The Challenges of Divine Determinism: A Philosophical Analysis 110848302X, 9781108483025
In this volume, Peter Furlong delves into the question of divine determinism - the view that God has determined everythi
483 58 1MB
English Pages 246 [256] Year 2019
Polecaj historie
Table of contents :
01.0_pp_i_ii_The_Challenges_of_Divine_Determinism
02.0_pp_iii_iii_The_Challenges_of_Divine_Determinism
03.0_pp_iv_iv_Copyright_page
04.0_pp_v_vi_Dedication
05.0_pp_vii_vii_Contents
06.0_pp_viii_viii_Acknowledgments
07.0_pp_1_13_Introduction
08.0_pp_14_33_A_Primer_on_Divine_Determinism
09.0_pp_34_59_Divine_Determinism_and_Free_Will
10.0_pp_60_85_Divine_Determinism_and_Free_Will
11.0_pp_86_104_Divine_Determinism_and_the_Author_of_Sin_Objection
12.0_pp_105_133_Divine_Determinism_and_the_Blameworthiness_Objection
13.0_pp_134_159_Divine_Determinism_and_the_Free_Will_Defense
14.0_pp_160_186_God_Determined_Agents_and_Love
15.0_pp_187_216_Divine_Commands_the_Divine_Will_and_Divine_Blame
16.0_pp_217_222_Conclusion
17.0_pp_223_234_References
18.0_pp_235_240_Index
Citation preview
THE CHALLENGES OF DIVINE DETERMINISM
In this volume, Peter Furlong delves into the question of divine determinism – the view that God has determined everything that has ever happened or will ever happen. This view, which has a long history among multiple religious and philosophical traditions, faces a host of counterarguments. It seems to rob humans of their free will, absolving them of all the wrongs they commit. It seems to make God the author of sin and, therefore, blameworthy for all human wrongdoing. Additionally, it seems to undermine the popular “Free Will Defense” of the problem of evil, to make a mockery of the claim that God loves us, and to make it inappropriate for God to blame and punish us. This work carefully formulates these and other objections to divine determinism and investigates possible responses to each of them, providing systematic and balanced discussion of this major philosophical and theological debate. is Professor of Philosophy at Valencia College in Florida. His work has appeared in journals including Pacific Philosophical Quarterly, Faith and Philosophy, Res Philosophica, Sophia, and International Philosophical Quarterly, as well as in Aquinas’s Disputed Questions on Evil: A Critical Guide (Cambridge University Press, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
THE CHALLENGES OF DIVINE DETERMINISM A Philosophical Analysis
PETER FURLONG Valencia College
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
University Printing House, Cambridge , United Kingdom One Liberty Plaza, th Floor, New York, , USA Williamstown Road, Port Melbourne, , Australia –, rd Floor, Plot , Splendor Forum, Jasola District Centre, New Delhi – , India Anson Road, #–/, Singapore Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/ : ./ © Peter Furlong This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data : Furlong, Peter, - author. : The challenges of divine determinism : a philosophical analysis / Peter Furlong, Valencia College, Florida. : [edition]. | New York : Cambridge University Press, . | Includes bibliographical references and index. : | (hardback : alk. paper) | (pbk. : alk. paper) : : Teleology. | Providence and government of God. | Free will and determinism. | Free will and determinism–Religious aspects. : . | –dc LC record available at https://lccn.loc.gov/ ---- Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
THE CHALLENGES OF DIVINE DETERMINISM
In this volume, Peter Furlong delves into the question of divine determinism – the view that God has determined everything that has ever happened or will ever happen. This view, which has a long history among multiple religious and philosophical traditions, faces a host of counterarguments. It seems to rob humans of their free will, absolving them of all the wrongs they commit. It seems to make God the author of sin and, therefore, blameworthy for all human wrongdoing. Additionally, it seems to undermine the popular “Free Will Defense” of the problem of evil, to make a mockery of the claim that God loves us, and to make it inappropriate for God to blame and punish us. This work carefully formulates these and other objections to divine determinism and investigates possible responses to each of them, providing systematic and balanced discussion of this major philosophical and theological debate. is Professor of Philosophy at Valencia College in Florida. His work has appeared in journals including Pacific Philosophical Quarterly, Faith and Philosophy, Res Philosophica, Sophia, and International Philosophical Quarterly, as well as in Aquinas’s Disputed Questions on Evil: A Critical Guide (Cambridge University Press, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
THE CHALLENGES OF DIVINE DETERMINISM A Philosophical Analysis
PETER FURLONG Valencia College
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
University Printing House, Cambridge , United Kingdom One Liberty Plaza, th Floor, New York, , USA Williamstown Road, Port Melbourne, , Australia –, rd Floor, Plot , Splendor Forum, Jasola District Centre, New Delhi – , India Anson Road, #–/, Singapore Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/ : ./ © Peter Furlong This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data : Furlong, Peter, - author. : The challenges of divine determinism : a philosophical analysis / Peter Furlong, Valencia College, Florida. : [edition]. | New York : Cambridge University Press, . | Includes bibliographical references and index. : | (hardback : alk. paper) | (pbk. : alk. paper) : : Teleology. | Providence and government of God. | Free will and determinism. | Free will and determinism–Religious aspects. : . | –dc LC record available at https://lccn.loc.gov/ ---- Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
THE CHALLENGES OF DIVINE DETERMINISM
In this volume, Peter Furlong delves into the question of divine determinism – the view that God has determined everything that has ever happened or will ever happen. This view, which has a long history among multiple religious and philosophical traditions, faces a host of counterarguments. It seems to rob humans of their free will, absolving them of all the wrongs they commit. It seems to make God the author of sin and, therefore, blameworthy for all human wrongdoing. Additionally, it seems to undermine the popular “Free Will Defense” of the problem of evil, to make a mockery of the claim that God loves us, and to make it inappropriate for God to blame and punish us. This work carefully formulates these and other objections to divine determinism and investigates possible responses to each of them, providing systematic and balanced discussion of this major philosophical and theological debate. is Professor of Philosophy at Valencia College in Florida. His work has appeared in journals including Pacific Philosophical Quarterly, Faith and Philosophy, Res Philosophica, Sophia, and International Philosophical Quarterly, as well as in Aquinas’s Disputed Questions on Evil: A Critical Guide (Cambridge University Press, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
THE CHALLENGES OF DIVINE DETERMINISM A Philosophical Analysis
PETER FURLONG Valencia College
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
University Printing House, Cambridge , United Kingdom One Liberty Plaza, th Floor, New York, , USA Williamstown Road, Port Melbourne, , Australia –, rd Floor, Plot , Splendor Forum, Jasola District Centre, New Delhi – , India Anson Road, #–/, Singapore Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/ : ./ © Peter Furlong This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data : Furlong, Peter, - author. : The challenges of divine determinism : a philosophical analysis / Peter Furlong, Valencia College, Florida. : [edition]. | New York : Cambridge University Press, . | Includes bibliographical references and index. : | (hardback : alk. paper) | (pbk. : alk. paper) : : Teleology. | Providence and government of God. | Free will and determinism. | Free will and determinism–Religious aspects. : . | –dc LC record available at https://lccn.loc.gov/ ---- Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
For my family
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Contents
Acknowledgments
page viii
Introduction
A Primer on Divine Determinism
Divine Determinism and Free Will: The Consequence Argument
Divine Determinism and Free Will: Manipulation Arguments
Divine Determinism and the Author of Sin Objection
Divine Determinism and the Blameworthiness Objection
Divine Determinism and the Free Will Defense
God, Determined Agents, and Love
Divine Commands, the Divine Will, and Divine Blame
Conclusion
References Index
vii
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Acknowledgments
I am grateful to Shawn Grant, Julie Montione, Travis Rodgers, Bob Warren, and Claire Yates, the members of an interdisciplinary writing group, for their advice and encouragement. I am grateful to Thomas Flint, Matthew Pietropaoli, Michael Staron, S. Matthew Stolte, and two anonymous referees for their comments on work that made its way into Chapter , to the Classical Theism Project for funding work on an article that appears in modified form in Chapters and , and to two anonymous referees who provided comments on that original article. I deeply appreciate the work of two anonymous referees for Cambridge University Press who gave careful and insightful comments on the entire manuscript. The book is much improved thanks to their comments. Finally, I wish to thank Hilary Gaskin, Jill Hobbs, and the team at Cambridge University Press. They have made the publication process of my first book a pleasure. I am very grateful to Tobias Hoffmann, Angela McKay Knobel, and Michael Gorman for their advice on publishing and especially for their mentorship over the past decade. Most of all, I am indebted to all my family members, who have taught and encouraged me, and whose sacrifices have made my career possible. This is a debt I will never be able to repay. Large parts of Chapter originally appeared in “Is God the Cause of Sin?” Faith and Philosophy (): – (). Large parts of Chapters and originally appeared as an article and are adapted and reprinted by permission from Springer Nature: Sophia, “Blameworthiness, Love, and Strong Divine Sovereignty,” Peter Furlong, . I thank both Faith and Philosophy and Sophia for granting permission to reprint material that originally appeared in the pages of their journals.
viii
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845
Introduction
During the past years, analytic philosophers have been busy exploring new ways to approach the perennially popular topic of determinism, especially as it relates to free will. Even more recently, some of these same philosophers have taken a much closer look at whether the acceptance of theism should change both how one views the prospects of determinism and how one views the relationship between determinism and free will. Central to such discussion is the question of whether divine determinism should be accepted. Divine determinism might be understood loosely to be the view that everything that occurs is a consequence of God’s will – a will both complete and irresistible, such that the entire history of the universe is settled by what God ordains. Different philosophers will provide different definitions of divine determinism; I discuss these and settle on one to guide this investigation in the next chapter. We might begin, though, by taking a look at some claims that seem to be paradigmatic of this view. In the history of philosophy and theology, the best way to interpret past authors is often a matter of some debate. This present matter does not constitute an exception. It has seemed to some that many of the major philosophers and theologians of the Western canon were divine determinists, including Augustine, Thomas Aquinas, René Descartes, and Gottfried Leibniz. We might begin by focusing on John Calvin, who seems to offer straightforward endorsements of divine determinism. He writes: [W]e ought undoubtedly to hold that whatever changes are discerned in the world are produced from the secret stirring of God’s hand. But what God has determined must necessarily so take place. (Calvin , ..)
For recent explorations of this topic, see the essays in McCann () and Timpe and Speak (). For discussion of Descartes, see Chapter , Section , of this book. For the claim that each of the others seems to endorse divine determinism (at least at times), see Vicens (). Calvin goes on to note that this divine decree imposes conditional – and not natural – necessity. Nevertheless, he indicates that given God’s decree, particular changes in the world are certain to come to pass.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
Calvin notes that many things in our experience seem to be merely random, but this is because the “necessity of those things which happen for the most part lie[s] hidden in God’s purpose” (, ..) He illustrates: Let us imagine, for example, a merchant who, entering a wood with a company of faithful men, unwisely wanders away from his companions, and in his wandering comes upon a robber’s den, among thieves, and is slain. His death was not only foreseen by God’s eye, but also determined by his decree (, ..).
This view has also been a matter of intense religious discussion, and sometimes finds its way into official religious statements. The Westminster Confession of Faith (.–), for example, seems clear on this matter. It declares: God from all eternity, did, by the most wise and holy counsel of His own will, freely, and unchangeably ordain whatsoever comes to pass. . . . Although God knows whatsoever may or can come to pass upon all supposed conditions; yet has He not decreed anything because He foresaw it as future, or as that which would come to pass upon such conditions.
Jonathan Edwards agrees, arguing that such a conclusion rests on a philosophical, rather than merely religious, basis. He summarizes his argument as follows: For, as the being of the world is from God, so the circumstances in which it had its being at first, both negative and positive, must be ordered by him, in one of these ways; and all the necessary consequences of these circumstances, must be ordered by him. And God’s active and positive interpositions, after the world was created, and the consequences of these interpositions; also every instance of his forbearing to interpose, and the sure consequences of this forbearance, must all be determined according to his pleasure. And therefore every event, which is the consequence of anything whatsoever, or that is connected with any foregoing thing or circumstances, either positive or negative, as the ground or reason of its existence, must be ordered of God; either by a designing efficiency and interposition, or a designed forbearing to operate or interpose. But, as has been proved, all events whatsoever, are necessarily connected with something foregoing, either positive or negative, which is the ground of its existence. It follows, therefore, that the whole series of events is thus connected with something in the state of things either positive or negative, which is original in the series; i.e. something which is connected with nothing preceding that, but God’s own immediate conduct, either his acting or forbearing to act. From whence it follows, that as God designedly orders his own conduct, and its connected consequences, it must necessarily be, that he designedly orders all things. (b, )
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
This view is defended by some contemporary philosophers as well. In his defense of divine determinism, Heath White writes: [Divine determinism says], roughly, that God’s will determines, settles, or fixes every other fact about the world that could have been other than it is. This means that there is no detail of the universe that is undetermined by God’s will. In particular, contingent facts about human wills, or about what humans would freely choose in such-and-such circumstances, or about other events in the creation like coin flips or gamma rays, are not ultimately or brutely contingent, in that they can be explained, completely, as the results of what God willed. (, )
Even if we put to the side questions about whether particular figures – Augustine or Aquinas, for example – are divine determinists or not, it is clear that this position has been attractive to philosophers in the past and continues to find support in the philosophical community today. Moreover, this position has played an important role in religious discussions, and indeed in religious movements. Most notable in this regard is the acceptance of divine determinism by a number of churches influenced by Calvin, and the arguments between Calvinists and Armenians on how to understand the extent of divine decrees, especially divine decrees about which humans will be saved. This book is a philosophical investigation of the challenges of divine determinism: It is an exploration of the problems that advocates of this position seem to face and a search for successful replies. Thus, this book is not merely a survey: I do not intend to present various positions in the detached manner of a historian of ideas. Instead, I plan to wade knee-deep into the problems that divine determinists must face. With each problem, I intend to turn common and often vague worries into nuanced objections and explore possible responses. One way to think of this project is as an exploration of philosophical topography. This work is not a defense of a single picture of the world, but instead a map of the logical landscape. Although the highly detailed proposals and spirited defenses of philosophers assured of their own idiosyncratic views can be enjoyable and thought provoking, there is also value – in some cases, more value – in projects like this one, which aim to break new ground by exploring varieties of some position in a critical way.
Some may also see the debate within Catholicism between Thomists and Molinists as a debate over whether a certain sort of divine determinism should be accepted. For the claim that at least some Thomist positions should count as divine determinism, see Vicens (), who thinks it is important to define this category in a way that includes some Thomist positions. Much of the voluminous work of Alfred Mele, which explores both libertarian and compatibilist accounts of free will without committing to either (although committing to many other positions
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
The viability of divine determinism is important for a number of reasons, even for those who have not adopted this position. The first is that many philosophers of religion are interested in exploring great-making features. For theists, at least, there is a worry about how investigations into such features should be conducted. Should theists, for example, presuppose that divine determinism must be false, and so be confident in rejecting any proposed great-making feature (or combination of such features) that suggests the truth of divine determinism? This question is most obviously pertinent to the issue of divine foreknowledge, more than one proposed account of which is motivated by a desire to sidestep divine determinism, but similar comments might be made about divine providence, aseity, and sovereignty. A related but distinct value of investigating the problems of divine determinism is that it can be an aid to theologians and religious believers in formulating and selecting particular versions of doctrines. Many religious doctrines come in deterministic and indeterministic flavors, and the choice between them might turn upon philosophical rather than theological considerations. Investigating the challenges of divine determinism also aids in ascertaining whether it is reasonable for theists to be compatibilists about free will and physical determinism. We will have occasion to look at physical determinism with more precision in Chapter , but we can loosely state that physical determinism is the view that past physical states necessitate all future physical states. I take a compatibilist to be someone who holds that the existence of free will is compatible with (some variety of ) determinism. Nevertheless, unless otherwise noted, I will speak in a more restricted way, understanding a compatibilist to be someone who holds that free will is compatible with physical determinism in particular, and an incompatibilist to hold that compatibilism (in this sense) is false. Many philosophers use “free will” to stand in for the sort of control required for moral responsibility, and I will here follow this usage. According to this formulation, compatibilism takes no stand on the question of whether determinism (either physical or divine) is true. There is no contradiction, then, in a philosopher denying both divine and physical determinism but accepting compatibilism. Nevertheless, it might be uncomfortable for a theist to categorically deny divine determinism and accept compatibilism. If
and argumentative moves), offers a nice example of the virtues of such an approach. See, for example, Mele (). Some philosophers, most notably Peter van Inwagen(, b), argue that we should understand free will in another fashion.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
physical determinism is compatible with free will, then it is curious that God would create a world that is not divinely determined. Almost universally, theists who deny divine determinism rely on incompatibilist claims in explaining God’s motive in giving up some control over creation. If compatibilism is true, then it seems that there was no need for a trade-off in creation between divine and creaturely control; thus, it is odd that God decided to create a world that is not divinely determined. Whether a theist should be a compatibilist, then, turns out to be tied up with the issue of divine determinism. Another reason to investigate the challenges of divine determinism comes from John Martin Fischer. He has repeatedly noted that one particularly worrisome aspect of ordinary libertarianism (that is, the conjunction of the claims that we are free and that free will is incompatible with physical determinism) is that on such a view free will “hangs by a thread” (b). What Fischer means is that according to such a view, the very existence of free will depends upon the falsity of physical determinism. If the laws of nature turn out to be probabilistic – even if in virtue of such laws, together with past states, each of our choices is overwhelmingly likely to occur in particular ways – then it is possible that we are free. Conversely, if it turns out that the natural laws governing the universe are not probabilistic but deterministic, as many physicists have thought, then we would be forced to conclude that we are not free. Our own self-image, it seems, is hostage to the possible discoveries of physicists. That this should be the case is not only worrying, but, Fischer maintains, implausible. Although some will contend that there is nothing implausible about such a claim, many seem willing to accept that it is worrying. This worry, Fischer has explained, provides the impetus not for believing compatibilism, but at least for examining compatibilism very closely in an attempt to discover whether it might be true.
Some have called into question whether incompatibilists should allow free choices to have probabilities governing their occurrence at all. For discussion, see Furlong (b), O’Connor (), Pereboom (), and Vicens (). There is, of course, debate about how much our own self-image would suffer if we were forced to conclude that free will is an illusion. Pereboom(), for example, argues that much of what we care about can survive even if belief in free will is abandoned. At times, Fischer seems to suggest that this worry does, in fact, provide reasons for believing in compatibilism. He writes: “I believe that theory choice is based on a holistic methodology in which one weighs pros and cons” (a, ). He admits that some arguments for incompatibilism have some force, but claims that the force of such arguments “must be put on the scales with the many pros of compatibilism, which include the fact that our moral responsibility status need not hang on a thread” (). The fact that this consideration counts in favor of compatibilism, one that can be put on the scales to help outweigh arguments against compatibilism, suggests that at times Fischer
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
Fischer claims that theists face an additional worry: Given ordinary incompatibilism (and some common theistic claims), theism hangs by a thread. He writes: Given that theism requires freedom, and that flip-flopping is not an acceptable option, it would seem that a libertarian theist would have to give up her belief in God, if she were to learn that determinism holds. But then one of the libertarian theist’s most central and fundamental beliefs will be held hostage to whether the natural laws have associated with them percent probabilities or something less. And this seems uncomfortable. (Fischer b, )
Such a worry does not provide theists with a positive reason for accepting physical determinism, but it does give them a reason to look very carefully at whether it at least might be compatible with their other views. Moreover, if theists are open to at least the possibility of physical determinism, then they should take a closer look at the possibility of divine determinism, since, if the universe is governed by deterministic laws, then there is some reason for thinking that God has so determined it. Suppose we discover that each physical state (other than the first) is determined by the one that came before it (together with the laws of nature). Suppose we then add that the universe was created by God. It seems that in creating the first instant, God has brought about a divinely determined world – a world completely governed by divine ordination in virtue of God’s having set up the initial conditions that necessitate all future states. There are ways for theists to avoid this line of reasoning, but they are unlikely to satisfy everyone. In particular, they might follow van Inwagen’s (, ) suggestion that God might be able to issue indeterminate decrees, such as “Let either X or Y be.” If this is possible, then the theist could maintain that a physically determined universe might be created but nonetheless not be determined by God. That is, God could create a deterministic world by decreeing “Let there be either deterministic universe A or deterministic universe B,” without determining which of the two was to come into existence. Other ways of avoiding the preceding line of reasoning might be explored, but given a finite past combined with the claim that God creates the first physical state, physical determinism seems to naturally suggest divine determinism. Thus, worries about hanging theism on a thread may give theists reason to take an especially close look at the prospects of divine determinism. does take the “hanging by a thread” worry to be more than a motivation for exploring compatibilism, but as a partial reason for adopting it.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
This work investigates the challenges of divine determinism, focusing on creating arguments against the view and exploring ways to avoid such arguments. It attempts to bring to light the costs of this view, not the benefits. The reader may wonder why I have refrained from discussing the entire matter – why I have not simply added a few more chapters that assess the reasons weighing in favor of divine determinism and offer a verdict on whether the benefits justify accepting the costs. This is a fair question, and the honest reader may be forgiven for being dissatisfied with “that is beyond the scope of the present work.” Nevertheless, it is. Here is why: Addressing the question of whether we have reasons to endorse divine determinism would require addressing a vast field of issues. It would require investigating certain divine attributes, establishing whether the most philosophically plausible accounts of these attributes rule out, require, or are neutral on the issue of divine determinism. Additionally, some of the virtues of divine determinism relate to its power in making sense of particular religious views – for example, the doctrine of grace within Christianity or divine governance within Islam. Reaching a final verdict on divine determinism would require singling out a particular religious tradition and then exploring both the requirements and desiderata of specific doctrines within that tradition. Not all Muslims agree on the minimal requirements for an account of divine governance, let alone desiderata for selecting among minimally acceptable accounts, and the situation is similar when we look at Christians on the issue of grace. Of course, the doctrines of grace and divine governance represent two of many such doctrines that would need to be explored. Investigating these issues thoroughly would not take a few chapters, but rather many volumes. In fact, even more complications must be considered. We may discover that physical determinism is true. Although we are far from a consensus on whether the fundamental laws of physics are deterministic or probabilistic, there are scientifically respectable deterministic models of quantum phenomena. Suppose that we discover that physical determinism is true. At that point, the question of divine determinism will take on a new urgency since, as noted earlier, some theists might think that physical determinism provides a reason for accepting divine determinism. Of course, it is not clear that determinism is true, but if it is possibly true, then any discussion of whether divine determinism should be accepted would require an investigation of the reasons for accepting theism in the first place, as well as an exploration of whether physical determinism should lead us to
For a philosophically oriented investigation of the current state of this question, see Lewis ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
believe in divine determinism, and, if it should, whether we should believe that physical determinism is true. This last investigation would likely take us out of the realms of philosophy and theology and require delving into arguments in physics. I beg the reader’s understanding, then, when I honestly plead that an investigation of the virtues of divine determinism is beyond the scope of this work. Nevertheless, some might think that there is no need for such an investigation to reach a verdict on whether divine determinism ought to be accepted. Some of the problems of divine determinism might be decisive; the costs of the view might be so high that no benefits could outweigh them. This is an important consideration. I once thought that more than one of the problems of divine determinism was decisive and that jointly the problems were obviously so. Today, although I am not unsympathetic to such a conclusion, I no longer accept it. In each chapter, I will explore one or more challenges, attempting to reach a conclusion on precisely what it will cost determinists to hold on to their view in face of the challenge, and each reader may consider whether the cost is so high that it could not possibly be accepted. In the conclusion, I will turn briefly to my own views of the relative values of each cost that divine determinism brings with it. In the end, it seems to me that although this view carries numerous non-negligible costs, none of the problems is obviously decisive (which, of course, is not to say that they are obviously not decisive). This book is best read from the first page to the last because some of the discussions in early chapters will be referenced in later ones. More importantly, some of the discussions of earlier problems inform the way that later ones are considered, even when this isn’t always explicitly noted. Nevertheless, the reader solely interested in one or more problems may navigate to the chapters focusing on those problems. To such readers, I suggest beginning with the first chapter, and, in light of the following overview, moving on to those chapters that are of most interest. In Chapter , I introduce a stipulated definition of divine determinism and address some worries about this definition. I then explore two distinct sorts of divine determinism, based upon the causal mechanism that explains the deterministic nature of our universe. In one of these accounts,
Some philosophers think that determinism should be considered a philosophical rather than scientific issue. If this is so, then perhaps the issues in physics could be sidestepped; see Steward (). Alternatively, some might think that there are theological reasons for rejecting determinism, such that questions of physics can be safely avoided.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
God creates the initial state of the universe governed by deterministic laws. Thus, God puts in place the first link of a deterministic chain, thereby determining the entire chain. In the second account, God is actively involved in each creaturely event by a more direct intervention. As this view has it, God ordains the actions of creatures not in virtue of having created an initial state of the world that will inevitably lead to creatures acting in particular ways, but instead by creating them in their acting, or by moving them to act, by a divinely caused motion. This view is sometimes illustrated by an analogy. God determines the history of the world not in the way that someone sets off a chain reaction, but instead in the way that an author writes a novel. Authors deliberately and directly act so as to create each element of the stories they pen; according to this view, God, too, is directly active in each and every element of the history of the universe, crafting every detail to match the divine plan, leaving no element unspecified, with the result that the finished product – the universe – is exactly as God had ordained. After outlining the differences between these two approaches, I turn to an overview of some of the reasons divine determinism has been proposed in the past. I refrain from critically assessing these purported benefits, instead merely introducing the reader to each, and thereby providing some context for the following chapters that investigate problems for this view. In Chapters and , I explore the relationship between divine determinism and human free will. Many of those who have argued against divine determinism have thought that if God determines all human actions, then no humans are morally responsible for anything they do. This problem has tended to be tied up with what we might call the compatibilist question: Is moral responsibility compatible with physical determinism? Many have thought not. Furthermore, many have thought that divine determinism (of either variety) is relevantly similar to physical determinism, and so have concluded that it, too, would undermine human moral responsibility. Divine determinists have utilized two different strategies in responding to such attacks. First, some have borrowed defenses developed by ordinary compatibilists or developed their own perfectly general defenses for ordinary compatibilism, and then extended these defenses to what we might call theistic compatibilism – that is, the compatibility of divine determinism and human free will. Others have argued that divine determinism – at
See, for example, Baker () and Edwards (b).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
least in some of its varieties – is relevantly different from physical determinism, and so avoids the problems the latter faces. The quest to settle the general compatibilism/incompatibilism debate in a few chapters would not merely be ambitious; it would be foolhardy. I will not take up this task. Instead, in Chapters and , I investigate whether divine determinists are in a better or worse position than physical determinists in responding to popular incompatibilist arguments. I will begin by investigating the Consequence Argument. According to this argument, if physical determinism is true, truths about the past and the laws of nature entail truths about what we will do in the future; thus, since we have no control over the past or the laws of nature, we should conclude that our own future actions are not up to us. Some have maintained that similar arguments cannot be used to show that divine determinism is equally problematic. In particular, they have suggested that those divine determinists who embrace the distinction between primary and secondary causality need not fear the worries that are embodied in the Consequence Argument. In Chapter , I investigate this claim. In Chapter , I turn to manipulation arguments against compatibilism. According to such arguments, agents who are manipulated by powerful neuroscientists, and thus determined to act in particular ways, are not morally responsible for acting in accord with their manipulators’ wishes. But, defenders of such arguments maintain, physical determinism is relevantly similar to the action of manipulators, so by parity of reasoning we should conclude that physical determinism undermines human moral responsibility as well. Thus, if physical determinism is true, then nobody is morally responsible for anything. Some have suggested that these arguments may be augmented with divine determinism (even if merely considered as a robust hypothetical consideration) to further strengthen them. It may seem that divine determinists are in a worse position to respond to such arguments than physical determinists, since they take this as more than a mere hypothetical possibility. Additionally, some common responses to manipulation arguments attempt to show that manipulation cases are not relevantly similar to physical determinism, at least in part because physical determinism does not posit an agent who determines all events. Since divine determinism does posit such an agent, it seems unable to utilize such responses. In Chapter , I examine both the worry that manipulation arguments are more powerful in themselves if we replace
See, for example, McCann () and Tanner () See van Inwagen (). See, for example, Pereboom (). See Rogers ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
determinism with a determining God and the worry that some ordinary responses to manipulation arguments will be unavailable to the divine determinist. Finally, I consider whether divine determinism is endangered by direct appeals to intuition concerning a manipulating God. In Chapter , I address the charge that divine determinists must admit that God is the cause of sin. It is not hard to see why some might think this is so: According to divine determinism, God ordains all that occurs, and, obviously, humans quite often perform morally evil acts. It is also not hard to see why this might be a worry: Theists often wish to keep God firmly removed from evil of any kind. Indeed, Satan is sometimes called the father of lies – but, if divine determinists are correct, it is not a demonic force that is ultimately responsible for human acts of lying, but rather God. In Chapter , I assess a response to this objection – the privation solution – that tries to avoid the undesirable conclusion that God causes sin. I also investigate a new twist on this old maneuver, which suggests that the evil of sin is not the sort of thing that can be caused, and so no position should lead one to think that God is its cause. I also evaluate whether the theist may reasonably bite the bullet, admitting that God is the cause of sin, while claiming that this assertion should not be thought to be especially damaging to theism. In Chapter , I turn to the question of whether God would be blameworthy for determining human wrongdoing. Would God act wrongly in determining a human agent to commit murder, for example? Once again, the basic problem is easy to see: It at least seems that human agents would act wrongly if they somehow managed to determine other agents to commit murder – perhaps even if the manipulating agents had perfectly laudatory reasons for acting as they did – and so it seems that God, too, must do something wrong in determining all the evils of human history. In Chapter , I investigate how this objection is best formulated and outline possible replies to it. In Chapter , I turn to the general problem of evil. Many philosophers, both theists and atheists, seem to think that the problem of evil constitutes the single most persuasive argument against theism. The problem, in short, is that the presence of the evil in this world seems to constitute evidence against the existence of God, at least given certain characterizations of the divine nature. Accepting divine determinism appears to make this problem even more daunting because it removes what is sometimes considered the most powerful rebuttal to (at least some forms of ) this problem: the Free Will Defense. Some philosophers, however, think that the Free Will Defense is available to all divine determinists, or at least to
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
those who are willing to adopt particular accounts of free will. If this is so, then divine determinists need not worry as much about this problem as is sometimes supposed. In Chapter , I investigate such claims, determining how much comfort they might provide for the divine determinist. In Chapter , I turn to the question of whether divine determinism poses problems for making sense of God’s love for humans and humans’ love for God. Theists often see God as a perfectly loving being, sometimes as a supernatural model of parental love. Divine determinism seems to undermine this picture, however, since it claims that God determines all the pains each human endures, and, perhaps more alarmingly, all the evil actions that scar each person’s character. Worst of all, many theists have claimed that at least some humans will eventually suffer eternal torment for their evil actions. We might reasonably wonder whether a perfectly loving God would dictate a personal history that unfailingly leads to such a painful fate. Although some aspects of worldly experience (and perhaps of religious belief, too) might make divine love difficult to comprehend, adding divine determinism to the mix seems to make it unbelievable. Divine determinists also face a problem concerning human love for God. According to many theists, humans should strive for a loving union with God. In such a union, lovers begin to identify with the beloved, accepting the cares, concerns, and values of the beloved as their own. Such a union seems to involve a sacrifice of the lover’s autonomy, and it seems important that this sacrifice of autonomy is itself an autonomous act. But if divine determinism is true, then humans can sacrifice this autonomy to God only if God causes them to do so. Because such causation seems to undermine the autonomy of humans, divine determinism seems to threaten the authenticity of union love. In Chapter , I address both of these worries, first sharpening them, then constructing and exploring a number of ways for determinists to respond. As with the other worries canvassed, I argue that determinists do have some options available to them, but many theists will find them less than ideal. In Chapter , I take up problems associated with divine commands, the divine will, and divine judgment. The first such problem is that it is reasonable to think that in giving commands, God is revealing the divine will. If God forbids killing, then God does not want humans to kill. But if divine determinism is true, then God sometimes does want humans to kill, as is clear from the obvious fact that humans sometimes do kill. This suggests that if divine determinism is true, then divine commands are deceptive.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
Introduction
Additionally, two related but distinct worries stem from human responses to the divine will. The first concerns human deliberation. To see this worry, consider an agent faced with the options of fulfilling a divine command and breaking it. Ordinarily, we would think it is easy to see what the agent ought to do: The agent ought to obey the divine command and acts wrongly in controverting it. If divine determinism is true, however, some odd cases will put pressure on this conclusion. Suppose, for example, that the agent has a true, justified belief that God wills this command to be broken in this instance. Given divine determinism, it is easy to see how such a belief could be true, and it could be justified just in case agents can be justified in having beliefs about their future behavior. In such a case, it seems that intending to obey divine commands and intending to fulfil the divine will are in tension with each other. If agents are aware of this tension, ordinary theistic modes of deliberation are called into question. This first worry is forward-looking: Divine determinism leads to oddities concerning how agents should think about future actions. A second, related worry is backward-looking: If divine determinism is true, then sorrow for past sins seems to involve a wish that the divine will had not been fulfilled. Ordinarily, theists tend to think that contrition involves both a wish that a sin had not been committed and a wish that God’s will had been done. It seems that divine determinists must choose between these wishes. The final worry I investigate in Chapter concerns a determining God’s standing to blame human agents. Suppose that agents are morally responsible for their behavior, even if God determines their every action. Even so, we may question whether God would be in a position to blame these agents. God is clearly in a position to know whether these agents have acted wrongly, and we might admit that God is in a position of authority over these agents. Nonetheless, there seems to be something perverse in one agent first determining another agent on a path to act wrongly, then passing judgment and punishing the agent for these determined actions. In Chapter , I suggest some options by which determinists might respond to this worry. Finally, I close the book with a brief conclusion, discussing which of the challenges of divine determinism I find the most worrisome, or, in other words, which of the costs of divine determinism I see as particularly high.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.001
A Primer on Divine Determinism
Marking the Borders of Divine Determinism This is a book about divine determinism, so perhaps it is best to start by discussing just what divine determinism (or theological determinism, as some authors have it) is. Immediately there is a problem: Not all parties in the discussion will be able to settle on a way to define this phrase. William Hasker writes: In agreement with Richard Taylor, I understand determinism to be “the general philosophical thesis which states that for everything that ever happens there are conditions such that, given them, nothing else could happen.” And theological determinism is determinism in which the relevant conditions have to do with the will and decrees of God. (, –)
Derk Pereboom offers the following account: Theological determinism is the view that God is the sufficient active cause of everything in the created world, either directly or by secondary causes such as human agents. (, )
Leigh Vicens notes that it might be best to define this phrase in terms of divine decree (or some similar notion). She writes: We might, for instance, take Feinberg’s definition of an “unconditional” decree as one “based on nothing outside of God that move[s] him to choose one thing or another” and then characterize theological determinism as the view that God unconditionally decrees every event that occurs in the history of the world. Such a view would exclude the possibility that God merely permits some events which He foresees will happen in some circumstances but which He does not Himself determine. ()
For Taylor’s account of determinism, which Hasker quotes, see Taylor (, ). See Feinberg (, ) for this author’s discussion.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Marking the Borders of Divine Determinism
David E. Alexander and Daniel M. Johnson try to capture what a number of authors mean when discussing divine or theological determinism – in particular, the sort of determinism entailed by Calvinist teachings. Discussing a variety of authors writing on this issue, they find that these authors tend to agree “that God decides everything that is going to happen and it happens because of God’s decision, and so God has exhaustive control over the world” (, ). The variety of definitions may make us worried that before we can begin exploring the challenges to divine determinism, we must first discover its true nature – that is, we must discover which definition is the correct one. Although it will be important to settle on a definition to guide our inquiry, I think it is misguided to search for the “true definition” of divine determinism. No set of criteria perfectly captures what divine determinism really is. Although I am sympathetic to forms of essentialism in metaphysics, we should be cautious about thinking that the words we use to categorize positions have true essences. Specifically, in the present matter we should be cautious of falling into what Peter van Inwagen (c) calls verbal essentialism. My choice of definition is pragmatic in two ways. First, while denying any sort of verbal essentialism, it is certainly useful to have clear meanings for terms, and (obviously) “divine determinism” is especially important for my present purposes. Second, pragmatic reasons have guided my selection of criteria, and thereby the borders for the category. Since this book is an investigation into the challenges that divine determinism faces, I have chosen borders that will be useful in delineating the category of views that may be reasonably thought to face a host of related worries. Perhaps an even more useful set of borders could be constructed based upon which views fall under the argumentative pressure of certain sorts of objections. Nevertheless, at least at the beginning of this book, I take it to be an open question which views can be successfully pressured by particular objections. I will rely on an account of divine determinism offered by Heath White (, ). According to White, divine determinism is the conjunction of two claims: . .
The facts about God’s will wholly determine every other contingent fact. The facts about God’s will explain every other contingent fact.
I do not seem to be alone in this contention. William Hasker indicates that this term may have more than one meaning and asks only that the sense of the word being used “be made tolerably clear” (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
White’s definition depends upon an account of what it means for one thing to determine another. He explains: Determinism is a form of conditional necessity: Given these facts or events over here, some other fact or event over there must be the case or must occur. The “must” can come in different flavors, depending on the type of determinism in view, but for our purposes, it is a “must” of metaphysical necessity. (White , )
I take “God” in this definition to refer to the sort of being often associated with Western monotheisms. In particular, I take “God” to refer to an omnipotent, omniscient, and perfectly good being. I use the term “physical determinism” in accord with White’s definition of “determinism,” with the understanding that the necessity in question is nomological. We can understand this sort of determinism in the following way: Given x, y is physically determined just in case, in every possible world with both x and the same physical laws as the actual world, either y occurs or y’s failing to occur constitutes a violation of physical laws (a miracle). Furthermore, we will say that x determines y just in case x is a fact or event that necessitates y. According to this strict account, divine determinists should say that facts about God’s will determine creaturely events. Whether, strictly speaking, we should also say that God determines creaturely events depends upon our ontology of facts and our views of divine simplicity. I will set such issues aside here, but for ease of expression I will often speak of God (or God’s will) determining various things. Such locutions should be understood as loose expressions that could be paraphrased away into the more precise (if also more cumbersome) language required by the preceding definition. Finally, a word about the metaphysical necessity flavor of “must.” White does not provide an account of this, but I interpret it in the following way: To say that given x, y must be the case or must occur, is to say nothing more than that in every possible world containing x, y is the case or occurs. With so many previous definitions on offer, one might reasonably wonder why I have chosen as I did. Briefly, let’s consider each of those accounts mentioned earlier. Hasker defines determinism (and thereby divine determinism) in terms of what “could” happen given certain facts. But how determinism affects what things (especially agents) can or could do is a controversial issue, and one I would sooner avoid if possible. Vicens offers an account in terms of divine decreeing. She does so to avoid relying on causal language, but I see no satisfactory way of defining divine decreeing that does not bring with it causal baggage of its own, unless
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Marking the Borders of Divine Determinism
we frame it in terms of metaphysical necessity, which would bring us back to White’s definition. Alexander and Johnson mean to provide a loose account, and I think that White’s definition, which I rely upon, is close to their own, if a bit tighter. I think that Pereboom’s definition is a good one, but I worry that it might inadvertently include too much. In particular, it may categorize a view as divinely deterministic even if it claims that God’s causing x is not explanatorily prior to x. It is not clear to me that such a view is helpfully categorized alongside the other views investigated in this book, at least for my purposes. Some might object to the definition I have chosen, claiming that it includes too much, that some views that fall within its borders should not be thought of as deterministic at all. Others will claim that it includes too little, unjustifiably excluding views that surely deserve to be categorized as divinely deterministic. I suspect that those most resistant to my definition will worry that it categorizes their own view as being divinely deterministic when they wish to resist such a characterization. Disagreement about whether one’s own position qualifies as a species of divine determinism is not new; to see this, consider the following introduction to an article by William Hasker: When I first set out to write this reply, it seemed obvious to me that Kathryn Tanner’s position . . . is a variety of theological determinism. This still seems obvious to me, but I have learned in the meantime that Tanner does not regard herself as a determinist. So it is up to me to provide some justification if I am going to refer to her view as theological determinism. (, )
I have already suggested my own justification for how I characterize divine determinism; perhaps some will be unimpressed and contend that whatever the value of my justification, it falls short. If I may be permitted to borrow from Peter van Inwagen: I’m happy to make a present of the phrase “divine determinism” to anyone who has strong feelings about its proper use and who thinks I am using it improperly. What I am interested in is the conjunction of White’s two claims. I invite those unhappy with my title for this conjunction to substitute their own. Others might worry that my definition relies on a characterization of God (as omnipotent, omniscient, and perfectly good) that leaves out
W. Matthews Grant (a) defends such a view, which he calls the extrinsic model of divine agency. See van Inwagen (a, ) for the original, which substitutes “‘property’ and ‘universal’” for “divine determinism.”
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
positions – whether ones endorsed by some philosophers or merely ones in logical space is perhaps unimportant – that deserve to be considered divinely deterministic. Such an objector might note, for example, that one could think of a stupid and evil being that determines every creaturely event. Such a position rejects the existence of an Omni-God, but surely accepts divine determinism. Nevertheless, according to my definition, it falls outside of this category. I am sympathetic to this objection, but again want to emphasize that I make no claim that the definition I will rely upon is the one true definition, nor that it boasts a high pragmatic value for all situations. In the present work, I wish to focus on a number of problems that seem particularly pressing for theists of a certain type. Perhaps “traditional Western theistic divine determinism” would be a better title for the view delineated here. Those who offer an objection along this line may take uses of “divine determinism” as shorthand for “traditional Western theistic divine determinism” if they wish. Before moving on, it is important to distinguish divine determinism from certain other views. Most importantly, I wish to set aside views that may seem deterministic in some sense, merely by virtue of the way they treat divine foreknowledge. Consider, for example, simple foreknowledge views. According to such views, God has perfect knowledge of all creaturely states, past, present, and future. The propositions that express God’s knowledge of creaturely events, together with the proposition that God’s knowledge is infallible, entail the propositions that express the state of the world at each moment. Some may consider this to be a sort of determinism. I have no battle with those who wish to use the word in this way, but it is not the way I propose to use it here. According to my use, such a view does not count as divinely deterministic merely by virtue of God’s knowledge, since this knowledge may not be a fact about God’s will and may not be explanatorily prior to their objects. Although this kind of view may turn out to face some of the same worries that divine determinists face – worries about making God morally responsible for evil or undermining human freedom, for example – it is outside the scope of the present work.
Motivations for Divine Determinism For reasons discussed in the Introduction, this book focuses on the challenges that divine determinism faces, rather than the benefits its acceptance offers. Nevertheless, I think it is important to be clear on at least some of the reasons why some people would be inclined to endorse
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Motivations for Divine Determinism
divine determinism. This view is not obviously attractive. In fact, to some theists, it is nearly unthinkable. In this brief section, I outline some of the reasons, both historical and contemporary, that have been given in defense of divine determinism, without weighing in on the value of such reasons.
The Principle of Sufficient Reason The first reason to be considered, unlike the others, is perfectly general and independent of particular claims about the nature of God. This reason springs from an acceptance of the principle of sufficient reason (PSR). According to this principle (at least on one simple formulation of it): [PSR]: For every fact F, there must be a sufficient reason why F is the case. (Melamed and Lin )
Different philosophers understand this principle in different ways, but it is often thought to suggest a thorough-going determinism. Hugh McCann discusses this point, noting that an undetermined event – an undetermined free choice, for example – seems to constitute a violation of this law: [If a free act is undetermined by physical states,] then there is nothing in my nature or experience, nothing about the setting in which my decision occurs, in terms of which an observer could reach a completely reliable conclusion as to how the decision would go. And, the argument runs, to the extent this is so we get a violation of the principle of sufficient reason. (, –)
He also suggests a solution for the theist: Suppose that God, as creator, is directly responsible for each of my decisions. If so, then even though my decision to vacation in Colorado was not determined by the rest of my nature, it still has an accounting – an accounting in terms of God’s plans, of the good He sees in my deciding as I do. That is to say, what fully accounts for my decision is not my reasons for it, but God’s. (, )
Although McCann resists calling this view divine determinism (keeping the term “determinism” tied to cases involving nomological necessity), his view seems to satisfy both of the conditions for divine determinism that I have borrowed from White. McCann appeals to facts about God’s will to avoid violations of the PSR, so these facts must be prior in explanation to my decision. Furthermore, it seems that these facts would need to necessitate the later action to meet McCann’s own standard by which to judge whether an explanation is sufficient; if no necessitation is involved, then
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
once again it would be impossible for an observer to “reach a completely reliable conclusion as to how the decision would go” (McCann , ). Moreover, if violations such as the one McCann is concerned with are to be avoided entirely, then God’s involvement with this particular decision cannot be unique but must be present throughout creation. Of course, not all theists, even those who find the PSR attractive, find this line of reasoning persuasive. Some theists who defend some version of this principle resist endorsing divine determinism, but adopting a deterministic view of the relationship between God and creation is certainly the path of least resistance. Divine Sovereignty Other reasons for endorsing divine determinism grow out of convictions about the nature of God’s relationship to creation. The first such reason has to do with divine sovereignty or governance. The idea here is a simple one: God has absolute sovereignty over all of creation. Of course, what constitutes sovereignty is a matter of some debate. Calvin describes divine governance this way: [T]here is no erratic power, or action, or motion in creatures, but that they are governed by God’s secret plan in such a way that nothing happens except what is knowingly and willingly decreed by him. ( I..)
It might seem that human actions, at least, should be thought somehow outside of such strict control, but Calvin stresses the importance of making divine governance complete. After rejecting the view that mere foreknowledge is sufficient for governance, he writes: Not so crass is the error of those who attribute a governance to God, but of a confused and mixed sort, as I have said, namely, one that by a general motion revolves and drives the system of the universe, with its several parts, but which does not specifically direct the action of individual creatures. Yet this error, also, is not tolerable; for by this providence which they call universal, they teach that nothing hinders all creatures from being contingently moved, or man from turning himself hither and thither by the free choice of his will. ( I..)
The idea seems to be that to whatever degree an action is not determined by God, then to that degree God is not sovereign. Thus, divine sovereignty suggests divine determinism. This line of reasoning may not
For a discussion of some ways to reconcile PSR with libertarianism, see Pruss ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Motivations for Divine Determinism
appeal to all theists, but it has seemed reasonable to some. Some theists, in fact, have gone further and argued that God must be the only true cause of all activity. All other apparent causal powers are merely apparent, and the apparent exercise of such supposed powers corresponds to changes only in virtue of divine decree. Divine Power Divine determinism might also be reached by considerations of divine power. Descartes seems to suggest as much: [T]he greater we deem the works of God to be, the better we observe the infinity of his power; and the better known this infinity is to us, the more certain we are that it extends even to the most particular actions of human beings. . . . I do not think that you have in mind some change in God’s decrees occasioned by actions that depend on our free will. No such change is theologically tenable. (, )
And again: As for free will, I agree that if we think only of ourselves we cannot help regarding ourselves as independent; but when we think of the infinite power of God, we cannot help believing that all things depend on him, and hence that our free will is not exempt from his dependence. (, )
Of course, once again, opponents of divine determinism will be eager to suggest that the falsity of divine determinism is consistent with the great power of God. Indeed, it might be pointed out that the power to do something is distinct from its exercise, so those who deny divine determinism while admitting that God had the power to determine everything cannot be said to impugn the power of God. Divine Knowledge and Aseity A second reason for endorsing divine determinism comes from considerations surrounding divine knowledge and divine aseity. I noted earlier that I intend to consider divine determinism (arising from the divine will) as
It is sometimes difficult to ascertain whether a particular philosopher embraces wholescale occasionalism, but prominent Muslim and Christian philosophers have been interpreted in this way, including al-Ash’ari, al-Ghazālī, and Nicolas Malebranche. As is common in the history of philosophy, it is controversial whether Descartes is truly a divine determinist. For discussion, see Collins (), Cunning (), and Ragland (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
distinct from worries that might arise because of perfect foreknowledge. Nevertheless, some theists have endorsed divine determinism specifically because of their views on divine knowledge. They begin by noting that God possesses full and perfect knowledge of all states – past, present, and future. We then move to issues concerning divine aseity. David E. Alexander frames the issue this way: If God is a se, then God is completely independent of everything. God is completely independent – God does not depend on anything for anything. God knows various truths (e.g., God knows that Obama is freely eating dinner). Is this bit of knowledge dependent on something other than God? If God’s knowledge of what Obama is doing right now depends in any way on Obama, then God’s aseity seems to be compromised. (, )
God’s aseity, then, gives us reason to reject the claim that God’s knowledge of creatures depends upon those creatures. If this is correct, then God’s knowledge of creaturely states is far different than ordinary human knowledge – so much the better, a defender of this view will claim. Ordinary human knowledge of another entity is dependent upon that entity. I know, for example, that my daughter Catherine is sitting because I see her sitting. My knowledge state, then, is dependent upon her, and in two ways. First, the mental state itself can be causally traced back to her. I have the mental state only because she is sitting, and, in some way, she seems to produce or bring about my mental state. Second, the status of the mental state as one of knowledge (rather than, for example, of mistaken belief ) depends at least partially upon her, since this mental state is true in virtue of her and, at least in ordinary cases, nothing I did made it true that she is sitting. According to the view of divine knowledge currently under consideration, God’s knowledge is radically different. God’s knowledge that Catherine is sitting does not depend upon Catherine in either way. First, God does not notice that Catherine is sitting and then form a belief that she is. Instead, God believes this in virtue of divine self-knowledge: God is causing her to sit, is aware of this state of affairs, and thus possesses the belief that she is sitting. Second, the status of God’s belief state as one of knowledge does not depend upon Catherine, or, if it does, only in a way that ultimately depends upon God. To see this, note first that God is ultimately responsible for the fact that Catherine is sitting. Even if we wish to say that Catherine’s sitting is the truth-maker for the proposition “Catherine is sitting,” Catherine’s sitting obtains only because God brings
See also Bonino () and McCann ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Motivations for Divine Determinism
it about, so even the truth-maker for the relevant proposition ultimately depends upon God. Such a view of divine knowledge – or at least the concerns that drive its acceptance – nicely dovetails with the concerns driving acceptance of the strong view of divine sovereignty discussed previously. Some people are driven to accept a view of sovereignty that includes divine determinism as a component because they wish to preserve the claim that creatures are wholly dependent upon God, not only for their being, but also for the determination of their actions. People are driven to accept this account of divine knowledge, at least in part, because it preserves the claim that God is wholly independent from creatures. Even divine mental states concerning creatures are not dependent upon creatures. Ultimately, then, we should think of this reason as springing from a desire to preserve the conjunction of both divine aseity and divine knowledge. Divine Providence An additional reason to endorse divine determinism comes from the idea that God providentially watches over creation. Derk Pereboom claims that “A reason to endorse theological determinism is that it provides an uncontested way to secure a strong notion of divine providence” (, ) He explains: Our lives are subject to pain, deprivation, failure, and death. How might we deal with these evils? To affirm a strong notion of divine providence is to accept everything that happens to us, to the last detail, is in accord with God’s wholly benevolent will. The aspiration is that this belief will result in great comfort in the face of evil. (, –)
It is important to keep the view discussed here distinct from Molinism. According to Molinists, God’s knowledge of creaturely states of affairs does not causally depend upon creatures, but instead depends upon God’s knowledge of counterfactuals of freedom. They insist that there are true propositions that express what a creature would freely do in particular circumstances, even though they deny that the creature is in any way determined to a particular choice, whether by God or by natural causes. They further claim that God knows these truths, and that, when combined with knowledge of divine actions, the truths are sufficient for yielding certain knowledge of what all creatures will do in the actual circumstances in which they find themselves. Although fatalistic worries may surround either the eternal truth of the counterfactuals of freedom or of God’s knowledge of all free actions, this view does not qualify as divinely deterministic according to my use of this phrase, since it does not claim that God’s will necessitates all worldly events. For recent discussions of this as a motivating reason for divine determinism, see Alexander () and Pereboom (, , , ). Heath White also links divine determinism to providence, although he also invokes concerns that seem most at home in discussions of divine sovereignty. He says that meticulous providence is “one of the primary attractions” of divine determinism (White , ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
Some think that we can secure this complete providence merely by allowing for complete and perfect foreknowledge. Others challenge the idea that foreknowledge is sufficient. In any case, it is easy to see that adopting divine determinism would solve any worries associated with how God would be able to ensure that the divine plan is carried out. Concerns about divine providence, then, give some a reason to take a closer look at divine determinism. In the preceding discussion, I collected a number of reasons that some have found to support divine determinism. In each case, the acceptance of divine determinism allowed for strong predications about God – that God is sovereign, powerful, a se, and provident in very strong senses. But this is not the only way in which divine determinism has been thought to be desirable: Acceptance of divine determinism also appears to make a number of other beliefs, whether unique to a particular Abrahamic religion or common to several, easier to understand. Prophecy Many theists believe that, from time to time, God sends forth prophets. It is important to stress that many prophets within the Abrahamic religions do not seem primarily focused on revealing future events. As William Hasker has noted, prophets more often are concerned with “witnessing to the people concerning God’s purposes and requirements and seeking to recall them to obedience” (, ). Nevertheless, prophets in the Abrahamic traditions are believed to have foretold the future on some occasions. It may seem that complete divine foreknowledge is itself sufficient to explain the possibility of such prophecy: God, knowing everything, knows what will happen and passes along a portion of this knowledge to a chosen few. Recently, however, challenges have arisen concerning whether foreknowledge is, in itself, sufficient to explain how prophecy is possible. In particular, Widerker
For discussion of this issue, which turns on whether accounts of foreknowledge other than Molinism can account for providentially useful knowledge, see Hunt () and Robinson (a, b). One such motivation that I have avoided diving into here – its complexity resists quick summary – concerns the depravity of sin and the function of grace in Christian doctrine. Some divine determinists contend that their views makes grace and sin much easier to understand. See Alexander (), Baker (), Johnson (), and White (). Such claims form the most common basis for explanations of prophecy, but they need not be used. Other possible explanations are sometimes given, including the possibility that the future is not set, such that God needs to strategize to ensure that a prophecy will become true, and that prophecy is somehow a natural or quasi-natural ability. For the former explanation, see Hasker (). Avicenna provides an example of the latter. See Gutas ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Motivations for Divine Determinism
() has argued that accounts of foreknowledge that depend upon God’s eternality do not provide the tools necessary to explain how prophecy is possible, at least not given certain religious beliefs. Freddoso () has argued that Ockhamist accounts face their own difficulties when it comes to prophecy. While it might be possible for these views to survive these attacks, or for other views of divine foreknowledge (such as Molinism) to readily account for the possibility of prophecy, divine determinism makes the matter entirely straightforward. Scriptural Authorship The path from scriptural authority to divine determinism is a little harder to see, but some have taken it. To follow this path, one begins by accepting that scripture is perfect; it is exactly as God willed it to be. Defenders of this path then add that the divine inspiration of scripture did not “violate or circumvent the integrity or personality of the author” (Alexander , ). If we accept both of these claims, then we cannot endorse an incompatibilist account of free will. That is, if we were to accept such an account of free will, then we must maintain that the authors of scripture had it within their power to write their books in any fashion they saw fit. But if this is so, it seems to be the greatest of coincidences that the books turned out exactly in keeping with the desire of God. Thus, some claim that to understand the divine inspiration of scripture, we must abandon incompatibilist views of free will and adopt at least the claim that God determined the relevant actions of these writers. If we embrace
Here is Widerker’s own summary of his argument: Suppose that God knows that Jack freely pulls the trigger at T, with the intention of killing Smith. Suppose further that wanting to save Smith, God reveals this fact to Smith at T by causing him to hear at T a voice telling him about what is going to happen. As a result, Smith by taking appropriate precautions is able to save his life. Now the eternalist concedes that if Jack has it within his power at T to refrain from his attempt to kill Smith, he also has the power at T to make it the case that God does not know that he attempts to kill Smith at T. On the other hand, we may assume that God’s knowing that Jack attempts to kill Smith at T is a condition which, in the circumstances, is causally necessary for the occurrences of the event of Smith’s hearing at T a voice telling him that Jack will attempt to kill him. It seems plausible to suppose that had God not known that Jack attempts to kill Smith at T, he would not have told Smith that Jack will try to kill him, and hence, Smith would not have heard the voice he in fact heard. But then it follows that by having the power to bring about the non-obtaining of that condition, Jack can also be said to have the power at T to bring about the non-obtaining of such genuine past facts as: XT: That Smith heard at T a voice telling him that Jack will attempt to kill him. (, )
Fredosso () raises a similar worry about Ockhamist approaches to prophecy. See Alexander ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
compatibilism, then there is no tension between saying that God determined the authors of scripture to write in particular ways and saying that they acted freely in writing. Of course, theists might potentially assert that although the human actions of writing scriptures are divinely determined, not all human actions are similar in this regard. Nevertheless, once it is admitted that God sometimes determines humans’ free actions, the first steps toward divine determinism seem to have been taken. Suggestive Scriptural Texts Some have also found various scriptural passages to be suggestive of divine determinism. These texts take a number of forms. Some of them may seem to suggest a straightforward endorsement of universal divine determinism, while others may be more limited, merely suggesting isolated cases in which God determines human action. Whether the scope of determinism is large or small, defenders of divine determinism have thought such passages are important: If a passage suggests that God determines even a single free human action, this suggests that human moral responsibility is compatible with such determinism. Moreover, if the action is an evil one, this further suggests that God’s goodness is compatible with determining evil actions. More precisely, a single instance of such a case is important, because it provides reason to accept each of the following suppositions: (a) God sometimes determines human beings to act in particular ways. (b) An agent may be morally responsible for an action even if the agent is determined by God to perform it. (c) Divine goodness is compatible with God determining human agents to sin. Even taken jointly, (a)–(c) do not entail divine determinism, but it is easy to see how acceptance of these three assertions paves the way for divine determinism. Indeed, once (b) and (c) are accepted, it seems that most of the costs of divine determinism have already been incurred. In keeping with the modest aims of this section, I will not aim to present all passages that have been taken to support or suggest divine determinism, but instead will be content to present a representative sampling. Given the controversial nature of the topic, it will surprise no one that each of these passages suggests a deterministic interpretation to some and a nondeterministic interpretation to others. I will refrain from detailing nondeterministic readings of the passages, and I intend to remain firmly agnostic on the way these texts are best understood.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Motivations for Divine Determinism
To begin, consider a passage from Genesis in which Joseph is sold into slavery by his brothers. It is clear that his brothers are doing something morally wrong: Their act is spiteful and cruel, and it seems that they perform this act with full knowledge of what they are doing and in full control of their action. In short, it seems that they have done something evil and are morally responsible for having done so. Later, when Joseph confronts his brothers and addresses what they have done, he says, “you meant evil against me, but God meant it for good” (Genesis :). This remark has led some to think that God not only brings good out of evil, but that he intentionally causes others to perform evil actions, in such a way that their agency remains intact. In short, for some, this passage provides scriptural reasons for accepting (a)–(c). Other texts seem even more suggestive, describing God as working in agents’ hearts to ensure that they act in particular ways. Consider, for example, the following three passages: And the Lord said to Moses, “When you go back to Egypt, see that you do before Pharaoh all the miracles that I have put in your power. But I will harden his heart, so that he will not let the people go.” (Exodus :) But Sihon the king of Heshbon would not let us pass by him, for the Lord your God hardened his spirit and made his heart obstinate, that he might give him into your hand, as he is this day. (Deut :) There was not a city that made peace with the people of Israel except the Hivites, the inhabitants of Gibeon. They took them all in battle. For it was the Lord’s doing to harden their hearts that they should come against Israel in battle, in order that they should be devoted to destruction and should receive no mercy but be destroyed, just as the Lord commanded Moses. (Joshua :–)
In each of these three passages, God seems to cause agents to act in particular ways [providing reasons for accepting (a)], the agents in question seem to retain their own agency in such a way that they remain responsible for their actions [providing reasons for accepting (b)], and the actions in question are evil [providing reasons for accepting (c)]. Some philosophers also find that the Christian New Testament hints at or at least is amenable to divine determinism. Hugh McCann (, ) points to the following passage from Philippians: Therefore, my beloved, as you have always obeyed, so now, not only as in my presence but much more in my absence, work out your own salvation
The English Standard Version of Biblical passages are used in this text unless otherwise noted.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism with fear and trembling, for it is God who works in you, both to will and to work for his good pleasure. (Philippians :–)
Of course, nondeterminists will be quick to point out that this passage does not unambiguously assert divine determinism. Once again, however, the acceptance of divine determinism would make the interpretation of this passage and other such passages a straightforward affair. Some passages from the Quran also seem to suggest (a)–(c). Consider, for example, the following: As for unbelievers, alike it is to them whether thou has warned them or hast not warned them, they do not believe. God has set a seal on their hearts and on their hearing, and on their eyes is a covering, and there awaits them a mighty chastisement. (Q.:–) If God had willed, He would have made you one nation; but He leads astray whom He will, and guides whom He will; and you will surely be questioned about the things you wrought. (Q.:) And He created every thing, then ordained it very exactly. (Q.:)
As with other texts we have considered, these texts have yielded various interpretations, many of which do not entail divine determinism. Yet it is clear that they are at the very least suggestive. Consider Q.:–, which says that Allah has placed a covering over disbelievers’ eyes. First, this passage says that these people will continue in their disbelief, even if they are warned. Their future seems settled, and settled by Allah, since Allah places the covering over their eyes and this covering keeps them in their disbelief. Thus, (a) seems to find support here. Furthermore, (c) seems equally supported, since disbelief is a paradigm instance of an evil action. Finally, the passage explicitly says that a great punishment awaits such people. Since moral responsibility is commonly understood in terms of a desert of punishment and blame, and assuming that the punishment is deserved (as it must be, since Allah is just), this passage seems to support (b). This is not to say that if one accepts the authority of the Quran, then (a)–(c) must be accepted on pain of irrationality – scriptural interpretation is a complicated business, after all. Nevertheless, such passages do make it clear why one might be tempted, based upon such holy texts, to adopt divine determinism.
