requirement by raising the following point in discussion: ‘From a you-world, you can see that there’s less value for you there [in a world without you] (none). So, why doesn’t it make sense to say that it’s worse for you?’ I have never been to Australia; so, as I look to Australia, I can see that there’s less value for me there than there is where I am, Great Britain. But it doesn’t seem to me that the right thing to say in the light of that is that Australia is worse for me than Great Britain. In the main text later, I do talk about what might follow if one was not as committed to the Full Comparative Requirement as I am. 10. It would not be the fullest sense of depends either if it was simply that God must refrain from destroying it, nor if it was simply that it depended on God for its initial existence and character. But see following note for how the divine properties of ultimate creatordom and omnipotence taken together generate the fullest sort of dependency. 11. One might suggest that one would have been able to block this linking argument if it had been premised on considerations arising from omnipotence alone. One might assert that whilst God’s omnipotence entails that anything non-God depends on God in the sense of depending on His permitting it to exist with the character that it has, it’s logically possible for something to depend on Him in this sense whilst not depending on Him in the sense of requiring Him for its having its initial existence and character (and if a part of this something’s character is its having a certain existential inertia, then it won’t depend on Him in the sense of requiring of Him that He act to sustain it, just that He omit from acting so as to annihilate it). If that’s right, then, if one was considering omnipotence alone, one might be tempted to say that

Agreeable Answer to Pro-/Anti-Theism Question 91 there is a logically possible world in which two beings are co-eternal, God and an angel; God did not bring this angel into existence, yet—in that God could annihilate that angel or change that angel’s character at any stage—He remains omnipotent over it. However, when one brings the property of ‘ultimate creatordom’ into the picture too, one can see that this is ruled out as a possibility, and the property of being the ultimate creator of everything other than Himself is as essential to the theistic concept of God as is omnipotence.   One might suggest that one would have been able to block this linking argument if it had been premised on considerations arising from ultimate creatordom alone; one might assert that whilst God’s ultimate creatordom entails that anything non-God depends on God for its initial existence and character, God could have given something a character such that it continued to exist of metaphysical necessity—such an object would be one which not only did not require His continued sustaining action to persist, but which not even He could destroy. But that, of course, would be repugnant to His omnipotence, which—I am taking it—is equally an essential part of the theistic picture of God. 12. See Mawson (2005, 86ff). 13. So, God minus might fail to know the most insignificant of truths but in other respects be like God; he is the smallest extent possible short of a Perfect Being. One could deny the coherence of God minus; perhaps it is not logically possible that there be a smallest margin by which one might fall short of perfection, rather as it is not possible that there be a smallest margin by which a number might fall short of another. If one thinks so, then this would be a reason to stick with Simulacrum minus God as the comparator. 14. Remember always, we have ruled out that the difference between the actual world and the closest in the doughnut might be one of bare identity by our specification of how theism and ‘God’ are to be understood; the nearest world in which theism is false is one in which no being has the full suite of properties that Perfect Being Theology says God has. 15. I cannot myself get to anything stronger, tied as I am by my commitment to my variant of the Full Comparative Requirement; any theists loosed of such a commitment may be able to posit our not existing in Nothingness as a harm, even a great harm, that we’re relatively a lot better off having avoided. From such a starting point, a stronger pro-theistic conclusion could be warranted iff Nothingness were in fact the right comparator, which I’ve argued, in any case, it is not. 16. Theists get to be right about what God is doing in the simulacrum with the causally active God, which makes it better for them, but—by the same token—atheists get to be wrong in thinking that God’s not doing anything, which makes it worse for them. Weighing these pros and cons against one another is the sort of issue I’m sweeping aside in the main text at this point. I return to a variant of the issue in a moment in the main text. 17. By the same token, atheists should judge such worlds to be better on this count for theists personally. Some weighing will then need to be undertaken to work out whether or not they are net better or worse on this count inter-personally. 18. Again, by the same token, they should judge it mildly better for theists. And again some weighing off would need to be done to determine whether it’s net better or worse inter-personally on this count. 19. I assume that putting it this way sets the constraint that the theist asking it must exist in the world he or she is then led to consider—for him or her to be wrong in his or her theism, he or she needs to exist. If that constraint isn’t yet built in by the phrasing of the question, tweaking it, I presume, could make

92  T.J. Mawson it so. Having something like this in play helps by straightaway removing the temptation to consider as cogent Nothingness (and, mutatis mutandis, for the atheist, Just God). 20. I haven’t actually given the argument for this in anything but a compressed form, earlier on. 21. I am grateful to Jason Carter, Klaas Kraay, Brian Leftow, Martin Pickup, Richard Swinburne, Roger Trigg, and Vincent Vitale for their comments on this paper.

References Kahane, G. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696. Kahane, G. 2012. “The Value Question in Metaphysics.” Philosophy and Phenomenological Research 85: 27–55. Kraay, K. J. and Dragos, C. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 57–178. Mawson, T. J. 2005. Belief in God. Oxford: Oxford University Press. Mawson, T. J. 2012. “On Determining How Important It Is Whether or Not There Is a God.” European Journal for Philosophy of Religion 4: 95–105. Mawson, T. J. 2013. “Is Whether or Not There Is a God Worth Thinking About?” In Ethics and the Challenge of Secularism: Russian and Western Perspectives, edited by David Bradshaw, 5–19. Washington, DC: The Council for Research in Values and Philosophy. Parfit, D. 1984. Reasons and Persons. Oxford: Oxford University Press. Rota, M. 2013. “Whether or Not There’s a God Is Worth Thinking About.” In Philosophy of Religion: An Almanac (2012–2013), edited by Vladimir K. Shokhin and translated by Kirill Karpov, 179–185. Moscow: Nauka-Vostochnaya Literatura.

Part II

Arguments for Anti-Theism

5 If There Is a Hole, It Is Not God-Shaped Guy Kahane

In a letter that Bertrand Russell wrote to his then lover Colette O’Neil, he confesses that The centre of me is always and eternally a terrible pain—a curious wild pain—a searching for something beyond what the world contains, something transfigured and infinite—the beatific vision—God.1 In an earlier letter to Lady Ottolline Morrell, another lover who was a religious believer, Russell wrote, Turbulent, restless, inwardly raging—I shall always be—hungry for your God and blaspheming him. I could pour forth a flood of worship— the longing for religion is at times almost unbearably strong.2 Russell was of course an avowed atheist, a lifelong critic of Christianity.3 But the state of mind he reports is not inconsistent or even uncommon. Russell rejected theism, yet found the godless universe that we inhabit grim and deeply unsatisfying. Something is missing—the thing that, for religious believers, is supplied by God. There is a religious yearning in him that cannot be fulfilled in a universe governed merely by blind, indifferent natural forces. So while Russell didn’t believe in God, he ardently wished that God had existed. We can call this combination of attitudes pro-theist atheism.4 The cliché says that the theist’s God-shaped hole is the atheist’s holeshaped God. If atheism is true, then both claims could be correct: we may feel a deep lack, something missing in our worldly existence. And wishing to believe that there is something out there that would fulfil this lack, some of us project a divine being, ‘shaped’ exactly to address that need, onto the cosmos.5 We could understand the metaphor of the ‘God-shaped hole’ in phenomenological terms, as merely a claim about the psychology of some people. I am not interested here in this psychological question. I’m interested in this metaphor as a substantive claim: the view that something

96  Guy Kahane essential is missing in the naturalist universe we inhabit, a universe that Samuel Beckett described as a “world of desolation”,6 and which is governed, as Richard Dawkins put it, by “nothing but blind, pitiless indifference” (2008, 133).7 Before ending with this phrase, Dawkins describes our universe as one with “no design, no purpose, no evil and no good” (2008, 133). I assume that Dawkins means that the naturalist universe itself is neither morally good or evil (or, indeed, morally anything), nor created by a being with such properties. But that isn’t the same as to hold that the naturalist universe doesn’t contain good or bad things, or that it cannot as a whole be said to be overall good or bad. Now some do hold that God is the source of all value, which would mean that in the absence of God it would make no sense to ascribe value to anything. But if there is no value then God’s absence cannot be a bad thing either, or make our lives worse. If nothing matters then that fact also doesn’t matter.8 Russell has no reason to feel such agony—not that his agony would matter anyway. So we can set aside nihilism here.9 Russell presumably thinks that our naturalist universe is in some way worse than the theist alternative that Lady Morrell believes in. But the mere fact that something is overall worse than some conceivable alternative is hardly a reason for existential anguish. Even the most extraordinary things could be bettered. Hamlet or the Sistine Chapel aren’t unimprovable. But this doesn’t make them any less wonderful. Shouldn’t we at least prefer that things had been better, if they could have? Even that isn’t obvious. Perhaps we should be satisfied with what’s very good, or even just good enough. And anyway, such a pale preference for a better alternative isn’t the same as thinking that this alternative really matters. Russell might hold that the naturalist universe is overall bad—in the sense that the overall balance of value in it is negative. That would be a form of pessimism in the Schopenhauerian mould. But Russell’s anguish is obviously not a response to all the agony around him. And it’s anyway hard to place an overall value on our universe. Things here on Earth aren’t going so well right now, to be sure, but can we really say with confidence that they have overall positive or negative value? And the future of humanity might extend billions of years into future, which could contain wondrous utopias, or endless misery. And that’s just referring to us humans. For all we know, there might be countless intelligent forms of life scattered around the universe, and our miserable performance might be an aberration. Or we might be the only ones in this vast cosmos.10 We don’t know how good or bad our universe is overall. Can we at least know that, in the absence of God, our own lives are bad? That would easily justify Russell’s despair, and his longing for God. But although Russell wasn’t a happy soul, his extraordinary life seems far from a life that is overall bad—the kind of life, as some would say, that isn’t worth living. Nor does Russell cite failure or misery in explaining why he longs for God—he is feeling that terrible pain because God doesn’t exist.

If There Is a Hole, It Is Not God-Shaped 97 It would be better, I think, to understand Russell’s anguish, or the metaphor of a ‘God-shaped hole’, not in terms of positive badness, or the complete absence of value, but in terms of the absence of sufficient good. It’s not that things are bad, not even that they lack overall positive value. But our lives aren’t good enough, they lack certain essential goods that are simply missing from the kind of universe in which find ourselves.11 This claim can be stated impersonally, at the level of the universe. You might say: the universe itself has a ‘God-shaped hole’, it’s not just that it could have been better, but it is missing something essential. After all, the fundamental axiological fact about our universe isn’t the misery down here in Earth but (setting aside aesthetic value) the utter absence of value in the vastness that surrounds us—outer space is an amazing waste of space.12 As experienced by Russell and others, however, the lack that is supposed to follow from God’s absence seems a self-focused matter: there are certain things that our lives need to be genuinely satisfying and that we simply cannot have if there is nothing beyond or outside nature.13 Such dissatisfaction with the world as we find it, and a sense that the lives we can lead in it are fundamentally lacking, is a recurring sentiment in the modern era. It is also, I think, an underappreciated source of social unrest and upheaval.14 I have to confess that, at least in some moments, I feel something like this lack, even though in nothing approaching the dramatic form that Russell reports. Now this sense of lack may well be illusory, but my aim here is neither to defend nor to dismiss it.15 What I shall argue is that if there is such a ‘hole’, it is not God shaped. If belief in God is a projection of that existential need, it is based on a double mistake—not just the mistake of the projection, but the further mistake of projecting the wrong thing. I should note, however, that my core argument doesn’t assume that our lives are lacking something essential. Even if there isn’t such a ‘hole’, it’s obvious that the actual world contains a great deal of misery, and could have been vastly better. Even if we ask how things might have been different, and better, more out of curiosity than out of Russell’s despair, our reflections shouldn’t lead us to wish that God had existed. While my argument will be developed from an atheist perspective, it also has implications for theists. If I am right—though I don’t expect theists to accept the value claims I will rely on—then theists have reasons to want God not to exist. To stretch the clichéd metaphor to the limit: theists should wish that the universe had a large hole where God now majestically looms. Even God cannot bring us true satisfaction.

Naturalism, not Atheism Let me start with some important preliminaries. Russell was an atheist, but this actually tells us little about his metaphysical views. The point

98  Guy Kahane is familiar: atheism is merely a negative claim, a denial of God’s existence. The range of views that count as atheist in this sense is large. These include polytheism, Manichaeism, solipsism, the idealism of Hegel or Fichte or of F. H. Bradley or McTaggart, Schopenhauer’s malignant cosmic will, Aristotelian teleology (perhaps sans Unmoved Mover), NeoPlatonism and other forms of axiarchism, Daoism, the metaphysics of karma and rebirth, and very many others.16 This list of views obviously doesn’t mark an interesting metaphysical grouping—except perhaps for a theist. Of course when someone is described as an atheist, we would today also assume that they are naturalists of some sort—that they reject not only the existence of God and His angels, but the supernatural more generally, and that they take the world to be, at least in its ground-level ontology, to be roughly as described by fundamental physics.17 Russell was a naturalist, broadly speaking, for large parts of his life (he had started out flirting with idealism), and I think that the anguish he describes makes sense only against this background.18 Now it’s obvious that the naturalist universe we inhabit is far from being the best of all possible worlds. There are so many ways in which things could have been better. That’s hardly surprising: it is precisely a key feature of such a universe that it’s not the product of a supreme designer, and not governed by any purpose or value. However, to find the universe around you unsatisfying isn’t yet to be dissatisfied with naturalism itself. After all, very many ways in which things might have been better would still count as naturalist, at least if we understand this term loosely enough.19 A steady state universe with no Big Bang would still be naturalist. The Copenhagen interpretation of quantum mechanics and Everett’s multiverse describe profoundly different cosmic orders, but both are (or can be) naturalist. Moving on to philosophy: behaviourism, the identity theory and functionalism (and even certain forms of panpsychism20 and dualism)21 in philosophy of mind, or ethical naturalism, constructivism, and noncognitivism in metaethics22—these and many other philosophical positions all attempt to describe the actual world, but even if they fail, they still describe possible naturalist worlds— which we may evaluate as more or less attractive than the actual one.23 It should be obvious, then, that to simply ask whether we should prefer theism to atheism, or whether one describes a superior universe to the other, is almost meaningless. Atheism does not refer to a positive metaphysical view, and the different views falling under it have little in common and are likely to wildly diverge in value. Moving from atheism to naturalism narrows down the options considerably, but as we just saw, naturalism can take multiple forms and conceivable naturalist universes will still differ in content and value. We can narrow down our focus to the actual naturalist universe we happen to inhabit—but now we are certainly no longer comparing theism and

If There Is a Hole, It Is Not God-Shaped 99 ‘atheism’. We are comparing the highly specific version of naturalism we find in the actual world and an alternative in which God does exist. Even this narrow comparison remains indeterminate, now in the other direction. Theism can also take very many forms. The deity described in early monotheist texts is distinctly unappealing and is rather different from the God of the Philosophers. In what follows, I will assume a fairly standard if not uncontroversial understanding of theism on which it refers to the claim that there exists an omnipotent, omniscient and omnibenevolent supernatural person who is the creator of the universe (many theists would add that God is a necessary being—a complication I will attend to only at the very end). I will also largely abstract away from the further elaborations of any concrete theist religion. Even so, the theist alternative can still be understood in at least two ways. On one, Russell might be comparing the actual naturalist world with the way the world would have been had a theist like Lady Morrell been right. So we envisage a world that overlaps with ours in many respects (Russell’s biography and prior human history stay more or less the same) but in which God does exist, and e.g., when Russell dies he might go to heaven and ask God why He didn’t provide more evidence.24 Such a comparison may be especially appropriate for an agnostic or wavering atheist since for such a person, both alternatives describe the way things may actually be. But it’s less clear why confident atheists, at least those who take the problem of evil seriously, should focus on this way of understanding the theist alternative. After all, here they would be conceiving of a possibility in which their various reasons for thinking that God exists—including all the suffering and misery around—turn out to be mistaken. But why focus on that as opposed to the very different, and far rosier world, in which God exists and, indeed, prevents all that suffering from occurring in the first place?25

The Axiology of Naturalism Russell seems unhappy with the actual naturalist world, which he compares unfavourably with a theist alternative. To make such large-scale evaluations of the naturalist universe is to engage in what we can call the axiology of naturalism—though our verdict needn’t be so negative, or so desperate. This is a strangely neglected project, especially given that so much of analytic philosophy since the mid-twentieth century (as well as earlier on) has been concerned with offering naturalist accounts of our core common-sense notions—morality, free will, consciousness, intentionality, and so forth.26 Presumably, it matters whether such accounts can be given and what shape they will take. It presumably matters, for example, whether we do have free will and if so, whether we have it in a variety worth wanting27—which is just to say, in a variety that it would be good or better to have, compared to alternatives.28

100  Guy Kahane Sketching what an axiology of naturalism will involve is beyond the scope of this chapter. It should be obvious by now that it can involve far more than comparisons between the naturalist world and theism. There are many relevant comparisons that we can make within a broadly naturalist framework, and even when we step beyond naturalism, there are numerous atheist metaphysical systems that may call on our attention. In fact, it’s far from obvious that theism has a privileged role—or indeed any interesting role at all—in an axiology of naturalism. After all, for a confident atheist, its only salience is historical and sociological. Why should Russell privilege it over, say, the idealist metaphysics of his youth? To the extent that you take theism to be a serious possibility, this higher probability may endow it with greater significance—since it might after all be true that things are better (or worse) in this way. But I take it that for the confident atheist, theism isn’t significantly more probable than McTaggart’s idealism or, for that matter, the Berkeleyian universe described in Borges’s (1962) fiction Tlön, Ukbar, Orbius Tertius (we can set aside spaghetti monsters . . .). This dismissal of theism may seem confused. God, it will be objected, is supposed to be the most perfect being we can conceive. Even if we deny that God exists, what we are denying is the existence of this perfect being. And this analytic point is sufficient to establish the axiological primacy of God within any space of metaphysical possibilities, however wide or wild. Whatever our location in metaphysical space, we have reason to turn our rapt attention to God, whether as actual or, with regret, as merely possible being. I will call this view Superior World Pro-Theism. In its strong form, the claim would be that any theist universe is superior to all atheist ones. For my purposes here, we can focus on the more qualified claim that at least some theist worlds are superior to all atheist ones. There is prima facie plausibility to this view. But I’ll argue that it is mistaken. God may be the perfect being, but His existence doesn’t make for the perfect universe. To put this differently: we shouldn’t assume that theism describes the best of all possible metaphysics.

Ambivalent Pro-Theism Needless to say, many atheists don’t share Russell’s tormented longing for God. But my impression is that (to the extent that they have considered the question) atheists tend to lean pro-theist. For example, Colin McGinn recounts that giving up religious belief was, for him, not a relief but a disappointment. He goes on to explain: I would have liked religion to be true. I’d like it to be true, because I’d like to be . . . I’d like there to be immortality, I’d like there to be rewards for those who have been virtuous and punishments for those who’ve not been virtuous. . . . You know, there’s not . . . there’s no

If There Is a Hole, It Is Not God-Shaped 101 justice in this world and it would be good if there was some cosmic force that distributed justice in the proper way that it should be. (as quoted in Miller 2004) This appears to express McGinn’s overall attitude: it is regrettable that there is no God. Yet he also immediately adds remarks that point in an anti-theist direction: there was some exhilaration too. I mean . . . Russell has a description that I think is kind of appropriate of a feeling of a Godless universe as a kind of exhilarating universe. There’s something hygienic about it. There’s something bracing about it. Whereas the idea that there’s this sort of. . . suffocating presence gazing at your every movement and thought . . . you know . . . and gauging everything you do . . . it’s a bit oppressive to think that way. (as quoted in Miller 2004) We can call McGinn an ambivalent pro-theist. He wants God to exist, but he also recognises that there is a serious downside to having God around—and that a godless world has its distinctive attractions.29 Paraphrased in explicit axiological terms, McGinn seems to hold that God’s existence would be bad in certain important respects yet, nevertheless, the world is overall made considerably worse by His absence, which is therefore regrettable—i.e., the cost of suffocating surveillance and constant assessment is well worth paying if we get immortality and cosmic justice in return. This gives us the following argument for pro-theism: Theism Is Better in Respect (TBR). If God had existed, this would have been better in important respects (because we could enjoy immortality and because cosmic justice would prevail). Theism Is Worse in Respect (TWR). If God had existed, this would have been worse in other important respects (because of the inescapable intrusive judging surveillance). Theism Is Better Overall. The ways in which God’s existence would make things better outweigh the ways in which would make it worse (because, for McGinn, immortality + cosmic justice > no divine surveillance). Therefore, Pro-Theism. We should prefer a world in which God exists to the actual natural universe. It would be hard to deny that the existence of an all-powerful and supremely benevolent being would make the world significantly better

102  Guy Kahane in some respects.30 The world would be a better place if cosmic justice was upheld, and (blissful) immortality does sound nice. Still, not everyone accepts McGinn’s conclusion. Thomas Nagel famously remarked: “I hope there is no God! I don’t want there to be a God; I don’t want the universe to be like that” (Nagel 1997, 130) Nagel is expressing here a strong antitheist sentiment. He is afraid of the possibility that God does exist. Nagel doesn’t elaborate on this remark or try to defend it. His attitude is certainly puzzling: rejecting cosmic justice seems callous, rejecting immortality foolish. In earlier work, however, I have argued that this challenge can be met, and that an antitheist attitude is defensible (see Kahane 2011). My argument made much more of the downside of God’s existence than McGinn does in the passage above. Although it would be hard to deny that a world in which God exists is overall impersonally better than a naturalist alternative, I argued that the impositions that God’s existence would make on our lives can justify a rejection of such a world, at least from our personal perspective.31 As I interpret Russell, McGinn and Nagel, they are all implicitly making what we can call actual world comparisons: they are comparing the actual world, understood in naturalist terms, with a theist alternative— which, as we saw, can be construed as more or as less distant. The opposing verdicts that they give on the basis of such comparisons we can call actual world pro- and anti-theism. The anti-theist argument I shall develop in this chapter will not defend anti-theism in this sense. In fact, although my argument here partly overlaps my earlier argument, it’s an independent argument that is compatible with siding with the answer that Russell and McGinn (and presumably most or all theists) give to the above question. The basic idea is simple. Notice that while the benefits mentioned in TBR are expected upshots of God’s existence, they do not inherently require God’s existence. The Karmic cycle is one mechanism of cosmic justice that could easily also operate in a godless world. Even heaven and hell (if one wants to tie immortality and cosmic justice in this way) could run just fine without a divine overlord. At the same time, God’s existence would entail a continuous intrusion into our privacy, on the standard understanding of the concept of God.32 In other words, you can’t have God without the negative stuff, but you can have the positive stuff without God—and thus without the negative stuff. So the best possible worlds are ones in which we still enjoy the benefits of a Godly world but where God doesn’t exist. That is to say, the most attractive possible worlds are atheist. Call this the Argument for Superior Atheist Worlds. Of course, the actual naturalist world isn’t one of these superior worlds (or even close to being one). And it’s likely that these worlds are supernaturalist atheist worlds. Though, depending on the relevant set of goods,

If There Is a Hole, It Is Not God-Shaped 103 and on how loosely we understand ‘naturalism’, it can hardly be ruled out in advance that we could enjoy many of these goods in a conceivable naturalist universe.

The Argument for Superior Atheist Worlds Let me set out this argument explicitly: (A) If God had existed, this would necessarily make things better in important respects (TBR). (B) If God had existed, this would necessarily make things worse in other important respects (TWR). (C) All (or at least most) of the benefits under (A) could be had without God. Therefore, (D) There are possible atheist worlds that offer all (or most) of the benefits of God’s existence but without the costs. Therefore, (E) Some atheist worlds are overall the best, or among the best, and are superior to all theist alternatives. Therefore, (F) The worlds we should most prefer are atheist worlds. Most of the rest of this chapter will be devoted to defending the premises of this argument. I take it that (A) is not under dispute here—at least not on the assumption that God is supremely benevolent. Premise (C) also seems to me fairly straightforward. Premise (B) is probably the most controversial and will require the most elaboration. As we shall see, it may require some fine-tuning. Both (A) and (B) are claims about the value of aspects of the relevant worlds. Premise (E), by contrast, is a claim about the overall value of worlds. I think that the conclusion in (E) can be made from both impersonal and personal perspectives, where the latter is understood not as referring just to mine or your individual standpoint but to that of all existing persons (that is, all persons who aren’t God . . .). But in order to support (F), it will be sufficient if some atheist worlds are better for us, and if theist worlds are not significantly impersonally better. Finally, in order to address complications due to the possibility of infinite value, (E) states that some atheist worlds are superior to all theist alternatives, even if they cannot be described as overall better.

104  Guy Kahane The conclusion of this argument amounts to what I call superior world anti-theism. This view is based on comparisons that range over the whole space of possible metaphysical views. It claims that the best atheist worlds are superior to the best theist worlds—that these atheist worlds are the ones we should prefer the most. This claim is obviously independent of our answer to the narrower comparison between the naturalist world that we seem to actually inhabit and a theist alternative. It is compatible with both pro-theism and anti-theism in the actual world sense. As we shall see later, one upshot of my argument here is that it puts that narrower comparison in a broader axiological context— and reveals it to be less important than it may seem at first.

Godly Goods Without God Let us start with the claim all (or at least most) of the benefits of God’s existence can be had without God. McGinn mentions cosmic justice and immortality (presumably of the heavenly kind), and these are high on most pro-theist lists. Some would also want to perhaps add a more enchanted, meaning-laden world, (true) mystic union with the universe as a whole, a harmonious, supernatural order, and so forth. And those who think that if God doesn’t exist and our universe is a naturalist one then nothing will have value, or that there could be no objective morality or meaning to our lives, or that we couldn’t have free will, will presumably add these to their list—though it’s far from obvious that naturalism has these implications. The point, though, is that it’s hard to see why God is needed for any of these goods. Again, cosmic justice and immortality in no way require a god—in fact, it is easy enough to conceive of thoroughly naturalist worlds that would offer these benefits.33 And even if you thought that value or objective morality or libertarian freedom are incompatible with naturalism, this doesn’t yet mean that they require God—there are numerous possible worlds that are godless but not naturalist.34 It might be objected, however, that the above list of divine benefits is too short. This objection can be made both at the level of impersonal value and that of personal value. At the level of impersonal value, there is God’s own value: if God is a perfect being then surely it follows that He must be of immense, even infinite value?35 Notice though that even if God is the most perfect being, and perfectly morally good, this doesn’t immediately entail a claim about the value that God adds to a world. But I agree that it would be hard to deny that God would possess immense value, a value obviously be missing from any atheist world. I’ll consider below how this bears on verdicts about the comparative value of theist and atheist worlds. Now few of those who yearn for God do so because they worry that His immense value is missing from our world. But it might be objected

If There Is a Hole, It Is Not God-Shaped 105 that there are fundamental personal goods that do inherently involve God: e.g., having a personal relationship with a supreme being who is also one’s creator, and the creator of the universe, and worshipping and being faithful to such a being, etc.36 It’s not surprising that theists consider these to be great goods, but I doubt that they will be attractive to many atheists. I don’t myself find the idea of such a relationship with a supreme being so appealing, and I see little plausibility in the idea that “final and perfect happiness can consist in nothing else than the vision of the Divine Essence” (Aquinas 2006, 1103; unlike Russell, I wasn’t brought up in a Christian school). By contrast, pro-theists who aren’t also devout believers tend to cite the afterlife and cosmic justice, as McGinn does. Freud, for example, wrote that “it would be very nice if there were a God who created the world and was a benevolent providence, and if there were a moral order in the universe and an afterlife” (1990, 215) and both Mill (1874) and Sidgwick (1907) hoped that there is such an afterlife. The physicist Steven Weinberg similarly writes: The more we reflect on the pleasures of life, the more we miss the greatest consolation that used to be provided by religious belief: the promise that our lives will continue after death, and that in the afterlife we will meet the people we have loved. (Weinberg 2012, 243) Theists who defend pro-theism also often appeal to such goods37— which are also included, for example, in the ‘good news’ that Christianity is supposed to deliver. But these goods are surely God-independent. So if the theist presses these supposed God-dependent goods, the atheist can just shrug in reply. In any event, it seems to me that even these goods don’t ineliminably involve God—as opposed to someone who is immensely powerful, knowledgeable, and nice.

Worse in Respect: The Example of Privacy If we could have the benefits of God’s existence without God, then this should be enough to make it permissible to prefer a godless world that contains these benefits over a Godly one; after all, the two worlds would be of broadly equal value. It’s the second premise of the argument that is supposed to support the stronger claim that such godless worlds are superior to Godly ones. To remind you, that premise was Theism Is Worse in Respect (TWR). If God had existed, this would have been worse in important respects.

106  Guy Kahane The next few sections will be devoted to defending this premise. Since TWR was also a key premise in my earlier argument for ‘actual world’ anti-theism, what I say here should also bolster that other argument. If God is a supremely good being, indeed the perfect being, how could His existence make things importantly worse in any way? To see how, it might be useful to start with an example already mentioned, that of loss of privacy. I quoted above McGinn’s remarks about a ‘suffocating presence gazing at your every movement and thought . . . and gauging everything you do’ (Miller 2004). And Sartre tells the story of how, as a child, he last felt the presence of God as a judging gaze observing him illicitly playing with matches in the bathroom, leading Sartre to respond with outrage at God’s ‘indiscretion’ (Sartre 1964, 102). This is a common anti-theist concern, but I must emphasise that I start with it not because it’s the strongest anti-theist concern—whether in force or plausibility— but because it’s the clearest. Some theists acknowledge this concern. Plantinga writes that “many people thoroughly dislike the idea of an omnipotent, omniscient being monitoring their every activity, privy to their every though, and passing judgement on all they do or think” (Plantinga 2000, 195). Elsewhere he remarks that ‘some will find in theism a sort of intolerable invasion of privacy: God knows my every thought, and indeed knows what I will think before I think it’ (Plantinga 2012), and he points to an ancient expression of this discomfort in the Old Testament: Before a word is on my tongue, you know it completely, oh Lord. . . . Where can I go from your Spirit? Where can I flee from your presence? (Psalm 139:4) To see how arguments for TWR work, it might be useful to spell out the argument from privacy more explicitly. It has two premises, one ‘theological’ and one evaluative: Theological. If God exists, then, necessarily, He will know all our innermost thoughts, feelings, and desires, as well as everything else we do. Evaluative. Having no privacy is inherently bad.38 And TWR pretty much follows. I don’t expect many to dispute the ‘theological’ premise. It just follows from omniscience, one of God’s essential attributes on the standard conception. It’s important here that if God exists, then He necessarily knows everything. Knowing our innermost thoughts and feelings isn’t something He chooses to do—or that He could choose not to do.39 It’s just the way things are—necessarily are—in a Godly universe. So this isn’t a case of

If There Is a Hole, It Is Not God-Shaped 107 God allowing some gratuitous evil.40 God isn’t allowing anything, except in the indirect sense I’ll discuss below. Notice also that no evil needs to be involved. I did phrase the evaluative claim in terms of badness, as seems appropriate here, but TWR only requires a comparative loss of good. The evaluative premise is obviously more controversial. It can be defended directly: having each and every of my innermost thoughts and feelings open to another’s view—even if it’s someone perfectly benevolent— strikes me as deeply unwelcome, and as the remarks above indicate, this isn’t an idiosyncratic response. This response is supported by many influential accounts of the value of privacy. On some of these accounts privacy is valuable in itself while on others privacy matters because it is required for something else that is valuable, whether our dignity, or our having control over information about us or over others’ access to us.41 To be sure, God isn’t some ordinary peeping Tom, but the total divine invasion of our privacy will still be regarded as unwelcome on most of these accounts. And many would hold that the privacy of our inner lives is of far greater importance than the mundane privacy of our home or briefcase. And notice that the worry isn’t that God would occasionally peep into our private matters but that, under God, we would have no privacy whatsoever. Now many theists won’t share these worries—this isn’t surprising, since they should be used by now to God’s constant prying presence. But—assuming that they don’t simply reject the value of privacy, or see it as mattering only as a way of protecting us from malevolent parties— I think that the challenge is on those who reject this evaluative claim to explain why we should make an exception for God. I’ll consider one such attempt below.

Worse in Respect: Structure The argument from privacy nicely illustrates how arguments for TWR work. As we saw, their structure is as follows: Theological. God’s existence entails X. Evaluative. X is bad (or worse than X’s absence). Therefore TWR. If God had existed, this would have been worse in important respects. In assessing such arguments, it’s important to distinguish questions about their abstract structure from questions about the evaluative claims they assume. The latter involves substantive moves within axiology and need

108  Guy Kahane to be assessed as such. You may reject these evaluative claims, but that isn’t the same as showing that arguments for TWR are incoherent. The theological premise raises other complications. What is entailed by God’s existence depends on how we conceive of God, and that is obviously a controversial matter. As I wrote above, I’ll be assuming the widely held theist view that God is a supernatural being that is the omnipotent, omniscient and perfectly benevolent creator of the universe. It would be easy, but also uninteresting, to show that the existence of the unpleasant deity portrayed by many monotheistic texts and traditions would have nasty consequences. Conversely, if by ‘God’ we just mean love then our axiological questions aren’t even worth asking. What is challenging is to show how TWR might be true even on the standard theist conception of God as omnipotent, omniscient, and perfectly benevolent. Significant departures from this standard conception need to be justified since they would no longer be concerned with the supernatural being that people actually long for (or fear). And if we radically revise this standard conception, it’s unclear why we should be concerned with God at all—why not directly pursue the project of trying to identify the metaphysical system that is axiologically the best? I stated the theological premise in terms of entailment. This is important since it explains how God’s existence can make things importantly worse even though He’s perfectly good. There is just nothing God can do about X. God is blameless. Job’s wife advises him to curse God and die (Job 2:9). But the badness that God’s existence brings to the world via X isn’t a ground to complain against him—again, the claim isn’t that God is wicked in some way. Now there is admittedly a complication. All the ways that I can think of arguing for TWR involve the existence of people, contingent existences. To make this explicit: Evaluative. X is bad (or worse) for (at least some) persons. So God’s existence would entail X only given the existence of the relevant persons, and that existence is a contingent matter—very much within God’s power. But if so, why should God create beings whose inner privacy He would then necessarily go on to violate? This point does complicate the argument, but it doesn’t block it. We can think of this issue in terms familiar from discussion of the problem of evil. You could say that, in creating the relevant persons, God is allowing them to suffer an evil they don’t deserve. Now theists often hold that for an evil to be permitted by God, it must either be needed to secure a sufficiently significant, otherwise unobtainable good, or to prevent a sufficiently significant, otherwise unpreventable evil.42 It seems to me that the good of creating persons easily meets this condition.43 For surely, a world

If There Is a Hole, It Is Not God-Shaped 109 with persons who endure such badness (or loss of good) would still be far better—within the range of options available to Good—than a world without any persons at all.44 Conversely, if there isn’t such a sufficient outweighing reason to create persons, then a Godly world would be one without persons at all which, besides involving very significant loss of impersonal value, would obviously be worse (or at least no better) for us. Why should we prefer such a world? So it seems the anti-theist has nothing to worry about here.45 Although TWR may be a surprising claim, I believe that it is compatible with standard conceptions of theism. Even if God is supremely good— the best conceivable being—His existence can still make things worse in significant respects. Perhaps there is no plausible evaluative premise that, when conjoined with God’s essential properties, would support TWR. But that is a substantive claim about value, not a denial of the coherence of the above form of argument. Before moving to such questions about value, let me consider one extreme way in which one might attempt to block the argument. It might be argued that if God’s very existence would make the universe worse, then God would respond by committing self-deicide—by ending His own existence.46 Even if we set aside the view that God exists necessarily, or that He is needed to sustain the universe, this suggestion still involves a radical departure from mainstream theism. In any event, the mere fact that God’s existence would make the world worse in some respect is hardly a good reason for Him to remove Himself from the scene. He would perhaps have reason to do so if His existence made the world overall worse compared to His non-existence, a far stronger claim. Indeed, even if His non-existence would make things overall better this won’t immediately give Him reasons to kill himself for us, so to speak (and of course in a much stronger sense than claimed by Christianity)—it would do so only on a consequentialist view. But let’s take the idea of self-deicide seriously for a moment. It would give us something close to the Deist universe where God created the universe and then removed Himself from the scene. Applied to the actual world, we don’t get a universe terribly different from a naturalist alternative—certainly not one that seems a great deal better. Or if it is better—say because God also left behind a working afterlife—then it’s hard to see how it could be interestingly better than the supernatural alternative in which we get the afterlife without an initial Godly push. Indeed, a world in which God, the supremely good being who is our creator, had to vacate the scene for our sake is one that is likely to arouse in us an immense sense of guilt. It doesn’t seem an appealing proposition.

