The Problem of Animal Pain: Trent Dougherty

PALGRAVE FRONTIERS IN PHILOSOPHY OF RELIGION
Series Editors: Yujin Nagasawa, University of Birmingham, UK, PALGRAVE FRONTIERS IN PHILOSOPHY OF RELIGION
and Erik J.Wielenberg, DePauw University, USA
The problem of evil constitutes the greatest challenge to rational belief

in the existence of God. Animal suffering constitutes perhaps the most
The Problem of Animal Pain

powerful version of the problem. Considerations that render human
suffering theologically intelligible seem inapplicable to non-human
animals. It is commonly held that they do not have morally significant
free will, they do not have immortal souls, and they do not have a direct
relationship with God. In this book, Dougherty defends radical possibilities
for animal afterlife that allow a soul-making theodicy to apply to animals.
He defends that animals have souls, and a novel model of materialist
resurrection if they don’t. He then proposes that animals will undergo
theosis and given the expanded cognitive resources to understand and
embrace their place in the scheme of salvation. Along the way we get
tours of probability theory, four-dimensionalism, and chimpanzee behavior.
From the split-brain experiment to the relationship between mammalian
and avian brains, this tour de force challenges conventional wisdom on
The Problem
the theology of animals.
Trent Dougherty is Assistant Professor of Philosophy at Baylor University,

USA. He is the editor of Evidentialism and Its Discontents, co-editor (with
Justin McBrayer) of Skeptical Theism: New Essays, and author of numerous
book chapters, essays, reviews, and reference works in epistemology and
philosophy of religion.
of Animal
Pain
Trent Dougherty
A Theodicy for All Creatures Great and Small
ISBN 978-0-230-36848-4
90101
9 780230 368484
Cover image: ‘The Animals Entering Noah’s Ark’ oil on canvas
painting by Giovanni Benedetto Castiglione called Il Grechetto
© ArtPix / Alamy
Trent Dougherty
Content Type: Black & White

Paper Type: White
Page Count: 212
File type: Internal
Palgrave Frontiers in Philosophy of Religion
Series Editors: Yujin Nagasawa and Erik Wielenberg
Titles include:
Zain Ali
FAITH, PHILOSOPHY AND THE REFLECTIVE MUSLIM
István Aranyosi
GOD, MIND AND LOGICAL SPACE:
A Revisionary Approach to Divinity
Trent Dougherty
THE PROBLEM OF ANIMAL PAIN:
A Theodicy for All Creatures Great and Small
Benedikt Paul Göcke
A THEORY OF THE ABSOLUTE
Yujin Nagasawa (editor)
SCIENTIFIC APPROACHES TO THE PHILOSOPHY OF RELIGION
Aaron Rizzieri
PRAGMATIC ENCROACHMENT, RELIGIOUS BELIEF AND PRACTICE
Aaron Smith
THINKING ABOUT RELIGION:
Extending the Cognitive Science of Religion
Eric Charles Steinhart
YOUR DIGITAL AFTERLIVES:
Computational Theories of Life after Death
Forthcoming titles:
Michael Sudduth
A PHILOSOPHICAL CRITIQUE OF EMPIRICAL ARGUMENTS FOR LIFE AFTER
DEATH
Palgrave Frontiers in Philosophy of Religion

Series Standing Order ISBN 978–0–230–35443–2 Hardback
(outside North America only)
You can receive future titles in this series as they are published by placing a standing order.
Please contact your bookseller or, in case of difficulty, write to us at the address below with
your name and address, the title of the series and the ISBN quoted above.
Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke,
Hampshire RG21 6XS, England
The Problem of Animal
Pain
A Theodicy for All Creatures
Great and Small
Trent Dougherty
Baylor University, USA
© Trent Dougherty 2014
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6–10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The author has asserted his right to be identified as the author of this work
in accordance with the Copyright, Designs and Patents Act 1988.
First published 2014 by
PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,
175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies
and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,
the United Kingdom, Europe and other countries
ISBN 978-1-349-34995-1 ISBN 978-1-137-44317-5 (eBook)
DOI 10.1057/9781137443175
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
Transferred to Digital Printing in 2015