Translations of the Quran are from Arberry (). For a brief overview of such passages and how they have been interpreted historically, see Attar (, –). This author seems to assume that (b) is self-evidently false, and so implies that such scriptural passages, if taken literally, rule out human moral responsibility.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Varieties of Divine Determinism
Varieties of Divine Determinism Thus far, we have considered what divine determinism is and which reasons have led some to embrace it. I turn now to considering how God might determine creaturely events. Broadly speaking, two types of causal mechanisms have been suggested. The first mechanism embraces creaturely determinism. According to such a view, ordinary worldly events, including human choices, are determined by prior worldly events. According to this account, human free choices are determined by prior psychological states. This sort of determinism is familiar; it is similar to the sort of determinism associated with Laplace. Although such a complete deterministic picture of the universe was called into question by work in quantum physics, it very much remains a live option according to physicists, and its implications continue to be hotly contested by philosophers. I hesitate to call this physical determinism, because adherents to this view need not endorse a fully naturalistic conception of the human person. I will call such views, which claim that all creaturely events are determined by prior creaturely events, natural determinism. If we adopt such a view, a full account of divine determinism is close at hand: One need only add the claim that God caused the first creaturely state. According to such a view, creating the first state (and any contingent causal laws that govern them) determines all the rest. Since this view is commonly associated with one of its champions, Jonathan Edwards, I will call it Edwardsian divine determinism. By embracing natural determinism, such a view seems to inherit all the problems that plague physical determinism, but it also comes complete with many responses to such problems that philosophers have developed.
Strictly speaking, this description of natural determinism should be modified in light of two considerations. First, if there was a first creaturely event, it could not have been determined by any other creaturely event. Thus, we should not require natural deterministic views to claim that all creaturely events are determined by prior ones, but rather that all of them besides the first are so determined. Second, this category of views should be taken to be agnostic on whether God might intervene into natural events, breaking the deterministic sequence. If God has such power, then even if it is never exercised, creaturely events do not, strictly speaking, determine later ones. With this in mind, we should take natural determinism to require that prior creaturely events determine later ones, barring supernatural intervention. See Edwards (). Natural determinism in a theistic setting is sometimes associated with Thomas Aquinas as well. His account of the relationship between the intellect and the will is complicated and sometimes unclear, so it is difficult to tell whether he thinks humans are determined by their prior psychological states or not. For discussion, see Hause (), Hoffmann (), Hoffmann and Michon (), MacDonald (), Pasnau (), Stump (), and Williams (). My own view is that he is not this sort of a determinist (Hoffmann and Furlong ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
To take just a single example, consider the Principle of Alternative Possibilities. According to this principle, an agent is morally responsible for a given action only if the agent had the ability to do otherwise. This ability is often thought to be ruled out by physical determinism (and, it seems, by natural determinism more generally). Since the principle seems true, and because physical determinism seems to eliminate the possibility of such an ability, it seems that those who accept physical determinism must deny that we are morally responsible for our actions. Divine determinists who embrace natural determinism seem to inherit this problem, but they also seem to inherit all of the work that has been done in past decades to solve this problem. They might, for example, help themselves to so-called Frankfurt scenarios, which suggest that the Principle of Alternative Possibilities is false. Such determinists, then, face familiar problems, but they seem able to utilize familiar solutions. Although divine determinists may adopt natural determinism, they need not do so. To see this, note that the entire causal picture presupposed by the Edwardsian determinist might be rejected on the grounds that there are fundamental differences between divine and creaturely causation – differences that are not respected if we think of God merely as bringing about the initial conditions of the universe. First, we might note that divine causation proceeds by creation (producing things from nothing), whereas creaturely causation always operates by modifying what already exists. Moreover, defenders of this sort of view will claim, we should not imagine that God creates the very first creaturely state, and then this state somehow unfolds on its own to produce the history of the universe. Instead, we should think of God’s creation as a single act in which the entire history of the universe is created out of nothing. God is not like a divine watchmaker who crafts a precision instrument, winds its mechanism, and then stands back and lets it work. Now this picture of divine causality need not constitute a variety of divine determinism. Whether it does so depends upon whether the view in question affirms both conjuncts in terms of which “divine determinism” was defined. In many cases, it will
For the original development of such scenarios, see H. G. Frankfurt (). In presenting this argument against compatibilism, I do not mean to suggest that it should keep one from embracing natural determinism. Nor, in mentioning Frankfurt scenarios, do I mean to suggest that such scenarios succeed in showing that the Principle of Alternative Possibilities is false. Instead, I merely mean to indicate that embracing natural determinism as a step toward divine determinism does not change the nature of this debate: The (purported) problems and (purported) solutions remain the same. For discussions of this sort of view, see Burrell (), McCann (), Tanner (, ), and White ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Varieties of Divine Determinism
simply not be clear whether a particular account that adopts this general view of divine causation constitutes a variety of determinism. If this view of divine causation is conjoined with the claims of divine determinism, we end up with an account according to which God is like the author of a novel who develops the entire story – setting, characters, and events – down to its finest detail. This metaphor of the author is telling in three ways. First, the characters in the novel can do nothing without the author. Although some writers may say that the characters take on a life of their own, that they end up dictating the story on their own terms, this is mere metaphor: The author is always truly in control. Similarly, although God’s characters (cats and cows, as well as human agents) do quite literally take on a life of their own, they, too, can do nothing without the divine author; they have no ability to alter events independently of God. Second, when discussing a novel, it seems appropriate to discuss the actions of both the author and the characters, but there is a clear difference in the sort of causal power wielded by each. We might, for example, claim that Fyodor Dostoyevsky exercised a number of causal powers in writing Crime and Punishment. He used his imagination to create the story and his fingers to put words to the page. We might say that Rodion Raskolnikov, the novel’s main character, exercises causal powers as well. For example, he kills Alyona Ivanovna, a pawnbroker. Although it seems reasonable to claim that both Dostoyevsky and Raskolnikov caused effects, it is clear that they do so in completely different ways. Indeed, this distinction is especially clear when we look at the way each is causally related to a single effect. Raskolnikov causes the pawnbroker to die. Dostoyevsky is also responsible for her death. How does the author bring about her death? Not by hitting her with an ax, as Raskolnikov does, obviously, but also not by somehow pushing Raskolnikov to do it; he doesn’t advise, threaten, or manipulate him, for example. In fact, it seems best not to say that Dostoyevsky causes Raskolnikov to hit her with the ax, but instead that he causes Raskolnikov’s hitting her. The entire event Raskolnikov killing Alyona is created by Dostoyevsky. Now we turn to the case of divine determinism. Suppose we look at a case of human killing – a criminal being hanged in nineteenth-century
Some primary divine determinists claim that God moves creatures to act through a “premotion.” Importantly, even in such cases, the process is nothing like physically forcing an agent to do something, nor like psychologically manipulating the agent to do so. See Oderberg () for a recent defense of such a view.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
A Primer on Divine Determinism
Philadelphia, for example. Suppose Ronald is the executioner on duty, and he does his duty. He pulls the lever, releasing the trap door and dropping Frank, the criminal, to his early death. Ronald clearly is a cause of Frank’s death (although it is important to note that there are others – the law, his crimes, a jury, and the rope may each be said to have caused his death in some sense). In this view of divine determinism, God is the cause of this same event, albeit in a completely different way. God does not enter into the narrative as one more character or element of the story that drives the action, no more than Dostoyevsky enters into Crime and Punishment to nudge Raskolnikov into killing. Nor need we suppose that God creates an earlier creaturely state that inevitably leads to Ronald’s action. Instead, God creates Ronald acting. Neither Raskolnikov nor Ronald can resist the narratives that have been created for them. Moreover, although both Dostoyevsky and Raskolnikov certainly brought about the death of the pawnbroker, in some contexts it would be odd to say that Dostoyevsky killed her. Moreover, it would be very odd indeed to say that Raskolnikov and Dostoyevsky together killed her. To do so seems to elide the important differences in the ways these two agents are causally related to the same event. Similarly, according to this view of divine determinism, it is true that both God and Ronald brought about the criminal’s death, but it would be potentially misleading to say that God and Ronald together killed him. To highlight the difference between the way in which God and creatures exercise their causal powers, defenders of this view sometimes say that God exercises primary causation, while creatures utilize secondary causation. Following Heath White (), I will call the sort of divine determinism under discussion primary divine determinism. These two models of divine determinism – Edwardsian divine determinism and primary divine determinism – differ in important ways. While Edwardsian determinism inherits a familiar set of problems and a familiar set of solutions, the terrain surrounding primary divine determinism is less explored. We might wonder, for example, whether such determinists would inherit the familiar challenges that plague physical determinism.
Although it seems open to authors – including, perhaps, to God – to do so. Christians think that the second person of the Trinity, who was fully God, entered into the creation narrative right alongside the other actors. Similarly, Dante writes himself into the Divine Comedy. One might worry that although it sounds perfectly ordinary to say Raskolnikov killed the pawnbroker, it sounds odd to our ears to say that Dostoyevsky killed her. I think that whether this sounds odd will depend upon context. If, for example, readers had fallen in love with the pawnbroker, it would not sound odd to hear them ask Dostoyevsky, “Why did you have to kill her?”
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Varieties of Divine Determinism
In some cases, they may; in others, they might not. Furthermore, if they do, they might not inherit all of the same solutions that have been proposed by physical determinists; some such solutions may be unworkable given their broader views. Before moving on to some problems that seem to face divine determinists, it is worth noting that nothing stands in the way of adopting a hybrid view according to which all creaturely states are determined by prior states and God directly brings each creaturely state into existence through primary causation. Although it is unclear whether such a view offers any particular advantages, one might certainly think that each part of the view is independently plausible and so affirm both. As far as I can tell, however, there are no particularly troubling problems that such a hybrid view uniquely faces, so I will not further consider such a view in this work.
To take just one example, consider the problem of alternative possibilities discussed earlier. The primary divine determinist may maintain that there is an important sense in which agents like Ronald can do otherwise, even if agents in physically determined worlds cannot. Thus, they will claim, a popular argument against compatibilism does not even arise as a challenge to their view. Perhaps, however, they are wrong. If so, can they utilize all the same responses that ordinary compatibilists have devised? Perhaps not, although some of them – for example, those utilizing Frankfurt scenarios – will still be available to them. One proposed solution to this problem (although it is considered fairly dated today) is to paraphrase “could have done otherwise” as “would have done otherwise if the agent had wanted to do so.” The primary determinist might not be able to utilize such a solution, since (returning to the example of Ronald) it is not true that Ronald would have done otherwise if he had wanted to. He might have done otherwise, but whether this is so depends ultimately not merely on what he wanted but on what God had decreed. Some may think that Thomas Aquinas adopts such a view. As noted earlier, some interpreters suggest that he adopts natural determinism; likewise, it is not uncommon to find Thomists who attribute primary divine determinism to him, although this is also controversial. For discussion, see Long, Nutt, and White (). I tend to think that Aquinas does not hold such a hybrid view (at the very least, I think he is not a natural divine determinist), but attributing such a view to him is not unthinkable.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.002
Divine Determinism and Free Will The Consequence Argument
Introduction
Divine determinism faces many challenges. These challenges claim that divine determinism threatens important claims about ourselves, about God, or about the relationship that obtains between ourselves and God. To begin, I wish to consider the challenges that divine determinism seems to pose to our view of ourselves – in particular, that divine determinism appears to threaten our view of ourselves as free agents, as masters over our actions and as morally responsible for them. While there is important controversy concerning some elements of our pre-philosophical understanding of ourselves as agents, it is uncontroversial that pre-philosophically we take ourselves to have significant control over our choices in such a way that how we act is within our power. Moreover, we take both ourselves and others to be morally responsible in what Derk Pereboom calls the basic sense. According to this view, we are sometimes fitting targets for blame and punishment, as well as praise and reward, not merely in virtue of consequentialist considerations, but because we deserve it. In fact, in the present work we can understand moral responsibility in these terms. That is, agents are morally responsible for their acts if and only if they would deserve blame or punishment for them (if the acts were bad) or praise or reward for them (if they were good). As the history of philosophy shows us, this picture is called into question when determinism is posited. Physical determinism seems to make a mockery of the entire concept of self-control. If past states determine what I will do tomorrow, then what possible domain remains within my power today? Of course, the history of philosophy also bears
Much recent work in experimental philosophy has focused on pre-philosophical intuitions about the free will issue, especially concerning the compatibility question. For an introduction to some of this work, investigated in relation to traditional approaches, see Björnsson and Pereboom (). See Pereboom ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Introduction
witness to frequent and sustained attempts to show that we need not call these pre-philosophical beliefs into question merely because of physical determinism. Some have found compatibilism intuitive, while others who first resisted were converted to this view under argumentative force. Some philosophers, however, have remained firm in their rejection of compatibilism. Some think that divine determinism is no better than physical determinism in this regard. They claim that the very same arguments (or close analogues) against ordinary compatibilism also apply against divine compatibilism. Two different questions might be raised at this point. First, is there any reason to worry about determinism at all? Are compatibilists correct in claiming that the apparent incompatibility between determinism and human free will is merely apparent? Second, are divine determinists differently situated in regard to the strongest arguments for incompatibilism? In other words, will divine determinists have an easier or harder time than ordinary compatibilists in responding to the arguments of incompatibilists? The first question has resisted definitive solution for millennia. Although I remain convinced that much progress has been made on this issue, especially in the last few decades, a single chapter of a book devoted to the philosophy of religion is hardly the place to attempt to settle it. In this and the next chapter, I will take up the second question. A fully comprehensive treatment would examine every argument for incompatibilism, investigating whether divine determinists are ever better (or worse) off than those who endorse physical determinism. I will not attempt such a complete examination. Instead, I propose to analyze arguments for incompatibilism that have recently gripped philosophers: the Consequence
William of Ockham () provides an early example of such a claim. He notes that if human actions are determined by the divine will, then its acts are as necessary as that of fire. See, for example, Vicens (). This view is not universally shared. Peter van Inwagen, for example, writes, “In my opinion – deeply prejudiced, to be sure – the philosophical literature on free will, subsequent to the publication of my book and Lewis’s essay, has degenerated into a sterile scholasticism (in the pejorative sense of the word; I do not mean to imply that scholasticism in the pejorative sense was any more common in the medieval schools than it has been in any other philosophical community). That is, there have been no new arguments or ideas of any real consequence” (, ). Others have gone even further than I have in praising the value of recent work. Saul Smilansky writes, “In the last generation or two, the free will debate has been one of the most exciting areas of philosophical endeavor, where undeniable progress has been made both in clarifying older theories and understanding their strengths and weaknesses, and in discovering new pertinent questions and positions. The progress has been so great that it might be claimed that here (unlike with most philosophical problems) there is not much worth reading that was written before, say, ” (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument Argument and manipulation arguments. These arguments are thought to be among the very best that incompatibilists can devise, so they deserve careful attention. In the present chapter, I take up the Consequence Argument, delaying consideration of manipulation arguments until the next chapter.
The Consequence Argument and the Revised Consequence Argument
The Consequence Argument (CA) aims to show that if physical determinism is true, then no one has ever had a choice about anything. Its aim is ambitious, and its credentials are excellent: Although it has certainly failed to convince everyone, it has stood for several decades as one of the most significant challenges to compatibilism. Informally, the CA begins with the claim that the past and the laws of nature are somehow outside of our control. It doesn’t matter, for the purposes of this argument, whether these are outside of our control because they are logically or metaphysically necessary, or merely because we are not in a position to exercise control over them. What is important is that nothing we can do can alter either the past or the laws of nature. The argument continues by noting that given determinism, propositions about the past together with propositions expressing the laws of nature entail propositions expressing everything we will ever do. Thus, it seems, we lack control over each and every one of our actions. The argument may be formulated in a variety of ways, but consideration of the following form is sufficient for the present purposes. To begin, we must understand Np. Van Inwagen, the creator of several important versions of this argument, tells us to read this as “p and no one has or ever had any choice about whether p.” To this we add the following two inference rules: α: □p implies Np. β: (Np & □ (p ! q)) implies Nq.
For van Inwagen’s original formulations of this argument, see his Essay on Free Will (). For the present purposes, we should understand this to be restricted to created entities. This differs from van Inwagen’s original formulation. This reformulation of Beta is utilized by Finch and Warfield () and Widerker (). Widerker refers to this principle as β’, and Finch and Warfield call it Beta to distinguish it from van Inwagen’s original formulation of Beta. Since I will utilize only this formulation of the transfer principle, I will refer to it simply as Beta. For other possible ways to formulate the CA, see Huemer () and O’Connor ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
The Consequence Argument & the Revised Consequence Argument Take F to be any true proposition, P to be a proposition expressing the complete state of the world at some point in the past before any humans existed, and L to be a proposition expressing the conjunction of the laws of nature. We may then formulate the CA in modal fashion: . . .
□ {(P & L) ! F} N(P & L) NF
Consequence of Determinism Fixity of the past and laws , , β
Determinists interested in resisting this argument apparently must do one of the following: reject α, reject β, or reject . Rejecting α seems unreasonable. If a proposition is necessary, it is unclear how anyone, even God, could have a choice about it. The second premise looks solid, too. It is unclear how anyone could possibly have a choice about what happens in the distant past, and the laws of nature seem beyond our reach as well. This leaves only the second option, denying β, but this inference rule seems intuitively obvious. The CA thus appears to present a significant challenge to compatibilism. Nevertheless, it has had its share of enemies. Most of those who attack this argument target β. They have undoubtedly had some success with this approach; van Inwagen () himself has admitted that, given his original formulation of N and β, β is invalid. Nevertheless, even McKay and Johnson (), who demonstrated the invalidity of β to van Inwagen’s satisfaction, offer a subtle modification that shores up this demonstrated weakness of the argument. As presented previously, this argument makes use of an improved formulation of β, which, as Alexander Pruss () has proven, follows from the weakening rule for subjunctive conditionals. Whether the force of this argument endangers divine determinism may depend on which sort of divine determinism is adopted. In Chapter , I distinguished between two broad categories of divine determinism. According to the first, which I call Edwardsian divine determinism, we might think of God as one more causal force in the universe, albeit the first and most powerful one. This God brings a universe into being, with each state of the universe then being sufficient for the next. According to such a view, God determines the world by bringing into existence a naturally deterministic universe. Such a view adopts an ordinary compatibilist account of one form or another, thereby adopting both its costs and its benefits. In particular, it is clear that whatever the merits of the CA, it threatens these divine determinists, at least those compatibilists among them, and ordinary compatibilists equally.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument According to a second sort of divine determinist, we should understand divine determinism in a way that is quite different from physical determinism, and we should do so because divine causation is radically different from ordinary creaturely causation. According to this second view, which we will call primary divine determinism, we should not think of God as being the first link in a causal chain that stretches through history and includes ordinary creatures as causes and creaturely events as effects. God is not merely one more cause – even if the first and most important one – operating within the same sphere and in the same sense as rivers, cats, and humans. Instead, we should think of God’s causal role as radically different from the causal role played by all other things. If this is so, defenders of this view claim, we should not assume that divine determinism and physical determinism closely resemble each other. Most importantly, we should not assume that if physical determinism eliminates free will, then divine determinism does as well. If this position is correct, then one of the major challenges ordinarily thought to face divine determinism – namely, that it undermines free will just as surely as physical determinism would – may be misguided. The question to be investigated here is whether these divine determinists are in a position to reject the CA. In one sense, these divine determinists obviously should not fear the CA. This argument, after all, does not conclude anything at all about divine determinism; it merely concludes that if physical determinism is true, then nobody has ever had a choice about anything. Since they reject physical determinism, they may easily avoid the conclusion of the argument by denying the first premise. Nevertheless, although ordinary formulations of this argument do not target divine determinism, the general thrust of the argument – that if determinism is true, then things over which we have no control settle all our actions – seems to threaten divine determinism. This intuition is correct; the CA may be reformulated in such a way that it threatens even those divine determinists who eschew physical determinism. I will call this reformulation the Revised Consequence Argument (RCA).
For examples of this claim, see Crabtree () and Oderberg () Consequence-style worries for divine determinism have been highlighted by other authors in one form or another. For example, Luis de Molina, writing in the sixteenth century, seems to anticipate the use of Beta in the RCA: “[W]hen the will operates by necessity of an antecedent supposition that in no way depends on the will, it does not operate freely.” This passage, which is from Molina’s Censura, , is quoted and translated in Matava (, ). Leigh Vicens () provides an alternate formal version of a Consequence-style argument against divine determinism, see her ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
The Consequence Argument & the Revised Consequence Argument In the RCA, D is a proposition that expresses the complete sum of all divine causal activity, desires, intentions, and/or decrees that are explanatorily prior to the history of the universe, and that, taken together, are explanatorily prior to each and every creaturely event. Take F to be any true creaturely proposition – that is, a proposition concerning only creaturely events. . . .
□ (D ! F) N(D) NF
, , β
The RCA appears to threaten divine compatibilism to the same extent that the CA threatens ordinary compatibilism. It uses the same inference rules, so if β is the weakest link in the argument, then these two arguments stand and fall together. Proposition , like , seems unassailable by those whom the argument targets; divine determinists can no more deny than physical determinists can deny . Is the denial of (given divine determinist convictions) any more reasonable than the denial of , which says that humans have no choice about the distant past and the laws of nature? It does not seem to be. Proposition seems intuitively obvious; is different, but no easier to deny. Claiming that humans sometimes have a choice about some of God’s actions is not sufficient for denying . Determinists would need to claim that sometimes we have a choice about a very particular proposition – one expressing the divine actions, intentions, and decrees that are explanatorily prior to every worldly event. It is very difficult to see how any human could have a choice about this. One may hope to solve this difficulty by claiming that God causes us to act but in some way leaves it open to humans how and/or whether they act. According to this view, God’s causal contribution acts in one of two ways. In the first way, God’s causal contribution is specific but impedible: The human agent can act (or refrain from acting) in such a way that God’s causal contribution does not terminate in a human action. This view suggests that God could operate in such a way that even given the divine influence, we may still somehow resist acting at all. In such a case, all actions are caused by God, and all humans can do is somehow resist God’s
Perhaps some Christians will claim that the second person of the Trinity, who is united to a human nature, does have a choice about this. Whatever may be said of such a claim, it clearly is of no help in preserving ordinary human freedom. For a discussion of such views, see Bonino ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument influence by inaction. In the second way, God’s causal contribution is indeterminate in the sense that it may cause one of any number of different human acts, and which act it ends up causing is somehow within the control of the human agent and outside of God’s control. Such views may or may not succeed, but they do not qualify as divinely deterministic on my use of this term, since they deny that the facts about God’s will determine all creaturely events. Thus, it seems, the only avenues by which one might deny are cut off to precisely those who need them: divine determinists. The RCA, then, seems as formidable as its namesake. Both arguments depend on extremely plausible premises and on two principles that, in some form or other sufficient for the argument, seem quite plausible as well. Most importantly, the easiest way to avoid RCA – denying Beta – is exactly the path that has proved most popular in attempting to avoid the CA. If RCA can be avoided only in this fashion, then physical determinists and divine determinists are equally positioned: Each can maintain their views, but only if they are willing to deny this seemingly reasonable principle.
Determinist Responses: The Importance of Causal Differences
We might conclude that this settles the matter: The CA poses just as much of a threat to ordinary determinists as the RCA does to divine determinists. We should not be too quick to judge, however, since some divine determinists vocally deny that divine determinism is as threatening as physical determinism. They maintain that although physical determinism, if true, would give us sufficient reason to reject belief in free will, divine determinism is relevantly different. They deserve their hearing. The question I wish to examine is not whether they are fully correct; evaluating this would require settling the issue of whether physical determinism is compatible with free will – a question which readers familiar with the intricacies of the debate will agree is impossible to sufficiently address in
Maritain () suggests a view that bears a resemblance to this. This denial has become even more painful since Pruss () proved that it follows from the weakening rule for subjunctive conditionals. Thus, to deny Beta, one would also need to deny the weakening rule. This cannot be done while holding onto Lewisian semantics, so this will need to be discarded as well. Crabtree (), for example, self-identifies as a divine determinist. Some others hold views that at least seem to involve both elements of divine determinism as defined in Chapter (the facts about God’s will wholly determine every other contingent fact, and the facts about God’s will explain every other contingent fact), but do not use this phrase to characterize their views. See, for example, Garrigou-Lagrange (), McCann (), Oderberg (), and Tanner ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Determinist Responses: Importance of Causal Differences
anything short of a volume dedicated solely to this issue. Instead, I intend to zero in on the question of whether divine determinism is relevantly different from physical determinism when it comes to responding to the worries raised by Consequence-style arguments. Defenders of the sort of divine determinism under discussion are often content to point to important differences between their view of divine causality and the operations of physical causality given physical determinism. In examining whether they may resist the RCA more easily than ordinary compatibilists may resist the CA, we will need to examine these two sorts of determinism to see whether the differences between them, important as they may be, are relevant for the present discussion. Those writing about divine causality often emphasize how it is radically different from human causality. Indeed, this difference seems to be of fundamental importance in understanding the distinctions between physical determinism on the one hand and primary divine determinism on the other hand. Claims about the radical difference between divine causality and ordinary worldly causation are common to many theists, and such claims may enable us to see why the CA is more powerful than the RCA, whether the authors making such claims are divine determinists according to my definition or not (and often it is difficult to tell). Some divine determinists explicitly argue that physical determinism may well threaten free will, while divine determinism in some form or other does not. J. A. Crabtree, for example, writes: One cannot, without independent supporting evidence or argumentation, generalize from physical causation to any and every sort of causation. One cannot validly conclude that because every house in Mills City is made of wood, every house everywhere is made of wood. Neither can one validly conclude that because a choice made out of physical necessity is not a real choice, a choice made out of any sort of necessity is not a real choice. (, )
Some might very well be convinced that physical determinism precludes the possibility of free will, think that divine determinism seems relevantly similar because both involve a sort of causal necessity, and conclude that divine determinism, too, eliminates the possibility of free will. If the only worry about divine determinism stemmed from such an argument, then
Some authors call this theological or divine determinism, while noting that the uniqueness of the causation involved in this view may make this phrase misleading. See, for example, McCann and Johnson (). As stated, this reasoning takes the form of an argument from analogy, although similar reasoning might take the form of a generalization, as Crabtree suggests.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument although legitimate concerns might exist, we would be hard-pressed to consider them very serious. As we have seen, the worries that plague divine determinism are not so vague. The very sort of reasoning captured by the CA – generally considered one of the most powerful arguments against compatibilism – can be shown to threaten the claim that divine determinism is compatible with human freedom. Importantly, nothing in the argument turns on generalizing about causation at all. No step of the argument makes any claims about the nature of divine causation beyond those made by divine determinism itself. Instead, both arguments turn on what we might call the transfer of powerlessness, which is captured by Beta. Next, we might look to divine transcendence in the hopes that it can provide a justification for the claim that natural determinism threatens free will while primary divine determinism does not. On the view of transcendence in question, God’s causal influence should not be thought of as one more link in the worldly chain of causality. Kathryn Tanner explains this as follows: Picture the plane of nondivine existence (which is the whole of the world as we know it) suspended in existence at each and every one of its points, and therefore in its entirety, by God’s creative action.
Heath White () discusses the difference between ordinary creaturely causation (secondary causation) and transcendent divine causality (primary causality) in terms of vertical and horizontal axes. In this depiction, we should think of God’s causality as operating along a vertical axis, and as creating the entire horizontal causal chain from nothing, holding it in existence, and underpinning each causal link in the horizontal chain. God is, therefore, not in competition with ordinary causes, but instead is necessary for them; they can do nothing – and indeed would be nothing – without divine causal support along the vertical axis. Since their causal roles are fundamentally different, no worries should even arise about one supplanting the other. According to some authors, since God is the causal source of everything along the vertical axis, there should be no worries that divine causality renders all apparent contingent events necessary. Joseph Trabbic claims: God predetermines not only the specific action of the created agent through physical premotion, but also its general modality – whether it is necessary, contingent, and/or free. (, )
Other divine determinists use this language as well. See, for example, Helm (, –). For this sort of defense, see Tanner (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Determinist Responses: Importance of Causal Differences
David Oderberg further explains, describing the divine motion by which creatures are moved into act: It is predetermining in the sense that the secondary cause infallibly does what God moves it to do, but its modal status qua secondary cause is not affected. (, )
Tanner writes: It is true that if the necessary infallibility that characterizes God’s calling forth of a human choice were to necessitate that choice absolutely, it would not be possible to say that the creature has the ability to choose otherwise vis-à-vis the created conditions of choice. A theologian holding to our picture has to deny that the necessity with which a human being’s choice follows from God’s will for it makes that choice simply and in all respects necessary. (, )
Before moving on to the question of whether this sort of consideration provides divine determinists with a unique way of avoiding the RCA, it is worth considering whether we should count this view as deterministic at all. There seem to be two reasons to think that the views of three of the four authors considered earlier are not divinely deterministic. (The fourth – White – both identifies the view he defends as divine determinism and creates the definition of “divine determinism” adopted in this book.) First, neither Oderberg nor Trabbic identifies their views as divinely deterministic, and Tanner does not consider herself to be a determinist. Second, each of these three authors denies that all creaturely events are necessary, but necessity was built into my definition of “divine determinism.” Let’s consider each worry in turn. In response to the first worry, I once again reiterate that I am not making an effort to police philosophers’ use of the phrase “divine determinism” or the ways in which they prefer to classify their own views. Nevertheless, whatever words these philosophers use to categorize their views, it at least seems that each is committed to divine determinism as characterized in Chapter . If this is so, and if these philosophers have offered defenses of their views that may enable us to see problems with the RCA, then they are worth considering. Before finally dismissing this worry, consider the following passages that suggest to me that these authors are divine determinists as understood in this book. Tanner writes: God’s creative activity calls forth or holds up into being throughout the time of its existence what has its own integrity as a nondivine existence, and
See Hasker (, ) for the claim that Tanner does not consider herself to be a determinist.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument this nondivine existence has to be considered the consequence of God’s creative calling forth and holding up as a whole, in its order and in its entirety, in every detail and aspect. (, )
She also claims: . . . God’s creative intention for the world cannot be hindered, diverted or otherwise redirected by creatures. Lacking any defect or internal principle of corruption, moreover, God’s creative calling forth is indeed unconditionally and necessarily efficacious. (, )
Trabbic defends a view that includes the claim that . . . God efficiently and predeterminately moves all created agents – that is, all agents other than God himself – to act, every time that they act and in the precise way that they act . . . (, )
Finally, Oderberg talks of the “eternal and infallible divine decree” and also notes that the motion by which God moves creatures is causally prior to that which it moves. I take it that if it is causally prior, it is also explanatorily prior. W. Matthews Grant (a) espouses a view that is similar in some respects to those discussed here, but that clearly falls outside of divine determinism, since it rejects the claim that God’s causal activity is prior in explanation to creaturely actions. As far as I can see, however, none of the authors discussed joins Grant in this rejection. If they do, then like Grant, they would not count as divine determinists according to the definition set out in Chapter . The second worry with treating these three authors as divine determinists stemmed from issues of necessity: Each of the authors claims that in some sense some creaturely actions remain contingent, but the definition of determinism references necessity. Recall that following White (), I take divine determinism to be the conjunction of the following two claims: The facts about God’s will wholly determine every other contingent fact, and the facts about God’s will explain every other contingent fact. I take x to determine y when given x, y must be the case, exist, or occur, where the “must” is a must of metaphysical necessity, and metaphysical necessity is understood in terms of colocation in possible worlds. Given these stipulations, to say that given x, y must be the case or must occur is to say nothing more than that in every possible world containing x, y is the case or occurs.
For the quoted phrase, see Oderberg (, ); for the claim about causal priority see Oderberg (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Determinist Responses: Importance of Causal Differences
Is the conditional metaphysical necessity discussed here compatible with the claims of these three authors? There is reason to think that it is. Tanner, for example, does not deny that human choices are conditionally necessary, but merely states that God’s causal influence “necessitate[s] that choice absolutely” (, ). But divine determinism does not suggest that human choices are absolutely necessary – just that they are conditionally necessary. Moreover, she says, If the claim that human choice is not necessitated by any given conditions must be taken in an absolute sense, with a universal reference that includes God’s creative working among those conditions, then the theologian can only say that such a thing does not exist. (Tanner , )
Oderberg is careful to limit his claims of contingency to the level of secondary causes, speaking of the “modal status qua secondary cause” (, ). No matter how exactly we understand such a modal status, it is not clear it rules out necessity conditioned upon God’s primary causation – in fact, it seems to be explicitly carving out a modal category that is compatible with such conditional necessity. Of the three authors under discussion, Trabbic is the hardest to pin down on the issue of modality, largely because it is not the theme of his work. He talks of God predetermining an action’s “general modality,” but does not precisely define this phrase. In fact, he admits that the problem of predetermined contingent events is “a hard nut to crack,” and adds, “I’m not going to pretend to crack it here” (Trabbic , ). What he does offer in “the aim of making the doctrine intelligible” is a discussion of diachronic ability, noting that an agent predetermined to do x at time t retains the ability to do other than x at times other than t. This discussion does not even suggest that human actions are not conditionally metaphysically necessary. I do not claim that this discussion demonstrates that Tanner, Oderberg, and Trabbic defend views that fall within the scope of divine determinism; I merely mean to indicate why it seems to me that their views fall within this scope. Assuming I am correct, we may then turn to their shared claim: Determined human actions remain contingent. This claim – that God creates not only choices but also their status as free and contingent – is apparently intended to assuage the concerns of indeterminists. Can it help us respond to the RCA? Many indeterminists will be suspicious of the claim that God has a free hand in establishing the modal status of entities by fiat in a way that is detached from their causal history. As Trabbic notes, “For many people, talk of a predetermined
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument contingent and/or freely willed action is oxymoronic” (, ). Indeed, whether we can make sense of predetermined contingent actions depends upon on what we mean by “contingent.” Let’s assume that, as these divine determinists tell us, our actions are contingent, even though they are determined by God. Does this provide an obvious way to avoid the RCA? It doesn’t appear so. The argument nowhere claims that determining divine causality bestows necessity on human free choice. Reflection on Beta suggests that determined actions and necessary actions are similar insofar as they presumably share a freedom-undermining property, that they are entailed by propositions whose truth-values are outside the influence of our choices. Thus, merely denying an action’s necessity provides no comfort whatsoever to those troubled by the RCA. Doubts about divine determinism remain, but determinists think that more can be said on this subject. They claim that although divine determinism does seem to impose something similar to necessity or even a kind of necessity upon an action, it is of a perfectly harmless sort. If they are correct, then it seems that there must be something wrong with the RCA, and it would be reasonable to hope that this same reflection upon types of necessity might reveal that problem. Trabbic points us to a distinction, often made by Thomists, between two sorts of conditional sentences that include modal operators, claiming that this distinction can help us understand that created agents in divinely determined worlds retain the ability to do otherwise (and thus, it seems, remain free). Consider the following two sentences: Stands: If God causes Ken to stand, then necessarily Ken stands. Sits: If Peter sits, then necessarily Peter sits.
Let’s begin by considering Stands. The first thing to note is that this sentence can be interpreted in two ways. Consider the following two disambiguations: Stands: If God causes Ken to stand, then the proposition “Ken stands” is a necessary truth. Stands: The proposition “If God causes Ken to stand, then Ken stands” is a necessary truth.
For discussion of this distinction, see Matava () and Trabbic (). Note that Matava provides a more detailed discussion of this distinction, especially concerning its relation to claims of necessity. Thomas Aquinas does appeal to these distinctions, but as is often the case with Aquinas, his own position on how these distinctions relate to the issue at hand is controversial. For his discussion of this distinction, see Thomas Aquinas ( q. , a. , ad ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Determinist Responses: Importance of Causal Differences
Note that these claims are not equivalent. In fact, a divine determinist may wish to deny the first and affirm the second. Such a position would not be unjustified. We might note that there is no possible world in which God causes Ken to stand and yet Ken does not stand. Thus, Stands is true. However, even if God causes Ken to stand in the actual world, there are (we may suppose) possible worlds in which God refrains from causing Ken to stand and as a consequence Ken does not stand; in this scenario, “Ken stands” is not a necessary truth, and Stands is false. We shall call the necessity involved in propositions like Stands absolute necessity and that involved in cases like Stands necessity of supposition. Domingo Báñez – a key figure in the Thomist tradition – and Kathryn Tanner make two key claims, and, although it is not perfectly clear, Trabbic seems to join them. First, divine determinism imposes only necessity of supposition upon human actions, not absolute necessity. Second, the necessity of supposition constitutes no threat to free will. If they are correct on these points, then maybe we can find our way to a solution to the RCA. To argue for this claim, these authors point to cases of suppositional necessity that clearly pose no threat to free will. To see this, let’s return to Sits: If Peter sits, then necessarily Peter sits. Once again, we can distinguish two possible meanings of this sentence: Sits: If Peter sits, then the proposition “Peter sits” is a necessary truth. Sits: The proposition “If Peter sits, then Peter sits” is a necessary truth.
Sits seems to be obviously false, but Sits seems to be obviously true. We might say, then, that whatever account of human action we give, human acts are marked by the necessity of supposition. We can then link this account of suppositional and absolute modality with different senses of the ability to do otherwise. The authors we are considering don’t name these different sorts of abilities, but let us adopt some for ease of expression: Complete ability to do otherwise: An agent has the complete ability to do otherwise only if the action the agent performs is neither absolutely necessary nor suppositionally necessary.
Trabbic doesn’t provide the names for these sorts of necessity, but the names are used within the Thomistic tradition he is drawing from. See, for example, Matava (, ). See Matava (, –) for an overview of Domingo Báñez’s view. See Tanner (, –) and Trabbic (, –) for more recent suggestions along these lines, although Tanner is more explicit on this point than Trabbic.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument Modest ability to do otherwise: An agent has the modest ability to do otherwise only if the action of the agent is suppositionally necessary, but not absolutely necessary.
Tanner and Trabbic claim that possessing what I call the complete ability to do otherwise is neither desirable nor even possible, whether divine determinism is true or not. After all, “If I x, then I x” is necessarily true for any value of x, whatever account of causation – natural or divine – we accept. Of a freedom requiring the complete ability to do otherwise, Trabbic writes: But this . . . sort of freedom should not be defended by anyone, because what it amounts to is a freedom to overturn the principle of noncontradiction, which is not a real freedom at all but a mere fiction. (, )
Tanner explains that the lack of the complete ability to do otherwise should not worry us, and so (it seems) the necessity imposed upon actions in divinely determined worlds should not worry us either: A theologian who holds to our picture of God as creator might claim that this is an inability like that whereby I say that insofar as I am choosing to do something I cannot choose not to so choose. This inability to do otherwise would be like the creature’s inability to do other than what God wills in that it would hold even under the most libertarian of circumstances. All it means is that I cannot choose and not choose to do the same thing at the same time. If I am choosing to do it at t, I cannot choose not to at t. This sort of inability does not indicate any real inability not to choose at t what I do choose. (, )
It seems, this line of reasoning continues, that divine determinism does not eliminate the ability to do otherwise any more than any other view does. If this is correct, then it seems the RCA demands closer scrutiny after all. The RCA apparently shows that if divine determinism is true, then no human has a choice about anything. The line of reasoning discussed previously (let’s call it the Necessity of Supposition Argument [NSA]) suggests that divine determinism is compatible with the strongest sort of ability to do otherwise that is coherent. Which one is mistaken? Whatever is ultimately concluded about the RCA, the NSA fails. The NSA distinguishes between two ways we might think of the ability to do otherwise, rules one out, and concludes that the other is sufficient.
It is logically possible that neither is mistaken, and thus the strongest sort of ability to do otherwise that is coherent is still insufficient for humans to have a choice about anything. In short, a third possible lesson is that free will is incoherent. The NSA does not maintain that possessing some sort of ability to do otherwise is sufficient for free will, but it does assert that the sort of ability to do otherwise that is compatible with divine
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Determinist Responses: Importance of Causal Differences
Unfortunately, the NSA never provides any reason for thinking that the two sorts of abilities it discusses are the only sorts we might reasonably desire. Incompatibilists will rightly point out that two importantly different sorts of actions are suppositionally necessary. The first sort are necessary in this sense only in virtue of propositions containing antecedents, the truth-values of which are within the relevant agent’s control. Thus, the sort of necessity we find in Sits (“If Peter sits, then Peter sits” is a necessary truth) seems perfectly harmless. But consider the following: Kneels: “If George Washington crossed the Delaware in , then Michael kneels in ” is a necessary truth.
Kneels seems to be fundamentally different. Supposing the antecedents of both Sits and Kneels are true, it seems that Peter may have a far more robust freedom than Michael has. Merely pointing out that neither Peter’s sitting nor Michael’s kneeling is absolutely necessary and that both are suppositionally necessary provides no reason whatsoever to doubt this intuition. Let’s look closer at Sits and Kneels. I suspect that many incompatibilists would be more worried about Michael’s freedom than they would be about Peter’s. Why? On the one hand, we have good reasons for thinking that Michael does not have a choice about the antecedent of the relevant conditional (“George Washington crossed the Delaware in ”). On the other hand, we may assume that Peter does have a say over the relevant conditional in his case. The discussion of complete and suppositional necessity gives us no reason to reject thinking this difference matters. This worry was expressed more than years ago by Luis de Molina: “[W]hen the will operates by necessity of an antecedent supposition that in no way depends on the will, it does not operate freely.” The cases of Michael and Ken seem relevantly similar. Ken’s action is necessitated by God’s will and Michael’s action is necessitated by Washington’s boating. Ken has no more say over God’s will than Michael does over the actions of long-dead leaders. The analogy with Peter and the suppositional necessary that attaches to his sitting fails, since Peter’s sitting is necessitated by his sitting, over which (we may suppose) he has control. Importantly, both the CA and the RCA seem to point to the issue of whether agents have a say over the conditions that necessitate their actions.
determinism is sufficient for meeting whatever “ability to do otherwise” requirement one might reasonably propose. This translation of Molina’s Censura, is found in Matava (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument In fact, we might think of transfer of powerlessness principles as claiming that in cases of suppositional necessity, if agents have no control over the supposition – the truth of the antecedent of the relevant conditional propositions – they likewise lack control over the action discussed in the consequent. Since the NSA offers us no reason for thinking that Michael and Peter are similar in all relevant respects, it cannot give us a reason to reject the RCA.
Assessing the Options in Responding to the RCA If what I have said is correct, then the RCA continues to be a threat to divine determinism. None of the claims made by divine determinists – or at least none discussed so far – provides clear guidance in how to respond to this argument. In general, it seems that divine determinists must choose among four broad options in responding to the RCA. First, they may claim that all Consequence-style arguments fail and that the worries that motivate them are misguided. The second option is to attempt to show that while the CA succeeds, the RCA does not. The third option is to attempt to show that the CA and the RCA share some flaw, and so both fail, but that the worries motivating the CA do prove serious, while those motivating the RCA are misguided. Finally, as a fourth option, the divine determinist may admit that both the CA and the RCA succeed, but deny that much real harm is caused by such an admission. If the first option is embraced, divine determinists will ultimately face the same difficulties as ordinary compatibilists (even if they resist affirming physical determinism). Thus, this response cannot be evaluated apart from a larger evaluation of ordinary compatibilism, which, as I have said, is firmly beyond the scope of the present work. I don’t see good prospects for the second option. As noted earlier, there simply does not appear to be a good way for divine determinists to resist the RCA without giving up either their adherence to determinism or Beta. If Beta is rejected, then the CA fails just as surely as the RCA does. The third and fourth options require more investigation. Some divine determinists’ suggestions seem to point toward the possibility of embracing the third option. The clearest proponent of such a view is David Oderberg. Oderberg defends premotionism, the view that for
It seems to me that some of Hugh McCann’s comments point in the same direction, but his own position is difficult to pin down. For this reason, I will not discuss his position at length. See McCann (, ) for comments that seem to suggest this third option. If this is what he means
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Assessing the Options in Responding to the RCA
any creature (free or unfree, sentient or insentient) to act in any way, God must move it by imparting a premotion. This premotion is distinct both from God and from the creature’s movement, is prior in the order of causation, and infallibly moves the agent to do what God has ordained. Oderberg provides an overview of a Consequence-style argument that seems to threaten his own view. He writes: [An objector] might propose an operator such as “X never has a choice about whether or not P”, or “X has no power causally to falsify P”, for some agent X and proposition P. He might then argue as follows. If God moves S to perform A, for particular agent S and action A, then S never has a choice about whether or not God so moves him. (Reason: God’s total independence of all creatures.) If God moves S to perform A, this logically entails that S performs A. (Reason: the infallible efficaciousness of the divine decree.) By transfer of necessity (the necessity of the operator just introduced), S never has a choice about whether or not to perform A. (, )
After considering the objection that this Consequence-style argument plagues divine determinism, Oderberg offers the following response: For all that this version of the famous “consequence argument” as applied to premotionism looks like a bulldozer to metaphysically substantial freedom, appearances are deceptive. It might not be possible here fully to define metaphysically substantial freedom, but let us suppose the following: (i) it involves leeway that is independent of secondary causes; (ii) it involves in addition some set of agential properties F. . .Fn up to but not including independence of divine predetermining premotion. In other words, think of metaphysically substantial freedom as being as substantial as can be short of independence of the sovereign divine providence. The premotionist claims that God moves every free human act in the sense of “freedom” just delineated. Given this sense, the appropriate transfer principle (framed in
to do, then McCann’s position seems to be the following: Beta is invalid because logical relations between propositions do not destroy freedom; only certain sorts of ontological relations can do that. In particular, freedom is destroyed in virtue of deterministic natural or quasi-natural processes of event causation. Since on his own view, this is not how God causes creaturely actions, God does not destroy human freedom. Physical determinism does involve such deterministic processes of event causation, and so does destroy freedom. What is missing from this version of what I am calling the third option in reply to the RCA (if McCann intends to be offering this sort of position at all) is any discussion of why Beta should be reformulated in keeping with McCann’s comments. Beta does not purport to be about ontological relations at all, so the distinction between divine causality on the one hand and event-causal processes on the other hand, while perhaps important to other arguments about incompatibilism, seems out of place in a discussion of transfer principles. Moreover, the claim that logical relations cannot destroy freedom, while obviously relevant to the issue of whether Beta should be accepted, is too weak a claim upon which to rest the denial of Beta. No one, as far as I know, claims that the logical relations themselves destroy freedom – merely that they give us conclusive reason for thinking that freedom is destroyed.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument terms of a necessity operator such as those given above) will be invalid. For no agent will, ex hypothesi, have a choice about whether God moves him freely to perform act A. God’s moving an agent so to act will entail his freely performing A. But every agent will still have a choice about whether freely to perform A. (, )
In these comments, Oderberg first sets out a Consequence-style objection and then suggests rejecting the transfer of powerlessness principle it utilizes. Before turning to evaluation, it is important to note a certain ambiguity in his presentation of the argument. Notably, Oderberg never fully sets out a transfer principle. He says that one might rely upon a principle such as “‘X never has a choice about whether or not P’, or ‘X has no power causally to falsify P’, for some agent X and proposition P” (, ). As stated, it is unclear which agents and propositions this principle governs. Nevertheless, his use of the principle in the argument suggests that Oderberg has Beta (or something relevantly similar) in mind; for this reason, I will assume that his imagined objection is relevantly similar to the RCA. With that much assumed, we can turn to Oderberg’s response. The response considered here consists of an outline of an account of free will together with the claim that the principle in question is invalid. It is true that given the account of freedom delineated, divine determinism does not seem to threaten freedom. This should not come as a surprise, as Oderberg’s account of freedom seems tailor-made to avoid conflict with divine determinism. This does little to blunt the force of Consequence-style worries, however. As noted previously, Beta seems intuitively plausible, and both the CA and the RCA derive force from this plausibility. The fact that a given account of free will entails the falsity of Beta does nothing to show that Beta is false; the entire point of Consequence-style arguments is to call into question such accounts. Even beyond such intuitive plausibility, Beta has been shown to follow from the weakening rule for subjunctive conditionals. It seems, then, that any divine determinist who rejects the RCA by rejecting Beta should provide some explanation for why the intuitive plausibility of Beta is misleading and some argument sufficient to justify rejecting the weakening rule for subjunctive conditionals. Those who take what I am calling the third option in responding to the RCA also should provide some reason for thinking that despite the problems with Beta, physical determinists will still face problems arising from some similar principle. Oderberg cites Pruss’s argument that Beta follows from the weakening rule for subjunctive conditionals, but never addresses how his own
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Assessing the Options in Responding to the RCA
response to Consequence-style arguments handles this problem. Furthermore, he notes that we should not adopt the view that free will is compatible with natural determinism, but offers little in the way of argument for this claim (which is unsurprising, given that the focus of his article is on divine premotion, not natural determinism). Oderberg does, however, have more to say in defense of the rejection of Beta: The theological necessitarian will, no doubt, respond that the premotionist begs the question: for how can God infallibly move an agent freely to perform an act? Surely something has to give? (, )
The idea is, I think, simple enough. The Consequence-style argument that Oderberg discusses begins with Beta as well as premises that he accepts; it concludes that agents in divinely determined worlds lack a choice over whether to perform A. In his response, Oderberg helps himself to an account of freedom consistent with divine determinism, claims that agents with such freedom do have a choice over how they act, and concludes that Beta must be false. Of course, whether a satisfactory account of freedom can be compatible with determinism is exactly what is at issue. Thus, in helping himself to such an account, this author seems to be begging the question against Consequence-style arguments. At best, it seems that we simply have arrived at a dialectical standstill, with one side beginning with Beta and ending with a rejection of human freedom in a divinely determined world, and the other side starting with the affirmation of human freedom in a divinely determined world and ending with a rejection of Beta. Oderberg writes: The premotionist response, apart from repeating the very considerations adduced earlier, is to urge “tu quoque” upon his opponent. For what could it mean for God to “move an agent to act”? Does it mean “move an agent’s body to move in certain ways corresponding to what we call action”? Does it mean “move an agent in such a way that the agent has no freedom over the act he is moved to perform”? The premotionist sees such construals of divine predetermination as either radically incomplete or else as amounting to a flat-out (and question-begging) denial of the premotionist theory. It is no surprise that if the opponent formulates the relevant transfer principle in a way that embodies such conceptions of action, the principle will turn out valid and the agent will lack metaphysically substantial freedom. But the premotionist refuses such formulations in the first place. (, )
It is unclear who precisely qualifies as a theological necessitarian, but the objection seems equally powerful, no matter who voices it.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument Oderberg’s imagined interlocutor apparently begins with some idea of divine predetermination, then somehow builds that into the relevant transfer principle. Thus, Oderberg concludes that it is actually the interlocutor who begs the question. Perhaps this is so in this imagined dialogue, but it is important to notice that Beta, which the RCA utilizes, was not formulated to take into account any particular views on divine action or causality. For this reason, whatever value Oderberg’s response has to the imagined interlocutor, it doesn’t seem sufficient to move the conversation I am interested in – the conversation about the perfectly general principle of Beta – forward. Even if we set aside the worry that Oderberg’s response to Consequence-style arguments begs the question, we will still be left with the apparent dialectical standstill. In such a case, it might seem that Oderberg and other divine determinists have the dialectical advantage. They propose a view, and others later present a Consequence-style argument, whether the RCA or something similar, that purportedly reveals a problem with the original view. Each side begins from a different premise, and consequently each side ends up with a different conclusion. Importantly, the opponents of divine determinism have taken upon their shoulders the heftier dialectic burden, since it is they who assume the task of arguing against the other view. For this reason, it might seem that even if divine determinists don’t definitively win the day, at least they can leave with their heads held high. I think, however, that this picture of the dialectical situation is incomplete. The proponents of the RCA do not begin with some particular account of freedom, but instead start with a general principle that seems highly plausible in its own right. They then add that it follows from an even more general principle – the weakening rule for subjunctive conditionals – which also seems plausible. Later, it is pointed out that divine determinism (if we add the claim that humans are morally responsible) is inconsistent with this principle, and so seems implausible. Simply denying these principles because they are inconsistent with one’s preferred account of divine causation and human freedom does nothing to weaken their plausibility. If my assessment is correct, then Oderberg’s response does not represent a successful example of the third option, which rejects Beta and replaces it with a transfer principle that may be used in the CA but not the RCA. I turn now to the fourth option: admitting that both the CA and the RCA succeed, while denying that such an admission is as costly as it appears. The work of John Martin Fischer, although generally focused upon physical determinism, has paved the way for such a response. Fischer
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Rejecting Moral Responsibility
argues in favor of what he calls semi-compatibilism. According to this view, determinism may well be incompatible with the ability to do otherwise, and thus with the sort of free will that requires such an ability. It is open, then, to the possibility that the CA is sound. Nonetheless, semicompatibilism maintains that physical determinism is compatible with moral responsibility. Thus, although some of our own self-image may need to be adjusted, most or all of our practices remain justified. For example, we may continue to hold that the practices of praise and blame, reward and punishment, are justifiable apart from consequentialist considerations. Can divine determinists adopt a form of semi-compatibilism, claiming that divine determinism does rule out the human ability to do otherwise, while maintaining that humans are responsible for their actions in the strongest ways? Ordinary semi-compatibilism faces difficulties (as do, perhaps, all interesting philosophical positions), but I see no reason to think that the addition of divine determinism would raise any difficulties other than those that all divine determinists face. As far as I can see, then, this should remain a live option for the divine determinist.
Rejecting Moral Responsibility There is one final move open to divine determinists, which I have not yet addressed: They may simply concede that humans are never morally responsible for anything they do. This position, labeled hard divine determinism, has been explored recently by Derk Pereboom (, , ), who contends that it has much to offer. One of the major advances of Pereboom’s work is a development of accounts of reactive attitudes that are still available to the hard divine determinist. According to this author, hard divine determinism “might be thought to threaten the attitudes of blame, guilt, forgiveness, repentance, gratitude, and love, since these attitudes are traditionally held to be core features of the ethical conceptions of the major theistic religions” (Pereboom , ). He argues, however, that this threat can be defused. Consider, for example, his treatment of blame. Clearly, the hard divine determinist must reject the form of blame of the basic desert variety. Nevertheless, several other varieties of blame remain available. George Sher () offers a two-component account of blame. In this view, agent A blames agent B when A judges that B has committed a bad act or
Fischer’s work on this topic is as expansive as it is impressive. See Fischer () for an earlier overview of his position. For a briefer and more recent discussion, see Fischer ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument possesses a bad character, and A wishes that B had not performed that act or possessed that character. If we consider blame along these lines, then the hard divine determinist may accept blaming as legitimate. Even if the hard determinist is correct, people nonetheless perform bad actions, albeit ones for which they are not responsible, and it would at least seem appropriate for us to wish they did not do so. One may object that bad actions cannot be committed in such a world. Pereboom claims that this is implausible, noting, “even if we came to believe that a perpetrator of genocide was not morally responsible in the basic desert sense because of some degenerative brain disease, we could still legitimately maintain that it was extremely bad that he acted as he did” (, ). A further objection – perhaps not properly philosophical, but important nonetheless – is that it is difficult to see how followers of the Abrahamic faiths can adopt such a view, since it seems to require the rejection of the existence of sin. If they are to accept hard divine determinism, they must provide a new analysis of sin, divorced from any notion of moral responsibility in the basic desert sense. I will not set forth a new account of sin, as I think that a number of different accounts on offer could be modified in light of the required change. Two points should be kept in mind, however. First, as noted earlier, there seems to be a strong intuitive pull toward saying that an agent can commit a bad act without being responsible for it. Second, the distinction between intentional evil acts and unintentional ones still carries important weight. Even if the hard divine determinist is correct, we should not begin thinking that all acts of a given type are equal. To see this, consider two individuals, John and Jessica, each of whom kills an innocent man. Assuming hard divine determinism, neither is morally responsible (in the basic desert sense) for the killing. Nevertheless, much might still be said about their agency. Suppose that John is a normally functioning adult, and Jessica, although also a generally normally functioning adult, was drugged on the night of the incident. Suppose this drug caused her mind to become impaired so that she was unable to clearly understand what was happening around her or to deliberate well over her potential courses of action. Although neither individual is a fitting recipient of blame, we can evaluate their acts differently. John’s action may still be understood as worse, since it flows from a worse character (or so one might presume) and was deliberately chosen. Such considerations
One might, of course, contend that deliberation is impossible if determinism is true, or at least that agents must assume that determinism is false in order to deliberate. This is not a special problem for
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Rejecting Moral Responsibility
matter, even in a world without responsibility. They matter because they show a particular kind of moral depravity: a willingness to knowingly take the life of an innocent person. This is not to say that Jessica is less responsible than John; neither is responsible. If we accept hard divine determinism, our relationship to our own actions ought to change. Pereboom argues that we may still show sorrow for our bad deeds and repent of them. I find his account plausible and do not wish to recount it here. But note that God’s relationship to us will also need to change in light of the change in our relationship to our own actions. The theist ought to claim that God treats us in light of our connection to our acts. In turn, hard divine determinists have more work to do to show that their view is a viable option for Abrahamic believers. Consider first divine judgment. According to traditional interpretations of the Abrahamic religions, God has judged the world to be evil and will continue to judge it accordingly. This judgment includes condemnation of individuals, not just their actions. Even if we are to allow for the eventual salvation of all, as universalists affirm, talk of judgment and condemnation does not fit well within the framework of the hard divine determinist. Followers of Abrahamic faiths have two general options. They may be strongly revisionist, arguing that we must do away with divine judgment, or they may be less revisionist and provide a close analogue that can do all (or most) of the work we expect judgment and condemnation to do. I want to investigate which resources the weak revisionists have at their disposal. Is there a close analogue of judgment and condemnation available to them? The first option is to consider judgment and condemnation of acts, rather than of individuals. The hard divine determinist may say that God’s harsh judgment of an individual is nothing but a judgment of the actions the person performed and that God does not pass judgment over agents, since they are all innocent of blame in the basic desert sense. This view would be quite revisionary, and the hard divine determinist may hew a little closer to traditional interpretations of Abrahamic faiths than this.
divine determinists, but is one that – as far as I can see – threatens all determinists equally. For this reason, I will set aside this worry, important as it is, as being outside of the scope of this work, which focuses on worries that arise specifically for divine determinists. Hard divine determinists must deal with this worry, however, even if it threatens other determinists as much as it threatens their own view. Insofar as this requires accepting implausible claims or denying plausible ones, it takes upon itself further costs. For discussion of the relationship between determinism and deliberation, see Kapitan (), Levy (), Pereboom (a), and van Inwagen ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will: The Consequence Argument The divine hard determinist may instead claim that in divine judgment of an individual, God judges the character and history of the agent in question. God may then say that this person is wicked and has done despicable acts. Hard determinists, after all, still allow for moral evaluation, just not moral responsibility. This view is, of course, still revisionary. God may pass judgment on an agent in the same way we may pass judgment on a fig tree: Both may be found wanting when measured against some standard. We may judge a tree to be a bad tree, but it is not reasonable for us to blame the tree for this state. Similarly, God may find an agent wicked, but could not reasonably blame the agent for this state. Of course, if agents are not morally responsible for their evil acts, then they do not deserve punishment (although perhaps it would not be unjust to punish them for consequentialist reasons). What sense, then, can we make of God’s mercy? One may (perhaps) claim that God owes nothing to sinners, and therefore should be praised for reaching out to call them. If this view is taken, then divine mercy turns out to be nothing more than a species of divine generosity. It seems, then, that the practice of revising divine attributes would also require revising away some borders between such attributes. Whether this constitutes an additional cost is an issue I here leave to the reader. Hard divine determinism does face some challenges, but it also possesses resources for responding to a number of problems that divine determinism ordinarily faces. Nevertheless, some problems become even more acute. For example, some worry that divine determinists must admit that God is the author of sin. In a later chapter I will consider this problem further, but notice that if hard divine determinism is true, then God is the only agent morally responsible for human sins. Surely hard divine determinists should deny that God is blameworthy for human sins, but it seems odd that human sins would be more closely connected to God’s agency than to that of humans. Whatever its ultimate value, very few have defended hard divine determinism. Thus, for the remainder of this book, I will assume that divine determinists wish to retain the belief that we are morally responsible for at least some of our actions.