110  Guy Kahane

Worse in Respect: Substance Having argued that it is coherent to hold that God’s existence might make things worse in important respects, I turn now to the substantive question of whether we can identify unavoidable implications of God’s existence—things that would simply follow from God’s essential properties (and the existence of persons)—and which are also bad, or would lead to considerable loss of good, compared to relevant alternatives in which God doesn’t exist. I have already given one example of how one might fill out such an argument, appealing to the essential property of omniscience and to the badness of loss of inner privacy. There are other ways of filling out such arguments for TWR—for example, divine providence might undermine genuine freedom, and God’s existence might mean that our ability to understand the universe is severely constrained—both undesirable implications. The existence of an omniscient God and heaven and hell may also make it practically impossible to act purely for the sake of virtue, and some have even argued that God’s existence is incompatible with morality (Maitzen 2009). It seems to me, however, that the deepest argument for TWR revolves around a set of related concerns relating to the hierarchical character of a theist universe or, in Nagel’s (1997) words, about the idea of a ‘cosmic authority’.47 In such a universe, we necessarily occupy a subordinate position in relation to a being that is vastly superior to us in every respect. That this supreme being is utterly benevolent is beside the point. The worry is about the position we occupy in such a world, not that God might abuse His power over us. This basic idea can be developed in a number of more specific ways. To begin with, I think that the worry about loss of privacy can be assimilated to this larger concern. It isn’t by accident that the idea that our inner lives will be under constant surveillance reminds us of a dystopian totalitarian regime.48 The worry about privacy is generated by God’s omniscience. A further worry is generated by God’s status as creator. If our existence, and the existence of everything that matters to us, is due to God, then it seems we owe Him a debt. It may also mean that it would be wrong for us to do as we will with God’s creation, and perhaps even with ourselves. Many theists also hold that God isn’t only the creator but also continues to sustain everything that exists. That would mean that our momentby-moment existence causally depends on Him, and that we are, in a way, constantly being acted upon by another person (for better or worse, there’s no real solitude in a theist universe). This worry about dependence goes deeper. If God created the universe, and us within it, He presumably did that for a purpose.49 If there is such a cosmic plan, and we (and others) are here to play our part in it, this

If There Is a Hole, It Is Not God-Shaped 111 severely constrains our ability to lead our lives according to our own plan. Even if it is permissible for us to just reject our role in God’s plan (which is far from obvious), that rejection would surely have a cost. In acting in this way, we might be interfering, even if only marginally, in the cosmic plan. To reject one’s role in the divine plan isn’t the same as being entirely free from the pressures of such a plan. Perhaps most fundamentally, if God exists, His moral status would be vastly higher than ours. We would not be equal members in a Kingdom of Ends but somewhere fairly low in the cosmic moral hierarchy. This seems undesirable in itself but is also likely to have concrete normative implications, such as duties to obey God and to worship Him (as the hymn says: ‘Thy way, not my, oh Lord’). There is something deeply undignified in occupying such a subservient position, in surrendering one’s will to that of another, however supremely benevolent.50 If God exists, I should add, we would also be utterly insignificant compared to Him. By contrast, as I’ve argued elsewhere, in a naturalist universe, we have at least a shot at being truly cosmically significant (Kahane 2014). The list could be continued, but I think it conveys the core idea. It obviously ties to the worry that God’s existence is in tension with our autonomy.51 But I find it more useful to describe this cluster of worries as the broader concern that our independence would be severely curtailed if God exists. In such a universe, our ability to shape and govern our own lives, to define our own purposes and stand independent, our own masters, would be dramatically limited. There is a sense in which, in a theist universe, we remain in a child-like state, unable to fully grow up to become completely independent persons who are fully in charge of their own lives and who are the moral equals of all other persons. We can put this schematically as follows: (I) If God exists, then we will necessarily always be in relation to Him as young children are to their parents. (II) Remaining forever in such a child-like state is undesirable. Adam Smith thought there was “nothing more melancholy than the suspicion of a fatherless world” (Smith 2002, VI.ii.3.). But for the antitheist the idea of a divine parent is precisely the problem. The analogy to childhood needn’t be taken too literally. But it does capture, it seems to me, the basic relation between us and God. Indeed, in many respects the analogy doesn’t go far enough—the gap between a child and her parents is not remotely as vast as that between us and God. You might even say that, in a Godly world, the parent-child relation is a faint echo of the relation between God and us. But it’s worth noting that, on common theist views, it would literally be true that we are God’s children, and that any aspiration to complete

112  Guy Kahane independence and maturity is misguided. For example, the Catholic author Peter Kreeft writes that all of us, even in Heaven, are children. And by the standard of the infinite, inexhaustible perfection of God, we remain children forever. Happy children, fulfilled children, but children . . . would we choose the childlikeness of Heaven or the promise of ‘maturity’, of ‘humanity come of age’ in Hell? Will we suffer gladly the blow and shock to our pride that is Heaven’s gift of eternal childhood (thus eternal hope and progress) or will we insist on the ‘successes’ of ‘self-actualization’ that Heaven denies us and Hell offers us? (Kreeft 1990, 28–29) If you remove the references to Heaven and Hell, this is actually a nice statement of the core antitheist concern. Maturity and self-actualisation is precisely what the antitheist chooses—though, importantly, the antitheist needn’t hold that we should attempt to fully realise these values if God exists; that would be futile, even absurd. Of course, Kreeft and other theists think that this is the wrong choice. For them eternal childhood is a wonderful gift. This isn’t surprising. A traditional theist outlook is not especially conducive to placing great value on us being autonomous agents plotting an independent course through the world based on our own conception of the good. It’s not by accident that recognition of the value of privacy, autonomy, and independence historically emerged in the Enlightenment, when the hold of traditional religious ideas started to wane. Kant famously described the Enlightenment as our emergence from ‘self-incurred immaturity’ (Kant 1970).52 The antitheist thinks that in a theist universe our immaturity isn’t only self-inflicted but also God-imposed. As Heinrich Heine wrote, “we are free, and we want no thundering tyrants; we have reached majority and can dispense with paternal care” (Heine 2012, 79).53 These antitheist worries aren’t idiosyncratic concerns marshalled out of mere metaphysical anxiety. They reflect familiar and deep-rooted moral ideas, even if these are applied in an unusual context. The egalitarian idea that there is something deeply unattractive in being in a subservient position—and that we should reject such a position even if it brings with it considerable benefits—is a central moral idea in Western thought from the Enlightenment onwards. It underlies modern conceptions of democracy, and our rejection of slavery, caste, sexism, racism, and other hierarchical moral systems. Think, for example, of our distaste for servility—the wife who follows her husband’s every order, or butler whose entire life is devoted to serving his master—importantly, a distaste that remains even if the master is utterly benevolent and confers great benefits on the servile person. Such servility involves a failure of self-respect, a failure to uphold one’s

If There Is a Hole, It Is Not God-Shaped 113 dignity. This is usually taken to be due to failure to recognise one’s true moral status, a moral status of equality with others (Hill 1973). Now in our relation to God, if God had existed, treating ourselves as vastly inferior to God wouldn’t involve such an error. A kind of servility would, I assume, be appropriate. The antitheist isn’t denying that. The point is rather that they find this a profoundly unattractive state to be in. To discover that God does exist would therefore be, in this respect, a deep disappointment. Related concerns underlie the view that liberty requires non-domination (Pettit 1997).54 It’s not enough to have the negative liberty to pursue a wide enough range of opportunities. The slave of a benevolent master can have such a range of opportunities, and the master needn’t ever restrict this range. It remains the case, however, that the master is in a position of domination vis-à-vis the slave. It’s in his power to take away these options as he wills, whether or not he actually does. It should be obvious that we cannot fully possess such liberty if God exists—we are utterly dominated by Him. God’s supreme benevolence does nothing to alter this fundamental hierarchical relation. (As Frederick Douglass wrote, “it was slavery—not its mere incidents—that I hated” [Douglass 1855, 161].)55 Notice that while a lack of complete inner privacy is directly entailed by God’s omniscience, many of the antitheist concerns about independence admittedly involve less straightforward claims about the normative implications of God’s existence, and then the further claim that these are undesirable. That we are inferior in status to God, or should worship Him, aren’t directly entailed by God’s omnibenevolence or omnipotence.56 They are based in substantive normative claims.57 Still, these normative implications (if not the evaluative claims about their undesirability) are widely accepted by theists, and seem to me plausible. The core antitheist argument for TWR, however, doesn’t require that they are all correct or that they take exactly the form assumed by theists. And it seems to me very doubtful that they are all false. Is it really remotely plausible that we have the same moral status as God—that we are His moral equals?

Four Objections The core argument for TWR rests on a cluster of evaluative claims relating to independence. These claims are based on direct reflection on what a theist world would be like but they are also grounded in familiar and widely held moral ideas—ideas that are central to modern moral thought. Now there is a limit to what can be done to persuade those who reject these evaluative claims. But that is true of any fundamental evaluative disagreement. And, again, it’s not hard to explain why theists wouldn’t place great value on independence. Moreover, some prominent theists do

114  Guy Kahane concede that the implications of God’s existence aren’t entirely welcome. Peter van Inwagen admits that a life “where I am free to live my life according to my own desires” rather than a cosmic purpose is attractive to some people (van Inwagen 2009, 203–204), and Alvin Plantinga remarks on the “the serious limitation of human autonomy posed by theism,” and which he sees as motivating atheism.58 Elsewhere he concedes that Theism severely limits human autonomy. According to theism, we human beings are also at best very junior partners in the world of mind. We are not autonomous, not a law unto ourselves; we are completely dependent upon God for our being and even for our next breath. (Plantinga 2012) Plantinga doesn’t dismiss these concerns and, for all I know, might endorse TWR. There is little I can do to persuade those who place no value at all on independence. But I wish to consider several possible objections that could be made by those who do accept that independence matters. Objection 1 It might be objected first that we do possess a significant degree of autonomy in a Godly world. We are free to align ourselves with God’s will or even reject it, to live our lives according to His plan or to ignore it or even rebel against it.59 Conversely, even if naturalism is true, we would be far from being absolutely independent. Perhaps there is no supernatural being who sustains our existence, but we are fragile and dependent on other people, and our liberty to do as we please is severely constrained by both nature and human law. Now it has been argued that God’s existence is simply incompatible with genuine autonomy (Rachels 1997). But my argument doesn’t depend on any such strong claim. The point is rather that there is an inherent constraint on how independent we can be in a Godly world. This, as I hope to have shown, is not a marginal difference, having a bit less of a good thing. The difference is fundamental. Children have a measure of freedom, but they aren’t independent adults. Indentured servants can have plenty of options to choose from, and can disobey or even rebel, but they aren’t free persons. The hierarchical character of a Godly world is its most fundamental fact, the fact around which life must revolve. There is nothing remotely comparable in the naturalist world. Even slaves and servants are, nevertheless, the moral equals of their masters. And the majority of us do not live lives centred around an all-powerful superior—not, that is, if we

If There Is a Hole, It Is Not God-Shaped 115 aren’t theists. We are dependent on many others, but, again, these others are our moral equals. And the institutions and laws that constrain us still ultimately rest, if they are legitimate, on our joint wills. Finally, even if we must depend on each other, humanity as a collective isn’t constrained by any others’ will. In any event, the comparison made in TWR needn’t be true of all atheist worlds, not even of all naturalist ones. In fact—and this marks an important difference from the antitheist argument in Kahane (2011)— although I think that TWR also applies to the actual naturalist world, it’s enough for the argument for superior atheist worlds if TWR is true with respect to some atheist worlds. In these conceivable worlds, we could be even more independent than we are here—and without an overbearing divine father in the background. Objection 2 The critics might go on the offensive at this point. They might argue that things are actually significantly worse, with respect to independence, in a naturalist universe. To begin with, in a theist universe we will possess full-blown libertarian freedom, yet there is no space for such freedom in a naturalist world. One naturalist who ridiculed the idea of such freedom was Friedrich Nietzsche. But Nietzsche also argued that the idea of moral equality is itself a relic of the religious worldview, a fantasy to be discarded by clear-eyed atheists.60 If Nietzsche were right, then this would puncture the antitheist argument. But few if any accept these claims of Nietzsche (though even if they were correct, the argument would still have force for those of us at the top of his unpleasant hierarchy). As for libertarian freedom, it is not obvious that we cannot have it in a naturalist universe (that we do not in fact possess it), and if we can’t, it’s not obvious that we can have it anywhere—the very idea might be incoherent. It is also unclear whether such freedom is necessary or even important for independence in the sense outlined above. Finally, there is the last point made in reply to the previous objection: even if libertarian freedom was both a genuine possibility and needed for independence, then if we don’t and can’t have it here, we could have it in a counterfactual atheist world (naturalist or not). And that world would be better for independence for any theist counterpart. Objection 3 It’s important not to confuse a concern about independence with a desire for anarchy61 or chaos.62 It may be that some antitheists wish to be a God-like source of everything—the narrator of Dostoevsky’s Notes From the Underground complains about the immutability of 2+2=4. And the extent to which antirealist naturalist accounts of value, morality,

116  Guy Kahane mathematics and the like are correct is an open question (though if you want to be the source of everything you should hope for idealism, or even solipsism, not for naturalism). But desiring everything to depend on you isn’t the same as wanting to be independent. Independence requires a resistant reality external to the will (you need something to be independent of). Otherwise the idea of pursuing one’s own projects and life plan makes no sense. An independent humanity faces the challenge of mastering blind, indifferent nature—as well as its own many deficiencies. There could be no such challenge, or achievement, if nature were merely the projection of our own wills. In fact, it seems to me that the naturalist universe—a mindless universe that allows no magic or (efficacious) prayer—is superior in this respect to a theist one. Objection 4 It might be argued, finally, that limits to our independence (and to our privacy) are a good thing in some cases—and that our subservient relation to God is one such case. God is, after all, our creator, and a supremely benevolent being. We can dismiss the appeal to benevolence. As we saw, the worry isn’t that our superior will misuse or abuse that position. What is unwelcome is the subservient position itself. And appealing here to God’s supreme status doesn’t really address the concern since, in a sense, the concern is precisely about that status. Kraay and Dragos attempt to defend the suggestion that it isn’t “always bad to be morally subordinate to some person’s demands” by pointing out that “children . . . are properly considered moral subordinates to their parents.” They similarly suggest that “some restrictions on privacy are not bad at all, but entirely appropriate and justified”, again giving the example of “certain restrictions that parents may place on their children’s privacy” (Kraay and Dragos 2013, 165). However, you cannot address a concern about forever remaining a childlike subordinate by appealing to what is appropriate for children! Moreover, children do not yet possess the capacity for genuine independence while adult humans do. And childhood is but a stage on the way to maturity; but in a theist universe we remain child-like for eternity. Kraay and Dragos also fail to distinguish here between value and justification. The argument was that the inherent constraints on our independence in a Godly world make such a world worse. The claim isn’t that God would be wrong, or unjustified, in expecting our allegiance—again, our duties of allegiance would simply follow from the fact that we inhabit a world in which God exists, and I’m assuming that God would be justified in bringing us into existence despite that unwelcome imposition. (I should add that past a certain age, impositions on the privacy and will of children seem to me to involve a degree of badness despite being overall justified.)

If There Is a Hole, It Is Not God-Shaped 117 Now I know that some people would choose blissful eternal servitude over a short, and far from blissful, (imperfect) independence. I don’t want, or need, to criticise such a preference. I do want to highlight the considerable costs involved in such a life, and the aim of my earlier antitheist argument was only to show that a rejection of this cost—the rejection of eternal childhood or servitude—is rationally and morally permissible. In any event, this point has no bite against the current argument. If you wish for eternal bliss, you should wish for eternal blissful independence.

Overall Verdicts I’ve argued that God’s existence would make things worse in important respects and that most of the benefits that are supposed to follow from God’s existence can also be had in many godless worlds. Together, these premises suggest that there must be atheist worlds that offer the benefits of God’s existence without the cost—and that these should therefore be the best worlds. But there are number of complications which we need to attend to before we can assert this conclusion. We can start by considering whether this overall verdict can be made at the impersonal level. As we saw earlier, one obstacle is God’s own immense value, which is obviously missing from all atheist worlds. From that immense positive value, we need to deduct the badness or loss of good that God’s existence would also bring about in worlds that contain persons. To that we may also add the badness for God of being the source of all that badness. But I cannot pretend to be able to confidently add these up to an overall verdict. To complicate things further, it might be held that God’s value isn’t merely immense, but infinite. It’s important to see, however, that this won’t in itself tip the overall balance in favour of Godly worlds. Atheist worlds may be spatiotemporally infinite in size and thus potentially contain an infinite amount of value; in fact, we cannot even rule out that the actual naturalist universe is infinite in this way.63 So to claim that God’s value makes theist worlds superior to all atheist ones requires us to rank some infinite worlds as better than others. But even if we could produce such a ranking I don’t see why we must assume that Godly worlds must rank here as superior to all atheist ones. Luckily, the Argument for Superior Atheist Worlds doesn’t require us to resolve these questions. For us to be justified in preferring some atheist worlds to all theist alternatives it’s enough, I believe, if (i) these atheist worlds are not impersonally overall worse in any significant way and (ii) these atheist worlds are overall superior to all theist alternatives from our personal perspective, in the sense of containing the personal benefits of a theist world while avoiding its harms. I think (i) would be hard to deny. The relevant atheist worlds would be incredibly good. They can contain perfect justice, no suffering or other

118  Guy Kahane evils (at least not of the gratuitous kind), eternal blissful life, etc. The only significant thing they lack is God Himself. So even if one insisted that a Godly world is somewhat better, we would not be rejecting, on purely selfish grounds, some incredibly rosy universe in favour of a miserable, unjust one—as we seem forced to do if we try to defend actual world anti-theism. So let me turn to (ii). If, as I had argued, God’s existence makes our lives considerably worse by severely constraining our independence (and potentially for further reasons) while we can get the benefits that would follow from God’s existence in some atheist worlds, then surely our lives in these atheist worlds would be better, and what we should most prefer. Worries about infinite value again complicate things, though now in the other direction. If people enjoy immortality in both of these alternatives, then perhaps they enjoy lives that are infinitely good in each case. So my argument seems to commit me to rank certain infinitely good lives as better than others. However, I think that such a ranking is straightforward here. A life that is eternally blissful and independent is surely superior to one that involves eternal bliss and submission. In fact, it seems to me that we have strong reasons to most prefer the former worlds even if we hesitate to call them better. We can think of these reasons as quasi-deontological in character: we should prefer those worlds where we don’t suffer certain indignities even if these worlds are not overall better for us. Indeed, considerations relating to threats to our autonomy and independence are widely taken to be the primary ground of deontological reasons to act in ways that may be suboptimal, and it’s only natural to extend this to attitudes such as preference. For my argument, I just need the weak claim that such considerations can tip the balance in cases where the competing options are, or may be, of equal value. But if we accept such deontological constraints on attitude then they are likely to extend further, allowing or even instructing us to reject worlds that are axiologically superior but nevertheless compromising. If so, this will greatly extend the range of atheist worlds that would count as preferable to all theist alternatives—arguably, including the actual miserable naturalist world, thereby also giving us a strong argument for actual world anti-theism. But I won’t defend this stronger claim here.64

The Best of All Possible Metaphysical Systems This concludes the Argument for Superior Atheist Worlds. As we saw, it is strongest when understood as a claim about which worlds are best, and most preferable, from our personal perspective. Parallel claims about the overall impersonal value of worlds are harder to resolve one way or the other, though it can hardly be ruled out that some atheist worlds are superior to all theist ones even in the impersonal sense.

If There Is a Hole, It Is Not God-Shaped 119 If this is so, then Leibniz was wrong to claim that God must create the best of all possible worlds: God would be absent from the best worlds, and God obviously can’t create those worlds that weren’t created by God . . . (Leibniz’s claim will hold only narrowly: God must create the best world out of those that He can create.) It’s important not to misunderstand my aim here. This isn’t a niceness competition between theism and atheism (and again, there is little in common between naturalism and some of the more fantastic atheist metaphysical systems). The point is that if you feel dissatisfied with what the actual naturalist universe has to offer, if you feel deprived of something further, then you should think seriously of what the best (or at least more satisfactory) universe would look like. But God needn’t figure prominently in such reflections, let alone be assumed to figure in their outcome. To assume that is to dramatically underestimate the relevant range of options.65 Indeed, why expect that the ancient musings that are codified in the main world religions—musings shaped by factual ignorance, social utility, and psychological need, and by evaluative and moral assumptions we now find deeply wrong, often repugnant—not to mention, musings aiming to offer a true view of the actual world—be any kind of deep (let alone conclusive) guide to what might be best? This is a question we would do well to sever from the question of what is actually the case. Now religious ideas can certainly provide material for reflection here— if, for example, the idea of heaven is an attempt to imagine a perfect form of life, then it is useful input. But the idea that we first need to go through a vale of sorrows in our mortal life before going to heaven seems to me to have little plausibility except as a feeble way of addressing the obvious point that the actual world clearly isn’t anything like heaven . . .

Two Kinds of Anti-Theism I have argued that that there are some atheist alternatives that are superior to all theist ones (superior world anti-theism). This conclusion is compatible with holding that a theist alternative is superior to the actual naturalist world (actual world pro-theism). But it makes that view rather uninteresting, at least for a confident atheist. After all, there are tons of alternatives that are better in various ways from the actual world—not to mention, tons of alternatives that are far worse. Except for historical, sociological, and perhaps dialectical reasons, there is no special reason to single out the theist alternative. It’s not the best, or what we should most prefer. And compared to many atheist alternatives, it brings with it considerable costs—costs that make it significantly worse even compared to the actual world.66 Even if you prefer the theist alternative to the actual naturalist world, it hardly follows that you want God to exist. Indeed, you may dread this

120  Guy Kahane prospect. (Compare: if you prefer a toothache to a broken leg, this hardly entails you want to have a toothache!) What you should really want is for one of the superior atheist alternatives to be true. In a perfectly good sense, you should still want God not to exist. But it’s enough for my purposes that you have powerful reasons not to want God to exist.67 Moreover, to the extent that this argument is successful, then it has the implication that theists should be anti-theists—not in the sense of wishing for the godless naturalist universe of Dawkins and Hitchens, but of wishing for these superior godless alternatives. Conversely, if you hold that we should prefer the actual naturalist world to a theist alternative, this hardly means that you’re in any way satisfied with the way things are. You don’t want a world governed by a God. But you may still feel that something is missing in this world. You may wish that we had inhabited one of these superior atheist worlds instead. Just as atheism isn’t synonymous with naturalism, anti-theism certainly needn’t commit us to pro-naturalism.68

If God Is Impossible The question of whether we should prefer a naturalist version of the actual world to a theist version, or even more generally the truth of theism to some naturalist alternative, is plagued by a now-familiar modal problem.69 If God is a necessary being, there is no atheist alternative to theism. If the very idea of God is incoherent, there is no theist alternative to atheism.70 But does it make sense to prefer (or disprefer) what is anyway impossible, perhaps even incoherent? Now, both theists and atheists routinely make comparisons between theist and atheist alternatives. For example, William Craig declares: “in a universe without God, good and evil do not exist—there is only the bare valueless fact of existence” (Craig 2008, 75), while Dawkins asserts that “[y]ou can’t escape the scientific implications of religion. A universe with a God would look quite different from a universe without one.”71 If such comparisons make no sense then neither does much philosophy of religion. It would be better, of course, to have a principled account of what is involved in such comparisons.72 And perhaps the problem doesn’t even arise: plenty of atheists regard the idea of God as at least coherent, and some prominent theists deny that He is a necessary being.73 But here I just want to briefly relate this issue to the argument of this chapter. Since I’m not a theist, I’ll focus on what would follow if theism isn’t a genuine possibility.74 It’s important to distinguish several issues here: (1) Are the alternatives being compared both genuine metaphysical possibilities? (2) If one is impossible, can we intelligibly ascribe value properties to, and within, such ‘impossible worlds’? (3) Even if we can ascribe such value properties,

If There Is a Hole, It Is Not God-Shaped 121 do such impossibilities matter? That is to say, are they fitting objects for attitudes such as fear, relief, or even just bare preference? But suppose that these challenges can’t be met and that theism isn’t even conceptually possible or describes a metaphysical impossibility to which we cannot ascribe value or that such impossibilities don’t matter. This is obviously not something atheists need to worry about—for one, if this is correct, then they needn’t worry about arguments purporting to show how bleak a naturalist universe is compared to the theist alternative. And such a conclusion would do no damage to the axiology of naturalism. No one seriously thinks that naturalism is necessarily true, let alone that the actual universe had to be that way. As we have seen, there is a vast space of possible atheist worlds, both natural and supernatural, to consider. Nor is there anything incoherent in the idea of cosmic justice or immortality. Of course, to the extent that we cannot intelligibly compare a theist universe to various alternatives, then many of the claims made in this chapter need to be reconfigured. But the main point could then be stated in even stronger terms: to the extent that people find the actual naturalist universe grim, or that they deeply regret that God doesn’t exist, then they are doubly mistaken. It’s not just that they are longing for the wrong thing, as I have argued, but for something that it makes no sense to long for. To the extent that they should wish that things had been different, they should focus not on theism but on those atheist counterfactuals in which we would have enjoyed the benefits associated with God’s existence. We arrive at the same conclusion; it’s just that the argument is much simpler.

Conclusion Commenting on Nagel’s remark that he doesn’t want God to exist, Plantinga writes: “here we have discomfort and distress at the thought that there might be such a being as God; but this discomfort seems more emotional than philosophical or rational” (Plantinga 2012).75 But Nagel may have good reasons for wanting God not to exist. What attitude we take to the possibility that God exists isn’t merely a psychological fact about us. Anti-theism and pro-theism require philosophical and rational justification, which must revolve, in large part, around claims about the difference in value made by God’s existence—I am assuming here, and Nagel and Plantinga will surely agree, that such evaluative claims aren’t themselves merely expressions of emotions. In debating such claims, we may eventually reach deadlock, a point where an axiological gulf divides us from those who disagree with us. But that can happen with any disagreement about value. When pro-theism is defended, this is usually by reference to the various goods that would follow from God’s existence—immortality, cosmic

122  Guy Kahane justice, meaning, etc. In this chapter, I have argued that if you want God to exist because of such extrinsic goods then what you should really want is a godless world that contains these goods without the adverse implications of God’s existence. I’ve developed this argument from a naturalist, atheist standpoint. But I think it has disturbing implications for theists. It supports the conclusion that theists have strong reason to find it regrettable that God exists—and that they have this reason even if actual world pro-theism is successful. The axiology of theism leads to blasphemous conclusions. . . I suspect it is anyway somewhat blasphemous for a theist to want God to exist simply because of the extrinsic benefits His existence would bring about. Shouldn’t the devout believer want God’s existence for its own sake, because she wants God to exist rather than because she fancies immortality? Perhaps even: regardless of what benefits God may bring about, indeed, even if His existence would make things worse. The protheism of theists should be pure and unquestioning . . . but this comes close to saying that theists (and perhaps also those who seriously think that God might actually exist) shouldn’t even raise these axiological questions. Perhaps the axiology of theism is permissible only to the unwavering atheist . . . Naturalists don’t face this problem. The indifferent and pitiless universe around us isn’t something we can disrespect, let alone blaspheme against. Still, as I said earlier, we cannot read what attitudes we should have, overall, simply by considering the value of options. Even if we recognise that things could have been better in a zillion ways—that we have reason to prefer a supernatural (if atheist) alternative to the naturalist universe we inhabit—it doesn’t follow we should dwell on these pale possibilities. Such daydreaming can seem immature, and may even be a kind of vice: a failure to pay proper respect, not, of course, to the miserable universe itself, but to the things in it that do possess value.76 Whether this is so, however, depends on a question I have not addressed here: on whether there is such a normative ‘hole’ in our universe or lives. Our world is rather miserable, and the evil that motivates the problem of evil doesn’t go away when God ‘departs’. Our lives may be lacking in some fundamental way. Just as it isn’t wrong for the oppressed citizens of a dystopia to reject the remaining glimmers of good, then we would also be entitled, if this is the case, to be dissatisfied with the universe we find around us.77

Notes 1. 23 October 1916. Cited in Monk (1996, 316). 2. 27 December 1911. See Griffin (2013, 394). Nagel (2009) includes Russell in the list of analytic philosophers who lack what Nagel calls ‘religious temperament’. As this letter shows, that isn’t quite right.

If There Is a Hole, It Is Not God-Shaped 123 3. Russell sometimes described himself as an agnostic, but by this he just meant that he didn’t have proof that God doesn’t exist. He claims to be similarly agnostic about the Homeric gods, and to regard neither to be “sufficiently probable to be worth serious consideration”. See Russell (1950/1997, 92). 4. See Kahane (2011); Rescher (1990) calls a similar view ‘axiological theism’. Notice that I understand pro-theism and anti-theism to be claims about attitudes to God’s existence, not about its value. While what we are justified in preferring, regretting, or fearing often reflects differences in the balance of value, the former needn’t be a simple function of the latter. Some responses to my article have understood these terms in straight axiological terms, ignoring this further step between value and attitude. 5. Of course, for the theist the explanation goes in the other direction. As Augustine wrote, “you made us for yourself, O Lord, and our heart is restless until it rests in you” (Augustine 2008, 3). 6. Beckett attributes the desolation to the discovery that we are “a speck of dust in a vast universe” (See Buttner 2002, 115). A. S. Byatt describes “a sense of lack which becomes intolerable. If there is not more love in the world than human beings supply, life is unlivable; if life has no purpose or meaning beyond work, charity and the acquisition of knowledge, one is better dead; if there is no authority to dictate, judge, punish or forgive our acts, our lone moral responsibility becomes so haphazard and at the same time so strenuous that we cannot act” (Byatt 1967, 71). See also Mavrodes (1986). 7. A E. Housman had earlier complained about the ‘heartless witless nature’ (1922/2010, 132) and Nietzsche spoke of a “de-deified world that has become stupid, blind, crazed and questionable” (Nietzsche 2001, 220–221). 8. See Nagel (1971). I elaborate this point in Kahane (2017). 9. This isn’t to deny that some atheists do worry about these supposed nihilist implications of God’s absence. Camus may be one example and, following Dostoevsky, Sartre (2007) claims that we humans are ‘forlorn’ because in the absence of God everything is permissible. But I think this is confused. See Kahane (2017). 10. See Kahane (2014). 11. We can understand what is missing in terms of quantity: our brief, discontented mortal life is obviously less than an immortal blissful one. But we might also be missing certain kinds of goods that are lexically superior to what’s around us—perhaps sacred beings and (true) mystical visions. 12. Whatever one makes of the fine-tuning argument, our universe is at best finetuned to allow for a tiny bit of value, perhaps only in some miniscule remote corner, as opposed to numerous naturalist alternatives that would have contained none. It is most certainly not especially fine-tuned to promote value, let alone to maximise it. 13. Nagel describes a similar view: “Outrageous as it sounds, the religious temperament regards a merely human life as insufficient, as a partial blindness to or rejection of the terms of our existence. It asks for something more encompassing, without knowing what that might be” (Nagel 2009, 6). Notice that we can make parallel claims about meaning. While it’s similarly implausible that in the absence of God there can be no meaning in human life (cf. Wolf 2010), and Russell’s life certainly seems meaningful, one may nevertheless hold that naturalist lives don’t have enough meaning. 14. Since modernity involved both a shift in metaphysical view and a radi cal transformation of our form of life, such dissatisfaction may be due to the latter at least as much as to the former. Perhaps the so-called ‘hole’ is tribe-shaped. . .

124  Guy Kahane 15. For further discussion of the idea that something fundamental might be missing in a naturalist universe, see Taylor (2007); Habermas (2010). 16. Some of these views can admittedly be interpreted as highly non-standard forms of theism: Neo-Platonists believe in ‘the One’, Idealists in the Absolute, and Hegel had his peculiar conception of God. As will be explained below, I will understand theism here along more orthodox lines. 17. The identification of atheism with ‘materialism’ goes a long way back. Cudworth already wrote that “all Atheists are mere Corporealists, that is, acknowledge no other Substance besides Body or Matter” (Cudworth 1678: I, iv, 187). 18. See especially Russell (1903). 19. I will not offer any very precise definition of ‘naturalism’ but I will understand the view to involve a metaphysical claim about the kinds of things that exist—and the kind that don’t. But for my purposes, it doesn’t matter if these metaphysical claims depend on some prior methodological view. I’ll also just assume that attempts to interpret theism in naturalist terms are really just atheism in disguise. 20. See Strawson (2008). 21. See Chalmers (2007). 22. To the extent that certain forms of non-naturalism about normativity and mathematics don’t have ontological commitments, as some claim, then they might also be compatible with metaphysical naturalism (Cf. Parfit 2011; Scanlon 2014). 23. Our present inability to resolve most of these debates adds a further limit to our ability to confidently speak about the overall value of the actual universe. 24. For further discussion of how to state such a comparison more precisely, see Kraay and Dragos (2013). 25. See Kahane (2011, 680, fn. 11). When we compare worlds in this way, we might be concerned with the impersonal comparison of their overall value. Or we can make these comparisons from the personal perspective, asking which of the relevant possibilities would be (or would have been) better or best for us. Needless to say, the latter comparisons only make sense if we exist in both possibilities. Spelling out the conditions under which we can make this assumption with any degree of plausibility is beyond the scope of this chapter. But it should be obvious that we can make personal comparisons only of a small subset of the total space of atheist and theist worlds. 26. See, e.g., Jackson (1998). 27. Dennett (1984). Within specific debates, there’s certainly work that makes an important contribution to the axiology of naturalism. But this question is rarely treated generally or systematically. 28. Kahane (2014). 29. It’s not clear whether McGinn thinks that a godless (naturalist) universe is ‘hygienic’ and ‘bracing’ simply because it frees us from the bad aspects of theism or because of its further positive features—features not to be found, for example, in many supernatural atheist metaphysical systems. 30. See Kahane (2011). That said, antitheists may well reject the specific advantages that McGinn finds in a theist universe. Many find the idea of divine punishment unpleasant, and one may find, with Bernard Williams (1973), the prospect of immortal life unappetising. 31. For responses, see Mawson (2012); Luck and Ellerby (2012); Kraay and Dragos (2013); Kraay (2013); Penner (2015); Dumsday (2016); Penner and Lougheed (2015). For criticism of Nagel, see McLean (2015). 32. By contrast, while some form of cosmic justice is probably entailed by God’s existence, this is less obvious for immortality—so far as I can see, theists believe in immortality on the basis of revelation; it’s not a simple entailment

If There Is a Hole, It Is Not God-Shaped 125 of God’s omnibenevolence; perhaps, for example, we humans possess greater value to God as more ephemeral beings. 33. Klaas Kraay has suggested to me that the character of these goods may be different, and more desirable, in a Godly world. Cosmic justice, for example, has a different character when meted out by God than by an impersonal force that cannot blame or forgive. But while these goods may be different in character in a Godly world, this difference doesn’t seem to me importantly better. And if it is, we may be able to get it without God—for example, cosmic justice can have its noumenal source in the collective verdict of all good persons. 34. While I can’t see any argument that would exclusively tie free will or meaning to a divine source as opposed to a supernatural one, views such as the Divine Command Theory do attempt to do so for morality or even value. There is obviously no need to rehearse here the arguments against such a view, arguments that many theists accept. In addition, it needs to be shown that some non-divine supernatural source can’t do the relevant work. 35. See Kraay and Dragos (2013, 168); MacLean (2015) and especially Penner and Lougheed (2015). 36. See Kraay and Dragos (2013, 168). 37. See, e.g., Kraay and Dragos (2013); Davis (2014); MacLean (2015). Adams’s (1979) moral argument for theism also appeals to the idea of a cosmic moral order. 38. To simplify things, I’ll ignore here the distinction between personal and impersonal value. I’ll assume here and below that the badness involved would be both personal and impersonal. 39. But can’t God restrict His knowledge in order to respect our privacy? It’s rather unclear, however, what picture of divinity we get if we permit this move. If God needs to find out about our inner states only from publicly available information (which exactly?) as well as refraining from using His perfect intellect to still precisely infer our inner states even from that limited information, then we get a pretty dramatic revision in the standard theist picture of God’s relation to His creatures. Now the standard picture may just be wrong. But I just want to highlight how revisionary such a move will be. 40. Contra Kraay and Dragos (2013). 41. For such accounts, see Parent (1983); Gavison (1980); Moore (2003); Bloustein (1964). 42. See Kraay and Dragos (2013). 43. Kraay and Dragos (2013) seem to me to conflate this fairly limited point with the very different claim that God must necessarily compensate for this loss of good compared to a world in which God doesn’t exist and persons don’t have to endure such invasion of privacy. The claim that God won’t allow gratuitous evil—which operates within a world—in no way entails this much stronger claim. 44. So the full argument really goes like this: (1) God’s existence entails X. (2) X is bad (or worse) for (at least some) persons. (3) The badness of X is outweighed by the value of the existence of persons (and what would logically and causally follow from that existence). (4) If God exists, He will create the relevant persons, or allow them to come to exist (follows from (3) and God’s essential properties). Therefore, (5) TWR. If God had existed, this would have been worse in important respects.