Conclusion Where does this leave the divine determinist? Although there are important differences between physical determinism and divine determinism,
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Conclusion
Consequence-style arguments – which are currently considered among the strongest arguments for incompatibilism – seem to threaten both equally. Although the CA itself does not pose any direct threat to divine determinists, the RCA does present such a danger. Although options in responding to the RCA remain available, there is little reason to think that divine determinists will have an easier time than physical determinists in confronting Consequence-style worries.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.003
Divine Determinism and Free Will Manipulation Arguments
Introduction
In Chapter , I began the task of considering whether arguments attempting to show the incompatibility of free will and physical determinism retain their power when physical determinism is substituted with divine determinism. In particular, I considered the Consequence Argument and concluded that the concerns it embodies pose similar threats to ordinary compatibilism and to divine determinism. Thus, divine determinists may not reject such worries as breezily as they often have. In this chapter, I continue the task of investigating the threat that divine determinism poses to free will by turning to manipulation arguments. Arguments of this sort begin with the intuition that thorough-going manipulation undermines freedom. They continue by arguing that physical determinism and thorough-going manipulation are relevantly similar, and conclude that physical determinism undermines freedom as well. Like the Consequence Argument, manipulation arguments (in at least their most common form) pose no direct threat to divine determinists intent on preserving free will. These arguments do not discuss divine determinism at all, so such determinists need not deny their conclusion. Nevertheless, some who are suspicious of divine determinism claim that the very worries that motivate manipulation arguments can be formulated into an argument that does threaten divine determinists, at least those who are unwilling to let go of their belief in human free will. The purpose of this chapter is to investigate this possibility. I intend to target my investigation just as I did in Chapter . Thus, I will not take up the question of whether manipulation arguments are ultimately successful, since doing so would require me to choose between
For examples of such arguments, see Ginet (), Kane (), Mele (), and Pereboom (, , ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Four-Case Manipulation Arguments
the unhappy options of doing so in an irresponsibly cursory fashion or giving up on an investigation into those challenges of divine determinism that are focused on issues other than free will to focus single-mindedly on those that are. Instead of embracing either of these options, I aim to more carefully limit my investigation by focusing on whether divine determinists are worse off than ordinary compatibilists in responding to manipulation arguments. My conclusions in this chapter are markedly different than those in Chapter . In the previous chapter, I concluded that Consequences-style arguments are equally powerful whether they focus on divine or physical determinism. When it comes to manipulation arguments, matters are more complicated. Although some similarities arise, I will argue that some of the replies to manipulation arguments available to divine determinists are unavailable to ordinary compatibilists; likewise, some of the argumentative resources available to ordinary compatibilists cannot be appropriated by divine determinists. Moreover, divine determinists face a manipulation worry that ordinary compatibilists do not face: a direct appeal to intuition. I begin by briefly introducing manipulation arguments, concentrating on an especially powerful version introduced by Derk Pereboom. I then turn to the question of whether the worries this argument embodies can be used to motivate a form of the argument that considers divine rather than physical determinism. Unsurprisingly, I suggest that such a modified form can be constructed and deserves careful attention. Next, I consider whether this new argument, which I will call the Revised Manipulation Argument, is as powerful as the well-known argument that inspired it. I also investigate the legitimacy of the objection that it is intuitively obvious that the actions of a divine manipulator undermine human moral responsibility.
Four-Case Manipulation Arguments
Manipulation arguments come in many forms, but in one way or another share certain features and a single goal. The goal is one they share with other arguments for incompatibilism: to establish (to some degree or
My view of the options available betrays an underlying claim that I am happy to embrace: Either the appropriate response to manipulation arguments is not obvious, or it is obvious but not obviously correct. Thus, the full assessment of these arguments should be done carefully, and we should not expect such an assessment to be quick. Although I will not defend this view here, some of my reasons for believing it will become clear throughout this chapter. Additionally, the mere quantity and quality of recent work on these arguments lend support – indirect to be sure, but support nonetheless – to my contention.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
other) that determinism is incompatible with free will and/or moral responsibility. Manipulation arguments are further unified by two common features: () reliance on the conviction that agents who are completely manipulated into acting as they do are not morally responsible for their actions and () reliance on the conviction that for one reason or another the conviction described in () works against the compatibility of determinism on the one hand and moral responsibility on the other hand. The term “manipulator” deserves attention here. As used in these arguments, it refers to an agent who intentionally determines (whether physically or divinely) the actions of another agent. It need not imply any moral judgment of either the manipulator’s actions or of the moral responsibility of the manipulator. It must be granted that the term does have a negative connotation in ordinary use. One might worry that our intuitions about manipulation cases will be illicitly colored by the ordinary negative connotations of the term used to categorize them. Perhaps this is so, but at this point in the philosophical debate, the word has taken on a life of its own in its more specific technical meaning, so I doubt that philosophers need to worry about their own intuitions being colored in this way. Finally, some may worry that the term “manipulation” implies a sort of independence between the manipulated agent and the manipulator – an independence that requires the manipulator to intervene into a state of affairs that exists independently of the manipulator’s will. Thus, even if divine determinism is true, it would not be appropriate to think of God as a manipulator. I think that the term “manipulation” is a rich one that includes a number of things left out of the technical use of this word.
Concerning the moral judgment of a manipulator’s action, we might imagine a case where the manipulator has sufficient reasons for determining the actions of another. We might even imagine a case where the manipulated agent has granted the manipulator the right to do so. If we are inclined to think that such actions are always wrong, perhaps based upon the Kantian idea that we must always treat others as ends, we might at least say that the term “manipulator” is not meant to imply that the actions of such an agent are wrong. Similarly, a pacifist might say that all killing is wrong, while admitting that the term “killing” itself includes no moral judgment as its content in the way that, say, “cowardly” or “gluttonous” do. As for the second claim, that manipulators need not be morally responsible for their actions, we might look to Alfred Mele’s (, n. ) case of Diana, who is insane and so not morally responsible for what she does. There may be other reasons to worry more generally about philosophers’ intuitions, but none of these seem specific to cases of divine determinism or free will. See, for example, Alfred Mele (), who worries that focusing on some issue long enough might lead to intuition blindness. See also recent discussions of the value of philosophical intuitions more generally (Andow ; Liao ; Vaesen, Peterson, and Van Bezooijen ). See Rogers (). Of course, some think this is entirely appropriate. Jerry Walls calls divine determinism “the most metaphysically majestic account of manipulation ever devised” (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Four-Case Manipulation Arguments
It seems to involve a moral condemnation of the manipulator, a presupposition that the manipulator is acting against or at least not in keeping with the goals and desires of the agent who is manipulated, and, perhaps, a sort of independence between the agent who manipulates and the one who is manipulated. Perhaps “controller” would be a better term to adopt, at least in some cases, as Katherin Rogers () suggests. However, this term has its own connotations that limit it, not least of which is that for many it will call to mind a quasi-mechanical process by means of which control is established and exercised, and this might distort our understanding of some accounts of divine causality. Since “manipulator” (and other related terms) have already been established as terms of art in these discussions, and possible replacements bring their own difficulties with them, I will continue speaking in terms of manipulation. I invite those worried that this phraseology unduly distorts our thinking to rephrase relevant instances in terms of control or some other term of their choice. Recently a number of philosophers have put forward different versions of manipulation arguments, but none has received more attention than that of Derk Pereboom. Pereboom’s Four-Case Argument (FCA) invites the reader to consider a first case in which a particular intuition is elicited: that the agent in question is not morally responsible for acting. Each of the next three cases is slightly different than the one that precedes it. In each case, the reader is supposed to judge that although the cases differ, they do not differ in any way that is relevant to attributions of moral responsibility. Thus, the reader is led, one case at a time, to continue judging the agents in question as lacking responsibility for their actions. In the fourth and final case, the reader considers an agent in an ordinary deterministic world and is once again invited to judge that the agent is not relevantly different than the agents in the previous cases, and so is not morally responsible for anything. Although Pereboom’s cases are long, it is worth considering the first three in full. We begin with his first case: Case : A team of neuroscientists has the ability to manipulate Plum’s neural states at any time by radio-like technology. In this particular case, they do so by pressing a button just before he begins to reason about his situation, which they know will produce in him a neural state that realizes a strongly egoistic reasoning process, which the neuroscientists know will deterministically result in his decision to kill White. Plum would not have killed White had the
Pereboom (, , , b, )has developed this argument over the past few decades.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
neuroscientists not intervened, since his reasoning would then not have been sufficiently egoistic to produce this decision. But at the same time, Plum’s effective first-order desire to kill White conforms to his second-order desires. In addition, his process of deliberation from which the decision results is reasons-responsive; in particular, this type of process would have resulted in Plum’s refraining from deciding to kill White in certain situations in which his reasons were different. His reasoning is consistent with his character because it is frequently egoistic and sometimes strongly so. Still, it is not in general exclusively egoistic, because he sometimes successfully regulates his behavior by moral reasons, especially when the egoistic reasons are relatively weak. Plum is also not constrained to act as he does, for he does not act because of an irresistible desire – the neuroscientists do not induce a desire of this sort (Pereboom , –).
Pereboom notes that most will have the strong intuition that Plum is not responsible for his actions. Plum simply does not deserve blame or punishment for his actions, since they originated in a source that was outside of his control. Pereboom then asks us to consider a second case: Case : Plum is just like an ordinary human being, except that a team of neuroscientists programmed him at the beginning of his life so that his reasoning is often but not always egoistic (as in Case ), and at times strongly so, with the intended consequence that in his current circumstances he is causally determined to engage in the egoistic reasons-responsive process of deliberation and to have the set of first- and second-order desires that result in his decision to kill White. Plum has the general ability to regulate his actions by moral reasons, but in the circumstances, due to the strongly egoistic nature of his deliberative reasoning, he is causally determined to make his decision to kill. Yet he does not decide as he does because of an irresistible desire. The neural realization of his reasoning process and of his decision is exactly the same as it is in Case (although their causal histories are different) (Pereboom , ).
Pereboom maintains that there is no principled reason to admit that Plum lacks moral responsibility in Case , but is responsible in Case . Thus, he maintains, our initial judgment about Plum should transfer from the former to the latter.
One might wonder whether the agential requirements discussed may be satisfied in any determined worlds. In these cases, we might just take it as granted for the sake of argument that they can be. If they cannot, then it seems that incompatibilists will have little need for manipulation arguments, because determined agents could not even meet the conditions suggested by compatibilists.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Four-Case Manipulation Arguments
Pereboom then presents us with a third case: Case : Plum is an ordinary human being, except that the training practices of his community casually determined the nature of his deliberative reasoning processes so that they are frequently but not exclusively rationally egoistic (the resulting nature of his deliberative reasoning processes are exactly as they are in Cases and ). This training was completed before he developed the ability to prevent or alter these practices. Due to the aspect of his character produced by this training, in his present circumstances he is causally determined to engage in the strongly egoistic reasons-responsive process of deliberation and to have the first- and second-order desires that issue in his decision to kill White. While Plum does have the general ability to regulate his behavior by moral reasons, in virtue of this aspect of his character and his circumstances he is causally determined to make his immoral decision, although he does not decide as he does due to an irresistible desire. The neural realization of his deliberative reasoning process and of the decision is just as it is in Cases and (Pereboom , ).
Pereboom claims that Case is relevantly like Case , and so our judgment about the former should be duplicated for the latter. Plum, it seems, is not morally responsible for his action. Pereboom goes on to consider a fourth case in which Plum is an ordinary human in a deterministic world. He concludes by claiming that this case is relevantly similar to the three that preceded it, so we should conclude that Plum is not morally responsible in Case either. The argument may wrap up in either of two ways. First, we might simply conclude that since there is no relevant difference between Plum in Case and ordinary humans in a world governed by physical determinism, no humans in such a world are morally responsible for anything. Such an argument makes no attempt to highlight the responsibility-undermining feature of the four cases, but instead merely establishes that there is one. A second way to wrap up the argument, and the one that Pereboom prefers, transforms it into an argument to the best explanation (Pereboom , –, n. ). Taking this path involves casting about for the best explanation for Plum’s lack of responsibility across the four cases. Pereboom concludes, “The salient factor that can plausibly explain why Plum is not responsible in all of the cases is that in each he is causally determined by factors beyond his control to decide as he does” (, ). Manipulation arguments in general and the FCA in particular clearly threaten any version of divine determinism that adopts physical
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
determinism – this much is clear. It is more difficult to determine whether the FCA would threaten other versions of divine determinism. If we cast the argument as an inference to the best explanation, then it very well may pose such a threat, as long as we interpret causal determinism broadly enough to include the determining forces of a transcendent creative God. Thus, although the FCA is not meant to target divine determinism, it may do so incidentally. Instead of exploring whether such versions of divine determinism are incidentally threatened by Pereboom’s argument, I suggest a revised version that directly sets its sights on divine determinism; I will call this argument the Revised Four-Case Argument (RFCA). The RFCA begins with Cases – as detailed previously. We then add the following case: Case *: Everything that happens in our universe is determined by God. Plum is an ordinary human being, raised in normal circumstances, and again his reasoning processes are frequently but not exclusively egoistic, and sometimes strongly so (as in Cases –). His decision to kill White issues from his strongly egoistic but reasons-responsive process of deliberation, and he has the specified first- and second-order desires. The neural realization of Plum’s reasoning process and decision is exactly as it is in Cases –: He has the general ability to grasp, apply, and regulate his actions by moral reasons, and it is not because of an irresistible desire that he decides to kill.
Hard divine determinists, of course, will not be threatened by the argument, since they embrace its conclusion. As mentioned in Chapter , I will not further address this view. A single line in the Introduction to Pereboom’s book suggests that he interprets his manipulation argument as doing so. He writes: “An action will also be causally determined by factors beyond the agent’s control if its occurrence is ensured by a causal process that originates in God’s timeless willing and ends in the occurrence of the action” (Pereboom , ). Even if we accept Pereboom’s claim, however, some versions of divine determinism may escape his claim. McCann (), for example, thinks it is imperative to see that divine determinism does not involve a causal process at all, and thus escapes ordinary problems of determinism. The reader may wonder if the inference to the best-explanation version of the FCA would represent a more significant challenge to divine determinism than the revised version of the argument that I pursue. I doubt that it would, because it would be difficult to determine whether the best explanation for Plum’s lack of responsibility in Cases – is to be found in determinism taken broadly (so as to include divine determinism) or in a more limited fashion (limited to natural determinism, for example). To justify the former response, one would need to conclude that divine determinism and natural determinism are relevantly similar, which is ultimately the same difficulty that, I will argue, faces the revised argument I explore. Thus, I think that the revised argument and the inference to the best-explanation reading of the standard FCA have similar dialectical force, but that the former more clearly reveals the premises needed to reach a conclusion about divine determinism. This case is a slightly modified version of Pereboom’s Case (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Responding to the RFCA
The RFCA concludes by claiming that however large the differences between Case * and Cases –, none of the differences is relevant for evaluating Plum’s moral responsibility for killing White. Since Plum is clearly off the hook in Case , and this judgment should transfer across Cases – to Case *, he is off the hook in Case * as well. Finally, since Plum in Case * is relevantly similar to ordinary humans in a divinely determined world, if Plum is not guilty for killing White in Case *, then no humans are morally responsible within a divinely determined world. Thus, the RFCA concludes that divine determinism is incompatible with human moral responsibility.
Responding to the RFCA What should we make of the RFCA? Is it any stronger or weaker than the FCA? To answer this question, we must briefly explore possible responses to the FCA and the RFCA, noting those that are common to both and those that have force only against one or the other. Broadly speaking, there are two categories of replies to the FCA: soft-line replies and hard-line replies. Defenders of the soft-line reply claim that some relevant difference exists between the manipulation cases used to motivate manipulation arguments and the cases of ordinary agents in deterministic worlds. Thus, they claim that although manipulation does undermine moral responsibility, physical determinism does not. As they have it, Plum in Case is morally responsible, while he lacks responsibility in Cases –. Defenders of the hard-line reply admit that agents in manipulation cases, at least in those that are carefully formulated, are relevantly similar to agents in deterministic worlds and suggest that we should conclude that manipulated agents are indeed morally responsible for their actions. As they have it, Plum is morally responsible for killing White in each of Cases –. We might distinguish hard- and soft-line replies to the RFCA along similar lines. A hard-line reply will argue that we should admit that Plum is responsible not only in worlds governed by divine determinism (as in
The RFCA thus eschews any attempt to establish the best explanation for Plum’s lack of responsibility across the four cases. Although establishing the responsibility-undermining feature of the four cases would be instructive, doing so is unnecessary for reaching the conclusion. These phrases were coined by Michael McKenna (, n. ), although he loosely based them on comments made by Robert Kane (). It is possible to adopt a hard-line reply to some arguments and a soft-line reply to others. Indeed, it is possible to suppose that Plum is responsible in more than one of Cases –, just not in all. For ease of discussion, however, I will treat hard- and soft-line replies taken neat.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
Case *), but also in other cases with covert manipulation (Cases –). The soft-line reply will maintain that some difference between Cases – on the one hand and Case * on the other hand explains why Plum is not morally responsible for killing White in the former but is in the latter. Both ordinary compatibilists and divine determinists must choose between hard- and soft-line replies to manipulation arguments, so it might be tempting to conclude that they are in the same dialectical position in regard to such arguments. Matters are more complicated than they first appear, however. The hard-line replies to the FCA and the RFCA are relevantly similar. Both seek to avoid an unwelcome conclusion by asserting (in the face of intuitions to the contrary) that Plum is responsible in Case . Once this move is made, both the FCA and the RFCA fall apart. Thus, we are right to conclude that for hard-liners, these two arguments have identical force. Soft-liners should not reach the same conclusion, since most soft-line replies to the FCA (at least those with a hope of success against carefully nuanced cases) will be inapplicable to the RFCA, and most soft-line replies to the RFCA will be inapplicable to the FCA. The problem, in short, is that soft-line replies to the FCA will need to make use of one of two strategies. First, the soft-liner could maintain that some subset of Cases – fails to paint Plum in such a way that he instantiates what McKenna () calls the Compatibilist-friendly Agential Structure (CAS), even though the CAS is not itself incompatible with complete intentional manipulation by another agent. So, for example, one might argue that in one or more of Cases –, Plum cannot be reasons-responsive due to the mechanism by which he is manipulated, even though reasonsresponsiveness is not itself incompatible with determinism or manipulation. Second, the soft-liner could maintain that Plum (in Cases –) fails to satisfy the CAS because some necessary condition for the CAS is incompatible with manipulation. The first strategy is at best a way to push off certain formulations of the manipulation argument, but it cannot hold out any promise of final success. If some subset of Cases – fails to
We might be tempted to think that only certain sorts of manipulation – morally odious ones – undermine moral responsibility, and since divine determinism does not involve this sort, then it does not undermine responsibility. Alm () is the only philosopher who develops a view similar to this thought. He develops an account according to which manipulation (unlike physical determinism) is responsibility-undermining because we have a cause to resent it. Unfortunately for the divine determinist, Alm notes that manipulation that deserves gratitude instead of resentment also precludes moral responsibility, and for the very same reasons: It renders agents passive in relation to their own actions. For discussion, see Barnes (), Haji and Cuypers (), Mickelson (), and Yaffe ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Responding to the RFCA
paint Plum in the appropriate way, but not merely because of the presence of manipulation, then in principle there should be no reason that the case(s) could not be modified to satisfy all of the soft-liner’s demands. At most, then, this sort of soft-line reply can push off the inevitable, and so must eventually yield to the second strategy of the soft-line reply or be supplemented by a hard-line reply. Thus, a thorough-going soft-line reply to carefully formulated manipulation cases will eventually need to apply the second strategy. The second strategy, it is clear, will be of little use to those interested in rebutting the RFCA, since divine determinism posits manipulation. Promising soft-line replies to the RFCA will appeal to the radical differences between divine causation and anything that neuroscientists (even ones populating wild science-fiction stories) can offer. Such a line of response will have little to offer those searching for a reply to the FCA. Thus, soft-liners should conclude that these two arguments, whatever their ultimate value, should be treated separately. If they were to have equal argumentative force, this would be merely coincidental. I turn now to a soft-line reply to the RFCA. Recall that the RFCA claims there is no relevant difference between the manipulation performed in Cases – and that present in Case *. I suspect that divine determinists, especially those adamant in proclaiming the radical differences between human and divine causality, will maintain that many relevant differences distinguish the two cases. They may point, for example, to some subset of the following claims: (i) (ii) (iii)
God has authority over human lives, and the manipulators in Cases – do not. God’s actions, including those acts in which human choices are determined, do not involve any event-causal processes. Manipulators and their actions in Cases – are beings in the universe that should be thought of as additional to Plum and his action, whereas God should not be thought of as one more being that might be added to Plum and his action.
See McKenna () for similar remarks. Soft-line replies that merely pave the way for hard-line ones need not be ultimately pointless. The soft-line reply may show that modifications need to be made to manipulation cases before they should be accepted as relevantly similar to cases of ordinary agents in deterministic worlds. If these revised cases differ substantially from those initially used in manipulation arguments, they may fail to produce the same intuition concerning Plum’s lack of responsibility. In such a case, the soft-line reply will have paved the way for a more palatable hard-line reply than was originally possible. For an explication of this claim, see Sokolowski (). For an example of its use in understanding human freedom in light of divine causality, see Burrell ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
(iv)
God’s action is required for human action of any kind, whereas the actions of manipulators are not. Manipulators in Cases – operate through ordinary causation exercised by creatures, which some call secondary causation, whereas God operates through primary causation. (This claim may be joined to claims of God’s radical transcendence.) God is responsible for both the human action performed and its modal status and, therefore, is able to determine humans’ free actions while also causing them to be contingent. Manipulation in Cases – renders Plum’s action necessary.
(v)
(vi)
I think that the divine determinist should not point to these or any other differences between Cases – and Case * in an effort to demonstrate that some relevant difference obviates the force of the RFCA. To do so, it seems to me, would be to accept a larger dialectical burden than is necessary in responding to this argument. Since the RFCA is an objection to the compatibility of divine determinism and moral responsibility, just as the FCA is an objection to ordinary compatibilism, it takes upon itself a hefty dialectical burden. This task entails providing reasons for thinking that divine determinism imperils human moral responsibility. If it cannot do this, then it fails. In responding to the RFCA, divine determinists need not demonstrate that divine determinism is compatible with human freedom and that it is relevantly different than Cases –. Instead, they must merely show that the RFCA fails to accomplish its task. To accomplish this task, divine determinists need not show that there is a relevant difference between Cases – and Case *. Instead, they may rest content having shown that the RFCA fails to provide good reason for thinking that no such difference exists. The point here is not that one may reply to the RFCA while remaining finally agnostic about the existence of a relevant difference. Divine determinist responders must indeed deny such a difference: otherwise, they should rationally become agnostics (at least) about the compatibility of their favored brand of determinism and moral responsibility. Nevertheless, because of the distribution of dialectical burden, they need not feel compelled to demonstrate its existence. The question divine determinists should focus on (at least if they are focused merely on evaluating the RFCA) is not whether there is a relevant
See Oderberg (), Thomas Aquinas ( q. , ad ), and Trabbic (). My comments here parallel those offered by Michael McKenna () regarding the dialectical situation concerning Pereboom’s original argument.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Responding to the RFCA
difference between the cases, but whether it is evident that there is no difference. To accomplish this, they need not defend the claim that some subset of (i)–(vi) presents such differences. Instead, they may merely claim that consideration of some subset of (i)–(vi) (or other considerations entirely) provides reason to be skeptical about whether Cases – are relevantly similar to Case * after all. This move may be accomplished in one of two ways. First, they might note that some subset of (i)–(vi) provides a plausible candidate for a relevant difference, and no part of the RFCA tells against its plausibility. Thus, until the RFCA is supplemented, it can be safely and rationally ignored. Second, they might note that some subset of (i)–(vi) provides us with reasons to be skeptical about our ability to discover any relevant difference between Cases – and Case * even if it exists. The idea here is that some subset of (i)–(vi), if true, makes it plausible that God’s causality is sufficiently different from what we are accustomed to that we would be unlikely to spot relevant differences if there were any. In this case, too, they should be unmoved by the RFCA’s claim that they are relevantly similar, and since the RFCA takes upon itself the heavier dialectical burden, the divine determinist comes out unscathed. I tend to think that the latter reply is the stronger of the two, because it does not require arguing that some subset of (i)–(vi) is a plausible relevant difference between Cases – and Case *, which may prove difficult. These strategies will be equally valuable in rebutting any manipulation argument that seeks to transfer intuitions from cases with human manipulators to those with divine ones. Setting aside the question of dialectical burden, we might wonder whether a divine determinist would be satisfied with the strategy noted previously. I think this depends upon what we mean by “satisfied” and, perhaps more importantly, the sensibilities of the determinist in question. We might take a response to an argument to be “satisfying” if it gives us good reason to think that the argument in question did not live up to its dialectic burdens. Of course, we might also hope for something more robust. We might wish for a response to an argument to both give us good reasons for thinking one of the premises of the original argument is false and provide an explanation of why it is false. This difference seems on display in possible responses to the problem of evil. Consider an argument from evil that uses as a premise the claim that there are no reasons that justify God in allowing evil in the world. The skeptical theist – who denies that we should be confident in our ability to discern what reasons God might have for allowing evil – argues that we
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
have reason to refrain from affirming this premise. In contrast, someone with a robust theodicy – someone who points to specific reasons God has for allowing evil in the world – argues that we have reason to think the relevant premise is false and provides an explanation for why it is false. Just as a philosopher might think that the skeptical theist’s argument (if it is a good one) is sufficient for responding to the problem of evil, yet unsatisfying, so too might a philosopher think that the modest strategy I outline in this section (if it is a good one) is sufficient for responding to this manipulation argument, yet unsatisfying. Before moving on to a judgment concerning the merits of the RFCA, I wish to consider one objection, which might take the following form: It may seem convenient to retreat to mild claims that we might know of some relevant differences between the cases (the first strategy) or that we should not be confident there are none, even if we haven’t found them (the second strategy), but this will not do the work you need it to do. The RFCA rests upon the perhaps obvious claim that Cases – and Case * are relevantly similar. In each case, Plum is caused to act in a particular way by a manipulator, and given the manipulator’s action, Plum is determined to so act. Since Plum has no control over how and/or whether the manipulator acts, he cannot possibly be responsible for killing White in Cases –; just as clearly, he cannot possibly be responsible for it in Case *.
This objection misses the role of manipulation arguments. If we are permitted to assume that agents are morally responsible only if they have control over the factors that determined their later actions, then there is no need to consider manipulation cases in the first place. Manipulation arguments, at most, seek to show via an argument to the best explanation that such a claim is true, but they cannot begin with it. Indeed, the reasoning presented here seems to reduce to Consequence-style arguments, which were discussed in Chapter . Although opponents of divine determinism have been quick to claim that this form of manipulation undermines moral responsibility, we should not be too hasty in judging that this is so. Although the FCA (and, indeed, any manipulation argument that relies on intuitions about human manipulators) can be revised to target divine determinism, their strength is reduced. Divine determinists may appropriate the hard-line reply used by ordinary compatibilists, but they also have a promising soft-line reply at their disposal. Although they must forfeit any use of soft-line replies offered by ordinary compatibilists, this should not cause them much distress, both because they have a soft-line reply of their own, and because
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
A One-Case Argument against Divine Determinism?
many ordinary compatibilists seem to be unconvinced of the efficacy of such replies.
A One-Case Argument against Divine Determinism? In the preceding discussion, I considered a way to modify Pereboom’s FCA in such a way that it threatens divine compatibilism. I now turn to a more direct attack that might be made against it. Katherin Rogers has suggested a manipulation argument against ordinary compatibilism that invokes intuitions about divine beings. She is not attempting to show that divine determinism is incompatible with moral responsibility, but her argument seems especially useful in considering manipulation worries aimed at divine determinism. Rogers suggests that our intuitions about manipulation cases – like the first three cases of the argument considered earlier in this chapter – might fail to be sufficiently robust because of certain elements of the stories themselves. She claims that the manipulators in such cases are “thin,” meaning that we lack a robust understanding of the people doing the manipulating. Indeed, we recognize that there is nothing there to understand, since they consist solely of a few sentences meant to pump up our intuitions. The problem, Rogers maintains, is that these cardboard characters are insufficient to bring about the very intuitions they were created to inspire, or, if they do inspire such intuitions, their value may be suspect. She also claims that they are bizarre and evanescent: They are bizarre figures, so it can be argued that their weirdness is doing much of the heavy lifting in eliciting the looked-for intuition. And they are evanescent. We do not really have to worry about them since they exist only as fanciful hypotheses, safely confined to the pages of philosophical literature. (Rogers , )
What we need, Rogers suggests, is to replace the neuroscientists of Cases – and the social engineers of Case with God. The image we have of such a being, whether we are theists or not, is much richer than anything we might have of those who manipulate Plum in the original cases. Moreover, God is much easier to take seriously, since, even if we are not theists, we see so many others who are. Even nontheists can see how stories about God
This judgment about the current status of the debate is supported by Tognazzini (, ), who notes that the hard-line reply is the most common one. Alfred Mele () has also put forward a manipulation argument in which a divine being – Diana – manipulates atoms in the creation of a zygote, all in an attempt to bring about an agent who will act according to her plans. Although Mele’s argument has been far more influential than that of Rogers, I think the latter’s is more useful in the present context.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
impact our world in ways that neither manipulating neuroscientists nor telling stories about them could hope to achieve. This difference may not change the substance of the arguments concerning them, but it will make the relevant intuitions both richer and more reliable – or so Rogers suggests. Rogers recommends that we replace Plum-style cases with the following (I shall call it Case *): Say that murder is wrong and that God has commanded you not to commit murder on pain of punishment. (Let us leave Hell out of the hypothesis, since the specter of an infinite punishment for a finite crime might affect our intuitions.) Suppose that the punishment involved is a long period of suffering, equivalent to a life sentence without parole – the sort of punishment which is standard in our society today. And now suppose that God directly causes in you a choice to commit murder, which [divinely determines] your subsequently committing murder. Do you deserve that God should punish you? Surely not! But why not? Because He made you choose to murder! Could our intuition be driven by repugnance at God himself punishing you for what He made you do? God is, by hypothesis, perfectly just. If you deserve to be punished, He ought to punish you. But allow that the punishment will not be done by God, but by human agents. Still, God made you choose, and so it seems that you do not deserve to be punished. (, –)
With this case in hand, Rogers thinks that we will have a reliable intuition that divine manipulation undermines moral responsibility. She then moves to the claim that divine determinism is relevantly similar to physical determinism and finally concludes that physical determinism undermines moral responsibility. Thus, compatibilism is false. Rogers contends that this argument is an improvement over ordinary manipulation arguments. Whether this claim is true or not, it surely has some force. For our purposes, we may consider the premises of the argument to have the following structure: . If God determined a created agent A to perform action x, then A is not morally responsible for x. . Divine determinism and physical determinism are relevantly similar.
I have removed “necessitates” from the original and replaced it with “divinely determines,” to clarify how I understand the relationship between God and the human action in this case, since various sorts of necessity are discussed in this work. As far as I can tell, this does no damage to Rogers’s original argument. It is difficult to be sure, however, because we do not get a strict definition of her view of necessity. See Rogers (, ). See Rogers (, ) for a formulation of the argument in terms of necessity.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
A One-Case Argument against Divine Determinism?
Each premise may be called into question, and the argument depends on the truth of both. Whatever the merits of this argument may be, it clearly has more force against divine determinists than against ordinary compatibilists. Ordinary compatibilists have two options to avoid the conclusion: They may deny the force of either the first or the second premise. If we wish to modify Rogers’s argument to apply solely to divine compatibilists, then we will be left with a one-case argument. The entire argument would consist of Case * together with the conclusion that divine determinism is incompatible with human moral responsibility. A one-case manipulation argument might seem to be illicit in some sense. Whatever the real value of such an argument, it clearly has more force than Rogers’s original version. After all, neither of the premises of the original argument is beyond question. Thus, the argument may fail in either of two ways. This one-case argument has only one way to fail, and, significantly, it is identical to one of the ways in which Rogers’s original argument may fail. In short, we should conclude that this one-case argument has more force than Rogers’s original argument since the success of Rogers’s argument requires the success of the one-case argument, whereas the success of the one-case argument does not rely on the success of Rogers’s argument. It must be granted that a direct appeal to intuition – the essence of this one-case argument – does not possess the appearance of tight reasoning that arguments built upon such intuitions enjoy. Nevertheless, we must not let this color our judgment. Direct appeals to intuition are less likely to fail than arguments built upon those same intuitions. So far, then, it seems that the direct appeal to intuition in Case * is powerful for two reasons: The first reason is that the intuition about Case * seems more reliable than intuitions about Cases –, insofar as the former provides a more richly developed case, appealing as it does to God, about whom we have elaborately detailed traditions, rather than appealing to thinly developed neuroscientists. Of course, on this issue, philosophers will differ. While Rogers claims that her intuition is stronger if we replace human manipulators with God, others will find no intuition at all about whether this divine manipulation undermines human freedom. For such people, this new manipulation argument will not even get off of the ground. The second reason is that the direct appeal to intuition is more independent than ordinary manipulation cases, since intuitions about Case * need not be conjoined with the truth of another controversial claim to reach the desired conclusion. This direct appeal to intuition seems more threatening than manipulation cases against ordinary compatibilists for another reason. If ordinary
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
manipulation arguments succeed, then compatibilists are forced to admit that physical determinism is like manipulation. Divine determinism seems to be a sort of manipulation, which leaves divine compatibilists in a worse position than ordinary compatibilists, or so it seems to some incompatibilists. William Hasker, for example, writes: If I were to discover that some action of mine, which I thought was freely chosen, was in fact the result of some random combination of physical circumstances, I should probably be somewhat upset. But I should be a great deal more upset were I to discover that my “free” action was the result of the deliberate control of me by some other person, someone who caused me so to act while leaving me unaware of that fact. Nor is it clear to me . . . that things are a great deal better if it is God rather than a human being who is doing the controlling. (, )
Responding to the Direct Appeal to Intuition How worried should divine determinists be about the one-case argument? Case * may fail to reliably elicit the relevant intuition in all, but it very commonly succeeds in doing so. The goal for divine determinists should be to show that this initial intuition is unreliable. In other words, they should be intent on making it clear to Hasker that whatever may be said about human manipulators, he shouldn’t be so upset about divine determinism. The best strategy for divine determinists is three-pronged. First, they should work to show that some of the force of the initial intuition, and indeed of intuitions concerning more standard manipulation cases, arises from irrelevant features of the case. Second, they should work to weaken the initial intuition by further detailing the case to closely align with their preferred views of divine causality and the relationship between creation and the creator. Finally, with the “no-responsibility” intuition suitably weakened, they should call into question the reliability of human intuitions about the consequences of divine activity. Let’s begin with the first prong. In developing this part of their response, divine determinists have a valuable resource in the work done in developing hard-line responses to ordinary manipulation arguments. The first thing to notice about agents in divinely determined worlds is that we must assume that they meet all compatibilist-friendly conditions of moral responsibility. Thus, we should consider them to have rich and
I suggest divine determinists pay especially close attention to work of Michael McKenna (, a, , , ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Responding to the Direct Appeal to Intuition
thoughtful mental lives. They deliberate about how to act, and their deliberations include moral reasons. They believe that their deliberations and choices lead to their actions, and they are right in believing this. Rogers notes that we must be careful not to paint manipulators as thin and bizarre, but we must be careful in how we think of the manipulated agents as well, for it is tempting to imagine the agents of manipulation cases as life-less sprites. In this spirit, then, let’s consider a more fully fleshed-out case with a more ordinary agent. Case **: Jonathan lives in a divinely determined world, and is relevantly similar to fully developed humans living in nondetermined worlds. He matured both intellectually and morally in the normal way. For example, he learned as a child that some acts are unjust, or dishonest, or cowardly, and others are just, or honest, or courageous. Jonathan was encouraged to act in virtuous ways, and often did so, although he sometimes failed to live up to these standards – standards that were initially set by others, but came to be self-imposed. When he failed to live up to his obligations, he felt guilty and attempted to make up for what he had done. Jonathan rightly believes that his actions are divinely determined, but also believes that this fact does not alter his own responsibility for his actions. After a particularly long day at work, Jonathan is eager to get home when his wife calls and asks him to stop at the bank on his evening commute. Jonathan rudely agrees before hanging up on her. This action is not typical for Jonathan, but it is not widely out of character either. Later than night, Jonathan apologizes, accepting responsibility for his rudeness and admitting that he deserved the blame his wife leveled at him.
What should we make of Jonathan? I suspect that although some might have the intuition that he is not morally responsible for his action – that he need not have accepted responsibility for what he did – this intuition will be both less widely shared and generally of a weaker intensity than the intuition produced from Case *. Why this difference? I suggest it is due to a combination of several points. First, Jonathan is described in ways that makes clear his similarity to ordinary humans and pushes out ideas that manipulated agents are little more than robots obeying orders. Second, Case ** concerns a third party, while Case * concerned ourselves; our own desire for exculpation (for murder, no less, which of course we wouldn’t have committed if not for divine determinism!) may be pushing us to absolve ourselves in Case *. Third, Jonathan understands the causal underpinnings of his actions and accepts them; nothing in Case * suggests a parallel. Fourth, the action performed by Jonathan is much more ordinary than the murder in Case *, and also described in such a way
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
that relates to his own agency. Fifth and finally, the wrong action in Case * is far more serious than that of Case **, which likely leads us to err on the side of innocence in the former. The first four differences should lead us to think that our intuitions about Jonathan are more reliable than those about Case *. The first feature merely ensures that we call to mind the strong ways in which Jonathan is like ordinary agents. Since manipulation cases may illicitly prime our intuitions by falsely suggesting that manipulated agents are mindless machines, this difference should make our intuitions more closely track the details of the case. The second feature ensures that our intuitions are not guided by either mere self-interest or a conviction that divine determinism works in such a way that agents act against their characters and considered views. The third feature, that Jonathan believes in divine determinism and accepts responsibility anyway, should not be relevant to assessing Case ** unless we think that believing one is responsible (while simultaneously understanding all relevant etiological features of one’s action) is a necessary condition for being responsible. Although some compatibilist accounts of moral responsibility suggest something along these lines, this claim is unlikely to be adopted by incompatibilists about moral responsibility, so it will be of little use to those arguing against divine determinism. The absence of such a belief in Case * might, however, improperly color our intuitions. Keeping the agent in Case * in the dark about divine determinism may have the effect of convincing us on some level that the agent is not really acting at all – that the only true cause of action is God. By pointing to Jonathan’s belief, we are forced to recognize that it is possible for agents to believe themselves to be responsible in spite of their belief in divine determinism, which forces us to at least take seriously the presupposition that they are agents. Moreover, whether divine determinism is true or not, there are people like Jonathan who both believe in divine determinism and believe that they are responsible for their actions. Thus, such a detail calls attention to the rich internal life of agents that is possible in divinely determined worlds.
For a discussion of philosophers’ intuitions changing based upon whether the cases are first personal or third personal, see Tobia, Buckwalter, and Stich (). To be sure, the existence of people in our world who believe both in divine determinism and in their own responsibility does not establish the possibility of such people in divinely determined worlds. Perhaps there are no divinely determined possible worlds, for example. Nevertheless, it does make it plausible that if there are divinely determined possible worlds, there are people in them who believe that they reside in such a world and also believe that they are responsible for their actions.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Responding to the Direct Appeal to Intuition
The fourth difference, that Jonathan’s action is much more ordinary and described so that it relates to his own agency, might also suggest that our intuition in Case ** is more reliable than our intuition in Case *. In general, we seem to trust our intuitions about ordinary things more than those about extraordinary ones. Moreover, Case ** describes Jonathan’s ordinary action in a way that seems continuous with his own agency. In Case *, there is no discussion of how the choice to murder arose, other than that it was caused by God. But the divine determinist will deny that this is the whole story, and may reasonably suggest that the omission of any connection between this choice and the murderer’s own deliberation unfairly primes us for the "not guilty" intuition. One might worry that Case ** unfairly primes us for the “guilty” intuition, but I don’t think that is so. The case as presented notes both that he is determined to act as he does and that his action is connected with his own character and behavior. What of the fifth difference – that Rogers’s case involved murder while mine involved rudeness? We might be tempted to think that the requirements of moral responsibility become stronger as the severity of the action in question increases, but I see no principled reason to believe this. Instead, I suggest an error theory for our willingness to adjust our criteria of responsibility based upon the severity of the offense. Our practices of holding people responsible become more harmful in proportion to the gravity of the offense to which they respond. This is obviously true of punishment, but the point applies to cases of blame as well. It is painful to be blamed for something, whether we deserve it or not, and it is generally more painful to be blamed for something serious than for something slight. We desire the respect of our peers, and if they respond to our actions with extreme moral indignation, we are usually hurt. Thus, we (rightly) tend to be more careful when assigning blame for serious offenses than for slight ones, since we would rather err by mistakenly refraining from placing strong blame when it is deserved than by mistakenly assigning strong blame when it is not deserved. If this is so, then we should not be surprised if we too quickly excuse murderers, but also, perhaps, not be too surprised if we are too slow to excuse those who are rude, like Jonathan. In sum, we should conclude that four differences between Rogers’s case and the case of Jonathan lend support to the reliability of our intuitions in the latter case, and a fifth difference points to possible problems with both. If this is correct, then it should soften our intuition about Case * or, if this doesn’t naturally happen, it should at least lead us to assign a lower
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
level of credence to our intuition. Although this is not sufficient for rebutting this problem entirely, this first prong of response does provide the beginnings of a full reply and paves the way for the next prong. The second prong of my recommended reply is to weaken the initial intuition by further detailing the case to closely align with one’s preferred views of divine causality and the relationship between creation and the creator. Instead of producing a new version of the Jonathan case, I suggest the following as an addition to it. From this point on, consider this addition to be an official part of Case **: Suppose that that from all eternity, God has created and sustained Jonathan in being, together with all the other beings in the world. Suppose that in virtue of this creative act, God has a certain sort of authority over human lives that individual humans do not possess over others, perhaps modeled on the authority that parents exercise over their children, but of an even greater degree. In creating, God has intentionally brought into being a world, including Jonathan and his actions, and God’s doing so could not have been impeded by Jonathan or anyone else. Nevertheless, this divine creation did not involve any sort of event-causal process whatsoever. In keeping with this scenario, it is best to avoid thinking that God caused Jonathan to act, and better to conceptualize God as having produced Jonathan as an acting agent. This creative act is a precondition for Jonathan’s acting. Without creation, there is no possibility of Jonathan or his act at all. Moreover, although tempting, we must not think of God as being one more thing in Jonathan’s universe, albeit one both extraordinary and extraordinarily weird. Instead, we should think of God as somehow infinitely different from all other beings, and indeed different from the universe of creatures entirely – as so radically different that it is misleading to think of God as a being, rather than as being itself and the ground of the existence of all beings. This radical transcendence from the created realm requires a modification of how we understand “God creates Jonathan acting.” Perhaps the best device for considering this – although surely imperfect – comes from the world of fiction. In thinking about God producing Jonathan together with his act, we should think about the relationship between an author and a character in a novel. God is not like another being causing Jonathan to act, but rather like an author writing Jonathan in his acting. For this reason, we must think of God’s action as involving a different sort of causality than is normally investigated. Sherlock Holmes is caused to investigate a crime both by his own curiosity and by Arthur Conan Doyle, but it would be an elementary mistake to suppose that these two causes operate on him in anything approaching a common way. So too, we should think of God’s causal production of Jonathan’s act as unlike anything else that has a role in bringing about Jonathan’s action (such as
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Responding to the Direct Appeal to Intuition
his character, his state of exhaustion, or his weak will). Of course, there is a limitation to this image: While Conan Doyle is incapable of producing a true agent, God – divine determinists contend – is not similarly limited.
What does this supplement to Case ** add? The goal here is not to provide such a precise exposition of divine causality that it is clear that Jonathan is responsible. Instead, the goal is to fill out the case in such a way that the initial intuition might be weakened. I have already suggested that for some philosophers, Case ** (in its original form) will elicit a weaker “guilty” intuition than that produced by Case *. Although different philosophers will react differently, I now suggest that the modifications to Case ** will – at least for some philosophers – further weaken their intuitions concerning Jonathan’s guilt. Perhaps for some the intuition that Jonathan is guilty will vanish entirely, but for most it will remain in some form or other. For this reason, I suggest moving on to the third prong of the reply. In this third and last stage, I suggest that the divine determinist should call into question the reliability of intuitions concerning Case **. The use and value of intuitions within philosophy have become increasingly controversial. It has been controversial, for example, whether philosophers do rely on intuitions, whether they should do so, and whether their intuitions are more reliable than those of nonphilosophers. Certain contested positions within these debates may be of help to divine determinists in formulating their reply to direct-intuition cases like the one under consideration: After all, if philosophers shouldn’t be appealing to intuitions, then manipulation cases cannot even get off the ground). Nevertheless, I suggest a different reply. Even if philosophers do appeal to intuitions (as seems clear), are right to do so, and generally produce reliable intuitions, there are reasons to decrease our credence in the reliability of intuitions elicited from Case **. Some intuitions involve nothing extraordinary, nothing far outside of our everyday experience. For example, I have the intuition that it is wrong to kick cats for fun. Although I do not regularly perceive people kicking cats for fun, I do regularly see cats, people having fun, and instances of
This final claim – that God can produce true agents in this way – is too closely connected to the issue of whether divine determinism undermines moral responsibility to settle in one way or the other in this case, for it is part of (or closely connected with) what is at stake in the argument. To begin by either assuming it or denying it would likely incite complaints of question-begging.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
both kicking and pain. The fact that the relevant intuition does not involve anything radically outside of my experience gives me some reason to be confident in it. Would it be wrong to kick a cat for fun whose brain had been modified by our mythical neuroscientists so that every time it was kicked, it experienced a jolt of pleasure instead of pain? In this case, an immediate “moral” or “immoral” intuition would be less reliable, since I have no experience with creatures that experience pleasure instead of pain when they are physically harmed. This does not mean that there is no way for us to investigate the matter – we need not rope it off as an area of unexplorable mystery. Instead, we could reason from general principles about, for example, when it is wrong to injure another creature for its own benefit, and whether mere pleasure counts as a benefit. We may also consider long-term consequences, both to ourselves and to the cats in question. I suggest that the direct appeal to intuition may be relevantly similar to trying to solve the moral conundrum of kicking suitably modified cats by a direct appeal to intuition. Divine determinism, even if true, does not reveal itself in our everyday experience. For this reason, our direct intuitions about it, although not completely worthless, should not be trusted as much as other, more ordinary intuitions. This does not suggest that reasoning about God in general or divine determinism in particular is off-limits. Instead, it means that intuitions immediately concerning divine determinism deserve limited confidence. It also suggests that other ways of considering the matter – ones that appeal to general principles connected with everyday experiences – might be more telling.
Assessing Manipulation Arguments and Bullet Biting
In this chapter, I have developed two manipulation arguments, each of which concludes that divine determinism is incompatible with human moral responsibility. In each case, I developed what I think are promising lines of response for the divine compatibilist. Before closing, I wish to turn briefly to two additional questions: First, how do manipulation arguments play out differently depending upon whether they take up physical or divine determinism? Second, what, if any, costs must be accepted by divine determinists in light of their confrontation with manipulation arguments? In answer to the first question, I conclude that the prospects of responding to ordinary manipulation arguments do not exactly match those of responding to manipulation arguments targeting divine determinism for two reasons. First, ordinary and divine compatibilists cannot freely make
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Assessing Manipulation Arguments and Bullet Biting
use of each other’s soft-line replies. Both have their own responses available, but those arguments must be assessed on their own merits. Second, a direct appeal to intuition can motivate manipulation worries with divine determinism; nothing similar can be done to call ordinary compatibilism into question. Consider, now, the second question: Do manipulation arguments reveal new costs for divine determinists? It is sometimes said that manipulation arguments may be avoided, but only if the compatibilist is willing to bite some bullets. Must divine determinists bite some bullets in responding to manipulation-style arguments? I think many will need to do so. To begin, consider one particular way of understanding “bullet biting.” On this use of the phrase, a philosopher – Rita, say – bites a bullet when each of the following is true: a. b. c. d.
Rita has an intuition that ~P. Rita sees that some set of beliefs she has commits her to belief in P. On the basis of b, Rita begins believing P. Even after adopting P, Rita continues to have the intuition that ~P.
A divine determinist may well be forced to bite bullets in responding to both the RFCA and the direct appeal to intuition. The RFCA contends that divine manipulation is relevantly similar to cases of (comparatively) ordinary and extraordinary human manipulation. It also contends that human manipulators, through their manipulation, undermine the responsibility of their victims. Plum, in Cases –, is not responsible for killing White, or so the RFCA contends. Each of these claims is intuitively plausible. Divine determinists – at least those unwilling to give up belief in human moral responsibility – must take either a hard-line or a soft-line response. The hard-line response follows a path marked out and explored by ordinary compatibilists, which makes the surprising claim that Plum is responsible for killing White even in cases populated by human manipulators. The soft-line response I developed earlier contends that there is a relevant difference between Cases – on the one hand and Case * on the other hand, and maintains that the RFCA does not threaten the tenability of this claim. In each of these cases, it seems to me, some divine determinists must bite a bullet while giving their preferred response. Let’s suppose our
This claim is generally made when considering hard-line replies to manipulation arguments. For a discussion, see Mele (). This is inspired by Mele’s () use of the phrase, but is not identical to it.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and Free Will: Manipulation Arguments
philosopher, Rita, believes in divine determinism and also has the intuition that Plum is not responsible for murdering White in Cases –. Unfortunately for her, she begins thinking about the RFCA and comes to think that her two beliefs are in tension with each other. After consideration of her options, Rita decides to follow ordinary hard-line compatibilists in admitting that Plum is responsible in spite of human manipulation. Nevertheless, it is likely that Rita will continue having the intuition, even if weakened, that Plum is not responsible. In the sense outlined previously, Rita bites a bullet in responding to the RFCA. The soft-line response to the RFCA may appear to fare better – and perhaps it does in this regard – but some divine determinists will find a bullet here as well. Consider Rita’s friend Riya. Suppose that Riya is a divine determinist and begins considering the RFCA. She finds both claims of the argument intuitively plausible: It seems that Plum lacks moral responsibility in the early cases and that Case * is relevantly similar to those that preceded it. Nevertheless, Riya thinks that her reasons favoring divine determinism (together with belief in human moral responsibility) outweigh anything the RFCA has to offer, and she decides that the best response is the soft-line one. For this reason, she denies that Case * is relevantly like those that preceded it. Riya (honestly) claims that the RFCA has not defended a claim to the contrary, and concludes that since she has good reason to adopt belief in divine determinism, human moral responsibility, and Plum’s guiltiness in Cases –, there must be some relevant difference between divine and human manipulation. Nevertheless, suppose that Riya continues having the intuition (even if weakened) that no relevant difference exists. In the sense discussed earlier, Riya must bite a bullet in adopting the soft-line response. Consider, now, the direct appeal to intuition. In response to this objection to divine determinism, I detailed a three-part response. This response aimed to weaken the intuition that divine determinism eliminated human moral responsibility before casting doubt on the reliability of whatever remained of this intuition. Consider a third divine determinist, Rika, who has the same intuition as Rogers. She thinks it is intuitively obvious that if God determined her to kill someone, she would not deserve to be punished for it. Nevertheless, she also has come to believe that there are good reasons for accepting divine determinism, and she is firmly committed to the belief that mature and well-functioning humans are generally morally responsible for their actions. Rika’s intuition weakens as she walks through the first two steps of the three-pronged reply; in the third step, she reaches the conclusion that her (now weakened) intuition
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Assessing Manipulation Arguments and Bullet Biting
that divine determinism eliminates moral responsibility deserves less credence than other intuitions that concern more ordinary matters. Nevertheless, her intuition remains. Like Rita and Riya before her, Rika is forced to bite a bullet. Surely some divine determinists will not respond like Rita, Riya, or Rika. Most likely, some divine determinists will claim no intuition whatsoever that either human manipulation and (suitably detailed) divine manipulation are relevantly similar, or that divine determinism undermines moral responsibility. Nevertheless, the stories of Rita, Riya, and Rika do not seem particularly fanciful. To the extent that they are not, I conclude that even if both the RFCA and the direct appeal to intuition can be successfully rebutted, such manipulation cases do reveal a cost for divine determinism – a cost that must ultimately be weighed against whatever reasons one has favoring this view.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.004
Divine Determinism and the Author of Sin Objection
Introduction
In this and the next chapter, I take up the question of whether God is the author of sin. I suspect that most theists wish to deny such an attribution, but it is particularly difficult for divine determinists to do so. Before moving into particular difficulties with divine determinism, however, it is important to clarify the meaning of “author of sin,” since it is a particularly fuzzy attribution. I wish to distinguish between three different meanings of this phrase. In claiming that God is the author of sin, one might be expressing any of the following propositions: Determined: God has determined human agents to sin. Caused: God causes human sins. Blameworthy: God is blameworthy for human sin in virtue of determining its occurrence.
It is clear that divine determinists must accept the truth of Determined (although they might well argue that the phrase “author of sin” is misleading and should not be used as a way of expressing this point).
Although I have no statistical data to present, the dearth of theist philosophers who admit that God is the author of sin seems to provide an indication that few theists are willing to accept this attribution. One thing that is especially telling is the way in which divine determinists, who have a difficult time not admitting that God is the author of sin, still argue against it. Indeed, the Westminster Confession of Faith III. is clear in denying that God is the author of sin. For a contemporary determinist case against the claim that God is the author of sin, see White (). Welty () responds to Molinist complaints that Calvinists make God the author of sin by arguing that even if determinists must accept that God is the author of sin, so too must Molinists. Implicit in this exchange is the presumption (even if defeasible) that it is preferable to avoid claiming that God is the author of sin. See Edwards (, Part , Section ): “But if by ‘the author of sin,’ is meant the permitter, or not a hinderer of sin; and at the same time, a disposer of the state of events, in such a manner, for wise, holy and most excellent ends and purposes, that sin, if it be permitted or not hindered, will most certainly and infallibly follow: I say, if this be all that is meant, by being the author of sin, I don’t deny that God is the author of sin (though I dislike and reject the phrase, as that which by use and
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Introduction
Divine determinists are also in agreement about Blameworthy: It must be denied, since it, unlike either of the other two claims, charges God with blameworthiness, which is incompatible with perfect goodness. Although they deny Blameworthy, there is an important question of whether such a denial is compatible with their other commitments. Some think, for example, that knowingly causing an agent to sin is itself evil. If this is so, then divine determinists will have a difficult time justifying their denial of Blameworthy. I will take up this difficulty in Chapter . Divine determinists are in agreement about Determined (they accept it) and about Blameworthy (they deny it), but there is disagreement about whether they should accept Caused. At first glance, it appears that Caused cannot be avoided. If God determines all creaturely actions, and some acts are sins, then (seemingly) God causes sins. Some, however, think that a careful analysis of the nature of evil will reveal that divine determinists may deny Caused without undermining the integrity of their determinist credentials. The basis of such an analysis is the claim that evil is not a being, but rather a privation. For this reason, I will call this defense of the determinist position the privation solution (PS). The key claim of the PS – that evil is a privation – is a claim with a fine pedigree, having been endorsed by both Plotinus () and Augustine (). The PS goes on to claim that sin is “composed” of two elements: the act of sin and a defect. The act of sin is itself a being, but, the PS claims, the defect is not itself an entity, but rather a mere privation. If one grants that one of the two elements of sin is not an entity, then it is easier to see how God might not be the cause of sin. I will begin by focusing my attention on one variety of privation solution, one which I will call the modest privation solution (MPS). I call this account “modest” because it does not deny that privations can be caused, but merely that the privations that are constituents of sin are caused by God. I maintain that several arguments provide reason for
custom is apt to carry another sense), it is no reproach for the Most High to be thus the author of sin.” Of course, some philosophers may think that an evil divine being has determined all creaturely events. Although I have no objection to such philosophers calling themselves divine determinists, I have chosen to use this phrase in a more circumscribed way, as I noted in Chapter . In earlier work, I called it the privation defense (Furlong ), but W. Matthews Grant (b) rightly pointed out that this term may be misleading, since it might suggest that it is a defense in the sense discussed by Plantinga (). In the past (Furlong ), I called this view “unadorned” because it did not rely on any claims about the unique nature of divine causality. In this present work, I merely call it “modest,” because I think it is most important to distinguish the MPS from a more robust solution I discuss later in this chapter.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
thinking that the MPS is insufficient to allay the worry that divine determinists must accept the truth of Caused. Although not all of these arguments are equally forceful against all versions of divine determinism, taken together they suggest that divine determinists should look beyond the MPS if they hope to avoid Caused. After consideration of the MPS, I will discuss the robust privation solution (RPS), another privation solution that claims the defects of sins lack causes entirely. This solution, I maintain, has better prospects than the MPS. I will close by considering the costs of accepting Caused, as some divine determinists have done.
The Modest Privation Solution
Privation solutions, of one sort of another, are not new. Several such solutions have their roots in Thomas Aquinas’s discussion of the nature of sin. Whether the account offered by Aquinas is really a privation solution is not clear, but it has seemed so to some of his followers. Whatever the case may be, the PS deserves to be discussed simply on its merits. What I am calling MPS has had a number of proponents, but none as clear as W. Matthews Grant. For this reason, I will especially draw from his description of this solution. According to the MPS, we can distinguish the act of sin from the sin’s defect, which together form the sin itself. Consider the following case: Paul desperately wants a new Rolex, but is unable to afford its high price. One day he notices that someone in his office has carelessly left one on a nearby desk. Paul, realizing that no one will spot his action, takes the watch for himself. Paul’s act of stealing the watch is the sin itself. The act of sin is
For Aquinas’s discussion of whether God is the cause of sin, see Thomas Aquinas (, I–II, q. , a. ). For solutions rooted in Aquinas’s thought, see Garrigou-Lagrange (), Grant (), and Nicolas (). It is possible to argue that Aquinas does not present what I am calling a PS because he might not be trying to use his account of the privation of evil as a solution to the present issue at all. Indeed, some have argued that he denies the truth of divine determinism, even as I use this phrase. See, for example, Maritain () and Torre (). Grant’s account is particularly suitable to be taken as a characteristic MPS for several reasons. First, although he does try to show that his account is in keeping with Aquinas’s work, he is more concerned with putting forward a viable solution. Notice that although Grant’s own view avoids divine determinism because he denies that God’s causal activity is explanatorily prior to human acts, his presentation and defense of the PS does not make use of this point. For discussion of his view in relation to the PS, see Grant (b). It is unclear what sorts of descriptions defenders of the MPS would be willing to admit as legitimate. Could one describe the sin in each of the following ways: Paul’s taking the watch, Paul’s taking the watch that does not belong to him, Paul’s taking the watch that does not belong to him without the
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
The Modest Privation Solution
Paul’s act of taking the watch insofar as this has being. This may seem identical to the sin itself, and indeed there is no positive difference between the act of sin and the sin itself. Nevertheless, the defender of the MPS will insist that the act of sin does not include the defect of sin, since this is a privation and does not have being. What is this defect? Grant describes it as “a lack of conformity to moral rule or order” (, ). The idea seems to be that we can distinguish the actuality or being of Paul’s act from its failure to accord with the demands of morality that govern property. Having drawn this distinction, the defender of the MPS affirms that God is the cause of the act of sin, but not of the sin itself. The claim that God is the cause of the act of sin is both important and common ground for many theists, determinist and indeterminist alike. Aquinas (who is claimed both by determinists and indeterminists), for example, claims that “The act of sin is both a being and an act; and in both respects it is from God” (, I–II, q. , a. ). Since God is the cause of all other beings, and the act of sin is itself a being, it is clear that God is causally responsible for it in some way. Nevertheless, this claim does not entail the further claim that God is causally responsible for the sin itself. Although one constituent of the sin itself is caused by God, defenders of the MPS deny that the other constituent, the defect, is caused by God. According to defenders of the MPS, the defect is not caused by God, but neither is it caused by any action of the agent. Instead, this view claims that the defect, which is a lack of conformity with the moral rule, is caused by an omission, which is the agent’s not attending to the moral rule. The agent does not bring about this defect by performing some other action, but rather by not acting in a particular way. Since this omission is not itself an entity, it is not obviously the case that God is responsible for it. This answer seems to give rise to a new objection to this position. The problem is that if agents are considered causes of defects because they failed to act in certain ways, then it seems that God, too, can be considered a cause of the defect since God, too, failed to act to prevent the defect from
permission of its owner? Some of these descriptions do not indicate that the act is defective, but perhaps they may still be used as descriptions of the sin; all descriptions, after all, under-describe. In any case, I do not think the MPS requires a specific answer to this question. Grant (, –) is explicit on this point. By comparison, this position seems implicit in other privation accounts, such as that of Garrigou-Lagrange (). Defenders of the MPS do not agree among themselves about how to understand the nature of this omission, although this detail will not be important for our purposes. For a discussion of the controversy, see Grant (, –).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
occurring. Given divine determinism, it seems that beings exist if and only if God causes them, and thus the privation of a being seems to be caused by God’s failure to cause the being. Grant formulates this worry: “Given that God’s causing guarantees the sinner’s considering, and that God’s notcausing guarantees the sinner’s not-considering, does it not follow that it is also caused by God in virtue of God’s not causing the sinner’s consideration?” (, ). To respond to this objection, Grant, following Aquinas, suggests a list of necessary and sufficient conditions that must be met for an agent to be a cause of something in virtue of not acting: Effect e is caused by being S in virtue of S’s not f-ing if and only if
(I) S’s f-ing would have ensured or at least made it likely that e not occur, and (II) S had the power to f, and (III) S ought to have f-ed. The plausibility of these criteria is strengthened by considering a situation Grant discusses. Suppose that you come home from work and notice that the fish food, which you sprinkled into the aquarium hours earlier, remains uneaten. What might have caused this? Consider the following three possibilities: (I) The food is still floating because your goldfish didn’t eat it. (II) The food is still floating because the plants didn’t eat it. (III) The food is still floating because the water didn’t dissolve it (Grant , ). Each possible explanation has recourse to the inactivity of some substance. In each case, the substance did not, in fact, act; if any had acted, then the food would not be floating. Surely if the plants did manage to eat the food, then it would no longer be floating in the aquarium. Despite this, it is absurd to
The source of this position in Aquinas (, I–II, q. , a. ) is as follows: “Now one thing proceeds from another in two ways. First, directly; in which sense something proceeds from another inasmuch as this other acts; for instance, heating from heat. Secondly, indirectly; in which sense something proceeds from another through this other not acting; thus the sinking of a ship is set down to the helmsman, from his having ceased to steer. But we must take note that the cause of what follows from want of action is not always the agent as not acting; but only when the agent can and ought to act. For if the helmsman were unable to steer the ship or if the ship’s helm be not entrusted to him, the sinking of the ship would not be set down to him, although it might be due to his absence from the helm.” See Grant (, ). I have replaced Grant’s term “substance” with the term “being” since it is less controversial that God is an being than that God is a substance.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Three Arguments for the Insufficiency of the MPS
think that we should consider all of these inactivities to be causes. In particular, it seems that we say that something is a cause through inactivity only if there is some reason to think that the thing should have acted. Grant persuasively argues that the notion of “ought” in the criteria is not that of moral obligation, but rather that of natural inclination. This allows us to say, concerning the case under discussion, that the fish are the cause of the food remaining, since they did not consume it. With this set of criteria in hand, the defender of the MPS is in a position to argue that God is not the cause of the defect of sin through inactivity, because it is not the case that God ought to have acted so as to prevent it. In defense of this claim, one might say that, first, God cannot fail to act as required because no external rule governs divine action. Thus it cannot be the case that God should have caused the agent to attend to the moral rule and failed to do so. Second, perhaps a thing ought to do something only if the activity in question is necessary for the thing to reach its end. Since God’s end is nothing but the divine existence, it is clearly not the case that God ought to cause humans always to attend to the moral rule. The point of this discussion is not to provide a lengthy defense of God’s innocence in the face of evil in the world, nor is it meant to show that God is governed by norms (moral or otherwise) that forbid causing sins but permit allowing them. In fact, the two reasons noted previously for thinking that is it false that God ought to cause all agents to attend to the moral rule suggests that God’s activity is not governed by any norms at all. Instead, the point is to show that condition (c) (S ought to have f-ed) is not met when God refrains from causing sinners to attend to the moral rule. Perhaps surprisingly, on this view we must first determine whether God ought to have caused sinners to attend to the moral rule; only once we have settled this issue can we discover whether God is the cause of the defect of sin. I will address the issue of whether God ought to always cause agents to avoid sins in Chapter ; in any case, none of the arguments I will produce against the MPS will turn on this question.
Three Arguments for the Insufficiency of the MPS I now wish to turn from an overview of the MPS to a criticism of it. In this section, I present a series of arguments that suggest the MPS is incapable of
See Grant (, ). Grant is not the only one to place emphasis, within a PS, on the claim that it is not the case that God should always prevent sin from occurring. See, for example, GarrigouLagrange (, :).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
entirely resolving the worry that a divinely determining God is the cause of sin. Some of these arguments more naturally target one or the other of the two sorts of divine determinism I have been investigating (Edwardsian and primary divine determinism). Taken together, however, these arguments suggest that we should look elsewhere for a solution to the present problem. The Author and the Novel I will start with an argument that targets primary divine determinism. To begin, consider the following story: Eleonora is a famous novelist, but she has grown tired. Fifteen years ago, Eleonora created the first Isaac Storm novel. Overnight, she rose from obscurity to fame. Her earlier work had never caught on with the public, but her fortunes changed when this new title was released. It became an instant bestseller and fans clamored for more. Eleonora wrote more. Fans wanted a series of films, so she sold the rights. The series of movies was even more popular than her books. Eleonora now had adoring fans, respect from her fellow authors, and, most importantly, financial independence. But she was tired, tired of working in the fantasy genre of the Storm novels, tired of writing simply to meet the expectations of fans, and most of all, tired of Isaac Storm. She decided that the seventh book would constitute the end of the Isaac Storm saga. At the end of this work, Isaac renounces his former heroism, slays his father, and burns down his hometown. He even watches as the flames take his brother’s life. Fans were livid – Eleonora had not only ruined this book, but she had sullied each of the others. Eleonora responded to the criticism, saying, “Let’s be clear on one thing. I caused Isaac to kill his father. And I caused him to set fire to the town. I even caused him to stand by, watching the flames encompass his brother when he could have easily saved him. But I didn’t cause him to refrain from saving his brother. No. That is merely an omission, and although Isaac can be said to have caused that omission, the same cannot be said in reference to me.”
What should we say about Eleonora’s response? To my ears, it sounds preposterous. She has caused the existence of a narrative. She admits that she has created the positive elements of that story, including the state of affairs in which Isaac stands still, watching his brother die. It might be granted that the omission of saving his brother is simply a privation – not something Eleonora added to the story, but merely something she passively allowed to occur in it. But, importantly, the complete description of what she created entails the truth of the claim that Isaac refrained from saving his brother. The privation in question follows necessarily from what she admits to creating. It is important to be clear on this point: The claim
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Three Arguments for the Insufficiency of the MPS
is not that what Eleonora wrote at the beginning of the book settled Isaac’s character in such a way that this later omission was inevitable. Instead, the claim is that Eleonora caused all positive elements of the state of affairs at the same (fictional) time in which Isaac’s omission occurred. Isaac’s omission follows from the positive act of standing still, watching his brother die, which is itself caused by Eleonora. Importantly, the idea is not that the positive features constitute the first link in a deterministic chain that will lead to the eventual omission. Instead, the omission follows from the positive features in the way that a hole follows from the positive shape of a doughnut. It might be tempting to say that Isaac’s omission is constituted by his act of standing still, but this might give it more metaphysical standing than the MPS can allow. Perhaps Eleonora could claim that she didn’t cause any of the fictional happenings, since nothing fictional can be caused. Alternatively, she might suggest that omissions, not enjoying any being of their own, are not the sorts of things that can be caused by anyone. Importantly, however, her response commits her to the falsity of both these claims. I contend that Eleonora owes us a further defense of why she is not a cause of Isaac’s omission. In particular, she needs to show why she is not a cause of it given that she is a cause of all the positive features of this world, the omission follows from the positive features, and omissions can have causes. It seems that a world divinely determined through primary causation is relevantly similar to the preceding story. God (like a novelist) creates an entire world down to the smallest detail. According to the MPS, the omissions of agents in this world can be effects. Moreover, a proposition completely describing every being of this world (including all their features) – a world that God designed – entails a proposition describing every omission it contains. In virtue of this, it seems that God is the cause of human omissions, including the omissions that lead to the defects of sin. Since God is the cause of all positive elements of sin, if God is also a cause of the omission, and thereby the defect, then God is responsible for sin itself. Because the MPS has failed to show that God is not the cause of the defect of sin in virtue of causing the omission, it has also failed to show how primary divine determinists can avoid admitting Caused. Internal Relations: From Rules and Primes to the Cause of Sin I have developed the preceding argument in terms of God’s responsibility for the omissions that lead to the defect of sin. A similar argument might
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
be constructed that shows God is directly responsible for the defect of sin itself. Remember, the defect of sin is the sin’s lack of conformity with the moral law. Such a description, however, might seem to suggest that this defect is a positive being (a relation) and so seems incompatible with the general strategy of privation solutions. We must be careful here, since the MPS claims that this defect has no being whatsoever. Thus, if we think of this lack of conformity as a relation, we must be willing to admit either that some things properly called relations lack all being or that we are extending the use of this word in this instance. In either case, we might then ask what sort of “relation” this is. This relation seems to be what we might call a transcategorical relation: Its relata hail from very different categories. The first relatum is the human action insofar as it has being. The second relatum is the moral law. However we understand the defect, most divine determinists will likely wish to say that the relation of inconformity between act and law obtains (when it does obtain) merely in virtue of the intrinsic natures of the relata. In other words, the relation of inconformity follows merely upon the nonrelational characteristics of the relata. Such relations are sometimes called internal relations. Although philosophers provide various accounts of internal relations, nothing of importance for the present discussion turns on which account is accepted. I will follow van Inwagen (, ) in claiming that a relation is internal if the intrinsic characteristics of the relata “[settle] the question whether that relation holds between (or among) them.” Most divine determinists, I suspect, will say either that the character of the moral law is settled by God or that moral norms are necessarily true. When this point is combined with the claim that the relation of inconformity is an internal one, we have the makings for an objection to the sufficiency of the MPS. To see this, consider the following two cases. Suppose I create a new game that is played between robots. I also create a robot and challenge others to create robots to enter into matches of this new game. In the first match, my robot faces off with an opponent. During the match, I press a button on my control box and the robot immediately acts in a way that violates the rules. My opponent (the human, not the robot) contends that I caused the robot’s action to have the characteristic of being in violation of the rules.
Some might contend that this argument is simply a reformulation of the first one considered. Whether this is so depends upon how we individuate arguments, which, it seems to me, has no bearing on the force of the arguments under consideration. One possibility is that the character of the moral law is settled by the natures of the creatures that exist. Given divine determinism, such a possibility still points back to God.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Three Arguments for the Insufficiency of the MPS
Before considering this case in more detail, let’s look at a second case. Suppose I work in a factory that produces model airplanes. My job is to hand paint the collector’s edition models. Thirty minutes before quitting time, my boss, Sandy, comes to evaluate my work. She is stunned by how productive my day has been. Both Sandy and I are currently math majors, working at the factory to pay our tuition bills. We are also a bit nerdy. This comes out in some of our interactions, including this one. Sandy tells me that since I caused the painted models to be greater in number than the quantity of prime numbers less than , I can quit early today. Each story provides us with a causal claim: Rules:
I caused the robot’s action to have the characteristic of being in violation of the rules. Primes: I caused the painted models to have the characteristic of being larger in quantity than the number of primes less than .
There are a number of similarities between these cases. Each claim involves an internal, transcategorical relation. In each case, I am causally responsible for the first relata. There is an importance difference, though. In the first case, I am also responsible for the second relata: I determined the rules of the game. In the second case, the character of the second relata is not caused by me, but neither is it caused by anyone else. Instead, it is necessary that there is a particular quantity of prime numbers less than . We can now turn to each case. Is my opponent correct in asserting the truth of Rules? At the very least, the claim seems plausible. I caused both the robot to act in a certain way and the rules to be a certain way. Given the ways in which I caused both of the relata, an inconformity followed. Is my boss right in asserting Primes? Once again, it at least seems plausible that she is. I am responsible for there being a particular number of painted airplanes. Although it is true that I am not causally responsible for the prime numbers less than , it would be odd to invoke this point in arguing against Sandy. If the number of primes less than was not fixed, and someone was changing it while I was working on the airplanes, then perhaps I would have a basis upon which to object to Primes. Let’s return to the issue of divine determinism and the MPS. Recall that our defender of the MPS claims that there is a component of sin that is a mere privation – a lack of conformity to the moral rule. This component, the argument goes, is caused by sinners (through an omission), but not by God. But is this claim plausible? In the picture under consideration, God is the cause of all positive aspects of the world – anything that has being is caused to be precisely as it is by the unstoppable power of God. Now
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
suppose that God causes a human act of sin. Proponents of the MPS are happy to make this claim, since the human act has being. Given the nature of the moral law, and holding fixed the intrinsic characteristics of this human act, the defect of nonconformity with the moral law follows. Following van Inwagen’s account of internal relations, we might say that the nature of the moral law and the intrinsic character of this action settles whether there is a defect of inconformity. In a divinely determined world, it seems that either the nature of the moral law is caused by God (if, for example, either divine command theory is true or the content of the moral law is caused by the existence or actions of creatures) or it is necessarily as it is. If God causes the moral law to have specific features (for example, forbidding certain sorts of actions), then a human’s sin is similar to the case involving Rules: In both cases, we have a state of affairs that is caused by an agent (God causes the human’s act of sin, and I cause the robot’s act), and this state fails to conform to a standard set by that same agent. If the moral law has its characteristics necessarily, then a human’s sin is similar to the case involving Primes: In both cases, we have a state of affairs that is caused by an agent (God once again causes the human’s act of sin, and I cause the planes to be painted), and this state bears a certain internal relation to necessary truths. Thus, if Rules and Primes are plausible – and it seems to me that they are – then divine determinists have some explaining to do. Importantly, the MPS does not seem to provide the needed explanation. The MPS claims that the defect of sin has no being, but the same can be said of internal relations generally. Moreover, it explicitly claims that the defect of sin is the sort of thing that can have a cause – indeed, it claims that humans cause these defects all the time. Defenders of the MPS will need to either provide some argument for thinking that God’s determining all of creation is different from the cases surrounding Rules and Primes or create some argument for thinking that Rules or Primes is false. Before moving on to the third and final argument for the insufficiency of the MPS, it is worth considering whether the force of the present objection equally targets Edwardsian and primary divine determinists. The first argument, which focused on the example of an author writing a novel, more naturally targets primary divine determinism. I see no reason for thinking the present argument is relevantly similar. Instead, the cases discussed in this argument seem to provide reason for thinking that in virtue of causing the characteristics of both relata, or the characteristics of
For discussion of this point, see Armstrong (), Campbell (), Fisk (), Heil (), Lowe (), and Simons ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Three Arguments for the Insufficiency of the MPS
one relatum when the other relatum has its characteristics of necessity, an agent is also the cause of the internal relations that follow upon these characteristics. If this thought is correct, both Edwardsian and primary divine determinists have reason to worry. States of Affairs That Cause Omissions I now turn to an argument that most naturally targets Edwardsian divine determinists. According to this argument, God is the cause of a sinner’s omission – the cause of the defect of sin – in virtue of having caused a state of affairs that both caused the omission and preceded it in time. To begin, notice that it is clear that we sometimes do consider agents to be causally responsible, in virtue of their activity, for the omissions of other things. Consider the following case: Joan and Robert have a single child, John. Joan and Robert consider many social conventions to be impositions on human freedom, and thus they do not act in keeping with them, nor do they teach John to do so. In particular, they consider the use of the word “please” to be a particularly damaging social convention, and they teach John to have a similar attitude. In their eyes, the use of this word involves inappropriate docility. John does not learn, as many other children do, that he ought to use the word “please” when he makes requests. On one particular occasion, he asks his teacher to be excused without saying “please.”
Suppose for the sake of the example that John likely would have learned to say “please” from others, if his parents had simply failed to discuss this term. This being the case, it seems inappropriate to say that John’s omission was caused by an omission on the part of the parents to teach John appropriate manners. Instead, it seems that their positive act of teaching him that such actions are inappropriate was a cause of his omission. This, it seems, is in keeping with our normal way of providing explanations. Thus, it is at least possible for agents to cause the omissions of others by creating a state of affairs – in this case, the character of the couple’s child – that later caused the omission. Although it is clear from the preceding example that an agent can be the cause of an omission through the agent’s own activity, it is not at all clear which criteria must be used to determine when an agent can be considered a cause in this way. To aid our investigation, consider again the case Grant discusses about the pet fish. Suppose that you come home from work and notice that the fish food, which you sprinkled into the aquarium hours earlier, remains uneaten. The
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection cause of the state of the fish food is the inactivity of the fish. The fish should – in at least some sense of “should” – have eaten the food.
What might have caused their inactivity? Presumably, quite a number of answers might be possible. Perhaps the fish gorged themselves the day before and did not desire to eat again. Perhaps the temperature of the water in the aquarium dropped, which caused a state of lethargy in the fish. Perhaps someone had spilled some coffee into the box of fish food, and the fish did not like the way this made the food taste. Consider the second explanation, that the fish remained inactive because the water temperature in the tank dropped. Suppose, for the sake of the example, that the temperature drop was a sufficient condition for the inactivity of the fish. If the temperature in the tank were to drop, then the fish would unfailingly refrain from eating it, barring your special intervention by, say, introducing a chemical additive to the water that would make them more active. In this case, we can still say that the fish food remains floating in the aquarium because of the inactivity of the fish. Their not-eating the food is the cause of it being the case that the food remains floating on top of the water. Clearly, in this example, we would not say that you caused the fish’s inactivity if you failed to introduce the chemical additive to the water. Grant’s criteria help explain this; it is not the case that you both can and should add this chemical to the tank. Even so, it seems perfectly in keeping with our ordinary way of speaking to say that the drop in temperature caused both the fish’s not-eating and the food remaining on top of the water. Moreover, if you are the one who lowered the aquarium’s thermostat, then you thereby seem to become the cause of (A) the temperature dropping, (B) the fish’s not-eating, and (C) the food remaining in the tank. I say that this is in keeping with our ordinary usage in much the same way that Grant says that it is in keeping with our ordinary usage to say that the food is still floating because the fish did not eat it rather than because the plants did not do so. An expansion of the example might help show the plausibility of my claim. Suppose the fish food you use is guaranteed to be liked by your fish. The manufacturer is so confident in its product that the company will refund your money if your fish do not eat the food. Now imagine that you seek a refund for this product because your fish did not eat it. Once the complete circumstances become known to the company, would it not be appropriate for the manufacturer to refuse a refund because you caused the fish not to eat it? We would, in fact, say that you caused the fish’s not-eating. Reflecting upon this example, I do not think it is possible to derive a set of necessary
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Three Arguments for the Insufficiency of the MPS
and sufficient conditions for determining when an agent causes the omission of another through activity. I do think, however, that reflection upon this case can yield conditions that are jointly sufficient: The not f-ing of agent S is caused by agent R in virtue of R’s acting if (i) (ii)
R is causally responsible for state of affairs p and p causes S’s not-f-ing.
In the preceding example, it is stipulated that you lower the temperature of the water, thereby becoming responsible for the state of affairs termed “the fish are lethargic.” This state of affairs causes the fish to refrain from eating the food. In virtue of these conditions being met, you can be considered the cause of the inactivity of the fish. These are, it would seem, the pertinent details that the company would reference in its decision not to give you a refund. Now consider again the question of whether God is the cause of the defect of sin. This argument is meant to suggest that Edwardsian divine determinists will face a particular difficulty, even if they appeal to the MPS. According to the view under consideration, the current complete worldly state of affairs is determined by the one that preceded it, which was itself determined by the one before it, and so on. God comes into the causal picture by creating the initial state of affairs, together with the laws that govern how the world develops over time. Now let us turn back to the criteria just suggested. Both criteria are met in the case of sin. The agent’s not-attending to the moral rule is caused by a state of affairs that can be causally traced back to God’s activity of creation. I contend that this condition is satisfied in the case of an agent’s nonapplication of the moral rule, where R is replaced with God, S with a human agent, f with the application of the moral rule, and p with the complete creaturely state of affairs immediately prior to the agent’s relevant nonapplication of the moral rule. Consider an agent, Paul, who commits some sin. I contend that the complete creaturely state of affairs a few moments prior to Paul’s relevant nonperformance causes Paul’s nonapplication of the moral rule (which causes the later action’s nonconformity to the moral rule). In virtue of causing the state of affairs prior to Paul’s nonperformance, God is also the cause of the nonperformance itself. Let’s take a closer look at the case of Paul, who fails to apply the moral rule and performs an action that does not conform to it. Suppose that Paul is speaking with Anne. At t, Anne recounts all the work she has done on their shared residence and asks Paul how he has spent the day. At t, Paul
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
begins to consider the matter. At t, embarrassed that he has simply taken a nap, Paul lies to Anne. Between t and the eventual lie, Paul does not even consider the moral rule. He does not apply this rule to his deliberations, his considered judgment of what should be done, or his choice to lie. This omission is responsible for the nonconformity between his utterance and the moral rule. But what caused this omission? I contend that, given Edwardsian divine determinism, this omission can be traced back to God in virtue of God’s action. We can begin with the plausible claim that if God is the cause of the initial creaturely state of affairs (the first instant of the universe), and a proposition expressing this state of affairs (together with descriptions of the laws of nature) entails a proposition expressing a future state of affairs, then God is also responsible for the future state of affairs. Given Edwardsian divine determinism, the proposition expressing the state of the world at t (the first moment of the Paul and Anne case) is entailed by the proposition expressing the initial state of creation together with the laws of nature. God, then, is the cause of the state of affairs at t. Moreover, given the relationship between any two complete worldly states of affairs in a world governed by deterministic laws, it seems that the state at t causes the state at t. If this is correct, then criteria (i)–(ii) have been satisfied, and we must conclude that God is the cause of Paul’s omission and, since the omission causes the defect, of his sin as well. Notice that the MPS does not seem to provide any unique resources that would help dispel this worry. Although this argument has been developed to target Edwardsian divine determinists, the previous two arguments are not limited in this way. The first targets primary divine determinists, and the second is perfectly general. It seems, then, that divine determinists of either stripe have reason to worry that the MPS fails to demonstrate that they can avoid admitting the truth of Caused. If they are unwilling to admit that God is the cause of sin, they should seek other solutions. Toward Another Solution In the preceding paragraphs, I argued that divine determinists face significant difficulties, even if the MPS is adopted. A second, and more promising, solution involves modifying the PS, making it a little less
In years past (Furlong ), I developed a version of this argument specifically targeting primary divine determinists; see Grant (b) for a reply. Although I still think that such an argument can be made plausible, I now appreciate that it turns on both identity conditions of omissions and a controversial account of what it means for one thing to account for another. The arguments discussed here depend, I think, on premises that are at once simpler and more widely held.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Three Arguments for the Insufficiency of the MPS
modest – hence the qualifier “robust” in the robust privation solution (RPS) term. This reply begins with the familiar claims that God is the cause of all that has being and that evil is not a being. Furthermore, it relies on the same constituent analysis of sin offered by the MPS: The sin itself comprises the act of sin (a being) and the defect of sin (a privation). To these claims, we add a twist: Neither the agent nor anything else causes the defect of sin. That is, the defect of sin is without a cause at all. This may seem to make sin a mystery, as some philosophers and theologians have done. Kathryn Tanner, for example, claims that sin “is ultimately without explanation” (, ). Some might think that such an appeal to mystery is fitting in the case of sin, while others think it is a great and perhaps unacceptable cost of any view that depends upon it. Whatever the merits or faults of such a view, the MPS need not – and indeed, should not – claim that sin is without explanation. If we wish for such an explanation, we may appeal to intrinsic characteristics of the beings involved (most importantly, the sinner and the sinner’s action) and the content of the moral law. To avoid making sin inexplicable, defenders of the RPS should appeal to the following: Explanations: Internal Relations:
An explanation of x may be satisfactory without appealing to a cause of x. Internal relations are, strictly speaking, incapable of being causal relata.
Defenders of the MPS might, of course, try to avoid Internal Relations by claiming that the case of sin is unique, and that although some internal relations can be causal relata, the relation of inconformity that sinful actions bear to the moral rule cannot, but this would appear ad hoc. Instead, the best course of action would be to defend Internal Relations by appealing to ordinary considerations about causes. Such a defense might begin by noting that causation seems to be what Bigelow (, ) calls a “genuine two-place relation”; that is, it is a relation that requires the existence of its relata. Internal relations, this current view claims, have no share in existence. If this is the case, then Internal Relations may be
This view has something in common with a view developed by Heath White (). It differs, however, in that White wants to claim that some privations are caused. White suggests that privations can be caused in two sorts of ways: One involves change over time and the other involves varying from some independent causal pattern, but God’s causality does not involve either. Once causes of privation are posited, I see no reason to restrict such cases to the two possibilities that White suggests. For this reason, I think that determinists tempted by White’s view should go further and reject causation of and by privations entirely.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
motivated without appeal to considerations of divine action or human morality. Explanations, too, may be independently motivated. First, noncausal explanations have been defended in areas of philosophy other than philosophy of religion and ethics. More importantly, Explanations seems particularly appealing when the explandum is an internal relation (as the defect of sin is). Notice that Explanations does not entail that for any x, an explanation of x may be satisfactory without appealing to a cause of x. Perhaps explanations of things like substances and intrinsic properties must appeal to the causes of the entities being explained, but the case might be different with internal relations. Suppose we want to know why Michelangelo’s David is larger than Verrocchio’s bronze of the same subject. It is true that Michelangelo’s David bears the relation larger than to Verrocchio’s David. What explains this? An explanation of this relation would explain why Michelangelo sculpted his statue to be more than meters tall, while Verrocchio made his own creation a mere centimeters. It might discuss the properties of marble and bronze and the purposes of the artistic commissions. In short, it would provide the causal stories that lead each individual statue to have its unique set of intrinsic features. This, it seems, would be sufficient for providing a satisfactory explanation to the original question. It would be unreasonable, according to this view, to respond, “I understand why Michelangelo’s David is the size it is, and I understand why Verrocchio’s is the size it is, but why is the former larger than the latter?” In explaining why each statue’s size is what it is, one also explains why the relation holds between them. Thus, even though no cause of the relation is posited, the explanation seems both complete and nonmysterious. Even if we can make sense of complete explanations of some internal relations without reference to causes, we might worry about doing so in the case of the defect of sin. This worry arises because, unlike the case of the statues, the case of sin requires reference to the moral law. I suggested that a complete explanation of the larger than relation discussed earlier could be achieved without reference to the cause of this internal relation (which is fortunate, since, according to this view, internal relations lack causes). Instead of pointing to the cause of this relation, the explanation could point to the causes of the intrinsic characteristics of the two statues. There is reason to worry that the same cannot be done in the case of sin. The
This is especially the case in philosophy of science and philosophy of mathematics. For a recent discussion, see Lange ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Conclusion
defect of sin is a nonconformity between an action and the moral law. If we grant that the moral law has a cause – that it has the content it does in virtue of the actions of God, for example – than there is no problem. Divine command theorists can reasonably claim that the case of sin is analogous to the case of the statues. In turn, the defect of sin can be fully explained by pointing to the causal explanations of the characteristics of the human action and the moral law. But what if true moral norms are necessarily true? What if this leads us to conclude that the moral law has no cause? In such a case, either a full explanation can be provided for the moral law, despite its lack of a cause, or it cannot. If it can, then an explanation of the defect of sin can point to a causal explanation of the human action and a noncausal explanation of the moral law. Conversely, if no explanation can be given for the moral law, then the explanation of the defect of sin can point to the causal explanation of the human action and to the moral law itself, as a sort of foundation of explanation. It is a commonplace that explanation must stop somewhere, and an explanation of the defect of sin that hits foundational bedrock in virtue of partially resting upon necessary truths seems none the worse for doing so. It is important to clarify exactly what the RPS may reasonably be thought to accomplish. If it is successful, it provides divine determinists with both the space and a reason to reject Caused. The RPS does not easily allow the divine determinist to get God “off the hook” for the evil in the world. As Heath White has noted, “If God creates a piece of Swiss cheese, it may be strictly correct to say that he does not cause the holes in the cheese, but it would defy credulity to say that God is not responsible for the holes in the cheese” (, ). Nevertheless, some divine determinists (including White), view the battle over Caused as one worth fighting, even if victory has limited consequences.
Conclusion In this chapter, I have considered two possible ways for a divine determinist to defend a denial of Caused. I argued that the most well-established such defense, the MPS, is not sufficient for showing that divine determinists can resist accepting Caused. I conclude that the RPS is the best strategy for the determinist intent on denying that God causes sin. Of course, divine determinists have another option at their disposal: They may affirm Caused. Some divine determinists have done so, admitting that God is the cause of sin, but arguing that this is not as bad as it first appears. They
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Author of Sin Objection
maintain that although Blameworthy (God is blameworthy for human sin in virtue of determining its occurrence) is unacceptable, Caused is harmless. Some may suspect that embracing Caused would force them to accept Blameworthy as well. I will consider this suspicion in Chapter . Even if accepting Caused does not lead to any further commitments, it is not difficult to see why it troubles philosophers. The core worry seems to be that Caused involves God far too closely with evil. If God is wholly good, the idea goes, that must involve not only an innocence of moral wrongdoing, but also a kind of purity that would be sullied by causing sin. This claim has been intuitively plausible to many theists – philosophers and nonphilosophers alike – and even many divine determinists are clearly uncomfortable with the idea that God causes sin. Accepting Caused, then, seems to bring with it a certain cost. Those who accept it may attempt to mitigate it by suggesting that there is no reason to think that Caused entails that God lacks some great-making feature, but in the end it still appears that some cost will remain.
This is evident both from the existence of divine determinists who deny that God is the cause of sin and from the perennial refrain – sung even by divine determinists – that God merely permits or allows sin.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.005
Divine Determinism and the Blameworthiness Objection
Introduction In Chapter , I took up the question of whether divine determinists can deny Caused (God causes human sins). I argued that they may do so, but not in the manner that is sometimes suggested. Moreover, some divine determinists seem content to affirm Caused, maintaining that this is not as bad as it appears. In this chapter, I turn to a more worrisome proposition, one that no traditional Western theist can endorse: Blameworthy: God is blameworthy for human sin in virtue of determining its occurrence.
Although divine determinists deny Blameworthy, some of their opponents suspect that they are not in a position to do so. The language of this accusation may vary: Some may claim that divine determinism entails that God violates a moral duty, or that God is guilty of the sins committed by humans. For the purposes of the present discussion, I will take these contentions as identical to the claim that divine determinists are committed (whether they wish to admit it or not) to the truth of Blameworthy. I will refer to this claim – that divine determinists are committed to Blameworthy – as well as the collection of reasons for endorsing it, as the blameworthiness objection. The basic thrust of the blameworthiness objection is not difficult to see: If divine determinism is true, then God arranged for all the evils of human history – not only natural disasters such as floods and diseases, but also human sins such as slavery and murder – and did so in such a way that these evils were determined to occur. The history of our species is not just spotted with evil; it is riddled with it. As Marilyn McCord Adams says, “history is a slaughter bench” (, ). To arrange for such evils seems quite clearly to be morally odious. Thus, divine determinism seems to quickly lead to Blameworthy. Importantly, this is not simply one more cost
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
that divine determinists can accept. Since the divine determinism we are investigating in this work involves the determining will of an all knowing, all powerful, perfectly good being, divine determinists are committed to the falsity of Blameworthy. If divine determinism forces us to accept Blameworthy, then divine determinism (at least the sort I am addressing) is false. Like many of the objections to divine determinism discussed in this book, the blameworthiness objection is intuitively plausible and thus forceful even in an informal, rough and ready form. Nevertheless, a careful examination can be undertaken only after more nuanced versions of the objection have been examined. For this reason, I will begin by constructing a more carefully developed version. I will then spend the remainder of the chapter surveying a number of responses to the blameworthiness objection. I will take up two responses – one based upon the claim that God doesn’t intend evil, the other based upon the claim that God has the authority to cause evil – in some detail. The contours of such views are not obvious, so they deserve careful attention. I will briefly survey a large number of other possible responses, quickly indicating the general shape of each as well as some possible difficulties that defenders of each will need to address. The goal is not to develop any particular response in great detail, but rather to briefly assess their promise as well as their costs. In the end, I will suggest that this objection can be overcome in a number of ways, and I recommend a complex response that draws force from each of the replies discussed.
Toward a Nuanced Blameworthiness Objection If humans are often blameworthy for their actions, then what can be said of a God who has determined this occurrence? It at least seems that we would be blameworthy if we caused others to carry out the worst that human history has to offer. If such action would be a blot on our pasts, then how is it possible for a perfectly good being to determine every evil ever performed? For divine determinists, the awkward question is not why God permits evil, but rather why God actively brings it about. Because of the nature of this question, it is helpful to think about the blameworthiness objection as separate from the problem of evil as generally
Some think that the distinction between doing and allowing is morally relevant, even in the case of God; others disagree. For discussions about whether there is any relevant difference between God’s permission of evil and God’s determining it, see Byrne (), Cowan (), and Helm ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Toward a Nuanced Blameworthiness Objection
formulated (which will be discussed in Chapter ). There is more than one way to distinguish these two issues. First, one could begin by noting that the general problem of evil is often considered in terms of God’s ability to have created a world better than this one. The presence of easily realizable alternatives seems to indicate that a perfect being did not create this one. The blameworthiness objection, in contrast, does not appeal to such alternatives in any way. It suggests that no matter which alternatives are possible, it was impermissible for God to create this one, since doing so includes an impermissible cooperation with evil. Alternatively, we could distinguish the two problems by looking to the Pauline principle that evil cannot be done in the pursuit of the good. Formulations of the general problem of evil tend not to rely on this principle, but instead claim that some evils are not necessary for bringing about goods, or instead that the goods that do presuppose such evils (or at least the possibility of such evils) are not sufficient to justify the evils. The blameworthiness objection, in contrast, seems to rely at least implicitly on the Pauline principle or something similar to it. Although the formulation of the blameworthiness objection I develop here does not appeal to this principle by name, it does rely on an intuition of blameworthiness that seems to presuppose something like it. A final way to distinguish these questions is by appeal to the ground for the supposed evil property that characterizes God’s creative action. As the general problem of evil is often considered, the normative property of God’s action seems to derive from a complex state of affairs consisting of God and the act of creation together with God’s omnipotence and omniscience, available alternatives, the actual consequences of the evil in the world, and the modal relationship between certain sorts of goods and evil (that is, whether these sorts of goods necessarily depend on the existence of evil). The blameworthiness objection, in contrast, claims that the evil character of God’s action is grounded in a far simpler state of affairs: the divine will and its relation to human sins. To begin closer inspection of the blameworthiness objection, we should note that the charge that God is blameworthy for human sin presupposes another, more innocent claim: Responsible: In virtue of determining the existence of human sin, God is in some way morally responsible for it
Responsible does not suggest that human sin is a blight on God’s otherwise clean ledger. As I noted earlier, I am following Derk Pereboom’s account
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
of what it means for an agent to be morally responsible, and this account does not even presuppose that the act for which the agent is morally responsible is morally relevant. As Pereboom explains: For an agent to be morally responsible for an action in this sense is for it to be hers in such a way that she would deserve to be blamed if she understood that it was morally wrong, and she would deserve to be praised if she understood that it was morally exemplary . . . This characterization leaves room for an agent’s being morally responsible for an action even if she does not deserve to be blamed or praised for it – if, for example, the action is morally indifferent. (, )
This characterization discusses what some philosophers think of as the primary object of moral responsibility – namely, individual actions. Nevertheless, one might also say that agents are responsible for the effects of their actions, as well as what follows from their omissions. In fact, once we see what “morally responsible” means in this context, divine determinists seem obviously committed to Responsible. After all, as they have it, human sins are a consequence determined by divine action, and surely God had a sufficient understanding of such consequences. Seemingly, the only way to avoid this conclusion is to deny that God is a moral agent at all. This understanding of Responsible allows room for some theists to claim – and perhaps most others should follow – that Responsible is true and add that the appropriate response to this instance of divine responsibility is praise. To take a single example, they may claim that God is responsible for Pharaoh’s sinful response to Moses and that God’s faithful should praise God (but blame Pharaoh) for this response. Defenders of the blameworthiness objection deny the plausibility of such a response. Instead, they claim that if Responsible is true, then Blameworthy is also true. Since it seems impossible for divine determinists to avoid Responsible, the objection continues, divine determinism unavoidably leads to the conclusion that God is to blame for human sin. What can be said of the claim that if divine determinism is true, then our sins are God’s fault – that is, not simply that God is morally responsible, but that creation is a ground for blame? This assertion has a long tradition, tracing back at least to Anselm. And whether an argument could be advanced for it, it has seemed to some to be “intuitively obvious” (Rogers , ). Often the problem is left at this level, but it will be helpful to offer some formulations of arguments in defense of this claim.
Some theists take this route. See Davies (, ) and Holland (). See Rogers ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Toward a Nuanced Blameworthiness Objection
Some have thought that we might use rather straightforward universal moral precepts to show that a God who determined evil actions would be evil. But this is more difficult than it first appears. Consider the following moral claim: (MC) There is at least one evil act E, committed in the actual world, such that if an agent A intentionally both caused and determined agent B to cause event E, then A has acted wrongly.
MC is attractive, because its acceptance does not commit one to strong versions of the Pauline principle that claim no evil may be committed for the sake of the good. Instead, it merely supposes that at least one of the evils in the history of the world is such that it ought not to have been caused, irrespective of the consequences. Moreover, it seems that if both MC and divine determinism are true, then God has acted wrongly. Why suppose that MC is true? Begin by supposing, as many theists do, that some actions are universally prohibited. Murder and rape are common examples; no matter what one’s intention or circumstances, one acts wrongly in intentionally murdering or raping another individual. One likely reason for thinking this is that in murdering and raping, we greatly harm other beings who possess a great degree of dignity. But there is also a long and veritable tradition, tracing back at least as far as Socrates and accepted by many theists down through the centuries, that holds that the greatest harm comes to an agent not in being killed or otherwise physically or psychological abused, but in performing evil oneself. If this is true, then it seems that in causing anther agent to perform an evil action (murder, say), especially if one thereby determines the action, one causes a double harm: One brings about the death of another and does an even greater harm in making someone a murderer. If murder and rape are always and everywhere wrong in virtue of the level of harm they cause, then it seems that causing and determining others to perform such acts would be wrong as well. As I noted in Chapter , some divine determinists claim that although God has determined everything that happens in the world, God does not cause sins. Such determinists can easily avoid any instance of the blameworthiness objection that relies on MC. Nevertheless, this objection may be formulated to avoid any such reliance. Consider the following moral claim:
This principle bears great similarity to ones endorsed by Byrne () and Walls ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection (MC) There is at least one evil act E, committed in the actual world, such that if an agent A intentionally determined that an agent B perform E, then A has acted wrongly.
This principle gains plausibility from the very same considerations that lend support to MC. Nevertheless, this is a stronger claim (the truth of MC entails the truth of MC, but the reverse is not true), and it is far from clear that it is true. Suppose that an agent has to choose between eating a bowl of ice cream and eating an apple. Now suppose that this agent cannot refrain from acting entirely and that the agent’s choice, no matter which is made, will be indeterministic. Finally, suppose that indeterministic events are extremely rare, with one occurring only every century or so. Our agent’s choice will have far-reaching consequences. Should we count our agent’s choice as an act of determining every event that was not determined to happen the moment before the choice but was determined afterward? Notice that given the account of determinism given in Chapter , we would say that the complete state of affairs in this world, including this agent’s choice, determined the later events. Things become even more complicated in fully deterministic worlds. If determinism is true, then each of our own actions plays a role in determining the actions of many other agents. Must we say that if physical determinism is true, then an action determines every later event to which it is causally related in any way? If so, then perhaps MC is false. If we limit the principle to cases where the agent in question is responsible for the entire state of affairs that determines a later action, then we seem to have a principle that applies only to deterministic Gods, and thus seems question-begging. It is difficult, then, to find a universal moral principle that is noncontroversial and that, together with divine determinism, obviously entails that God is blameworthy. Nevertheless, despite the problems with MC, it seems that a divine act that determines (even if it doesn’t strictly cause) the entire course of a world filled with evils such as ours would be an immoral act. Why might this be so? Even if MC does not apply in this case and MC is false, it is still useful to return to the reasons motivating the acceptance of each moral claim. If it is always and everywhere wrong to murder or rape because of the harm it brings to others, then it likewise seems wrong to bring about events that include such evil actions. We might think that ordinary agents can avoid wrongdoing while determining
This point is rarely commented upon, but see Todd (, ) for a discussion of this point in relation to whether agents may blame others whose actions they have determined.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Toward a Nuanced Blameworthiness Objection
that evils occur if their acts are far removed from the evils that follow from them, insofar as their actions form only a tiny part of the circumstances that somehow lead to the evils. In the case of divine action, this sort of exception simply does not arise. If God’s action determines all later events, whether through the laws of nature, via primary causation, or directly and solely through divine action, then no circumstances are truly independent of the divine will. This being so, it seems that the harm brought about by such actions falls at God’s feet. A further argument for this claim is based upon an analogy with humans. Suppose we developed a way to manipulate the choices of others without undermining their moral agency. Suppose further that one individual, Jack, decided to do so by manipulating his neighbor Thomas to become a serial killer. If Thomas’s moral agency is truly retained, then he is surely responsible for his murders, but what of Jack? It seems obvious that he, too, should be held accountable for both Thomas’s murders and turning Thomas into an evil agent. Furthermore, it is far from clear how one could claim that Thomas committed an evil act (since murder is always and everywhere wrong), but that Jack may not have, depending upon his motivations. Perhaps there are conditions under which it is permissible to allow such evils to occur, but if one is committed to the claim that it is impermissible to do evil even for the sake of a good outcome, then divine determinists seem unable to simply point to some greater good that would exonerate the agent determining the evil action. We might further modify our example to more closely mimic the case of divine determinism. Suppose that through centuries of technological innovations, humans eventually gain the ability to modify infants to such a degree that they can design entire human biographies, down to the smallest detail of the smallest choice. Now suppose that Alice does so with a whole host of individuals, living within a closed environment of her own creation. In such a scenario, she has determined how each agent will act in a given situation, and has completely controlled which situations each will encounter. Now suppose that she does so in such a way that the moral agency of each person is retained. In Alice’s little world, murder and rape are, if not common, at least not exceedingly rare. Since these agents retain their moral agency (or so we may assume for the example), they are to blame for their own acts. But what of Alice? Putting aside the rather questionable action of creating designer human biographies in the first place, it seems quite evident that she, too, is accountable for all of these rapes and murders.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
This argument continues by noting that the cases of Jack and Alice are relevantly similar to that of God if divine determinism is true. Since Jack and Alice are evidently guilty of wrongdoing, the same is true for a God who infallibly ordains the evils of our world. I think the blameworthiness objection takes on an especially powerful form in this sort of argument from analogy. This argument depends upon an initial intuition about Alice and the claim that there is no relevant difference between Alice and a determining God. Structurally similar claims were discussed earlier in regard to manipulation arguments. There, we saw that some compatibilists take the initially surprising route of defending the so-called hard-line response. The hard-liner denies the conclusion of manipulation arguments by denying the seemingly obvious claim that agents who are thoroughly manipulated are not responsible for their actions. They resist the argument’s conclusion by resisting the powerful initial intuition that it relies upon. Before turning to promising responses to the blameworthiness objection – each of which will attack the claim that the cases of Alice and a determining God are relevantly similar – I wish to highlight the strength of the crucial intuition that Alice has acted wrongly. The hard-line reply to this argument focuses not on God, but on Alice. In particular, it rejects the intuition that she has acted wrongly. Although hard-liners may be consequentialists, their response collapses into the general consequentialist response discussed later in this chapter, and so will be put aside here. A non-consequentialist hard-liner (for brevity, I will hereafter use “hard-liner” and associated phrases to refer exclusively to this sort of responder) will admit that at least some actions derive their normative properties from factors other than consequences – in particular, that some wrong-making features of an action are independent of consequences. Moreover, a hard-liner will claim that some or all of these wrong-making features are decisive, in the sense that the presence of such a feature renders the relevant action impermissible, whatever the consequences. The hard-liner will claim that although some actions are impermissible in virtue of wrong-making features independent of consequences, Alice’s action is not characterized by any such features, although perhaps the
The one possible exception to this is an appeal to consequentialism. This reply notes that all actions, including those of Alice and God, must be evaluated in light of all their consequences. If this is so, then Alice’s story would require far more fleshing out before we could assess the permissibility of her action.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
An Intention-Based Reply
murders and rapes within the world she designed (that is, the actions of the murderers and rapists, as distinct from her causings of such actions) do possess such features. For this reason, the argument based upon an analogy between Alice and God cannot get off the ground. The trouble with this response is that it is difficult to see a way to acquit Alice without falling into full-blown consequentialism. To do so, one would need to begin by outlining which wrong-making features are both independent of consequences and decisive. One would then need to show that some of those features are present in some cases, but absent in the Alice case. Moreover, one would need to be sensitive to the fact that the case of Alice may be modified in a number of different ways while still retaining its similarity to the case of a determining God. This task is, perhaps, not impossible. Nevertheless, it is undoubtedly a large undertaking, and one that – as far as I am aware – no divine determinist has attempted. Until some attempts are forthcoming, this possible reply cannot be taken seriously. Given the plausibility of the initial intuition concerning the Alice case – that Alice has done something wrong and deserves blame – I conclude that the divine determinist should avoid the hard-line response. Although the problem of divine blameworthiness seems formidable, it is not obviously unanswerable. In the following sections, I explore a number of different possible responses to the blameworthiness objection. Some of them, I argue, are less than plausible. Others, although viable, include what many will consider to be serious costs. I begin by exploring two replies in some detail before turning in the fourth section to a quick investigation of a number of other replies.
An Intention-Based Reply One way to avoid the blameworthiness objection is to deny that God intends the evil present in the world. Heath White explains this view: [Divine determinism] holds, indeed, that God’s will determines every other contingent fact. It does not follow that God intends every other contingent fact. All of what happens is a consequence, indeed a deterministic consequence, of God’s action of primary causation, but not all of what happens
The Alice case is relatively recent; it first appeared in Furlong (a). Nevertheless, the basic idea behind the blameworthiness objection it appears in – that it is blameworthy for anyone, even God, to act in the way Alice does – is not new.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection need be an intended consequence. Some of what God brings about may be foreseen but not intended. (, )
White notes that one way of understanding what I am calling the blameworthiness objection is in terms of the principle of double effect. White provides an outline of this principle, which claims that an action is permissible only if it meets these four conditions: . . . .
The action is not intrinsically wrong (it is good or at least indifferent). At least one good effect is intended. No evil is intended, as an end or a means. The good done provides a good enough reason to produce the evil.
If God intends the sins of determined creatures, then God’s action is impermissible. White’s argumentative move, denying that God intends every aspect of the world, allows him to sidestep this objection. This reply comes in different varieties, and each carries its own costs. Notably, although White invokes primary causation in his description, the same general reply – that God avoids blameworthiness in virtue of not intending the evil in the world – may be made by either primary divine determinists or Edwardsian divine determinists. The particular replies they are able to give, however, look radically different. The Edwardsian Divine Determinist Variation Recall that the Edwardsian divine determinist claims that God has determined all creaturely events in virtue of having caused initial conditions that evolve according to deterministic laws. Given this view, the intentionbased reply is straightforward. It begins by noting that the initial conditions themselves involved no evil (or at least no moral evil). God intended these initial conditions and foresaw all of the consequences that would inevitably follow from them. In claiming that the foreseen evils are unintended, however, the divine determinist is constrained in formulating the reasons for creating the initial conditions in the first place. Nevertheless,
White’s analysis of causation is simply meant to rebut the claim that God intends evil. For the broader claim that God is not blameworthy, he also relies on the claims that God does not cause sin and that “God has no duties or responsibilities to cause goods in his creatures” (, ). I take it, however, that the claim that God does not intend sin is necessary, on his view, for rebutting the blameworthiness objection, since he writes (albeit in the formulation of an objection to his view, but with apparent approval for this element) that “any intention to bring about evil would impugn the goodness of God” (, ). For this formulation, see White (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
An Intention-Based Reply
some options remain. As Alexander Pruss notes, the Edwardsian divine determinist “can say that God might have a reason to produce a set of initial conditions that are valuable in ways independent of the fact that they necessitate sin. Thus, these initial conditions might be particularly simple and hence intrinsically valuable, or they might give rise to a particularly elegant set of evolutionary developments” (, ). If God merely foresees that sins follow as an unintended side effect of creation, then the blameworthiness objection likely fails. The determinist must still deal with the general problem of evil, of course, but dispatching this objection would still constitute an initial victory. Nevertheless, I doubt many theists will be willing to embrace this view because of its costs. Notice that given this reply, God’s intentions in creation seem nearly exclusively focused on nonpersonal elements. Since all humans’ lives are so deeply intertwined with past sins, it is hard to see how many elements of such lives could be intended, given the constraints of this reply. To see this, consider a single action by a single human living in a modern country today. Take, for example, my six-year-old daughter’s offer to read my three-year-old son a story. Could God intend this? Given the reply under consideration, it is unlikely. If God intended this act, then God must have also intended the means that caused this act. Given the commitments of Edwardsian determinists, these causes will stretch back to the initial conditions of the universe. The problem, of course, is the winding path leading from those initial conditions to my daughter’s act. It is overwhelmingly likely that my very existence – upon which my daughter’s existence, and thus my daughter’s act, depends – can be traced back to the initial conditions only by having recourse to evil actions along the way. Of course, if any of my ancestors was conceived as a result of rape, then this is easy to see. Even if my complete ancestry is clean of any such blot, it is exceedingly likely that a key pair of ancestors (a pair who together produce another ancestor) would never have met if not for some evil, whether this be unjust war or impermissible colonization or some much more mundane action. Furthermore, my very existence depends not only upon my pairs of key ancestors meeting, but also upon them conceiving at very specific times. These considerations do not imply that there is no possible world free of all sin but containing me. This may be true, but my worry with this reply does not depend upon anything so metaphysically grand. Instead, it merely depends upon the claim that my existence in the actual world came about in virtue of prior evil actions. If this is correct, then God could not have created the initial conditions as a means to the intended goal of my
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
existence without also intending these evil acts as a means. Of course, there is nothing particularly special about my existence, other than, perhaps, that I am not a very early human, and that my ancestors were not incredibly isolated. Those who lived far earlier or in isolated environments may have smaller chains of events relevant to their existence, but it is still likely that they, too, will exist in virtue of particular evils in the past. I leave the assessment of this cost to the reader. Some theists may accept or even revel in the claim that God did not intend creation, even in part, for us. I suspect, however, that many more theists will find this price too steep to pay. The Primary Divine Determinist Variation At first, it seems that the primary divine determinist will have a more difficult time defending the claim that God does not intend evil acts. After all, the primary divine determinist claims that God is immediately responsible for all that has being in creation. If this is so, then it seems that God must intend each aspect of creation. If evil is a mere privation, then there is no need to say that God intends evil itself, but the blameworthiness problem will still remain if God intends evil actions. White contends that this problem can be avoided by considering how God comes to select which world to create. To begin, he thinks we should consider the act of creation as one in which God creates a four-dimensional block of spacetime: How might he choose which block to create, from among the infinity of possible blocks of spacetime? We can imagine God contemplating a large number of schematic blueprints of such blocks. He takes note of the good aspects and the bad aspects of each block. Any good aspect he would see as a reason to create the block (i.e., that universe); any bad aspect he world see as a reason not to create the block . . . His reasons for creating the world he
It may be true that I am a relatively early human if the species continues far into the future. Whether this is so is not relevant; what is relevant is that I exist far enough along that my own existence depends upon countless generations before me. Some primary divine determinists, especially those within the Dominican tradition, may modify this claim a little. Some primary determinists say that God immediately creates a certain real entity, which has come to be called a physical premotion. This entity then brings about certain changes within the human being, in particular moving the human from a potential to act to actually acting. It is typical for these philosophers to claim that human choices are not unique in this regard, and that all actions require such divine assistance. Perhaps such a view requires us to say that God is mediately responsible. Even if this is so, because the immediately created premotion has only one purpose, to move a substance into action, it will be of little help in solving this particular problem. For discussion of this view, see Matava () and Oderberg ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
An Intention-Based Reply
does are simply the world’s good aspects, and all these aspects are intended. The bad aspects he foresees but does not intend. On this picture of God’s creative activity, no evil is intended. (White , )
White continues by arguing that there is no need to think that God intends an evil as a means to the goods for which it is a necessary condition, relying once again on the author analogy. Within a fictional world, one event may be causally connected to another, but this connection does not justify the belief that the author intended one as a means to another. White’s image is a provocative one, but it seems to me to crucially misrepresent the way most theists – indeed, most primary determinists – view the act of creation. Given White’s image of world selection, it is indeed plausible that God does not intend the evil of the world. Given this account, God’s act of creation is similar to my act of selecting which snack to buy from a vending machine: I survey the vast options available; I notice that different offerings bring with them different pros and cons; I then select one, intending each of its pros (its taste and its ability to satisfy my hunger, perhaps), while merely foreseeing its cons (its deleterious effects on my body and the fact that the company that created it supports causes I oppose, say). But there is a key difference between how most theists view creation and this vending machine model. In creation, God’s activity is not fully captured by the idea of selection. Here the author analogy is telling. Authors do not merely select a book to write out of the many they can imagine. Instead, each author is directly responsible for bringing about each word that shows up in the work. Each and every sentence is something considered, and the presence of each is an indication that the author intended this very sentence to be added to the work. Whether a particular part of a novel is intended for this or that purpose is indeed something difficult for the reader to ascertain, but that each part was intended is assured – at least, assuming the author is in complete control of the work and meticulous in attention to detail, as God is supposed to be. One might think that the solution gains necessary support from appealing to God’s eternity. If God is eternal, then we should suppose that God creates in a single act. But it seems that this fact is important to the present discussion only by pointing to a weakness in the author analogy. It is certainly true that no human author creates a novel by a single discrete act. Instead, each word must be added, one at a time. Some
It might also be relevant for investigating the question of whether God employs means–end reasoning in creation, but claims about such reasoning make up no part of my objection to this reply. For a discussion of this use, see Pruss ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
might think this is a relevant difference between a human author and God, but it does not seem relevant to the present use of this analogy. Would a human author creating a work in a single act somehow avoid intending each part? I think not. Consider John Cage’s ’”, which is a musical composition in which performers refrain from playing entirely for the length of the piece. Although the performance of this work involves a number of discrete acts, we can imagine Cage conceiving the work in a single moment. Suppose that he did so. From this mere fact, can we suppose that some of the moments of silence – perhaps the th second or the th – were not intended by the creator? I can see no reason why we should reach this conclusion. Both the author analogy and the selection model are mere images, and we should not expect to reach precise conclusions upon them. Nevertheless, the more God is taken to be actively involved in the creation of the world – that is, the more the author analogy takes precedence over the selection model – the less plausible this reply is. I cannot help but think that the solution works only if we take God’s act of creation to be far too similar to my act of pushing F- and watching a bag of Doritos tumble down the front of the vending machine.