126  Guy Kahane 45. Notice also that if this claim is correct, then the anti-theist argument turns into an instance of the problem of evil when applied to the actual world: since we do exist, God probably doesn’t. 46. See Kahane (2011). 47. This is a common antitheist concern, though it’s often stated in unhelpful ways. For example, although Russell longed for God, he was also a profoundly ambivalent pro-theist, writing that “[t]he whole conception of God is a conception derived from the ancient Oriental despotisms. It is a conception quite unworthy of free men” (Russell 1957, 23). Christopher Hitchens (2001, 64) similarly describes the theist heaven in unflattering terms, a place of “[e]ndless praise and adoration, limitless abnegation and abjection of self; a celestial North Korea.” Elsewhere Hitchens explains that he doesn’t want theism to be true because it would amount to a “permanent, invigilated, regulated dictatorship which you are told is for your own good” (Brown 2011). And in a discussion with Nasrallah, Julian Assange asked “shouldn’t you as a freedom fighter also seek to liberate people from the totalitarian concept . . . of a monotheistic God?” (on Assange’s RT show The World Tomorrow, 17 April 2012). These claims all seem to express the antitheist concern I’ll be developing. But they fail to distinguish concern about the existence of a supremely benevolent being from far more obvious concern about some petty, vindictive deity. 48. See Bloustein (1964) for a relevant attempt to ground the value of privacy in Kantian concerns about dignity. 49. The Baltimore Catechism teaches that “God made me to know Him, love Him, and serve Him in this life and to be happy with him for ever in the next” (Anonymous 2010, 8). 50. C. Stephen Evans writes that if God exists, then our relation to Him “carries with it distinctive obligations. . . . A proper social relation with God is one that requires humans to recognise the enormous debt of gratitude they owe to God, as well as the value of an on-going relation to God. Most religious believers have seen this relation to God as one in which God rightly has authority over them.” This authority, Evans adds, might even be due to “God’s ownership rights as creator”! (Evans 2013, 28; cited in Penner 2015, 332). 51. Rachels (1997). Rachels deploys this concern in an interesting but ultimately unpersuasive argument against God’s existence. But I think Rachels’ claims have more force when interpreted in antitheist terms. For another argument that has strong affinities with the antitheist argument I’m developing here, see Johnston (2011, 50–52), though Johnston goes on to defend a form of panentheism that is supposed to be compatible with metaphysical naturalism. 52. Concern about independence has an admittedly Kantian flavor, and Kant himself held that “kneeling down or groveling on the ground, even to express your reverence for heavenly things, is contrary to human dignity.” Kant (1994, 99; cited in Rachels 1997, 109). 53. Heine, however, was a pantheist idealist rather than a naturalist in the contemporary sense. 54. Similar concerns can be made in terms of ‘externalist’ accounts of autonomy. See Oshana (1998). 55. Lemos (2017) has relatedly argued that if theism is true, then we humans are treated by God merely as means to some good end—a conclusion that Lemos sees as supporting a kind of Kantian antitheism. 56. Though if Rachels (1997) is correct that it’s constitutive of God that we have a duty to worship Him, then this would simplify my argument. For an argument against such a duty, see Bayne and Nagasawa (2006).

If There Is a Hole, It Is Not God-Shaped 127 57. Kraay and Dragos (2013, 164ff) also highlight this point. But I think they fail to see that the same applies to many familiar theist claims. They describe the view that God won’t allow gratuitous evil as entailed by God’s essential properties—but that claim similarly involves a substantive (if less controversial) unpacking of what it would mean to be omnibenevolent. Setting aside revelation, standard theist claims about cosmic justice or the afterlife are also not mere logical entailments of God’s essential properties. 58. In Gutting (2014). 59. Cf. Kraay and Dragos (2013); Davis (2014); Penner (2015); MacLean (2015). 60. See e.g. Leiter (forthcoming). 61. Kraay and Dragos (2013, 166). See also Penner (2015). 62. van Inwagen (2009). 63. See Bostrom (2011) for discussion of this possibility as well as of obstacles for claiming that one world containing infinite value is better than another. 64. A third line of argument would claim that the evils we incur if God exists (or goods we lose) are lexically more important than even infinite amounts of ‘lower’ goods such as bliss. 65. The idea that we should explore alternatives to both (mainstream) theism and (hard) naturalism has been recently defended by others—see, e.g., Schellenberg (2016); Nagel (2012). But these authors hold that some such alternative might actually be true. My claim is rather that such alternatives describe the best worlds we can conceive; it’s incredibly unlikely that our world is anywhere close to being the best. 66. Given that McGinn started out a theist, it was reasonable for him to feel disappointed when discovering that the world is considerably worse than he thought it was—even though it was also better in important respects. The question is what, now that (as I assume) McGinn is a confident atheist, he should feel about naturalism and theism. My argument is that he needn’t pay much attention to theism, axiologically speaking. To the extent that regrets the way things are, or reflects on how they could have been better, he should focus on atheist alternatives. There is no reason to find God’s absence regrettable. 67. As I indicated several times, these considerations shift if we treat the truth of theism as a serious option. That could justify singling it out within that large space, despite it being inferior to some atheist possibilities—if these aren’t taken to be serious options. 68. Nagel himself expresses both anti-theist and anti-naturalist sentiments—it’s clear that he wants neither to be true. Since he also appears to deny that ‘hard’ naturalism succeeds in correctly describing the actual world, it’s not obvious whether he’d prefer the truth of hard naturalism over that of theism. 69. See Kahane (2011; 2012). See also Mugg (2016). 70. Martin and Monnier (2003). 71. In a debate with Archbishop Habgood, 1992. Cited by Johnston (2011, 46). 72. For discussion, See Kahane (2011); Kraay and Dragos (2013). 73. See, e.g., Swinburne (1994). 74. See Findlay (1955). 75. The assumption that disagreements on such matters are no more than temperamental differences is common. Peter van Inwagen (2009, 203–205) suggests that either ‘chaos’ or ‘logos’ can be emotionally attractive for certain people. Peter Godfrey-Smith (2013) similarly writes, in a review of Nagel’s book, that “Darwinism offers a view according to which the evolution of awareness and reason is, in a broad sense, accidental. Some will respond by hoping for more, for a universe in which we are supposed to be here. Others might find that our deep contingency brings with it a peculiar sense of freedom”. And Adam

128  Guy Kahane Gopnik (2014) remarks, of the wildly different responses of major poets to the absence of God, that “in the end, these seem questions more of temperament than of argument.” 76. Nietzsche held an extreme version of this view, rejecting any transcendent otherworldly longing as life denying, and asking us to unwaveringly affirm the actual. For a much milder version of this claim, see Calhoun (2017). 77. This chapter was presented at the philosophy of religion work-in-progress group at Ryerson University. I’m grateful to the participants for useful suggestions, and to Klaas Kraay and Kirk Lougheed for extremely helpful written comments.

References Adams, Robert, 1979. “Moral Arguments for Theistic Belief.” In Rationality and Religious Belief, edited by Cornelius Delaney, 116–140. Notre Dame: University of Notre Dame Press. Anonymous. 2010. Baltimore Catechism Two. Charlotte, NC: TAN Books. Aquinas, Thomas. 2006. Summa Theologica. Raleigh, NC: Hayes Barton Press. Augustine. 2008. Confessions. Translated by Henry Chadwick. Oxford: Oxford World’s Classics. Bayne, Tim, and Nagasawa, Yujin. 2006. “The Grounds of Worship.” Religious Studies 42: 299–313. Bloustein, Edward J. 1964. “Privacy as an Aspect of Human Dignity: An Answer to Dean Prosser.” New York University Law Review 39: 962–1007. Borges, Jorge Luis. 1962. “Tlön, Uqbar, Orbis Tertius.” Chapter 1 in Labyrinths. London: Penguin Books. Bostrom, Nick. 2011. “Infinite Ethics.” Analysis and Metaphysics 10: 9–59. Brown, Mick. 2011. “Godless in Tumourville: Christopher Hitchens Interview.” The Telegraph, March 25. Büttner, Gottfreid. 2002. “Schopenhauer’s Recommendations To Beckett.” Samuel Beckett Today/Aujourd’hui 11: 114–122. Byatt, A. S. 1967. The God I Want, edited by James Mitchell. London: Bobbs Merrill. Calhoun, Cheshire. 2017. “On Being Content with Imperfection.” Ethics 217: 327–352. Chalmers, David. 2007. “Naturalistic Dualism.” In The Blackwell Companion to Consciousness, edited by Max Velmans and Susan Schneider, 359–368. Oxford: Blackwell Publishing. Craig, William Lane. 2008. Reasonable Faith (3rd Edition). Wheaton, IL: Crossway Books. Cudworth. Ralph. 1678. The True Intellectual System of the Universe. London: Printed for Richard Royston. Davis, Stephen T. 2014. “On Preferring that God Not Exist (or that God Exist): A Dialogue.” Faith and Philosophy 31: 143–159. Dawkins, Richard. 2008. River Out of Eden: A Darwinian View of Life. New York, NY: Basic Books. Dennett, Daniel C. 1984. Elbow Room: The Varieties of Free Will Worth Wanting. Cambridge, MA: MIT Press. Douglass, Frederick. 1855. My Bondage and My Freedom. New York and Auburn: Miller, Orton and Mulligan.

If There Is a Hole, It Is Not God-Shaped 129 Dumsday, Travis. 2016. “Anti-Theism and the Problem of Divine Hiddenness.” Sophia 55: 179–195. Evans, C. Stephen. 2013. God and Moral Obligation. Oxford: Oxford University Press. Findlay, John N. 1955. “Can God’s Existence Be Disproved?” In New Essays in Philosophical Theology, edited by Antony Flew and Alasdair MacIntyre, 49–55. New York, NY: Palgrave Macmillan. Freud, Sigmund. 1990. The Future of an Illusion in Civilization, Society and Religion, Penguin Freud Library, Vol. 12. London: Penguin. Gavison, Ruth. 1980. “Privacy and the Limits of Law.” Yale Law Journal 89: 421–471. Godfrey-Smith, Peter. 2013. “Not Sufficiently Reassuring.” London Review of Books, January 24. Gopnik, Adam. 2014. “Bigger than Phil: When Did Faith Start to Fade?” The New Yorker, February 17. Griffin, Nicholas. 2013. The Selected Letters of Bertrand Russell, Volume 1: The Private Years 1884–1914. London: Routledge. Gutting, Gary. 2014. “Is Atheism Irrational?” New York Times, February 9. Habermas, Jurgen. 2010. An Awareness of What’s Missing: Faith and Reason in a Post-Secular Age. London: Polity Press. Heine, Heinrich. 2012. Religion and Philosophy in Germany. Albany, NY: State University of New York Press. Hill, Thomas E. 1973. “Servility and Self-Respect.” The Monist 57: 87–104. Hitchens, Christopher. 2001. Letters to A Young Contrarian. New York, NY: Basic Books. Housman, Alfred Edward. 1922/2010. “Tell Me Not Here, It Needs Not Saying.” In A Shropshire Lad and Other Poems: The Collected Poems of A.E. Housman. London: Penguin Classics. Jackson, Frank. 1998. From Metaphysics to Ethics. Oxford: Oxford University Press. Johnston, Mark. 2011. Saving God: Religion After Idolatry. Princeton University Press. Kahane, Guy. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696. Kahane, Guy. 2012. “The Value Question in Metaphysics.” Philosophy and Phenomenological Research 85: 27–55. Kahane, Guy. 2014. “Our Cosmic Insignificance.” Noûs 48: 745–772. Kahane, Guy. 2017. “If Nothing Matters.” Noûs 51: 327–353. Kant, Immanuel. 1970. “What Is Enlightenment?” In Kant: Political Writings. Cambridge: Cambridge University Press. Kant, Immanuel. 1994. Ethical Philosophy. Translated by James W. Ellington. Indianapolis: Hackett. Kraay, Klaas, and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 153–178. Kreeft, Peter. 1990. Everything You Ever Wanted to Know About Heaven But Never Dreamed of Asking. Ignatius Press: San Francisco. Leiter, Brian. Forthcoming. “The Death of God and the Death of Morality.” The Monist. Lemos, John. 2017. “God’s Existence and the Kantian Formula of Humanity.” Sophia 56: 265–278.

130  Guy Kahane Luck, Morgan, and Ellerby, Nathan. 2012. “Should We Want God Not to Exist?” Philo 15: 193–199. Maitzen, Stephen. 2009. “Ordinary Morality Implies Atheism.” European Journal for Philosophy of Religion 1: 107–126. Martin, Michael, and Monnier, Ricki. 2003. The Impossibility of God. New York, NY: Prometheus Books. Mavrodes, George I. 1986. “Religion and the Queerness of Morality.” In Rationality, Religious Belief, and Moral Commitment: New Essays in the Philosophy of Religion, edited by Robert Audi and William Wainwright, 213–226. Ithaca, NY: Cornell University Press. Mawson, Tim. 2012. “On Determining How Important It Is Whether or Not There Is a God.” European Journal for Philosophy of Religion 4: 95–105. McLean, Graeme. 2015. “Antipathy to God.” Sophia 54: 13–24. Mill, John Stuart. 1874. “Theism.” In Three Essays on Religion. New York, NY: Henry Holt & Co. Miller, Jonathan. 2004. The Atheism Tapes: Colin McGinn, BBC. Monk, Ray. 1996. Bertrand Russell: Spirit of Solitude. New York, NY: The Free Press. Moore, Adam, D. 2003. “Privacy: Its Meaning and Value.” American Philosophical Quarterly 40: 215–227. Mugg, Joshua. 2016. “The Quietist Challenge to the Axiology of God: A Cognitive Approach to Counterpossibles.” Faith and Philosophy 33: 441–460. Nagel, Thomas. 1971. “The Absurd.” The Journal of Philosophy 68: 716–727. Nagel, Thomas. 1997. The Last Word. Oxford: Oxford University Press. Nagel, Thomas. 2009. “Secular Philosophy and the Religious Temperament.” In Secular Philosophy and the Religious Temperament: Essays 2002–2008. Oxford: Oxford University Press. Nagel, Thomas. 2012. Mind and Cosmos. Oxford: Oxford University Press. Nietzsche, Friedrich. 2001. The Gay Science. Translated by Josefine Nauckhoff. Cambridge: Cambridge University Press. Oshana, Marina A. L. 1998. “Personal Autonomy and Society.” Journal of Social Philosophy 29: 81–102. Parent, William. 1983. “Privacy, Morality and the Law.” Philosophy and Public Affairs 12: 269–288. Parfit, Derek. 2011. On What Matters, Vol. II. Oxford: Oxford University Press. Penner, Myron. 2015. “Personal Anti-Theism and the Meaningful Life Argument.” Faith and Philosophy 32: 325–337. Penner, Myron, and Lougheed, Kirk. 2015. “Pro-Theism and the Added Value of Morally Good Agents.” Philosophia Christi 17: 53–69. Pettit, Philip. 1997. Republicanism: A Theory of Freedom and Government. Oxford: Oxford University Press. Plantinga, Alvin. 2000. Warranted Christian Belief. Oxford: Oxford University Press. Plantinga, Alvin. 2012. “Why Darwinist Materialism Is Wrong.” (Review of Nagel’s Mind and Cosmos.) The New Republic, November 2012. Rachels, James. 1997. “God and Moral Autonomy.” In Can Ethics Provide Answers? And Other Essays in Moral Philosophy, 109–123. Boulder: Rowman & Littlefield Publishers.

If There Is a Hole, It Is Not God-Shaped 131 Rescher, Nicholas. 1990. “On Faith and Belief.” In Human Interests, 166–178. Stanford: Stanford University Press. Russell, Bertrand. 1903. “The Free Man’s Worship.” In Russell on Religion: Selections from the Writings of Bertrand Russell, edited by Louis Greenspan and Stephan Andersson, 31–38. London: Routledge. Russell, Bertrand. 1950. “Am I an Atheist or an Agnostic?” Reprinted in John G. Slater, ed., 1997. The Collected Papers of Bertrand Russell: Last Philosophical Testament, 1943–68, 89–92. London: Routledge. Russell, Bertrand. 1957. Why I Am Not a Christian and Other Essays. New York, NY: George Allen and Unwin. Sartre, Jean-Paul. 1964. The Words. New York, NY: George Braziller. Sartre, Jean-Paul. 2007. Existentialism is a Humanism. New Haven: Yale University Press. Scanlon, Thomas M. 2014. Being Realistic About Reasons. Oxford: Oxford University Press. Schellenberg, J. L. 2016. “God for All Time: From Theism to Ultimism.” In Alternative Conceptions of God: Essays on the Metaphysics of the Divine, edited by Andrei Buckareff and Yujin Nagasawa, 164–177. Oxford: Oxford University Press. Sidgwick, Henry, 1907. The Methods of Ethics. London: Hackett Publishing. Smith, Adam. 2002. The Theory of Moral Sentiments. Cambridge: Cambridge University Press. Strawson, Galen. 2008. Real Materialism and Other Essays. Oxford: Oxford University Press. Swinburne, Richard. 1994. The Christian God. Oxford: Oxford University Press. Taylor, Charles. 2007. A Secular Age. Boston: Harvard University Press. van Inwagen, Peter. 2009. Metaphysics (3rd Edition). Boulder, CO: Westview Press. Weinberg, Steven. 2012. “Without God.” Chapter 25 in Lake Views, Cambridge, MA: Harvard University Press. Williams, Bernard. 1973. “The Makropulos Case: Reflections on the Tedium of Immortality.” In Problems of the Self, 82–100. Cambridge: Cambridge University Press. Wolf, Susan. 2010. Meaning in Life and Why It Matters. Princeton: Princeton University Press.

6 The Problem of Magic Stephen Maitzen

1. Introduction In previous work (e.g., Maitzen 2009; 2010a; 2010b; 2013; 2014), I argued that theism undermines moral value because theism dissolves moral obligations that form the core of ordinary morality. Here I describe some destructive implications of theism, and supernaturalism in general, for epistemic and more broadly axiological kinds of value. Various philosophers and theologians—including Aquinas, Leibniz, Bernard Lonergan, and Hugo Meynell—see evidence for the existence of God in the fact that the universe is intelligible to us, i.e., the fact that we can understand to an impressive degree how the universe works. As Lonergan says, “If the real is completely intelligible, God exists. But the real is completely intelligible. Therefore, God exists” (2004, 5). In Meynell’s words, If there were not something analogous to human intelligence in the constitution of the world, the world would not be intelligible. But the world is intelligible. . . . The intelligibility of the world . . . is perfectly to be accounted for if the world is due to the fiat of an intelligent will which conceives all possible worlds, and wills the one which we actually inhabit. (1982, 68, 70) I’ll argue very much to the contrary. First, I’ll try to make clear what “intelligibility” means in this context and in particular, what it had better not mean. Second, I’ll examine and reject some reasons philosophers have given for thinking that the intelligibility of the universe requires or suggests supernatural—and more particularly, theistic—backing. Third, I’ll argue that supernaturalism, including theism, would in fact threaten the complete intelligibility of the universe. Fourth, I’ll argue that supernaturalism, including theism, would impose non-epistemic costs as well.

2. The Kind of Intelligibility at Issue When philosophers and theologians invoke the existence of God in order to explain the intelligibility of the universe, what they mean by

The Problem of Magic 133 “intelligibility” of course depends on which philosopher or theologian you ask. But one thing I hope they don’t mean by it is that the universe obeys the laws of logic and in that sense is a “logical” or a “rational” place. Yes, the universe obeys the laws of logic: in that sense, the universe is indeed a logical and a rational place.1 But to explain the operation of the laws of logic by invoking God’s existence is to venture beyond the bounds of intelligibility. For no genuinely theistic explanation can assert merely that God’s existence is sufficient for the operation of the laws of logic, because everything is (trivially) sufficient for the operation of the laws of logic: the operation of any law of logic is necessary in the strongest sense and therefore implied by any proposition at all. The existence of cheese is therefore just as (trivially) sufficient for the operation of the laws of logic as the existence of God is. Instead, any theistic explanation of the operation of the laws of logic must say at least this: if God didn’t exist, then the laws of logic wouldn’t hold. But no sense at all can be attached to the consequent of that conditional. What could it mean for the laws of logic not to hold? Would it “mean” that the laws of logic never hold and yet sometimes hold? Would it “mean” that the laws of logic sometimes hold, never hold, and neither sometimes nor never hold? If it wouldn’t have either of those pseudomeanings, why not? Presumably not because the laws of logic would prevent it! No one can make any sense of what would be implied by the failure of the laws of logic, and therefore no one can make any sense of the supposition that the laws of logic might not hold.2 In the face of that senselessness, one might retreat to the claim that without God only some rather than all of the laws of logic would fail. But which laws would fail, and why only those? I can’t see any plausible answer to those questions. Admittedly, intuitionistic logicians do claim to accept all of the laws of classical logic except for excluded middle and double-negation elimination. If they mean what they claim, then they allow that for some proposition (not merely some sentence) p, it fails to be the case that either p or not-p. But the example of intuitionistic logicians is instructive. For it would seem that they don’t, after all, mean what they claim: they don’t wish to be construed as saying that the world might be neither a particular way nor not that way, presumably because they recognize the senselessness of saying such a thing. So, instead, they redefine assertion and negation as the constructive provability of, respectively, a proposition and the proposition’s contradictory. There’s nothing senseless in asserting that neither a constructive proof that p nor a constructive proof that not-p exists, but the assertion doesn’t come close to rejecting the law of excluded middle or any other law of logic. In sum, it makes literally no sense to credit God for the fact that logic works. Therefore, anyone hoping to give a theistic explanation for the intelligibility of the universe presumably means something else by “the intelligibility of the universe.” So far as I can tell, recent proponents of the argument for God’s existence from the intelligibility of the universe

134  Stephen Maitzen do mean something else by it, namely, the fact that the universe is susceptible to empirical, scientific discovery and explanation.3 At any rate, I’ll interpret them that way in what follows.

3. Knowledge Needs No Supernatural Backing First, however, I want to critique a different but related theistic argument also premised on our ability to know the world around us. One version of the argument occurs in the Meditations, where Descartes argues that we have reason to trust our cognitive equipment—i.e., our senses, memory, and reasoning—if and only if we believe that a benevolent God created us. Among the things we need God to guarantee, Descartes notoriously includes any legitimate certainty we have that the laws of logic hold (1641/2008, 49–51, 92–93). That way, however, lies peril, for reasons we’ve just seen. Because it’s literally unintelligible to suppose that the laws of logic ever fail, it’s likewise unintelligible to claim that we might not be legitimately certain that they never fail. For what is it that we’re not legitimately certain never happens? Presumably the answer is “The laws of logic fail.” As I argued, no one can understand that answer any better than he or she can understand gibberish. Therefore, no one can understand the claim that we might not be legitimately certain that the answer isn’t true: what answer?4 So it makes no sense to suppose that God might underwrite our trust in logical reasoning, which leaves Descartes to argue that we have reason to trust our other cognitive faculties—our senses and memory—if and only if we believe that a benevolent God created us. We can make short work of this argument. Unfortunately for Descartes, the “if” part of his biconditional fails because it depends on a premise that his own contemporaries and many later commentators have roundly criticized, namely, the premise that a benevolent God couldn’t ever deceive us (Descartes 1641/2008, 33). I regard those criticisms as decisive.5 There’s no good reason for presuming that a benevolent God couldn’t deceive us, and therefore, Descartes fails to show that God must have given us reliable senses and memory. Consequently, the belief that a benevolent God created us gives us no more reason to trust our senses and memory than the reasons we have independently of that belief. Indeed, there’s a sense in which it gives us less reason, as I’ll argue below. In any case, the existence of God is at best irrelevant to the trustworthiness of our senses and memory. At the outset I mentioned Lonergan’s argument for the existence of God from the complete intelligibility of the universe. His fellow theologian R. M. Burns rejects Lonergan’s argument, but the grounds Burns gives for rejecting it are interestingly similar to the grounds Descartes gives for using God to underwrite the reliability of our cognitive equipment. According to Burns, The ultimate reason for [Lonergan’s] failure is that any attempt at proof of the human capacity to attain objective knowledge must beg

The Problem of Magic 135 the question by assuming such a capacity. . . . The general conclusion to be drawn is that despite all efforts to remove it, or to ignore it, an abiding question-mark remains at the basis of all human cognitive endeavour. . . . It follows that any human being who chooses to affirm his capacity to know is making an act of trust in his nature which goes beyond any possible evidence. (Burns 1987, 146, 149) This act of trust, according to Burns, has theistic implications: If so, then the basic trust at the root of human cognitive activity must be understood as, implicitly at least, the hope of created minds that they can participate in the uncreated light of the divine intellect by grace. . . . [H]uman reason must be grounded in a trusting dependence upon God. . . . By contrast, the attempt of the human mind to understand itself as cognitively self-sufficient can lead only to an oscillation between unjustifiable dogmatism and a paralysing skepticism. (Burns 1987, 151, 154) I’ll examine this argument at some length. But an obvious objection to it arises immediately. Burns claims the following: (B) Any attempt at proof of the human capacity to attain objective knowledge must beg the question by assuming such a capacity. What’s the status of claim B itself? Burns argues for B, as if he wants us to accept B and, indeed, as if his aim is to prove that B is true and, given that his argument for B is entirely a priori, to prove it a priori. Must Burns therefore beg the question in his attempt to prove that B? If he must, then his attempt must fail, because arguments that commit the fallacy of begging the question don’t prove anything. Therefore, if Burns doesn’t mean to imply that his own argument is a failure, he must allow that at least proposition B can be proven to be true. In that case, it’s hard to see what else is necessary yet unattainable in order for us to know that B and to know that B unconditionally. Indeed, it’s for this reason—i.e., the sufficiency of proving for knowing—that a proof of global scepticism, a proof that no one knows any proposition at all, looks to be impossible. As decisive as this objection to Burns seems to be, let’s waive it. The objection refutes any alleged proof that we have no propositional knowledge at all, not even a priori knowledge: the objection shows that any such proof undermines itself. But perhaps Burns is merely guilty of overreaching. As I remarked earlier, it’s more charitable to construe theistic arguments from intelligibility as focused on our empirical beliefs. Construed that way, Burns’ pessimism about our cognitive self-sufficiency has been echoed by such nontheistic epistemologists as Laurence BonJour

136  Stephen Maitzen and Barry Stroud.6 These epistemologists observe that we routinely take ourselves to know various empirical propositions, such as the proposition that we have bodies, that Toronto is north of New York, and that species evolve by natural selection. That is, we routinely regard our cognitive equipment as successful in achieving such knowledge. But how do we know that we know such propositions—how do we know that our cognitive equipment in fact succeeds in achieving empirical knowledge— if all we can rely on to answer that epistemological question is that very cognitive equipment itself? As Burns says, if nothing transcending our cognitive equipment ensures its reliability, then mustn’t we beg the question if we use that equipment to verify that we’ve attained empirical knowledge? The worry is that without some transcendent validation of our cognitive equipment, any empirical knowledge that we claim for ourselves will always be “merely conditional” knowledge: it will always be the consequent of a conditional proposition whose antecedent we can never discharge.7 But this worry rests on a mistake.8 To see why, suppose that the “JTB analysis” of propositional knowledge is correct: suppose that propositional knowledge is correctly analyzed as justified, true belief. I choose the JTB analysis merely in order to have a simple example; the point I’ll make doesn’t depend on the details or the correctness of the analysis. According to the JTB analysis, I know that Toronto is north of New York if and only if I have a justified, true belief that Toronto is north of New York. I know that the JTB analysis is correct if and only if I have a justified, true belief that the JTB analysis is correct. I know that I possess knowledge if and only if I have a justified, true belief that I possess knowledge. And so on. At no point must the JTB analysis appeal to anything that transcends what human cognitive equipment can grasp on its own. In what sense does the JTB analysis, or any analysis of knowledge, make all knowledge “merely conditional”? One might suppose that the JTB analysis makes all knowledge merely conditional because (1) only if the JTB analysis is correct in the first place does anyone know that it’s correct, and (2) only if the JTB analysis is correct in the first place does anyone know a proposition p just by virtue of attaining a justified, true belief that p. But notice that (1) and (2), although perfectly true, are perfectly trivial. The truth of (1) stems entirely from the conceptual truth that only what’s correct can be known to be correct. Given that the JTB analysis by definition analyzes knowledge as justified, true belief, (2) is just the truism that any explanation implying the JTB analysis will be a correct explanation only if the JTB analysis is itself correct. If (1) or (2) implied that knowledge is always merely conditional, then every proposition q would be merely conditional, on the grounds that (1*) q is known to be true only if q is true and that (2*) if q implies r, then q is true only if r is true. But to call every proposition “merely conditional” on those

The Problem of Magic 137 grounds is to misuse language. Therefore, neither (1) nor (2) implies that knowledge is always merely conditional. Indeed, nothing of interest follows from the trivial truths (1) and (2). In particular, neither (1) nor (2) implies that we can’t know that the JTB analysis is correct, know that we know it’s correct, know other propositions, and know that we know those other propositions. Crucially, neither (1) nor (2) implies that we must always merely assume that the JTB analysis is correct or that we must always merely assume that we possess knowledge. If one demands that proponents of the JTB analysis show that they have knowledge without invoking the JTB analysis itself, then one’s demand is obviously unfair: it’s unfair to ask proponents of the JTB analysis to show that they have knowledge without letting them invoke what they regard as the correct analysis of knowledge. “But,” one might ask proponents of the JTB analysis, “how do you know it’s the correct analysis of knowledge?” To that question the proponents can reply, entirely legitimately, that they have a justified, true belief that it’s the correct analysis. One might then challenge them to show that they fulfil the conditions required by the analysis: that they really do accept the analysis, that their acceptance of the analysis is justified, and that the analysis itself is correct. But none of these latter challenges would expose any defect in their reply to the question, “How do you know that your analysis is correct?”9 Nevertheless, the objector might then ask them: “How can you be sure that the JTB analysis you invoked is the correct analysis in the first place?” Here, however, the objector’s use of the word “sure” smuggles in the false assumption that knowledge must be infallible. For the objector doesn’t mean to ask the empirical, psychological question “How can you achieve full confidence, or subjective certainty, that your analysis is correct” That question fails to capture the objector’s epistemological worry. Nor does the objector mean to ask how proponents of the JTB analysis can rule out the epistemic possibility that their analysis is wrong, because that question is just a fancier way of asking them how they know that their analysis is right, a question to which we’ve already seen the answer. Instead, the objector means to ask proponents how they can rule out the logical possibility that their analysis is wrong. To that question, it suffices to reply that nothing in the concept of knowledge requires the logical impossibility that the knower is mistaken: the concept of knowledge allows that I can know that p based on evidence that’s logically consistent with the falsity of p.10 In sum, no general problem of the kind described by Burns, BonJour, or Stroud afflicts the analysis of human knowledge.11 Someone wanting to criticize the project of analyzing knowledge must challenge the correctness of the particular analysis on offer, whether it’s the JTB analysis or some other, in the usual way by proposing counterexamples. It doesn’t suffice to charge, falsely, that any analysis leaves us

138  Stephen Maitzen with only “conditional knowledge” or that any analysis is known to be correct only if it’s infallibly known to be.

4. Supernaturalism Threatens Human Knowledge We’ve seen no good reason to think that human knowledge needs backing by anything transcending our cognitive equipment, including anything supernatural. In this section, I’ll argue that supernaturalism, including theism, is worse than unnecessary for human knowledge: it positively threatens human knowledge.12 Philosophers have long been aware of the possibility that theism poses a sceptical threat. In the First Meditation, Descartes briefly wonders whether believing in the existence of an all-powerful God might undermine our reasons to trust our cognitive equipment. He then cites God’s reputation for goodness in order to dismiss this worry and move on to consider the Evil Demon hypothesis instead (1641/2008, 15–16). I’ve already said why I think Descartes moves too fast here: God’s goodness wouldn’t preclude God’s engaging in deception. As I’ll now explain, theism (and supernaturalism more generally) poses a threat to human knowledge for reasons having nothing to do with deliberate deception. As I use the terms, “supernaturalism” denotes the view that at least some supernatural things exist or occur, whereas “naturalism” denotes the view that no supernatural things exist or occur. Admittedly, that way of putting the contrast isn’t very informative! Can we define “naturalism” more informatively than as “the denial of supernaturalism”? I think so. Any informative definition of “naturalism” is likely to be contentious (see Williamson 2011a; 2011b), but according to one promising definition, naturalism asserts “in the simplest terms, that every mental thing is entirely caused by fundamentally nonmental things and is entirely dependent on nonmental things for its existence” (Carrier 2007). While this definition is fine as far as it goes, it rests on a distinction between the mental and the nonmental, a distinction that’s roughly as contentious as the definition of “naturalism” itself, to say nothing of the definition of “caused by,” another concept it invokes. Therefore, I propose to define “naturalism” in a way that avoids those controversies by invoking a concept that I think we understand at least as well as, and probably better than, the distinction between the mental and the nonmental or the concept of causation: to wit, the concept of a purpose. As I construe it, naturalism is the view that purposes aren’t fundamental: every being, action, or whatever, that has a purpose (a goal, a telos) arises from things that have no purpose.13 In a phrase, “Purposes don’t go all the way down.” I wrote the previous paragraph with the purpose of introducing my proposed definition of “naturalism.” According to naturalism as I define it, that purpose of mine must have arisen, somewhere down the line, from things that have no purpose at all—neither a purpose supplied by

The Problem of Magic 139 any of those things themselves nor a purpose given to them by something else. Maybe those purposeless things are the neurons in my brain, the molecules that compose those neurons, or the atoms that compose those molecules. My neurons function when they generate my purposes, but my neurons can function without functioning for some purpose. Going downward (so to speak) in the chain of ontological dependence, purposes eventually cease to exist and never return. This definition makes naturalism compatible with contemporary science and, I believe, with common sense as well; I don’t think that common sense takes a stand on whether purposes go all the way down. It allows the naturalist to accept the reality of phenomenal, first-person experiences; to accept that purposive explanations, although never fundamental, are often useful and correct; to regard mathematics as discovering objective truths about an independent, even Platonic, reality; and to regard the laws of logic as holding necessarily in the strongest sense (which they do) and as more fundamental than the truths discovered by empirical science (which they are). Yet the definition has enough content to make naturalism incompatible with traditional theism, substance dualism, animism, other versions of supernaturalism, and perhaps also panpsychism. Most importantly in the present context, only positions that are compatible with naturalism as I’ve defined it allow human discovery to be limitless in depth. For according to supernaturalism—the contradictory of naturalism—purposes do go all the way down, at least sometimes if not always. According to supernaturalism, at least some things have a purposive explanation that’s fundamental. According to naturalism, by contrast, no purposive explanation is fundamental: any purpose has a purposeless, purely mechanistic explanation. Again, there’s a purposive explanation for my writing what I wrote three paragraphs ago: I wrote it in order to propose a definition of “naturalism.” But according to naturalism, the explanation of my action doesn’t bottom-out with any purpose, because purposes themselves always arise from things without purpose. I hasten to add that this view doesn’t regard purposes as epiphenomenal: purposes make a difference in the world. My having a particular purpose played an essential role in my writing what I did: had I lacked that purpose, I wouldn’t have written what I wrote. But a purpose can play that indispensable role even while arising from something purposeless. According to supernaturalism, by contrast, the world contains genuine magic, not just the pseudo-magic practiced by illusionists like David Kotkin, a.k.a. “David Copperfield,” whose tricks all rely on mechanisms and therefore can be explained in mechanistic terms. We know that there’s some manner in which Copperfield makes the Statue of Liberty disappear from view, some mechanism by which he accomplishes the illusion, and furthermore a mechanism whose operation we could, at

140  Stephen Maitzen least in principle, understand as thoroughly as we might like by asking ever-deeper questions about it. Things are very much otherwise with the magic posited by supernaturalism. The priest who performs an exorcism isn’t supposed to be exploiting a mechanism whose operation we could scrutinize with arbitrary thoroughness. On the contrary, he calls on purposive agents: beseeching God to give him the strength required to perform the rite, and commanding the evil spirit to depart the victim. To see why it can’t be a case of mechanism, consider the following questions. In order for an exorcism to work, exactly how loud, in decibels, must the priest’s words be when they reach the ears of the possessed person? Indeed, why must the priest make any noise: what function do vocalizations serve? Could he use only sign language, or no language at all? Would a speech defect diminish the effectiveness of his incantation? What if he leaves out a word of the prescribed rite or switches the order of two words without violating grammar or syntax? At least in Latin, word-order usually doesn’t affect sentence-meaning, but would it affect the power of a spell? If exorcism worked by way of some mechanism, there would be a way the mechanism worked, and all of these questions would be worth taking seriously. But I presume that nobody would take all, or perhaps any, of them seriously. In sum, if something literally magical occurs, then there’s no way it occurs—no manner in which it occurs, no answer to how it occurs, no mechanism by means of which it occurs—that we could scrutinize ever more deeply. What holds for supernaturalism holds all the more so for theism in particular, which says that something magical stands at the foundation of our universe: a nonphysical God who created the universe from nothing, without exploiting any laws of nature, and without relying on any mechanism. On the contrary, God had free rein over which natural laws, if any, to create in the first place. Adolf Grünbaum puts it this way: The Book of Genesis tells us about the divine word-magic of creating photons by saying “Let there be light.” But we aren’t even told whether God said it in Hebrew or Aramaic. I, for one, draw a complete explanatory blank when I am told that God created photons. This purported explanation contrasts sharply with, say, the story of the formation of two photons by conversion of the rest-mass of a colliding electron-positron pair. Thus, so far as divine causation goes, we are being told . . . that an intrinsically elusive, mysterious agency X inscrutably produces the effect. (Grünbaum 1991, 235) If the universe is at bottom magical, then our inescapably nonmagical ways of figuring it out are doomed to fail eventually. According to

The Problem of Magic 141 naturalism, by contrast, nothing magical stands at the foundation of our universe, so there’s no reason in principle why science can’t make our knowledge of the origin and workings of the universe ever deeper.14 It’s a commonplace that whenever we can explain a phenomenon entirely in mechanistic terms, we don’t invoke the intentions of agents to explain it. Now that we can explain thunder entirely in mechanistic terms, we’re not at all tempted to explain it by invoking the intentions of Thor. Indeed, the progress of natural science has largely consisted in removing intentional agents from our explanations and predictions of phenomena. But theism regards God’s intentions as fundamental to the universe, rather than as the product of anything nonintentional. Augustine makes the point this way: [T]hey are seeking to know the causes of God’s will, when God’s will is itself the cause of everything there is. After all, if God’s will has a cause, then there is something that is there before God’s will and takes precedence over it, which it is impious to believe. So then, anyone who says, “Why did God make heaven and earth?” is to be given this answer: “Because he wished to.” . . . Anyone though who goes on to say, “Why did he wish to make heaven and earth?” is looking for something greater than God’s will is; but nothing greater can be found. (Augustine 2002, 42)15 On theism, therefore, we can’t hope to understand the universe as deeply as we might want by means of our most reliable (i.e., natural-scientific) methods. Unlike naturalism, theism puts a barrier in the path of our everdeeper knowledge of the universe. At best, we could hope to approach that barrier asymptotically, with scientific discovery becoming everslower, which would represent a surprising and discouraging reversal of the pattern shown by science since the Enlightenment. At this point, one might object that it’s not only supernaturalism that puts a barrier in the path of our ever-deeper knowledge of the universe: so does quantum mechanics. At least in the form of its most popular interpretation, quantum mechanics says that (3) events such as individual particle-decays are fundamentally, ontologically indeterministic, there being no sufficient reason why they occur rather than not; and that (4) some pairs of quantum values can’t, even in principle, be measured with arbitrary precision (Faye 2014; Hilgevoord and Uffink 2014). If (3), then we can’t hope to know why, for example, a particular atom of uranium decayed during a given interval of time rather than not; if (4), then we can’t hope to know, for example, both the position and the momentum of an electron at a given time as precisely as we might wish. Where quantum phenomena such as these are concerned, maybe we’ve already hit

142  Stephen Maitzen the barrier to our ever-deeper understanding of the world. Indeed, maybe this barrier constitutes evidence that supernatural magic does underlie the world that science can observe and understand! One reply to this objection is that (3) and (4) belong to a particular interpretation of the quantum mechanical formalism, the Copenhagen interpretation developed in the early years of quantum theory. Some surveys suggest that the Copenhagen interpretation may still command the allegiance of a plurality of the experts—although, significantly, not a majority of them (Schlosshauer et al. 2013). Nevertheless, several other interpretations of quantum mechanics exist that fit with the formalism and with all the experimental data, including interpretations that emphatically reject ontological indeterminism and uncertainty (Wolchover 2014). In short, neither the formalism nor the data imply any barrier to further discovery about the mechanisms underlying quantum phenomena. Nor, therefore, do they provide any evidence that we’ve already hit a barrier separating mechanism from magic. Finally, one might object that supernaturalism, including theism, allows for more robust inquiry than my criticisms of it suggest. For supernaturalism still allows us to ask indefinitely many scientific questions, even though it limits the depth of the scientific questions that have answers, and, unlike naturalism, theism in particular allows us to ask indefinitely many nonscientific questions about God. But how many and how deep can those latter questions really be? Monotheistic theology has been a going concern since 1200 CE, at the latest; Augustine, Boethius, Avicenna, Anselm, Averroes, and Maimonides had all made their contributions by then. Empirical science has been a going concern since perhaps 1543, the date of Copernicus’s De Revolutionibus. Even theologians must admit that few things in human history are as glaringly unequal as (c) the growth of our knowledge about God since 1200 and (d) the growth of our knowledge about nature since 1543. If the prospects for further discoveries about God were as robust as the objector suggests, then we should expect (c) and (d) to be less glaringly unequal. Even if, for some reason, further growth in our knowledge of God can’t occur in this world but only in the next, that waiting period must nevertheless count as an epistemic cost of theism as compared to naturalism.