The Authority Reply Another way to respond to the blameworthiness objection is to point to the important role that authority plays in the permissibility of an action. It is clear that in some cases, whether one is in a position of authority plays a key role in determining whether an action is permissible. For example, it seems that a professor is morally permitted to adjust a student’s grade, but that it would be impermissible for the professor’s spouse, upon seeing an open gradebook, to adjust it, even if the spouse possessed all relevant expertise and had evaluated all of the student’s work. The role of authority seems especially important in cases where one agent directly affects the life prospects of another. This can most easily be seen in the parent–child relation, where parents have morally permissible options that are (ordinarily) unavailable to others. To respond to the blameworthiness objection, the divine determinist can maintain that the key difference between Alice and God is that God possesses a level of authority that Alice lacks. Remember, Alice’s actions harm people in two ways: She harms those whom she determines to be murderers and rapists, and she also harms those who are murdered and raped. If divine determinism is true, then God seems to be in a similar
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
The Authority Reply
position, on the hook for both the harms endured by those who commit horrible crimes (since God determined them to do such things) and the harms endured by the victims of such crimes (since God determined that they be harmed). This reply begins with the idea that the wrong-making feature of Alice’s action is causing harm to others without the authority to do so. If this is so, then the moral wrong of harming others is structurally similar to that of stealing. Stealing is wrong not in virtue of what is done – the taking of property – but in virtue of its being done by someone without the authority to do so. If this view is accepted, then the wrong of killing is not found simply in taking life, but in taking life without the appropriate authority. Divine determinists might even note that consideration of self-harm supports this view. Some think that killing oneself is not immoral, at least not if certain conditions are met. Similar claims may be made about other self-inflicted harms. If these claims are plausible, their plausibility seems to come from the idea that we have an important sort of ownership or authority over ourselves and our lives. According to this view, it is wrong to act in such a way that certain harms are imposed, just in case one does not bear the appropriate relation to the one harmed. Once the contours of this view are motivated, the further question of who bears this relation to whom may be investigated. We might stay with the motivating idea that agents bear this authority relation to themselves and their lives, and nothing else bears this relation to them. Alternatively, some may contend that some agents, perhaps those charged with caring for the common good, bear this relation to other agents. The divine determinist seeking for a way to avoid the blameworthiness objection can contend that whatever is said of our authority over ourselves and others, God bears the relevant authority relation to each creature. Given common theistic commitments, it is plausible that God does bear a relation to humans that allows for a wider range of permissible actions than is available to ordinary human agents. Consider, for example, a case of human private property. It would be impermissible for another human
Indeed, I suspect that many theists may claim that only God bears this authority relation to innocent human lives. Notice that even these theists may legitimately use consideration of self-harm to motivate the idea that harm may be permissibly inflicted by those with authority. They may claim that although it is false that individuals may permissibly harm themselves in particular ways – by taking their own lives, for example – the reason many find this false claim plausible is that they are correctly supposing that those with certain sorts of authority may permissibly kill, but incorrectly supposing that human agents have this sort of authority over themselves.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
agent to take this property without permission (setting aside cases of truly extreme circumstances). However, many theists – both those who accept and those who reject divine determinism – will be comfortable with the claim that God could permissibly take away this property. Indeed, it is not uncommon for theists to claim that everything belongs to God; human private property is not, in the last analysis, a matter of an ownership relation, but rather one of stewardship. If this is correct, then there is little difficulty in explaining why God may permissibly take human property, but in ordinary situations, one human may not take the property of another. Such an explanation will depend upon the following: Weak Authority Thesis (WAT): In virtue of the authority relation God bears to creation, God may permissibly determine that any divine gift granted to humans be taken away from them at some point, as long as there is good reason to do so.
The divine determinist may accept WAT, then add that life is among the many things that God gives to humans. It follows that God may permissibly arrange for it to be taken away. Indeed, such a thought is common to theists of many stripes and is often invoked to make sense of the place of death in the world. Some philosophers, even those without determinist leanings, view life and death according to something like the WAT. Richard Swinburne writes: If there is a God, he has made it abundantly clear that the “gift” of life is a temporary one which he makes as long or short as he chooses. To use an analogy, I may lend you a book, saying, “you can have this until I want it back”. I don’t wrong you if I let other people use a book for a longer period than I let you use a book. (, –)
Of course, the application of the WAT to death is ordinarily limited to certain sorts of deaths. In particular, deaths caused by murderers are often excluded. Determinists, however, should apply the thesis to these cases, too. They should explain that even the deaths of murder victims are determined by God’s plan, and so the events of their deaths should also be thought of as ones in which life – borrowed from God for only a short time – becomes due.
See, for example, Job :– and Q.:. See, for example, Pruss (, ), who claims that the prohibition against killing the innocent does not apply to God, for “God is the owner of our lives, and has no duty to continue sustaining us in existence.”
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
The Authority Reply
If this view is accepted, then we can begin to exonerate God of certain sorts of wrongdoing. Consider again the analogy of Alice. Alice determines other human agents to become murderers. One of the key worries about this issue is that it seems to make Alice guilty of the very same crime as the murderer. Both, after all, have brought about the death of another; the mere fact that Alice has used another agent as in intermediary seems to be of little or no importance in assessing her guilt for this death. Since a determining God is similar to Alice, it seems that God, too, is blameworthy. This response maintains that God, in determining a human agent to commit murder, avoids blameworthiness in the death of the victim, because God has loaned life to the victim and possesses the right to establish its due date. The divine determinist, and indeed nondeterminists like Swinburne who claim that God possesses authority over life and death, must explain why God possesses this authority while Alice (and indeed ordinary human parents) lack it. Perhaps this is due to the fact that God creates ex nihilo, while Alice does not; or that God creates the individual life, whereas Alice merely arranges appropriate circumstances for it; or that life and indeed existence are nothing but a participation in the life of God, a share in the divine existence, and that nothing similar can be said of Alice and her relation to her little world. Different answers are possible here, and at least some of them are not ad hoc, but some theists may find that each answer brings with it unwanted costs. Another difficulty is that appeal to the WAT is sufficient for clearing God of blameworthiness for determining the harm done to the murdered victim, but not for the harm done to the murderer. Theists typically regard wrongdoing – rather than pain and suffering – as the worst sort of fate that can befall one. Consider, for example, Socrates‘s claim that his accusers are harming themselves more than him or Kant‘s claim that it would be better for the whole world to perish than for a single act of injustice to be done. For the determinist reply under consideration to be viable, it must also defend the permissibility of God determining that Jeffrey Dahmer and Heinrich Himmler developed into the sorts of people they became. The borrowing and returning analogy used to account for the permissibility of bringing about death will not work in these cases. Instead, the divine determinist would do well to rely on the claim that the perpetrator of evil does not receive an undeserved harm.
This particular response will likely require the rejection of materialism, but for many theists such a rejection will be independently motivated and thus not ad hoc. For Socrates’s claim, see The Apology in Plato (); see Kant () for that author’s claim.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
Suppose that God has ordained from all eternity that an agent, Alan, murder another human, Alicia. Because of this divine ordination, Alan cannot but commit the murder, and so undergoes the harm of becoming a murderer. It is true that God has determined that this harm be done to Alan, but it is open to the determinist to claim that this harm is not undeserved. First, note that this harm is nothing other than the act of killing itself, together with the moral consequences (the change in Alan’s character and the fact that he is now in a blameworthy state) that follow from it. Alan assuredly did not perform an earlier action by which he came to deserve these harms, but it seems that he deserves them in light of the action itself. If we assume that Alan retains his full agency, then he is morally responsible for his act of killing, and is thus morally responsible for the harm that he brings upon himself. If this is so, then this great harm is not undeserved; indeed, no such moral harm could ever be undeserved. Although the magnitude of the harm to the perpetrator of wrongdoing is great – sometimes thought to be far greater than the suffering which it causes others – the fact that it is not undeserved makes it less problematic to the determinist than it initially appears. A third difficulty arises when we move from cases of killing to other sorts of sins. It is, perhaps, not shocking to think of God possessing the authority to lend life and demand its return. Such a thought helps absolve God of determining agents to kill – but what of agents who torture or rape? Such cases do not fit easily into the model of temporary gifts or lending items and requiring their return. Using such a model to defend God’s innocence in the face of human lives cut short depends only on the relatively plausible claim that those who own something can permissibly take it back. Let us pause to review the authority reply as it currently stands. It begins by suggesting that there is a relevant difference between a determining God and Alice, even though both determine human agents to sin and suffer. This reply then notes the plausibility of the WAT, which is comparatively modest. This thesis merely claims that God may permissibly take back divine gifts, since not every gift must be permanently given. The WAT provides a basis for showing that God can permissibly decree that humans’
Perhaps an exception might be made to this because of the Christian doctrine of original sin. According to Christian tradition, all humans inherit a sort of undeserved moral defect due to original sin. Understandings of such a defect vary among Christians, and the difficult problems raised by this doctrine are not unique to divine determinists, so I will set such worries aside. If one does claims that the harm is undeserved, then this claim must be dealt with through an appeal to the Strong Authority Thesis, which will be discussed later in this section.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
The Authority Reply
lives are cut short. In contrast, this thesis does not provide a basis for exonerating God for decreeing that humans endure pain or become evil. It fails to do so because being in pain seems to consist of more than a removal of the divine gift of painlessness, and being a rapist seems to consist of more than a removal of the divine gift of not being a rapist. To extend the authority reply to cover the occurrence of such cases, a more uncomfortable principle must be used. The most straightforward way of extending this defense to cover sins such as rape and torture would rely on the claim that agents may permissibly treat that which they possess as they wish, or at least that the only constraints on such action would be consequentialist ones. In particular, most divine determinists utilizing this strategy must rely on the following claim: Strong Authority Thesis (SAT): In virtue of the authority relation God bears to humans, God may permissibly determine that humans endure undeserved harms of immense intensity, including harms of both physical and mental torment, as long as there is good reason to do so.
Determinists will likely wish to add that the authority relation God bears to creatures is necessarily unique and, therefore, no human may permissibly act as God does. For this reason, God, unlike Alice, may permissibly bring about the pain and suffering, both emotional and physical, of human beings. If one thinks that the harms done to evil-doers are undeserved, then the SAT may be interpreted as saying that God may permissibly bring it about that human agents develop deeply vicious characters, creating lives that are tragic both for themselves and for those around them. Once the SAT is accepted, the divine determinist has the makings of a reply to the blameworthiness objection. Alice and God are relevantly different, even in the acts by which they determine other agents to sin, because God alone has the authority to treat humans in this way. Even granting the SAT, God would still need good reasons for creating sinful agents, but the search for such reasons would bring us beyond the blameworthiness objection and into the problem of evil. The SAT may be necessary for an authority-based response to the blameworthiness objection, but it will seem to many to be deeply problematic. Perhaps we may accept that merely allowing such wrongdoing and suffering may be permissible in certain cases, but actively bringing about such acts will seem, at least to some, to be significantly worse. Some may contend that the SAT not only pushes the bounds of credence, but is based upon epistemically problematic and deeply offensive
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
assumptions. In proposing the SAT, the determinist would be allowing for the permissibility of a wide array of harms; indeed, it covers all harms any human has ever suffered. Presumably, when the permissibility of harms is under discussion, the particular character of the harms in question is of central and ineliminable importance. Thus, it seems, any evaluation of the permissibility of some harm presupposes a deep understanding of the harm in question. Of course, no determinist – indeed, no human – has such an understanding of the wide array of harms people have suffered. To discuss the permissibility conditions of an action which one hasn’t experienced, and thus an action which one doesn’t fully understand, seems foolhardy. What is more, it bespeaks an offensive level of presumption, a presumption that one can fully comprehend the anguish endured deep in the hearts and minds of those who suffer – a presumption that one can, from brief armchair reflection, “know what it is like” to be a child suffering silently under relentless abuse, or a political prisoner who has just witnessed a beloved parent mercilessly beaten to death, or a dying elderly laborer who has known nothing but a life of cruel slavery. Such a presumption, it might be charged, cannot possibly hope to support the divine determinist position; it does little more than reveal that its defenders are willing to defend the innocence of a seemingly malevolent God at all costs. I am sympathetic to elements of this response. I think that, as Eleonore Stump () has noted, any discussion of God’s allowance of suffering must be sensitive to the deep and often opaque lived experiences of those who suffer the most. Nevertheless, it is important to note that an affirmation of the SAT need not presuppose that the suffering of others is fully understood. Instead, a defender of the SAT would do well to note that the academic discipline of ethics – or at least wide areas of it – must accept that judgments of permissibility may be made without a full understanding of the lived experiences of those involved in particular acts. Thus the defender of the SAT may follow a long tradition of ethicists in claiming that admittedly limited first-person experience of particular sufferings is not a total impediment to grand ethical theorizing. Moreover, the defender of the SAT might note that the blameworthiness objection itself relies on the claim that the omnipotent, omniscient, transcendent source of all being is relevantly similar to a slightly weird and fantastical human named Alice. If first-person knowledge of the internal lives of agents is a prerequisite for ethical judgments, then those proposing the blameworthiness objection, while having no knowledge of what it is like to be a transcendent divine being, ought to remain silent as well.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Other Replies
Notice that this rejoinder is not an objectionable appeal to divine mystery. The objector notes that in assessing the permissibility of an action, we must presuppose an intimate knowledge – perhaps even firstperson knowledge – of what it is like to be particular agents. But if this is so, the objection continues, we cannot say how God may permissibly treat humans because we lack intimate knowledge of what it is like to suffer various sorts of evils. Such an objection, if allowed to stand, would indeed force us to reject the SAT. Just as surely, it would force us to reject the blameworthiness objection itself, since such an objection purports to tell us what is permitted to another being we have no intimate knowledge of: God. The point here is not to defend the plausibility of the SAT by appeal to a divine mystery. Instead, the objection itself supposes a sort of epistemic bar that must be reached before we may make moral pronouncements – a bar set so high that not only does the authority reply fail, but so too does the blameworthiness objection. In the end, the plausibility of the authority reply depends upon the SAT. Many divine determinists will be comfortable endorsing it, while others may find that accepting it would be a last resort, and the necessity of doing so would constitute a cost for divine determinism. In the next section, I suggest that the authority reply may be used even by those unsure of its success, if it is combined with other possible replies. If this is so, then it should be welcomed even by those who are not fully convinced of the SAT.
Other Replies In the previous sections, I considered two replies in some detail. It is not possible to consider each possible reply at such length. In this section, I consider five additional replies that the divine determinist might utilize; in each case, I will limit myself to indicating the basic line of reasoning each reply makes. The first three replies depend upon the acceptance of a particular general ethical framework. The following two replies may be incorporated into more than one such framework. Appeal to Consequentialism The simplest way to deal with the blameworthiness objection is to embrace a thorough-going consequentialist ethic. Recall the formulation of the blameworthiness objection that appeals to the case of Alice, who, through technological innovations, modifies whole hosts of infants to act in various
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
ways, including various evil ways. That formulation of the argument began with the intuition that Alice’s action was wrong, independent of the later consequences of her action. Once a thorough-going consequentialist framework has been adopted, this intuition must be rejected. If it is, then the blameworthiness objection cannot even get off the ground. Like many of the other responses investigated in this section, this approach will not immediately resolve the more general problem of evil. The problem of evil – whether in its logical or evidential versions – avoids any reliance on the claim that some acts are impermissible irrespective of the consequences. For this reason, the consequentialist will find no premise that can be breezily dismissed. A thorough evaluation of the merits and difficulties of consequentialism is clearly beyond the scope of the present work. Nevertheless, it is important to note that at least some theists have been wary of adopting it. Thus, although this view is popular among ethicists, and it provides a clean response to the blameworthiness objection, some divine determinists will be unwilling to embrace it, even if doing so would help them defend their position. Appeal to Teleology Another reply to the blameworthiness objection turns on the idea that oughts are intimately related to a subject’s end. W. Matthews Grant discusses this claim as follows: “It makes sense to say that a substance ‘ought’ to perform certain activities only on the supposition that those activities are needed, either instrumentally or constitutively, for the substance to attain its end” (, ). Moral oughts, being a species of oughts taken broadly, are likewise constrained. The moral oughts governing human action are just those precepts (or some subset of them) that should be followed if human agents are to attain their end. To discover whether God has violated some moral precept, we must first know what the end of such a being is. Only then will we be able to investigate whether, by acting in some way, God’s end has been put into jeopardy. According to the view under consideration, God’s end is nothing other than the divine existence. The key idea is that God is self-sufficient and perfectly fulfilled; no creaturely states of affairs affect God’s fulfillment
I chose to highlight this formulation because I think it is the most persuasive, but the other formulations also rely on the claim that some acts are impermissible irrespective of the consequences.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Other Replies
in any way. We may then ask, does God’s act by which murderers are determined to commit their crimes endanger the divine existence? Obviously not. Indeed, no act, creaturely or divine, could do so. Thus, even granting divine determinism, no human atrocity we could discover could provide any evidence whatsoever that God is blameworthy. This response handily dispatches each of the arguments we considered earlier and also provides a nice error theory explaining why some go astray in thinking that determining sin would constitute a divine fault. First, we are tempted to say it is intuitively plausible that such determining would be blameworthy because we think in terms of attaining our own ends; if we had determined so much horror and evil, it seems very likely that we would endanger our own flourishing. But since God has a different end than ours, the same considerations simply do not apply. The same line of reasoning dispatches the argument from analogy with Jack and Alice. Finally, the great harms caused by, for example, murdering someone, provide reason to think that murder is wrong. But this is so, according to the view under consideration, only because by causing such a great harm, we move away from our own flourishing. Killing innocent humans does not endanger a lion’s flourishing, so it is false that a lion should avoid killing us. Similarly, God’s end is not endangered by determining the great harms that arise in virtue of determining human wrongdoing, so it is false that such determining ought not to have been done. This reply quickly dispatches the blameworthiness objection, but two distinct worries may arise for some theists. First, some theists will think that ethics should disentangle itself from the Aristotelian tradition and its reliance on ends. For those of this opinion, this reply will be a nonstarter. Second, some theists may think that examining an agent’s end is crucial to understanding it, but nevertheless assert that we cannot understand claims of divine goodness in a way that reduces to the mere fact of divine existence. Instead, God’s voluntary action, perhaps especially in that action related to created persons, must be examined to first assess and then fully understand divine goodness. Appeal to Divine Command Theory According to some theists, moral oughts derive from divine commands. As McCann explains, “the sinfulness of sin consists in our placing our own projects above God’s decrees, by defiantly willing what he has commanded us not to do” (, ). He continues, “But then it turns out to be impossible for God to sin. The reason is simply that no one can be in
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
moral rebellion against himself, for no one has moral authority over himself” (, ). If we add that only sins are deserving of blame – at least in the sense relevant to the current discussion – then the blameworthiness objection is readily dispatched. Once more, not only does this view provide a clean response to the problem of divine blameworthiness, but it also provides the basis for a satisfactory error theory concerning why we might think that God would act wrongly by determining sin. We have tended to internalize God’s commands to us (whether we recognize them as divine commands or not) such that we think that anyone performing such actions would act wrongly. This is why we think that causing such harm would constitute an evil action, even for God, and why we think that there is no relevant difference between the cases of Jack and Alice on the one hand and God on the other hand. Theists may plausibly suppose that this internalization is normally beneficial: By forming themselves to see the world through a lens of divine commands, they presumably will act in keeping with them more often, readily, and easily. But, if this line of reasoning is cogent, then it could also cloud judgments concerning the nature of divine goodness. This reply, like the two before it, makes quick work of the blameworthiness objection, but it also inherits all the problems that divine command theory has faced. Given the reticence with which many theists have approached this view, it is likely that its use in replying to the blameworthiness objection will win it few new converts. Nevertheless, many theists have endorsed this view, both in the past and today, and their moral framework is ideal for responding to this objection. The three replies just discussed function by adopting a particular moral framework that allows for a ready-made solution to the blameworthiness objection. The next two responses take what is perhaps a more difficult road, arguing that the solution to this objection cannot be found merely by examining basic ethical commitments. God Is Not a Moral Agent Another response to the blameworthiness objection, which is equally useful to theists who endorse divine determinism and those who do not, begins by denying that God is morally good at all. According to this view, God is simply not the kind of being susceptible to moral evaluation in the first place (Davies , ; Holland ). If this is true, the problem of divine blameworthiness arises only because of confusion. God is no more an appropriate target of moral blame (or moral praise) than a cat, and
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Other Replies
thus no defense of divine actions is needed or even appropriate. Indeed, to treat God as in need of an excuse for this or that action is to fall prey to a category mistake: [A]n end-of-term report on his moral standing (whether favorable or unfavorable) would simply fail to engage with what he is. It would be confusing the Creator with some of his creatures. It would be grounded in a category mistake, since it would take God to be something that he could not be. Suppose I complain that a particular tennis player is not scoring many goals. You would rightly reply that tennis players are not in the business of scoring goals . . . By the same token, so I am arguing, we have reason to deny that God should be thought of as morally good or bad and we therefore have reason to deny that God is a moral agent. (Davies , –)
According to this view, God is exceedingly good, but this does not reduce to or even entail His being well behaved. Instead, we might highlight other kinds of goodness. For example, God is good insofar as it is impossible for such a being to be improved (Davies , –). Furthermore, since God is the source of all goodness in creation, and effects always resemble their causes, we have good reason for thinking that the creator must be good (Davies , –). It is worth dwelling briefly on how this position relates to the appeal to divine command theory and the appeal to teleology. It seems that divine command theorists may not only accept the claim that God is not a moral agent, but also have an easy explanation for it: A moral agent is an agent to whom a divine command has been given, but no divine commands have been given to God. Those who endorse the appeal to teleology may claim that God is not a moral agent, but need not do so. Instead, they may claim that God is a moral agent, just one governed by very different moral precepts. Although some overlap might arise between those who endorse this view and those who appeal to either teleology or divine command theory, it is important to note that one can endorse this view without appealing to any particular moral framework. Those who do so can simply contend that calling God a moral agent involves an unjustified anthropomorphizing of the divine. Like each of the answers discussed before it, this view handily dispatches any worries associated with divine blameworthiness. Moreover, we can easily identify why we were led to think that there was a substantial problem here: We have become too accustomed to overly anthropomorphizing the divine. Once we have remedied this failing, the problem of divine
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
blameworthiness is not so much solved as dissolved. Although the success in responding to the blameworthiness objection is clear, this response’s costs will likely keep many theists from adopting it. Most importantly, most theists seem to think that God is well behaved, and in fact, that divine behavior is to be taken as a model for human conduct. An Appeal to the Unusual Nature of Divine Causality The final reply I wish to consider here once again returns to the claim that divine causality is radically different from human causality. As noted earlier, this claim is sometimes invoked to explain why physical determinism would undermine free will, but divine determinism does not. It is also sometimes used to explain away the claim that a determining God is blameworthy for bringing about human sins. This sort of reply can take one of two forms, both of which begin by recalling that the blameworthiness objection relies on the claim that the case of Alice is relevantly similar to that of God. The first form of this reply notes that we should have no confidence in this claim, since we are ignorant of much of the divine nature. This reply does not seek to offer a substantive reason for thinking that Alice and God are relevantly different, but instead suggests that the radical difference between divine and human causality, as well as our deep ignorance about the nature of the former, give us reason to be agnostic about whether they are relevantly similar. I will call this the “skeptical form” of this reply, since it depends on the claim that we should be skeptical of our ability to determine whether God and Alice are relevantly similar. The second form of this reply suggests that we have reason for thinking that they are relevantly different. I will call this the “ambitious form” since it sets itself a more ambitious goal. This form has recourse to the now familiar metaphor of creation as an act of novel writing. McCann () notes that when we think of human agents causing others to sin, we should think of them as influencing preexistent people in such a way that they are led toward evil. Nothing similar happens in the
Claiming divine action as a model for human conduct is limited to neither philosophers nor those of Abrahamic religions. The popularity of the phrase “What would Jesus do?” provides evidence of the former, whereas Euthyphro’s appeal to Zeus’s behavior in defense of his own provides evidence for the latter. See Euthyphro in Plato (). Such determinists may believe that the cases must be different, since they believe in both divine goodness and divine determinism, but correctly recognize that it would be question-begging to use these considerations to respond to the blameworthiness objection.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Other Replies
case of a determining God. Instead, we should think of God as writing a novel in which characters are created in their acting. God does not cause a person to sin, but instead creates the individual sinning. If we continue with this metaphor, we might say that although Alice is similar to God insofar as she determines people to act as they do, she does not create them as God does. As McCann would have it, Alice is not the cause of anyone’s existence, since this prerogative is reserved for God alone. Thus, she causes people to murder and rape. God, by contrast, creates people murdering and raping. This, the reply continues, is a relevant difference. Both the skeptical and ambitious forms of this response have challenges. The skeptical form is subject to the worry, familiar from other skeptical arguments, that it is difficult to isolate the skepticism to targeted areas. In particular, defenders of the skeptical form must be able to explain why we should be skeptical about this issue, but confident about other areas of so-called natural theology – among them, whether divine causality is deterministic at all. The ambitious form faces a different challenge: It is difficult to see why the dissimilarity between Alice and God is a difference that makes a difference. It is relatively easy to see that the ambitious defense points to an important difference between Alice and God, but it is unclear how this difference relates to the moral norms that govern their behavior. The metaphor of the novel appears to help only if we take it too far. It is certainly true that Arthur Conan Doyle does nothing impermissible in creating Moriarty; it is also true that it would be misleading to say that Doyle caused him to sin. But this is not due solely to Doyle’s role as a creator – a role that is analogous to that of God – but to Moriarty’s fictional existence. If Moriarty somehow took on a full-blooded existence (as he does in some science-fiction stories), and Doyle was responsible for determining this being’s crimes, then presumably Doyle would be on the hook for them. Thus, although God’s role might be analogous to that of Doyle, it is far closer to that of a Doyle whose creations come to life than to the historical author who is innocent of Moriarty’s behavior. A mere appeal to a radical difference in causality or to the author metaphor is insufficient to establish the key claim of the ambitious reply. Thus, both the ambitious form and its less ambitious sibling face challenges, though this is not to say that they are inevitably doomed to
See the episodes “Elementary, Dear Data” and “Ship in a Bottle” of Star Trek: The Next Generation, in which Moriarty achieves sentience. There is no indication in these stories, however, that his actions are determined.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Blameworthiness Objection
failure. The blameworthiness objection’s claim about the relevant similarity between Alice and God is made breezily if not hastily, given the proposed differences between the two agents. Demanding closer inspection of this claim, as defenders of this reply do, is surely important, though this inspection has thus far been inconclusive.
Toward a Unified, Noncommittal Strategy In the preceding section, I discussed a large number of possible replies to the blameworthiness objection. Each of those might be endorsed by a particular determinist to respond to this objection, but many determinists may be unwilling to commit to any of them. Alternatively, a theist may be confident in one such reply, but even more confident in divine determinism. In such cases, determinists should work toward a unified, albeit noncommittal strategy of response. The determinist might utilize the previously described responses without committing to any in particular. Instead, the supposed determinist could propose what I will call the disjunctive response to the blameworthiness objection. According to this response, the individual plausibility of each of the preceding responses might not be particularly high. Nevertheless, the plausibility of each individual response seems to be independent of the others. In some cases, in fact, the plausibility of one reply detracts from the plausibility of another. Consider, for example, a divine determinist considering whether some form of consequentialism should be accepted. This determinist may assign the truth of consequentialism a relatively low credence level, perhaps as low as .. This low credence level may be a consequence of the credence levels that the determinist assigns to a classically teleological position and to one grounded upon divine commands. Suppose our determinist assigns a . credence to each of these three positions. In that case, no single position is successfully poised to ground a response to the blameworthiness objection. Nevertheless, if the determinist’s credence levels are justified, and if the positions are mutually exclusive, then it would be reasonable for the determinist to propose the following reply to the blameworthiness objection: Disjunctive Proposal: An ethical framework based solely on consequences, classical teleology, or divine commands is true. Disjunctive Conditional: If Disjunctive Proposal is true, then the blameworthiness objection fails.
The determinist uncommitted to any particular response need not even have complete confidence in Disjunctive Proposal. Given the dialectical
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Toward a Unified, Noncommittal Strategy
situation, the burden seems to be upon those making the objection to show why this response is implausible. The particular replies based upon general ethical frameworks are especially useful in constructing a disjunctive reply because they are plausibly mutually exclusive. Nevertheless, the plausibility of the disjunctive proposal will increase if it is modified to include other possible responses (and, of course, Disjunctive Conditional should be modified accordingly). Thus, the proponent of the disjunctive reply should endorse the following as probable: Disjunctive Proposal*: An ethical framework based solely on consequences, classical teleology, or divine commands is true, or God avoids wrongdoing because human sins fall outside of the divine intention, or God’s authority makes determining human sins permissible, or God is not a moral agent, or the nature of divine causality renders Alice and God relevantly dissimilar.
This proposal, it seems to me, is not unreasonable. If I am correct about this, then the divine determinist may reasonably resist the force of the blameworthiness objection. Of course, even if I am correct, the blameworthiness objection is important and should not be ignored. After all, many theists will find each of the individual replies discussed previously costly. These costs do not magically disappear when individual replies are eschewed in favor of a disjunctive one. If each individual reply brings a cost, then the disjunctive proposal has its own price tag. We might think of this cost as being itself disjunctive. To avoid the blameworthiness objection using the disjunctive solution involves accepting that either the implausible implications of one of the three general frameworks are true, or the implications or implausible elements of one of the more particular replies (such as the truth of the SAT) is true. The blameworthiness objection, then, is powerful insofar as it lays bare additional costs of divine determinism.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.006
Divine Determinism and the Free Will Defense
Introduction
Of the many challenges that theism faces, the problem of evil is often considered the most pressing. According to many theists, the following three theses are true: i. God is omniscient. ii. God is omnipotent. iii. God is perfectly good. To these three, we can add the following, which is clear from our experience: iv. There is evil in the world, The problem of evil is the problem of reconciling the existence of evil with the existence of a being who is at once omnipotent, omniscient, and perfectly good. Arguments for atheism from evil come in many shapes and sizes, and are popularly distinguished into two types. The first, which are called logical problems of evil, claim that there is a logical inconsistency in theses i–iv. In Michael Almeida’s words, this sort of argument “advances the strong modal thesis that, necessarily, God does not coexist with evil” (, ). The evidential argument from evil refrains from such strong claims and instead seeks to show that the presence of the sorts or amount of evil in the actual world makes the existence of God unlikely. This chapter will focus almost exclusively on the former, although the latter will briefly be considered as well. It is widely thought that the so-called Free Will Defense (FWD), the most famous version of which was created by Alvin Plantinga (), has successfully shown a defect with the logical problem of evil. It is also often thought that this defense is unavailable to
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
The Logical Problem of Evil and the Free Will Defense
determinists. In particular, it is commonly thought that this defense relies upon a libertarian view of free will, which itself requires the absence of determinism. Some have argued that this relationship represents a mark against divine determinism. Other philosophers, however, have argued that this matter is more complicated than it appears. They maintain that the FWD is available to divine determinists, too. If this is so, then whatever other troubles divine determinists face, they need not fear the loss of this seemingly powerful defense. In this chapter, I examine whether divine determinists should fear the loss of the FWD by considering the work of three authors: Steven Cowan, A. A. Howsepian, and Jason Turner. The work of each seems to offer hope to determinists, and I will seek to determine whether that hope is false. I will begin by providing a very brief overview of the logical problem of evil, together with a similarly brief sketch of the FWD. After presenting these preliminary considerations, I will consider different desiderata seemingly sought in the FWD by different authors. Finally, I will move to the heart of the chapter – the investigations of the proposals of Howsepian and Turner.
The Logical Problem of Evil and the Free Will Defense Formulations of the logical argument from evil share a conclusion: The existence of evil is incompatible with the existence of God. They differ, however, in how they defend this conclusion. John Mackie, whose work becomes a key point of reference for Plantinga’s examination of the problem of evil and eventual formulation of the FWD, writes: In its simplest form, the problem is this: God is omnipotent; God is wholly good; 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. (, )
See Pawl and Timpe (), Plantinga (), Sennett (), van Inwagen (), and Walls (). Some have argued that libertarianism is compatible with particularly strong views of divine governance, which seem to fall within divine determinism as I have defined it. See, for example, McCann (). Perhaps this disagreement is ultimately a matter of classification rather than substance, as even McCann suggests. Nevertheless, even if we do allow that libertarianism is compatible with divine determinism, it seems that acceptance of divine determinism will nonetheless threaten the success of the FWD, since, even in views like McCann’s, God could have ordained that all humans refrain from all evil. See, for example, Pawl and Timpe (, ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
Other authors suggest more structured versions of this argument. In his overview of the problem of evil, Michael Tooley () offers the following as an example of a concise formulation: . If God exists, then God is omnipotent, omniscient, and morally perfect. . If God is omnipotent, then God has the power to eliminate all evil. . If God is omniscient, then God knows when evil exists. . If God is morally perfect, then God has the desire to eliminate all evil. . Evil exists. . If evil exists and God exists, then either God doesn’t have the power to eliminate all evil, or doesn’t know when evil exists, or doesn’t have the desire to eliminate all evil. . Therefore, God doesn’t exist. Almeida presents the argument in a slightly different way. He writes: The argument is straightforward, even if not always presented in a straightforward way. There exists a possible world w that is morally perfect and naturally perfect. Since all agree that, necessarily, an omniscient being can do anything that is metaphysically possible, necessarily, an omniscient being can actualize w. So, necessarily, an essentially omnipotent, omniscient, morally perfect, and necessarily existing being actualizes w. Therefore, necessarily, there are no instances of evil. But of course there are, indisputably, instances of evil. Therefore, God does not exist. (, )
The FWD can be understood, in its simplest form, as providing a reason to reject Tooley’s sixth premise. According to this defense (and, in general, “greater good” defenses), an omnipotent, omniscient, perfectly good being may well bring about a world with evil in it, if evil is a precondition for some great good. The FWD proceeds by suggesting that exercises of free will (and goods for which such exercises are a precondition) are especially valuable. In particular, free choices to act in keeping with the moral law – what we might call upright morally significant choices – are of great value. Prior evil acts need not be a prerequisite for such choices, although in some cases they might be, but their possibility does seem to be such a precondition. In short, the key claim is that creating a world without any evil requires, or at least may require, eliminating the human capacity to choose freely. But once this is taken away, so too is the possibility of upright, morally significant choices. The FWD does not require that there be no possible worlds in which humans act freely and in which there is no evil. In fact, Plantinga sometimes suggests that there may be such worlds, and indeed many
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
The Logical Problem of Evil and the Free Will Defense
possible worlds with, on balance, a better ratio of good to evil. The problem, according to the FWD, is that these possible worlds might be outside of God’s ability to bring about. Plantinga identifies this claim as of central importance: “The heart of the Free Will Defense is the argument that this is indeed so” (, ). Why might we think that there are worlds that an omnipotent being cannot actualize? The idea, in short, is that a particular human free action is not the sort of thing that God can determine to occur. Although there may be possible worlds in which humans do use their own agential powers to avoid all evil, God cannot keep them from such evil while still offering them the possibility of making upright morally significant choices. Why think that the FWD is not open to divine determinists? The defense turns upon the idea that it is impossible for God to actualize a world where agents are free and there is no moral evil. Plantinga writes: “The Free Will Defense obviously presupposes a libertarian or incompatibilist conception of freedom. If freedom were compatible with causal determinism, then God could have his cake and eat it too: He could create significantly free persons and cause them always to do only what’s right” (, ). I have largely been assuming that divine determinists will seek to preserve the existence of human free will, and so will embrace compatibilism. Alternatively, they could embrace incompatibilism and admit that humans lack free will. In either case, however, it seems that the FWD will be off the table. Before moving to proposals for a determinism-friendly FWD, I want to bring to light what I perceive to be a difference in opinions concerning the desiderata for a successful defense. To highlight this difference, I will briefly recap an exchange that occurred in a recent series of articles between Timothy Pawl and Kevin Timpe on the one hand, and Steven B. Cowan on the other hand. Pawl and Timpe kicked off the argument that led to further discussion with what seems to be an innocent observation: It is hard to see how the Free Will Defense would go on the assumption of compatibilism. On the supposition that free will is compatible with determinism, God could actualize the good of free will (as well as those additional goods which logically require that some agents have free will)
There are important questions about why God cannot make use of omniscience to determine which possible world to create, and in particular why God could not foresee which world contained no evil and actualize that one. These problems are not directly relevant to the present discussion, so I have chosen to set them aside. Note that Plantinga once thought that his defense could be utilized by compatibilists, but later changed his mind. See Plantinga (, –) for discussion.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense without the possibility of moral evil by determining all free creatures never to do evil. Thus the Free Will Defense to the logical problem of evil is rendered impotent. (, )
This claim fits well with what Plantinga himself said in the passage quoted earlier, and seems to voice a widely held view. Cowan, however, thinks that the FWD is open to compatibilists. He notes that, as a defense, the FWD is not in the business of providing God’s actual reasons for allowing evil, but instead of showing that theses i–iv are not inconsistent. He writes, “In order to utilize the FWD, one must assume a libertarian notion of freedom at least as a working hypothesis. However, I do not believe that this admission would disallow me, a compatibilist, from utilizing the FWD and claiming that it actually works – as a defense” (Cowan , ). This seems odd. How can a compatibilist utilize a libertarian claim in any argument whatever? It would seem that any response to the logical problem of evil would take the form of an argument, and that in utilizing an argument one thereby at least implicitly and tentatively asserts the truth of each statement it contains. Cowan seems to think that the logical problem of evil has a notable aspect that allows for an unusual response. This problem of evil claims that theses i–iv are inconsistent. To show this is false, it is sufficient to show that i–iii, together with other propositions (which are not themselves inconsistent with i–iii), entail iv. Cowan claims that there is no need for these additional propositions to be true. The point of this defense, he recalls, is simply to show that i–iv are not inconsistent, not that each of i–iv is true, and so the truth-values of the propositions utilized is immaterial. He concludes, “Compatibilists can happily embrace the FWD as providing a set of propositions that make coherent God’s permission of evil, even though (as a compatibilist believes) one or more of those propositions is actually false” (Cowan , ). Perhaps, though, if the additional propositions proposed are believed to be necessarily false (and, crucially, compatibilists tend to believe that incompatibilism is necessarily false), then the defense may fail, as Cowan suggests (, ). Still, he maintains, the compatibilist has two options in utilizing the FWD: First, “say that the FWD is successful insofar as one lays aside the question of
The popularity of such a view is hard to gauge definitively, but A. A. Howsepian () and Jason Turner () – both of whom think that some compatibilists can make use of the FWD – judge that many philosophers think compatibilists cannot avail themselves of it. Cowan sets up the initial propositions slightly differently, combining my i–iii into a single proposition, but the substance is the same.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Determinism, Middle Freedom, and the Free Will Defense
whether libertarianism or compatibilism is the better account of free will, or [second,] say that, for all we know, libertarianism is true, and thus the FWD shows that [i–iv] are compossible for all we know” (, ). While one might consider the disagreement between Pawl and Timpe on the one hand and Cowan on the other hand to be over whether the compatibilist may utilize the FWD, I think it may be useful to think of the debate as actually a disagreement about a desideratum of such a defense. Pawl and Timpe seem to suggest that one desideratum of a successful FWD is that it utilizes propositions that are not denied by the one giving the defense; Cowan suggests otherwise. I suspect that many will think that if determinists can utilize the FWD only in the two ways that Cowan suggests, then the acceptance of divine determinism does, indeed, incur a cost in this regard. For this reason, more ambitious proposals for a compatibilist version of the FWD are worth investigating. Consideration of two such proposals will occupy us for most of the remainder of the chapter.
Determinism, Middle Freedom, and the Free Will Defense The first view to consider comes from A. A. Howsepian. In a pair of articles, Howsepian first lays out a unique compatibilist account () and then argues that those who accept this account may still make full use of the FWD (). To understand Howsepian’s account, which he calls the Theory of Middle Freedom (TMF), we must first turn to counterfactuals of creaturely freedom (CCFs). A CCF is a proposition stating what a particular agent would freely choose to do in certain circumstances. Although CCFs are flexible in that they may be utilized by philosophers advocating a wide variety of views on free will, Howsepian is clear that he intends them to be understood in terms of agents possessing free will as understood by agent-causal libertarianism. Such CCFs are generally thought to take the following form: “If libertarian free agent S were in circumstances C at time t, then S would (incompatibilist) freely perform action A in C at t” (Howsepian , ). Howsepian’s view is unique insofar as it invokes the importance of such CCFs, but does not require free actions to be agent-caused. In fact, the
Cowan doesn’t think so because he doesn’t think that the FWD delivers the desideratum Pawl and Timpe seek even if incompatibilism is true, because one of the propositions it utilizes is “almost certainly false.” But exploring this claim would take us too far outside the scope of the present work. For his argument for this claim, see Cowan (, –).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
TMF does not depend most crucially upon a causal relation at all, but instead upon a noncausal one. According to this view, agents act freely, not in virtue of agent-causing acts, but instead in virtue of the fact that they would have agent-caused the acts in these circumstances, as noted in the relevant CCF. More precisely: Principle of Middle Freedom (PMF): Any determined voluntary act A that is performed by an agent S at time t in circumstances C in world W is an act that is, ultimately, under S’s control and, hence, is an act that S freely performs, if and only if (a) S possesses agent-libertarian powers at t in C in W, and (b) if S were permitted to exercise his agent-libertarian powers in C* at t in W* (where ‘C*’ denotes circumstances in a possible or non-actual world W* that are at least ‘relevantly similar’ to C, and where W* is a world closer to W than any other world in which S is permitted to exercise his agent-libertarian powers at t) S would agent-libertarian freely perform A at t in C*. (Howsepian , –)
We can see, then, that the actual causal history of an agent’s action is not as relevant as might otherwise be thought. In particular, an act may be free even if it were caused by some fantastic mad neuroscientist. Crucially, whether such a manipulated act is free depends upon the relevant counterfactual; that is, it will depend on whether the manipulated agent would have acted similarly in the absence of manipulation. Although Howsepian develops this view with greater depth than we have explored here, this brief overview is sufficient for our purposes, for we are now in a position to see why compatibilists who adopt this view may utilize the FWD. Crucially, the FWD claims that there are possible worlds that God cannot simply set out to create. Instead, these worlds include agents who have their own robust abilities – abilities whose exercise in some way fall outside of God’s control. This key claim seems to presuppose incompatibilism, for seemingly only incompatibilists can explain why God cannot simply determine that each creature always freely acts in keeping with the moral law. Nevertheless, compatibilists like Howsepian can explain this as well. On their view, God can determine agents to freely
The TMF can be used to respond to both the Consequence Argument and manipulation arguments, which were discussed in Chapters and , respectively. It can respond to the former by noting that although the occurrence of the action itself is not under the agent’s control, whether it is free or not is, since this depends upon the truth of a contingent CCF, which, Howsepian claims, is under the agent’s control. It can respond to the latter with a similar claim: Although the manipulator determines the agent to perform the act, it is not up to the manipulator, but rather to the agent, whether it is free. I delayed discussing this view until this chapter, since it is rather idiosyncratic and brings with it significant costs, as I note later. Consequently, it not the most attractive way to respond to those arguments discussed in earlier chapters.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Determinism, Middle Freedom, and the Free Will Defense
perform good actions only if those creatures would indeterministically agentcause them in similar circumstances if they were permitted to exercise their powers. But whether agents would do so is something that is not up to God, so God is constrained in world-making by the set of true CCFs. Thus, depending upon which CCFs turn out to be true, it is possible that, given the TMF, God is not able to create a world populated with many humans whose acts are always in keeping with what morality requires. Two worries – rather substantial ones, I think – arise with the TMF. Even if these objections can be met, some aspects of the TMF will make it unattractive to many divine determinists. The first worry is that the position takes upon itself the problems that plague Molinism. These problems have been debated elsewhere at great length, but one continuing problem should be highlighted: Molinists are committed to the existence of true CCFs, but it is unclear how their truth might be grounded. Although Molinists have expended considerable effort in attempting to solve this problem, I join many others who remain unconvinced that this issue has been satisfactorily resolved. The second problem for defenders of the TMF concerns agential control. According to Howsepian, although free agents need not possess control over whether they act or which acts they perform, they do need to possess control over the valence of their actions – that is, over whether those actions are performed freely. Control over an action’s valence seems to constitute the sole control condition on free actions, on this view. For this reason, the theory’s ability to explain how agents control the valence of their actions is of central importance. Unfortunately, it falls short in this regard.
To be complete, this defense can invoke a doctrine of transworld depravity similar to that suggested by Plantinga (). Since transworld depravity is no more difficult to defend based on the divine determinist view, given the TMF, than it is in typical libertarian models, I leave this issue to the side. As Howsepian notes, the TMF does not involve any commitment to theism. For this reason, adoption of the TMF alone will bring with it only some of the problems of Molinism. For example, it need not worry about how anyone, even a divine being, could know the truth value of a CCF. If the theist adopts this theory as a way of reconciling the existence of a determining God with the presence of evil, then, of course, the rest of the problems plaguing Molinism will be brought along with it. An enormous amount of recent work has addressed this and other problems with Molinism. For a discussion providing an overview of much of it, see Perszyk (). This shares some similarity to what Otsuka () calls the Principle of Avoidable Blame, which he recommends as a replacement for the Principle of Alternative Possibilities. Molinists in general have faced problems accounting for how – and indeed whether – an agent controls the truth-value of relevant CCFs, and, relatedly, whether true CCFs are explanatorily prior to the actions they concern. For a discussion of these challenges, see Adams (), Flint (),
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
Consider, first, that control over the valence of an action is achieved when a noncausal relation obtains between an action and a true CCF. Howsepian is explicit that this relation grounds freedom in virtue of grounding control. On his way to establishing the sufficient condition of freedom, he provides what he calls the Preliminary Principle of Middle Control: “I propose (to a first approximation) that any determined voluntary act A that is performed by agent S at time t in circumstance C is an act that is, ultimately, under S’s control if and only if S would have performed A at t in C if S were permitted agent-libertarian freely to perform A at t in C” (Howsepian , ). We might think of it as a relation of a particular sort of transworld similarity: Both the action performed in the actual world and the one referenced in the true CCF are of the same type, although key features of the world in which the action is performed and the world(s) referenced by the CCF are quite different. The control condition seems to rest on the existence of three things: the true CCF, the act that the agent performs, and the relation of transworld similarity that obtains between them. Plausibly, so as to meet the control condition, agents must have control over at least one of these three things. Before diving into each of these three elements, it is important to get a sense of the sort of control at issue, since control comes in many varieties. Howsepian is clear on this matter: “The Theory of Middle Freedom does not entail that my doing A is under my control (in Fischer’s regulative sense), but only that my doing A freely – and, correlatively, my being blameworthy for A – is under my control (in that same sense)” (, ). Howsepian here references John Martin Fischer’s account of regulative control, which involves an ability to determine which of multiple possible alternatives becomes actual (Fischer ). Regulative control is contrasted with guidance control, which concerns whether an agent guided a particular event in coming about, regardless of whether alternative possibilities were available before the control was exercised. Defenders of the TMF must explain how agents settle the valence of a particular action, the performance of which is not under their (regulative)
and Hasker (, , , ). The TMF faces a worse problem in this regard for two reasons. First, ordinary Molinists can deny that agents do control the truth-values of CCFs; see Basinger () for discussion. Although such a view may involve problems, it is at least on the table; those who endorse the TMF must assert that agents control the truth-values of CCFs. Second, ordinary Molinists may claim that agents control the truth-values of CCFs through the actions they perform. While this option is not particularly promising, it seems to me that defenders of the TMF lack even this option, as I will argue later. It seems that they could not be token-identical, since their causal histories are so different.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Determinism, Middle Freedom, and the Free Will Defense
control. At this point, we might look at each of the three elements – the act, the truth of the CCF, and the relation that obtains between them. Howsepian, in the passage just quoted, denies that the act itself is under the regulative control of the agent, and it seems that he must do so. Of course, one might appeal to an account of regulative control and aim to show that such control over actions is not ruled out by determinism. Nevertheless, any attempt to utilize such an account by a defender of the TMF is likely to raise a problem: If such an account is sufficient for regulative control, then there is no need for any appeal to CCFs to account for control, and so the TMF is unmotivated. We may also dismiss the possibility that agents possess control only over whether the relevant relation obtains between the act and the CCF. Since this is a similarity relation, it seems clear that it obtains if and only if the two relata possess particular features. Thus, the only way for agents to control whether the relation obtains is to exercise control over one or more of the features of one or more of the relata. Inevitably, the defender of the TMF must turn, as Howsepian does, to the truth of the CCF itself. Now Howsepian claims that the truth of a particular CCF is a contingent affair, so it seems to be the sort of thing that an agent might control. Moreover, Howsepian claims that true CCFs – at least the ones relevant to the present discussion – “partly comprise one’s individual essence” (, ). Thus, in having control over the truthvalues of certain CCFs, agents possess control over their individual essences. This may seem impossible to some, but Howsepian doesn’t seem to think so. He notes that “the agent herself could not possibly directly and straightforwardly control the truth values of her counterfactuals of creaturely freedom prior to her existence” (, ). This suggests to me that Howsepian thinks agents can do so after their existence, but it is not clear how they could do so. At this point, we are in a position to ask an awkward question: How do agents exercise regulative control over the truth of the relevant CCFs? Before investigating possible answers to this question, I wish to introduce two distinctions. The first is between direct and indirect control. Indirect control=df An agent S possesses indirect control over an entity E if and only if S controls E and does so in virtue of controlling some entity other than S or E. Direct control=df An agent S possesses direct control over an entity E if and only if S controls E and S does not possess indirect control over E.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
Indirect control is a common feature of this world. I possess (at least some) control over my blood pressure in virtue of control I possess over such things as my diet, stress level, and exercise regimen. Direct control is a bit more mysterious, but every account of control must stop somewhere, and so every account of control must appeal to instances of direct control. The next important distinction is between basic and non-basic control: Non-basic control=df An agent S bears a non-basic control relation R to an entity E if and only if S controls E and does so only in virtue of bearing another relation R either to E or to some other entity. Basic control=df An agent S bears a basic control relation R to an entity E if and only if R is a control relation S bears to E, and it is not the case that S bears R to E in virtue of bearing another relation to E or to some other entity. Non-basic control may be either direct or indirect, depending on whether the relation upon which it depends is itself a relation of control. If it is, then the control in question is indirect and non-basic. In contrast, if the control relation obtains in virtue of another relation, but one that is not a control relation, then the control is direct and non-basic. All cases of indirect control are non-basic. With these distinctions before us, we can return to the question of how agents possess control over the truth-values of the relevant CCFs. First, we must ask whether this control is direct or indirect. Ordinarily, it seems, an agent exercises control over the truth-value of a proposition in virtue of controlling something else. For example, the only way I can control the truth value of the proposition “Peter kayaks at : PM on July , ” is in virtue of controlling my actions. In fact, it seems that the only way we might go about controlling the truth-value of a proposition is through controlling something else, since it does not appear that there is any other mechanism by which we might control a proposition’s truth-value. To see this, consider what happens if we suppose agents do possess direct control over the truth-value of a proposition – in particular, of a CCF. We must then ask whether this control is basic or non-basic. If it is non-basic, then we must appeal to some other relation that holds between agents and other entities (whether the CCFs or something else). But what could this relation be? Perhaps it is the relation that agents bear to CCFs in virtue of () the relation obtaining between themselves and their individual essences and () the constitution relation that obtains between the CCFs and their individual essences. If, as we are supposing, the control relation between the agent and the truth of the CCF is direct, then the agent has no
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Determinism, Middle Freedom, and the Free Will Defense
control over either () or (). There is a problem, however. Assuming an agent has no control over () or (), it is unclear how a regulative control relation could obtain in virtue of them. This is especially unclear if we assume that regulative control relations cannot be accounted for within ordinary compatibilist frameworks, since this seems to raise the bar on the level of settling power required for regulative control. It is key, however, that defenders of the TMF maintain that ordinary compatibilists cannot account for regulative control, since otherwise the TMF itself would be largely unmotivated. Suppose, instead, that the direct control an agent possesses over the truth-value of a CCF is basic. If this is the case, then an agent possesses a sort of metaphysically fundamental control over whether a particular proposition is true. This control is not exercised in virtue of the agent’s actions or omissions, nor through volitions or the possession of explanatorily prior features of any kind. Instead, with the assertion that agents control the truth-values of these peculiar contingent propositions, we have reached ontological bedrock. All metaphysical accounts must do so at some point; we may not reductively analyze everything, of course. Nevertheless, to think that a control relation, especially one between an agent and a proposition, could be a candidate for such bedrock is rather mysterious. Moreover, this mystery is additional to that borne by other Molinists and agent-causalists. Other agent-causalists seemingly must claim that the causal relation between an agent and an action is basic and constitutes a control relation. Many have cried foul over the idea that a causal relation could obtain between a substance and an action, and others have questioned whether such a relation might be control conferring, but defenders of the TMF are just as committed to such worrisome claims as other agentcausalists are. Moreover, ordinary Molinists need not say that we possess control over the truth-values of CCFs in a basic way. They may either deny that we control them at all (insisting that since CCFs play no causal role in the production of our actions, they do not eliminate freedom) or contend that we control them in virtue of the actions we do perform or the sorts of people we freely form ourselves into.
They are committed to the idea that a causal relation can obtain between substances and actions since it is key to their account that there are possible worlds in which agents are permitted to exercise agent-causal powers and since, if we are free, we must possess such powers, even if we are not permitted to exercise them. They are committed to the idea that such a causal relation is control-conferring, since it would be preposterous to think that agents that do exercise their agentcausal powers control their actions not in virtue of having agent-caused them, but in virtue of controlling a CCF that describes the possible world they are actually in.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
It seems, then, that the only viable option in defending the TMF is to claim that we exercise indirect control over the truth-values of CCFs. If this is so, then we control these truth-values in virtue of controlling something else. What might this be? The only options seem to be our actions and our habits. We can quickly dismiss the possibility that we control our actions, since the TMF makes clear that this is not so. In turn, since we control our habits through our actions, we can dispatch the second possibility as well. It seems, then, that there is no way, given the TMF, for agents to control the truth-values of CCFs. For this reason, I am inclined to think that the TMF lacks the resources to provide a satisfactory account of agential regulative control superior to that available to ordinary compatibilists. I noted earlier that the TMF takes upon itself all the troubles of Molinism, and I then argued at some length that it is unsatisfactory as a theory of agential control. Although Howsepian seems correct in thinking that defenders of the TMF may make free use of the FWD, I think the problems with the view are so great that the TMF itself must be rejected as an inadequate account of free will. Even if my worries with this view are set aside, I suspect that few divine determinists would be willing to accept the TMF, for it runs counter to some of the motivations inspiring divine determinism in the first place. Divine determinists are motivated to accept their view for many and varied reasons, including that only divine determinism captures the complete sovereignty and aseity of God. Accepting the TMF threatens both God’s sovereignty and aseity. First, it threatens divine sovereignty because it admits that contingent realities (the counterfactuals of freedom) exist that are in some sense independent of God. Second, it threatens divine aseity
Perhaps, however, the TMF can be recast in a way that doesn’t suggest it offers enhanced control relative to other compatibilist accounts. According to this version of the TMF, whether an act is under the agent’s control is simply a matter of meeting some compatibilist-friendly conditions of one sort or another. The defender of the TMF could then contend that ordinary compatibilists can explain how agents satisfy the control condition for moral responsibility, but suggest some other condition might still be unsatisfied. We might think of this as something like a self-expression condition, distinct from issues of control. As this idea has it, for an action to be free, it must meet the control condition and also reflect the deep self of the agent. The defender of this sort of TMF would then go on to argue that existing deep-self views [see, for example, Arpaly (), Sher (), Wolf (), and especially Sripada ()] are insufficient, and must be supplemented with a condition appealing to CCFs, much as the PMF does. Whether this modified version of the TMF could be developed is something I leave to the reader to judge. In any case, it has a clearer path toward success than a version that depends upon establishing that fully determined agents might have control over the CCFs that partially constitute their essences. Even if successful, this alternative version of the TMF would still take upon itself the costs of Molinism, and would also be unappealing to many divine determinists for a reason to be discussed shortly.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism, Independent Compatibilism
because God’s knowledge of the valence of particular human actions (that is, whether they are free) depends on knowledge of the counterfactuals of freedom, which are not themselves dependent upon the divine will. The desire to uphold strong claims about God’s aseity and sovereignty explains why many divine determinists have traditionally avoided libertarian positions – even Molinist ones – in the past, so it is unlikely that they will find great reassurance in the claim that they may utilize the FWD if they adopt the TMF.
Divine Determinism, Independent Compatibilism, and the Free Will Defense Next, I will consider a proposal from Jason Turner. Turner, like Howsepian, considers a particular form of compatibilism and argues that compatibilists of this sort, at least, may adopt the FWD. Before turning to the substantive position, it is important to note a distinction between the projects pursued by Howsepian and that investigated here. Howsepian was interested in two distinct projects: arguing that compatibilists who adopt the TMF can make use of the FWD, and arguing that the TMF is the correct account of free agency. In fact, his published work outlining and defending the view preceded his work linking it to the FWD by several years. In Howsepian, then, we have a compatibilist whose preferred account of free will is perfectly compatible with the FWD. Turner’s project is more modest: . . . I’m not trying to argue that the view outlined is true. Rather, I’m outlining a (not obviously hopeless) position in logical space to show that an apparent incompatibility – between compatibilism and the free will defense – isn’t. (, )
With the contours of his ambitions before us, we can turn to the proposed account of compatibilism. The account of compatibilism that Turner put forward, which he calls independent compatibilism, is best understood by beginning with a disagreement pitting compatibilists against each other, with historical compatibilists on one side and nonhistorical compatibilists on the other. Nonhistorical compatibilists claim, roughly, that whether an act is free depends solely on facts about the world at the time in which the agent is
For defense of historical compatibilism, see, for example, Fischer and Ravizza () and Haji (). For defenses of nonhistorical compatibilism, see, for example, Berofsky (), Frankfurt (), McKenna (a, b), and Vargas ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
acting. According to this view, facts about the past, including facts that would explain the causal history of the agent and the agent’s mental states, are not relevant to evaluating whether the agent acted freely. Historical compatibilists, in contrast, claim that evaluations of freedom (and/or moral responsibility) are at least sometimes history sensitive – that is, a correct evaluation of an act may be possible only with knowledge of particular historical facts. One way to see this difference, and in fact a way to motivate historical compatibilism, is to recall manipulation scenarios. Following Turner, we can look to Alfred Mele’s Zygote Argument, which highlights the disagreement between the two sorts of compatibilists. This argument asks us to reflect upon two agents, Anna and Betty, who act in identical ways, but in different possible worlds. Anna and Betty are conceived at the same time, live full (and sometimes exciting) lives, and die at the same time. In fact, these agents are exact duplicates of each other. If we look only to the intrinsic features of each, no one – not even God – could detect any difference. We might even expand the similarity by noting that the portions of the possible worlds they inhabit – that is, the series of time slices between their conceptions and deaths – are intrinsically identical to each other. Within Anna’s life, one moment sticks out: the moment (t) in which she robs a bank. Of course, Betty similarly robbed a bank at the very same moment. Although their lives are identical, the histories of their worlds differ in one extremely noteworthy way: In Betty’s world, a demigoddess named Diana sometimes intervenes – in a way that does not respect the laws of nature – to change the course of worldly events. In fact, before Betty was conceived, Diana decided that she wanted her world – the world that Betty would soon come to inhabit – to have a particular sort of devil-may-care bank robber running around. She surveyed the state of her world, reflected on the laws of nature, and noticed that there was to be no such woman – at least, none that quite fit her desires – conceived in the near future. Saddened but resourceful, Diana took matters into her own hands. She ever so slightly fiddled with her world, knowing that even minor modifications could bring about great differences. After her brief intervention, she let the world run its course, knowing that Betty would be conceived in an environment that ensured that her life would be intrinsically identical to that of Anna.