5. Some Axiological Costs of Magic I’ll conclude by describing some more broadly axiological threats posed by supernaturalism, including theism. These threats would arise if genuine magic were a feature of our world, and particularly if it were, as theism says, a fundamental feature of our world. For reasons similar to those I’ve already sketched, genuine magic isn’t a progressive research program, to use Lakatos’s term. Consider again

The Problem of Magic 143 the case of exorcism, and compare two hypotheses: (5) the naturalistic hypothesis that behaviour attributed to demonic possession can be explained in purely mechanistic terms (invoking, say, neurological disorders such as epilepsy or psychosis); and (6) the supernaturalistic hypothesis that the behaviour results from being possessed by demons. If (5) is true, then we can realistically expect to cure or altogether prevent the unwanted behaviour if we focus sufficiently on the neurological mechanisms that underlie it. But if instead (6) is true, then we can’t realistically expect to get any better at treatment than we’ve been for centuries; much less can we expect to prevent the behaviour entirely. As the Harry Potter stories seem to recognize, a genuinely magical world, as enchanting as it may sometimes seem, is perversely unpredictable, frustrating, and scary. To use a mundane example: In a naturalistic, law-governed, fundamentally mechanistic world, if I want to read a book on my shelf, I pick it up and start reading. If I find that I can’t accomplish that, then I consult a physician about my immobility, or my poor reading vision, or my inability to comprehend text. But in a supernaturalistic world, I may need the book’s cooperation to begin with in order to read it: maybe the book has its own personality and needs to be appeased before it will sit still for me. Or maybe the book has been bewitched by a spell that someone else cast on it. Even if I can get the book to cooperate by casting a counter-spell, there will be no mechanism by means of which my counter-spell works. There will be no mechanism by means of which any spell or counter-spell works, not merely a mechanism we don’t yet understand. Consider how much better our lives go because the inanimate things in our world aren’t animated by spirits. Otherwise there would be no limit to how unpredictable, frustrating, and scary life could get. Suppose we had to appease the capricious spirit of every part of every meal before we could eat it, with a different spirit animating every type of food or, yet more chaotically, every morsel of every type of food. Our world isn’t as chaotic as that, of course, even if theism is true. But if theism is true, then our world rests on a foundation that’s just as inscrutable as magic.

Acknowledgment For helpful comments, I thank my audience at the conference on the Axiology of Theism held at Ryerson University in September 2015.

Notes 1. Regrettably, people don’t always reason as the laws of logic imply they ought to: people sometimes make invalid inferences. But they don’t thereby violate the laws of logic, any more than I violate the laws of arithmetic when I make a calculation error. The laws of logic and of arithmetic govern even if I fail to infer or calculate as they require.

144  Stephen Maitzen 2. One might respond that, because the laws of logic are necessary, their failure is impossible and therefore their failure implies every proposition: the failure of the laws of logic implies their total success; their total success and their total failure; their total success, their total failure, and neither their total success nor their total failure; etc. Because it relies on drawing inferences, I’m not convinced that this response genuinely assumes the failure of the laws of logic. But even if it does, I don’t see how it counts as making any sense of the failure of the laws of logic. 3. A complication arises in the case of Meynell, who at times seems to treat a thing’s existence as logically equivalent to its intelligibility: “[W]hat is at issue is the general property of intelligibility which one might say that things have to have to be things, that the world has to have in order to be a world” (Meynell 1982, 83, as noted in Sherry 1983, 130; italics mine). If simply being a world necessitates being intelligible, then it’s hard to see how the existence of God would be needed to explain the world’s intelligibility. Henceforth I’ll simply ignore this unfortunate feature of Meynell’s presentation. 4. Because the answer is no more intelligible than gibberish, the claim in question is no more intelligible than “We might not be legitimately certain that it isn’t true that @#$%^&*,” a claim that should strike anyone as unintelligible. Here it’s crucial to distinguish (a) the semantic claim that “@#$%^&*” expresses a truth from (b) the first-order claim that @#$%^&*. I grant that (a) is intelligible, but (b) isn’t, and it seems clear that Descartes’ position requires (b): Descartes isn’t making a semantic claim in the metalanguage; he’s making a first-order claim in the object language. 5. For recent criticism, I recommend Wielenberg 2010. 6. Stroud is explicitly pessimistic when he writes that “it still seems to me [that] scepticism will always win going away” (Stroud 1994, 307). BonJour’s pessimism is less explicit, because he seems to allow that one can achieve cognitive self-sufficiency if one has internal, first-person access to the fact that one’s empirical beliefs are reliably produced (see Kornblith 2001, 185–186). Because I see no reason to think that we have such access, I classify both Stroud and BonJour as pessimists. 7. This way of putting the worry, including the phrase “merely conditional,” is found in Kornblith 2001, who goes on to criticize the worry as unfounded. The specific context of Kornblith’s discussion, and also that of Stroud 1994 and Sosa 1994, is externalist analyses of knowledge and justification, rather than the JTB analysis of knowledge. But that difference doesn’t affect the point I’m making here. 8. As Ernest Sosa (1994) and Hilary Kornblith (2001) have pointed out. 9. A similar lesson applies in the case of explanations. When the fire investigator concludes that a short circuit in poorly installed wiring explains why the fire started, she has given a paradigmatic explanation. Her explanation isn’t “merely conditional” because it doesn’t also explain why the wiring was poorly installed. 10. As Peter Klein emphasizes, empirical knowledge “does not rest upon entailing evidence” (Klein 1981, 14–15, citing Chisholm 1973, 232 to the same effect). Similarly, possible-worlds analyses of knowledge, such as “truth-tracking” and “safety” analyses, impose requirements on the knowing subject’s belief of p in nearby possible worlds rather than in all possible worlds. See Ichikawa and Steup 2014. 11. Compare Kornblith 2001, 197. 12. Christine Overall has argued that miracles, in the usual sense of violations of natural law, “would be a ‘cognitive evil’ . . . because they would seriously compromise the human capacity to understand the universe” (Overall 2006,

The Problem of Magic 145 356). I’ll argue that supernaturalism in general, and theism in particular, would seriously compromise the human capacity to understand the universe even if miracles never occurred. 13. Alternatively, we can say that on naturalism, anything’s having a purpose is realized in or constituted by one or more things lacking any purpose, in the sense of “realized” and “constituted” used in discussions of mental causation. 14. Alvin Plantinga has argued that naturalism itself, when combined with evolutionary biology, provides grounds for doubting the reliability of our cognitive equipment. I ignore his argument here because it has been thoroughly and trenchantly criticized by the contributors to Beilby 2002. 15. I thank Mike Ashfield for the reference. See also O’Connor 2013 for an argument that no contrastive explanation of God’s choice to create a world (i.e., rather than not create one) is even possible.

References Augustine. 2002. “On Genesis: A Refutation of the Manichees.” In On Genesis, edited by John E. Rotelle, translated by Edmund Hill. Hyde Park, NY: New City Press. Beilby, James, ed. 2002. Naturalism Defeated? Essays on Plantinga’s Evolutionary Argument Against Naturalism. Ithaca and London: Cornell University Press. Burns, R. M. 1987. “Bernard Lonergan’s Proof of the Existence and Nature of God.” Modern Theology 3: 137–156. Carrier, Richard. 2007. “Defining the Supernatural.” Blogpost on January 18. http://richardcarrier.blogspot.ca/2007/01/defining-supernatural.html. Chisholm, Roderick M. 1973. “On the Nature of Empirical Evidence.” In Empirical Knowledge: Readings from Contemporary Sources, edited by Roderick M. Chisholm and Robert J. Swartz, 224–249. Englewood Cliffs, NJ: Prentice-Hall. Descartes, René. 1641/2008. Meditations on First Philosophy: With Selections from the Objections and Replies. Translated by Michael Moriarty. New York, NY: Oxford University Press. Faye, Jan. 2014. “Copenhagen Interpretation of Quantum Mechanics.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. http://plato. stanford.edu/archives/fall2014/entries/qm-copenhagen. Grünbaum, Adolf. 1991. “Creation as a Pseudo-Explanation in Current Physical Cosmology.” Erkenntnis 35: 233–254. Hilgevoord, Jan, and Uffink, Jos. 2014. “The Uncertainty Principle.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. http://plato. stanford.edu/archives/spr2014/entries/qt-uncertainty. Ichikawa, Jonathan Jenkins, and Steup, Matthias. 2014. “The Analysis of Knowledge.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. http://plato.stanford.edu/archives/spr2014/entries/knowledge-analysis. Klein, Peter. 1981. Certainty: A Refutation of Scepticism. Minneapolis, MN: University of Minnesota Press. Kornblith, Hilary. 2001. “Does Reliabilism Make Knowledge Merely Conditional?” Philosophical Issues 14: 185–200. Lonergan, Bernard J. F. 2004. Collected Works of Bernard Lonergan, Vol. 17. Toronto: University of Toronto Press. Maitzen, Stephen. 2009. “Ordinary Morality Implies Atheism.” European Journal for Philosophy of Religion 1: 107–126.

146  Stephen Maitzen Maitzen, Stephen. 2010a. “Does God Destroy Our Duty of Compassion?” Free Inquiry 30: 52–53. Maitzen, Stephen. 2010b. “On Gellman’s Attempted Rescue.” European Journal for Philosophy of Religion 2: 193–198. Maitzen, Stephen. 2013. “Atheism and the Basis of Morality.” In What Makes Us Moral? edited by A. W. Musschenga and Anton van Harskamp, 257–269. Dordrecht: Springer Publishing. Maitzen, Stephen. 2014. “Agnosticism, Skeptical Theism, and Moral Obligation.” In Skeptical Theism: New Essays, edited by Trent Dougherty and Justin P. McBrayer, 277–292. Oxford: Oxford University Press. Meynell, Hugo A. 1982. The Intelligible Universe: A Cosmological Argument. London and Basingstoke, UK: The Macmillan Press Ltd. O’Connor, Timothy. 2013. “Could There Be a Complete Explanation of Everything?” In The Puzzle of Existence: Why Is There Something Rather Than Nothing? edited by Tyron Goldschmidt, 22–45. New York, NY: Routledge. Overall, Christine. 2006. “Miracles, Evidence, Evil, and God: A Twenty-Year Debate.” Dialogue 45: 355–366. Schlosshauer, Maximilian, Kofler, Johannes, and Zeilinger, Anton. 2013. “A Snapshot of Foundational Attitudes Toward Quantum Mechanics.” Studies in the History and Philosophy of Modern Physics 44: 222–230. Sherry, Patrick. 1983. “Review of Hugo A. Meynell, The Intelligible Universe: A Cosmological Argument.” Philosophy 58: 129–130. Sosa, Ernest. 1994. “Philosophical Scepticism and Epistemic Circularity.” Proceedings of the Aristotelian Society, Supplementary Volumes 68: 263–290. Stroud, Barry. 1994. “Scepticism, ‘Externalism’, and the Goal of Epistemology.” Proceedings of the Aristotelian Society, Supplementary Volumes 68: 291–307. Wielenberg, Erik J. 2010. “Sceptical Theism and Divine Lies.” Religious Studies 46: 509–523. Williamson, Timothy. 2011a. “On Ducking Challenges to Naturalism.” The Stone, New York Times, September 28. http://opinionator.blogs.nytimes.com/2011/ 09/28/on-ducking-challenges-to-naturalism. Williamson, Timothy. 2011b. “What Is Naturalism?” The Stone, New York Times, September 4. http://opinionator.blogs.nytimes.com/2011/09/04/what-is-naturalism. Wolchover, Natalie. 2014. “Fluid Tests Hint at Concrete Quantum Reality.” Quanta Magazine, June 24. www.quantamagazine.org/20140624-fluid-testshint-at-concrete-quantum-reality.

7 The Absurdity of Life in a Christian Universe as a Reason to Prefer that God Not Exist Erik J. Wielenberg It is sometimes claimed that life in a godless universe is absurd. That claim has been the subject of considerable philosophical debate. More recently, Thomas Nagel’s remark that he doesn’t want God to exist has triggered a debate about whether it is rational to prefer God’s non-existence to His existence. Much of the literature on that issue concerns the axiological impact that God’s existence or non-existence would have. Here, I make the case that the existence of the Christian God who has commanded us to love our neighbours as we love ourselves makes life absurd in the sense that a universe in which such a God exists includes truths that most human beings would find it difficult to accept and live happily. I then connect that result with the debate about the rationality of desiring that God not exist. I argue that the fact that the existence of the Christian God makes life absurd in the sense just explained is a reason to prefer that such a God not exist.

1. Introduction It is sometimes claimed that life in a godless universe is, in some sense, absurd. That claim has been the subject of considerable philosophical debate. More recently, Thomas Nagel’s remark that he doesn’t want God to exist (1997, 130) has triggered a debate about whether it is rational to prefer God’s non-existence to His existence. Much of the literature on that topic concerns the axiological impact that God’s existence or nonexistence would have. Here, I make the case that the existence of the Christian God (as defined below) makes life absurd (in a sense defined below). I then connect that result with the debate about the rationality of desiring that God not exist. I argue that the fact that the existence of the Christian God makes life absurd is a reason to prefer that such a God not exist. What does it mean to say that life is absurd, and what is it exactly about a godless universe that is supposed to make life absurd? William Craig is a prominent defender of the absurdity claim; he seems to have in mind the claim that in a godless universe, it would be very difficult

148  Erik J. Wielenberg or impossible for human beings to face certain aspects of reality without experiencing significant negative psychological consequences such as madness or despair. Craig asserts that it is impossible for us to live “consistently and happily” in a godless universe (1994, ch. 2). Accordingly, I shall work with the following definition of absurdity: Claim C makes life absurd = df. Claim C’s truth makes (or would make) true at least one claim C1 such that most (actual) human beings are such that if they were to accept C1 they would experience negative psychological consequences that would make it difficult or impossible for them to be happy (without also failing to accept at least one entailment of C). Some comments on this definition are in order. First, I take the making true relation employed in the first part of the definition to be distinct from entailment. I will not attempt to define what it is for one claim to make true another claim; a rough intuitive grasp of the concept will suffice for the argument I will give. An example from Michael DePaul may help to illustrate the distinction between entailment and making true. According to DePaul, William Paley maintained that the properties of being felicific (i.e., being productive of a greater balance of pleasure over pain than any alternative) and being commanded by God are both necessarily co-extensive with the property of being morally obligatory. But Paley also maintained that it is being commanded by God that makes a given action obligatory. Thus, being felicific entails that an action is obligatory as does being commanded by God; however, only having the latter property makes it true that an act is obligatory (DePaul 1987, 436–437).1 So, the making true relation employed in the definition above is a stronger relation than entailment. Second, the word ‘actual’ in parentheses in the definition is intended to indicate that the definition is indexed to human beings as they actually are in our world. When evaluating whether a given claim is absurd, we look to human beings as they actually exist. Third, the last parenthetical component of the definition is intended to capture Craig’s talk of living consistently. According to Craig, one can sometimes avoid the negative psychological consequences of accepting a certain claim by failing to recognize the full implications of that claim. Craig offers the case of Sartre as an example, claiming that Sartre accepted the claim that (a) God does not exist while at the same time holding that (b) one can create meaning in one’s own life. According to Craig, (a) entails that (b) is false and hence Sartre was able to live happily only by failing to accept a significant entailment of (a). Putative sources of the absurdity of life in a godless universe include the fact that life in such a universe lacks ultimate meaning and ultimate purpose (Craig 1994, ch. 2), the absence of any reason to be moral (Craig 1994, ch. 2), and the existence of moral obligations that, if fulfilled, make one worse off overall (Mavrodes 2002). I have argued elsewhere that at

Absurdity of Life in a Christian Universe 149 least some of these absurdity arguments fail because the absence of God does not have the alleged problematic implication; for example, I think that Craig is mistaken that human life in a godless universe is meaningless in any important sense (Wielenberg 2005; 2014). Here I argue that the claim that the God of traditional Christianity exists makes life absurd in the sense defined above. I further argue that such a result is a reason to desire that the God of traditional Christianity does not exist. Most people would find it quite difficult to live happily and consistently if they grasped the full implications of the traditional Christian view of reality, and that is a reason to desire that the Christian worldview is false.

2. The Main Argument Throughout this essay, I use ‘God’ to indicate the God of traditional Christianity. God so construed has at least the following features: He is omnipotent, omniscient, and morally perfect; moreover, He has these features essentially, meaning that it is impossible for Him to exist and not have these features (in terms of possible worlds, there is no possible world in which God exists and lacks any of these properties). He is also essentially triune, meaning that He is three persons (Father, Son, and Holy Spirit) in one God. Finally, I will assume that it is an essential feature of the Christian God that He has commanded us to love Him with all our hearts and to love our neighbours as ourselves (see Matthew 22:36–40). I am happy for this assumption to be taken as stipulative; my aim is to show that the existence of an essentially perfect being who has commanded us to love our neighbours as ourselves makes life absurd. I use the term ‘God’ to refer to such a being. With that in mind, let us consider my main argument. The Absurdity Argument 1. Necessarily, if God exists, then whenever a person P experiences undeserved involuntary suffering, P is better off overall than P would have been without the suffering. 2. So: Necessarily, if God exists, then whenever a person A causes another person B to experience undeserved involuntary suffering, B is better off overall than B would have been without the suffering (from 1). 3. God’s existence makes it true (or would make it true) that each of us is morally obligated to pursue the good of others. 4. Necessarily, if (i) A is morally obligated to pursue B’s good and (ii) A’s performing act X would make B better off overall, then (iii) A has a fact-relative reason to perform X. 5. So, God’s existence makes it true (or would make it true) that C: each of us has a fact-relative reason to cause others to experience undeserved involuntary suffering (from 2, 3 and 4).

150  Erik J. Wielenberg 6. Most human beings are such that if they were to accept (C), they would experience negative psychological consequences that would make it difficult or impossible for them to be happy (without also failing to accept at least one entailment of (C)). 7. Therefore, the claim that God exists makes life absurd (from 5 and 6).2 In the rest of this section, I defend the premises of this argument. Following that, I consider and respond to some objections. Finally, I make the case that if the conclusion of the argument is true, that constitutes a reason to desire that God (as defined above) not exist. Defence of (1) This premise is accepted by many (but not all) contemporary Christian philosophers as well as many (but not all) contemporary philosophical critics of Christianity.3 Part of the justification for the principle is that a morally perfect God would not permit the existence of any gratuitous evil, evil that is not necessary in order to prevent an equal or worse evil or necessary to produce some great good. However, the claim that God’s existence entails the non-existence of gratuitous evil does not by itself entail premise (1). Premise (1) depends on the further claim that if God permits a certain evil to befall a particular individual, God’s moral perfection requires not merely that the evil be compensated for somewhere in the universe but that the evil be compensated for in the life of the very person who endures that evil. Premise (1) deals with the specific evil of suffering, claiming that a morally perfect God allows a person to experience involuntary and undeserved suffering only if such suffering ultimately makes that person better off overall than she would have been otherwise (call suffering of this sort “patient-benefitting suffering”, or simply “pb-suffering”). One justification for this stronger claim is advanced by Maitzen, who argues that if God allows a person to experience involuntary and undeserved suffering in order to prevent a great evil or attain a great good but not for the sufferer, then God is treating the sufferer as a mere means to an end, which is incompatible with God’s moral perfection (2009, 115–117). I take it that Maitzen provides at least a strong prima facie case for (1), so that anyone who would reject it has some explaining to do. And indeed, the idea that suffering ultimately benefits the sufferer is a prominent theme in traditional Christianity.4 Furthermore, the fact that (1) is so widely accepted among contemporary philosophers of religion means that drawing out its full implications is of philosophical interest for that reason alone. Because an omniscient and omnipotent God is aware of and could prevent any instance of involuntary and undeserved suffering that occurs, any such suffering that does occur is permitted by God. Since the God of traditional Christianity is essentially omniscient, omnipotent, and

Absurdity of Life in a Christian Universe 151 morally perfect, it follows that the existence of God entails that all involuntary and undeserved human suffering is pb-suffering. Defence of (2) Suppose that A causes B to experience some involuntary and undeserved suffering. From (1), God’s existence entails that B’s suffering is pb-suffering. Thus, one consequence of A’s act is that B is made better off overall than B would have been otherwise. God’s existence means that human beings can exploit a kind of divine “fail-safe device” (Jordan 2004, 174) to benefit one another by causing each other to experience undeserved and involuntary suffering. We can imagine a universe in which benevolent prayers for others are 100 percent effective. In such a universe, praying for others is a guaranteed way of making them better off overall. Similarly, premise (2) tells us that inflicting involuntary and undeserved suffering on others is a guaranteed way of making them better off overall. Defence of (3) According to Jesus, the two greatest commandments are to love God with all your heart and to love your neighbour as you love yourself (Matthew 22:26–30). Various views have been advanced regarding the full meaning of the second of these commandments. However, a central element of the traditional Christian view is that that loving someone includes desiring the good for that person.5 An important wrinkle here is that our desires are not under the direct control of our wills in the way that, for example, many bodily movements are. Under normal circumstances, I can make my arm go up simply by deciding to raise it; however, I often cannot make myself have a certain desire simply by deciding to have that desire. Recognizing this point, some thinkers have suggested that what we ought to do in light of the command to love our neighbours is to act as if we love them in the (reasonable) hope that over time we will come genuinely to love them. C.S. Lewis puts the idea this way: “The rule for all of us is perfectly simple. Do not waste time bothering whether you ‘love’ your neighbour; act as if you did. As soon as we do this we find one of the great secrets. When you are behaving as if you loved someone, you will presently come to love him” (1955, 114). In light of this, we can say that among the moral requirements imposed by the command to love others is the obligation to do what we can to promote what is genuinely good for others (which is what I mean when I speak of an obligation to pursue the good of others). Thus, the existence of God (as defined earlier) entails that God has commanded each of us to love others as we love ourselves and hence the existence of that God makes it true that each of us is morally obligated to pursue the good of others. Furthermore, Jesus’s words make it clear that such an obligation

152  Erik J. Wielenberg is not merely one among many. Rather, it is one of our most important moral obligations. Defence of (4) Consider Derek Parfit’s distinction between fact-relative and evidencerelative reasons (2011, 150–151). Parfit’s idea is that sometimes we have reasons to act in a particular way that are not revealed to us by the evidence we have available to us. If A in fact could act in such a way that would make B better off overall than B would be otherwise, then A has a fact-relative reason to act in that way. However, if A’s evidence does not reveal to A that she could benefit B in this way, then A lacks an evidencerelative reason to cause B to suffer. Accordingly, premise (4) claims that if A is morally obligated to pursue B’s good and there is something A could do such that if A were to do it, B would be better off overall, then A has a fact-relative reason to perform X. This fact-relative reason is defeasible because the fact that A could make B better off overall by performing some act X might be outweighed or overridden by other features of the situation so that A’s all-things-considered moral obligation is to perform some other act. For example, perhaps X, while benefitting B, would harm a million other people; or perhaps A has made an important promise not to perform X, and so on. However, as I noted in the defence of (3), Jesus’s words indicate that our obligation to promote the good of others is a particularly weighty one, so the fact-relative reason to benefit another will not be easily outweighed. Defence of (5) Suppose that God exists. (3) tells us that we have an obligation to pursue the good of others and (2) tells us that one effective way of promoting another’s good is to cause her to experience involuntary and undeserved suffering. From (4), it follows that each of us has a (defeasible) factrelative reason to cause others to experience involuntary and undeserved suffering. Ordinarily, we think that the default position with respect to others is that we should not cause them to suffer—we need a good excuse for causing another to suffer. Claim (C) says that if God exists, then the opposite is true: the default position with respect to others is that we should cause them to suffer—we need a good excuse for refraining from causing another to suffer.6 Defence of (6) Any attempt to establish that a given claim makes life absurd in the sense defined above will consist of two main parts. One part will be philosophical: that part consists of making the case that the target claim has certain

Absurdity of Life in a Christian Universe 153 salient implications. The other part will be psychological: that part consists of making the case that acceptance of the salient implications will have significant negative psychological consequences for most human beings. Once these two components are distinguished, it is apparent that any plausible version of the second component will be informed by work in psychology. Premises (1)–(5) of my argument are devoted to bringing out the salient implications of the existence of God. It is here in defence of premise (6) that psychological considerations come to the fore. To begin, we can note that it seems likely that most human beings would find (C) to be depressing and demoralizing. I take it that (C) strikes most of us as morally perverse. It runs contrary to the natural moral tendencies of most human beings; our natural inclination is to believe its opposite, namely that we all have reason to avoid inflicting involuntary and undeserved suffering on others. Thus, those who accept (C) would likely find themselves continuously struggling to hold onto a depressing belief that contradicts what they are naturally inclined to believe. This is an obstacle to happiness that would afflict the vast majority of humans who believe (C). To see some additional obstacles to happiness, we can divide believers in (C) into two groups: those who succeed in acting in conformity with their belief in (C) and those who fail. Let us consider the first group first. Those who accept (C) will be forced to conclude that they ought to be inflicting extreme suffering on a great many people. If (C) is true, then whenever one can inflict involuntary and undeserved suffering on another—no matter how extreme—one has a fact-relative reason to do so. In light of the weightiness of this reason, those who accept (C) will often be forced to conclude that they have an actual moral obligation to inflict such suffering. In many cases this will require the infliction of suffering in an up close and personal way. Putting aside sociopaths (who constitute less than 5 percent of the population), human beings find it unpleasant to witness or even imagine the suffering of others. This aspect of human nature has been widely recognized since at least the fourth century BC when Mencius claimed that “all men have a mind which cannot bear to see the suffering of others.” Most people find it quite unpleasant to inflict significant suffering on others, even when they believe that doing so is justified. As a mundane example, many parents find it unpleasant to help (or even permit) doctors to give their children important but painful vaccinations even when such parents fully grasp the justification behind such vaccinations. A less mundane example involves the general reluctance of soldiers in combat situations to kill the enemy. Dave Grossman declares: “[T]he vast majority of combatants throughout history, at the moment of truth when they could and should kill the enemy, have found themselves to be ‘conscientious objectors’ ” (1995, xv). Grossman argues convincingly in his extended study of killing (and the avoidance of killing) in war that one of the psychological costs of up-close killing—even when seen by its perpetrators

154  Erik J. Wielenberg as justified—is guilt.7 Indeed, the guilt associated with up-close killing often produces “physical revulsion and vomiting” on the part of the killer (Grossman 1995, 115). Furthermore, the guilt of such killing is a significant contributor to psychiatric casualties, much more so than fear of death or injury to oneself: “Balancing the obligation to kill with the resulting toll of guilt forms a significant cause of psychiatric casualties on the battlefield” (1995, 91). To accept and live in accordance with (C) is to make inflicting suffering on others a central aspect of one’s life; unless one is a sociopath, this is a recipe for severe psychological trauma. Thus, most people who act in conformity with (C) will experience significant negative psychological consequences that will make happiness difficult or impossible to obtain. Of course, not all who recognize the existence of an obligation to impose suffering will actually do so; some will instead experience weakness of will, failing to do what they genuinely believe they morally ought to do. People in this category will be susceptible to what psychologists have labelled “moral injury”, a psychological affliction that occurs when one acts in a way that runs contrary to one’s most deeply held moral convictions. Journalist David Wood characterizes moral injury as “the pain that results from damage to a person’s moral foundation” and declares that it is “the signature wound of [the current] generation of veterans [of war]” (2014).8 As Wood notes, moral injury is sometimes a result of killing that one judges to be wrong; however, it is also sometimes a result of failing to fight as or when one thinks one ought to. It is this second sort of moral injury to which those who genuinely believe they ought to inflict involuntary and undeserved suffering on others but fail to do so will be susceptible. The effects of moral injury are comparable to those of PTSD; chronic and repeated moral injury will make it difficult to be happy. Additionally, Stanley Milgram’s infamous obedience experiments dramatically illustrate the psychological costs of failing to do what one believes one ought to do. Milgram’s subjects were instructed to administer what they thought were increasingly severe electric shocks to an increasingly upset (and ultimately unconscious) victim. Most relevant here is the fact that most obedient subjects exhibited a state of conflicted obedience—they obeyed the instructions to inflict the shocks but were greatly upset at doing so because they believed that what they were doing was wrong. Most of Milgram’s fully obedient subjects experienced severe internal conflict. One subject broke into hysterical laughter each time he administered a shock; afterward, he explained his behaviour this way: This isn’t the way I usually am. This was a sheer reaction to a totally impossible situation. And my reaction was to the situation of having to hurt somebody. And being totally helpless and caught up in a set of circumstances where I just couldn’t deviate and I couldn’t try to help. This is what got me. (Milgram 1974, 54)

Absurdity of Life in a Christian Universe 155 Another obedient subject, upon learning after the experiment that the apparent victim was really just an actor, made the following remarks to him: I’m exhausted. I didn’t want to go on with it. You don’t know what I went through here. A person like me hurting you, my God. I didn’t want to do it to you. Forgive me, please. I can’t get over this. My face is beet red. I wouldn’t hurt a fly. (Milgram 1974, 82) A life in which one is essentially a life-long obedient subject in a Milgram experiment is one in which happiness will be elusive. Therefore, those who accept (C) face a terrible dilemma: face the psychic trauma associated with inflicting severe suffering on others or endure chronic moral injury. In this way, most people who accept (C) will find it difficult to be happy. Since the existence of the Christian God makes true (or would make true) (C), it follows that the claim that such a God exists makes life absurd.9 I turn now to some objections to the absurdity argument.

3. The Lack of Knowledge Objection In attempting to reconcile God’s goodness with human suffering, C.S. Lewis proposes that God uses suffering to improve our character in various ways, thereby benefitting us in the long run (2001, ch. 6). One worry raised by critics of Lewis’s proposal is that it would license the conclusion that human beings ought to participate in God’s good work by inflicting suffering on others. This worry has obvious similarities with the argument I have defended, so attending to how Lewis’s defenders respond to the objection to his view may point us toward potential weaknesses in my argument here. James Petrik defends Lewis’s view by proposing that differences in God’s knowledge and that of human beings imply that while it is permissible for God to use suffering as part of a character-building project it is not permissible for humans to do so. As he puts it, “an individual’s spiritual and moral development is an extremely complex affair” and so for us to be in a position to produce a benefit for another by causing her to suffer we would have to “know much more about the individual’s character, history, and present state of mind than we can possibly know” (1994, 51–52).10 However, because the issue at hand is whether (C) would make life absurd in the sense defined above, our question is not whether ordinary people are typically in a position to recognize the truth of (C) (if it is true) but rather what would be the psychological consequences for ordinary people if (C) were true and they recognized its truth. Consequently, any differences that exist between God and ordinary people in the actual

156  Erik J. Wielenberg circumstances with respect to knowledge of (C) is beside the point in the present context. Additionally, the fact that God sometimes uses suffering to produce good consequences does not entail that we have any reason to inflict suffering on others—but of course (C) does have that implication. Therefore, Petrik’s response to the objection to Lewis’s view does not apply to my Absurdity Argument.

4. The Character Damage Objection Eleonore Stump also considers an argument similar to the one I have advanced here. Adapting her suggestion to fit my particular argument yields the thought that inflicting involuntary and undeserved suffering on others has a negative impact on our own characters (even if it yields a net benefit for the sufferer) and that is a strong reason to refrain from inflicting such suffering. Many Christian thinkers take character formation to be extremely important. Lewis, for example, holds that it is the nature of one’s character that determines one’s ultimate fate, and that makes the impact of our actions on the character of others and ourselves a particularly important feature of actions (1955, 68–69; see also Hick 1966). Thus, the claim that the fact that a given act would harm one’s own character is a strong reason not to perform it coheres well with traditional Christianity. This proposal, if correct, suggests a way of rejecting premise (6) of my argument. The objection to that premise has it that much of (C)’s sting is lost if each of us has a compelling reason not to inflict involuntary and undeserved suffering on others—a reason so compelling that the reason we have to inflict suffering is trumped and hence we are never actually obligated to inflict suffering on others. One tempting reply to this objection is that the objection founders on the fact that those who understand that inflicting involuntary and undeserved suffering on others provides such others with a net benefit and who inflict such suffering precisely for that reason will not undergo character damage as a result of their actions; if anything, their characters will improve since their actions are altruistically motivated. While tempting, I think that this response is overly simplistic. In particular, it fails to take into account the destructive psychological effects of inflicting suffering that I discussed in my earlier defence of premise (6). It is plausible that such psychological effects may damage even the character of those who inflict suffering out of altruistic motivations. A more serious problem with the objection stems from the fact, mentioned previously, that in traditional Christianity the obligation to love others as we love ourselves is not merely one obligation among others; rather, it is identified as one of the two most important obligations, second only to loving God. In light of that, it is plausible that the fact that inflicting involuntary and undeserved suffering on another would

Absurdity of Life in a Christian Universe 157 damage one’s character is outweighed by the fact that one could promote the other’s good by inflicting such suffering and hence, all things considered, one ought to inflict the suffering. At best, those who accept (C) while recognizing the potential for character damage would find themselves in a state of moral uncertainty such that no matter how they acted they would have serious doubts about whether they had acted rightly— thereby making themselves vulnerable to moral injury. So it appears that considerations of character damage do not alleviate the dilemma faced by those who accept (C).