See Mele () for his discussion of this argument. See Turner (, –) for his use of this argument. My discussion of the argument loosely follows Turner’s presentation, which departs in some minor ways from Mele’s versions.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism, Independent Compatibilism
The argument then turns from the scenario to claims about the agents in the scenario. Specifically, it argues that Betty is clearly not responsible for robbing the bank; after all, her entire life was planned before her birth with the sole intention of her becoming the bank-robbing villain she eventually became. No moment of her life, including the act of robbing the bank, was truly hers; instead, each was designed by Diana for the sake of satisfying the demigoddess’s desires. But, the argument continues, since Betty and Anna are intrinsically identical, and if Betty’s act of robbing the bank is not free, than neither is Anna’s. Thus, Anna’s action is not free. Incompatibilists, of course, will be quite happy with this conclusion. Our purpose here, though, is to see how this argument sheds light on different sorts of compatibilism. As long as the case of Anna is created carefully, compatibilists will be forced to deny the conclusion of the argument, but they may challenge the argument in different ways. The nonhistorical compatibilist will claim that, as long as the story is fleshed out in the right way, Betty is indeed free, and so, of course, is Anna. The historical compatibilist, in contrast, will claim that a crucial difference between Betty and Anna explains why Anna is free while Betty is not. The difference is not one of intrinsic dissimilarities, but rather of the histories of their worlds. Although their worlds are identical during their lives, an important past feature of Betty’s world, historical compatibilists will contend, makes all the difference: Diana arranged the world so that Betty’s action of robbing the bank was inevitable; moreover, she did so with the intention of determining that Betty act in this way. This history – this distant etiology of her action – renders Betty unfree. Since nothing similar can be said for Anna, and since we have no other reason to think her free will has been undermined, we must conclude that she is free. Betty and Anna, then, although intrinsic duplicates, must be evaluated differently. This evaluation of Betty relies upon the following principle, formulated by Turner: Independence: If S’s arranging matters in way w would result in T’s being causally determined to A, and if S knows this and arranges matters in way w in order to get T to A, then T does not freely A. (, )
Thus, the nonhistorical compatibilist takes the hard-line response to manipulation arguments that we discussed in Chapter . I am oversimplifying the matter, since some of the details of the distinction between these two sorts of compatibilists need not concern us here. Some compatibilists may say that there are historical requirements on free action, but that Betty fulfills all of them. These historical compatibilists will join with their nonhistorical cousins in responding to this particular pair of cases, but will differ from them in the analysis of other intrinsically identical pairs with different histories.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
We might call any compatibilist account that endorses this thesis independent compatibilism. We might think that Independence is a strictly incompatibilist principle, so there cannot be any sort of independent compatibilism. But this is not so. The easiest way to see this is by looking at determined worlds without God. Consider such a world that is similar to our own: It contains human beings who act and think just as we do. In this world, all human actions are determined, but they are so determined by past states of the universe together with the laws of physics. Even if we accept Independence, we will still not be in a position to judge whether humans in this world truly act freely. Although all agents are determined in this world, there is no being who knowingly arranges for humans to deterministically act in particular ways. Independence tells us nothing about whether such mindless determinism poses a threat to free will. While it might seem legitimate for atheists to adopt both determinism and Independence, we might think it is impossible for theists to do so. After all, we might think, if human agents are determined, then God must be the one doing the determining, and if God is doing the determining, then, given the truth of Independence, humans are not free. Matters are not quite this simple, though, for two reasons. First, as noted previously, some think that God could give indeterminate commands like Let there be deterministic world A or deterministic world B. In such cases, either world A or world B would be created, but neither would be a divinely determined world, since God’s command left it open which was to be created. Since we are interested in divinely determined worlds, we can set such a possibility aside. Second, God might create a divinely determined world containing determined agents, without having done so in order to have agents act in particular ways. This discussion reveals that the combination of divine determinism and independent compatibilism leaves us with a seemingly odd result: Humans can be free only if God did not arrange the world as it is in order to get agents to act in particular ways. In a divinely determined world, God has certainly determined that each creature acts exactly as it does; there is no denying this. Nevertheless, Independence considers not simply determinism caused by an agent, but determinism caused by an agent in an effort to get a particular result. Thus, it seems divine determinists may accept independent compatibilism and claim that humans are free.
If God necessarily exists, there are no such possible worlds. Nevertheless, we can surely consider such worlds whether they are impossible or not.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism, Independent Compatibilism
As an illustration of such a view, consider an Edwardsian account according to which God creates the initial state of the universe, together with deterministic laws, without intending each of the particular events that will occur as a result of this initial creation. According to such a view, the particular actions of humans might be nothing more than foreseen but unintended side effects of divine action. This is similar to a version of the intention-based reply to the blameworthiness objection considered in Chapter . Recall that this reply said that God avoids blameworthiness in determining the evils of the world, because evil actions fell outside of the divine intention in creating. Independent compatibilism provides a nice motivation by which to explain why human actions – both good and evil – are outside the scope of the divine intention: Given divine determinism, they must be outside of the divine intention for them to be free. In Chapter , I considered whether divine determinists can claim that God does not intend some of what occurs in divinely determined worlds. I argued there that making such a claim carries significant costs, and that these costs differ depending upon whether Edwardsian or primary divine determinism is adopted. The Edwardsian version, I argued, would likely require claiming that good actions, and indeed the entire lives, of humans at the present point in history fall outside of the divine intention as well. Primary divine determinists will, I claimed, need to concede that the act of creation is less like God’s writing of a novel – in which every element is thoughtfully created by the author – and more like an act of selection in which some elements are not intended. It should be readily apparent that the independent compatibilist can make full use of the FWD. According to this version of the defense, God could create a world filled with agents, each of whose acts were good. Furthermore, it is possible for God to create a world where such agents are determined to be morally perfect and to act freely. Crucially, however, God could not create such a world with the intention of bringing about these morally perfect, free agents. If God were to set out to determine a world with the intention of bringing about morally perfect agents, Independence would entail that the created agents do not act freely. Thus, even a morally perfect, omnipotent, omniscient being cannot set out to determine that evil is never done in the world while still allowing for the free actions of creatures. Independent compatibilism brings with it significant benefits for divine determinists. By accepting this view, they retain the ability to utilize the FWD. They also have a natural response to the blameworthiness objection as well as the manipulation argument. In regard to the latter, they
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
can claim that manipulation – at least the variety that undermines free will – requires the manipulator to act with the intention of making the manipulated agent act in particular ways. Because God does not so act as (this sort of ) manipulator, human actions remain free. Although adopting this position brings clear benefits, there are reasons to be careful. First, although independent compatibilism can be motivated independently of considerations of divine determinism, historical varieties of compatibilism are not obviously superior to nonhistorical ones. Second, in adopting independent compatibilism, divine determinists would be forced to give up on the idea that human lives are governed by the providential and intentional care that is sometimes supposed. God’s perfect sovereignty would remain intact, since nothing would be outside of the divine plan, but the parts of the divine plan governing human action would take on a peculiar, detached quality. I doubt many determinists would be happy to assume this added cost. Despite these worries, I think it is clear that Turner succeeded in what he set out to do. Independent compatibilists clearly can make use of the FWD. Moreover, whatever worries might be associated with the view, either concerning its viability as a theory of freedom or given its implications for divine determinists, it is not obviously hopeless.
I am inclined to agree with McKenna (b) in thinking that nonhistorical compatibilism is more resilient than sometimes thought, and in fact is more appealing than historical compatibilism. Turner (, –) develops an account of divine providence compatible with independent compatibilism, but it relies upon God governing through probabilistic causality, which is incompatible with divine determinism. Some nonphilosophical reasons also explain why some divine determinists will be unlikely to adopt independent compatibilism. Recall that some of the reasons some are tempted to adopt divine determinism come from specific religious texts or doctrines. Determinism (arguably) makes the interpretation of these texts and doctrines both easy and natural, which is something nondeterminists cannot boast. Consider, for example, the following biblical passage: “The Lord said to Moses, ‘When you return to Egypt, see that you perform before Pharaoh all the wonders I have given you the power to do. But I will harden his heart so that he will not let the people go’” (Exodus :). In this passage, it is clear that the Lord acts on Pharaoh precisely so that Pharaoh acts in a particular way. Given divine determinism and independent compatibilism, Pharaoh’s act cannot be a free one. Of course, the imagined determinist could simply deny that this particular act is a free one, but then incompatibilists may say the same thing; the divine determinist, in adopting independent compatibilism, has lost the upper hand. An even more problematic case arises in regard to the Christian doctrine of divine grace. According to (some) Christians, God grants at least some humans grace to perform good actions, grace without which it is impossible for humans to act well. Given the combination of divine determinism and independent compatibilism, this doctrine becomes rather peculiar. If God gives grace with the intention of prompting humans to act well, then the ensuing human acts are not free; in fact, in such a case, morally praiseworthy free human action is impossible, even with grace. To avoid this conclusion, one might say that God gives grace without the intention of prompting humans to act well, but then the very intelligibility of grace is called into question.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Other Options and Assessing the Costs
Other Options and Assessing the Costs What other ways might determinists find to dispatch the logical problem of evil? Although I do not have sufficient space to develop or assess any proposals at length, it is worth briefly considering some possibilities. One option is to endorse a strategy more commonly used to deflect the evidential problem of evil: skeptical theism. Although some might contend that the skeptical theist reply is not suitable for responding to the logical problem of evil, others, such as Daniel M. Johnson (), will disagree. Still others will opt for some sort of “greater good” defense that does not rely on indeterminist or incompatibilist claims. The existence of evil is incompatible with the existence of an omnipotent, omniscient, perfectly good God only if there are no goods that are at once incompatible with the nonexistence of evil and sufficiently good to justify the existence of some evil in the world. The idea common to this family of replies is that some such good might be plausibly identified. For example, determinists seem able to claim that evil plays a role in character formation, and so might claim that the existence of certain sorts of good character traits requires the existence of evil (Byerly ; Johnson ). Alternatively, they might appeal to the good of a law-governed world (Johnson ). The greater goods canvassed here are likely to be invoked by both divine determinists and indeterminists. Another greater good reply tends to be endorsed predominately by determinists: the divine glory defense. Such a reply begins by noting that the greater good in question is (or might be, for all we know) the revelation or manifestation of divine characteristics such as justice or mercy. The reply would then need to justify both the claim that the revelation or manifestation of divine glory is a great good (and thus could justify the existence of at least some evil in the world) and the claim that, in some sense, it requires the existence of evil. A full exploration of this reply would require carefully teasing apart and testing each element it includes. In this short overview, I will concentrate on only two questions: () What sort of good is being invoked? and () In what sense does this good require the existence of evil? Those who defend the divine glory reply disagree on just what sort of good is being invoked. Some theorists provide an analysis in terms of an epistemic good. Christopher Green (), for example, develops a reply to the problem of evil based upon the idea that the manifestation of God’s characteristics to humans is an epistemic good for humans. Matthew Hart () proposes a similar solution, arguing that the epistemic benefit to those in heaven justifies the pains of those damned to hell. Others,
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
however, suggest that epistemic goods are only part of the picture. According to Johnson (), for example, the action by which God manifests divine characteristics is a good distinct from the epistemic good of coming to understand these characteristics through their manifestation. In short, divine self-expression is a good distinct from that of creaturely understanding. Do the manifestation and revelation of divine glory require evil? At first it might seem that the epistemic model would be in trouble, since any knowledge humans might gain of divine characteristics could potentially be imbued in them directly by God. Green () suggests that although God could reveal the same information about divine characteristics without the need for evil, alternative manifestations (for example, through nightmares) would not constitute the same demonstrative goods that evil allows. Thus, punishing Hitler is not the same manifestation of divine justice as revealing that, if anyone like Hitler were to exist, that person would be punished, since the modes of presentation are different. Accounts of divine glory that posit the good of divine self-expression have a more straightforward way of meeting the present challenge. According to Johnson, several divine characteristics could not find expression in action without the existence of evil. He claims, for example, that “God cannot express his mercy and forgiveness without sin to forgive and undeserving sinners to whom mercy can be shown, and God cannot express his justice by punishing wrongdoers without wrongdoing to punish” (Johnson , ). What, in the end, should we make of the divine determinist’s purported loss of the FWD? We begin by noting that whether it is indeed lost depends upon one’s desiderata for a successful defense. Supposing the modest aim noted by Cowan, the determinist can use it as much as anyone can. If the determinist has more robust goals, though, then matters become more complicated. Cowan suggests that even if one has robust aims for the FWD, divine determinists and indeterminists still face similar problems. Divine determinists cannot utilize the FWD if they are not permitted to utilize propositions they believe to be false, but indeterminists fare no better: One of the propositions the FWD uses is “almost certainly false” even if libertarianism is granted (Cowan , –). Let us suppose, however, that divine determinists wish for a robust reply to the logical problem of evil, but are not convinced that the robust FWD is in as much trouble as Cowan suggests.
In addition to Green, Hart () discusses ways in which the suffering of the damned provides goods that could not have been achieved, or achieved as well, without suffering. The proposition in question is “Free will is a great good that justifies the existence of the moral evil that will occur if it exists” (Cowan , ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
A Brief Look at the Evidential Problem of Evil
We must then turn to the question of whether divine determinists may make free use of the robust FWD. The arguments of Howsepian and Turner suggest a positive answer, but Howsepian’s account runs into significant difficulties, and Turner’s account is unlikely to appeal to many divine determinists. I suggest, then, that few divine determinists will be able to utilize the FWD, at least in its robust and ambitious form. The divine determinist’s difficulty in utilizing the ambitious FWD does seem to represent some cost – though the size of this cost depends on how one views the value of determinist-friendly defenses relative to that of the FWD. Thoroughly investigating the value of skeptical theism alone would take a volume, to say nothing of exploring other possible replies. Whatever the eventual verdict on these other replies, losing a respected reply to a difficult problem represents some additional theoretical cost for the divine determinist.
A Brief Look at the Evidential Problem of Evil I have concentrated on the logical problem of evil because this problem has been given a solution that, although not without important critics, has won significant praise and seems incompatible with divine determinism. The same cannot be said concerning the evidential problem of evil. If divine determinists are forced to reject a widely supported solution to a troubling problem, then that seems to represent a real cost. Since there is no widely supported solution to the evidential problem of evil, it might seem that divine determinists have little to lose. But this is not so. Although the FWD has achieved greater success in regard to the logical problem of evil than any freedom-based response has achieved in regard to the evidential problem of evil, this is partly because the logical problem of evil is easier to deal with. Losing a great response to a mediocre objection may involve no greater cost than losing a mediocre response to a powerful objection. Views on both the success of the FWD and the relative weight of the two problems of evil will surely diverge. However they are judged, we shouldn’t be too quick to dismiss the cost of the loss of freedom-based replies to the evidential problem of evil. For this reason, it is worth dwelling, even if only briefly, on divine determinism and the evidential problem of evil. To begin, it might be helpful to think of the dialectical framework in which the evidential argument from evil is situated. The atheist begins by
I am grateful to an anonymous reader for pointing this out and encouraging me to discuss it.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
objecting to theism by pointing to certain sorts or amounts of evil in the world. The presence of such evil, the atheist maintains, provides a solid, if inconclusive, reason for thinking that God (at least the God of traditional Western theism) does not exist. The theist responds by arguing that the highlighted evil does not lower the probability of theism, or does not lower it as significantly as the atheist supposes, or does not make belief in theism, given all available information, irrational. Given this dialectical situation, the question is whether divine determinists are any worse off than other theists. We might think they are for two reasons. First, God’s involvement with evil may be closer and more problematic given determinism than would otherwise be the case. For example, the atheist may maintain that permitting a certain amount of suffering seems incompatible with perfect goodness, but that arranging things so that the same amount of suffering is determined to occur is even worse. Ultimately, this is simply another way of approaching the blameworthiness objection, which was discussed in Chapter . As I showed there, the determinist is not without resources in combatting this objection. The second reason for thinking that divine determinism might exacerbate the evidential problem of evil is that it might eliminate freedom-based replies to this objection. Theists might respond to this assertion in several ways. First, they might claim that no substantial cost is incurred here, because freedom-based replies to the evidential problem have little value anyway. They might make this response because they think we have good reasons, even independent of motivations for divine determinism, to think that the theory of freedom supposed by such replies is intellectually bankrupt. If one thinks that libertarianism is obviously absurd, one will likely think that replies based upon it are of little value. Thus, the loss of such replies carries no significant cost. Second, some might think that freedom-based replies do have some value and that their loss constitutes a cost. How costly this loss is will depend upon several other issues. The first and most obvious matter related to the cost of losing freedombased replies concerns the value of other currently available replies. If other responses to the evidential problem of evil are possible – ones that are not significantly worse than freedom-based ones, or perhaps even ones that are superior to them – then determinists need not fear too steep a cost in this case. So, for example, theists may respond to this problem by arguing that because of human epistemic limitations, we are not in a position to judge
As noted earlier in this chapter, this might not be the case with replies to the logical problem of evil.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
A Brief Look at the Evidential Problem of Evil
that the presence of evil constitutes evidence (or much evidence) against the existence of God. This skeptical theist response has many defenders, but also many critics. If one judges this response to be of a value not significantly below that of freedom-based replies, then the assessed cost of losing the latter replies will be relatively low. Alternatively, determinists might construct a divine glory response to the evidential problem of evil, claiming that epistemic and/or selfexpressive goods justify not only the existence of some evil, but also the sorts and amounts of evil we find in the actual world. Although such views are not as popular as they once were, they have found recent support. The second matter related to the cost of losing freedom-based replies concerns the nature of those replies themselves. We are granting that, as formulated, these replies are not available to divine determinists. At least some accounts, such as those of Hick () and Swinburne (), rely, at least in part, on libertarian accounts of free will. Although some divine determinists may feel comfortable invoking libertarianism to solve the logical problem of evil – as Cowan, for example, does – there is little value in their doing so to solve the evidential problem of evil. Putting aside the question of how successful these accounts are, we can ask how deeply they rely upon libertarian views. Ryan Byerly () has recently argued that the main argumentative strategies of several major freedom-based theodicies can be utilized by divine determinists. For example, divine determinists might agree with Hick in thinking that virtues developed through free decisions sometimes require that agents face challenges, hardship, and even misery. If this is so, then we have the root of a freedom-based, greater goods reply that is open to the divine determinist. Byerly () even suggests that the determinist might extend this theodicy to cover some moral evils, if such moral evils are necessary for greater goods such as forgiveness and contrition. He also tries to show that the theodicies of Swinburne and Plantinga, although explicitly libertarian, might be modified for the use of determinists. The present consideration of possible determinist responses to the evidential problem of evil has been swift, but it does allow us to reach a preliminary conclusion about the costs incurred in this area by divine determinists. In particular, it suggests that the costs to determinists will not be very significant if any of the following are true:
See Dougherty and McBrayer () for a collection of recent work on this topic. See Green (), Hart (), and Johnson () for recent work that defends robust uses of the greater glory account.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Divine Determinism and the Free Will Defense
(a) Freedom-based replies to the evidential problem of evil are of little worth. or (b) Freedom-based replies to the evidential problem of evil are not significantly better than other individual replies. or (c) The core argumentative strategies of freedom-based replies like those of Hick, Swinburne, and Plantinga can be utilized in modified form (and without major loss of plausibility) by divine determinists. Evaluating the truth of each of these disjuncts is not possible here, but many theists will likely judge at least one of them to be true. In any event, the case for the high cost of determinism in this matter is not as strong as it might initially appear. Of course, this does not mean that there is no cost involved: Surely every respectable reply to a challenge as powerful as the evidential problem of evil is worth something, and even if the core argumentative strategies of libertarian theodicies can be utilized, the determinist cannot adopt them in their entirety.
Conclusion It is sometimes thought that the FWD is incompatible with compatibilism. In this chapter, I have surveyed the positions of three authors – Cowan, Howsepian, and Turner – who claim that this common thought is wrong. Cowan maintains that the FWD, since it means to show merely that a number of propositions are logically compossible, may utilize false propositions. Moreover, even if we worry that it cannot utilize propositions thought to be necessarily false – as incompatibilism is often thought to be by compatibilists – divine determinists might still contend that incompatibilism is true for all they know. I have suggested that we treat Cowan’s discussion, together with Pawl and Timpe’s disagreement with it, as revealing different possible desiderata for a successful FWD. If divine determinists are willing to settle for Cowan’s weaker desiderata, then they may still utilize the FWD without qualms. Conversely, if they desire a different sort of FWD, which utilizes only propositions that they believe to be true, or at least only those
Even if (b) is true, some cost may still be incurred by determinists. One way for theists in general to respond to the evidential problem of evil is through a disjunctive reply, uniting many different responses into one. Insofar as the determinists cannot avail themselves of freedom-based replies, their disjunctive reply will be weakened. If we are sympathetic to disjunctive replies as an argumentative strategy, we will also judge the weakening of such a reply to be a cost, even if not a very significant one.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
Conclusion
that are not believed to be false, then a more robust proposal for the compatibility of divine determinism and the FWD is necessary. Both Howsepian and Turner rise to the task of offering such proposals. Howsepian provides an account of free will, the TMF, that is perfectly compatible with the FWD. He sets out to do more than demonstrate the logical space for such a view; he defends this view of free will as well. While he does succeed in showing the possibility of a compatibilist endorsing the FWD without appealing to Cowan’s weakened desiderata, his account is ultimate implausible. It takes upon itself the costs of Molinism, fails to account for enhanced agent control relative to other compatibilist accounts, and would force some divine determinists to cast away some of the beliefs that led them to adopt determinism in the first place. Turner’s project is more modest and more successful. He doesn’t succeed in showing that the proposed account of free will, independent compatibilism, is correct, but he doesn’t set out to do so. Instead, he shows that one compatibilist account is compatible with the FWD, even when more ambitious desiderata are sought, and that this account can be motivated independent of theistic concerns, and is not “obviously hopeless.” I conclude, then, that Turner has succeeded in what he set out to do. Still, divine determinists will likely find themselves looking elsewhere for a reply to the logical problem of evil. In this search, they will have options. They might turn to skeptical theism or invoke a greater good response that does not rely on incompatibilist accounts of free will. In particular, they might point to the goods of character formation, orderly laws of nature, or the manifestation of divine attributes. Losing the ability to invoke the FWD is not as bad as it is sometimes made out to be, but losing such a powerful response to a familiar objection seems to bring with it some cost for the divine determinist. Another cost comes from consideration of the evidential problem of evil. Here, too, divine determinists will find themselves with options, but once again they will lose the ability to invoke some of the most popular responses to a common and forceful objection. In neither of these cases are determinists forced to accept implausible theoretical baggage; instead, they simply lose the ability to utilize familiar argumentative moves. We might say that the loss of the FWD and some freedom-based replies to the evidential problem of evil are dialectical costs, rather than theoretical ones. This is not to say that they are unimportant; if, in the end, these dialectical costs make it impossible to respond to the problem of evil, divine determinism may prove unworkable. I tend to think that some deterministfriendly responses are not without value, but that is an issue too large to address here.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.007
God, Determined Agents, and Love
Introduction
In previous chapters, I took up challenges to divine determinism arising first from concerns about the status of humans and later about the status of God. In this chapter, I turn to decidedly more interpersonal matters: the love of God for humans and the love of humans for God. Divine determinists face some special problems in accounting for each of these. We might begin with God’s love for humans. Among theists – certainly those who are laypeople but also, it seems, among philosophers – God is often taken to be not only morally perfect in some sort of abstract way, but also personally concerned for the welfare of human beings. Indeed, God is often thought of as the paradigm of a loving individual; in some traditions, God is even identified with love itself. Yet divine determinism is sometimes thought to call this image into question. Certainly the sorts and amounts of evil around us have provided grounds for suspicion about the reality of a loving God, but some think that divine determinism makes matters significantly worse. One may begin to suspect that even if divine determinists can avoid saying that God causes sins, and even if they can further maintain that God is morally blameless, their defenses of God amount to nothing more than appeals to technicalities, and that no such technicalities can help us make sense of a loving God who inflicts such evil on the world. There is, then, a worry that divine determinism endangers the view that God loves us. Additionally, divine determinists face a problem in accounting for how humans can love God, at least with the sort of union-seeking love that leads to or partially constitutes being in a genuine loving relationship with God. If divine determinism is true, then God has determined at least some human agents to seek after a relationship with their creator. Augustine famously calls upon God, saying, “You have made us for yourself, and our heart is restless till it finds its rest in you”
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
The Problem of Divine Love
(Augustine , sec. ..). Perhaps there is nothing odd about God creating beings who yearn for the divine, but divine determinism adds something more to this. I discussed manipulation arguments in Chapter , and we need to consider another of these arguments here. Let us return to the fanciful case of the team of scientists. Now, instead of imagining them manipulating an agent who goes on to perform terrible crimes, we can imagine them manipulating an agent who will become their close friend. Suppose we were observers of both the original manipulation and the life of this determined agent. After seeing the determined agent slowly develop into the manipulators’ friend, we might think that the manipulated agent’s love for the manipulators isn’t genuine love at all, but rather some sort of worthless counterfeit. In this chapter, I investigate whether divine determinists may continue to maintain that God loves humans and at least some humans love God. I begin by considering several images of divine love, focusing and developing one of these for the purposes of this chapter. I then turn to an argument that concludes if divine determinism is true, then God fails to live up to this image. Next, I canvas several tempting ways to respond to this argument, concluding that each of these is ultimately unsatisfactory. I then lay out several more responses, each of which will prove successful in making sense of divine love, but each of which, unfortunately, brings with it significant costs. After these considerations of the problem of divine love, I turn to the problem of creaturely love, addressing whether divine determinism undermines the ability of human agents to enter into a relationship of union love with God.
The Problem of Divine Love Many theists seem to think of being loving as a great-making feature. Instead of focusing on what sort of love, if any, might qualify as a greatmaking feature, I propose to look at the images of a loving God that have been common in Western monotheisms and seem to be embraced by many contemporary theistic philosophers. A number of such images present themselves. First and most basically, God is seen as the source of all benefits. According to this view, God creates all things, and in doing so is a source of great goodness in the world, and indeed, the ultimate source of all goodness. Some of these created entities are good for others – as food and sunsets are good for humans – and insofar as God is the source of the former, God displays a love for the latter. Moreover, since existence is itself a good, and since God is the ultimate source of all contingent existence, the divine love reaches to all contingent entities.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
A second image of divine love is substantially richer. It seems to many modern-day theists, and indeed has seemed to many of their predecessors in the past, that God continuously watches over and cares for the lives of creatures. According to this view, God’s love is not exhausted merely in bestowing existence and providing an environment in which creatures may thrive. Instead, the divine love is caught up with the concerns of creatures, so God continuously cares for creation. This idea is often captured by claiming that God’s love is reflected, albeit dimly, in the love of parents for their children. If this is so, then we may begin modeling God’s love on human love in our ordinary experience. Like human parents at their best, God wants what is best for us and understands what this entails. Some theists go further in developing the idea of divine love, particularly divine love for humans. Christians claim, for example, that the divine love extends to sacrificing for the objects of this love. Some have also claimed that in some sense God is love. Finally, some theists claim that the love of God is displayed in an offer to share the divine life with creatures in eternal bliss. I wish to turn from these images to focus in on one particular view of divine love: that captured by the image of God as a loving parent. If God’s love for creation is exhausted by bestowing some good, such as the good of existence, on creatures, then I see no reason to think that divine love is endangered by divine determinism. I will set aside claims about divine sacrificial love, as well as claims about God being identified with love, since such claims would take us too far into particular theological traditions. The relevance of the final image of divine love, that of a God offering a gift of eternal life to humans, to the current project is more complicated. There is certainly no incompatibility between divine determinism and this sort of love, but the possibility of an afterlife may be relevant to this discussion in another way: It may provide the tools for a possible reply to the objection that a determining, loving God would not create a world with as much evil as this one. For this reason, the possibility of a glorious afterlife will have some place in the following considerations. Most centrally, however, I will focus on the image of divine love as perfect parental love. This image might have roots in considerations of perfect being theology or in particular religious traditions. We might reach a view of
The classical Christian source of this claim is John :. For our purposes, the offer to share the divine life with humans is the most important element of this claim. Some theists extend the offer of eternal blessedness to other creatures, too. Claiming that angels enjoy heaven is a staple of various traditions, and some theists extend the population of heaven even further – for example, to animals other than humans (Dougherty ).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
The Problem of Divine Love
God’s parental love by noting that omnibenevolence is a great-making feature and, therefore, must be a feature of God. Given this, it seems that God’s love for creatures would manifest itself in meticulous care for creatures, similar but far superior to that of human parents for their children. Of course, there may be very reasonable ways of resisting the claim that a perfect being must show this sort of love to humans, but it seems possible to reach such a view of divine love from perfect being theology. Most theists, however, have this view of divine love – if they have it at all – for reasons rooted in specific religious traditions. References to God as a father of creatures, or creatures as children of God, occur clearly in both Jewish and Christian traditions. Likewise, other religious traditions have strong views on the care that God bestows on creation – a care that in some cases might be considered a model for human parents to follow. Obviously, the relationship between God and humans is a complicated matter, and claims will be controversial not just between religious traditions, but even within them. Nevertheless, the idea that God is in some way a loving provider is a common one. Different individuals will qualify this account in a variety of ways, and if the account of divine love is qualified enough, then the problem to be discussed later in this chapter may not even arise. I will return to this issue when I discuss revising one’s account of divine love in response to that problem. Of course, some theists might not need to revise their view of divine love because it is sufficiently different from the parental view of divine love. Something similar might be said in relation to other problems for divine determinism. For example, if we adopt certain sorts of views of divine goodness, the problems of evil and divine blameworthiness will not arise. Why think that divine determinism is incompatible with the parental view of divine love (hereafter simply “divine love”)? Some might simply find it intuitively obvious that divine love is incompatible with determining the horrors endured by victims of unspeakable crimes. Moreover, if we think that it is preferable to endure evil rather than to perpetrate it, we may simply find it obvious that divine love is incompatible with determining individuals to become twisted and cruel. This intuition may be defended with an argument from analogy. God’s love is like human parental love at its best. Since the best human parents would not determine their children to become wicked, God would not do so either. Since divine determinism implies that God does so, either God’s love is false or divine determinism is.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
This problem is significant as it stands, but the doctrine of hell makes this problem even more difficult. We might be tempted to think that the very best parents would permit and even determine their children to undergo great suffering, and perhaps even become temporarily wicked if it is somehow aligned with the children’s long-term good. But if some are damned to hell, as many Western theists suppose, and every moment of the lives that led toward this fate was determined by God, then it is difficult to see what the greater good for them could be.
Some Unsatisfactory Ways of Responding to This Problem I wish to begin the assessment of responses by highlighting some that, I will argue, are incapable of resolving the problem of divine love. The first few come from Hugh McCann, who has provided a number of brief remarks that point toward responses to the divine love problem. To conclude this section, I turn to the question of whether divine determinists may avoid this problem by appealing to skepticism concerning what is good for us. Respecting Autonomy Hugh McCann recognizes the problem of divine love for those who espouse divine determinism, and particularly focuses on the issue of hell. He writes: “Above all, there is the question of how a God who truly loves his creatures, and is in absolute control of all that occurs concerning them, could consign anyone to damnation” (, ). In responding to this question, he moves between responding to two objections that are best dealt with separately. First, what justifies God in condemning people to hell? Second, how could God as a perfectly loving parent ensure that people earn this fate in the first place? McCann claims that the act of damnation is not inconsistent with the idea of a loving God. He argues that God does not do anything unfair in honoring the choices of those who have rejected the possibility of salvation. In fact, he suggests, “it is hard to see how a loving God could do anything but honor their choice in the matter” (, ). This claim seems to rest upon two key ideas: Honoring: Loving parents should honor the autonomy of their children by respecting their choices. Damning: In damning agents, God merely respects their choices.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Some Unsatisfactory Ways of Responding to This Problem
Damning relies upon a particular soteriological account, and I have no interest here in establishing which such account should be adopted. For this reason, I will grant Damning for the sake of discussion. But what of Honoring? The truth of this claim is difficult to establish, both because “respecting their choices” is painfully vague, and because there are likely limits to what loving parents ought to respect. In any case, even if Honoring is also accepted, this solution cannot succeed in showing how divine love is compatible with a determining God damning human agents. This response provides a reason for thinking that a loving God, faced with a hardened sinner – one who has freely chosen to reject salvation – might have recourse to damnation. But even if it can succeed in this regard, the divine determinist should not look solely at the moment of damnation in addressing the problem of divine love. The problem of divine love, if framed in terms of damnation, is not why a loving God would justly condemn the wicked to hell, but why one would design a world in which certain people are determined to become wicked and then remain obstinate in their wickedness in the first place. Focusing on God’s judgment of those who have already become wicked obscures the real problem. To see this point, we might return to the parental analogy. Suppose a parent arranges matters so that a child is brought up in an environment that glorifies violence, selfishness, and disrespect for others’ rights and the rule of law. Now suppose that this child, predictably, develops into a criminal. The parent then publicly states that the child should be punished for the criminal actions. If we assume that the parent could have raised the child in a better environment – that the parent understood the problem with the environment in which the child was raised and had the means and opportunity to provide a better one – then it seems that this parent falls short of perfect parental love. Perhaps the parent’s love could be defended in some way, but not by merely noting that punishment is an act of love, responding to the autonomy of the child’s choices by respecting the free criminal decisions. To respond in this way involves mistakenly focusing solely on the question of whether the punishment itself could be in keeping with love, instead of addressing whether a loving parent might arrange for the child to become deserving of punishment.
It does not, of course, solve problems unrelated to issues of soteriology – for example, how a loving God could condemn a child to a painful battle with cancer – but it doesn’t even attempt to solve such issues, so it cannot be faulted in this regard.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
Pointing toward Future Autonomy McCann (, ) also argues that moral autonomy is itself a good thing, and insofar as even those destined for damnation possess such autonomy, we have evidence of divine love. This may be so. It seems reasonable to grant that moral autonomy is a good thing, and we might accept that God’s giving such a gift to the reprobate constitutes a loving action. Furthermore, McCann notes that we should not presume that God would show greater love by not creating the reprobated at all, for it is meaningless “to think the lost would be better off had they not existed” (, ). McCann’s claim might be taken to involve a different sort of love than the one at issue; it might be thought that this claim is plausible only if we set aside interpersonal love and consider some sort of abstract love of being. Recall that the original objection claims that a determining God fails to demonstrate consistent parental love precisely because it seems that God does not act in a loving way toward creatures. It seems that God does not act in loving ways toward creatures precisely because God seems to create a world in which specific creatures lack specific goods or endure specific evils. In particular (this version of the objection assumes), some creatures are damned to hell. McCann responds by claiming that we should not say that the damned individual would be better off lacking existence entirely. Now, we might expect him to defend this by saying that existence is a good thing, and that in some abstract way this damned individual contributes to the collective goodness of the universe. But this is not what McCann has in mind. Instead, his claim turns on the idea that we can meaningfully contrast two possible worlds only in certain ways. Most importantly, if world w contains an agent, Agnes, and world x does not, then there is no sense in saying that world x is better for Agnes than world w. Thus, although it might be true that Agnes contributes to the goodness of world w, while the same cannot be said of world x, this is not the point upon which McCann’s response rests. If McCann is correct in thinking that there is no sense in saying that God would have loved Agnes more in creating a world in which she never existed than in creating a world in which she exists but is damned, then the determinist may be better positioned to respond to the present objection. When we add together the gifts of existence and moral autonomy, perhaps we can admit that God does act lovingly toward these lost souls.
The question of whether the damned would be better off if they didn’t exist has a long history. For an overview of a number of historical discussions of whether this is so, see Hoffmann (). I am indebted to an anonymous reader for urging me to discuss this issue.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Some Unsatisfactory Ways of Responding to This Problem
But this is no solution to the problem of divine love. As framed, this problem deals with how a love modeled on human parental love at its best, but exceeding it in generosity, could be compatible with reprobation. Surely, perfectly loving parents would not simply provide a few good things to their children. Consider a human mother who, having brought her son into the world, continues to feed him to preserve his life. Furthermore, she provides her son with many good things. Nevertheless, without diminishing her son’s moral autonomy, the mother knowingly ensures that her son develops a wicked character (perhaps she surrounds him with evil friends and teaches him false moral precepts). Although we might have grounds for saying that this mother shows a rather limited degree of love to her son, we would not think of her as a model of parental love. Authorial Image McCann may respond by noting that God is not related to the reprobate as this imagined mother is to her son. To understand the relationship between God and creatures, McCann suggests we consider an image to which we have continuously had recourse: that of an author writing a book. The author creates characters in their very acting, rather than causing others to act in the way our imagined mother might. One effect of this is that, in creating a sinful agent, God “is no more guilty of sin in the matter than Mozart was guilty for creating Don Giovanni a seducer and murderer” (McCann , ). McCann admits that the image has limits, most notably that an author cannot truly interact with the characters within the novel. Could it nonetheless help us solve the problem of divine love? It is not clear how it could. If we are to take acts of divine generosity and sacrifice as signs of love, then why shouldn’t we likewise take God’s determining harms as signs of a lack of divine love? There may be other reasons for reading the signs in some other way, but if we are entitled to see some of God’s actions as rather like a loving parent, then it is unclear how we would deal with other actions that challenge this image. Skepticism about Our Good Some might attempt to solve the problem of divine love by appealing to something like skeptical theism. This response begins by noting that we are likely ignorant of many sorts of goods. Additionally, we might be confused about whether certain events that we judge to be bad truly are bad. Finally, we might correctly judge that certain events are bad for us, but misjudge precisely how bad they are. Because of the likelihood or
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
probability of these three epistemic problems, this response continues, we should not be surprised that we cannot tell how the divine love of a determining God is compatible with the evils in this world. It may be that the suffering we endure is necessary for our very own later fulfillment, so a loving God would arrange for it. Alternatively, we might judge certain experiences to be bad that are either themselves good or indifferent. It seems that God’s making John into a serial killer is a sign of a lack of parental love. But, just as a child cannot see why a loving parent would arrange for a painful medical procedure, so too we might not be able to see why a loving God would design such a human life. While it seems reasonable that we are ignorant of what is good for each of us, it is far from clear that a satisfactory defense of God’s love can be based upon such a claim, at least not if the reply aims to defend the combination of divine determinism and the doctrine of damnation. Although we might not be aware of many good things, defenders of divine determinism do have an idea of what is bad for us. Separation from God, often taken as the fundamental bad-making property of the state of damnation, is undeniably a terrible fate. For this reason, those who adopt this view cannot remain skeptical concerning whether damnation is actually bad for the reprobate.
Some Costly Answers I have argued that several avenues of response are not successful. This is not to say, however, that no replies could succeed. Indeed, a number of options are open to the defenders of divine determinism. In this section, I outline several such lines of response and argue that they do succeed in solving this problem. I also argue that, in each case, many defenders of divine determinism are likely to find the costs of the response rather high. Thus, although the problem of divine love can be overcome, it represents an important objection to divine determinism because it forces its defenders to bite a bullet, even if they have a choice about which to bite. Skepticism without Hell Earlier in this chapter, I considered the response that the problem of divine love might arise merely because we cannot tell what is good for us. I argued there that if the doctrine of hell is adopted, this form of skepticism would
This response may say, for example, that given our epistemic situation we cannot tell whether the Stoics are correct in thinking that pain is indifferent.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Some Costly Answers
be unable to solve the problem of divine love. If the doctrine of hell is not accepted, then this form of skepticism shows real promise. This reply again makes the claim that God does love each person and acts in a loving way toward all. The true evils each person endures (and indeed, the evils each performs) somehow will lead to our own benefit in the long run. This reply comes in two forms. In the first form, the response begins by endorsing universalism about salvation. This beginning does not immediately answer all questions about divine love, but it does make the problem much less daunting. It is hard to see how a loving parent would arrange for children to become people worthy of enduring unending pain, but it is considerably less difficult to begin imagining such a parent arranging for a series of painful trials before bringing them to fulfillment. Of course, some account of why these trials were necessary at all would be desirable, but the acceptance of universalism at least opens up space for greater good accounts to be plausible. In the end, however, this answer may prove unsuccessful. It may turn out that the great pains that some undergo are unnecessary for their fulfillment, or that it is unloving to inflict such pains even if it is for the good of those who suffer. Whether universalism will be of much use or not, there is a problem: Many Western theists have hesitated to adopt universalism, sometimes tying the worries to particular religious views, sometimes grounding them in more general considerations. For example, Hugh McCann notes: The scriptural support, though real enough, is thin – and not nearly as weighty as what can be brought to bear against it. Moreover, universalism could seem to trivialize the moral experience of life, to belie the weight most theists attach to sinners finding redemption in their earthly existence, and the importance of assisting them in so doing. (, )
I have no interest in evaluating these concerns. Instead, I merely note that a number of worries are associated (at least for some) with adopting universalism. Thus, although it provides the beginnings of a response to the problem of divine love, universalism brings with it significant costs as well. The second version of this response denies the existence of an afterlife entirely. While it is evident to many people that their past pain has provided an overall benefit to their lives, it is hard to see how a painful and tragic death might be good for the one who endures it. This version of the skeptical response maintains that each of our sufferings is somehow in
See, for example, McCann () and Timpe ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
our own interest, just as a painful vaccination might be; we simply cannot see how. If we adopt the universalist attitude toward salvation, then it might be relatively easy to be skeptical about our ability to see how present suffering is connected to future fulfillment. The skepticism, in fact, might be of a fairly mild variety: We simply cannot see how present pains will eventually help bring us to the best sort of eternal life. If, however, we deny the existence of an afterlife entirely, then the required skepticism becomes much more robust: If the evil events we endure do have some connection to promoting or even constituting our own good, we shouldn’t expect to see that this is so. An Appeal to Controversial Metaphysics The second possible response I wish to consider, like that just discussed, assumes that God’s love for us ensures that a concern for our good guides divine action. We might begin by granting that it is false that nonexistence is better for an individual than any state of existence, no matter how bad. The next step would be to deny the existence of any modal properties of individual persons. Take the life of John, a serial killer. Consider John’s life story – one may fill it out as one likes – and add to it his post-death biography. It at least seems that things might have gone differently for John. Even if we suppose some sort of divine determinism, we might still think that at least God could have ensured that John’s life went differently. But what if this is wrong? What if any change in John’s story would have wrought a change in whose story it is? What if every aspect of John’s life and death were essential to his personal identity? The metaphysics of this account may be filled out in a variety of ways. As an illustration of such a possibility, consider a version of the worm account. According to this account, individual people are spacetime worms, made up of temporal parts. John, it turns out, is simply an aggregate of a very large number (perhaps an infinite number) of temporal parts. In the version of this account under consideration, John’s identity conditions are rather severe: Add, remove, or change a single temporal part, and the aggregate that is John is replaced by a different aggregate. Now consider again the worry underlying the problem of divine love. The major concern is that loving parents surely would never ensure that their children would perform heinous actions and then be punished (even if justly) by eternal suffering. If God loves John, then why not create a
For discussions of personal identity and temporal parts, as well as a discussion of the worm account (if not an endorsement of it), see Sider ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Some Costly Answers
better life for him? More importantly, why not ensure a better existence after death? Given the account of personal identity discussed earlier (or any other similar in the relevant respect), it was impossible for God to create John with any other destiny than the one he has. God had only two options concerning John: either create him, as a serial killer, destined (the examples assumes) for damnation, or not create him at all. Since nonexistence (this position claims) is not better for John than existence (even this existence), God’s parental love does not rule out creating him. Moreover, we cannot even reasonably ask why God couldn’t simply create the serial killer but deal differently with him – perhaps, for example, by choosing to withhold punishment. Doing so would change the post-death temporal parts, which would entail that John never existed at all. If John is created, then God has given him the best possible life he could have had. The cost, obviously, is again rather high. These requirements for identity conditions are extremely strict, and it is far from clear that any such conditions should be accepted. Moreover, some may blanch at the idea that nonexistence is never better for an agent than other possibilities. Sacrificing One Loved Object for the Sake of Another A third possible line of reply begins by granting that despite God’s love for individuals, sometimes a creature’s good must be sacrificed (even by ensuring that that being will never reach fulfillment) for other goals. Various goods may be filled in here, but most notable is perhaps the manifestation of God’s justice. Reprobation, then, is no indication of a lack of divine love, but rather an indication that God loves some things more than we might expect. Here, a worry arises. Consider again our image of human parental love. Suppose James, a father of three, deeply loves his children. Suppose, further, that James also cares deeply about his collection of model trains. In fact, although he would sacrifice his own life for his children, he consciously chooses to expand his collection even when he knows that this has detrimental effects on the well-being of his children. When his friends confront him, claiming that if he loved his children, he would spend less time and money on his trains and more time and money on his family, James responds by pointing out that he does love his children, and indeed would die for them, but he loves his trains as well and must balance these as best he can.
For discussion of this and other possible goods, see Green (), Hart (), and Johnson ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
We might be concerned about making James a paradigm of human parental love, especially if we decide to use him as a model for divine love. Nevertheless, we certainly do think that sometimes it is appropriate to sacrifice the goods of our children for other goods. We would, perhaps, not think badly of someone who allowed their child to die so that thousands of others might live. Consider, for example, a case in which a general must make decisions about which soldiers to send on a dangerous, but crucially important mission. The general considers various possible platoons and judges that the most prudent decision would involve sending in the platoon to which his child belongs. Based upon this judgment, he orders this platoon to carry out the mission. It seems evident that this action would be perfectly compatible with the general having the deepest love for his child. (Christians, at least, are already committed to the virtue of such actions since they praise the Father for sacrificing the Son for others. Other examples of praise for such sacrifice, if less central ones, appear in other religious traditions.) Some may find such behavior unconscionable and, therefore, deem this option to be a nonstarter. For others, the real question will come down to the object of love for which the good of one’s child might be sacrificed. Sacrificing the good of one’s child (at least, in any major way) for one’s train collection seems to undermine a parent’s professions of love. The question is whether the value of something, whether the manifestation of divine justice or something else entirely, could reasonably weigh against the good of one’s child. Here, perhaps, a skeptical response becomes useful: We simply do not know the value of many goods, so we should not expect to see what may reasonably cause loving parents to sacrifice the good of their children. In such a view, God’s love for human persons is surely great, just as parents’ love for their children, but that does not imply that there is nothing else of value that must be weighed against human fates.
Johnson () uses a similar image to motivate the same point. One might think that manifestations of divine justice are valuable for two reasons. First, one might think that they are valuable because of their usefulness. Other agents, perhaps glorified agents in heaven, gain from viewing such manifestations. Second, one might think that such manifestations are valuable in themselves, independent of their benefit to agents. For discussion, see Johnson (). One might go even further and say that the greatest good for any individual is carrying out God’s plan, even if it involves damnation. Some have suggested that our love for God should be so great that we would willingly sacrifice everything, even salvation, for the pleasure of God (Liguori ). Even if we should be willing to make such a sacrifice, the point remains that it would still be a sacrifice, precisely because our own good would surely not involve damnation.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Some Costly Answers
Perhaps this line of response is not unreasonable. Furthermore, the most extravagant proposition it foists upon us is that we are ignorant of the extent of possible goods that God might value. The real cost for this view comes from accepting that God’s great love for us is not sufficient to keep the divine plan from determining some to meet the worst of all fates. If this is so, then although God’s love may remain strong, it cannot bring the same level of assurance and comfort that many hope to find there. What assurance can parental love provide if we know that sometimes the parent willingly sacrifices the good of the child in a quest to secure other goods? Revising This View of Divine Love The final response worth considering is a revision of the common view of divine love. Theistic philosophers have long stressed the idea that we can have little understanding of divine characteristics. Although religious believers have found clues about divine love in their scriptures, perhaps they have taken metaphors of parental love too far. What is central to divine love is God’s willing the good of creatures. Insofar as each creature has been granted some goods, not least of all existence, it is clear that God does love everyone. Ideas about God loving each of us such that we should eventually reach fulfillment, and indeed the idea that God loves each of us equally, need to be jettisoned. At least at certain moments, important theistic philosophers seem to have adopted such a view. Aquinas considers the objection that God could not reprobate anyone, since it is impossible to reprobate one whom you love. He responds: God loves all men and all creatures, inasmuch as He wishes them all some good; but He does not wish every good to them all. So far, therefore, as He does not wish this particular good – namely, eternal life – He is said to hate or reprobate them. (Thomas Aquinas I, q. , a. , ad )
Defenders of this sort of view must replace the parental view of divine love with some other account, but they will, of course, have options in this regard. One option that they might consider is replacing the interpersonal or moral account we find in the parental view with something decidedly more metaphysical. According to this sort of view, God’s love of the universe is displayed in the being or existence of creation. Once this sort of love is brought into the picture, the metaphor of the author is again valuable. Like the great novelist, God shows concern for creation by
For discussion of this claim, see Helm (), Jordan (, , ), and Talbott (). I am indebted to an anonymous reader for suggesting that I explore this alternative view of love and for noting its connection to the authorial image and other responses.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
breathing life into a complex and interconnected web of beings with a range of abilities and perfections. Thus, although appeal to the authorial image will not, by itself, diminish the force of the present objection, it can be of aid to those willing to reject the parental view of divine love. Moreover, it seems well positioned to combine with other responses noted earlier. For example, we might combine both this more metaphysical view of divine love and the authorial image with the skeptical view or the claim that an object of love might be sacrificed for the good of another. Although this line of response is the cleanest and easiest, my sense is that many contemporary theistic philosophers will be uncomfortable with it. Others will think that this is one of the most promising lines of response discussed. Of course, the ranking of the plausibility of these views will depend upon the plausibility of related positions. For example, the plausibility of the first response (skepticism without hell) depends, in part, on the plausibility of universalism, while the plausibility of the third view (sacrificing one loved object for another) as well as the first depends, in part, on the plausibility of a sort of skeptical theism. The second response (appeal to controversial metaphysics of personal identity) will be seen by many as an appeal to ridiculous metaphysics. Finally, the fourth response (revising the image of divine love) will be seen by many as an abandonment of a core tenet of theism, and may be accused of producing an image of God unworthy of devotion. Many theists will be united in concluding that among these responses there is an obvious leader, but, I suspect, they will differ in their view of which response is the obvious choice. Are other options available to defenders of divine determinism? While I make no claim that the options I have offered here are exhaustive, there is some reason to think that the four promising lines of response I have considered are representative of all specific possible lines of response. Consider first that any possible response either will rely on the claim that evils are present because of God’s love for us or will avoid relying upon any such claim. If it does rely upon such a claim, then, if we focus on ultimate fulfillment, either everyone eventually reaches salvation (the first promising response) or God shows parental love for each person even in determining some to find their way to hell and never reach fulfillment (the second promising response). If the response does not rely on the claim that evils are present because of God’s love for us, the response must argue that sacrificing the fulfillment of humans either is consistent with our picture of divine parental love (the third promising response) or is inconsistent with it and forces us to revise our image of God’s love (the fourth promising response).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
The Costs of Divine Determinism and Its Alternatives
In addition to those four specific lines of response, one might offer a disjunctive reply, similar to that discussed when we looked at divine blameworthiness earlier. Some might claim that, although they aren’t sure which specific line of response to endorse, there is weighty reason to hold that either () all evils endured will eventually lead to an individual’s good, or () the nature of personal identity requires that specific people suffer in particular ways, or () God is justified in sacrificing the object of love for other goods, or () divine love should not be thought of in the way it often is. According to this reply, each specific line of response deserves some relatively small credence, but the truth of the disjunction involving all of them should be significantly higher, and high enough to successfully reply to this objection.
The Costs of Divine Determinism and Its Alternatives So far, I have put forward an objection to divine determinism and discussed some possible ways to respond. It seems that investigation of this issue highlights some costs for divine determinists: They are forced to adopt some seemingly costly way of responding to the objection from divine love. But it is important that we consider not only the costs of such a view, but also what we might think of as the relative costs of a view – that is, the costs of a view relative to its alternatives. Although an assessment of such relative costs in general is far beyond the scope of this book (as it would involve assessing the costs of atheism and indeterministic theism overall), it is worth considering whether the alternatives to divine determinism face very similar costs to those under consideration. Of course, atheism involves no costs related to the problem of divine love, but other versions of theism may. Consider, for example, the view that God is omnipotent, omniscient, and perfectly good, but does not determine every worldly event. Would those who endorse such a view incur costs concerning divine love similar to those faced by the divine determinist? If so, then the relative costs of divine determinism arising from consideration of divine love, at least relative solely to what many would think of as this close rival, would be low. Theists who reject determinism may, indeed, face something like the problem of divine love outlined in this chapter. Although fully exploring
This disjunctive reply may not be quite as valuable as that discussed in Chapter as a response to the problem of divine blameworthiness. In that disjunctive reply, there was good reason to think that the truth of some of the specific replies it involved were mutually exclusive, which allows for a robust disjunctive response. I see no reason to think that the individual replies discussed here are similar in this respect.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
this problem would take us too far afield, the basic thrust of the objection is that events in this world do not seem compatible with the providence of a perfectly good God, even if divine determinism is rejected. It seems, for example, that a loving God would intervene to prevent devastating floods (or would have set up the universe so that they didn’t happen in the first place) or to prevent individuals from developing vicious characters and harming each other. Indeterminists would surely have some explaining to do, but they might have a few more possible replies that are not open to determinists. Earlier, I rejected the idea that one might defend the love of God by appealing to God’s respect for human autonomy. Notice, however, that my complaints do not apply against an indeterministic version of that line of response. For example, the suggestion that human autonomy is both of very high value and incompatible with divine determinism fosters the beginnings of a reply to the problem of divine love. In this case, one might think God could not prevent people from becoming evil without violating their autonomy. What of preventing them from carrying out their evil wills, bringing pain and suffering to others? Perhaps, the indeterminist may respond, autonomy is not valued where the effects of evil actions are consistently prevented. These last two claims will be familiar to those who are acquainted with the “problem of evil” literature. While such a response does not even attempt to address issues related to natural disasters, it does at least suggest a response to some troubling aspects of our world, and this should not be discounted. Readers will likely be familiar with the objections to these responses as well. I have no intention of evaluating the cycle of objections and responses associated with these replies, but instead rest content noting that indeterminists have the option of taking up such a reply, if they are willing to fight for it. Determinists have no such option. Indeterminists might also claim that there is a difference between doing and allowing that is relevant to the present issue. They might, for example, claim that if God acts so as to bring about pain and suffering for creatures, then we have a good indication that God does not love creatures. In contrast, if God merely allows others to bring pain and suffering on creatures, then we have less evidence of God’s lack of love. The moral relevance of the distinction between doing and allowing is highly contested, and even if it is granted, the indeterminist might not be out of the woods. After all, even with such a distinction in place, we might be left with some reason (even if less) for thinking that God does not love creatures. This reply can be seen, however, as merely a first step that utilizes another reply – for example, the claim that other values can outweigh the good of the beloved.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
The Problem of Creaturely Love
Consider again the case of the general who sends his own child into the midst of an important but very dangerous battle. The indeterminist can claim that God is not like that general, but instead resembles the general who sees that his child has been ordered on an important and dangerous mission, and refrains from countermanding the order. If the moral relevance of the doing/allowing distinction is granted, then the indeterminist will be in a position to suggest that it is easier to see how God’s action could be reconciled with divine love, supposing that other things worth caring about sometimes conflict with the good of individuals. What, then, can be said of the cost of divine determinism in relation to the problem of divine love? In absolute terms, these costs depend upon how taxing one finds the possible responses. If one is already committed to something like the worm account of identity, then there will be no cost at all. For those who find each possibility ludicrous, the costs will be fairly high, even if a disjunctive reply is utilized. What of its costs relative to indeterministic theism? I have sketched only a few possible ways in which indeterminists might respond, so a final determination here is impossible. Nevertheless, it seems to me that indeterminists can utilize all the possible responses open to the determinist, as well as others available solely to them. The relative costs, then, depend upon how plausible the uniquely indeterministic replies are relative to those available to both parties. I leave this determination to the reader.
The Problem of Creaturely Love Earlier in this chapter, I raised and pondered what I called the problem of divine love. I now move on to what I will call the problem of creaturely love. Put succinctly, this problem is that divine determinism posits that God – like the fanciful neurosurgeons populating much contemporary philosophical work – plans each human action before it is performed, and, given the divine plan, the human action could not fail to occur. But there is some reason to think that love produced in this way – especially love directed toward the one pulling the strings – is not genuine love at all. To begin, it is necessary to specify the sort of human love that divine determinism calls into question. Clearly, certain sorts of human love are
Some divine determinists might resist talk of a divine plan, fearing that this would endanger divine simplicity. Such determinists are free to translate talk of the divine plan into language that suits their preferred views. The sense in which the divine plan precedes the action need not, of course, be temporal. Most importantly, given my definition of divine determinism, it must be explanatorily prior to the human action.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
compatible with divine determinism. For example, an agent’s love of oranges, skiing, or Stephen King novels is no less genuine if it was brought about by powerful neuroscientists than if it came about in the usual way. Similarly, divine determinism seems to pose no threat to such superficial loves. Another sort of love, however, does seem endangered. This sort of love is sometimes thought of as producing or being constituted by union. Agents love one another in this sense when they see each other as other selves, when they see the interests, concerns, values, and pains of the other as their own. Theists have often thought of such a union with God as the ultimate goal of human life. Although some see this union love as present when one agent desires union, even if the attitude is unreciprocated, the paradigm cases of this sort of love occur when both agents desire the union and, as Scruton says, “reciprocity becomes community: that is, just so soon as all distinction between my interests and your interests is overcome” (, ). Such a love, in seeking unity, seems to involve a loss of some individual autonomy. Some have even seen such relationships as involving the transference of individual rights. Since I wish to focus on this sort of love, for ease of expression, I will simply refer to this sort of love as “love.” I wish to turn, now, to two arguments, each of which concludes that humans cannot have (authentic) union love with God. The first relies on the following two principles: First Principle of Union Love: One agent, A, can love another, B, only by autonomously surrendering some autonomy to B. Second Principle of Union Love: An agent, A, who surrenders some autonomy to another agent, B, does so autonomously only if B does not control whether A does so.
Some might contend that we should not call such attitudes love at all, but perhaps likes. In everyday life, we certainly do use the English word “love” to cover such attitudes. Whether these attitudes are differentiated from other, deeper attitudes, by explaining them as different sorts of love or as differentiated into likes on the one hand and loves on the other hand does not, it seems to me, constitute a substantive issue. Indeed, focus on such issues would betray a sort of verbal essentialism. For discussions of this sort of love, see Fisher (), Nozick (), Scruton (), and Solomon (). Scruton is talking specifically about romantic love. Often union love seems to take on romantic connotations, although it need not. In any case, some theists and some religious traditions sometimes use the metaphor of romantic love to describe the relationship between God and creature. For a discussion of the claim that this sort of love requires merely the desire for union, see Nozick (). See, for example, Nozick (). Perhaps this sort of love is best thought of as sacrificing autonomy or transferring rights to a newly created third entity, which is the product of the love. The same problems seem to arise if we allow for this understanding, so I have not overcomplicated the formulation with this possibility.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
The Problem of Creaturely Love
If these principles are true, it seems that divine determinism is incompatible with humans loving God (in this sense, of course). This conclusion follows given the claim that if divine determinism is true, then God controls every action and omission of each human agent. Before turning to a closer consideration of this argument, I wish to discuss another argument for the same conclusion. This argument is based upon an analogy, and to explore it we turn to a new case: Paul, a powerful wizard, deeply loves Diya, a girl who has shown no interest in him. Paul cares deeply for Diya, and identifies with her cares, counting both her pains and her pleasures as his. Paul wants Diya to reciprocate his love; he wants, in short, to form the sort of union that typifies union love. He knows, moreover, that he can use his powerful magic to control whatever Diya does. He spends long hours working in his cold basement in the production of a powerful potion. When finished with his work, he secretly pours the potion into some of Diya’s soup. Diya eats the soup, and before she knows it, finds herself loving Paul. She does not see these feelings as foreign, nor does she experience them as an uncontrollable urge (in the way that, for example, some addicts experience their urges), even though the potion does its work in a deterministic manner. Instead, she identifies with the desire to seek a relationship with Paul, carefully examines it, and decides to yield to this desire. Moreover, she meets all (nonhistorical) compatibilist conditions for freedom, and is (it is stipulated) free. Diya enters into a relationship with Paul, sacrificing some of her autonomy by merging her own interests and values with his.
The argument begins with this case, and invites the reader to conclude that because Paul controlled whether Diya chose to sacrifice her autonomy, there is something wrong with the relationship. The “something wrong” may be explicated in a number of ways. One might say, for example, that Diya doesn’t truly love Paul; or that her relationship with Paul does not count as union love; or that if it does, it is somehow defective. I will examine the argument given the conclusion that Diya does not love Paul, but the reader is invited to substitute whichever conclusion is judged to be the most plausible. The argument continues by noting that the case of Paul and Diya is relevantly similar to that of a human seeking a relationship of union with God in a divinely determined world. It concludes that divine determinism is incompatible with a human entering into a loving relationship with God.
The problem of creaturely love is, I think, best envisioned as distinct from that of creaturely freedom, rather than as a particular instance of it. It is, I contend, a problem even if we grant that human agents can be free in divinely determined worlds. For this reason, the problem of creaturely love cannot be isolated unless we suppose that the human agent in question is free.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
I have put forward two arguments, each of which reaches the same conclusion. I put off discussion of the first until the second had been formulated, because I think they are best considered together. The First Principle of Union Love gains plausibility from the nature of union love itself. In uniting with the object of one’s love, in taking upon oneself foreign cares and values as one’s own, one inevitably gives up some autonomy. This fact has even led some to reject the value of union love, finding the loss of autonomy to be an unacceptably high cost. This principle does not simply say that autonomy is lost in this sort of union; that is, perhaps more controversially, it requires an agent to autonomously surrender some autonomy. If the union was entered into in certain ways – for example, as a result of coercion or ignorance – then we would be hardpressed to consider the result to be a loving relationship (of this type). One thing that makes this sort of love so unique is that it includes a willingness to give up some of one’s autonomy. If this willingness is not itself autonomous, then while possibly valuable, the ensuing relationship seems to be of a different sort. The second principle, that the autonomy relevant to the first condition requires autonomy from the object of one’s love, gains plausibility from the case of Paul and Diya. If we grant, at least for the sake of argument, that Diya is a free agent, even after ingesting the potion, then we might think that she possesses the autonomy required for entering into a relationship of union with Paul. If the intuition that she does not love Paul is correct, however, then this cannot be right. Some might find in this case a reason to reject nonhistorical compatibilism. Perhaps this is the moral of this case, but I don’t think so; I think only a weaker conclusion is justified. I think this case suggests (and at this point merely suggests) that whatever the conditions of ordinary freedom, the conditions for the sort of autonomy required for union love are in some sense historically sensitive. In particular, I think this case points directly to the Second Principle of Union Love. For this reason, the argument from analogy is foundational for the argument from the two principles. If the former argument can be avoided, then so, too, can the latter.
Responses to the Problem of Creaturely Love The divine determinist might reply to the problem of creaturely love in several different ways. Since the argument from analogy is the more
See, for example, Singer () and Soble ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Responses to the Problem of Creaturely Love
fundamental argument, I will concentrate on that version here. Three sorts of responses are possible. First, the divine determinist could deny that the case of Paul and Diya is relevantly similar to that of a determining God and a creature. Second, the determinist could accept a hard-line reply, claiming that Diya really does love Paul after all. Third, the determinist could claim that humans are incapable of having the sort of union love under discussion, and (presumably) propose a close analogue that humans can have toward God and that allows theists to say all or nearly all of what they wanted to say about human relationships with the divine. The Cases Are Relevantly Different The first response I wish to consider is that the cases are relevantly different. This response comes in two forms. The first works best for primary determinists; the latter fits more naturally into the Edwardsian model. The first form of this response has recourse to the claim that God’s role as primary cause is fundamentally different than any role a creature might assume. According to this response, Paul’s manipulative action does undermine the possibility of his entering into a relationship of union love with Diya, but the same cannot be said of a determining God and creature. Paul pushes Diya to have a particular attitude toward him, but God does not push creatures any more than authors push the characters of their fictions to act as they do. Instead, God creates and sustains creatures having particular attitudes. Because of this difference, we are not justified in concluding that a determining God’s action eliminates the possibility of a divine-creature relationship of union love. Readers unsympathetic to this view may be growing tired of seeing it trotted out to deal with any and every problem that divine determinists face. I think there is something to this complaint: Primary divine
The work of Derk Pereboom provides a model of this sort of general response, where one attitude or practice is ruled out as impossible but a close analogue is proposed to replace it. Pereboom () generally utilizes this strategy to show that many things we care about – or close analogues of them – can be retained even if we lack free will. There is an interesting question of how Christian determinists should think of grace. Depending on which account is given, grace might end up looking more like Paul’s potion than God’s ordinary causal activity does. Given such an account, this response would need to be abandoned for another strategy. This response is best thought of in the following weak form: We are not justified in concluding that Paul and a determining God are relevantly similar. Of course, the theist may assert that they are relevantly different, thinking that creatures can have relationships of union love with God, but I think this particular response to the problem of creaturely love should be cast as undermining an inference to the claim that God’s action violates the required creaturely autonomy, rather than establishing that it does not do so.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
determinists sometimes do use the image of the authorial model for purposes to which it is ill suited. In this case, however, the determinist’s reply does have some plausibility. Paul’s action is clearly manipulative in a nonphilosophical sense: He violates the moral law by infringing upon Diya’s autonomy for his own good. He violates her autonomy because she preexists his intervention, and he, with duplicity and without permission or authority, alters her deepest and most personal attitudes. God, according to this model, does not treat creatures in a similar way. Creatures do not preexist divine influence, but instead come to be and continue to be merely as a result of such influence. Moreover, their attitudes do not change as a result of divine causation, but instead exist only as effects of divine causation. Finally, (we might presume) God causes creatures to seek union with the divine with the intention of fulfilling the creature, not merely out of selfish motives. Thus, this account posits significant differences between God and Paul, specifically in regard to how they affect the autonomy of those who are acted upon. This response might be further strengthened by an appeal to participation, but only if the determinist is willing to embrace further theoretical commitments. Theories of divine participation vary, but share a commitment to the claim that contingent beings are what they are – that is, have their various attributes and indeed their very existence – merely by sharing in the perfections of being itself. If we accept such a claim, then not only are there no creatures preexisting divine influence, but even after they exist, such creatures have little independence from God. God creates them and sustains them, and their existence and each of their features – their mass and shape, just as much as their thoughts, emotions, and desires – is nothing but a participation in the being of God. If this is so, then the divine influence on creatures is much less foreign than Paul’s influence on Diya. Moreover, it is plausible that God’s influence differs from Paul’s, specifically in terms of the impact on the autonomy of the targets of their actions.
For example, the Consequence Argument cannot be weakened by an appeal to the unique causal role of God. Some theists hold that God is the end of all divine action. Even such theists, however, can claim that the immediate aim of creating creatures yearning for union with God is the creatures’ fulfillment, and that this fulfillment is itself directed to God’s good, by, for example, manifesting divine goodness. In fact, Thomas Aquinas thinks that God is the end of all divine action, but claims “God, in willing his own goodness, wills things other than himself to be in so far as they participate in his goodness” (, chap. ). Thus, God is still importantly dissimilar to Paul within this account.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Responses to the Problem of Creaturely Love
We still might wonder how this reply helps deal with the first formulation of the problem of creaturely love – the one targeted by the two principles of union love. It does so by suggesting that the second principle should be limited to cases of relationships between two creatures. This restriction of the principle isn’t ad hoc, since the second principle is motivated by appeals to cases like that of Paul and Diya. Nevertheless, this reply argues, their case is not clearly relevantly similar to that of a deterministic God and creature. For this reason, the cases can only support a principle restricted to cases of creatures. Edwardsian divine determinists also claim that important differences distinguish God’s causal role and that of Paul, but these differences are less robust than those to which primary determinists can point. They may, however, note that God does not intervene to alter an already existing being, but instead merely puts in place conditions that will bring about creatures who act in a particular way. This does seem to be somewhat less troubling than what Paul does to Diya. Another strategy they might use has been considered in Chapter : They might deny that God creates with the intention of determining creatures to perform specific actions (or even to have certain attitudes and desires). In this case, a creature’s action of seeking union love with God would have been determined by the initial conditions God created; importantly, though, God would have set those initial conditions for other reasons, so the creature’s action would be merely a foreseen but unintended consequence of creating the initial conditions of the world. If we accept this claim, then the cases of God and Paul would seem importantly different, and the force of the argument from analogy would be significantly weakened. Further, the first principle of union love would no longer threaten divine determinism, since on such an account, it seems that there is an important sense in which God does not control creatures. In the end, this reply seems to have considerable power to solve the problem of creaturely love, but, as noted in Chapter , it brings with it considerable costs. Hard-Line Reply Instead of complaining that the cases of Paul and God are relevantly different, divine determinists might simply reject the proposed intuition concerning Paul and Diya in the first place. In other words, they might claim that Diya does retain the ability to enter into a relationship of union love with Paul after she eats the soup and, in fact, exercises this ability shortly thereafter. They might motivate their rejection of this intuition by highlighting important mental states that Diya has after eating the soup.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
Perhaps the potion need not have operated by clouding her mind to Paul’s features, but instead by making her notice various qualities he possesses that she hadn’t noticed before. Moreover, they might claim that as long as the potion works to integrate her new feelings for Paul into her overall psychological makeup, we should think of them as expressing Diya’s current true self. In fact, the potion may leave her in a condition that is qualitatively identical to the one she would have experienced if she had developed loving feelings for Paul all on her own. If this is so, this reply continues, then we might judge that although the method by which the love developed is morally suspect (at the very least), the resulting relationship should be judged on its own features, and these features are sufficient for it to qualify as a relationship of union love. Divine determinists need not say that Paul and Diya have an ideal relationship, or that it is similar to God’s relationship with creatures in all important respects. They should claim that Paul did violate Diya’s autonomy, and so acted contrary to the moral law, in sneaking his potion into her food. Crucially, this violation of autonomy does not render Diya incapable of freely surrendering some of her own autonomy later down the road. Paul’s act may have been despicable, but it doesn’t take away Diya’s abilities governing her autonomy. Such a reply might be considered only partially successful. After all, even if Paul’s act fails to undermine Diya’s ability to love him, it does seem that, because it violates Diya’s autonomy, it is incompatible with his unselfishly and (therefore) genuinely loving her. Similarly, divine determinism might not be perceived as undermining the ability of humans to seek union love with God, but it might erode the possibility of this love being reciprocated. I think that this concern is misguided. God, determinists should claim, does not violate human autonomy, and therefore does no harm to humans in determining them to seek union with the divine, since God does not act so as to subvert the normal functioning of humans. Unlike Paul, God does not radically and artificially alter an agent’s preexisting mental condition by inserting a desire for union. Instead, God either creates agents together with all their mental states (as primary determinists have it) or brings them about by creating the initial conditions of the universe, which deterministically bring them into existence and slowly bring about their mature mental lives in a perfectly natural way.
This further objection is not merely a retreading of the problem of divine love, since that issue concerned whether the caring love of a parental God might be undermined by divine determinism, rather than the union-seeking love we are now considering.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Conclusion
Impersonal Love of God In this chapter, I have discussed two general strategies for solving the problem of creaturely love: deny that Diya and divinely determined humans are relevantly similar or deny that Diya loses the ability to love Paul. A third avenue for dealing with this problem is to deny that the conclusion is to be avoided. This response accepts the conclusion of the arguments presented earlier: Given divine determinism, God and humans cannot share a relationship of union love, at least not if we characterize this relationship alongside cases of human interpersonal relationships. Human–divine love might sometimes be characterized, metaphorically, as that between friends, or even that between lovers, but these metaphors inevitably fall short of the truth; indeed, they can be misleading if interpreted too strictly. The love between a human agent and a divine one can never be one between two autonomous individuals – for among all agents, only God is truly autonomous. The human yearning for union with the divine cannot be thought of as that of a perfectly autonomous creature freely and spontaneously surrendering some of its autonomy, but instead of a radically dependent being seeking a return to its source and natural end. In seeking union with the divine, creatures do not seek union with something that is truly foreign and other, as two human agents are relative to one another, but instead seek union with a creator that has been, necessarily, imminently present to them throughout their existence. For this reason, to worry about a lack of autonomy on the part of the human agent is to be caught up in confusion about the sort of union that is sought. Once this is recognized, the problem of creaturely love reveals itself not as a concern for divine determinists, but rather as a useful reminder that union with God is not, in the end, just one more interpersonal relationship alongside others.
Conclusion In this chapter, I have investigated two ways in which divine determinism may be thought to imperil important sorts of love. First, divine determinism, combined with what we know of the evils of the world, forces us to conclude that God does not love us – at least in a way that satisfies the model of perfect parental love that is often suggested. What should we make of this problem? Several responses are failures, but others show promise. Unfortunately, each of the four promising responses, which seem representative of the available options, engenders significant costs. Thus, the problem of divine love must be taken seriously; even if one of the
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
God, Determined Agents, and Love
suggested responses is adopted, it is important to remember that divine determinism exposes some of its costs when confronted with this problem. The second problem, that of creaturely love, can be more easily handled. Of the solutions discussed, only the Edwardsian attempt to deny the relevant similarity between Diya and ordinary humans seems costly or forced. Others may also balk at the hard-line response to this problem, but this still leaves two satisfactory responses. The first of these is best utilized by only some divine determinists, specifically those who adopt primary divine determinism. The second, which asks that we think of loving union with God as merely analogous to ordinary relationships of union love, is open to determinists of every stripe.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.008
Divine Commands, the Divine Will, and Divine Blame
Introduction According to many theists, God gives commands that are often broken. But, if determinism is true, then God causes these commands to be broken. Since God’s commands must be in keeping with the divine will, it seems that God gives commands and wills that they be broken. There is something deeply troubling about this claim. Further issues arise on the side of the one commanded. Theists typically think that humans are morally obligated to obey divine commands, but what of cases in which a human agent (justifiably) believes that God wishes the divine command to be broken? Should the human agent act in accord with the divine will or the divine command? Given divine determinism, such cases need not be rare. Another problem arises from considerations focused on an agent’s view of past sins. It is not uncommon for theists to think that humans should aim to achieve a loving relationship with God together with a perfect surrender to the divine will. Now, if someone is in such a relationship with God, then acts in keeping with divine commands will follow; thus, it is clear that God willed the agent to act in keeping with divine commands. There are no worries, then, about present or presumably future actions for such agents. Nevertheless, a problem remains about how such virtuous agents ought to view their past sins. Since these sins existed, it is clear that God willed that they be performed. On the one hand, agents, in wishing that such sins not occur, would be wishing to have (per impossibile) violated the divine will. On the other hand, if agents refrain from wishing that such sins had not occurred, then they fall short of repentance, which many theists find to be a necessity for
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
union with God. Divine determinists, it seems, face a dilemma concerning virtuous agents with checkered pasts. The final set of worries concerns divine blame and punishment. According to divine determinism, God has infallibly ordained that agents act in particular, sometimes immoral, ways. Many theists also think that at some point, whether during agents’ earthly lives or after, God will pass judgment over them and inflict punishment upon them. But there seems to be something particularly perverse about such divine behavior. Divine determinists – at least those interested in maintaining traditional views of divine punishment – must resist such a claim, but it is far from clear how they should go about doing so.
Divine Determinism and the Giving of Divine Commands Katherin Rogers has recently argued that divine determinists face a difficulty in making sense of divine commands. In particular, she argues that in giving what she calls a sincere command, an agent intends the subordinate to gain two sorts of beliefs. First, the command “must instill the belief that the commander wants to be obeyed” (, ). I will call such a belief a “desire belief.” Second, “the command must intend to instill the belief that it is possible for it to be obeyed” (, ). I will call such a belief a “possibility belief.” If Rogers is correct, and if divine determinism is true, then it turns out that God, in giving commands, becomes a deceiver. For, given divine determinism, Rogers claims, “when God commands He deliberately deceives in that He knowingly conveys that He wants to be obeyed and He believes it is possible that He should be obeyed, when in fact, in the cases where He causes sin, He does not really want to be obeyed and He knows it is not possible for the sinner to have obeyed in any case” (, ). I wish to consider possibility beliefs first. Rogers claims that serious commands aim to instill the belief that the action commanded is possible for the agent to perform. In defense of this claim, she points to the principle that “ought implies can” (OIC). Her use of this principle seems to imply that it makes sense to give a command only if, given the command, the one commanded ought to obey. It is not obvious to me that this is true, but most theists will admit that divine commands ought to be obeyed.
This assumes, of course, that the agents in question have sinful pasts. In some traditions, some human agents are considered to have nothing to repent from. In some traditions, for example, Mary, St. John the Baptist, and infants who died before becoming morally responsible agents are thought to have nothing of which to repent. Rogers is specifically targeting the views of Hugh McCann in this article, but her worries generalize.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Determinism and the Giving of Divine Commands
Should determinists be swayed by these considerations? I don’t think they should. Determinists will already need to deal with a structurally similar argument against moral responsibility, one that suggests since agents cannot avoid the actions they perform, it is false that they ought to have avoided performing them. Compatibilists have typically dealt with this worry in one of two ways. First, they might simply deny the OIC principle. Alternatively, they might offer some compatibilist-friendly account of “can” according to which agents can do particular things that they are determined not to do, even if they are so determined by factors beyond their control. Compatibilists have long been focused upon worries concerning what agents can and cannot do and how this relates to moral evaluation. They already have to deal with the OIC principle in one way or another – either by denying it or by arguing that in some important sense agents can do things they are determined not to do. Thus, any argument against compatibilists that appeals to the principle for support will have little dialectical power. In the present argument, OIC is cited as a reason to endorse the claim that those giving commands must intend to instill the belief that the command can be obeyed. Rogers argues that, under divine determinism, many of these commands cannot be obeyed, so God is involved in deceit. But some compatibilists already deny OIC, in which case the giving of a command would not be meant to instill any possibility belief at all. Alternatively, they might accept OIC and think that determined agents can perform acts that they have been determined not to do. In this case, each divine command can be obeyed, so there is no need to conclude that God is a deceiver. My conclusion is not that Rogers’s conclusion is wrong, but rather that it doesn’t represent a distinct problem for divine determinists; it is merely a particularized version of a problem that all divine determinists – at least those intent on preserving some sort of robust set of moral obligations – face. If one is a compatibilist, then this larger problem cannot have failed to present itself; thus, it is safe to assume that any particularized version of it poses no additional threat.
Some recent work in experimental philosophy supports the idea that adoption of the OIC principle is not as widely held among nonphilosophers as is sometimes thought (Turri ). For a discussion of the history of the rejection of this view among influential theists, most notably Augustine, see Couenhoven (). This is the route taken by classical compatibilists. Although this path has lost its status as the default view among compatibilists, it or something similar to it still has able defenders. See, for example, Fara (), Smith (), and Vihvelin (, ). New and compelling arguments that OIC is true, or that compatibilist-friendly accounts of “can” fail, might do so, but additional arguments appealing to OIC and merely claiming that determined agents cannot do otherwise cannot.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
Although claims about possibility beliefs will be of little use in arguments against determinists, Rogers need not use claims about possibility beliefs to reach her conclusion that divine commands given by a determining God are deceitful. Remember, she also contends that serious commands are meant to instill the belief that the one giving the command wants to be obeyed. How might determinists respond to this claim? One might either deny that commands carry with them an intent to instill desire beliefs, or assert that there is an important sense in which God does wish for all divine commands to be obeyed, and this sense is sufficient to alleviate worries about deceit. McCann, in a response to Rogers’s worries, combines these approaches. He claims, “At best there is a widespread conversational implicature that commanders want their commands to be obeyed” (, ). He defends this claim with an appeal to the divine command that Abraham sacrifice Isaac, and also provides the following case to support his view: “A parent might command a rebellious child to perform an action A, not because he wants A done but because he knows the command will induce the child to perform B instead, which is the action the parent really wants performed. And again, this situation does not take away the child’s obligation to do A” (, ). Unfortunately, as Rogers (, ) notes, the latter case clearly involves deceit: The parent gets the rebellious child to act in a particular way through an act of deceit in which the parent’s true desires are hidden. The case of Abraham and Isaac is doubly problematic, both because it is among the most controverted passages in sacred texts and because the deception involved in such divine commands is precisely the issue at hand. Luckily, the success of McCann’s response does not depend solely on these problematic cases. He also argues that there is a sense in which God wills that each divine command be obeyed. Following a long theological tradition, McCann distinguishes between God’s antecedent and consequent will. This distinction is based upon the idea that, when we look simply at the nature of something, it might be good and thus desirable. For example, I might rightly judge that a trip to the beach is a good thing, and so desire it. Since this desire is simply addressed to the goodness of the thing itself, and is prior or antecedent to any consideration of the specific circumstances in which I find myself, this is called my antecedent will. Suppose, however, that I have much needed work to attend to and need to
McCann alludes to this in more than one place (, ; , ). See John of Damascus () for an early discussion of this distinction and Thomas Aquinas (, I, q. , a. , ad ) for a discussion of this distinction more congenial to McCann’s use of it.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Determinism and the Giving of Divine Commands
take one of my children to the dentist’s office in two hours. In that case, I judge that today is not a good day to go to the beach, so my desire, consequent to a full consideration of circumstances, is to avoid the beach today. In other words, it is my consequent will or desire that I not go to the beach. We might be tempted to think that the distinction between antecedent and consequent desires is not applicable to God, at least if classical theists are correct in thinking that God is both simple and immutable. This temptation arises from the thought that the distinction between antecedent and consequent desires applies only to beings capable of successive reasoning, which requires both complexity and mutability. But this temptation should be fought. The distinction between antecedent and consequent desires may apply to any being capable of distinguishing the goodness (and thus desirability) that a thing or event has in isolation from the goodness of a world in which the thing in question is possessed or the event occurs. I can recognize that the event of going to the beach is a good thing (thus having an antecedent desire to go to the beach), while simultaneously recognizing that, overall, things would be worse if I were to go to the beach. Similarly, I see no reason to deny that God can recognize that an event is good (perhaps, say, a human life lived with no need to struggle to succeed), while recognizing that, all things considered, things would be better if humans did sometimes need to struggle to succeed. In support of this conclusion, we can note that even in the case of humans, antecedent desires need not preexist consequent ones, nor must they cease when consequent desires arise. Upon seeing a new laptop, I might immediately form both the antecedent desire to possess the laptop and the consequent desire to save my money for other purchases. This need not even require any successive reasoning. Indeed, it seems that consequent desires concerning an object can preexist antecedent desires for it. I may, for example, see an announcement for a new laptop without having any antecedent desire for it at all – not knowing anything about its capabilities. Nevertheless, I will have a consequent desire not to buy it, since I know my budget is tight. Later, upon hearing a friend sing the laptop’s praises, I might gain an antecedent desire for it, while never altering my consequent desire to refrain from buying it. We might now look at a case of a disobeyed command in light of this distinction. Suppose divine determinism is true and that there is a divine command against stealing. Now imagine Kira, who believes that God has given this command, and believes that this command has been directed to all people, including her. Nevertheless, she freely and knowingly breaks
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
this command. In this case, it is clear that God antecedently wills that Kira refrain from stealing, since stealing is evil in itself. It is equally clear that God consequently wills that Kira steals at this moment, since this is exactly what happens as a direct or indirect result of God’s will. McCann does not tell us how this distinction is supposed to help solve the problem that Rogers has raised. Presumably, the idea is that even if we assume that commands generally inculcate a belief that the commander desires the command to be obeyed, commanders do not deceive so long as they have at least an antecedent desire to be obeyed. McCann’s response seems to require such a claim, but it is difficult to determine whether it is true. Surely most commanders give commands precisely because they have a consequent desire that the commanded action be performed, but neither McCann nor Rogers considers ordinary cases where the commanders have merely an antecedent desire for their commands to be fulfilled. Consider Matthew, who is in charge of overseeing basic training to a group of raw soldiers. He gives many orders to those under his charge, each of which is carefully designed to help recruits develop the skills they will need to succeed. Some of these orders are meant to develop in the recruits a greater respect for the dangerous instruments they will use throughout their careers. With some past groups, one or more of these orders were disobeyed, and the disobedient party was injured as a direct result of failing to obey. Matthew has noticed that such events bring about a net benefit for both the injured recruit and the larger group that witnessed both the disobedience and the resultant injury. These recruits develop a far higher respect for the dangerous weapons than they would have otherwise. With conclusions from his past experience in mind, Matthew gives these new recruits the same orders he has given many others in the past. These are good orders and, antecedent to all other considerations, Matthew desires that they be obeyed. Yet, all things considered, Matthew desires that a few of them are broken at least a few times. Does Matthew intend to deceive when he gives his commands? I think not. Moreover, Matthew is like a determining God in several important ways: Both give orders to groups of people, antecedently desiring that each order is always obeyed, but consequently willing that some commands are, in some instances, broken. Nevertheless, there are also some important differences. As the case is set up, Matthew does not have a desire – even a consequent one – that some particular recruit disobey a particular order at a particular time. In contrast, a determining God does have a desire that particular agents perform particular sins at particular times. This makes the worry that God deceives in divine commands worse than that Matthew
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Determinism and the Giving of Divine Commands
deceives in issuing his orders. There is no single recruit, who, if learning the whole truth about Matthew’s orders, could stand and say, “But you don’t really want me to follow this order, at least not this time.” Thus, it seems that no single recruit is being deceived. Given determinism, the case with God is different. Knowing the whole truth, certain agents, upon hearing a divine command, could stand and say, “But you don’t really want me to follow this rule this time, so by commanding me you are just pretending to want this done.” This complaint does not seem unreasonable. For this reason, I think that the divine determinist will need to do more to answer this objection than to gesture toward the distinction between antecedent and consequent desires. Instead, I suggest, the divine determinist should explain that divine commands must be thought of within a larger context of morality and divine–human relations. Presumably divine commands are given not merely to make clear the divine will, but to instruct agents in how to live a good life. Thus, divine commands are meant to instill the belief that obedience to the command will aid in the pursuit of fulfillment. This belief is true, even given divine determinism. Additionally, theists often admit that we should be humble about our ability to know the contents of the divine mind, and at least initially skeptical about why God might do or allow this or that. If this is so, then theists should not be quick to judge that particular commands reveal the complete set of divine desires, both antecedent and consequent, for all creatures. Instead, they should reach initial, rather tentative conclusions, which are subject to revision based upon further consideration. Divine determinists will say that such consideration (whether based on reason alone or dependent upon revelation as well) will lead to the conclusion that God’s consequent will is in accord with what actually occurs. Since divine commands are broken, we must conclude that God wills them to be, in at least some sense. Are these considerations sufficient to deflect the claim that a determining God is a deceiver? Should we conclude that if divine commands do instill true beliefs about human fulfillment and the nature of the good, and if they are in accord with the divine antecedent will, and if we should be (at least initially) cautious about reaching conclusions about the divine mind, and if further possible consideration would lead to an accurate conclusion about God’s will in relation to these commands, then
McCann adopts a divine command theory, but also thinks that these commands reflect ways of reaching fulfillment given the natures that God has created. See McCann (, ). Of course, many theists – even those who say it is reasonable to be cautious in this regard – are quick to reach conclusions about divine intentions, but this sort of action does not shed doubt on the advice to be cautious.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
giving these commands is not deceptive? Even if such considerations do not decisively rebut Rogers’s objection, they do offer the possibility of a reasonable response. In any case, such considerations are the best that the divine determinist can do.
Following Orders or Carrying out God’s Will? In the preceding section, I sketched out a proposed response by which the divine determinist might resist the charge that a determining God’s commands are deceitful. Unfortunately for determinists, this is not the only worry about divine commands they must deal with. We have considered difficulties with whether God should give such commands, given the divine will. Now we must turn to the question of whether agents ought to obey such commands when they believe that God wills them to do otherwise. We might think that it is impossible to know the will of God, but, given divine determinism, this is not as difficult as it seems. God’s antecedent will might be difficult to ascertain, but God’s consequent will is revealed in the way events play out in our world. If Alejandro is born, then God has willed that Alejandro be born; if Danielle dies, then God has willed that Danielle die. God’s will concerning future events is, of course, a more difficult matter. Nevertheless, agents may still make plans, knowing that, if they succeed in their plans, they have done God’s will. This ability to make plans, being sure that if they are fulfilled, they are in accord with the divine will, presents a problem. It provides a way for agents to justify any action they wish to perform in light of God’s will. Suppose, for example, that James is a divine determinist. Suppose, furthermore, that James’s belief in divine determinism is both true and justified. James finds himself in a difficult position at work. His superior has offered him $, to falsify some paperwork and thereby hide some of the superior’s past misdeeds. James wishes to take this offer, and has no worries about its worldly consequences. But he is a committed theist and, in deciding how to act, is guided by a desire to act as God wills. In fact, he is so committed to his beliefs that if he decides that this act is against God’s will, he will not perform it. At first, James plans to reject the offer, but then the following line of reasoning presents itself: Although God’s antecedent will is revealed in divine commands, his consequent will is not. God’s consequent will is difficult to discern ahead of time, but does reveal itself in the way events play out. What does God will in this case? James isn’t sure, but he is not without resources. If God wishes that he take the offer, then
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Following Orders or Carrying out God’s Will?
he will; if God wishes that he decline the offer, then he will do that instead. James decides on a plan of action: He decides to take the offer, knowing that if he succeeds in so doing, then he is doing God’s will, and can be content. If he fails to carry out his plan somehow, then he can still rest content that God wanted him to decide to take the offer, since that is, in fact, what he has done. Since his action is motivated by and in keeping with his justified and true beliefs about the divine will, he concludes that he does not sin. We might think that there is something amiss in this case. What sense does it make for someone who believes in divine determinism to adopt the deliberative principle of following God’s will? After all, whatever action is performed by anyone at any time is willed by God. I don’t think that James, in adopting such a principle, is being practically irrational. Even if he will inevitably follow God’s will, James is not guaranteed to perform each action because of a motive to follow God’s will. Supposing (as we are assuming is the case) that divine determinism does not undermine human moral responsibility, it seems more praiseworthy to act out of a motive of doing God’s will than merely to follow God’s will in what, from the perspective of the human agent, would seem to be a happy accident. Although, given divine determinism, humans always follow God’s divine will, a difference, even a moral difference, is still possible between acts in keeping with the divine will and acts performed out of a desire to follow the divine will. Given this conclusion, the case of James presents a problem for the determinist. Such a case does not show that people do not sin, but it does suggest that divine determinists have a way to avoid blame on every occasion. Notice that James is not like sinners who try to justify their actions at some later date by complaining that these actions were out of their control or in keeping with God’s will anyway. Instead, James is intentionally guiding his deliberation in light of his beliefs about the divine will – beliefs that, in this case, are both justified and true. Finally, his case generalizes: James can use similar reasoning in every case that ever presents itself. He need not even walk through this line of reasoning continuously; instead, he can adopt methods of action based upon this line of reasoning, such that doing so will always ensure that both his deliberative processes and his acts are guided by a concern for the divine will. How should determinists respond to this case? Surely they should not praise James’s reasoning or accept that he doesn’t sin. This leaves them with two possible lines of response. First, they might say that James is merely fooling himself. If he truly cared about the divine will, he would
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
simply follow the commands God has given. His supposed deliberation is nothing but sham reasoning, an attempt to justify a course of action that he intends to pursue anyway. If it is stipulated that James’s reasoning is not a sham, that he is dedicated to following the divine will, then, this response continues, such cases are impossible. As a matter of our psychology, or perhaps of divine grace, agents committed to the divine will cannot reason in this way. I find this line of response unpersuasive. Although it might be true that many people find ways to convince themselves that their immoral actions are the will of God, I see no reason to think that the case of James is impossible. I conclude that this line of response is not viable. This leaves the determinist with no choice but to say that deliberating in light of the divine will is not sufficient; agents must look also to divine commands. As McCann says, “it is God’s commands that are crucial here, not his will” (, ). This might seem to suggest that divine commands take precedence over the divine will. Perhaps such a conclusion is unsettling – we will look further into this issue in the next section – but there is no need for the determinist to weigh the value of following the commands or the divine will in response to the present difficulty. Instead, they should point out that James should guide his deliberation with the thought that no matter what he does, he will fulfill God’s consequent will. If he chooses to decline the offer, he additionally fulfills God’s commands. This argument does not suggest that the standing intention of fulfilling the divine will is empty of value, nor is the determinist forced to say that obeying commands is more important than following the divine will. Instead, it merely states that human action should be at least partially guided by the desire to fulfill divine commands. Crucially, this principle of action guidance need not be seen as either replacing or governing the principle that agents should aim to act in accord with the divine will. Of course, there is some sense in which agents sometimes cannot both follow divine commands and act as God wills. This is clear from the fact that they do sin and, therefore, God must have willed that they sin. While such a conclusion is unavoidable for divine determinists, this claim need not infect agential deliberation. If James knew ahead of time that God
I do not mean to deny that there may be some sense – perhaps a crucially important one – in which some determined agents can do what they are determined not to do. Even if this is so, surely there is some sense in which they cannot do other than they do.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Following Orders or Carrying out God’s Will?
willed that he take the offer, then his deliberative process would require a consideration of how to weigh divine commands against the divine will. Since James lacks all such knowledge, there is no need for him to weigh these two factors against each other, even if it is true that in this case God does will him to sin. What the determinist must deny is that agents may rightly weigh the divine will over divine commands. If they were to do so, then they might adopt the aim to follow divine commands in all cases except those in which God wills them not to do so. Imagine what James’s deliberation would look like if he accepted a principle of action guidance in according with such an aim. He faces a decision between accepting the offer, which violates a divine command, and refraining from doing so. He guides his actions, and in turn his deliberation, by the principles that he should act in keeping with the divine will and that he should follow divine commands unless God wills that he not do so. He notices that if he accepts the offer, then he would be successful in following both of these principles. If he succeeds in accepting the offer, then he has (given the truth of divine determinism) strong evidence that he acted as God willed and that God willed that he break the divine command on this occasion. If, in contrast, he succeeds in declining the offer, then he also is successful in following both of his principles of action guidance. In this case, he obeyed the divine command, and once again (as is always the case) his act is in keeping with the divine will. Since both possible actions satisfy both of his principles, James seems free to choose between them as he wishes. Because this conclusion is an odious one, the determinist must reject the principle that we should follow divine commands unless God wills otherwise. Following the divine will and obeying divine commands may be seen as equal or perhaps incommensurable goods, or divine commands may be seen as the more important guide to human behavior. Earlier in this chapter, I suggested that determinists can avoid the problem under discussion by claiming that intentionally fulfilling the divine will is insufficient for avoiding blame. Some might not take this to be much of a cost, whereas others will demur. In any case, it is easier to assess such a cost if its contours are clearly delineated. I suggest that the cost of responding to the problem under consideration is captured by noting that divine determinists must accept the following: Principle of Non-Excuse: An agent A may sin (or is not excused) in performing φ even if A intended to φ with the goal of fulfilling the divine will and
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
Before performing φ A is truly and justifiably believes that God will’s A to φ and A would not have performed φ if A believed doing so was against the divine will and The goal with the strongest motivational power to influence A’s actions is to conform all acts with the divine will and A’s deliberation is governed by the desire to obey the divine commands in all cases in which God wills that the command is obeyed.
As I have continuously noted, different philosophers assess costs differently. Nevertheless, before turning to other concerns, I wish to note that there is something uncomfortable about accepting the Principle of NonExcuse together with the claim that human agents should strive for union love with God. Recall that union love involves the adoption of another’s values, cares, concerns, and desires as one’s own. Clearly, one’s own deliberations should be guided by desires (even if second- or higher order desires). If human agents strive to adopt God’s desires as their own, then they should be striving to let God’s own desires inform their deliberations. In such a model, rightful human action is that which is successfully guided by accurate beliefs about God’s values, cares, concerns, and desires. Such a model does not naturally suggest the Principle of Non-Excuse.
The Problem of Contrition Earlier, I considered a problem concerning how agents might deliberate about future actions in light of the belief that God sometimes wills for divine commands to be broken. I now wish to turn from forward-looking agential considerations to backward-looking ones. In particular, I wish to address how virtuous agents ought to consider their own past sins. Ordinarily, theists tend to think that virtuous agents should have sorrow or contrition for their past faults. They ought, it seems, to be sad that they had committed such sins in the past, wishing that they had not done so. But many theists also think that perfectly virtuous agents should hope that God’s will be fulfilled. Unfortunately, these wishes are in tension with each other.
A may be justified in this belief because of the possession of good evidence that A will φ. Such evidence might consist of A’s strong desire to φ together with a strong habit of following such desires, as long as they do not conflict with a belief that it is against God’s will. Such an attitude is highlighted in the Christian Lord’s Prayer, where agents petition God, requesting “thy will be done.” Such an attitude is, of course, not unique to the Christian tradition.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Contrition
Consider again James, who was asked to deceitfully modify documents. Let’s suppose that James accepts the offer, and so sins. Later, James reflects on his act, concludes that he ought not to have done this, and seeks reconciliation with God. Presumably, James should intend to avoid such actions in the future, but also should be sorry that he performed this act in the first place. Upon accepting the wrongness of his action, he finds sorrow naturally welling up within him. Being a divine determinist, though, he wonders whether this is the correct response. Although James recognizes that his act was wrong and that he acted immorally, he also recognizes that his sin was part of the divine plan – not just something that God might work for the good, but rather an element that was willed as part of the divine plan “from the beginning.” If James wills that God’s will had been done in the past, then James’s will is in keeping with his past sin after all. In such a case, at t, the point of repentance, he continues to will that he sins at t, the moment when he sinned. This does not seem like contrition. If, in contrast, he wishes that he had not sinned in the first place, then he seems to wish that God’s will had not been carried out. This seems to be impiety. The only other option, it seems, would be for James to will a state of affairs in which he did not sin and in which God did not will for him to sin. But to will that God’s will be different seems to be presumptuous. Each of these options seems less than ideal. Theists who reject divine determinism can appeal to a relationship between the divine will and divine commands that seems in all ways more natural: God wills that each divine command is followed. When a command is disobeyed – that is, when evil is done – God can work to bring good out of it. However, in the ideal world, the world God wants, there is no need for any backup plans. Moreover, when agents sin, they should repent of their sin, wishing that they had not sinned and that they had carried out God’s will. Such a view seems natural, and nondeterminists can say what seems reasonable: Wishing that God’s will had been done constitutes a wish that the sin had not been performed. Not only is there no contradiction between the two wishes, but there need not even be a distinction between the two. Determinists are at a decided disadvantage here. In their view, a wish to have avoided a past sin constitutes a wish for God’s will not to have been done (or to have been otherwise than it is). This might not be terribly problematic if this constitution relation goes unrecognized, but for agents who accept divine determinism – for an agent such as James – the problem cannot be ignored.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
The path open to the divine determinist is to embrace one horn of the trilemma and make it seem as palatable as possible. Of course, this leaves us with three options. The first option is to admit that repentant sinners should not wish that they had not sinned in the first place. Here the determinist may appeal to a distinction we discussed earlier, that between one’s antecedent and consequent will. This will be important, because even if determinists admit that repentance does not require desiring in some particular sense that an earlier sin had not been performed, they should surely claim that repentance requires recognizing the evil of sin and desiring – again, in some sense – that no such actions should be performed. These senses of desiring are captured by the distinction between antecedent and consequent desires discussed earlier. Repentants ought to antecedently will that they had never committed any evil acts whatsoever. In light of the role these sins play in the divine plan, as well as their property of being willed by God, sinners should will for their pasts to be precisely as they are, sins and all. In other words, consequently, they ought to will that their pasts be just as they are, however full of sins. This response might be further developed by noting that repentance lies not in consequently willing that the past had been different, but in having a set of responses – notably, recognition and abhorrence of the past sin, asking for forgiveness and working for restitution, and an intention to avoid all such further actions. An agent adopting such a stance would, our imagined determinist maintains, qualify as repentant. If such a stance seems somehow deficient, this appearance arises not because of any true defect, but rather because of a mistaken assumption about the relationship between divine commands and the divine will. It might be claimed that once one sets aside the assumption that divine commands and the divine will are always in accord, it becomes clear that this account of repentance both offers all that can be reasonably hoped for and is entirely adequate. The second option is to say that repentant sinners should wish that they had not sinned in the past, even though their doing so was in accord with the divine will. Determinists who pursue this option can also make use of the distinction between antecedent and consequent wills. They may say that the repentant sinner antecedently and consequently desires that God’s
This position bears some resemblance to an idea put forward by Chrysippus, and noted and affirmed by Epictetus, that if we knew we were fated to be sick, we should move toward this, rather than away from it. See Epictetus (). The present consideration goes further than this, since the Stoics see illness, unlike wrongdoing, as an indifferent.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Contrition
antecedent (but not consequent) will be satisfied. In this view, God’s commands (or the moral law taken in some other way) trumps allegiance to the divine consequent will. Mark Murphy () adopts such a position, albeit not in any attempt to make sense of divine determinism. Instead, Murphy makes these claims in the context of discussing the biblical story of the slaughter of the Jerichoites. Murphy argues that even if this story is literally true, and God did order the slaughter of all these people, this fact would not force us to conclude that God acted immorally. Although most of his time is spent defending his claim, he also considers how we should react to this event. Murphy suggests that there is no need for us to will the destruction of these people, even if we believe that God did so. Although we must formally will what God wills – that is, we should will everything under the aspect of universal goodness – there is no need for each of the particular things we will to correspond to each of the particular things God wills. This response fits well with a claim to which divine determinists – at least those who might be concerned with the problems of this chapter – are already committed: Moral obligations of created agents are not determined (or solely determined) by a criterion of conformity to the divine (consequent) will. Moreover, such determinists already admit that agents are sometimes morally obligated to perform actions whose performance would contravene the divine (consequent) will. James was morally obligated to turn down the immoral offer, even though his doing so was against the divine (consequent) will. The present difficulty – that repentant sinners should will that God’s consequent will had not been fulfilled – fits well with this earlier claim, but is not identical to it. The present difficulty is that this response foists upon us the claim that some existent good actions actively involve willing that the divine will not have been carried out. Earlier determinist claims might make this pill easier to swallow (a fact, if it is one, that determinism’s enemies will hold up as further evidence against the position), but it remains an additional pill just the same. The third option in dealing with this trilemma is to claim that repentant sinners should wish that they had not sinned, while additionally willing
Murphy points to Aquinas’s claims in ST, I-II, q. , a. , but it is unclear whether he can utilize this text in the way he wishes. In the reply to the first objection, Aquinas () writes, “But in the state of glory, every one will see in each thing that he wills, the relation of that thing to what God wills in that particular matter. Consequently he will conform his will to God in all things not only formally, but also materially.” The point seems to be that if we know what God materially wills, or what He wills in particular circumstances, then we must do so as well. In the case of the Jerichoites, we are supposing that we do know what particular event God wills, not just the general aspect (the good of the universe) in which all divine volitions are willed.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
that God’s will had been different. If God’s will had been different, if God had willed that they never sin in the first place, then there would have been no tension between God’s will and the continual innocence of these agents. This reply requires that repentant agents consequently will each of the following: (a) That, contrary to fact, their pasts had been free from sin (b) That God have no unfulfilled consequent volitions (c) That, contrary to fact, God had consequently willed that their pasts had been free from sin Willing (a) and (b) seem perfectly reasonable. Willing (c), however, is troubling. Willing (c) is problematic because in so willing, agents seem to place themselves in a position of judgment over the divine plan, established to bring about the good of the universe. Two problems arise here. First, assuming such a position seems to be the height of intellectual presumptuousness. In assuming to pass judgment over the divine plan – in willing that it had been otherwise – human agents seem to suppose that they are in an epistemic position to evaluate the weaknesses of the plan and suggest repairs for any deficiencies they find. But it is unlikely that any of us is in such an epistemic position. A second problem appears based upon the sort of relation that creatures bear to God. It seems that willing (c) violates the norms of various sorts of relationships that many theists think humans and God either should or do enter into. If we suppose that God bears a parental relation to humans, and particularly if we think that God is an especially good or perfect parent, then willing (c) seems to constitute an act of filial impiety. Alternatively (or in addition), if we suppose that God and humans ought to enter into a relationship of union love, then we must look again to the nature of this union. Recall that it involves viewing the other as another self, accepting the concerns, goals, and desires of the other as one’s own. Consequently willing things that are opposed to the
Recent arguments against skeptical theism might be suggested to provide a reason to reject this claim, but this is not so. Responses to skeptical theism propose that we are in something of a position to judge whether, if there were good reasons for an omnipotent, omniscient, perfectly good being for allowing evil, we would be able to discern them. But such responses – correctly, given the dialectical context in which they appear – do not presuppose that such a being exists. Our repentant sinners are presupposing that such a being exists. They are not in a position of trying to discern whether there might be justifying reasons for evil, but instead already believe that such reasons exist. The task seemingly presupposed by willing (c) is of determining whether the divine plan could be even better – whether, even though God’s willing of their past sins must have been justified, God’s (consequently) willing that they had not sinned would have been even better.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Divine Blame
consequent will of the beloved, together with willing that the beloved have different volitions, seems to betray a defect in the union. Such a response may be justified if the volitions of the beloved are themselves immoral, but surely the repentant sinner is not to think this of God’s will. Ultimately, the defender of this option must appeal to claims similar to those used to defend the second option: There is no need for us to will what God wills, and indeed, there is no need to believe that God wills us to will every particular aspect of the divine will. Instead, we act as we should when we will (b) and satisfy the divine commands. The defender may claim that filial piety is found not in acting in accordance with a parent’s will, but instead in obedience. Since God commands repentance for sins (we may suppose), the pious child will obey, even if this means willing that God’s plan had been different. Moreover, the defender might note that even perfect union love, at least of the sort God and humans might share, should respect some remaining distinctions between volitions and allow for some conflict in certain cases. In particular, conflict is acceptable and even required in cases where one’s own good conflicts with the good rightly willed by the beloved. Each of these responses carves out a stance that makes sense of repentance within a divinely determined world. Each, however, sacrifices something that nondeterminists can easily secure, whether it is that repentant agents should wish they had never sinned in the first place or a complete correspondence between the actual consequent will of God and the consequent will of those that repent. Insofar as they are forced to sacrifice one of these desiderata, their view brings with it an additional cost.
The Problem of Divine Blame According to many theists, God will first pass judgment and then inflict punishment on the wicked. Some find this outcome troubling in itself, but the notion of a retributive God becomes far harder to defend when it is
Antecedently willing things that happen to be against the consequent will of the beloved does not represent any problem. After all, we antecedently will things that go against our own consequent desires. This claim has some plausibility apart from divine determinism, as demonstrated by nondeterminists’ willingness to adopt it. See, for example, Rogers (, –). For an ideal case, we might imagine agents who are currently in a relationship of union love with God and learn that God wills that they eventually betray this relationship and are damned. Should such agents will that they eventually betray God and are damned? Thomas Aquinas (, I–II, q. , a. , ad ) considers whether those who are to be damned are obligated to will their own damnation, and concludes that they are obligated only to will that God’s justice be carried out in general. It is unclear, however, whether in the case he is considering, the agent in question recognizes what God wills in this particular case.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
conjoined with divine determinism. The idea that God first determines humans to sin, then sits in judgment over them, self-righteously listing the sins they were determined to perform, and finally condemning them to torment – perhaps endless torment – is too much for many to bear. Sometimes philosophers refrain from constructing detailed arguments against a God who acts in such ways, instead pointing to the intuitive inhumanity of such activity. For example, Antony Flew writes: It is not only intellectually scandalous but also an outrage against justice and humanity to maintain that someone whom God arranges shall, freely and eagerly, act thus, may properly by Him be punished for so acting. (, )
And in another work he writes: Certainly it would be monstrous to suggest that anyone, however truly responsible in the eyes of men, could fairly be called to account and punished by the God who has rigged his every move. All the bitter words which have ever been written against the wickedness of the God of predestination – especially when he is also thought of as filling Hell with all but the elect – are amply justified. (, )
Rogers has the same view: It is God who creates the reprobate in their sins. Having made the vessel fit for wrath, it is difficult to see that the creator is just in blaming the vessel for being fit for wrath. It is being punished for doing as God made it to do. Is this justice? (, )
In what follows, I hope to turn away from such brief reactions responding to a particular picture of God, and toward the construction of arguments that conclude that the behavior of such a being is objectionable. Although we might worry both about divine blame and divine punishment falling on those who are determined to sin, the bulk of my comments will concern the former. After investigating the permissibility of divine blame, I will briefly turn to how these considerations apply to divine punishment. We might worry that, given divine determinism, God acts wrongly in blaming sinners for one (or more) of three reasons: . If God determines an agent x to perform evil act y, then x is not blameworthy for having performed y. . If God determines an agent x to perform blameworthy act y, then God loses the standing to blame anyone for performing acts of the same type as y. . If God determines an agent x to perform blameworthy act y, then God loses the standing to blame x for y.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Divine Blame
According to the first reason, given divine determinism, sinful agents are not responsible for their actions, and thus no one should blame them. Presumably, this objection stems from the idea that determinism undermines human freedom. This objection was discussed at length earlier, and will be set aside here; I will assume for the sake of the present considerations that these objections can be met. Many philosophers think that blame is appropriate only when the blamer meets particular conditions, and the next two conditions challenge the claim that a determining God has the moral standing to blame humans. Before moving forward to examine objections and , it is important to get a little clearer about just what they mean. First, as I will use the term, “blame” includes not only a judgment that a particular act was wrong, but also a negative reactive attitude such as indignation. Second, we might take blame to involve only a feeling of indignation (internal blame) or to also involve some expression of blame to the one who is blameworthy (external blame). Both sorts of blame will be relevant to our investigation, so we must further distinguish the two reasons we had before us: *: If God determines an agent x to perform blameworthy act y, then God loses the standing to internally blame anyone for performing acts of the same type as y. **: If God determines an agent x to perform blameworthy act y, then God loses the standing to externally blame anyone for performing acts of the same type as y. *: If God determines an agent x to perform blameworthy act y, then God loses the standing to internally blame x for y. **: If God determines an agent x to perform blameworthy act y, then God loses the standing to externally blame x for y.
With these distinctions before us, we might turn to specific arguments that make use of each objection.
The view that appropriate blaming requires a special standing to blame has become very common. For examples of work defending such a view, see Cohen (), Duff (), Fritz and Miller (), Sabini and Silver (), Scanlon (), Smith (), Wallace (), and Wertheimer (). This use is typical in discussions of the moral standing to blame. See, for example, the discussion by Todd (in press). I set aside as irrelevant to the present discussion cases where blame involves an internal component and an external expression to parties other than the one who is blameworthy. Some authors, however, write as if a minimum condition of blame is an outward expression of a moral judgment. See, for example, Bell (). Those who hold such a position might think of internal blame as being blame only in a deficient or analogous way.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
Let’s begin by considering * and **. Each seems to rest on the claim that it would be hypocritical for a determining God to blame anyone for anything, since whatever sins constitute a blot on a human’s moral record also appear in some way on God’s. The basic idea underlying this claim is that those who commit some act cannot blame others for committing similar acts. Earlier, we addressed the question of whether a God who determines an agent to murder thereby becomes a murderer. Divine determinists, I take it, are committed to answering this question in the negative, and I will not rehearse their defenses here. Nevertheless, such an answer is not itself sufficient to respond to the larger question of whether a determining God would be hypocritical to blame a murderer, given God’s own involvement in so many previous murders. After all, one may be hypocritical in blaming another for doing something that is merely similar to what one has done in the past. The objector may claim that there is surely a sense in which a determining God and a murderer are similar. After all, a murderer arranges events in such a way that people’s lives are cut short; if divine determinism is true, then this is just what God does, too. Thus, we might think that the murderer, upon hearing a divine rebuke, might reasonably reply: “And just who are you to blame me for this? You have been involved in every murder in history, and here I stand guilty of one and you presume to sit in judgment?” Various philosophers have suggested reasons for thinking there is such a hypocrisy condition on the standing to blame – that is, for thinking that hypocrites lack the standing to blame. Some think that hypocritical blamers pretend to care about some moral value that has been violated, but their own past actions suggest otherwise. R. Jay Wallace () contends that blame involves a commitment to self-scrutiny, and hypocrites lack such a commitment. Divine determinists may utilize either of these views on hypocrisy to support the claim that God retains the standing to blame. They might begin by endorsing either view of why hypocrisy sometimes undermines the standing to blame, and then note that it cannot support the idea that God is a hypocrite in a way that undermines this standing. God is committed to the various sorts of moral values that are ignored by those worthy of blame. Moreover, it is not clear what a commitment to self-scrutiny might amount to in a being who
See Todd (in press). Or so divine determinists will say. I addressed whether divine determinism leads to God’s blameworthiness in an earlier chapter.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Divine Blame
is necessarily good and omniscient, but it would stretch credulity to say that such a being lacks the standing to blame because of a lack of selfscrutiny. So determinists have a way to avoid *, but what of ** (if God determines an agent x to perform blameworthy act y, then God loses the standing to externally blame anyone for performing acts of the same type as y)? Here it seems that the divine determinist is on firm footing. If God is not guilty of a sort of hypocrisy that undermines the standing to internally blame an agent, then it is unclear why it would undermine the standing to externally blame. We might point to some criterion of the speech acts involved in external blaming and suggest that this criterion is not met in God’s case. But to make such an objection stick, we must both find such a criterion and provide some sort of defense for it. Perhaps the criterion might just be that those who have performed some act cannot blame others for performing similar acts. But what might explain this criterion? It seems unlikely that it could be a fundamental fact about such speech acts that they ought (or can) be performed only by those who have not acted similarly to those whom the blame is targeting. To evaluate whether this criterion rules out God’s moral standing – an unusual case if ever there was one – we would need to know more precisely which sorts of hypocrisy undermine the standing to blame, and it is unlikely that we can reach such precision without knowing what grounds the criterion in the first place. Of course, if one points to either of the two reasons discussed previously – that hypocritical actions betray a lack of commitment to the relevant value or that they are evidence of a lack of commitment to self-scrutiny – then the determinist may point out that whatever might be said of God, divine blame is not hypocritical in these ways. In the absence of a ground for the hypocrisy condition that suggests God lacks the standing to blame, divine determinists should not be overly concerned about **. Now let’s turn to *: If God determines an agent x to perform blameworthy act y, then God loses the standing to internally blame x for y. This principle is based upon a more general principle, sometimes called the non-involvement condition. According to this condition, one cannot blame another for an action if one is involved in particular ways in this action. Cohen, a defender of such a condition, explains that one is involved in a relevant way if one is “implicated in the commission of this
Cohen () suggests that standing to blame is related to the nature of blame as a speech act. See Bell (), Cohen (), and Todd (in press) for discussion. Notice that one also might endorse * merely because one endorses * (and likewise, might endorse ** because of **).
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
very act, as its co-responsible stimulus, commander, coercer, guard, assistant, or whatever” (, ). If we accept such a condition, then it is plausible that, by determining humans to act badly, God has lost the moral standing to blame them. The theist might try to avoid this claim by pointing to the important difference between the way that God is involved in human sin and the way that other human agents are often involved in the sins of their peers. If I coerce or command another to sin, then, presumably, I sin as well. God, theists maintain, is not in a similar position. However one defends the claim, theists (at least those who believe in the sort of God associated with Western monotheisms) are committed to God’s innocence. Such a claim will not avoid worries associated with the non-involvement condition. Cohen notes that one’s action might itself have been blameless, yet still result in the loss of standing to blame. In defense of such a principle, Matt King offers the following case: Suppose Charlie knows that Linus, who has a weakness for sweets, is trying to lose weight. Nevertheless, he takes Linus to a place for dinner that (he knows) is located next to an incredible ice cream shop. Quite predictably, after dinner Linus visits the shop next door and has some ice cream. Charlie may have acted quite permissibly, if a bit unkindly. But even if Linus is to blame for reneging on his commitment to diet (e.g., his family and friends might rightly hold him responsible on this score), Charlie may not be in a position to legitimately blame Linus for it. And this is because Charlie can be partly responsible for Linus’s failing even while Linus is wholly responsible for it. For those who may be apt to think Charlie is in a perfectly good position to blame Linus, consider what such an exchange would look like. Imagine Charlie scolding Linus for sheepishly ordering a scoop of mint chocolate chip. Linus’s indignation might rightly flare at this point. He might insist, “But you brought me here knowing just how difficult it would be for me not to get ice cream. You know I am trying to avoid sweets; you know how important losing weight is to me. And not only do you throw a huge temptation at me, you have the gall to blame me for succumbing to it!” It seems to me that something is problematic about Charlie’s blame here, and it plausibly has something to do with the fact that he knowingly placed Linus in a position to fail. Still, it need not be the case that Charlie has done anything wrong. He need not be ill-intentioned in proposing the dinner spot. He might reasonably believe that Linus will resist the temptation, or that it is important for Linus to meet these challenges head-on. Even so, it seems as though Charlie’s involvement reduces his ability to fairly blame his friend. (, –)
See also Matt King (), who agrees, and Lippert-Rasmussen (), who seems to do so as well.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Divine Blame
King concludes from this case that Charlie has lost his moral standing to blame, and that this implies it is possible to lose the standing to blame by performing actions that are not themselves blameworthy. If he is right, then the claim that God does nothing wrong in determining agents to sin will have little force in the present context. King even suggests a link between his case and divine determinism. According to this author, a determined agent’s response to God can be like that of Linus to Charlie. Both can say, “‘Gee, thanks a lot! I know I screwed up, but do not act like you had nothing to do with it’” (King , ). The point seems to be that blaming another either constitutes or presupposes a claim that one is not involved in the particular wrongdoing under discussion. How, then, might a divine determinist reply to this case and the accompanying non-involvement condition on the standing to blame? I think the determinist has at least two good options. First, the determinist might embrace skepticism about the idea of moral standing to blame. According to this view, a blaming agent need not possess any particular moral standing to appropriately blame someone. If this is so, then the case of Charlie and Linus should have little force. Charlie can fairly blame his friend. Likewise, God can fairly blame human sinners. In response to the reply offered by Linus and the determined agent, God and Charlie can reply, “You have this wrong. Nothing in my stance of blaming suggests that I wasn’t involved. Of course, I had something to do with it, but my involvement doesn’t render you blameless. In blaming, I merely take a stance against the evil of your action; I don’t suggest anything about myself.” The idea here is that blame is an appropriate response to blameworthiness, and its appropriateness does not depend upon the “larger normative network in which this action takes place.” The second avenue of response involves embracing an analysis of the case of Charlie and Linus offered by Patrick Todd (in press). Todd argues that if Charlie lacks the moral standing to blame Linus, this lack is not grounded in Charlie’s involvement with Linus’s action. Charlie’s involvement does not itself bring about the loss of the moral standing to blame,
See, for example, Bell (). This is not to say, however, that the target of blame need not meet particular conditions. Moreover, it is plausible that to appropriately blame someone, one must be justified in thinking that the target of blame is blameworthy. Such an epistemic condition is importantly different from the conditions for moral standing that are under discussion. For a defense of this last claim, see Todd (in press). I am indebted to an anonymous reader for this expression, which captures this aspect of this response nicely.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
but is instead evidence of something else that does so. Charlie’s actions betray a lack of regard for Linus and his goals, and it is this lack of regard or commitment to particular values that does all the work. Todd concludes, “when ‘involvement’ removes someone’s standing to blame, it does so only by indicating something further about that agent, viz., that he or she lacks commitment to the values that condemn the wrongdoer’s action” (in press). At this point, we might note that the story of Charlie and Linus is underdescribed. Importantly, we are not told why Charlie took Linus to this particular restaurant, nor whether Charlie took the proximity to the ice cream shop, together with Linus’s battle with sweets, to be a reason against the choice. If Charlie dismissed such thoughts as irrelevant or too easily defeasible – or worse, took it as a reason in favor of the choice, hoping that Linus would fail and he might humiliate his friend – then Charlie clearly lacks the relevant commitment that would give him the moral standing to blame (assuming any such commitment is required at all). If, in contrast, Charlie considers the proximity to the ice cream shop to be a serious reason against his choice, but finds that other reasons outweigh it – perhaps it is the only restaurant open or the only one that serves food both friends can afford – then his choice need not provide any evidence of his lack of commitment to particular values. If we fill out the story in the latter way and accept Todd’s analysis of the case, then we have no reason to think that Charlie blames his friend unfairly. Similarly, divine determinists should claim, independent of whatever they might think of the practice of blaming, that God’s actions do not reveal a lack of commitment to any moral values. Like Charlie in the expanded case, God is in a position to blame appropriately. The determinist, then, can avoid *. But what of **? This objection claims that if God determines an agent x to perform blameworthy act y, then God loses the standing to externally blame x for y. Here, I think, the determinist should build off of whichever response was used to counter *. If God’s involvement in human sin does not make it inappropriate for God to internally blame, then why should we worry about external blame? Someone skeptical about the moral standing to internally blame might nonetheless think it is dialectically inappropriate for those who lack a commitment to the moral values at issue to externally blame, but divine determinists should say that God is rightly committed to all moral values. Determinists might add that God is each agent’s creator and the governor of the universe and, therefore, is in a perfect position to externally blame.