5. The Limits on the Right to Inflict Suffering Objection There is another objection that, like the previous one, aims to take the sting out of (C). According to Richard Swinburne, “[t]o allow someone to suffer for his own good . . . one has to stand in some kind of parental relationship towards him” (2004, 297). Based on this thought one might argue that the scope of (C) is limited to cases of parents inflicting involuntary and undeserved suffering on their own children. Maitzen likens involuntary and undeserved suffering in a theistic universe to a painful vaccine that produces a net benefit for anyone who receives it (2009, 111). That example supports the objection at hand as it is plausible that while parents have an obligation to subject their own children to painful vaccinations, this obligation does not apply to other people’s children. Thus, when it comes to inflicting involuntary and undeserved suffering on people to whom we have no parental relationship, we have a compelling reason not to inflict suffering on them that overrides our reason to make them suffer. This objection does little to mitigate the force of the argument. Indeed, focusing our attention on the case of parents and children serves instead to bring out the troubling moral implications of God’s existence. We ordinarily think that parents have a particularly strong obligation not to inflict involuntary and undeserved suffering on their own children; however, my argument suggests that in a theistic universe, parents have a fact-relative reason to do precisely that and, to the extent that they grasp the full implications of God’s existence, have a powerful reason to make their children suffer. In fact, precisely because they have the primary responsibility for promoting their children’s good, their obligation to impose the relevant sort of suffering on their children is enhanced. As before, those who genuinely accept such a moral view will find it difficult or impossible to be happy. Parents who accept (C) will understand that inflicting undeserved involuntary suffering on their children (the more the better!) is a guaranteed way to secure a net benefit for their children So the Swinburne-inspired objection, like the others we have considered, fails to defuse the absurdity argument.11

158  Erik J. Wielenberg

6. The Axiological Impact of God’s Existence and Nagel’s Fear of Religion Let us turn now to consideration of the implications of the conclusion of the absurdity argument. One natural thought is that the fact that the existence of God makes life absurd means that the existence of God somehow makes (or would make) the universe worse and that that is a reason to prefer that God not exist. To explore this suggestion, first consider the thesis that Kraay and Dragos call wide impersonal anti-theism (and which I shall call simply “anti-theism”): Anti-Theism: For each theistic world which is sufficiently similar to the actual world in the relevant ways, all else equal, the axiological downsides of God’s existence suffice to make that world far worse overall than it would otherwise be, on naturalism. (2013, 8) Suppose that the absurdity argument I have given succeeds. It might be thought that one could use the conclusion of that argument to argue for the truth of anti-theism. A natural way of developing such an argument would be as follows. Any theistic world that is sufficiently similar to the actual world in the relevant ways is one in which there is a truth that makes life absurd and hence is a world in which most human beings cannot live happily and consistently. Such absurdity constitutes a significant axiological downside of God’s existence. On the other hand, if naturalism is true, then the problematic absurdity is absent, and so is the axiological downside. I doubt that any argument of this sort can succeed. Such an argument faces two significant obstacles. The first is that God Himself is tremendously (perhaps infinitely) valuable, and so God will greatly (perhaps infinitely) enhance the overall value of every world in which He exists (Kraay and Dragos 2013, 16). It is plausible that the value contributed to a world by God’s existence outweighs the axiological downside of the absurdity of life. The second obstacle is that since God’s existence entails the non-existence of gratuitous evil, any evil that is a consequence of God’s existence will not make the world worse overall (Kraay and Dragos 2013, 15). Consequently, if the absurdity of life is a genuine evil, God’s existence entails that it does not make the world worse overall. I am also pessimistic about the prospects of appealing to the absurd nature of a theistic universe to reach the conclusion that God’s existence makes the world worse for some people. Kraay and Dragos call such a conclusion wide personal anti-theism and formulate it this way (I shall call this view simply “personal anti-theism”): Personal Anti-Theism: There is at least one person P such that: for each theistic world which (i) is sufficiently similar to our world in the

Absurdity of Life in a Christian Universe 159 relevant ways, and (ii) in which P exists, all else equal, the axiological downsides of God’s existence suffice to make the world far worse overall for P than it would be otherwise, on naturalism. (2013, 15)12 In defending premise (6) of my argument, I made the case that anyone who accepts (C) is on the horns of a dilemma, facing psychological distress resulting either from inflicting suffering on others or from moral injury. One might be tempted to argue that any psychological distress that results from this dilemma is an axiological downside of God’s existence, and hence people who experience such distress are made worse off overall than they would be in a naturalist world. However, because it is plausible that all such psychological distress (a kind of suffering) is involuntary and undeserved, it is plausible that the first premise of my argument entails that in a theistic universe all such psychological distress produces a net benefit for the sufferer and hence does not make the sufferer worse off overall. Therefore, as far as I can see, the absurdity argument I have defended here does not provide the materials for a promising argument for either anti-theism or personal anti-theism. However, recall that much of the debate over the axiological impact of God’s existence (or non-existence) appears to have been inspired by Thomas Nagel’s description of what he calls his “fear of religion.” Nagel writes: “It isn’t just that I don’t believe in God and, naturally, hope that I’m right in my belief. It’s that I hope there is no God! I don’t want there to be a God; I don’t want the universe to be like that” (Nagel, 130).13 It is significant, I think, that Nagel nowhere claims that God’s existence makes the universe worse. Noticing this points us toward the thought that the fact God’s existence makes life absurd is itself a reason to desire that God not exist (at least for those who are somewhat uncertain about whether God exists).14 Suppose you are uncertain as to whether p is the case. One reason you might have to desire that p not be the case is that if p were the case, then the universe would be worse overall (or worse for someone) than it would be if p were not the case. But that is not the only reason you could have to desire that p not be the case. I submit that if p’s being the case would make life absurd, then you have a reason to prefer that p not be the case—a reason that is entirely independent of whether p’s being the case would make the universe worse overall (or worse for someone). When I reflect on the nature of the theistic universe that is revealed by the absurdity argument I have defended, I find that I want our universe not to be like that. It seems to me that something is wrong with a universe in which human beings cannot face the important truth that God exists and live consistently and happily. I don’t want the universe to be absurd; that the Christian God would make the universe absurd is a reason to desire that the Christian God not exist.

160  Erik J. Wielenberg One objection to this line of thought starts with the claim that if God does exist, then a world containing God is the best of all possible worlds. Since it is not reasonable to desire that the best of all possible worlds not obtain, it is not reasonable to desire that God does not exist. In response, I offer the following example. Suppose you know that if you will suffer extreme agony for the next ten years, then the best possible life for you is one in which you suffer such agony. I think that it is rational for you to want not to experience the agony and to want the best possible life for you not to be one in which you suffer extreme agony for the next ten years. Similarly, if you know that if you will suffer extreme agony for the next ten years, then the best possible world is one in which you suffer such agony, then it is reasonable for you to want not to experience the agony and to want the best possible world not to be one in which you suffer the agony. Also similarly, even if we know that if God exists then this is the best of all possible worlds, it may be rational to desire that God not exist and to desire that the best of all possible worlds be one in which human beings can live consistently and happily while facing reality.15 Though I cannot argue for it here, I think that arguments for the view that the non-existence of God is absurd are unsuccessful (for critiques of some of these arguments, see Wielenberg 2005; 2014). If the claim that God exists makes life absurd and the claim that God does not exist does not make life absurd, then I think that this constitutes a reason to prefer God’s non-existence to His existence, for it is rational for us to want the best of all possible worlds to be one in which human beings can live consistently and happily.16 Of course, there may be other reasons for preferring that God exists.17 However, the existence of a reason to desire that God does not exist that has no corresponding reason for desiring that God does exist opens the door to the possibility that it is rational allthings-considered to desire that God not exist—even if God’s existence does not make the world worse overall (or worse overall for someone).18

Notes 1. I ignore here the historical question of whether DePaul’s interpretation of Paley is correct. For present purposes, it is enough if the view DePaul attributes to Paley is coherent. 2. This argument owes much to Jordan (2004) and Maitzen (2009). 3. See Stump (1985, 411) and Maitzen (2009, 109). 4. See Young (1998) for a helpful discussion of this theme. 5. See, for example, Thomas Aquinas, Summa Theologica I–II 26.4 and Lewis (1955, 113). 6. One possible example of a prominent Christian figure who has incorporated this sort of view into her thinking is Mother Teresa. As Christopher Hitchens notes, a number of observers and former volunteers in care facilities run by Mother Teresa’s organization claim that it was standard practice in such facilities to withhold life-saving care and pain relief that could easily have been provided (1995, 37–50; see also Jeffrey, O’Neill, and Burn 1994). While

Absurdity of Life in a Christian Universe 161 Hitchens perhaps goes too far in suggesting that Mother Teresa’s goal is “the promulgation of a cult based on death and suffering and subjection” (1995, 41), Mother Teresa’s remark to a person dying of cancer that “You are suffering like Christ on the cross. So Jesus must be kissing you” (1995, 41) together with the reports of intentional withholding of painkillers from the terminally ill documented by Hitchens points toward the acceptance of something like premise (5) by Mother Teresa. 7. While Grossman’s work focuses on killing, it is clear that much of what is true of killing is also true of inflicting non-fatal suffering. Furthermore, killing a given person’s loved ones is an extremely effective way of causing that person to suffer, so those who fulfill their obligations to inflict suffering will often engage in killing. 8. See also Shay (2014). 9. A natural objection here is that belief in the existence of the Christian God is also a source of tremendous comfort, including as it does the promise of forgiveness and hope of eternal union with God. Thus, it might be claimed that the tremendous comfort derived from belief in God will make it possible for those who accept (C) to attain happiness despite the negative psychological consequences of accepting (C). My response is that for this reply to work, an empirical case must be made that belief in God provides sufficient comfort to actual human beings to overcome the extreme psychological effects of accepting (C). Furthermore, the existing empirical research on the relationships between religious faith, guilt, and happiness is complex and as far as I can see, makes it unlikely that the sort of counter-argument described here could succeed. 10. Eleonore Stump makes a similar argument (1985, 412–413). 11. Still another strategy for depriving (C) of its sting has it that various elements in Christian moral thought (e.g., God’s commandments, the example of Christ) suggest that we have compelling fact-relative reasons not to inflict involuntary undeserved suffering on others. One problem with this strategy is that it appears to imply the existence of incoherence (or at least oddness) in Christian moral thought: if inflicting involuntary undeserved suffering on others always benefits them, it would be inexplicable or misguided for God to forbid inflicting such suffering (see Maitzen 2009, 114). 12. I have modified Kraay and Dragos’s formulation slightly to change it from the first-person to the third-person perspective. 13. The context in which Nagel makes these remarks suggests that his fear of religion is not as closely linked to the existence of God as the quoted passage suggests. In the discussion that precedes the quoted passage, Nagel characterizes as “alarming” the idea that there is “a natural sympathy between the deepest truths of nature and the deepest layers of the human mind” (Nagel, 130). He goes on to explain: “The thought that the relation between mind and the world is something fundamental makes many people in this day and age nervous. I believe this is one manifestation of a fear of religion which has large and often pernicious consequences for modern intellectual life” (Nagel, 130). Thus, Nagel does not view his own fear of religion as a good thing; he describes it as having “pernicious consequences.” The textual evidence suggests that in fact Nagel sees his fear of religion as an irrational prejudice that he is unable to free himself from rather than as a sensible, justified preference that God not exist. Nagel here is describing what we might label an “irrational fundamentality-of-the-relation-between-mind-and-world phobia” that he thinks afflicts many of us. These remarks also indicate that when Nagel speaks of hoping that there is no God and of not wanting the universe to be

162  Erik J. Wielenberg “like that”, the target of his hope and desire is a broader (and somewhat more nebulous) proposition than that God exists. This is confirmed later when Nagel writes that “the feeling that I have called the fear of religion may extend far beyond the existence of a personal god, to include any cosmic order of which mind is an irreducible and nonaccidental part” (1997, 133). 14. The caveat in parentheses is important because it’s not clear that it is reasonable for S to prefer that p be false if S knows that p is true; in that case, it may be reasonable instead for S to regret that p. (Thanks to Tim Mawson for this point.) 15. It might be objected that such a combination of desires is irrational because if it is true that the best of all possible worlds is one in which God’s existence makes life absurd, then it is a necessary truth that such a world is the best of all possible worlds. According to this objection, if a given proposition p is necessarily true if true at all, then it is not rational to desire that p is false. But notice that it is widely held that if God exists, then it is a necessary truth that He exists, and that feature of the claim that God exists does not seem to render it irrational to desire that it be false. So the objection rests on a false principle. 16. Remember that ‘God’ here refers to the Christian God as defined earlier; thus, even if successful, my argument does not establish that it is rational for us to prefer the truth of naturalism to the truth of any sort of theism. 17. See, for example, McLean (2015). 18. I presented earlier versions of this paper at the University of Birmingham (UK), Oxford University, and Ryerson University; I thank the audiences on those occasions for their helpful feedback.

References Craig, William Lane. 1994. Reasonable Faith (Revised Edition). Wheaton, IL: Crossway Books. DePaul, Michael R. 1987. “Supervenience and Moral Dependence.” Philosophical Studies 51: 425–439. Grossman, Dave. 1995. On Killing: The Psychological Cost of Learning to Kill in War and Society. New York, NY: Back Bay Books. Hick, John. 1966. Evil and the God of Love. New York, NY: Palgrave Macmillan. Hitchens, Christopher. 1995. The Missionary Position: Mother Teresa in Theory and Practice. London: Verso. Jeffrey, David, O’Neill, Joseph, and Burn, Gilly. 1994. “Mother Teresa’s Care for the Dying.” The Lancet 344(8929) (October): 1098. Jordan, Jeff. 2004. “Divine Love and Human Suffering.” International Journal for Philosophy of Religion 56: 169–178. Kraay, Klaas, and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 157–178. Lewis, Clive Staples. 1955. Mere Christianity. London: Fontana. Lewis, Clive Staples. 2001. The Problem of Pain. New York, NY: HarperCollins. Maitzen, Stephen. 2009. “Ordinary Morality Implies Atheism.” European Journal for Philosophy of Religion 2: 107–126. Mavrodes, George I. 2002. “Religion and the Queerness of Morality.” In God (2nd Edition), edited by Timothy A. Robinson, 74–89. Indianapolis: Hackett. McLean, Graeme. 2015. “Antipathy to God.” Sophia 54: 13–24.

Absurdity of Life in a Christian Universe 163 Milgram, Stanley. 1974. Obedience to Authority. New York, NY: Harper & Row. Nagel, Thomas. 1997. The Last Word. Oxford: Oxford University Press. Parfit, Derek. 2011. On What Matters, Vol. 1. Oxford: Oxford University Press. Petrik, James. 1994. “In Defense of C. S. Lewis’s Analysis of God’s Goodness.” International Journal for Philosophy of Religion 36: 45–56. Shay, Jonathan. 2014. “Moral Injury.” Psychoanalytic Psychology 31: 182–191. Stump, Eleonore. 1985. “The Problem of Evil.” Faith and Philosophy 2: 392–423. Swinburne, Richard. 2004. The Existence of God (Second Edition). Oxford: Clarendon Press. Wielenberg, Erik. 2005. Value and Virtue in a Godless Universe. Cambridge: Cambridge University Press. Wielenberg, Erik. 2014. Robust Ethics: Metaphysics and Epistemology of Godless Normative Realism. Oxford: Oxford University Press. Wood, David. 2014. “The Grunts: Damned If They Kill, Damned If They Don’t.” The Huffington Post, March 18. http://projects.huffingtonpost.com/ moral-injury/the-grunts. Young, Alexey. 1998. “Natural Death and the Work of Perfection.” Christian Bioethics 4: 168–182.

8 Recasting Anti-Theism Toby Betenson

In this chapter, I aim to offer a new defence of anti-theism. I will do this by casting doubt on the notion that anti-theism should be understood as the judgement that God’s existence ‘makes the world worse’. I will argue instead that anti-theism should be understood in terms of reasonable preferences that are not necessarily connected to rational judgements about the comparative value of possible worlds. I defend this form of anti-theism by showing that it is reasonable to detach our preferences from our rational judgements about which possible worlds are ‘better’ or ‘worse’. I will offer what I take to be some plausible analogies, analogies that show that we can reasonably prefer things that we judge to make our world worse, disprefer1 things that make our world better, and hold that, whilst we accept that something is an admittedly good thing, we would rather the world be different. If these analogies are plausible on their own, and if they are appropriately analogous to the axiology of theism, then it can be reasonable to prefer the non-existence of God even whilst accepting that God’s existence would be2 a good thing, or even that God’s existence would make the world better than it would otherwise be. In section 1, I briefly summarise the arguments against the coherence of anti-theism. I take these arguments to be entirely conclusive, such that anti-theism, understood as the judgement that God’s existence makes the world worse, will always fail. But because this argument, although seemingly successful, also seems to remain unpersuasive to the anti-theist, I consider whether it might be missing an important point (section 2). In section 3 I begin to offer a new defence of anti-theism, one that attempts to capture the point that was missed in the previous arguments. I claim that the previous arguments against anti-theism have been based upon a conditional, a conditional that has been taken to be representative of anti-theism: one can reasonably prefer the non-existence of God only if God’s existence makes the world worse than it would otherwise be. Previous arguments against anti-theism work to show that this necessary condition is not satisfied, since God’s existence does not make the world worse. But I argue that it’s not a necessary condition at all: God’s

Recasting Anti-Theism 165 existence ‘making the world worse’ is not the only reasonable justification for anti-theism. In section 4, I argue via analogy, and attempt to show that our rational preferences are often not aligned with our judgements about which worlds are ‘better’ or ‘worse’. The conclusion is that it can be perfectly reasonable to recognise that something ‘makes the world better’, recognise that such a judgement in some way rationally compels you to prefer to have that thing (rather than not have it), and yet still reasonably prefer that the world be different. If anti-theism can be seen in this way, then it can be reasonable to simultaneously recognise that the existence of God would make the world better than it would otherwise be and yet still reasonably not want the world to be like that!3 In section 5, I reflect on how we might respond to such a re-cast antitheism. It seems, on reflection, that many of these anti-theistic preferences, although reasonable, might appear trivial. I consider whether there might be anti-theistic preferences that are not vulnerable to an accusation of triviality, concluding that there might be. I take this non-trivial form of anti-theism to present the greatest challenge to pro-theism and other rival views: not, perhaps, sufficiently challenging to make anti-theists out of us all, but nonetheless leaving us with difficult questions about how to effectively respond to such an anti-theist.

1 It seems to me that anti-theism—understood as the preference for the non-existence of God—is a very difficult position to justify. Regardless of the reasons cited in its support—concerns about privacy, about independence and autonomy, about moral authoritarianism, about the downsides of immortality, about the comprehensibility of the natural world, etc.4—when anti-theism comes into contact with the debate concerning the problem of evil, its reasons must be seen to collapse. In the debate concerning the problem of evil, it is commonly granted by all parties that a good, powerful, and knowledgeable God would eliminate any bad state of affairs that it could, unless doing so would cost the loss of a greater good or the permission of a worse state of affairs. Such a God would look after us and would not let us suffer unnecessarily. God would not permit ‘gratuitous’ evils—evils for which there is no morally sufficient reason. But if this is true, then any ‘axiological downside’ (Kraay and Dragos 2013, 162) that the anti-theist chooses to offer in defence of their anti-theism must be (all things considered) justified by a greater good or prevention of worse evil. This, note, extends to all possible axiological downsides that the anti-theist can identify:5 whatever negatives are associated with the existence of God, if God exists, then they cannot be needless negatives: If God exists, ‘Whatever is, is right.’ If

166  Toby Betenson this ‘no gratuitous evil’ restriction is further strengthened by a ‘patientcentred restriction’—such that the morally sufficient compensation for the axiological downside must directly benefit the individual who suffers the negative effects of the downside—then this ‘philosophical judo move’ (Kraay and Dragos 2013, 167), which uses the force of anti-theists’ motivation against them, is conclusive against both ‘impersonal’ and ‘personal’ anti-theism. As such, the anti-theist is left with no good reason to prefer the non-existence of God. Anti-theism and the problem of evil are inexorably connected: If there is a good reason to prefer the non-existence of God, then there is an unjustified bad state of affairs that follows from God’s existence. And if there is an unjustified bad state of affairs, then God cannot be both perfectly good and perfectly powerful, and hence cannot exist. If God does exist, or if we are to assume for the sake of argument that God can exist, then there can be no legitimate reason supporting anti-theism; any such reason must be compensated by a good, powerful, and knowledgeable God. It seems to me that this conclusion is unavoidable: There can be no good reason to support the notion that God’s existence makes the world worse than it would otherwise be. God’s existence must be admitted to be a good thing, something we should want to be the case, regardless of whether we think it is likely to be the case or not. This means that many self-proclaimed anti-theists are mistaken in their preference for God’s non-existence. We should call on them to be more rational. After further reflection on the philosophical discussion of the problem of evil and the connections it has with the reasoning behind anti-theism, they ought to accept that their stated preference was not fully rational. Given the circumstances, they would in fact ultima facie prefer that a good and powerful God exist, even if such a prospect offers prima facie cause for concern.

2 Such has always been my view. And yet, when faced with this conclusion, I am inclined to say that it does a disservice to the anti-theist: it does not seem to be a fair account of their position. What prompts me to say this? Well, for one thing I am pessimistic that such an argument—pointing out the network of entailments identified in the problem of evil, and connecting that network with the reasons offered for anti-theism—would in any way change the anti-theist’s mind.6 For another, it seems to me that the previous arguments against anti-theism, although simple and (I think) correct, in some important sense miss the point. My purpose in this chapter is to try to capture the point that the former arguments missed. To do this, I aim to offer a defence of anti-theism that evades the previous objections and does greater justice to the anti-theist view, and I will do so

Recasting Anti-Theism 167 by recasting anti-theism in what I take to be more phenomenologically accurate terms: that is, less in terms of judgements about whether God ‘makes the world worse’, more in terms of wants and desires, or likes and dislikes.

3 Up to now, anti-theism has been understood largely in terms of God’s existence ‘making the world worse’. Guy Kahane, for example: (5) It would be far better if God exists than if He does not.  . . . The truth of (5) turns on the comparative value of possible worlds . . . would it be good and better if God exists and bad, and worse, if He does not? (2011, 677) And Tim Mawson: “To decide that it’s important whether or not A [God’s existence] obtains, we have to find a significant value difference between worlds in which A is true and worlds in which not-A is true” (2012, 102). And Klaas Kraay and Chris Dragos: “Some philosophers have held that God’s existence would make things worse, and that, on this basis, one can reasonably prefer God’s non-existence” (2013, 157). This seems perfectly reasonable, because likes and dislikes are prima facie understood to align with judgements about what ‘makes the world worse/better’, in some sense. For example, I like mint-choc-chip ice cream. From my point of view, mint-choc-chip ice cream makes the world better. I prefer mint-choc-chip to rum and raisin: I prefer the taste, it is more pleasurable to me, so having mint-choc-chip instead of rum and raisin increases the amount of pleasure in the world, and this makes the world better. If I were so inclined, I could look at the two possible worlds—one in which I have mint-choc-chip and one in which I don’t—compare them, and judge that the one in which I have mint-choc-chip is better. I don’t mind rum and raisin; it’s okay, but I’m not wild about it. Having rum and raisin would make the world worse, from my point of view, compared to having mint-choc-chip, even if having rum and raisin might make the world better than having no ice cream at all. Intuitively, likes and dislikes almost always align with judgements about whether the world is ‘better’ or ‘worse’ given certain circumstances. It would be counterintuitive to say that I prefer rum and raisin over mint-choc-chip, even though I judge that my having rum and raisin makes the world worse compared to having mint-choc-chip. The natural reaction would be to ask why on earth you prefer it over the alternative if it makes the world worse than the alternative. Because of this common intuition, anti-theism—understood as the preference for the non-existence of God—has been aligned with the

168  Toby Betenson judgement that God’s existence makes the world worse. And once you are working with that definition, you will immediately run into the network of entailments surrounding the problem of evil (as has been mentioned) resulting in the ‘philosophical judo move’ whereby the strength of your proffered axiological downsides is used against you. This route will always end in failure for anti-theism: God’s existence cannot coherently ‘make the world worse’; God’s existence must be admitted to be a good thing. Therefore, to defend anti-theism—understood as the preference for the non-existence of God—what is needed is to show a way in which it can (sometimes) be reasonable to prefer the non-existence of an admittedly good thing. This is broadly the position I will attempt to defend in this chapter. It seems to me that all of the arguments against anti-theism, up to now, have understood anti-theism to be fairly represented by a conditional: one can reasonably prefer the non-existence of God only if God’s existence makes the world worse than it would otherwise be. Previous arguments against anti-theism work to show that this necessary condition is not satisfied, since God’s existence does not make the world worse than it would otherwise be. But I argue that it’s not a necessary condition at all: God’s existence ‘making the world worse’ is not the only reasonable justification for anti-theism. My strategy here will be to argue by analogy. I offer various examples, cases in which it seems to be reasonable to prefer the non-existence of admittedly good things, cases in which our preferences can reasonably go against our judgements about whether the world is better or worse. These examples are intended to be analogous in the relevant ways to the case of anti-theism. If the examples manage to show that it can be reasonable to prefer the non-existence of admittedly good things, and the examples are relevantly analogous to anti-theism, then it might be reasonable to prefer the non-existence of God.

4 Example 1: Dieting I like cake (a lot). But I also like to be lean and trim. When I am dieting, I decide to prioritise my preference for being lean and trim over my preference for eating cake. I do this because I recognise that it is not really possible both to eat cake (in sufficient quantity) and to be lean and trim. These are the circumstances I am faced with, given the biological facts of humans and the nutritional facts of cake, so I order my preferences, arrange my behaviour in response to my likes and dislikes, in light of those circumstances. There’s nothing rationally compelling me to choose to (try to) be lean and trim, but what is rationally forced upon me, due to circumstances over which I have no control, is the notion that if I want to be lean and trim then I ought not to eat cake (much).

Recasting Anti-Theism 169 So, when dieting, am I ‘pro-cake’ or ‘anti-cake’? What is the appropriate characterization? In a sense, I am anti-cake, because I recognise it as something that (given my goals of being lean and trim) ‘makes the world worse’. But I would still describe myself as ‘pro-cake’, even in the deepest state of dieting: I like cake (a lot), and my occasional desire to be lean and trim doesn’t change that fact! There are two lessons from this, admittedly trivial, example: First, we can clearly detach preferences from judgements about what makes the world worse/better. It is entirely usual to have fairly firmly entrenched preferences that are unrelated to rational appraisals about what makes the world worse/better. Secondly, not only can we detach our preferences from our judgements about what makes the world worse/better, but we can even find ourselves being ‘pro’ things that we rationally judge to ‘make the world worse’. We very often prefer things that we fully accept make the world worse: Cake is bad for me, but I prefer to eat it anyway. Alcohol is probably bad for me, but I prefer to drink it anyway; smoking is very bad for you, but lots of people prefer to do it anyway; etc. Further, this doesn’t even seem to be particularly counter-intuitive; it’s a very common occurrence. And further still, just as we can prefer things that make the world worse, we can prefer the non-existence of things that make the world better. Again, examples of this are widespread and easily found. Consider health food. Consider security checks at airports. We do not like these things, we would prefer that we not have to have these things, but we recognise that they make the world better and so we put up with them (happily or not). This suggests that quotidian preferences are not particularly aligned with rational judgements. And so to argue against anti-theism—understood as the preference for the non-existence of God—by appealing to the rational judgement that God’s existence makes the world worse might start us on the path towards missing the point. Analogously, one could accept that God’s existence ‘makes the world better’ and yet still prefer that God not exist. But perhaps these lessons are not all that helpful. If I were perfectly rational—if that is a philosophical ideal—then perhaps I would be capable of orienting my preferences with what makes the world better. Perhaps it is my failing to remain ‘pro-cake’ in light of the circumstances. Perhaps if I were perfectly rational, I would shun my love of cake, just as I would encourage a smoker to shun his or her love of smoking: I would appeal to the smoker’s rationality in an attempt to change his or her mind, and I would be right to do so. Example 2: Vital Medication Imagine you suffer from an incurable disease. You have a very tough life. You are entirely dependent on daily medication for your survival. But taking the medication is unpleasant, an onerous routine that severely

170  Toby Betenson impacts upon the quality of your life. Taking the medication undoubtedly improves your life (indeed, it makes it possible); the medication is undeniably something that makes the world better. It would seem to follow that you ought, rationally, to prefer to have your medication. But must you? The point here is that if you were to suffer from an incurable disease, requiring unpleasant daily medication to survive, you would obviously want to have your medication, given your circumstances. But you would rather your circumstances were different! You would rather not need the medication in the first place. You would rather not have an incurable disease. And this would be, surely, an entirely reasonable preference. It seems, then, that though you are rationally compelled to be ‘pro’ your medication, because it makes your life possible, you are nonetheless ‘anti’ the world in which you need your medication. That is, you are ‘anti’ the world in which you are compelled to be ‘pro’ medication. You recognise that your medication makes the world better, but you don’t want the world to be like that! The second-order nature of this preference could be seen to be applicable to some of the other examples that have been mentioned. Consider security measures at airports. We all accept that they are necessary, that they prevent terrible things, that security is a good thing, that security measures make the world better: we are rationally compelled to prefer the existence of security measures. But we are, surely, ‘anti’ the world in which they are necessary. We don’t want the world to be like that! And when dieting, I am rationally compelled to recognise that cake is something that makes the world worse. But I don’t want the world to be like that! I am fiercely pro-cake, and I wish we could all eat cake and not get fat. These are strange preferences: they are preferences regarding possible worlds in which we are rationally compelled to accept certain judgements about which possible worlds are better or worse. Further, if there were a strong alignment of rational judgements with rational preferences, such that a rational judgement that a particular world is better/worse entails that we ought to prefer/disprefer that world, then these are also preferences regarding possible worlds in which we are rationally compelled to accept certain judgements about what we ought to prefer. Although the case is more complicated, I think we can plausibly recast anti-theism as a preference of this kind. Anti-theists are rationally compelled to accept that God’s existence makes things better than they would otherwise be, but they might still rationally prefer that the world not be like that. That is, they might prefer the possible world in which they are not rationally compelled to be ‘pro-theists’. In summary, anti-theism could look like this: ‘I accept that God’s existence would make the world better, but I don’t want the world to be like that! I accept that God’s moral authority entails universal justice, and

Recasting Anti-Theism 171 for this reason a Godly world is the better world. But it also entails that I have no ultimate privacy, no ultimate autonomy, no ultimate freedom. I wish the world were arranged in such a way that we could have a guarantee of universal justice but without having to forgo our full autonomy as human beings.’ God’s existence is seen to be a good thing, but we would rather the world be different. Analogously: ‘I accept that security measures make the world better, but I don’t want the world to be like that! I wish the world were such that security measures were not necessary.’ Or: ‘I accept that my medication is a good thing, but I don’t want the world to be like that! I wish I didn’t need my medication; I wish I didn’t have an incurable disease.’ Anti-theism simply becomes a recognition that there are axiological downsides to the existence of God, that the goodness of God’s existence does not come without cost. In this way, anti-theism can still be distinguished from pro-theism and can still provide a helpful distinction, even though both positions agree that God’s existence makes the world better. By way of illustration, consider this real-world example: When giving a talk to an undergraduate philosophy society on the topic of anti-theism,7 I encountered two clear reactions: One side of the room (the anti-theists) firmly did not want God to exist because they understood God’s existence to entail an objective morality; but they did not like the restrictions that this entailed, and they were anti-theistic because they took it that these restrictions implied that God is some kind of authoritarian moral dictator. They wanted to be free to choose what was right, what was wrong, how they should live their lives, etc. They did not want there to be one ‘right’ answer to the question ‘how should I live?’. The other side of the room (the pro-theists) firmly wanted God to exist for precisely the same reasons! That is, they wanted God to exist because they did not want the freedom to choose what was right, what was wrong, how they should live their lives, etc. They wanted there to be an objective ‘right’ answer to the question ‘how should I live?’, which they could simply discover and follow. Both sides, you could say, agreed on what follows from the existence of a good and powerful God. The anti-theists might even accept, with a little reflection, that they ought to rationally judge that God’s existence would ‘make the world better’. But they didn’t want the world to be like that! They wanted to be free to do what they wanted to do, and for what they wanted to do to be the right thing to do. And they wanted that world to be the better world, and for that better world to be the actual world. What differentiates pro-theists from anti-theists here is the recognition of the axiological downsides of God’s existence as being significant downsides. Pro-theists do not see God’s moral authority as being a bad thing; they like that there is an ultimate guarantor of universal justice, and they like that there are objectively right and wrong answers about how we should live our lives. Pro-theists do not see ‘violation of privacy’

172  Toby Betenson as being a cost, an evil necessary for some other greater good. Pro-theists typically like that theism makes possible intimate relations between God and creatures. Any of these preferences are perfectly compatible, in any combination, with any rational judgement about which world is ‘better’ or ‘worse’. For example, an anti-theist might (must) accept that God’s existence makes the world better than it would otherwise be, because, due to the existence of God, sufferings in this life are compensated in the next. However, they might still prefer that the world be such that sufferings in this life are compensated in the next, but not because of God— perhaps they prefer the idea of some kind of purely naturalistic process, or karma. These do not seem to be unreasonable preferences. My conclusion, therefore, is that a suitably recast anti-theism can be considered no less reasonable than our preferences regarding medication, security measures, healthy food, and the like. That is, whilst some people enjoy healthy food, some people eat healthy food because they know they must (but they would rather eat cake). Some people might prefer the feeling of safety and reassurance that comes with security measures, others might get very annoyed at having to take their shoes off at airports. But despite these differences, we can all agree that security measures are a good thing and that eating healthy food makes the world better (rather than the alternatives). Likewise, some people like certain features of God’s existence—the lack of privacy, the moral authority, etc.—and some people do not like these features. These preferences do not seem unreasonable. That these preferences might run contrary to rational judgements about which world is ‘better’ or ‘worse’ is not definitively relevant. What we are left with is a kind of reversal of the aforementioned ‘philosophical judo move’:8 The anti-theist attacks by positing an axiological downside of theism. The pro-theist counters by using the force of the anti-theist’s motivation against them, appealing to the ‘no gratuitous evil’ assumption and the ‘patient-centred condition’, showing that God would ensure that such a downside would be adequately compensated. The anti-theist accepts this, but further counters by pointing out that there is a possible world in which the compensation is achieved without the axiological downsides associated with God’s existence: there could be all the good with none of the bad, and wouldn’t that be better? Both agree that God’s existence makes the world better than it would otherwise be, but the anti-theist would rather the world be better without God.9 It appears that, just as with the original philosophical judo move, a suitably recast anti-theism can use the force of the pro-theist’s motivation against them: For any axiological upside associated with God’s existence, the anti-theist can always reply: Why could we not have that without God? If this is coupled with any notional axiological downside of God’s existence, then it would seem that one might be left with adequate reason to support anti-theism.

Recasting Anti-Theism 173

5 How could a critic respond?10 It would not be easy, since all axiological upsides of God’s existence are liable for judoistic reversal. One option would be to question, not the rationality of the preferences, but their depth and seriousness. For instance, how might we respond to someone who complains about having to stand in a queue and take off their shoes at the airport? There’s nothing irrational about their preference— standing in a queue and having to take your shoes off is annoying. But it would be a trivial reason on which to base a far-reaching preference about the existence of security measures. So whilst the call to ‘be more rational’ might miss the mark, the call to ‘be more serious’ might hit home. Such an appeal is a call to recognise depth in an otherwise flat conceptual domain. Imagine you are having a serious discussion about security measures, over the course of which you discuss the tragic loss of life that occurs when security fails or is absent, the sometimes discriminatory practices that seem to go hand-in-hand with a targeted security response, the plight of refugees who are barred from travel on the pretence of security . . . and one of your interlocutors raises the point that having to take your shoes off at airports is really annoying. Such a bathetic statement, whilst not being irrational or false, would show anyone who raised it to be shallow. We might not criticise them for being irrational—what they said is not wrong, after all: having to take your shoes off at the airport really is annoying—but we would ask them to recognise that there is a difference between such a shallow or superficial issue, and the deeper or more profound issues at play in the debate. So if the best that anti-theism can do is to insist, whilst recognising that God’s existence is a good thing, that they want all of the good but none of the bad, can this be seen as a really serious, non-trivial preference? Example 3: An Old Dog Can there be non-trivial preferences that are detached from rational judgements about the comparative value of possible worlds? I think so. Consider a less trivial example. Many of us have dogs and other animal companions. Dogs do not normally live as long as humans, so almost all of us who have dogs must deal with their decline into old age, ill health, and death. This is always traumatic. We care deeply about our animals, and it is right, or at least reasonable, that we do so. Only a cold-hearted person, or a person aspect-blind11 to the meaning of human-animal relationships, could deny the meaningfulness of this aspect of human life. But many of us who must deal with the decline of an animal will be faced with a difficult decision: Should the animal be put down? And if so, at what point? Whatever else might be said in favour or against such

174  Toby Betenson a decision, it is not reducible to rational judgements about what ‘makes the world worse’. Imagine if it were: We might ask, worse for whom, and in what respect? For me? My dog’s ill-health makes the world worse for me: I suffer to see it suffer. Perhaps I would be better off without my dog. Indeed, rationally it seems clear that I would be. Veterinary bills can be expensive, after all! But this seems to miss the point: it’s not really about me and my wellbeing. Perhaps how the situation affects me ‘personally’ is not as important as how objectively bad the situation is? Although it might be bad for me to spend a lot of money on veterinary bills, presumably it is good for the vet. And if the veterinary medicine is effective, presumably it is good for the dog too. Should I strike the balance between my, the dog’s, and the vet’s wellbeing, and at the moment that the suffering of one outweighs the benefit to the others, we kill the dog? I can get a new, younger, healthier dog, after all, one that can go on long walks and make the world much better overall (for me). The absurdity of this over-rationalisation is, I hope, obvious. In The Philosopher’s Dog, Raimond Gaita speaks movingly about a situation in which he and his wife are caused to work extra jobs in order to earn money to pay veterinary bills for their injured dog. He remembers, with shame, considering whether his dog would be better off being put down rather than having a leg amputated, or considering at what point he and his wife would be forced to say “No more. This time the dog will have to be put down” (2003, 25–27). Once again, whatever else might be said about such difficult decisions, it would not be reducible to rational judgements about what makes the world better or worse. If a helpful bystander were to call on such a dog owner to ‘be more rational; you are suffering, your dog is suffering, it’s costing you lots of money, it’s making the world far worse overall, you ought to end the situation and put it out of its misery—it’s just a dog!’, we might rightly respond that they were missing the point. Regardless of whether or not these considerations might feature in our decision, such considerations are not the only feature in the justification of our value judgements. It’s a similar point (though with important differences) that allows us to see dark humour in jokes about ‘putting granny down’ if she becomes a bit of a burden in later life, or in D. Z. Phillips’ talk of “rescuing a prostitute from degradation by telling her to charge higher fees” (2004, Chapter 3, section 7, para. 18).12 To my mind, this is clearly a case in which a) rational judgements about the comparable value of possible worlds are not aligned with preferences (such judgements are, in fact, entirely beside the point), and b) the call to ‘be more rational’ is misplaced. Rational judgements about the ‘value difference between worlds in which A is true and worlds in which not-A is true’ are incidental, at best. It is not an intellectual failing to act ‘irrationally’ with regard to beings we care about. I would say it is entirely

Recasting Anti-Theism 175 reasonable to prefer to give all you can for a dog, even if it makes your world worse personally, impersonally, in certain respects, or overall. The applicability of this analogy to the axiology of theism is probably easiest to see, in this case, via an analogy to pro-theism. Consider the pro-theistic version of the conditional at issue: one can reasonably prefer the existence of God only if God’s existence makes the world better. Is this really a necessary condition? Imagine that God’s existence did not, in fact, make the world better; imagine God’s existence made the world worse. Could it still be reasonable to prefer the existence of God? Well, by analogy, can it be reasonable to prefer to not have your dog put down, to prefer to go the extra mile to care for a friend, even if the rational judgement would be that doing so ‘makes the world worse’? Of course! Why? Because you love your dog, you have a duty to your dog, you respect your dog’s dignity and rightful place in your life, even if there are axiological downsides. Likewise, why might you prefer that God exist, even if He made the world worse? Perhaps because you love your God, you have a duty to your God, you respect your God’s rightful place in your life, even if there are axiological downsides. We can easily break the connection between the preference for God’s existence and the rational judgement that God’s existence makes the world better, and we can do so in a way that appeals to truly non-trivial preferences: but it does require that we are capable of seeing the meaning that these pro-theistic concepts can hold. Can we do the same for anti-theism? I think it is possible, but we would need to have an appreciative ear for the meaning that certain concepts might have for anti-theists. We would need to hear the objection to God’s moral authority as a sincere violation of individual autonomy, rather than as a childish rebellion. It would, to a certain extent, be up to anti-theists to show us that such a preference could be non-trivial, but I do not think we could dismiss it out of hand: We would have some minimal duty to take the anti-theist at their word, and at least try to come to see the value of individual autonomy under that aspect. Whatever else might subsequently be said against the anti-theistic preference, pointing out that God’s existence makes the world better will not touch such a suitably recast anti-theism, and hence neither will the call to ‘be more rational’. These appeals will miss the point. What is required here is an attempt to bring someone to see something from a different aspect.13

Notes 1. ‘Disprefer’ is probably not a real word, but I take it to mean simply the opposite of ‘prefer’. 2. ‘Would’ here indicates the possible world in which God exists. For the sake of argument, because we are comparing between possible worlds, I make no assumptions about whether or not that possible world is our world.

176  Toby Betenson 3. The oft-quoted Thomas Nagel (1997, 130): “I hope there is no God! I don’t want there to be a God; I don’t want the universe to be like that.” 4. Some of these axiological downsides (e.g., immortality) are not inherent to theism, but are commonly seen to follow from theism or are otherwise be closely associated with the existence of God. 5. I am assuming here that ‘evil’ is understood to mean any ‘bad state of affairs’. Perhaps not all bad states of affairs should really warrant the designation ‘evil’, but if these trivially bad states of affairs are so insignificant as to be incapable of playing a role in the problem of evil, then they will surely struggle to provide sufficient motivation for the anti-theist. 6. Having presented such arguments to a few anti-theists over the past few years, I can confirm that this is not idle speculation. 7. At Ryerson University, Toronto, in 2015. 8. A kaeshi-waza, if you will. 9. Consider the same attack-counter-reversal procedure in terms of one of our analogies: Attack: ‘I don’t like security measures. I hate having to queue for ages and take off my shoes.’ Counter: ‘But security measures make us safe! And that’s better than the alternative, surely?’ Reversal: ‘Okay, I accept that . . . security measures are a good thing. But I’d rather be safe without security measures.’ 10. There is the possibility of appealing to the intrinsic value of God’s existence— but I expect this is reasonably accepted to be ‘off the table’ in terms of a response to anti-theism (it would not leave much room for debate, after all). 11. I draw here on Wittgenstein’s (admittedly problematic) notion of aspect- or meaning-blindness: Could there be human beings lacking in the capacity to see something as something—and what would that be like? What sort of consequences would it have? Would this defect be comparable to colour-blindness or to not having absolute pitch?—We will call it ‘aspect-blindness’ . . . Aspect blindness will be akin to the lack of a ‘musical ear’. The importance of this concept lies in the connexion between the concepts of ‘seeing an aspect’ and ‘experiencing the meaning of a word’. For we want to ask ‘What would you be missing if you did not experience the meaning of a word?’ (1953, 213ff) For a helpful discussion of the problems that come with the concept of aspectblindness, see Wenzel (2010). 12. Phillips echoes Simone Weil’s thoughts on the language of rights: “relying almost exclusively on this notion, it becomes impossible to keep one’s eyes on the real problem. If someone tries to browbeat a farmer to sell his eggs at a moderate price, the farmer can say: ‘I have the right to keep my eggs if I don’t get a good enough price.’ But if a young girl is being forced into a brothel she will not talk about her rights. In such a situation the word would sound ludicrously inadequate” (1962, 21). 13. Many thanks to Klaas Kraay for his extensive and extremely helpful comments on earlier drafts of this work, to the participants of the ‘axiology of theism’ conference at Ryerson University in 2015, and to the John Templeton Foundation for supporting this project.

References Gaita, Raimond. 2003. The Philosopher’s Dog. London: Routledge. Kahane, Guy. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696.

Recasting Anti-Theism 177 Kraay, Klaas, and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 153–178. Mawson, Tim. 2012. “On Determining How Important It Is Whether or Not There Is a God.” European Journal for Philosophy of Religion 4: 95–105. Nagel, Thomas. 1997. The Last Word. Oxford: Oxford University Press. Phillips, D. Z. 2004. The Problem of Evil and the Problem of God. London: SCM Press. Weil, Simone. 1962. “Human Personality.” In Selected Essays, 1934–1943: Historical, Political, and Moral Writings. Oxford: Oxford University Press. Wenzel, C. H. 2010. “On Wittgenstein’s Notion of Meaning-Blindness: Its Subjective, Objective and Aesthetic Aspects.” Philosophical Investigations 33: 201–219. Wittgenstein, Ludwig. 1953. Philosophical Investigations. Translated by G. E. M. Anscombe. New York, NY: Palgrave Macmillan.

Part III

Connections between the Existential and Axiological Debates

9 Triple Transcendence, the Value of God’s Existence, and a New Route to Atheism J.L. Schellenberg It is generally assumed that existential and axiological issues—questions about whether God exists and questions about the value of God’s existence— can be kept apart in the debate that is the catalyst for this book. The stance most discomfiting for theists in that debate is supposed to be antitheism, not atheism. And anti-theism, so it is held, does not entail or collapse into atheism. The line of inquiry I want to develop challenges this view. After exposing the challenge, I will build on it in an attempt to determine whether there is anything new about the anti-theistic route to atheism. But for now: the challenge.

I I start with a picture. It’s a simple picture. Way over here, on one side, is what I shall call God-ness, the property of being an omnipotent, omniscient, and morally perfect creator of any universe there may be. Way over here, on the other side, is some particular property (perhaps a conjunctive property) at some time attributed to God and considered to be such as would make a world or our world—or perhaps someone’s world— worse with God in it than it would be if it were a naturalistic world, such as God’s inspection of the thoughts of other persons without their permission. (For present purposes, it won’t matter just what property we imagine, or whether it really makes a good starting point for anti-theism.) There’s a considerable distance between the two items in my picture. God-ness is obviously rather important, so let’s inspect it. Here’s my question about God-ness: is God-ness supposed to be a religious property? If we say no, then the arguments we develop about God may have no relevance to theistic religion. But I assume that anti-theists think their view is relevant to theistic religion. It would be odd for the anti-theist to say “Yes, I’m talking about a view of God, but not a religious view of God.” Or: “Yes, I’m talking about God, but what I have to say may have no relevance to religious believers.” In any case, a discussion with relevance to religion is, for me, the more interesting prospect. So let’s not say no. Let’s say yes: yes, God-ness is a religious property. If you disagree,

182  J.L. Schellenberg then please take the rest of what I say hypothetically: if we take God-ness as religious, then . . . So what if God-ness is taken as religious? Well, then God-ness must be seen as realizing what I have elsewhere called triple transcendence. What is triple transcendence? Start with the notion of a transcendent reality. The alleged reality providing the focus for many religious lives past and present is conceived as transcending—as being something more than or deeper than or greater than—mundane reality, where by ‘mundane reality’ or ‘the mundane realm’ I mean (to quote part of an earlier discussion of mine that remains relevant) “those aspects of human life and its environment to which just any mature human always has quick and natural cognitive and experiential access, what might (in two senses) be called the common elements of human life, which all who eat, drink, sleep, play, think, relate, and so on, will explicitly know and regularly encounter” (Schellenberg 2005, 11). Now the transcendent reality of religion is certainly conceived as something ‘more’ than anything mundane in factual or metaphysical terms, but if that’s all that can be said about it, this reality might very well turn out to be something discoverable by science and completely at home in a secular picture of the world. As so many examples suggest, what the religious have in mind is also something ‘more’ in inherent value and in what we might term importance, by which I mean its value for us. Let’s call these three sorts of transcendence metaphysical, axiological, and soteriological, respectively (I use the term ‘soteriological’ advisedly, recognizing that it is often employed in contexts narrower than mine). To say that something is metaphysically transcendent is to say that its existence is a fact distinct from any mundane fact and in some way a more fundamental fact about reality than any mundane fact (more fundamental in a broadly causal and explanatorily relevant sense). To say that something is axiologically transcendent is to say that its intrinsic value—its splendour, its excellence—exceeds that of anything found in mundane reality alone. And to say that something is soteriologically transcendent is to say that being rightly related to it makes for more well-being, fulfilment, wholeness, and the like for creatures than can be attained at the mundane level alone (this leaves open the possibility that spiritual well-being might in some way be attainable through mundane things). The different realities believed by practitioners to be at the heart of Jewish, Christian, Islamic, Hindu, Buddhist, and Taoist practice certainly appear to be regarded as ‘more’ in all three of these ways (and the same goes for other forms of religion as different from one another as North American aboriginal and ancient Greek); they are regarded as transcendent not just metaphysically, but also axiologically and soteriologically. It is at least in part by embracing these three together—by embracing triple transcendence—rather than the first sort of transcendence alone, so

Transcendence, Value, and a New Route to Atheism 183 I suggest, that religiousness is instantiated. And this is why, earlier, I said that, if religious, God-ness has to be seen as realizing triple transcendence. Having established a role for triple transcendence, we now need to notice the further point that in some examples of religion, one or more of the three sorts of transcendence are given a strong interpretation according to which the divine reality is ultimate in the relevant sphere(s) of transcendence: the existence of the metaphysically transcendent is also regarded as in some way being the ultimate fact; the axiologically transcendent is regarded as in some way having ultimate intrinsic worth; or the soteriologically transcendent is regarded as such that being rightly related to it makes for the ultimate good of creatures and the world. And—you guessed it!—theism is one of those cases. Indeed, theists through their worship and in other ways show that they regard God as ultimate in all three spheres of transcendence. So here we have not just triple transcendence but triple transcendence in a special form: triple ultimacy. If religious in a relevant way, this is what God-ness has to be seen as realizing. But how logically tight a link should we insist on here? Must Godness, on the grounds provided, be said to entail triple ultimacy? Let me say, first, that little in my challenge to the distinction I mentioned at the beginning will depend on how we answer the question. But more will hang on this later on. So let’s proceed with it. Notice that it is a real question. Religiousness and thus triple ultimacy might be regarded as belonging only contingently to God—say, only in worlds including God where creatures other than God exist who do not eternally luxuriate in finite perfection and fulfilment, and who are not restricted in their potential for well-being to states made available in that world’s counterpart for what I called ‘the mundane realm.’ Notice also that it’s only soteriological ultimacy that raises the question. God’s metaphysical and axiological ultimacy in every world in which God exists seem quite non-negotiable. Can the question be answered? I think it can, and in the affirmative. The advocate of entailment can, for example, say that in all worlds including God, a certain relation to God makes available an ultimate good for any creatures there may be who are capable of being fulfilled by such a relation. This seems to preserve and render applicable the idea of soteriological ultimacy even in worlds in which no one achieves or can achieve fulfilment in relation to God. I shall, at any rate, assume that it does; if you have qualms, I again ask you to set them aside and continue on with me hypothetically. So suppose that God-ness entails triple ultimacy. It follows that God’s existence rather evidently could not in any relevant way make a world or this world or someone’s world—the world of someone’s personal experience— worse than it would be if naturalism were true in that world. Axiological and soteriological ultimacy, in particular, must prevent this. What we have here Wittgenstein might have called a ‘grammatical’ claim about

184  J.L. Schellenberg theistic religion, and a theistic version of John Searle might view it as part of the ‘theistic Background.’ Indeed, if God is construed as providing for the flourishing of what exists apart from God in the thoroughgoing way that most theists would accept if queried, or even as making an opportunity for such flourishing generally available, then the idea of soteriological ultimacy alone shows that a world could not be made worse by God’s inclusion in it. But if all this is so, then, given the assumption we made when we said yes rather than no to the question whether God-ness is religious, the wind does appear to go out of the sails for the debate over the value of God’s existence; this debate then seems based on wrongheaded ideas and of relatively little interest. That might appear to make quite a good ending for any discussion of the matters addressed in this volume. But let’s push the inquiry a bit further. Drawing my attention to that particular property over at the other end of my picture, which we’ve been ignoring, you may insist that it is still the case that, given certain features of the world (and it is the world, our world, the actual world that anti-theists tend to focus on), that property is rightly causally linked to the world—or to someone’s world within it—being or becoming worse in the relevant way. And, moreover, you may point out that many theists include it in a description of the being to whom they give the name or title ‘God.’ At the risk of casting doubt on my identity as a Canadian, I am inclined to reply: “So what?” If that property really does make things worse, then theists are wrong to include it—just as it is wrong to think of God as literally being an old man with a long white beard. Indeed, believers now apparently have a proof that God does not possess that property. (God exists and is triply ultimate; no being with the property in question exists and is triply ultimate; so God doesn’t exist and possess that property.) The theists in question need to find a better description. Again, we seem to reach an early end to the inquiry, with little interest left to the debate with which we are concerned, except for any interest incidentally attached to such a change in someone’s description of God. But perhaps some theists will resist changing their descriptions of God in a relevant way. Perhaps the property in question will seem to them to really belong to God. What then? Well, then they will need to argue that what we think of as bad here, as making the world worse, isn’t really, or that although it is bad, there is or might be some—as we may say— countering fact consistent with triple ultimacy and preventing the world (or someone’s world) from being made worse by it, perhaps the existence of an appropriate outweighing good. If the latter is the route chosen, then of course what we see is just what we often see in response to the problem of evil, except that the bad thing, supposing it to be bad, isn’t something as ‘external’ to God as we tend to think of evil as being: rather it is something that God can’t help introducing because it is—allegedly— an inevitable product of God’s nature.

Transcendence, Value, and a New Route to Atheism 185 Now this move on the part of the theist either succeeds or fails. If it succeeds, the debate about the value of God’s existence again turns out to have limited interest—except for the new ripple sent through the problem of evil. Certainly in that case, the theist is not stuck with accepting the rationality of anti-theism. Indeed, then the anti-theist herself should cease to be one! If, on the other hand, the move fails, as presumably antitheists will think it does, then the theist who is alert to what theistic religiousness entails should, however reluctantly, start looking for a better description of God—and we are back to where we were before, without an intellectual prescription that makes the debate over the value of God’s existence more than incidentally interesting and significant. But you may still want to push me further. What if this move fails and a better description is difficult or impossible to find? Suppose that the property in question is part of the theist’s favourite form of theism, or of a traditional or culturally significant form of theism. Or suppose, worse, that it is entailed by every description involving a personal being with the attributes earlier mentioned in connection with bare God-ness. This is where things start to get interesting. Now we have a problem! But notice what sort of problem it is—and what sort it isn’t. We still don’t have something suggesting the rationality of anti-theism, as distinct from atheism—the validation of axiological concerns distinct from existential. No, for this problem to exist the former have to collapse into the latter. For now, a particularly important or indispensable theistic description or disjunction of personalist theistic descriptions of a reality purported to be Divine cannot be instantiated in the world in conjunction with triple ultimacy, and so, since God-ness depends on that conjunction, the nonexistence of God can be inferred. The point can be formalized as follows. Let ‘D’ stand for the claim that a particularly important or indispensable theistic description or disjunction of personalist theistic descriptions of a reality purported to be Divine is instantiated, ‘W’ for the claim that a relevant worse-making state of affairs causally linked to a property inescapably tied up with D is instantiated, ‘T’ for the claim that a triply ultimate reality exists, ‘&’ for conjunction and ‘→’ for logical entailment. Then given the scenario that alone makes the debate over the value of God’s existence interesting and significant, we have a swift and simple proof by reductio that the theist’s God does not exist which should be quite effective even for believers in that scenario (the first two lines are in bold to indicate that the conjunction of their contents is being assumed for reductio): (1) D & W. (2) T. (3) (D & W) → ~T. (4) ~T. (1, 3) (5) T & ~T. (2, 4)

186  J.L. Schellenberg Now we might question whether the theist is committed to (1), but that would simply be to question whether the anti-theist has got as far in the dialectic as we have allowed her to go; we might question whether the theist must embrace (2) as belonging to her idea of God, but that would be, stunningly, to question whether her view is religious; and we might question (3), but that would involve questioning whether triple ultimacy really is triple ultimacy. So it doesn’t seem that there is any way to make the present debate significant and interesting that doesn’t lead us straight into the arms of this reductio. Anti-theism collapses into atheism.

II But does it at least do so in an interesting way? Do we have here a new route to atheism? This is the next question I want to consider. It may seem that we should answer in the negative. Someone may suggest that it is quite in keeping with how our inquiry has been developed thus far to say that what we see here is no more than a side road of the superhighway known as the problem of evil. After all, if the theist explicitly builds triple ultimacy into her idea of God, then, using ‘G’ to stand for the claim that God exists, we might also have given the reductio this rather familiar form: (1) G. (2) W. (3) G → ~W. (4) ~W. (1, 3) (5) W & ~W. (2, 4) Although W has a unique source, it is still an instantiation of what we could just as well have called evil and represented using ‘E’ for the proposition that a certain evil exists. In short: ‘E’ could replace ‘W.’ This interpretation succeeds if we suppose that theists will rightly think religiousness to be not just contingently included in God-ness but entailed by it, and if we further suppose that it is only in the actual world, or in some worlds, on account of their peculiar features, and not in just any world in which God exists, that anti-theists will rightly wish to say W obtains. The former we have been assuming, and the latter represents how it usually is with the problem of evil. Pain and suffering, for example, occur in the actual world and in some other possible worlds, but there are possible worlds in which they do not occur. And so it might be with W. Thus it seems that the charge, against the anti-theistic atheist, of conventionalism— of not really having a new route to atheism—can be made to stick. However now notice that because of W’s peculiar source in the very nature of what the theist calls God, things might be thought to be different here, and the latter point might be thought not to apply. Perhaps

Transcendence, Value, and a New Route to Atheism 187 we could have something bad that a being satisfying the description D inevitably brings into any world in which it exists. Suppose this is what we have. Then, arguably, we have a problem structurally different from the problem of evil. If theism succumbs to the problem of evil, we can assume, it is because although the theist’s full description of God (which includes triple transcendence) remains coherent, on account of E we cannot suppose this description to be instantiated in the actual world. But here it is different. Here it is because the full description turns out not to be coherent that God’s existence must be denied. We seek to prove that God does not exist by showing, in connection with D and W, that in no world is a personal being of the sort alleged by theists to be Divine triply ultimate—from which it follows that God-ness, taken as by its very nature religious, cannot be instantiated. In these circumstances, the proof might be given the following shape (here it is no longer a reductio): (1) D → W. (2) W → ~T. (3) D → ~T. (1, 2) (4) ~T → ~G. (5) D → ~G. (3, 4) Or, if a reductio is desired: (1) D. (2) T. (3) D → W. (4) W → ~T. (5) D → ~T. (3, 4) (6) ~T. (1, 5) (7) T & ~T. (2, 6) So does the atheistic anti-theist have anything strong enough to make this happen—does he have a filling for ‘W’ that is realized in every possible world in which D is true? It is hard to see how he does, or even could, have something so strong. This is because in some worlds a vital condition of ‘worse-making’ appears not to obtain, since a being satisfying D exists alone, without creating anything else which we might imagine as made worse by the D-satisfying being’s existence than it would otherwise be. In reflection on such worlds, one might still wonder whether reality would be worse with the existence of such a being than it would be if no concrete being existed at all, but it seems to me that one would have to have a low opinion indeed—and an unrealistically low opinion—of the existence of an omnipotent, omniscient, and morally perfect person to move from such wondering to positive belief!

188  J.L. Schellenberg Now it might be that the incoherence in question can be established in some other way, a way that does not depend on D entailing W, but the argument of the anti-theist requires that we establish it using W. And for the reasons given, it does not seem that this can be done. But perhaps we might try to do it in the following way. We could recall that the entailment of triple ultimacy—and in particular of soteriological ultimacy—was preserved through the device of saying that, in every world including God, for any creatures there may be whose own properties make a relation of the relevant sort to God possible, God is soteriologically ultimate. And then we could argue that in all worlds in which D is true, fulfilment in or through a D-satisfying being must be stymied because of the effect of some property of that being, such as an Orwellian Big Brother property, or, more generally, an autonomy-removing property, or a property ensuring too neatly happy an ending for creaturely stories, or some other such thing—so that even if there were in that world creatures who could relate to the D-being, they would not be fulfilled by the relationship. This is an initially promising suggestion. But I do not think it succeeds in the end. That is because even if in some relevant worlds, perhaps including ours, the argument of the atheistic anti-theist applies, there are still others in which it does not, or does not clearly do so, and so the incoherence result is not established. To see this, notice that there is no reason to suppose that the D-satisfying being must create humans or human-like creatures, and no reason to deny that in some worlds it creates beings to be fulfilled in relation to it—beings whose properties are such as to make fulfilment of any other kind impossible. In such worlds the D-being would be soteriologically ultimate even if in others it would not. Thus the result needed for a distinctive route to atheism cannot here be claimed by the anti-theist.

III It seems, then, that the attempt to provide for the anti-theist a distinctive— a truly new—route to atheism fails. Anti-theism, when thought through, appears to collapse into atheism supported by a version of the problem of evil. But I want to conclude by taking this as a new starting point and going just one step further, making a connection between such atheistic anti-theism and a theme not usually mentioned when we discuss it—a theme which provides its own distinctive perspective on the question of the value of God’s existence and allows us to broaden discussion of these matters in illuminating ways. This is the theme of cultural evolution. Tied to this theme, anti-theism suddenly becomes part of a larger venture that is new. It is the cultural evolution of the past few centuries that has brought to the foreground the concerns that animate anti-theism: concerns for

Transcendence, Value, and a New Route to Atheism 189 such things as independence, autonomy, equality, privacy, or an escape from meaninglessness through pursuit of a complete understanding of the world. Concerns like these can make someone think that it would be worse for them or for the world if there were a personal God than it would be if naturalism—itself boosted by the successes of science, which belong to cultural evolution—were true. So the anti-theist who feels able to give the sort of argument from evil outlined at the beginning of the previous section of this paper is in a position to say the following: there are new insights, afforded by recent cultural evolution, that lead us to see a triply-ultimate God as preventing certain things that are bad in ways we had not considered (e.g., things made bad because they constitute or create meaninglessness for some human beings), and this with atheistic consequences (since, e.g., in the actual world the existence of a personal God would bring meaninglessness for some human beings). Now it’s just the other side of the same coin to say this: there are new insights, afforded by recent cultural evolution, that lead us to see a triply ultimate God as producing certain things that are good in ways we had not considered, and this with atheistic consequences. Initially this may seem confusing, for if we discover new ways in which God’s existence would be good, (i) how could we get an argument going from that which disproves God’s existence, and (ii) how could it be an argument from evil? To take (ii) first: it wouldn’t be an argument from evil and doesn’t have to be, since the sort of argument from evil the anti-theist may be in a position to present is here subsumed under a larger category which we might as well call ‘the axiology of atheism’, which requires us to think about both dis-value and value. As for (i): what we need to notice is that the aspects of God’s value-related nature that cultural evolution makes salient for us may lead us to expect good things that are absent from the actual world—such things as, for example, every person other than God who is nonresistant to such a state of affairs being in a position to participate, at some level, in conscious relationship with God. Yes, that’s right: a connection can be made here to the hiddenness argument. The hiddenness argument is not an argument from evil, but it might be said to be an argument from unrealized good. And it clearly owes some of its present prominence to insights (or what the advocate of the argument will call insights!) that are owed to cultural evolution: openness to relationship is, even for masculine types, part of the best love; God should not in any case uncritically be thought of as masculine; fathers should avoid being ‘distant’ from their children; people can have all sorts of religious stances for causes having nothing to do with ‘sinfulness’, and so on. How many other such atheistic good-oriented arguments, underwritten by cultural evolution, are waiting to be discovered? There is a sort of research programme waiting to be undertaken in this neighbourhood. And, returning to the ‘problem of evil’ side of things: we can ask whether there are other new insights here, afforded by cultural evolution

190  J.L. Schellenberg but independent of those by which the anti-theist is moved, that lead us to see a triply ultimate God as preventing certain things that are bad in ways we had not considered. I think there are. We humans of the twenty-first century keep going to war and regularly show our moral immaturity in all kinds of ways. But we are slowly growing up, and at the level of what we tend to believe about values, and even in the realm of relevant feeling, there have been significant changes. Steven Pinker acquainted us with one of these in his recent book The Better Angels of Our Nature: deeper and more widespread empathy. His whole book represents another: weaker dispositions of violence, and even a growing recognition of the power and value of non-violence that may end up making the Mennonites with whom I grew up, who advocate non-violence as a matter of principle, less uncommon in that respect—not as obviously outliers. There are many other signs of what I have in mind, and this also in religious circles. Rowan Williams wrote recently, in a New Statesman review, that [t]he paradox of our era in the modern North Atlantic world is that while we are probably objectively more secure against the casual daily risk of violence than our ancestors, we are more anxious and more outraged by the prospect as well as the reality of violence, and more prone to extend its meaning to forms of offensive or menacing speech and action that would not have registered for those ancestors. (2015, 62–63) What can we do with all this? Well, I foresee more arguments about how empathy and the permission of horrors are related; arguments about how non-violent dispositions would be part of any perfectly good God’s character; arguments about how rudeness is just unacceptable, and not to be expected in any developed moral personality. And so on. If, as I’ve been suggesting we might, we take the notion of value-related insights afforded by recent cultural evolution and leading to atheism as providing a new framework for philosophical discussion, and if we interpret antitheism as ultimately atheistic in the way I’ve described, then the concerns of the anti-theists are naturally construed as one part of this broader new approach to the problem of evil, which is itself one part of a broader axiology of enlightenment atheism. One theoretical advantage of situating anti-theism here, it may be noted, is that even if its particular concerns turn out to be answerable by pro-theists, as I think they may be, we will have dealt with just one way in which the approach it exemplifies can be advanced. So, perhaps, we do in the end have a new route to atheism with which atheistic anti-theism can be associated. One final thought: the emphasis on triple transcendence and triple ultimacy that lies behind all the arguments I have developed not only facilitates such arguments but also opens up a possible way of escape

Transcendence, Value, and a New Route to Atheism 191 from them, which, when the religions of the world more fully absorb the results of cultural evolution, may be a way more often taken. Anyone who accepts atheism who wants to push religiousness further, at our relatively early stage of evolutionary development, might take as the framework for religious behaviour and religious investigation triple transcendence alone, or triple ultimacy without details. Triple transcendence alone or triple ultimacy alone will not—or will not as easily—succumb to any value-based critique of the sort I have contemplated in this paper.

References Pinker, Steven. 2011. The Better Angels of Our Nature. New York, NY: Viking. Schellenberg, J. L. 2005. Prolegomena to a Philosophy of Religion. Ithaca, NY: Cornell University Press. Williams, Rowan (2015). “Violence is an Unavoidable Part of Being Human.” The New Statesman, August 6. https://www.newstatesman.com/2015/07/ be-human-be-violent.

10 Arguments from Evil and Evidence for Pro-Theism Myron A. Penner and Benjamin H. Arbour

Introduction Recently, philosophers of religion have been paying increased attention to the axiological implications of the existence of God.1 Rather than focussing on the epistemic requirements for rational belief in theism or atheism, these philosophers address questions related to the difference that God’s existence would make to the value of a world. For example, if God were to exist, would God’s existence, all else being equal, make things better or worse?2 But suppose the distribution of goods in a God-inclusive world is not equal: Is it plausible to think that God’s existence could make things worse in some ways, but better in others, such that the overall value contribution God’s existence would make is determinate one way or the other? And while the impetus for this emerging literature is axiological and not epistemological, interesting epistemic issues lurk nearby. For example, how could one acquire knowledge or justified belief about these sorts of axiological judgements? Do one’s views on the axiological contribution (positive or negative) of God’s existence support the rationality of certain preferences about what is (or might have been) the case? And so on. Following Guy Kahane, we identify pro-theism as the belief that God’s existence would be good in that it increases the value of a world, whereas anti-theism is the belief that God’s existence would decrease the value of a world (2011, 694).3 In this paper, we develop an argument for pro-theism that takes into account the structure of certain arguments from evil. More specifically, we argue that if one accepts a crucial premise in certain arguments from evil—namely, that certain observed features of the world are deficiencies that count as evidence against God’s existence—then one ought to be a pro-theist.

I.  On Evil as Evidence for Atheism A.  The Logical Problem of Evil Although interest in the logical problem of evil has waned since Alvin Plantinga presented his modal version of the free will defence,4 it is useful

Arguments from Evil; Evidence for Pro-Theism 193 to compare both logical and evidential formulations of arguments from evil in order to note a broad structural similarity. Here’s J. L. Mackie’s classic formulation of the logical problem: In its simplest form the problem is this: God is omnipotent; God is wholly good; and yet evil exists. There seems to be some contradiction between these three propositions, so that if any two of them were true the third would be false. But at the same time all three are essential parts of most theological positions: the theologian, it seems, at once must adhere and cannot consistently adhere to all three. (1955, 200) Notice that Mackie appears to be committed to the idea that God’s existence has empirical consequences, and thus that God’s existence is something for which we can test. This is because for Mackie, whether one is theist, atheist, or agnostic, one must conjoin the claim that God exists to a particular auxiliary assumption about what things would be like, empirically, if God were to exist.5 By itself the claim “God exists” might not be testable because it’s unclear what is entailed directly by the existence of God. As far as testability goes, this puts God in an epistemic category similar to many respectable entities postulated by science. For example, we can test for the Higgs boson not because the claim “the Higgs boson exists” is testable all by itself. Rather, we can test for the Higgs boson only if we bring to the testing all sorts of assumptions about what the world is like and what the world would be like if the Higgs boson were to exist. With respect to the logical problem of evil, the operative auxiliary assumption seems to be the following: A1 If God were to exist, there would be no evil to be observed.6 Thus, if A1 is true and one observes evil, then one has falsified the existence of God because in this case, any evil whatsoever is direct evidence against God’s existence.7 B.  The Evidential Problem of Evil Whereas the logical problem of evil attempts to demonstrate that there is a logical contradiction between God’s existence and evil, evidential arguments hold that God’s existence, while logically possible given the existence of evil, is rendered extremely unlikely given the quality and quantity of evil we see. And as the logical problem of evil requires an auxiliary assumption in order for evil to count against God’s existence, evidential arguments from evil also require certain auxiliary assumptions in order for observed instances of evil to count as evidence against the existence of God. For example, in order for the observation of gratuitous

194  Myron A. Penner and Benjamin H. Arbour evils to count as evidence against the existence of God, something like A2 must be assumed: A2 If God were to exist, there would be no gratuitous evils to be observed. A2 is importantly different from A1 in that while theists will typically deny that A1 is true,8 theists differ widely with respect to A2. Some think that God’s existence is compatible with gratuitous evil and thus deny A2.9 Theists who are theodicists think that A2 is true, but also claim that we can tell there are no gratuitous evils because they possess a true and comprehensive theodicy. Theists who are sceptical theists tend to think that A2 is true, but also hold that because of our limited cognitive powers with respect to moral cognition, there’s no good inference from the appearance of gratuitousness for some evil to the conclusion that the evil is in fact gratuitous.10 Switching focus from the possible justification or lack thereof for some evil, one might focus instead on the sheer scope and magnitude of some observed evils. The observations of massive devastation wrought by hurricanes, tsunamis, earthquakes, genocides, and the like will count as evidence against the existence of God if something like the following is true: A4 If God were to exist, there would be no massive evils to be observed. Our purpose in this paper is not to argue for or against any of these auxiliary assumptions. However, it is important to note that in order for any arguments from evil to succeed, be they logical or evidential in nature, one must assume that God’s existence would prevent some negative state of affairs from obtaining.11 Thus, we can see that A1-A4 are particular instances of a more general type of assumption: A5 If God were to exist in a world, some possible really bad feature would be precluded from that world.12 The general assumption can serve as a premise in the following argument schema that captures what all of the above arguments from evil have in common: Schema for Arguments from Evil I f God were to exist in a world, some possible really bad feature would be precluded from that world. S2   But as we see, that possible really bad feature of a world is present in the actual world.