We are told some of the reasons he might have had, but his motive is never pinned down.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Divine Blame
King offers a reason independent of those discussed previously for thinking that God is not in a position to blame humans: There is nothing God could blame them for. King discusses this matter by pointing to a divinely determined human named Ernie. He writes, “Ernie’s action is also necessitated by the greater good, and so God cannot be angry with him, cannot think he acts wrongly (in the big picture) or that he should have done differently. Indeed, this is doubly true for God, because God cannot even be angry that circumstances necessitated the action, for God constructed those very circumstances” (King , ). King seems to presuppose that one can legitimately blame agents for performing x only if one consequently wills that they had not performed x. But King doesn’t defend this supposition, and divine determinists should deny it. Moreover, there seem to be good reasons for everyone to deny it. To see this, consider the following case: Erich is a Nazi commander working in a death camp. Erich has no love for the Nazi party in general or the genocide in which this camp participates in particular, but he feigns such love to be better positioned to help the Jewish prisoners. One night, a guard reports seeing something beyond the fence – perhaps an escaped prisoner, but he cannot be sure. Ordinarily, Erich would tell the guard that it couldn’t possibly be a prisoner and must have been a deer or perhaps a trick of the light. On this occasion, however, Erich knows it is a prisoner, and he knows the prisoner meant the guard to see him. The prisoner is creating a diversion so that other prisoners can take advantage of the confusion to escape from another part of the camp. For this reason, Erich passes on the report to Karl, who is a true believer in the Nazi cause. Erich knows that Karl loathes the prisoners in general and escapees most of all, and will spare no effort in following up this lead. Karl pulls guards from all over the camp, selecting the most capable, most devoted, and most vicious. Erich is glad that the prisoner’s plan is working this well, even if saddened for the heroic prisoner who sacrificed freedom and indeed life for others. Nonetheless, he internally condemns Karl, unable to understand how anyone can act in this way.
In this case, Erich consequently wills that Karl act in this heinous way. Doing so allows for the greater good of the escape of the many. Does this fact undermine Erich’s standing to internally blame Karl? In other words, does Erich do something wrong in internally blaming Karl? I must confess that I cannot see why we should think so. We might think (in keeping with a suggestion from King) that if Karl’s action promotes the greater
This case is modeled after that of Jonas, discussed by Todd in two works (, in press). Unlike Jonas, however, Erich consequently wills that the target of blame acts in a blameworthy way.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
good, then he cannot be blamed for it, but it seems that his ignorance about the greater good to be achieved eliminates any possibility of his seeking an excuse in this greater good. Todd () agrees that, given the moral responsibility of determined agents, a determining God can appropriately blame humans. But, he suggests, this is not a win for divine determinists. Instead, this should lead us to the conclusion that incompatibilism is true. Why? Todd thinks that if compatibilism is accepted, then there is no good reason to think that a determining God could not blame those who are determined to do evil. We might consider his argument as a reductio: Begin by assuming compatibilism, then note that given compatibilism, we should believe that a determining God can blame determined agents. Since this is absurd, we should reject the original assumption (i.e., compatibilism) and embrace incompatibilism. Incompatibilism, he says, offers the best explanation for why a determining God cannot blame humans, and it is “radically,” “severely,” and “massively” counterintuitive to think that God could blame humans if divine determinism is true (Todd , , , ). In fact, it is “considerably mysterious” how a determining God could rightly blame humans (). Although he firmly believes that such a God would lack the standing to blame, Todd writes: “I do not know how to argue for this claim. I simply say that it is eminently plausible, and that we would need some very good reason to deny it” (). Notice that this argument does not collapse into any of the manipulation arguments discussed earlier, since it turns not on an intuitive judgment about whether a determined agent of one kind or another is blameworthy, but rather on whether particular agents would be in a position to blame the determined agent, whether or not this agent is blameworthy. Of course, the conclusion of the argument reaches a verdict about the determined agent’s moral responsibility, but no part of the argument depends on an intuition about such responsibility. So far, I have been considering how to respond to objections that conclude that a determining God lacks the standing to blame. I now wish to briefly consider a way to build up to a positive claim that such a God does not lack such a standing. Earlier, I assumed that blame is more than a judgment of blameworthiness – that it involves a reactive attitude such as indignation or perhaps resentment. But maybe blame understood in this
Todd doesn’t think that the truth of divine determinism provides grounds for incompatibilism, but instead that consideration of divine determinism as a hypothetical case provides support for it. If compatibilists deny divine determinism, or even theism more generally, they would still, argues Todd, be committed to saying that in such a scenario God would be in a position to blame. This, Todd suggests, represents a cost for compatibilists.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
The Problem of Divine Blame
way is not what should be attributed to God. In fact, insofar as blame is taken to involve an emotional element, many theists, whether determinists or not, should deny that God blames anyone for anything, since they deny that God experiences emotions. Instead, they should say that God’s attitude toward sinners is somehow analogous to the attitude we have when we blame the wicked. Which features are analogous? Well, the judgment that someone has done something wrong, together (perhaps) with a commitment to the moral value the agent in question disregarded. But if this is all that is required for divine blame (or divine blame*, for those who think that this should not be regarded as blame, properly speaking), then there should be little reason for us to worry about it. In addition, this construction might help remove worries from objections. My suggestion is that once principles *–** are rejected, any remaining intuition against God possessing the standing to blame may be dissolved by focusing on the differences between divine and human blame – most notably, that divine blame is an emotionless affair. By removing emotion from our account, it becomes easier to see that even if we have concerns about certain humans lacking the standing to blame, this does not provide evidence that a determining God lacks the standing to blame. Thus, even if we might wonder whether it would be appropriate for Charlie to have certain sorts of judgmental emotions toward Linus, this should not make us worry about divine determinism because a determining God lacks all such emotions toward determined agents. Before closing this chapter, I wish to turn briefly to divine punishment. Although it is controversial whether a justified belief that another is blameworthy is sufficient to justify blaming another, it is clear that it is not sufficient to justify punishment. Perhaps some agents may be blameworthy without being worthy of any punishment whatsoever. More obviously, some agents may be worthy of punishment but many other agents might lack the moral standing to punish them. I will suppose that humans may be worthy of punishment even if divine determinism is true, and take the present objection to be that given divine determinism, God lacks the moral standing to punish them. In my experience discussing this matter with others, the intuition that God lacks the standing to punish seems to be even stronger than the perception that God lacks the standing to blame. Despite this, I see no reason to think that the arguments in support of such a claim will be any stronger. Of course, one might think that God lacks the standing to punish
Perhaps such views get even human blame wrong. For discussion, see Sher ().
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
even if divine determinism is false, perhaps because only those appointed to positions of authority (rather than those who assume such positions without consent) possess such standing. If, however, we are looking for problems specific to divine determinism, then such considerations must be set aside. Why think that in determining agents, God loses the standing to punish even if determined agents remain worthy of it? It seems we must once again return to the twin concerns of hypocrisy and involvement. I suggest the divine determinists respond to these worries in the same way that they responded to the worries that hypocrisy and involvement undermined God’s standing to blame: by pointing to ways that God’s actions neither constitute nor provide evidence for the sorts of hypocrisy and involvement that might plausibly be thought to undermine standing. Most crucially, God’s actions in determining agents do not reveal a lack of commitment to the relevant moral values, nor does God’s action of blaming or punishing deceptively suggest that God was in no way involved in the relevant human action. This response can help push back specific arguments that God lacks the standing to punish, but many will be left with the intuition that God does lack such standing. I think that this is best viewed as an incompatibilist intuition, since the best defense of its propositional content is that if determinism is true, then humans are not worthy of punishment because they are not responsible for their actions. I think Todd is correct in asserting that the problem of a determining God’s standing to blame and punish is not a special problem for divine determinists, but rather a general problem for compatibilists. Supposing that divine determinists do remain firm compatibilists, what should they make of their intuition (supposing they have one) that a determining God lacks the standing to blame and punish? I think they should treat it like other intuitions springing from rejected claims. We might have the intuition that flying is to be feared more than is driving, but we know this intuition is based on some sort of judgment that flying is more dangerous than driving. What should we make of the intuition if it persists even when we become convinced of the relative safety of flying? Presumably we should work to eliminate it, if we think this would be both possible and practical, and we certainly shouldn’t give it any evidentiary weight. But surely our evidence of the safety of flying is much more easily and confidently assessed than our evidence for compatibilism, whatever ones takes this evidence to be, and so we might still wish to allow incompatibilist intuitions to retain some force. Nevertheless, if one finds compatibilist arguments convincing, the evidentiary force of the intuition that a determining God lacks the
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Conclusion
standing to blame and punish should be radically diminished. Furthermore, I suspect that for many, the strength of the intuition may diminish the less one becomes convinced of incompatibilism or the more one becomes convinced of compatibilism. Given the diminished evidentiary weight of this intuition, together with its weakened intuitive force for many committed compatibilists, it cannot single-handedly support a decisive objection to divine determinism. Nevertheless, to the extent that the intuition should be granted some evidentiary weight and to the extent that the intuition remains even for committed compatibilists, divine determinism incurs yet another cost. We might still wonder why the intuition that a determining God lacks the standing to punish is stronger than that concerning the standing to blame. My suspicion is that this is an instance of a more general phenomenon: Our intuitions tend more toward excusing the agent from blame and punishment (or blame and punishment from particular agents) as the seriousness of the matter in question increases (the matter in question including the evil being attributed to the agent and the harm imposed upon the guilty agent). Since punishment, and especially the severe punishment often associated with divine punishment, is far more serious than blame, we tend toward excusing agents more readily.
Conclusion In this chapter, I have investigated four distinct problems that divine determinists face – at least those, who, like many theists, think that God issues commands and reacts with blame and punishment to at least some agents who break them. First, it seems that there is something deceptive in giving commands and then determining agents to break them. The former behavior implies that the content of the command expresses one’s will, whereas the latter behavior makes clear that it does not. Second, it seems that knowingly doing the will of God is sufficient for avoiding
Five years ago, I was a firmly committed incompatibilist. Over time, my commitment has waned; I have not embraced compatibilism, but have tended toward what Alfred Mele called agnostic autonomism. In essence, I take it that humans are morally responsible, but am agnostic on the question of whether determinism is compatible with moral responsibility. Mele has developed this view for decades over many works. For an early statement of the view, see Mele (). As my commitment to incompatibilism has slipped, so has my intuition that a determining God lacks the standing to blame and punish. Although experiences will surely vary, I suspect that the correlation between the strength of my commitment to incompatibilism and the strength of my intuition that a determining God lacks the standing to blame and punish is not an anomaly. I discuss this as an error theory for certain intuitions in Section of Chapter .
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Divine Commands, the Divine Will, and Divine Blame
wrongdoing, but divine determinists cannot say this. Third, it seems that good agents should wish that God’s will is always fulfilled and also will that they had not sinned in the past, but divine determinists cannot say this, either. Finally, it seems that a determining God lacks the standing to blame and punish humans who had been determined to sin, but many divine determinists wish to deny this. I have attempted to explore the logic of each objection and to develop options by which the divine determinist might reply. I have argued that the second problem can be successfully avoided with no cost. Each of the other three objections reveals something further about divine determinism that might be taken to constitute a cost: In each case, determinism (together with other positions to which many theists are committed) either leads to unintuitive results or constrains the possible positions that one might hold.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.009
Conclusion
Over the course of this book, I have investigated the challenges with which divine determinists must struggle. In the past, enemies of divine determinism often rested content with simply gesturing toward these issues, but I have sought to push beyond such generalities, first by presenting the arguments in a careful and thorough way and second by exploring possible responses to each. Judging by my experience, readers will respond to my investigations in very different ways. Some, of course, will respond by complaining that I have ignored the most powerful formulations of the anti-determinist arguments I have discussed, and still others will maintain that I have overlooked the best replies available. In response to such readers, I can only offer my assurance that I have not intentionally shrunk away from the most challenging objections and have posed the objections to divine determinism in the way that seems most powerful to me. Similarly, I have attempted to defend the prospects of such determinism in the face of these challenges. But, as is always the case with philosophical issues, some readers will disagree with me on which existing objections and replies are most powerful, and new objections and replies – ones that neither I nor perhaps anyone has yet considered – are sure to arise. But in this final section, I wish to highlight another sort of disagreement readers might have with each other. Some readers will find – indeed, already have found – at least some of my discussions of problems and possible replies satisfactory. They will find that my investigation of one problem or another gets to the heart of the concern, and that my exploration of possible replies reveals the best that divine determinists may provide in response. And they will find the result damning. The problems, they will think, are formidable, and the possible replies, if not downright ridiculous, at least carry with them a hint of desperation.
Parts of Chapter were originally published in Furlong (a). A positive referee for that article took my treatment of these issues to reveal a bleak landscape for the divine determinist: The options
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.010
Conclusion
Others will begin with at least partial satisfaction with my investigation, but reach the opposite conclusion: They will decide that the challenges, even if formidable, may be successfully met. They will judge each supposed problem to eventually dissolve upon careful consideration or at least pose no danger to those willing to adopt some perfectly innocent claims about human agents, God, and the relations between them. Divine determinism, they will conclude, is innocent as well. I suspect that some who hold the former view cannot help but think of those with the latter as self-deluded, or, at any rate, as forced into an uncomfortable position by nonphilosophical constraints. Patrick Todd seems to suggest as much: If it is really so massively counterintuitive to suppose that God can blame those he determines, exactly what explains the willingness of so many to endorse this picture? As I noted, there are many who presently accept (or claim to accept) such a theological picture. However, I am not inclined to regard these facts as evidence of the intrinsic reasonableness or defensibility of the view in question, but rather as testimony to the sometimes incredible power religious commitments can have in leading people to accept what would otherwise seem to be overwhelmingly implausible. (, –)
Although he makes these comments within the context of discussing whether a determining God has the standing to blame, I think many would be happy to apply the same sentiment to many of the other problems discussed in this book. I think we should be careful, however, in jumping to easy answers about why one philosopher is willing to accept the costs of a view, or indeed, is blind to the purported costs in the first place. We see this general phenomenon in many nontheistic contexts. We see it, for example, in the battle between compatibilist and incompatibilist intuitions. In this arena, we have some who find incompatibilism simply obvious – and yet others have told me that they lack the intuitions that support incompatibilism entirely. Some find desperation in the willingness of others to accept a hard-line reply to manipulation arguments – that is, the notion that an agent might be thoroughly manipulated and yet still
I canvassed, the referee thought, did represent the possible replies, but surely no theist would accept any of them. Of course, some theists do accept at least some of these responses, but in my experience the reaction of this referee is not unusual. Some might argue that this merely appears to be a nontheistic context by pointing to research that shows theists are overwhelmingly more likely to be libertarians than nontheists are (Bourget and Chalmers ). Note, however, that strong incompatibilist intuitions are not limited to libertarians. Hard indeterminists and many compatibilists share such intuitions, too.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.010
Conclusion
morally responsible. Others don’t see such a reply as even adding to the cost of compatibilism. This is obviously not the place to rehearse arguments about the value of intuitions. Instead, I merely mean to suggest that we should not assume that particular intuitions, or the lack thereof, should be immediately put down to religiously induced madness. First, even if it is madness, there is no reason to think that the general phenomenon of lacking obvious intuitions when it is convenient to one’s position is peculiar to religious matters. Second, it is often difficult to tell if it is madness or merely a radical disagreement so fundamental that both sides must resort to saying things like “I don’t know how to argue for this; it simply seems obviously correct.” I have nothing new to add about how to deal with such fundamental disagreement, just advice to be charitable in interpreting it. In part, my own commitment to this advice is born of experience. I once thought that incompatibilism was not only true, but obviously so, and the only legitimate question was whether one should embrace libertarianism (as I wholeheartedly did) or incompatibilism of a harder variety (which I rejected but respected). Moreover, I thought that if ordinary determinists cannot account for human freedom, then neither could divine determinists. I thought that if divine determinism was true, then human life was a sham. And the biggest sham of them all was the claim that this determinism was perfectly compatible with the beliefs and practices of many theists. But I no longer think compatibilism a cop-out or divine determinism a delusion. And yet I have become neither a compatibilist nor a divine determinist. I am convinced we are morally responsible, but have become agnostic on
I do not mean to suggest that Todd does make such general assumptions, merely that the sentiment expressed in his earlier quote might be taken too far. Alfred Mele discusses what he calls ‘‘intuition deficit disorder”: Some readers may not be familiar with the malady known as intuition deficit disorder (IDD). In one of its manifestations, the person who suffers from it is blind to counterexamples to his theories about things, except when the counterexamples are widely recognized to be real-world events, states, or beings . . . Sometimes, after years spent considering a wide range of cases and refining their provisional analyses in light of their intuitions, they wind up not only with an analysis they like but also with IDD. If someone has this malady, the fact that he or she lacks certain intuitions does not count for much. (, ) Mele admits that he may have IDD about some issues in moral responsibility, and I might as well. Mele might be correct in thinking that the lack of some intuition in those with IDD doesn’t count for much, but I think it is also important to be careful not to judge that all those lacking our intuitions suffer from IDD.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.010
Conclusion
the issue of the compatibility of determinism and free will. I am a theist, but I am now agnostic on the question of divine determinism. In this book, I have attempted to show the available responses with which divine determinists might answer various objections, without, in most cases, lending support for one response over another. I have endeavored to examine the logical space in which divine determinists might stake out their positions, rather than constructing, proposing, and defending a particular view. Nevertheless, I owe it to readers who have stuck with me this far to make my own views clear. In what remains, I wish to lay bare what I find to be the most promising sort of determinist position, together with those objections that I currently find the most worrisome. In several places in this work, I had recourse to a distinction between two sorts of divine determinism: primary divine determinism and Edwardsian divine determinism. I am attracted to primary divine determinism, but largely because of reasons independent of any benefits it provides in responding to objections to determinism. Instead, I think the image of creation that Heath White () offers, that of a timeless being creating a four-dimensional block, gets things importantly correct. If I were eventually to endorse divine determinism, it would be of the primary variety. In the early chapters of this book, I argued that divine determinists are no better positioned to respond to the Consequence Argument than other compatibilists are, but primary divine determinists are (at least in some respects) better positioned than others to respond to manipulation arguments. Although primary divine determinists do have additional resources in responding to manipulation arguments, the problems they face with the Consequence Argument suggests to me that the compatibilism of divine determinists and that of the garden variety might stand and fall together. Determinists, both divine and naturalistic, would do well to follow Harry Frankfurt in denying that alternative possibilities are required for moral responsibility. They should also seriously consider following Michael McKenna in offering a hard-line response to manipulation arguments, although I think the primary determinist soft-line reply to these arguments shows promise as well.
And does so even if divine determinism is false, although the image must be complicated in several ways if that is so. I do not mean that they should endorse the reasons these authors offer for their positions – on that I take no stand – but that they should accept these argumentative moves.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.010
Conclusion
I think that divine determinists should admit that there is a sense in which God causes the acts of sin, while noting that God neither performs sinful actions nor causes evil. In defense of the claim that God does not cause evil, I suggest an appeal to the classical claim that evil is a privation and, therefore, not the sort of thing that could be caused in a strict sense. In defense of the moral innocence of God, I offered a unified, noncommittal strategy – but to lay my cards on the table, I think the strongest element of such a strategy is the appeal to God’s authority over humans. I think that divine determinists should simply give up on the idea that the Free Will Defense will be of much use to them. They can still utilize the defense in a limited way, while admitting that one of the propositions it contains is false. Of course, if they are willing to have recourse to propositions that they think are false, then an endless array of defenses might be trotted out. Some versions of divine determinism are compatible with a strong use of the Free Will Defense, but they rely on historical compatibilist conditions, which I find implausible. Although the Free Will Defense has been useful in the past, I suggest divine determinists look elsewhere, in particular to skeptical theism, for resources to address the problem of evil. This brings us to the problem of divine love. For my part, I find this to be the greatest difficulty for divine determinism. I think the best response is to invoke a sort of skeptical “greater good” defense. The evil actions that agents are determined to perform serve some purpose in bringing about a greater good, whatever it might be. Universalism, if the determinist in question is open to it, makes this pill a little less bitter to swallow, but I suggest this view should be adopted by non-universalists as well. The problem of human love I find less concerning: Here, I think, the determinist has a wealth of good replies available. This brings us to the four problems discussed in Chapter . The two problems that worry me the most here are the problem of divine deception and the problem of repentant sinners. In regard to the former, determinists have little option but to say that we shouldn’t assume that divine commands reveal the divine consequent will. Insofar as this is surprising – and I think it is – divine determinists must bite a bullet. This bullet might seem an especially hard one, since it would make deliberation rather difficult for those seeking to fulfill the divine will, but determinists must also sever the goal of human life from doing God’s consequent will. This comes out most forcefully in what I called the problem of contrition. This problem has been unjustly ignored, but should be a matter of great concern for determinists. We might solve this problem by saying the
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.010
Conclusion
contrite should wish God’s will hadn’t been fulfilled or that God hadn’t had this will in the first place, but both suggest that our goals in life should stand apart from God’s will. In the past, many theists have thought that we should model our lives after the divine plan, and we act wrongly in straying away from it. Determinists seem forced to suggest a funhouse mirror image of this thought: We always do God’s (consequent) will (even when we sin), and in cases where we sin, we should have strayed from God’s (consequent) will. What should we make of the costs – both large and small – that determinists must pay? Some of them might be decisive: They might be such that divine determinism should be denied no matter what reasons might be put forward in its favor. Some might think that some of the costs examined in the previous chapters are decisive in this way. If they are not, then they must be weighed against the benefits of adopting divine determinism, benefits stemming from general philosophical considerations and matters of religious commitment. I highlighted a number of such considerations in Chapter , and I encourage readers to consider the many serious reasons divine determinists have put forward in favor of this position, reasons I largely set aside in the present work. Finally, it is worth noting that some of the costs of divine determinism might turn out to be nearly unavoidable; it may turn out that our universe is governed by deterministic laws after all. Since physical determinism is very much a live possibility given what we know of the world, I think that theists would be wise to consider matters carefully before dismissing the costs of divine determinism as obviously decisive.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.010
References
Adams, Marilyn McCord. . “A Modest Proposal? Caveat Emptor! Moral Theory and Problems of Evil.” In Ethics and the Problem of Evil, edited by James P. Sterba, –. Bloomington, IN: Indiana University Press. Adams, Robert Merrihew. . “An Anti-Molinist Argument.” Philosophical Perspectives : –. Alexander, David E. . “Orthodoxy, Theological Determinism, and the Problem of Evil.” In Calvinism and the Problem of Evil, edited by Daniel M. Johnson and David E. Alexander, –. Eugene, OR: Pickwick Publications. Alexander, David E., and Daniel Johnson M. . “Introduction.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Alm, David. . “Responsibility, Manipulation, and Resentment.” Social Theory and Practice (): –. Almeida, Michael J. . “Bringing about Perfect Worlds.” In Free Will and Theism: Connections, Contingencies, and Concerns, edited by Kevin Timpe and Daniel Speak, –. Oxford: Oxford University Press. Andow, James. . “Thin, Fine and with Sensitivity: A Metamethodology of Intuitions.” Review of Philosophy and Psychology : –. Arberry, A. J. . The Koran Interpreted: A Translation. New York: Simon and Schuster. Armstrong, D. M. . A Theory of Universals: Volume : Universals and Scientific Realism. Cambridge: Cambridge University Press. Arpaly, Nomy. . Unprincipled Virtue: An Inquiry into Moral Agency. Oxford: Oxford University Press. Attar, Mariam. . Islamic Ethics: Divine Command Theory in Arabo-Islamic Thought. New York: Routledge. Augustine, H. . Confessions, translated by Henry Chadwick. Oxford: Oxford University Press. Baker, Lynne Rudder. . “Why Christians Should Not Be Libertarians: An Augustinian Challenge.” Faith and Philosophy (): –. Barnes, Eric Christian. . “Freedom, Creativity, and Manipulation.” Noûs (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Basinger, David. . “Divine Omniscience and Human Freedom.” Faith and Philosophy (): –. Bell, Macalester. . “The Standing to Blame: A Critique.” In Blame: Its Nature and Norms, edited by D. Justin Coates and Neal A. Tognazzini, –. Oxford: Oxford University Press. Berofsky, Bernard. . “Global Control and Freedom.” Philosophical Studies (): –. Bigelow, John. . “Presentism and Properties.” Philosophical Perspectives (Metaphysics): –. Björnsson, Gunnar, and Derk Pereboom. . “Traditional and Experimental Approaches to Free Will and Moral Responsibility.” In Companion to Experimental Philosophy, edited by Justin Sytsma and Wesley Buckwalter, –. Oxford: Wiley-Blackwell. Bonino, Serge-Thomas. . “Contemporary Thomism through the Prism of the Theology of Predestination.” In Thomism and Predestination: Principles and Disputations, edited by Barry David, Steven A. Long, Roger W. Nutt, and Thomas Joseph White, translated by Stefan Jetchick, Barry David, and Steven A. Long, –. Ave Maria, FL: Sapientia Press of Ave Maria University. Bourget, David, and David J. Chalmers. . “What Do Philosophers Believe?” Philosophical Studies (): –. Burrell, David B. . “Divine Action and Human Freedom in the Context of Creation.” In The God Who Acts: Philosophical and Theological Explorations, –. University Park, PA: Pennsylvania State University Press. Byerly, T. Ryan. . “Free Will Theodicies for Theological Determinists.” Sophia (): –. Byrne, P. . “Helm’s God and the Authorship of Sin.” In Reason, Faith and History: Philosophical Essays for Paul Helm, edited by M. Stone, –. Burlington, VT: Ashgate Publishing. Calvin, John. . Calvin: Institutes of the Christian Religion, edited by John T. McNeill, translated by Ford Lewis Battles. Vol. . Philadelphia: Westminster Press. Campbell, Keith. . Abstract Particulars. Cambridge, MA: Basil Blackwell. Cohen, G. A. . “Casting the First Stone: Who Can, and Who Can’t, Condemn the Terrorists?” Royal Institute of Philosophy Supplement : –. Collins, Brian. . “Adding Substance to the Debate: Descartes on Freedom of the Will.” Essays in Philosophy (): –. Couenhoven, Jesse. . “The Indicative in the Imperative: On Augustinian Oughts and Cans.” In Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology, edited by Hugh J. McCann, –. Oxford: Oxford University Press. Cowan, Steven B. . “Compatibilism and the Sinlessness of the Redeemed in Heaven.” Faith and Philosophy (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Crabtree, J. A. . The Most Real Being: A Biblical and Philosophical Defense of Divine Determinism. Eugene, OR: Gutenberg College Press. Cunning, David. . “Descartes on God and the Products of His Will.” In Models of God and Alternative Ultimate Realities, edited by Jeanine Diller and Asa Kasher, –. New York: Springer. Davies, Brian. . The Reality of God and the Problem of Evil. London/ New York: Continuum. . Thomas Aquinas on God and Evil. Oxford: Oxford University Press. Descartes, René. . The Philosophical Writings of Descartes, Vol. : Correspondence, translated by John G. Cottingham, Robert Stoothof, Dugald Murdoch, and Anthony Kenny. Cambridge: Cambridge University Press. Dougherty, T. . The Problem of Animal Pain: A Theodicy for All Creatures Great and Small. New York: Palgrave Macmillan. Dougherty, Trent, and Justin P. McBrayer. . Skeptical Theism: New Essays. Oxford: Oxford University Press. Duff, R. A. . “Blame, Moral Standing and the Legitimacy of the Criminal Trial.” Ratio (): –. Edwards, Jonathan. . The Works of Jonathan Edwards: Freedom of the Will. New Haven, CT: Yale University Press. Epictetus. . Discourses and Selected Writings, translated by Robert Dobbin. London: Penguin Books. Fara, Michael. . “Masked Abilities and Compatibilism.” Mind (): –. Feinberg, John S. . No One Like Him. Wheaton, IL: Crossway Books. Finch, Alicia, and Ted A. Warfield. . “The Mind Argument and Libertarianism.” Mind (): –. Fischer, John Martin. . The Metaphysics of Free Will: An Essay on Control. Oxford: Blackwell. . “Semicompatibilism and Its Rivals.” Journal of Ethics (): –. a. “How Do Manipulation Arguments Work?” Journal of Ethics (–): –. b. “Libertarianism and the Problem of Flip-Flopping.” In Free Will and Theism: Connections, Contingencies, and Concerns, edited by Kevin Timpe and Daniel Speak, –. Oxford: Oxford University Press. Fischer, John Martin, and Mark Ravizza. . Responsibility and Control: A Theory of Moral Responsibility. Cambridge: Cambridge University Press. Fisher, Mark. . Personal Love. London: Duckworth. Fisk, Milton. . “Relatedness without Relations.” Noûs (): –. Flew, Antony. . “Divine Omnipotence and Human Freedom.” In New Essays in Philosophical Theology, edited by Antony Flew and Alasdair MacIntyre., –. New York: Macmillan. . “Splitting Hairs before Starting Hares.” The Personalist : –. Flint, Thomas P. . “A New Anti-Anti-Molinist Argument.” Religious Studies (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Frankfurt, Harry G. . “Alternate Possibilities and Moral Responsibility.” Journal of Philosophy (): –. . “Three Concepts of Free Action: II.” In Moral Responsibility, edited by John Martin Fischer, –. Ithaca, NY: Cornell University Press. Freddoso, Alfred. . “‘Introduction’ to Luis de Molina.” In On Divine Foreknowledge, edited by Alfred J. Freddoso, –. Ithaca, NY: Cornell University Press. Fritz, Kyle G., and Daniel Miller. . “Hypocrisy and the Standing to Blame.” Pacific Philosophical Quarterly (): –. Furlong, Peter. . “Is God the Cause of Sin?” Faith and Philosophy (): –. a. “Blameworthiness, Love, and Strong Divine Sovereignty.” Sophia (): –. b. “Libertarianism, the Rollback Argument, and the Objective Probability of Free Choices.” Pacific Philosophical Quarterly (): –. Garrigou-Lagrange, Réginald. . Predestination. London: Herder. . God, His Existence and His Nature, translated by Dom Bede Rose. Vol. . London: Herder. Ginet, Carl. . On Action. Cambridge: Cambridge University Press. Grant, W. Matthews. . “Aquinas on How God Causes the Act of Sin without Causing Sin Itself.” The Thomist (): –. a. “Divine Universal Causality and Libertarian Freedom.” In Free Will and Theism: Connections, Contingencies, and Concerns, edited by Kevin Timpe and Daniel Speak, –. Oxford: Oxford University Press. b. “The Privation Solution.” Faith and Philosophy (): –. Green, Christopher, R. . “A Compatibicalvinist Demonstrative-Goods Defense.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Gutas, Dimitri. . Avicenna and the Aristotelian Tradition: Introduction to Reading Avicenna’s Philosophical Works. Second, Revised and Enlarged Edition, Including an Inventory of Avicenna’s Authentic Works. Leiden: Brill. Haji, Ishtiyaque. . Moral Appraisability: Puzzles, Proposals, and Perplexities. Oxford: Oxford University Press. Haji, Ishtiyaque, and Stefaan E. Cuypers. . “Hard- and Soft-Line Responses to Pereboom’s Four-Case Manipulation Argument.” Acta Analytica (): –. Hart, Matthew J. . “Calvinism and the Problem of Hell.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Hasker, William. . God, Time and Knowledge. Ithaca, NY: Cornell University Press. . “God the Creator of Good and Evil?” In The God Who Acts, edited by Thomas Tracy, –. University Park, PA: Pennsylvania State University Press. . “Middle Knowledge: A Refutation Revisited.” Faith and Philosophy (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
. “A New Anti-Molinist Argument.” Religious Studies (): –. . “‘The (Non)-Existence of Molinist Counterfactuals.’” In Molinism: The Contemporary Debate, edited by Ken Perszyk, –. Oxford: Oxford University Press. Hause, Jeffrey. . “Thomas Aquinas and the Voluntarists.” Medieval Philosophy and Theology (): –. Heil, John. . The Universe as We Find It. Oxford: Oxford University Press. Helm, Paul. . The Providence of God. Downers Grove, IL: Intervarsity Press. . “God, Compatibilism, and the Authorship of Sin.” Religious Studies (): –. . “Discrimination: Aspects of God’s Causal Activity.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Hick, John. . Evil and the God of Love. London: Macmillan. Hoffmann, Tobias. . “Aquinas and Intellectual Determinism: The Test Case of Angelic Sin.” Archiv Fu¨r Geschichte Der Philosophie (): –. . “The Pleasure of Life and the Desire for Non-existence: Some Medieval Theories.” Res Philosophica (): –. Hoffmann, Tobias, and Peter Furlong. . “Free Choice.” In Aquinas’s Disputed Questions on Evil: A Critical Guide, edited by M. V. Dougherty, –. Cambridge: Cambridge University Press. Hoffmann, Tobias, and Cyrille Michon. . “Aquinas on Free Will and Intellectual Determinism.” Philosophers’ Imprint (): –. Holland, R. F. . Against Empiricism: On Education, Epistemology and Value. Oxford: Blackwell. Howsepian, A. A. . “A Libertarian-Friendly Theory of Compatibilist Free Action.” Southern Journal of Philosophy (): –. . “Compatibilism, Evil, and the Free-Will Defense.” Sophia (): –. Huemer, Michael. . “Van Inwagen’s Consequence Argument.” Philosophical Review (): –. Hunt, David P. . “Providence, Foreknowledge, and Explanatory Loops: A Reply to Robinson.” Religious Studies (): –. John of Damascus. . “Exposition of the Orthodox Faith.” In A Select Library of Nicene and Post-Nicene Fathers of the Christian Church, translated by S. D. F. Salmond. Vol. ., –. Edinburgh: T and T Clark. Johnson, Daniel M. . “Calvinism and the Problem of Evil: A Map of the Territory.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Jordan, Jeff. . “The Topography of Divine Love.” Faith and Philosophy (): –. . “The Topography of Divine Love: A Reply to Thomas Talbott.” Faith and Philosophy (): –. . “The ‘Loving Parent’ Analogy.’” International Journal for Philosophy of Religion (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Kane, Robert H. . The Significance of Free Will. Oxford: Oxford University Press. Kant, Immanuel. . The Metaphysical Elements of Justice: Part I of the Metaphysics of Morals, translated by J. Ladd. Indianapolis: Hackett Publishing. Kapitan, Tomis. . “Deliberation and the Presumption of Open Alternatives.” Philosophical Quarterly (): –. King, Matt. . “Manipulation Arguments and the Standing to Blame.” Journal of Ethics and Social Philosophy (): –. Lange, Marc. . Because without Cause: Non-Causal Explanations in Science and Mathematics. Oxford: Oxford University Press. Levy, Neil. . “Determinist Deliberations.” Dialectica (): –. Lewis, Peter J. . Quantum Ontology: A Guide to the Metaphysics of Quantum Mechanics. Oxford: Oxford University Press. Liao, S. Matthew. . “A Defense of Intuitions.” Philosophical Studies (): –. Liguori, Alfonsus. . The True Spouse of Jesus Christ, translated by A Catholic Clergyman. Dublin: John Coyne. Lippert-Rasmussen, Kasper. . “Who Can I Blame?” In Autonomy and the Self, edited by Michael Ku¨hler and Nadja Jelinek, –. Dordrecht: Springer. Long, Steven A., Roger W. Nutt, and Thomas Joseph White. . Thomism and Predestination: Principles and Disputations. Ave Maria, FL: Catholic University of America Press. Lowe, E. J. . “A Neo-Aristotelian Substance Ontology: Neither Relational nor Constituent.” In Contemporary Aristotelian Metaphysics, edited by Tuomas E. Tahko, –. Cambridge: Cambridge University Press. MacDonald, Scott. . “Aquinas’s Libertarian Account of Free Choice.” Revue International de Philosophie (): –. Mackie, J. L. . “Evil and Omnipotence.” Mind (): –. Maritain, Jacques. . God and the Permission of Evil, translated by Joseph W. Evans. Milwaukee: Bruce Publishing. Matava, Robert Joseph. . Divine Causality and Human Free Choice: Domingo Báñez, Physical Premotion and the Controversy de Auxiliis Revisited. Leiden: Brill. McCann, Hugh J. . “Divine Sovereignty and the Freedom of the Will.” Faith and Philosophy (): –. . “The Author of Sin?” Faith and Philosophy (): –. . “God, Sin, and Rogers on Anselm.” Faith and Philosophy (): –. . Creation and the Sovereignty of God. Bloomington, IN: Indiana University Press. . Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology. Oxford: Oxford University Press. McCann, Hugh J., and Daniel M. Johnson. . “Divine Providence.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/ spr/entries/providence-divine/.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
McKay, Thomas, and David Johnson. . “A Reconsideration of an Argument against Compatibilism.” Philosophical Topics (): –. McKenna, Michael. . “A Hard-Line Reply to Pereboom’s Four-Case Manipulation Argument.” Philosophy and Phenomenological Research (): –. a. “Defending Nonhistorical Compatibilism: A Reply to Haji and Cuypers.” Philosophical Issues (): –. b. “Moral Responsibility, Manipulation Arguments, and History: Assessing the Resilience of Nonhistorical Compatibilism.” Journal of Ethics (): –. . “Responsibility and Globally Manipulated Agents.” In The Philosophy of Free Will: Essential Readings From the Contemporary Debates, edited by Paul Russell and Oisin Deery, . Oxford: Oxford University Press. . “Resisting the Manipulation Argument: A Hard-Liner Takes It on the Chin.” Philosophy and Phenomenological Research (): –. . “Manipulation Arguments, Basic Desert, and Moral Responsibility: Assessing Derk Pereboom’s Free Will, Agency, and Meaning in Life.” Criminal Law and Philosophy (): –. Melamed, Yitzhak Y., and Martin Lin. . “Principle of Sufficient Reason.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/arch ives/spr/entries/sufficient-reason/. Mele, Alfred R. . Autonomous Agents: From Self-Control to Autonomy. Oxford: Oxford University Press. . Free Will and Luck. Oxford: Oxford University Press. . “Manipulation, Moral Responsibility, and Bullet Biting.” Journal of Ethics (): –. Mickelson, Kristin. . “The Soft-Line Solution to Pereboom’s Four-Case Argument.” Australasian Journal of Philosophy (): –. Murphy, Mark. . “Reply to Morriston.” In Divine Evil?, edited by Michael Bergmann, Michael Murray, and Michael Rea, –. Oxford: Oxford University Press. Nicolas, Jean-Hervé. . “La Permission Du Péché.” Revue Thomiste (): –, (): –, (): –. Nozick, Robert. . “Love’s Bond.” In The Examined Life: Philosophical Meditations, –. New York: Simon and Schuster. O’Connor, Timothy. . “On the Transfer of Necessity.” Noûs (): –. . “Probability and Freedom: A Reply to Vicens.” Res Philosophica (): –. Oderberg, David S. . “Divine Premotion.” International Journal for Philosophy of Religion (): –. Otsuka, Michael. . “Incompatibilism and the Avoidability of Blame.” Ethics (): –. Pasnau, Robert. . Thomas Aquinas on Human Nature: A Philosophical Study of Summa Theologiae a, –. Cambridge: Cambridge University Press.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Pawl, Timothy, and Kevin Timpe. . “Incompatibilism, Sin, and Free Will in Heaven.” Faith and Philosophy (): –. . “Heavenly Freedom: A Response to Cowan.” Faith and Philosophy (): –. Pereboom, Derk. . “Determinism Al Dente.” Noûs (): –. . Living without Free Will. Cambridge: Cambridge University Press. . “Free Will, Evil, and Divine Providence.” In God and the Ethics of Belief: New Essays in Philosophy of Religion, edited by Andrew Dole and Andrew Chignell, –. Cambridge: Cambridge University Press. . “Reasons-Responsiveness, Alternative Possibilities, and Manipulation Arguments against Compatibilism: Reflections on John Martin Fischer’s My Way.” Philosophical Books (): –. a. “A Compatibilist Account of the Epistemic Conditions on Rational Deliberation.” Journal of Ethics (–): –. b. “A Hard-Line Reply to the Multiple-Case Manipulation Argument.” Philosophy and Phenomenological Research (): –. . “Theological Determinism and Divine Providence.” In Molinism: The Contemporary Debate, edited by Ken Perszyk, –. Oxford: Oxford University Press. . Free Will, Agency, and Meaning in Life. Oxford: Oxford University Press. . “Libertarianism and Theological Determinism.” In Free Will and Theism: Connections, Contingencies, and Concerns, edited by Daniel Speak and Kevin Timpe, –. Oxford: Oxford University Press. . “Theological Determinism and the Relationship with God.” In Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology, edited by Hugh J. McCann, –. New York: Oxford University Press. Perszyk, Ken. . “Recent Work on Molinism.” Philosophy Compass (): –. Plantinga, Alvin. . God, Freedom, and Evil. Grand Rapids, MI: Eerdmans. . “Self-Profile.” In Alvin Plantinga, edited by James E. Tomberlin and Peter van Inwagen, –. Dordrecht: D. Reidel Publishing. Plato. . Complete Works, edited by John M. Cooper. Indianapolis: Hackett Publishing. Plotinus. . The Enneads, translated by Lloyd P. Gerson, George Boys-Stones, John M. Dillon, R. A. H. King, Andrew Smith, and James Wilberding. Cambridge: Cambridge University Press. Pruss, Alexander R. . The Principle of Sufficient Reason: A Reassessment. Cambridge: Cambridge University Press. . “Incompatibilism Proved.” Canadian Journal of Philosophy (): –. . “The First Sin: A Dilemma for Christian Determinists.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Ragland, C. P. . “Descartes on Divine Providence and Human Freedom.” Archiv Fu¨r Geschichte Der Philosophie (): –. . The Will to Reason: Theodicy and Freedom in Descartes. Oxford: Oxford University Press.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Robinson, Michael D. a. “Divine Guidance and an Accidentally Necessary Future: A Response to Hunt.” Religious Studies (): –. b. “Divine Providence, Simple Foreknowledge, and the ‘Metaphysical Principle.’” Religious Studies (): –. Rogers, Katherin A. . “Does God Cause Sin?” Faith and Philosophy (): –. . “God Is Not the Author of Sin: An Anselmian Response to McCann.” Faith and Philosophy (): –. . “Anselm against McCann: On God and Sin.” Faith and Philosophy (): –. . “The Divine Controller Argument for Incompatibilism.” Faith and Philosophy (): –. Sabini, John, and Maury Silver. . Moralities of Everyday Life. Oxford: Oxford University Press. Scanlon, Thomas. . Moral Dimensions: Permissibility, Meaning, Blame. Cambridge, MA: Belknap Press of Harvard University Press. Scruton, Roger. . Sexual Desire: A Moral Philosophy of the Erotic. New York: Free Press. Sennett, James F. . “The Free Will Defense and Determinism.” Faith and Philosophy (): –. Sher, George. . In Praise of Blame. Oxford: Oxford University Press. . Who Knew?: Responsibility without Awareness. Oxford: Oxford University Press. Sider, Theodore. . “Criteria of Personal Identity and the Limits of Conceptual Analysis.” Philosophical Perspectives : –. Simons, Peter. . “Relations and Truthmaking.” Aristotelian Society Supplementary Volume (): –. Singer, Irving. . The Pursuit of Love. Cambridge, MA: MIT Press. Smilanksy, Saul. . “What Choice Did I Have?” The Times Literary Supplement. www.the-tls.co.uk/articles/private/what-choice-did-i-have/. Smith, Angela M. . “On Being Responsible and Holding Responsible.” Journal of Ethics (): –. Smith, Michael. . “Rational Capacities, Or: How to Distinguish Recklessness, Weakness, and Compulsion.” In Weakness of Will and Practical Irrationality, edited by Sarah Stroud and Christine Tappolet, –. Oxford: Clarendon Press. Soble, Alan. . “Union, Autonomy, and Concern.” In Love Analyzed, edited by Roger E. Lamb, –. Boulder, CO: Westview Press. Sokolowski, Robert. . The God of Faith and Reason: Foundations of Christian Theology. Washington, DC: CUA Press. Solomon, Robert C. . About Love: Reinventing Romance for Our Times. New York: Simon and Schuster. Sripada, Chandra. . “Self-Expression: A Deep Self Theory of Moral Responsibility.” Philosophical Studies (): –. Steward, Helen. . “The Metaphysical Presuppositions of Moral Responsibility.” Journal of Ethics (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Stump, Eleonore. . Aquinas. London: Routledge. . Wandering in Darkness: Narrative and the Problem of Suffering. Oxford: Oxford University Press. Swinburne, Richard. . “Natural Evil.” American Philosophical Quarterly (): –. . “Reply to Morriston.” In Divine Evil? The Moral Character of the God of Abraham, edited by Michael Bergmann, Michael J. Murray, and Michael C. Rea, –. Oxford: Oxford University Press. Talbott, Thomas. . “The Topography of Divine Love: A Response to Jeff Jordan.” Faith and Philosophy (): –. Tanner, Kathryn. . God and Creation in Christian Theology: Tyranny or Empowerment? Minneapolis: Fortress Press. . “Human Freedom, Human Sin, and God the Creator.” In The God Who Acts, edited by Tom Tracey, –. University Park, PA: Pennsylvania State University Press. Taylor, Richard. . “Determinism.” In The Encyclopedia of Philosophy, edited by Paul Edwards, –:–. New York: Macmillan. Thomas Aquinas. . Summa Theologica, translated by Fathers of the English Dominican Province. London: R. and T. Washbourne. . The Disputed Questions on Truth. Vol. . Chicago: Henry Regnery Company. . Summa Contra Gentiles: Book One: God, translated by Anton Pegis. Notre Dame, IN: University of Notre Dame Press. . The de Malo of Thomas Aquinas, translated by Richard Regan. New York: Oxford University Press. Timpe, Kevin. . “Why Christians Might Be Libertarians.” Philosophia Christi (): –. Timpe, Kevin, and Daniel Speak. . Free Will and Theism: Connections, Contingencies, and Concerns. Oxford: Oxford University Press. Tobia, Kevin, Wesley Buckwalter, and Stephen Stich. . “Moral Intuitions: Are Philosophers Experts?” Philosophical Psychology (): –. Todd, Patrick. . “Manipulation and Moral Standing: An Argument for Incompatibilism.” Philosophers’ Imprint (), –. In press. “A Unified Account of the Moral Standing to Blame.” Noûs. Tognazzini, Neal A. . “The Structure of a Manipulation Argument.” Ethics (): –. Tooley, Michael. . “The Problem of Evil.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/fall/entries/evil/. Torre, Michael D. . God’s Permission of Sin: Negative or Conditioned Decree? A Defense of the Doctrine of Francisco Marin-Sola, O.P., Based on the Principles of Thomas Aquinas. Fribourg: Academic Press. Trabbic, Joseph G. . “Praemotio Physica and Divine Transcendence.” In Thomism and Predestination: Principles and Disputations, –. Ave Maria, FL: Sapientia Press of Ave Maria University.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
Turner, Jason. . “Compatibilism and the Free Will Defense.” Faith and Philosophy (): –. Turri, John. . “Compatibilism Can Be Natural.” Consciousness and Cognition : –. Vaesen, Krist, Martin Peterson, and Bart Van Bezooijen. . “The Reliability of Armchair Intuitions.” Metaphilosophy (): –. Van Inwagen, Peter. . An Essay on Free Will. Oxford: Oxford University Press. . “The Place of Chance in a World Sustained by God.” In God, Knowledge, and Mystery, –. Ithaca, NY: Cornell University Press. . Metaphysics. Boulder, CO: Westview Press. . “Free Will Remains a Mystery.” Philosophical Perspectives : –. . The Problem of Evil: The Gifford Lectures Delivered in the University of St. Andrews in . Oxford: Oxford University Press. . “How to Think about the Problem of Free Will.” Journal of Ethics (/): –. . “Some Thoughts on An Essay on Free Will.” Harvard Review of Philosophy : –. a. “Concluding Meditation.” In Being, Freedom, and Method: Themes from the Philosophy of Peter van Inwagen, edited by John A. Keller, –. Oxford: Oxford University Press. b. “The Problem of Fr** W*ll.” In Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology, edited by Hugh J. McCann, –. Oxford: Oxford University Press. c. Thinking about Free Will. Cambridge: Cambridge University Press. Vargas, Manuel. . “On the Importance of History for Responsible Agency.” Philosophical Studies (): –. Vicens, Leigh. . “Divine Determinism, Human Freedom, and the Consequence Argument.” International Journal for Philosophy of Religion (): –. . “Theological Determinism.” Internet Encyclopedia of Philosophy. . “Objective Probabilities of Free Choice.” Res Philosophica (): –. Vihvelin, Kadri. . “Free Will Demystified: A Dispositional Account.” Philosophical Topics (/): –. . Causes, Laws, and Free Will: Why Determinism Doesn’t Matter. New York: Oxford University Press. Wallace, R. Jay. . “Hypocrisy, Moral Address, and the Equal Standing of Persons.” Philosophy and Public Affairs (): –. Walls, Jerry. . “Why No Classical Theist, Let Alone Orthodox Christian, Should Ever Be a Compatibilist.” Philosophia Christi (): –. Welty, Greg. . “Molinist Gunslingers: God and the Authorship of Sin.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Wertheimer, Roger. . “Constraining Condemning.” Ethics (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
References
White, Heath. . “Theological Determinism and the ‘Authoring Sin’ Objection.” In Calvinism and the Problem of Evil, edited by David E. Alexander and Daniel M. Johnson, –. Eugene, OR: Pickwick Publications. Widerker, David. . “On an Argument for Incompatibilism.” Analysis : –. . “A Problem for the Eternity Solution.” International Journal for Philosophy of Religion (): –. . “Providence, Eternity, and Human Freedom: A Reply to Stump and Kretzmann.” Faith and Philosophy (): –. William of Ockham. . Predestination, God’s Foreknowledge, and Future Contingents, translated by Marilyn McCord Adams and Norman Kretzmann. Indianapolis: Hackett Publishing. Williams, Thomas. . “Human Freedom and Agency.” In The Oxford Handbook of Aquinas, edited by Brian Davies and Eleonore Stump, –. Oxford: Oxford University Press. Wolf, Susan. . Freedom within Reason. Oxford: Oxford University Press. Yaffe, Gideon. . “Indoctrination, Coercion and Freedom of Will.” Philosophy and Phenomenological Research (): –.
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.011
Index
ability to do otherwise, –, –, –, See also Principle of Alternative Possibilities complete ability to do otherwise, modest ability to do otherwise, Alexander, David E., Almeida, Michael, , Anselm, Armenianism, aseity, divine and the Theory of Middle Freedom, – Augustine, , author metaphor, –, –, and the problem of creaturely love, – author of sin objection disambiguation, authority, divine, – Báñez, Domingo, beliefs desire beliefs, – desire beliefs vs possibility beliefs, possibility beliefs, – pre-philosophical, – religious, –, – Beta, , Bigelow, John, blame, – divine, – God’s standing to, – and hypocrisy, – internal vs external, nature of, blameworthiness objection and the appeal to teleology, – and the argument from analogy, – and divine causality, – and divine command theory, – and general moral precepts, – and harm done to the wicked, – and harms other than murder,
consequentialist reply to, – disjunctive reply to, – hard-line response to, – and the harm of dying, informal characterization of, – intention-based reply to, – vs problem of evil, – and the question of divine moral agency, – bullet-biting, – Byerly, T. Ryan, , – Calvin, John, –, – Calvinism, causality, divine, –, –, and the problem of creaturely love, , vs human, – causation horizontal vs vertical, choices, morally significant, Cohen, G. A., – commands, divine, –, See also divine command theory and deception, – and desire beliefs, – and divine intention, – and the divine will, – and human fulfillment, – indeterminate, weighed against the divine will, commands, sincere, compatibilism and the authorship of scripture, – definition of, – historical, – independent, – nonhistorical, – reasons for investigating, Consequence Argument, the, – Consequentialism, –
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.012
contingency, –, –, contrition, , – and indeterminism, control basic, – direct, – indirect, – nonbasic, – regulative, – counterfactual of creaturely freedom, control over, – Cowan, Steven B., –, Crabtree, J. A., Davies, Brian, – deception, divine, – and desire beliefs, and possibility beliefs, , dependence God’s on humans, – humans’ on God, , of human knowledge, – Descartes, René, determinism, divine and Calvinism, definition of, , – as distinct from foreknowledge, and explanation, , – and the evidential problem of evil, –, and the giving of divine commands, – and the logical problem of evil, – and the problem of creaturely love, – and the problem of divine love, – as pragmatically defined, Edwardsian divine determinism, –, –, – hard, – importance of, – motivations for, – primary divine determinism, –, , –, –, – varieties of, determinism, natural, determinism, physical definition of, –, as a live option, – as a motivation for divine determinism, – worries concerning, –, Deuteronomy, Book of, dialectical burden, , disjunctive replies, –, divine command theory, –
Index Edwards, Jonathan, –, ends, human vs divine, – essentialism, verbal, eternality, divine, – Exodus, Book of, explanations, noncausal, – Fischer, John Martin, –, –, Flew, Antony, foreknowledge, divine as distinct from determinism, foreseen vs intended, – Four-Case Argument, – hard-line reply to, soft-line reply to, , – Freddoso, Alfred, free will. See also compatibilism; incompatibilism; libertarianism and the authorship of scripture, and the Consequence Argument, – hanging by a thread, and moral responsibility, and the Revised Consequence Argument, – Free Will Defense basic presentation of, – goal of, – and the inclusion of false claims, – and independent compatibilism, – and libertarianism, and the Theory of Middle Freedom, – Genesis, Book of, glory, divine, –, governance, divine, , See also sovereignty, divine grace, divine, Grant, W. Matthews, , –, – Green, Christopher, hard-line reply. See Four-Case Argument: hard-line reply to; problem of creaturely love: hard-line reply to; Revised Four-Case Argument: hard-line reply to Hart, Matthew, Hasker, William, , hell and divine love, – and the greater glory reply to the problem of evil, – and respect for human autonomy, – Hick, John, – Howsepian, A. A., –
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.012
Index identity, personal, – incompatibilism and the authorship of scripture, – definition of, Independence (principle), benefits for divine determinists, – costs of, and Edwardsian divine determinism, – and primary divine determinism, inference to the best explanation, intention, divine and the blameworthiness objection, and divine commands, – and independent compatibilism, – Johnson, Daniel M, Joshua, Book of, judgment, – justice, divine, Kant, Immanuel, King, Matt, – knowledge, divine, as different from human knowledge, – libertarianism, love, divine, as identical with God, as parental, – revising, – as sacrificial, as a source of benefit, love, human and manipulation, types of, – and union, love, union, – and autonomy, – and historical sensitivity, and manipulation, revised account of, Mackie, John, manipulation and the blameworthiness objection, – definition of, – differences between human and divine, – and the standing to blame, and union love, manipulation arguments. See Four-Case Argument; problem of creaturely love: versions of; Revised Four-Case Argument; Zygote Argument
McCann, Hugh, –, –, –, –, McKenna, Michael, Mele, Alfred, – mercy, divine, and the greater glory response, modest privation solution, – objections to, – Molina, Luis de, Molinism, moral agency, divine, – moral responsibility, , – and Consequence-style arguments, denial of, – God’s moral responsibility for sin, – and manipulation arguments, – for sin, – necessity absolute, – conditional, , metaphysical, nomological, suppositional, non-involvement condition, – Oderberg, David, , – omissions, as causes, as effects, – One-Case Argument, – responses to, – ought implies can, – Pawl, Timothy, – Pereboom, Derk, , –, –, –, Philippians, Letter to the, – Plantinga, Alvin, , –, – Plato, Plotinus, power, divine, premotionism, – Principle of Alternative Possibilities, and divine deception, Principle of Double Effect, Principle of Middle Freedom, Principle of Non-Excuse, – Principle of Sufficient Reason, – problem of contrition, – responses to, – problem of creaturely love, – and Edwardsian divine determinism, and the First Principle of Union Love, hard-line reply to, –
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.012
Index
problem of creaturely love (cont.) loose characterization of, and participation, – and primary divine determinism, and the Second Principle of Union Love, – versions of, – problem of divine blame, – problem of divine love and the author metaphor, costly responses to, – costs of, – disjunctive reply to, and future autonomy, – for indeterminists, – and personal identity, – and respect for human autonomy, – responses to, – and revising the account of divine love, – and sacrificing an object of love, – and skepticism, – and universalism, – unsatisfactory responses to, – problem of evil and character formation, evidential, – and the greater glory reply, greater good responses to, – and independent compatibilism, – logical, – logical vs evidential, and natural laws, and the Theory of Middle Freedom, – prophecy, – providence, divine, – Pruss, Alexander, , punishment, vs blame, – punishment, divine, – Quran, relations, internal, – Revised Consequence Argument, – Revised Four-Case Argument, – hard-line reply to, – responses to, – soft-line reply to, –, – robust privation solution, – Rogers, Katherin, –, , –, scripture as a motivation for divine determinism, – authorship of, –
Scruton, Roger, semi-compatibilism, – Sher, George, sin act of, – composition of, – defect of, –, – and the divine will, –, and hard divine determinism, – and the Principle of Non-Excuse, – privation account of, skeptical theism, – and the logical problem of evil, Socrates, sovereignty, divine, – and independent compatibilism, Strong Authority Thesis, – and the appeal to mystery, objection to, – Stump, Eleonore, Swinburne, Richard, , – Tanner, Kathryn, –, teleology, – Theory of Middle Freedom, – and agential control, – and Molinism, and the Preliminary Principle of Middle Control, and the Principle of Middle Freedom, worries about, – Thomas Aquinas, , Timpe, Kevin, – Todd, Patrick, –, Tooley, Michael, Trabbic, Joseph, – transcendence, divine, – Turner, Jason, – universalism and hard divine determinism, and the problem of divine love, – van Inwagen, Peter, , – Vicens, Leigh, – Wallace, R. Jay, Weak Authority Thesis, – insufficiency of, – Westminster Confession of Faith, White, Heath, , –, , –, – Widerker, David, will, divine antecedent vs consequent, –, – and the definition of divine determinism, –
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.012
Index and divine commands, – and human contrition, – and human deliberation, – human ignorance of, human knowledge of, – and human motivation, and human willing, and the Principle of Non-Excuse, –
and sin, – weighed against divine commands, will, human antecedent vs consequent, – and the divine will, – worm account. See identity, personal Zygote Argument, –
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.012
Downloaded from https://www.cambridge.org/core. Helsinki University Library, on 23 May 2021 at 18:30:52, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108696845.012