S1  

Arguments from Evil; Evidence for Pro-Theism 195 On the basis of S1 and S2, suitably filled in to take into account the specific really bad feature of a world that would be prevented if God existed but is observed in the actual world, one can derive the conclusion that there is no God—just what arguments from evil infer.13

II.  The Argument from Evil to Pro-Theism A.  The Evil Evidence Argument With “RBF” standing for a really bad feature of a world and “Wn” picking out some possible world, here is our main argument according to which proponents of the argument from evil are committed to pro-theism: (1) All else being equal, if W1 has some RBF and W2 lacks that RBF, then W2 is a better world than W1.14 (2) According to arguments from evil, the actual world contains some RBF. (3) According to arguments from evil, a world in which God exists would lack that RBF. (4) So, all else being equal, according to arguments from evil, a world in which God exists would be a better world than the actual world. Premise (1) is an intuitive principle that employs the common-sense notion that an RBF—that is, a really bad feature of a world—is, well, really bad. As we’ll demonstrate below, much will turn on whether all else is in fact equal when comparing God-inclusive worlds to the actual world from an anti-theist’s perspective, but set that aside for now. Premises (2) and (3) reflect the common structure of the arguments from evil surveyed above.15 And (4), which follows from (1)–(3), is just a different way of stating pro-theism: namely, that God’s existence would add value to a world. B.  Applications to Various Anti-Theisms The infamous “all else being equal” clause serves an important function in (1) and (4) above. However, if one is a settled anti-theist, one might think that all is not equal when comparing God-inclusive worlds to Godexclusive worlds, because according to the anti-theist, God’s existence makes things worse in some respect(s). In exactly what respect God’s existence would decrease the value of the world depends on what type of anti-theism we’re talking about. Building upon Guy Kahane’s taxonomy concerning the axiology of theism, Klaas Kraay and Chris Dragos have argued that both pro-theism and anti-theism can be subdivided to account for the value impact God’s existence would have to a particular person (personal scope), or all things

196  Myron A. Penner and Benjamin H. Arbour in a way that is cut off from personal considerations (impersonal scope), in some particular respect (narrow scope), or overall (wide scope). Here’s how Kraay and Dragos describe these distinctions: It would be far worse overall if God exists than if He does not. It would be far worse overall for me Wide Personal Anti-Theism: if God exists than if He does not. Narrow Impersonal Anti-Theism: It would be far worse in certain respects if God exists than if He does not. It would be far worse in certain Narrow Personal Anti-Theism respects, for me, if God exists than if He does not. (2013, 159–160) Wide Impersonal Anti-Theism:

Combining the descriptions of various anti-theisms with the concept of an RBF, we can see that the RBF picked out by Wide Personal AntiTheism is one that results in things being worse overall for a particular person, whereas the relevant RBF on Wide Impersonal Anti-Theism is one that decreases the value of a world tout court. And we can see that the RBF picked out by Narrow Personal Anti-Theism is one that results in things being worse in some ways for a particular person, whereas the relevant RBF on Narrow Impersonal Anti-Theism is one that decreases the overall value of a world in certain respects. Thus, all forms of antitheism assume the following: A6  If God were to exist in a world, that world would include a really bad feature. An interesting contrast now comes into view. Earlier we claimed that various arguments from evil required the following general, auxiliary assumption in order for certain observations to count as evidence against the existence of God: A5 

I f God were to exist in a world, some possible really bad feature would be precluded from that world.16

It turns out that both proponents of arguments from evil and anti-theists make claims about the implications God’s existence would have with respect to RBFs. According to arguments from evil, God’s existence would result in a world better overall than the actual world; according to wide versions of anti-theism, God’s existence would result in a world that is worse overall than a world in which God does not exist.17

Arguments from Evil; Evidence for Pro-Theism 197 We are now able to see the way in which an anti-theist might attempt to reject premise (1) of the Evil Evidence Argument. Recall that according to (1), (1) All else being equal, if W1 has some RBF and W2 lacks that RBF, then W2 is a better world than W1. But here the anti-theist might protest that because A6 is true—that is, because God’s existence entails that some RBF of a world is instantiated— therefore things are not equal in the relevant respects when comparing a God-inclusive world to a God-exclusive world. She might say something like the following: Suppose it’s true as is suggested by arguments from evil, that God’s existence would preclude some possible RBF from being instantiated. Okay—but that doesn’t mean that God’s existence would result in an overall better world, because according to my anti-theism, God’s existence would result in some other even worse possible RBF being instantiated. Any world that lacks some possible RBF because God’s existence precludes it would also possess some possible RBF—the RBF that justifies my anti-theism. So, it’s still rational for me to be an anti-theist because God’s existence would make things worse overall. The anti-theist who attempts to make this sort of move bears a significant burden of proof in order to show that the RBF entailed by God’s existence is worse than the RBF that, according to arguments from evil, would be precluded by God’s existence. Notice what RBF’s would be prevented according to arguments from evil if God existed. On the logical problem of evil, God’s existence would prevent any evil whatsoever. On various evidential arguments from evil, God’s existence would prevent gratuitous evil, or perhaps seemingly gratuitous evil, or horrendous evil or perhaps all of these. Imagine the following sorts of worlds which, according to various arguments from evil, God’s existence would be sufficient for bringing about: a world in which there is no gratuitous evil—no instances of suffering that serve no greater moral good such that it couldn’t be achieved some other way; or a world in which there is no seemingly gratuitous evil because in such a world, if there were some evil, one would always be able to see its purpose; or a world in which there were no evils of the sort of scope and magnitude such that they qualify as horrendous evils; or a world in which there is no evil whatsoever. A consequence of adopting arguments from evil, we argue, is accepting that God’s existence would be a sufficient condition for bringing about a world like any of these. Essentially, the anti-theist described above is saying that whatever RBF is entailed by God’s existence is worse than the RBF precluded by God’s

198  Myron A. Penner and Benjamin H. Arbour existence. But what could possibly qualify as an RBF such that it would be worse than gratuitous evil, seemingly gratuitous evil, horrendous evil, or any evil whatsoever? The emerging literature on the axiology of theism suggests the following candidates for an RBF that justifies anti-theism:18 (a) Loss of Privacy: God would know everything about S, including the content of S’s reflective and pre-reflective cognition. (b) Loss of Autonomy: S would be subject to God as creature to creator and unable to dictate her own future in relevant respects. (c) Loss of Understanding: God’s existence means that naturalism is false, and that non-natural explanations are possible. But a world in which non-natural explanations are possible calls into question the rational intelligibility of the world. (d) Inability to Pursue Meaningful Life: If pursuing goods the existence of which is precluded by God’s existence is essential for S to achieve a meaningful life, then God’s existence means her life lacks meaning. It’s plausible to hold that privacy, autonomy, and understanding a rationally intelligible world are goods. Moreover, it’s true that God’s existence would preclude both absolute privacy and complete, unfettered autonomy. It’s also true that God’s existence would preclude the possibility of a complete, total explanation of all things in terms of a rationally intelligible naturalism. But in order for anti-theism to be justified on the basis of a loss of these or any other property P, not only must it (i) be the case God’s existence precludes P, but it must (ii) be the case that the loss of P is a bad feature of the world. Furthermore, if the anti-theist adopts some version of an argument from evil, it must also (iii) be the case that the loss of P is an RBF worse than the RBF precluded, according to arguments from evil, by God’s existence.19 Establishing (ii) and (iii) will be a tall order with respect to the loss of privacy, autonomy, and understanding. First, consider privacy. Is God knowing our thoughts really an RBF worse than preventing, say, gratuitous evil? This seems wildly implausible, especially when considering that God, as a perfectly good moral agent, would never use the information God has about me in a way that is unjust, harmful, or less than perfectly loving. Second, consider autonomy. It is true that if God exists, there are aspects of our existence such that we are not in control of them and we lack the ability to self-determine our own future. But unless we are ourselves omnipotent, we lack complete and unfettered autonomy whether or not God exists. Moreover, God’s existence doesn’t preclude significant and robust amounts of personal autonomy. Finite and imperfect moral creatures seem capable of a great deal of self-legislating autonomy that is utilized for harmful ends, such that it calls into question whether unfettered autonomy would be a great good at all; it certainly doesn’t seem to

Arguments from Evil; Evidence for Pro-Theism 199 be a good more valuable than eliminating the RBFs which, according to arguments from evil, are precluded by God’s existence. As concerns about autonomy go, it is important to mention another crucial connection to issues relating to problems of evil. Anti-theists who find the presence of moral evil in a world to constitute evidence against the existence of God in that world should consider the degree to which unfettered personal autonomy would be a good feature of a world. Moral evil is evil for which persons are morally responsible, which differs from natural evils wherein pain and suffering are caused by the likes of earthquakes, tornados, and the like. As we pointed out above, if personal autonomy can be utilized for harmful ends, then unfettered personal autonomy itself might be a RBF of a world.20 Third, a high-degree of rational understanding of the universe seems possible even if God exists. Consider, for example, the ever-increasing glory of science. With incredible fine-grained and collaborative precision, humanity is extending the bounds of knowledge at both the sub-atomic and cosmic scales. Finally, if one sees a dichotomy between one’s life having meaning and a world in which all evil is defeated or removed, that seems to be evidence that one’s view what constitutes meaningfulness of life is mistaken.21

Conclusion According to various arguments from evil, certain observed features of the world count as evidence against the existence of God because of the auxiliary assumption that if God were to exist, the world wouldn’t contain those features. This, we argue, commits proponents of those arguments to pro-theism: the belief that God’s existence would increase the value of a world. We further argue that there’s no good anti-theistic way out of the Evil Evidence Argument; the anti-theist faces an incredible burden in attempting to justify the good of God’s non-existence over the evils that would be prevented if God were to exist. One might find arguments from evil or anti-theism compelling, but adopting one precludes the other.22

Notes 1. For example, see Kahane (2011); Mawson (2012); Luck and Ellerby (2012); Kraay and Dragos (2013); Davis (2014); McLean (2015); Penner (2015); Penner and Lougheed (2015); Dumsday (2016); and Mugg (2016). 2. If the concept of God includes necessary existence, then if God exists in any world, God exists in every world, in which case considerations of “if God were to exist,” or of “if God doesn’t exist,” take one into the realm of counterpossibles, and discerning the meaning and truth-value of counterpossibles is difficult. However, even if the proposition “God exists” is necessarily true or necessarily false, there is still a sense in which we can consider and compare the value of two epistemically similar possible worlds, with the relevant

200  Myron A. Penner and Benjamin H. Arbour ontological difference being considered is God’s existing at one, but not both, worlds. For example, suppose no one in the actual world ever proves or disproves Goldbach’s Conjecture according to which every positive even integer greater than 2 is the sum of two prime numbers. Now consider a possible world W1 which includes the existence of someone who proves Goldbach’s Conjecture, and another world W2 which includes the existence of someone who disproves Goldbach’s Conjecture. There is a sense in which it’s possible to compare W1 and W2 with the actual world, even though either W1 or W2 involves a counter-possible, for Goldbach’s Conjecture, if true, is necessarily true and if false, is necessarily false. 3. Just because a world is better in certain respects doesn’t necessitate that world’s being better overall. We account for these nuances in detail further below. 4. See Plantinga (1974), especially chapter nine: “God, Evil, and the Metaphysics of Freedom.” 5. For a helpful discussion of the concepts of observation, testability, and the relevance of auxiliary assumptions, see Sober (2008). 6. We add “to be observed” because of the emphasis in the sequel on the observed instances of evil. Also, one might be tempted to replace A1 with: A1′ If God were to exist, there would be no observed evil. But that wouldn’t be right, because A1′ would be consistent with God existing, evil existing, and creatures failing to observe any instances of evil because creatures lack the cognitive equipment to detect evil; the proponent of the logical problem of evil will want and need a stronger assumption like A1 in order to get her argument off the ground. 7. Proponents of the logical problem have tended to think that A1 is a conceptual truth following from the concept of God as a perfect being who is both omnipotent and perfectly good. Critics, meanwhile, have argued that A1 is false on the grounds that God might have a morally justifying reason to permit some evil to occur. 8. Some theists might object to A1 because of a way of understanding and adopting claims from Augustine and Aquinas according to which evil is not observed per se because evil is the privation of good. However, substituting “privations of good” for “evil” in A1 will work just as well. 9. For example, see Peterson (1982); Hasker (1992); and van Inwagen (1988; 1991). For a critique of Peterson and Hasker and a defense of van Inwagen, see Howard-Snyder and Howard-Snyder (1999). For a critique of van Inwagen, see Kraay (2014). 10. Even granting skeptical theism, one might still see the existence of evil as empirical evidence against the existence of God if one were to assume the following: A3 If God were to exist, there would be no seemingly gratuitous evils. Notice that A3 doesn’t commit one to the existence of gratuitous evil—only to the existence of seemingly gratuitous evil. Cf. Bergmann (2011). 11. We don’t mean “assume” in the sense of a claim that is adopted but not argued for. Rather, we mean “assume” in the sense of auxiliary assumption— a claim that must be conjoined to a hypothesis in order to generate an observational consequence. 12. In the logical problem of evil, X is the obtaining of any evil whatsoever; in the other versions of the argument above, X is the obtaining of gratuitous evil, seemingly gratuitous evil, and horrendous evil respectively.

Arguments from Evil; Evidence for Pro-Theism 201 13. The argument schema as presented is structured as a modus tollens argument to the conclusion that there is no God. Of course, this is not to deny that the conclusion could be expressed as the more moderate conclusion that probably there is no God, depending on how the premises of the schema get fleshed out in an actual argument. 14. Our use of “RBF” assumes for simplicity that multiple atomic RBF’s can be combined and referenced as a single, complex RBF. 15. Not every argument against the existence of God that employs observations of the experience of suffering and evils neatly fits the structure of the arguments surveyed here. For example, in Paul Draper’s probabilistic argument according to which the experience of sentient creatures with respect to pleasure and pain is more probable on the Hypothesis of Indifference fits than on theism, it’s less obvious what RBF is being depended on in the argument and would be preventable if God were to exist (perhaps a better epistemic state?). And not every argument against the existence of God that employs observed features of the world is an argument from evil. For example, in J. L. Schellenberg’s argument from divine hiddenness, the observed feature of the world that counts against God’s existence isn’t the observations of evil, but rather the observation of non-culpable non-believers. If Draper’s and Schellenberg’s arguments don’t fit the argument schema presented at the end of Section I, then they don’t fall within the scope of “arguments from evil” referenced in (2)–(4) of the Evil Evidence Argument. See Draper (1989) and Schellenberg (2006). 16. In the logical problem of evil, X is the obtaining of any evil whatsoever; in the other arguments from evil discussed above, X is the obtaining of gratuitous evil, seemingly gratuitous evil, and horrendous evil respectively. 17. We won’t dwell on the narrow forms of anti-theism in the sequel. This is because while it is interesting to consider in what certain respects God’s existence could perhaps decrease the value of a world, God’s so doing is consistent with God’s existence making things better overall; as such, one can adopt a narrow anti-theism (i.e., acknowledging that God’s existence makes things worse in certain respects) while also being a pro-theist (i.e., holding that God’s existence makes things better overall). 18. See Penner (2015) and Penner and Lougheed (2015). 19. Actually, (iii) follows even if the anti-theist doesn’t adopt any problem of evil, for (iii) follows even if the anti-theist adopts only A1–A4, or even if she merely adopts assumptions relevantly similar to A1–A4. 20. Alternatively, one might think that unfettered personal autonomy is highly valuable. Perhaps it is so valuable that a world in which agents enjoy unfettered personal autonomy is automatically considered superior to worlds in which agents lack such autonomy, even if the results of agents using their autonomy involves horrendous moral evils. However, this seems to amount to an anti-theistic offering of a version of the free will defense. If such moves are available for anti-theists, it’s difficult to see why theists cannot adopt the same defensive strategy against problems of evil. 21. This assumes a particular take on the moral landscape, and defending it is beyond the scope of this paper. However, the basic idea is this: if I privilege my own scale of goods such that in order to achieve a meaningful life, it would be better for me to live in a world without God, even though I’m committed to the claim that a world with God would, say, eradicate all evil, then it seems like the price for my life to be meaningful is too high. 22. This project was funded by the John Templeton Foundation “Axiology of Theism” grant. Also, we’ve benefitted from feedback we received when earlier versions of this chapter were presented at the Axiology of Theism capstone

202  Myron A. Penner and Benjamin H. Arbour conference at Ryerson University (2015), the Evangelical Philosophical Society (2015), and the Pacific Division Meeting of the American Philosophical Association (2016). Thanks in particular to Klaas Kraay, who provided excellent and detailed comments on an earlier draft.

References Bergmann, Michael. 2011. “Skeptical Theism and the Problem of Evil.” In The Oxford Handbook of Philosophical Theology, edited by Thomas P. Flint and Michael C. Rea, 374–402. New York, NY: Oxford University Press. Davis, Stephen. 2014. “On Preferring that God Not Exist (or that God Exist): A Dialogue.” Faith and Philosophy 31: 143–159. Draper, Paul J. 1989. “Pain and Pleasure: An Evidential Problem for Theists.” Noûs 23: 331–350. Dumsday, Travis. 2016. “Anti-Theism and the Problem of Divine Hiddenness.” Sophia 55: 179–195. Hasker, William. 1992. “The Necessity of Gratuitous Evil.” Faith and Philosophy 9: 23–44. Howard-Snyder, Daniel, and Howard-Snyder, Frances. 1999. “Is Theism Compatible with Gratuitous Evil?” American Philosophical Quarterly 36: 115–130. Kahane, Guy. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696. Kraay, Klaas. 2014. “Peter van Inwagen on Gratuitous Evil.” Religious Studies 50: 217–234. Kraay, Klaas, and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 153–178. Luck, Morgan, and Ellerby, Nathan. 2012. “Should We Want God Not to Exist?” Philo 15: 193–199. Mackie, J. L. 1955. “Evil and Omnipotence.” Mind 64: 200–212. Mawson, Tim. 2012. “On Determining How Important It Is Whether or Not There Is a God.” European Journal for Philosophy of Religion 4: 95–105. McLean, Graeme. 2015. “Antipathy to God.” Sophia 54: 13–24. Mugg, Joshua. 2016. “The Quietist Challenge to the Axiology of God: A Cognitive Approach to Counterpossibles.” Faith and Philosophy 33: 441–460. Penner, Myron. 2015. “Personal Anti-Theism and the Meaningful Life Argument.” Faith and Philosophy 32: 325–337. Penner, Myron, and Lougheed, Kirk. 2015. “Pro-Theism and the Added Value of Morally Good Agents.” Philosophia Christi 17: 53–69. Peterson, Michael. 1982. Evil and the Christian God. Grand Rapids: Baker Book House. Plantinga, Alvin. 1974. The Nature of Necessity. Oxford: Clarendon Press. Schellenberg, J. L. 2006. Divine Hiddenness and Human Reason. Ithaca, New York, NY: Cornell University Press. Sober, Elliot. 2008. Evidence and Evolution: The Logic Behind the Science. Cambridge: Cambridge University Press. van Inwagen, Peter. 1988. “The Magnitude, Duration, and Distribution of Evil: A Theodicy.” Philosophical Topics 16: 161–187. van Inwagen, Peter. 1991. “The Problem of Evil, the Problem of Air, and the Problem of Silence.” Philosophical Perspectives 5: 135–165.

11 Plantinga’s Defence and His Theodicy Are Incompatible Richard B. Davis and W. Paul Franks

According to Alvin Plantinga’s celebrated free will defence, although there are “sinless worlds” (that is, worlds in which creatures are significantly free but never go morally wrong), it isn’t within God’s power to actualize any of them. For it is entirely possible that we all suffer from transworld depravity—a curious modal/moral malady according to which if God tried to actualize a world in which we were free and always did what was right, we would in fact go wrong. That’s a possibility, says Plantinga. But then given that God has actualized a morally good world—a safe assumption on theism—it follows that evil does exist. Hence, the existence of God and the existence of evil are compatible. More recently, Plantinga has proposed an axiological extension of this conclusion. Not only is “any world in which God exists . . . enormously more valuable than any world in which he does not exist” (Plantinga 2004, 7)—a position that has come to be known as pro-theism1—“it is [also] plausible to think,” says Plantinga, that “the best possible worlds contain Incarnation and Atonement, or at any rate Atonement, and hence also contain sin and evil” (2009, 179).2 Incarnation and atonement (I&A) worlds are best. These worlds include God, sin, and evil. Thus we have the much-sought-after explanation—not simply for how evil could co-exist with God, but for why it is better that God actually permits it. What we have, then, is a rare attempt at theodicy by way of a distinctly Christian pro-theism. In this paper, we attempt to show that if Plantinga’s defence succeeds, his theodicy fails. For if every creaturely essence suffers from transworld depravity, then given that Jesus has a creaturely essence (as we will attempt to show), it follows that I&A worlds cannot be actualized by God, in which case we have anything but a felix culpa.

1. Plantinga’s Pro-Theism According to Plantinga, when faced with the question “Why does God permit evil?”, a theist “might like to have a theodicy, an answer to the question why God permits evil” (1974a, 10). That is to say, she might like

204  Richard B. Davis and W. Paul Franks to know God’s actual reason for allowing evil and suffering in the world. Given our limited epistemic situation, however, that’s a pretty tall order. Thus, as we all know, Plantinga offered something rather less: a defence. What’s the difference? Here is van Inwagen’s helpful explanation: A defense is not necessarily different from a theodicy in content. Indeed, a defense and a theodicy may well be verbally identical. . . . The difference between a defense and a theodicy lies not in their content but in their purposes. A theodicy is a story that is told as the real truth of the matter; a defense is a story that, according to the teller, may or may not be true. (2006, 7) Now this feature of a defence has left some philosophers unsatisfied. For example, John Hick, while agreeing that Plantinga has resolved the logical problem of evil, finds his methodology “disquieting.” He writes: “That he should so easily fill a gap in his theodicy by appealing to a mythological idea, on the ground that it is logically possible, emphasizes again the remoteness of Plantinga’s concern from all questions of plausibility and probability” (2007, 369). Sure, it’s possible (as Plantinga says) that natural evil could be the result of the free actions of non-human agents (demons). But as J. L. Mackie notes: “whether this offers a real solution of the problem is another question” (1982, 154). Still, this is surely unfair. Contra Mackie, Plantinga is, in fact, proposing a real solution to a real problem: the real problem of the (alleged) logical inconsistency between God and evil. If that is the problem, then a defence is all that’s needed. And generally speaking, the defence has met with a favourable response. According to William Rowe, “Plantinga’s argument for this conclusion is, I believe, fairly compelling” (1998, 115). No doubt thinking of Plantinga’s defence, van Inwagen concurs: “It used to be widely held that . . . no possible world contained both God and evil. So far as I am able to tell, this thesis is no longer defended” (1991, 135). This is high praise indeed. Nevertheless, in his recent work on the problem of evil, Plantinga hasn’t been content with mere defence (even recognized, successful defence). He has been tempted to theodicy. In his paper, “Supralapsarianism, or ‘O Felix Culpa’ ” (2004) he suggests that when God creates, he “considers all the uncountably many possible worlds, each with its own degree of excellence or value” (5). However, since there isn’t a best possible world to select (or at least no reason to think there is), God settles on the next best thing: creating “a really good possible world” (5). Of course, given the very nature of God as “perfectly good and holy, all-powerful and all-knowing” (7), it follows that “any world in which God exists is enormously more valuable than any world in which he does not exist” (7).

Plantinga’s Defence, Theodicy Are Incompatible 205 Plantinga, then, is a pro-theist. It would be far better if God existed, he thinks, than if he did not. Now consider all of those worlds in which God does exist. Even if we say that a world in which God exists is “in a good sense infinitely valuable” (Plantinga 2004, 8), still some are more valuable than others. In particular, there is a good-making feature that all “highly eligible worlds” have in common— one that isn’t present in all worlds—that towers enormously above all the rest of the contingent states of affairs included in our world: the unthinkable great good of divine incarnation and atonement. Jesus Christ, the second person of the divine Trinity, incomparably good, holy, and sinless, was willing to empty himself, to take on our flesh and become incarnate, and to suffer and die so that we human beings can have life and be reconciled to the Father. (7) More than that, “any world with incarnation and atonement is a better world than any without it” (10). And this furnishes us with God’s actual reason for permitting evil and suffering. For God wouldn’t create anything less than a “highly eligible world,” and all such worlds contain incarnation and atonement; hence all those worlds contain evil. So if a theodicy is an attempt to explain why God permits evil, what we have here is a theodicy—and, if I’m right, a successful theodicy. (12) It’s a bold and substantial claim—one that leaves the cautious but highly regarded free will defence far behind. In what follows, we argue that Plantinga cannot have it both ways. The details of the earlier defence undermine the new theodicy.

2. The “Depraved” Defence At the heart of Plantinga’s defence is his much-discussed notion of transworld depravity. This initially obscure concept gets unpacked in terms of essences (properties essentially unique to an object—e.g., being identical with Socrates) and worlds (maximal possible states of affairs). Where ‘E+’ is the instantiation of an essence E, the definition of transworld depravity (hereafter, TWD) goes as follows: TWD: For all essences E and worlds W, E has TWD iff E entails the world-indexed properties is significantly free3 in W and always

206  Richard B. Davis and W. Paul Franks does what is right in W only if there is a state of affairs T and action A such that: (1) T(W) is the largest state of affairs God actualizes in W, (2) A is morally significant for E’s instantiation (E+) in W and (3) If God were to strongly actualize T(W), E+ would go wrong with respect to A.4 We are first invited to consider R: God actualizes a world containing moral good, and every essence suffers from transworld depravity. Now according to Plantinga, R is consistent with G: God is omnipotent, omniscient, and wholly good and together with it implies E: There is evil. If he’s right, this shows rather handily that G and E are logically consistent, at least if we accept the elementary modal principle ◊ (p & r) and  ((p & r) ⊃ q); hence ◊ (p & q). So perhaps there are these worlds in which everyone is significantly free and always does what is right. It by no means follows that God can actualize (even weakly actualize)5 any of them. For it might be that we all suffer from Plantinga’s curious transworld affliction. At this juncture, Plantinga has been at pains to convince his critics that R can serve its modal purposes very nicely even if it happens to excite “incredulous stares” from the intelligentsia: R need not be true, or probable, or plausible, or accepted by the scientists of our culture circle, or congenial to “man come of age,” or anything of the sort . . . R can do its job perfectly well even if it is extraordinarily improbable or known to be false. (1985b, 43) To do its work, R need only be possible—a modest claim, to be sure, one befitting our meagre knowledge of what actually goes on with essences across worlds. Strangely enough, however, it turns out that R’s possibility is both a blessing and a curse: a blessing if the goal is to show that E is consistent with a minimal theism (i.e., the view that G is true); a curse if we’re aiming for consistency with something more robust—say, Christian theism and its doctrines of Incarnation and Atonement.

Plantinga’s Defence, Theodicy Are Incompatible 207

3. A “Depraved” Divinity? 3.1.  A Fatal Possibility But wherein lies the curse? R’s possibility doesn’t wear its problems on its sleeve. How does the argument go? Approximately as follows. We begin by noting that if R is possible, then so is its right conjunct—the claim that every essence suffers from transworld depravity (hereafter, TWD). So, initially, we have it that ◊TW: Possibly, all essences suffer from TWD. Now as we all learned at our mother’s knee, a Plantingean essence is a unique sort of property—a property without which a given object could not exist, but also such that nothing else could possibly have it. So consider Socrateity—Socrates’ essence. Socrateity entails each of the properties essential to Socrates: properties we all share such as being coloured if blue,6 but also properties essentially unique to Socrates (e.g., being identical with Socrates—what Plantinga calls his “thisness”).7 We should also note that essences, like numbers, propositions, and possible worlds, are abstract objects; they have a Platonic status.8 They are the sorts of things that are exemplified or not.9 And the thing to see is that everything whatsoever has an essence, and that includes Jesus Christ: “the second person of the divine Trinity . . . [who] was willing to empty himself, to take on our flesh and become incarnate” (Plantinga 2004, 7). For ease of reference, then, let J refer to Jesus’ thisness. The question at once arises: does J suffer from TWD? Naturally, being depraved is the very last thing one wants to attribute to Jesus—at least if we’re thinking of him (as Christians do) as the one who atones for human sin and wrongdoing. Thus we are told: You know that he [Jesus] appeared in order to take away sins, and in him there is no sin. (1 John 3:5) For we do not have a high priest who is unable to sympathize with our weaknesses, but one who in every respect has been tempted as we are, yet without sin. (Hebrews 4:15) As Plantinga says, “Jesus Christ . . . [is] incomparably good, holy, and sinless” (2004, 7). It almost goes without saying that anyone who is himself depraved stands in need of atonement, and thus cannot hope to atone “for the sins of the whole world” (1 John 2:2). Still, to be fair, Plantinga never says that Jesus’ essence suffers from TWD; he only says (by implication and at best) that this is possible

208  Richard B. Davis and W. Paul Franks provided that ◊TW is true. And in fact, even that is questionable. For ◊TW is true only if there is a world W* in which TW: (x) (x is an essence ⊃ x has TWD)

is true. But here, by all accounts, the domain of the quantifier ranges only over the set ψe(W*) of essences that exist in W*. And what TW says is that every essence contained in that set is such that it suffers from TWD. There isn’t the slightest suggestion that ψe(W*) includes J. But if not, then we don’t have it that Jesus’ essence is even possibly transworld depraved, in which case J, from all appearances, appears to be happily sheltered from the atonement-negating effects of TWD. Unfortunately, however, things are not as they appear. It turns out that there are a few complications, arising from particular features of Plantinga’s modal metaphysics. Consider, for example, what he says about Socrateity: Socrates is a contingent being; his essence, however, is not. Properties, like propositions and possible worlds, are necessary beings. If Socrates had not existed, his essence would have been unexemplified, but not non-existent. (2003, 116–117) Or again, The property which is in fact the thisness of Socrates would have existed, I hold, even if Socrates hadn’t—although then it wouldn’t have been a thisness. If Socrates had not existed, this property would not have stood in the is the thisness of relation to Socrates—just as, if I had not existed, my brother-in-law would not have stood in the is the brother-in-law relation to me. He could nevertheless have existed even if I had not. (1985a, 335) So the basic idea is that essences exist necessarily; it isn’t necessary for any of them to be exemplified to exist. They will exist even in worlds where they go unexemplified altogether. Accordingly, ψe(W*)—the set of essences that exist in W*—is identical with the set of all essences (whether exemplified or not). 3.2. Incarnation and Atonement? Interesting things follow. Note first that if J is an essence and if every essence exists in every world, then J will exist in W*—a conclusion that together with the truth of TW (in W*) entitles us to infer that Jesus’

Plantinga’s Defence, Theodicy Are Incompatible 209 essence has TWD is true in W*. Of course, this is not to say that J+, the instantiation of J (namely, Jesus himself), is transworld depraved but only that he could be. Even so, this bare possibility rapidly leads to grief. For consider what it would be for J to suffer from TWD. Where ‘A’ and ‘T’ are as before, it is for J to possess the following property: being such that TWDj: (W)((J entails is significantly free in W & J entails always does what is right in W) ⊃ (God strongly actualizes T(W) > J+ goes wrong with respect to A in W)).10 However, if J has TWDj in W*, then J actually has TWDj (by the principle ◊ p entails  p). It then follows (by universal instantiation) that TWDjα: (J entails is significantly free in α & J entails always does what is right in α) ⊃ (God strongly actualizes T(α) > J+ goes wrong with respect to A in W). But given that J entails is sinless and significantly free in α (the sole actual world), as Plantinga will presumably insist, we have the following fatal counterfactual: FATAL: God strongly actualizes T(α) > J+ goes wrong with respect to A in α.

Now FATAL is fatal for the simple reason that α is in fact actual; and so we know that there is a largest state of affairs T(α) that God has strongly actualized in weakly actualising α.11 Therefore, we cannot avoid SIN: J+ goes wrong with respect to A in α.

Certainly, this is no happy outcome for the Christian; for J+ is the instantiation of J—namely, Jesus himself, the incarnate son of God. How can he go morally wrong and yet still qualify as a perfect substitute for sin? No doubt Plantinga is right when he says: Atonement is among other things a matter of creatures’ being saved from the consequences of their sin; therefore if there were no evil, there would be no sin, no consequences of sin to be saved from, and hence no atonement. (2004, 12) But then—at least on the standard Christian scheme—the one doing the atoning isn’t supposed to have sins in need of atonement.12 Yet that’s precisely what is implied by ◊TW—the lynchpin in Plantinga’s defence. Hence, if God’s actualising “a really good possible world” implies “he

210  Richard B. Davis and W. Paul Franks will create a world containing incarnation and atonement” (2004, 11–12), then the actual world isn’t “really good.” To be sure, it contains sin and evil, as Plantinga says every “highly eligible world” does. But if SIN is true, then although it may include incarnation, the actual world lacks atonement. It doesn’t rank, therefore, among the best possible worlds. Those worlds with both incarnation and atonement will be better. That raises another question. Why didn’t God choose to actualize an I&A world? Of his I&A theodicy, Plantinga remarks: All it really requires is that among the worlds of great value, there be some that include incarnation and atonement . . . all that is really required, for my argument, is that incarnation and atonement be possible, i.e., that there be possible worlds that include them. Since, according to Christian thought, this state of affairs is actual, it is a fortiori possible. (2004, 11) So I&A worlds are possible and available for actualization because the actual world is an I&A world. This follows, at any rate “according to Christian thought,” though not as we’ve seen on ◊TW. But let that pass. Let’s stipulate that there are I&A worlds, that they are all “highly eligible,” better than non-I&A worlds, and that God, if he creates at all, will actualize one of them. The salient question is not whether there are such worlds, but whether Plantinga’s theodicy can make use of them. And there is reason to think not—at least if he wants to keep his defence. Perhaps we can see this as follows. To begin with, let’s recall what it means for Jesus’ thisness to be transworld depraved: TWDj: (W)((J entails is significantly free in W & J entails always does what is right in W) ⊃ (God strongly actualizes T(W) > J+ goes wrong with respect to A in W)). Notice that the antecedent here picks out a range of worlds: those in which J+ is free and always does what is right (FAR worlds, as we might call them). Now it seems plausible to suppose that some FAR worlds will include incarnation and atonement, while others will not. Perhaps some FAR worlds involve J+ being free and always doing right (being FAR, for short), but there is no sin and evil. Think, for example, of a world in which each instantiation (of an essence) has being FAR. Then of course J+ will also possess that property therein. But this won’t be an I&A world, since there is no sin or evil to atone for. The important thing to see, however, is that all I&A worlds will be FAR worlds. For if J+ either isn’t significantly free or doesn’t always do what is right, we don’t have atonement even if we have incarnation. If J+ doesn’t always do what is right, that is, if J+ sins, then (as we said before) we don’t have atonement. On the other hand, if J+ isn’t free with

Plantinga’s Defence, Theodicy Are Incompatible 211 respect to any morally significant action, then it is difficult in excelsis to grasp how anything J+ might do related to atoning for sin could count as praiseworthy, meritorious, or acceptable to God. Indeed, if atoning for sin in any way involves being “tempted as we are” and “yet without sin,” lacking significant freedom is a serious shortcoming. For presumably, resisting temptation involves a free decision not to succumb to sin’s ‘gravitational pull’. And now for the problem. If FAR worlds include all I&A worlds, then TWDj implies the following: ∼(I&A) For any I&A world W, if God were to strongly actualize T(W), then J+ (that is, Jesus—J’s instantiation) would have gone wrong with respect to A in W. Put another way: God cannot (weakly) actualize any I&A world—not if every essence suffers from transworld depravity. For if God were to actualize such a world, Jesus would have sinned, so that some I&non-A world would have been actual. While no doubt possible in the sense of being free of contradiction, worlds containing “the towering and magnificent good of divine incarnation and atonement” (Plantinga 2004, 9), and in which “sinful creatures are offered redemption and salvation from their sins” (9–10) are simply not within God’s power to create. So we may grant (with Plantinga) that “any world with incarnation and atonement is better than any without it” (10), and also that “a necessary condition of atonement is sin and evil” (12). But how does that explain why God actually permitted evil, if these “better” worlds can’t be actualized? Whence, then, Plantinga’s theodicy? Rather unexpectedly, it appears to falter in the face of his widely-touted defence.

4. Objections and Replies 4.1.  The Creation Objection The doctrine of TWD is indeed possible. However, its scope doesn’t include Jesus’ essence because Plantinga tells us quite explicitly that he has only creaturely essences in view, “i.e., every essence entailing is created by God” (1974b, 188, fn 1). But Jesus’ essence, J, doesn’t entail this property, since his essence (part of it at least) is divine, namely, being God the Son. Reply: It will be helpful to have before us an authoritative (or at least quasi-authoritative) account of the incarnation. Here we can scarcely do better than the Definition of Chalcedon (451): Following, then, the holy fathers, we unite in teaching all men to confess the one and only Son, our Lord Jesus Christ. This selfsame

212  Richard B. Davis and W. Paul Franks one is perfect, both in deity and also in human-ness; this selfsame one is also actually God and actually man, with a rational soul and a body. He is of the same reality as God as far as his deity is concerned and of the same reality as we ourselves as far as his human-ness is concerned, thus like us in all respects, sin only excepted. (Leith 1963, 35–36) For our purposes, the basic idea comes to this. Jesus is that person— God the Son, the second person of the Trinity—who, subsequent to the incarnation, has two kind essences: being human and being divine.13 As the Definition suggests, this implies that Jesus also possessed a concrete human nature: “a rational soul and a body.” Let ‘JHN’ denote this nature. We must be careful to distinguish JHN from the abstract property of being human. The latter is an uncreated, multiply exemplifiable universal.14 Every human being shares and exemplifies it. By sharp contrast, Jesus’ humanness is a trope—his concrete, particular instancing of this property. Naturally, JHN is distinct from your specific humanness (trope) and mine; still, as far as his humanity goes, Jesus is “of the same reality as we ourselves” (Leith 1963, 36). That is, he exemplifies the same kind essence as we do (namely, being human), along with everything that entails. 4.1.1. Is Jesus’ Essence Creaturely? Now if that’s right, we know that J—the property of being identical with Jesus—is conjunctive. It includes both being God the Son (GS) and being human (H). No one could have J without co-exemplifying GS and H. Among other things, this implies that J is a creaturely essence. It’s an essence because it’s a basic identity property. It’s creaturely because it happily meets Plantinga’s sole (sufficient) condition for being a creaturely essence: “every essence entailing is created by God” (1974b, 188, fn 1). Since J entails GS and H, it entails H. But it is a necessary truth (on theism) that every property instance of H (e.g., Adam’s humanness, Jesus’ humanness, and so on) is created by God. By the transitivity of property entailment, therefore, J entails is created by God. Hence, J is a creaturely essence. The reply, of course, will be that while this line of reasoning meets the “letter of the law,” it nevertheless runs roughshod over the spirit of the original objection. The issue at stake isn’t simply J’s creaturely status; we can grant that if we like. Rather, what we want to know is whether J is the sort of creaturely essence featured in TWD. And here, presumably, the essences Plantinga has in mind all include or entail the property possibly sinning (PS). We’re asked to envision the possibility of (creaturely) moral agents freely going wrong if God were to (weakly) actualize FAR worlds in which they exist. But this isn’t so much as possible if creaturely essences preclude PS. For then any such

Plantinga’s Defence, Theodicy Are Incompatible 213 essence would include the complement ∼PS of PS; that is, it would entail not possibly sinning. Well then, what about J? Is its instantiation, J+, the sort of object that can make free choices about morally significant matters (e.g., deciding whether to betray a friend for thirty pieces of silver), but also go wrong along these lines—even if, in fact, it doesn’t? At first glance, it is hard to see how J could entail PS. For consider, once again, J’s conjuncts: GS and H. Neither of these properties, it seems, entails PS. Take GS, for example. Not only does this property not entail possibly sinning; it is logically incompatible with it. For the sole instantiation of GS is GS+ (that is, God the Son)—a divine person who, by the very nature of the case, is incapable of sinning. As Plantinga observes, If, as most of the Christian tradition affirms, [God] could not have been powerless, or morally imperfect or without knowledge, then he has the complements of these properties essentially; being knowledgeable, morally perfect, and powerful will be part of his nature. (1980, 141) But then if the tradition is right, it’s part of the nature or essence of GS+ to be morally perfect, so GS won’t entail PS; it will entail its complement not possibly sinning. Next consider J’s other conjunct H—the property of being human. If this property entails PS, then it won’t be possible for there to be an object x such that x exemplifies H but not PS. But is that really true? Consider JHN. Surely it has or exemplifies H. How, otherwise, could it be considered a concrete human nature? That it also exemplifies PS is wholly unclear. For possibly sinning entails being a person, and that is precisely what JHN cannot be.15 For suppose that JHN were a person. Then God the Son’s becoming incarnate (that is, his taking on a human nature) would involve taking on another person, thereby plunging us into the Nestorian heresy; in which case (as Plantinga says in another connection) “there would be two persons here (one human and one divine) rather than the one person who is both human and divine” (2000, 319, fn 41). 4.1.2. Is Jesus’ Essence Conjunctive? But then doesn’t that clinch things for the objector? Neither of J’s conjuncts entails possibly sinning; hence, even if J is a creaturely essence, it’s not the sort of essence TWD has in view. Now the problem with this argument, at least in the present context, is that it relies on a questionable assumption. Where ‘C’ is any creaturely (conjunctive) essence and ‘P’ is any property, the general principle at work seems to be this: CONJUNCT: C entails P only if there is a conjunct C* of C such that C* entails P.

214  Richard B. Davis and W. Paul Franks While this principle surely holds in a wide range of cases, it is by no means clear that it always holds, or even that it would be acceptable to Plantinga. Consider, for example, God’s capacity to suffer: it “exceeds ours,” says Plantinga, “in the same measure as his knowledge exceeds ours” (2000, 319). In particular, “Christ’s suffering was no charade” (319). His suffering was real and (at least in part) physical in nature; for he “endure[d] the agonies of the cross” (319). This raises a question: how shall we handle the property suffering physically here? What does it get predicated of? Here one is initially inclined to think that Jesus suffered in his human nature; that is, it was JHN that experienced the physical pains associated with crucifixion. Interestingly, Plantinga considers (and promptly rejects) this possibility: Can we say that Christ qua human being (according to his human nature) suffered while Christ qua divine nature (according to his divine nature) did not? . . . I’m inclined to think this suggestion incoherent. There is this person, the second person of the divine trinity who became incarnate. It is this person who suffers; if there really were two centers of consciousness here, one suffering and the other not, there would be two persons here (one human and one divine) rather than the one person who is both human and divine. (2000, 319) This quotation is instructive along several lines. First, it gives us a handy recipe for incarnational predication. Since the property suffering physically is person-entailing,16 if JHN did suffer, it would have had its own “centre of consciousness” and been a person in its own right. But then the incarnation would have involved two persons “rather than one person who is both human and divine.” Second, although Plantinga doesn’t explicitly rule it out, he carefully avoids attributing suffering physically to Christ qua his divine nature. More precisely, he doesn’t predicate it of God the Son (GS+) in isolation. Perhaps it’s not hard to see why. For surely it isn’t possible for someone to experience physical pain and suffering if they lack a concrete physical nature altogether. If Kripke is right, pain isn’t to be identified with this or that physical structure or event (say, C-fibre firings). It doesn’t follow that physical pain can be experienced apart from these things. Thus when we say that God the Son could have suffered physically, we don’t mean this in an absolute, unqualified sense. What we mean, rather, is this. Consider the second person of the Trinity united with a concrete human nature: that individual could have experienced what we do: physical, human suffering. Accordingly, the proper ontic subject in this case isn’t GS+ simpliciter but rather GS+ incarnate: the “one person who is both human and divine.”

Plantinga’s Defence, Theodicy Are Incompatible 215 The important thing to see here is that this gives us the makings of a counterexample to CONJUNCT. According to Plantinga, J+ can possess a property when neither GS+ nor JHN alone does. But if that is right, then surely J—a conjunctive essence—can entail a property (e.g., possibly sinning) even if neither GS nor H does. To be sure, CONJUNCT enjoys a certain initial plausibility, but it fails in the case that presently matters most: the incarnation. It therefore poses no insuperable problem for our claim that J is rightly thought to be among TWD’s creaturely essences. Here, as we demonstrate below, it doesn’t help to reply that since J is singular and unique—that is, since it (alone) among creaturely essences has being God the Son as a conjunct—it must be treated differently. For being thus conjoined, one might think, renders J the sole exception to CONJUNCT. In point of fact, however, this is not the case. 4.2. The Cartesian Objection On your view, there are possible worlds in which Jesus goes wrong with respect to morally significant actions. But you haven’t shown this is possible; and in fact it isn’t. For if per impossibile Jesus could sin, it would surely be a consequence of his having a human nature. And while, for the rest of us, that alone is sufficient for possibly sinning, not so for JHN. For the sin-inducing potentialities JHN possesses get overridden or suppressed if JHN is assumed by a divine person. But then if being assumed by God the Son is essential to JHN (which it is), there won’t be any worlds in which Jesus sins. There are two crucial but questionable assumptions behind our objector’s reasoning here. Let’s examine each in turn. 4.2.1. Must JHN Be Assumed by God the Son? The first assumption has to do with JHN’s existence across worlds. It may be stated as follows: TRANSWORLD: Necessarily, for all worlds W, JHN exists in W only if GS+ assumes JHN in W. According to TRANSWORLD, any world in which JHN exists is an incarnation world. More exactly, it is a world in which GS+ (and none other) has assumed JHN. Under this assumption about JHN’s assumption, Jesus’ human nature has the essential property being assumed by God the Son. Now why believe that? Why not think instead that this is a contingent property of JHN? For this reason, perhaps. If ‘JHN’ is functioning as a Kripkean rigid designator, then it picks out the particular and unique human nature

216  Richard B. Davis and W. Paul Franks Jesus did have across worlds. Now that very object just in itself is ontologically incomplete. It isn’t, as we’ve said, a human person. Thus, Craig points out that it is preferable to speak of the Logos’s grounding, rather than assuming, a human nature, for the human nature of Christ becomes complete only in its union with the Logos. (2006, 63) So JHN must be “grounded” or “completed” in some way. In the incarnation that grounding is provided by a divine person (the Logos or second person of the Trinity). But why think that’s the only way JHN could be completed? Why couldn’t God bring it about instead that (say) a human person—or perhaps the angel Gabriel—serves as ontological ground here? That certainly seems like something an omnipotent being could do. At any rate, nothing we know or reasonably believe about JHN precludes it. Or take O’Connor and Woodward’s recent suggestion. JHN is indeed a “stand-alone” entity, as are all instances of being human. It’s not that something must be supplied to JHN in order for it to exist. Not at all; JHN’s existence is in the clear. Still, human personhood emerges only after a process of development is completed and a certain threshold of complexity is achieved.17 They write: Typically, an instance of human nature will include, in itself, a proprietary center of subjectivity and agency; that is, a properly formed and functioning human body is sufficient for the emergence of an autonomous, experiencing subject and agent. (2015, 233)18 This is “typically” how things go. In the incarnation, however, God the Son assumes (they use the term “absorbs”) JHN prior to this complexity “threshold,” thereby preventing the emergence of a functioning human person. And this, they think, neatly avoids Nestorianism. As always, we can quibble about the details; but we shouldn’t miss the forest for the trees. If either of these proposals—“human/angel grounding” or “emergent functionalism”—is even barely logically possible, then it’s not true that (necessarily) JHN exists only if assumed by GS+. At any rate, the burden of proof here is squarely on the objector’s shoulders. She must show that these proposals are impossibilities if she wants to advance TRANSWORLD. 4.2.2. Is Jesus Essentially Sinless? There is a second critical (but questionable) assumption behind the objector’s claim that Jesus (J+) cannot go morally wrong, if his human nature (JHN) is assumed by God the Son. Here is the assumption:

Plantinga’s Defence, Theodicy Are Incompatible 217 SINLESS: Necessarily, for all worlds W, if GS+ assumes JHN in W, then J+ is sinless in W. We must ask, of course, how it is that God the Son’s assuming JHN necessitates Jesus’ sinlessness. Proponents of SINLESS are not far to find. According to Craig, for example, “Regardless of which body the Logos chose to be united with, the body-soul composite which is the result of the Incarnation and is the individual human nature of Christ is incapable of sin” (2006, 63). Now so far as we can see, this way of thinking (to borrow one of Plantinga’s expressions) begins in a pious and commendable concern for Jesus’ impeccability. Unfortunately, it ends by casting doubt on whether Jesus was genuinely tempted, thus calling into question the very existence of I&A worlds. Just ask yourself: how is it meritorious or in any way atonementsecuring, if Jesus fails to do what is, for him, logically impossible—say, worshipping someone other than God when tempted to do so? Indeed, given SINLESS, Jesus cannot even want or will to sin because these, too, are morally wrong (mental) actions. Nor could he intend to want to sin. All of this, it goes without saying, bears little resemblance to his being “tempted as we are.” What would be admirable and praiseworthy is a case where Jesus could sin (where that was a real possibility for him), but then freely resisted the temptation to do so. That’s certainly the way things go with us. But if it isn’t so much as logically possible that Jesus fail to resist temptation—that is, if he is logically determined to resist it—then (on Plantinga’s scheme at least) he doesn’t do so freely, and so isn’t properly subject to praise and approval. Someone might reply (following Thomas Morris) that temptation doesn’t have to involve the logical possibility of sinning but only its epistemic possibility. That is, it need only involve the agent’s not knowing he couldn’t sin even if he wanted to, while also being unaware that he couldn’t want to. This is a subtle topic with a growing literature.19 For present purposes, however, it will suffice to note the following. Morris’ prime example for showing that temptation only requires epistemic possibility actually backfires. Indeed, it lends support to precisely the view he means to oppose. Here is his prime example: On reflection, we can see that it is the epistemic possibility of sinning rather than a broadly logical, or metaphysical, or even physical possibility that is conceptually linked to temptation. Jones can be tempted at t to go and lie to his department chairman, although, unknown to him, his chairman died an hour earlier, making it impossible for Jones or anyone else to go and lie to him at t or thereafter. (1986, 147) But this example doesn’t presently serve. For one thing, the origin of the impossibility isn’t right. In Jesus’ case, the impossibility of sinning

218  Richard B. Davis and W. Paul Franks is supposed to be de re; it’s an impossibility said to arise from his being essentially morally good. Jesus’ identity property not only precludes A: Possibly not doing what is morally right. it includes its negation. For if J includes A, says the objector, there will be possible worlds in which God the Son sins. And this is out of the question if, as we might think, Jesus has being morally perfect essentially. In Jones’ case, by contrast, the impossibility is conjunctive and circumstantial: It is not possible that (Jones is tempted to go and lie to his department chairman & Jones’ department chairman is already dead). In other words, it’s not that Jones cannot lie. No doubt he’s like the rest of us. There are certain features (weaknesses or limitations) of his nature that make his telling a lie a logical possibility. Circumstances permitting, it is within Jones’ power to lie. What he can’t do—and this is no surprise—is lie to someone who just isn’t there because they’ve died. But then, clearly, it’s the circumstances that make the lying impossible here, not some de re truth about Jones’ (concrete, human) nature. Notice, too, that the example turns on the fact that there are some things Jones doesn’t know—a fact, once again, to be ascribed to his limited nature (e.g., not being multiply located). He can be tempted because he doesn’t know his department chair is already dead. If he did know this, to go and lie to him wouldn’t so much as cross his mind. It wouldn’t be a temptation at all. But of course, there’s a problem here, if we’re thinking along the lines of our objector. On her view, and as “most of the Christian tradition” affirms, God’s being knowledgeable is on modal par with his being morally perfect. God—and by extension God the Son— has both of these properties—and has them essentially. Now if that’s right, then by our objector’s logic, Jesus cannot fail to know anything, in which case we can’t explain his being tempted by way of “the epistemic possibility of sinning”—that is, not knowing he cannot sin. There won’t be such a possibility. The bigger question here, however, is whether the objector is in a decent position to deny that Jesus’ identity property, J, precludes A. It is far from clear that she is; indeed, precisely the opposite seems to be the case. For suppose that God’s knowledge and moral perfection are on a modal par. Then recall the objector’s argument against J entailing A: “If J entails A, then it is possible that God the Son sins. But this isn’t possible.” Now consider this parallel (but problematic) argument. According to Plantinga, a successful doctrine of the incarnation must accommodate Jesus’ not knowing some things.20 “No one knows about that day or hour, not even the angels in heaven, nor the Son, but only the Father”

Plantinga’s Defence, Theodicy Are Incompatible 219 (Mark 13:32).21 Here Jesus refers to the time of his second coming. There is, presumably, a day and hour (that is, a time) at which this event will occur. If ‘t’ stands for that exact time, Jesus is saying he does not know the truth value of the following true proposition: TIME:  Jesus returns at t. Now let’s suppose that J entails B: Possibly not knowing TIME. If this is so, we might argue, it is possible that God the Son lacks knowledge. But this isn’t in fact possible. So J doesn’t entail B. Case closed. Of course the problem is: if (following Plantinga) we take the biblical texts at face value, we know there is at least one thing Jesus doesn’t know (i.e., TIME). So we know that J entails B. But doesn’t that give the objector a reason for thinking Jesus’ creaturely essence does include A, thus situating it squarely within the scope of TWD’s quantifier? There is a kind of parity argument here. By the objector’s logic, if J doesn’t entail A, then it doesn’t entail B either. But it does entail B. Therefore, it also entails A. In any event, if as the objector concedes a concrete human nature N isn’t a divine nature, then N must possess certain limiting features (e.g., having a brain with only a finite number of synapses, a body that gets hungry, tired, and so on)—features sufficient for anything that has N possibly lacking power, knowledge, and goodness. In which case, it is essential to anything with N that it have A. But then to say that Jesus is essentially sinless amounts to saying that God has eliminated or suppressed those limiting features in JHN. This is hardly better than John Locke’s once having asserted that while it is impossible for matter to think, God, being omnipotent, can nevertheless override de re modalities and bestow upon it the power of thinking. As the good Bishop Stillingfleet aptly observed, however, this is simply Cartesianism in disguise. The fact is: “God doth not change the essential properties of things while the things themselves remain in their own nature” (Stillingfleet 1987, 78). The application to JHN is patent. Of course, some will be nervous here about this claim that there are worlds in which Jesus isn’t impeccable. Still, for someone like Plantinga, there should be no cause for concern. We want I&A worlds because they’re best. But if atonement requires being tempted but freely resisting sin, there must at least be possible worlds in which Jesus doesn’t always do what is right. If there aren’t, then his always choosing the good in the actual world isn’t free (in Plantinga’s sense) but rather logically determined. So these worlds are there so to speak; however, by hypothesis, they aren’t I&A worlds; they’re I&non-A worlds. On Plantinga’s theodicy, then, God

220  Richard B. Davis and W. Paul Franks simply won’t create them because they’re not among the best. Rather, he’ll “choose around” them (and in fact has done so) using his middle knowledge.

5. Conclusion Plantinga is a serious, Christian pro-theist. On his view, all possible worlds are good, those in which God exists are better than those without, and the best worlds (including the one that is actual) all involve incarnation and atonement (and hence sin and suffering). These claims constitute the basic working parts of the theodicy—Plantinga’s actual explanation for the presence of evil in the world. But the theodicy only works if it’s true that our world is an I&A world God has actualized. And we’ve argued it isn’t, not if Plantinga’s free will defence succeeds. For the defence requires that (possibly) every creaturely essence suffers from transworld depravity. Unfortunately, so far as we can see, there is no solid basis for exempting Jesus’ essence from this moral misfortune. But then given various first principles in Plantinga’s modal metaphysics, we have SIN—the decidedly unwanted conclusion that Jesus has actually gone morally wrong. And if that’s right, our world is not an I&A world and the theodicy is a bust. As Christian philosophers, we can’t help but think that it may well be time to re-evaluate the defence—or at least its betrothal to the doctrine of transworld depravity.22

Notes 1. For further details on pro-theism, see Kahane (2011), as well as Kraay and Dragos (2013). 2. “Or, at any rate, Atonement”: this little hedge is perhaps ill-advised. For it opens up the possibility (epistemic at least) that the best possible worlds include atonement but not incarnation. This leads to an unhappy dilemma. Since, on Plantinga’s view, an incarnation has in fact taken place, it follows either that God has created a world that might not be among the best (that is, there are better); or alternatively, the actual world is one of the best, but (for all we know) the incarnation never happened. At any rate, Plantinga finds it “hard to imagine” what a world with atonement but no incarnation would look like, and thus proposes “that we ignore those possible worlds” (2000, 10). We owe this point to Klaas Kraay. 3. Plantinga’s account of freedom is libertarian. A person has significant freedom with respect to an action only if she could have done otherwise—that is, if it was within her power “at the time in question, to perform the action, and within [her] power to refrain.” Thus no one is free with respect to an action if she is causally (or logically) determined to take that action or refrain from it. See Plantinga (1974b, 166). 4. This is Plantinga’s most recent formulation of TWD (cf. 2009, 178–179). In both its early and later formulations, TWD involves quantification over all essences, though the use to which Plantinga puts this principle is specified to creaturely essences—essences, as he says, entailing the property is created by

Plantinga’s Defence, Theodicy Are Incompatible 221 God. See Plantinga (1974b, 188, fn 1). We shall have more to say about the implications of this specification below. 5. According to Plantinga, God weakly actualizes a state of affairs S just in case God strongly actualizes—that is, causes to be actual—“a state of affairs S* that counterfactually implies S” (1985b, 49). 6. See Plantinga (1974b, 60–62). 7. “Let us say that a thisness is the property for some object x (some actually existing object x) of being that very object x” (Plantinga 1985a, 335). 8. See Plantinga (1985b, 88). 9. See Plantinga (1974b, 70; 1985a, 335). 10. Here we use “>” to express the counterfactual connective. 11. Our argument here presupposes “Lewis’ Lemma”: “For every world W in which God exists, God could have weakly actualized W only if G(T(W)) >W” (Plantinga 1985b, 50). 12. For someone like Plantinga, the alternatives here are somewhat bleak: (1) hold that Jesus atones for our sins, but someone else atones for his (say, God the Father), or (2) hold that Jesus’ sins aren’t in need of atonement at all. Thanks to Klaas Kraay for drawing this point to our attention. 13. Compare Plantinga: “Jesus Christ, the second person of the divine Trinity . . . was willing . . . to take on our flesh and become incarnate” (2004, 7). 14. “What God has created are the heavens and the earth and all that they contain; he has not created himself, or numbers, propositions, properties, or states of affairs: these have no beginnings” (Plantinga 1974b, 169). 15. Objection: “You deny that JHN can exemplify possibly sinning on the grounds that it is a person-entailing property. But this falsely assumes that being a person is essential to whatever has it. It assumes that if you are a person in any world in which you exist, then you’re a person in every world in which you exist. However, as Thomas Flint points out, JHN—a non-person— could have existed without being assumed by GS+, in which case JHN would have been a person in its own right—a person who, like the rest of us, was capable of moral wrongdoing.” See Flint (2001, 312–318).   Reply: This reasoning is mistaken. If ‘JHN’ is a rigid designator, then what it refers to across worlds is the object that as things in fact stand (i.e., in the actual world) it does refer to. That object in any world it happens to exist, and just in itself, isn’t going to be a person. Perhaps it could be “assumed” by something else, or develop into something else that was a person. But that isn’t to the point. What ‘JHN’ does in fact denote won’t—just on its own—be a person in other worlds. 16. If animals also suffer physically (as it seems that they do), then they too would count as persons—not human persons, of course, but persons in their own right nonetheless. 17. Just to be clear, this is not to say that the object that is in fact Jesus’ human nature (what we’re refering to as ‘JHN’) is possibly a person. What’s possible here, rather, is only that JHN could develop into something else—something non-identical with what ‘JHN’ does in fact denote. What O’Connor and Woodward are saying is that that thing might count as a person if certain conditions are met. 18. O’Connor and Woodward state their position in materialist terms. But their point is perfectly general. Dualists (like Plantinga) can also accommodate a soul-body developmental process, perhaps citing Luke 2:52 in support— “And Jesus grew in wisdom and stature.” 19. See in particular Yandell (1988, 313–344; 1991, 169–182). See also Leftow (2014, 3–23).

222  Richard B. Davis and W. Paul Franks 0. See Plantinga (1999, 184). 2 21. Jesus’ reference to himself here as “the Son” is not to God the Son, but rather (in context) “the Son of man” (Mark 13:26, emphasis added)—an individual with a concrete human nature. 22. Our thanks to Stephen Maitzen, Michael Tooley, William Lane Craig, Scott A. Davison, Joseph Jedwab, and Myron A. Penner. We are especially grateful to Klaas J. Kraay for his detailed, painstaking, and insightful comments on earlier drafts of this chapter.

References Craig, William Lane. 2006. “Flint’s Molinist Christology Not Radical Enough.” Faith and Philosophy 23: 55–64. Flint, Thomas P. 2001. “The Possibilities of Incarnation: Some Radical Molinist Suggestions.” Religious Studies 37: 307–320. Hick, John. 2007. Evil and the God of Love. New York, NY: Palgrave Macmillan. Kahane, Guy. 2011. “Should We Want God to Exist?” Philosophy and Phenomenological Research 82: 674–696. Kraay, Klaas J. and Dragos, Chris. 2013. “On Preferring God’s Non-Existence.” Canadian Journal of Philosophy 43: 157–178. Leftow, Brian. 2014. “Tempting God.” Faith and Philosophy 31: 3–23. Leith, John H. 1963. Creeds of the Church. New York, NY: Doubleday Anchor. Mackie J. L. 1982. The Miracle of Theism. New York, NY: Oxford University Press. Morris, Thomas V. 1986. The Logic of God Incarnate. Ithaca: Cornell University Press. O’Connor, Timothy, and Philip Woodward. 2015. “Incarnation and the Multiverse.” In God and the Multiverse: Scientific, Philosophical, and Theological Perspectives, edited by Klaas J. Kraay, 227–241. New York, NY: Routledge. Plantinga, Alvin. 1974a. God, Freedom, and Evil. Grand Rapids, MI: Eerdmans. Plantinga, Alvin. 1974b. The Nature of Necessity. Oxford: Clarendon Press. Plantinga, Alvin. 1980. Does God Have a Nature? Milwaukee, WI: Marquette University Press. Plantinga, Alvin. 1985a. “Replies to My Colleagues.” In Alvin Plantinga, edited by James E. Tomberlin and Peter van Inwagen, 313–396. Dordrecht: D. Reidel Publishing Company. Plantinga, Alvin. 1985b. “Self-Profile.” In Alvin Plantinga, edited by James E. Tomberlin and Peter van Inwagen, 3–97. Dordrecht: D. Reidel Publishing Company. Plantinga, Alvin. 1999. “On Heresy, Mind, and Truth.” Faith and Philosophy 16: 182–193. Plantinga, Alvin. 2000. Warranted Christian Belief. New York, NY: Oxford University Press. Plantinga, Alvin. 2003. Essays in the Metaphysics of Modality. Edited by Matthew Davidson. New York, NY: Oxford University Press. Plantinga, Alvin. 2004. “Supralapsarianism, or ‘O Felix Culpa’.” In Christian Faith and the Problem of Evil, edited by Peter van Inwagen, 1–25. Grand Rapids, MI: Eerdmans.

Plantinga’s Defence, Theodicy Are Incompatible 223 Plantinga, Alvin. 2009. “Transworld Depravity, Transworld Sanctity, & Uncooperative Essences.” Philosophy and Phenomenological Research 78: 178–191. Rowe, William. 1998. “In Defense of ‘The Free Will Defense’: Response to Daniel Howard-Snyder and John O’Leary-Hawthorne.” International Journal for Philosophy of Religion 44: 115–120. Stillingfleet, Edward. 1697/1987. Three Criticisms of Locke. Reprint, Hildesheim: Georg Olms Verlag. van Inwagen, Peter. 1991. “The Problem of Evil, the Problem of Air, and the Problem of Silence.” In Philosophical Perspectives 5, Philosophy of Religion, 1991, edited by James Tomberlin, 135–165. Atascadero, CA: Ridgeview Publishing. van Inwagen, Peter. 2006. The Problem of Evil. Oxford: Oxford University Press. Yandell, Keith. 1988. “Divine Necessity and Divine Goodness.” In Divine and Human Action: Essays in the Metaphysics of Theism, edited by Thomas V. Morris, 313–344. Ithaca: Cornell University Press. Yandell, Keith. 1991. “Some Problems for Tomistic Incarnationists.” International Journal for Philosophy of Religion 30: 169–182.

Index

abstract object/abstracta 76, 207 absurdity of life 17 – 18, 25 – 26, 112, 147 – 163 actual world 3 – 5, 7 – 10, 22 – 24, 27, 42, 44, 49, 58, 71 – 79, 82, 84 – 87, 90 – 91, 97 – 99, 102, 104, 106, 109, 118 – 120, 122, 126 – 127, 158, 171, 184, 186 – 189, 194 – 196, 200, 209 – 210, 219 – 221 aetiological 75, 79, 81, 83 afterlife 15, 23, 62, 79 – 80, 84 – 86, 105, 109, 127 agnostic/agnosticism 2, 10 – 14, 20, 31, 99, 123, 193 Aquinas, Thomas 105, 132, 160, 200 Aristotle 41 atonement 29, 203, 205 – 211, 217, 219 – 221 Augustine 123, 141 – 142, 200 autonomy/autonomous 17, 24, 27 – 28, 111 – 112, 114, 118, 126, 165, 171, 175, 188 – 189, 198 – 199, 201, 216 benevolent/benevolence 32, 63, 99, 101, 103, 105, 107 – 108, 110 – 113, 116, 125 – 127, 134, 151 best possible world/best of all possible worlds 29, 48, 98, 102, 119, 160, 162, 203 – 204, 210, 220 Big Bang 8, 79, 80, 98 Buddhism/Buddhist/The Buddha 48, 57, 182 Bugs Bunny 6 cause/causal/causation 8, 26, 54, 74 – 75, 80 – 82, 91, 110, 125, 138, 140 – 141, 145, 149, 151 – 152,

154 – 155, 161, 182, 184 – 185, 220 – 221 chance 79 – 82 Christianity/Christian 17 – 18, 25, 29, 62, 95, 105, 109, 147 – 162, 182, 203, 221 commensurability 4, 9 comparability 4, 9 compatibilism 30 compensation/compensate 125, 150, 166, 172 concrete object 39 – 40 consequentialism 109 contingency/contingent/contingently 2, 8, 27, 30, 89, 108, 127, 183, 186, 205, 208 cosmic justice 15, 17, 24, 31, 101 – 102, 104 – 105, 121, 124 – 125,  127 counterpossible 5, 30, 70, 74, 90, 199, 200 Craig, William Lane 120, 147 – 149, 216 – 217 creation/creator 2, 8, 16, 23, 39, 40, 50, 62, 75, 78, 91, 99, 105, 108 – 110, 116, 126, 181, 198, 211; free will defence 19, 28 – 29, 99, 203 – 222 death 44, 57 – 59, 61 – 63, 105, 154, 161, 173 deicide 109 deontological 118 Descartes, Rene (Cartesian) 25, 134, 138, 144, 215, 219 design/designer 32, 96, 98 dignity 17, 107, 113, 126, 175 disposition 21, 40 – 41, 190

226 Index divine command theory 77, 125 dualism 98, 139 epistemology/epistemic/epistemologist/ epistemological 4, 6, 7, 9, 39 – 40, 70 – 71, 76, 132, 135 – 137, 142, 192, 193, 199, 201, 204, 217, 218, 220 equality 113, 115, 189 essence 28 – 29, 39, 105, 203, 205 – 213, 215, 219, 220 eternalism 30 eternal life 2, 15, 22, 24, 62 evil 5, 8, 15, 18, 19, 22, 25 – 29, 44, 48 – 50, 56, 58 – 61, 69, 96, 99, 107 – 108, 118, 120, 122, 125 – 127, 140, 144, 150, 158, 165 – 166, 168, 172, 176, 184 – 190, 192 – 202, 203 – 206, 209 – 211, 220; gratuitous 8, 15, 25, 26, 28, 107, 118, 125, 127, 150, 158, 165 – 166, 172, 193 – 194, 197, 198, 200; horrendous 58, 197 – 198, 200 – 201; moral 22, 28, 29, 59, 199, 201; natural 59, 199, 204; problem of 5, 19, 27 – 28, 99, 108, 122, 126, 165 – 166, 168, 176, 184 – 190, 192, 193, 197, 200, 201, 204 evolution 8, 27 – 28, 127, 145, 188, 189 – 191 forgiveness 16, 123, 125, 161 freedom 104, 110, 114 – 115, 127, 171, 211, 220 free will 30, 61, 99, 104, 125, 192 goodness 28, 43, 58, 61, 63, 138, 155, 171, 219 haecceity 39, 89 happiness 62, 105, 153 – 155, 161 heaven 15, 23, 62 – 63, 79 – 80, 82, 84 – 87, 99, 102, 104, 110, 112, 119, 126, 141, 218, 221 hell 62, 63, 80, 82, 85, 102, 110, 112 hiddenness 31, 60, 189, 201 Hinduism 57, 182 Hitchens, Christopher 120, 126, 160 – 161 hope 20, 26, 32, 102, 105, 112, 135, 141, 151, 159, 161 – 162, 176 idealism 98, 100, 111, 116, 124, 126 immortality 24, 61 – 63, 89, 100 – 102, 104, 118, 121 – 124, 165, 176

impossible worlds 44, 49, 74, 77, 120 incarnation 29, 203, 205, 206, 208, 210 – 212, 214 – 218,  220 incommensurability 13 incomparability 13, 205, 207 independence 17, 27, 28, 64, 111 – 118, 116, 165, 189 indifferentism 11, 20, 31 infinite/infinity/infinitely 32, 51 – 54, 95, 103 – 104, 112, 117 – 118, 127, 158, 205 intelligibility 5 – 6, 10, 13, 16 – 17, 24, 30, 121, 132 – 135, 144, 198 Jesus Christ 19, 29, 80, 151 – 152, 161, 203, 205, 207 – 222 Judaism 182 justice 15, 17, 22, 24, 31, 58, 61, 101 – 102, 104 – 105, 117, 121, 122, 124 – 125, 127, 166, 170 – 171 Kant, Immanuel (Kantian) 112, 126 karma 24, 48, 98, 172 knowledge 17, 25, 43, 66 – 67, 105, 123, 125, 134 – 138, 141 – 142, 144, 155 – 156, 165 – 166, 192, 199, 206, 213 – 214, 218,  220 Leibniz, Gottfried Wilhelm 119, 132 Lewis, C.S. 151, 155 – 156 Lewis, David 5, 74, 221 libertarianism/libertarian 30, 61, 104, 115, 220 life plan 17, 33, 116 logic, laws of 24 – 25, 133 – 134, 139, 143 – 144 love/loving 16, 25, 59, 62 – 63, 105, 108, 123, 126, 147, 149, 151, 156, 161, 169, 175, 189 magic 25, 116, 132, 139 – 143 meaning/meaningful/meaningfulness/ meaningless/meaninglessness 15, 17, 26 – 28, 33, 42, 56, 98, 104, 122 – 123, 125, 133, 140, 148 – 149, 151, 173, 175 – 176, 189 – 190, 198 – 199,  201 metaethics 57, 98 miracle 80, 144 – 145 modal/modality/modalities/modal space/modal ability/modal skepticism 7, 13 – 14, 23, 30, 74, 120, 192, 203, 206, 208, 218 – 220 monotheism/monotheistic 23, 46, 50, 63, 67 – 69, 99, 108, 126, 142

Index  227 morality 5, 15, 17, 32, 99, 104, 110, 115, 125, 132, 171 moral wrongdoing 10, 16, 121 multiverse 98 Nagel, Thomas 20, 32, 51, 64, 102, 110, 121 – 124, 127, 147, 158 – 159, 161 – 162,  176 naturalism 4 – 5, 7 – 10, 15 – 17, 22 – 23, 25, 27, 31, 32, 97 – 100, 103 – 104, 114, 116, 119, 120, 121, 124, 126 – 127, 138 – 142, 145, 158, 159, 162, 183, 189, 198 natural law/law of nature 79, 80, 83, 140, 144 necessity 6, 22, 32, 70, 75, 76 – 81, 83 – 85, 89,  91 neutralism 10 – 12, 14, 20, 71, 88 Nietzsche, Friedrich 115, 123, 128 nihilism/nihilist 31, 96, 123 non-resistant nonbelief 8, 19, 27, 189 Nozick, Robert 41 numbers 200, 207, 221 omnipotence/omnipotent 2, 4, 18, 22 – 23, 25, 47 – 48, 51 – 57, 59, 60 – 61, 65 – 68, 73, 78, 90 – 91, 99, 106, 108, 113, 149 – 150, 181, 187, 193, 198, 200, 206, 216, 219 omniscience/omniscient 2, 4, 16, 18, 22 – 23, 25, 43, 47, 50 – 53, 57, 59 – 60, 65 – 68, 73, 99, 106, 108, 110, 113, 149 – 150, 181, 187, 206 panpsychism 98, 139 Parfit, Derek 72, 124, 152 patient-benefitting suffering/ patient-centred restriction 15, 150, 166, 172 perfect goodness 2, 4, 18, 22 – 23, 47, 50 – 53, 56, 57, 59, 63, 65 – 67, 73, 108, 120, 166, 190, 198, 200, 204 Plantinga, Alvin 7, 19, 28 – 31, 106, 114, 121, 145, 192, 200, 203 – 222 Plato (Platonism/Platonic) 57, 77, 98, 137, 207 polytheism 98 possibility: epistemic 7, 137, 217 – 218; logical 5 – 7, 19, 23, 28 – 29, 54, 74, 78, 82 – 83, 89 – 91, 137, 193, 204, 216 – 218; metaphysical 7, 23, 74, 82, 100, 120 power 9, 14, 21, 22, 26, 32, 40 – 43, 50, 52, 54, 57, 67, 68, 101, 105,

108, 110, 113 – 114, 138, 140, 157, 165, 166, 171, 190, 194, 203, 204, 211, 213, 218 – 220 prayer 116, 151 presentism 30 privacy 16, 17, 22, 24, 27, 28, 32, 63 – 68, 102, 105 – 108, 110, 112 – 113, 116, 125 – 126, 128, 165, 171 – 172, 189, 198, 200 properties, unknown 58 – 60 providence 105, 110 psychology/psychological 18, 25 – 26, 95, 119, 121, 137, 148, 150, 153 – 156, 159,  161 purpose 8, 16, 25, 30 – 31, 83, 96, 98, 100, 110, 111, 114, 120, 123 – 124, 138 – 140, 145, 148, 197, 204, 206 quietism 10 – 14, 20 – 21,  30 rational preference 1, 20 – 21, 26, 165, 170 regret/regrettable 100, 101, 121 – 123, 127, 143, 162 reincarnation 48, 57 rigid designator 215, 221 Sartre, Jean-Paul 106, 123, 148 scepticism/sceptic/sceptical/skepticism/ skeptic/skeptical 14, 30, 135, 138, 144 science 139, 141 – 142, 182, 189, 193, 199 sin/sinner/sinless 10, 29, 31, 62, 203, 205, 207, 209 – 213, 215 – 221 skeptical/sceptical theism 61, 194, 200 solitude 64, 110 soteriology/soteriological 27, 182 – 184,  188 soul 41, 48, 57, 59, 96, 212, 217, 221 Stalnaker, Robert 5, 74 Stump, Eleonore 32, 156, 161 suffering 8, 15, 17, 18, 22, 25, 26, 28 – 29, 44, 50, 58 – 59, 61 – 62, 73, 99, 108, 112, 117 – 118, 149 – 151, 153 – 157, 159 – 162, 165 – 166, 169 – 170, 172, 174, 186, 197, 199, 201, 203 – 209, 211, 214, 220,  221 sustainer/sustaining/sustained 2, 8, 16, 32, 79 – 80, 83, 90 – 91, 109, 110, 114 teleology 8, 9, 98 telos 10, 25, 138

228 Index theodicy 19, 29, 61, 194, 203 – 205, 210, 211, 219 – 220 time 18, 22, 30, 40, 50, 52 – 55, 62 – 63, 66, 74 – 75, 77, 90, 95, 141 time travel 74, 90 transcendent/transcendence/ transcendental 27, 28, 136, 138, 181 – 187, 190,  191 trans-world depravity 28 – 29, 203, 205 – 211,  220 ultimacy 27, 183 – 186, 188, 190 – 191 understanding 28, 64, 67, 142, 189, 198 – 199 universalism 15, 23, 79 – 80, 82, 88, 151, 219 – 220

universe 16, 17, 24 – 25, 32, 50, 64, 72, 79, 81, 83, 84, 95 – 106, 108 – 112, 115 – 124, 127, 132 – 134, 140 – 141, 144 – 145, 147 – 151, 157 – 159, 161, 176, 181, 199 value: aesthetic 2, 39, 75, 97; intrinsic 2, 14, 21 – 22, 39 – 44, 63, 64, 176, 182 – 183; moral 132; prudential 2, 39 virtue/virtuous 100, 110 Wittgenstein, Ludwig 176, 183 worship 16, 23, 33, 64, 67 – 68, 79, 82, 86, 95, 111, 113, 126, 183,  217