I dedicate this book to Caesar, the best chocolate lab ever; Roxy,
the youngest kitten I ever rescued and the first cat the kids ever
got; Bagheera, the blackest cat who found us on the blackest
night; Ka, the goldenest snake, and to the memories of Zeke,
my first “very own” lab who was killed by a car on her second
birthday (I understand “Zeke” is a nontraditional name for a
female anything); to the memory of Schneeflocke, the first cat to
come to Sarah and I and whom we lost when Sarah was pregnant
with Fiona; and to the memories of Chickpea and Azure, whom
we lost to disease very recently to this writing, even after the
expert care of Fiona. They had short lives, but they were sweet,
thanks to her. Sadly, also I must mention Roxy, the part Maine-
coon we lost during the writing of this book. A tiny, abandoned
kitten, I saved her from Toms under the bushes of a building
on campus. She traveled with us from Texas to Minnesota to
Kansas City to South Bend in the car. She is very sorely missed.
The most recent loss during the writing of this book – alas this
acknowledgement has become a catalog of sorrow – is Lyndsay’s
Blitz. May he run with Caesar and Zeke on the golden shores.
We miss these companions dearly and daily. Ka died while I
was finishing the proofs of this book at St. Andrew’s. We are
very happy, though, for the new friends who have come our way:
Rocky, She-her-kahn, and Mr. Skittles, three rescued cats, and a
pair of sister-pups, Daria and Sophie, who in six months now
have never slept in any other configuration than a pile. We are
lucky to have Alli and Jordan to care for them while we are in
St. Andrew’s and Oxford. We miss them and are ready to see
them.
Contents
Series Editors’ Preface x
Acknowledgments xi
1 The Plan of This Book 1

1.1 Background 1
1.2 Epistemology 5
1.3 Not-so-bare bare theism 7
1.4 Two ways of learning what theism entails 12
2 The Problem of Animal Pain 16

2.1 Stating the problem 16
2.1.1 What is the problem? 16
2.1.2 Getting more precise 19
2.1.3 Theism as “hypothesis” 24
2.2 How shall I understand “pain” and “suffering” 25
2.3 The amount of suffering 31
3 The Bayesian Argument from Animal Pain 36

3.1 Formalizing the problem: a Bayesian approach 36
3.1.1 The formal structure of the argument from
animal suffering 36
3.1.2 The simplicity of the hypotheses 38
3.1.3 The Bayes factor 42
3.1.4 Taking stock thus far 43
3.2 A narrative approach to assigning probabilities 46
3.3 Criteria for success: between theodicy and defense 51
4 Is There Really a Problem?: The Challenge of

Neo-Cartesianism 56
4.1 A starting point: common sense and scientific consensus 57
4.1.1 Common sense epistemology 57
4.1.2 Scientific consensus 61
4.2 A statement of the objection 64
4.2.1 Introduction 64
4.2.2 Kinds of neo-Cartesianism and how they
function epistemically 65
vii
viii Contents
4.2.3 Some motivation for neo-Cartesianism 69

4.2.3.1 From philosophy: HOT of consciousness 71
4.2.3.2 From neuroanatomy: the lobotomy
argument 74
5 There Is a Problem: The Defeat of Neo-Cartesianism 77

5.1 Initial reply to neo-Cartesianism: moving
away from rationalism 77
5.1.1 Pain as emotion-like 78
5.1.2 Non-conceptual knowledge of pain 82
5.1.3 Recap 86
5.2 The problem of primate pain or seeing through
the mirror test 86
5.2.1 A statement of the problem 86
5.2.2 Objection from Carruthers 91
5.2.3 A final argument 94
6 The Saint-Making Theodicy I: Negative Phase 96

6.1 Some necessary ground clearing 96
6.2 Axiology and teleology 103
6.3 The defeat of evil 108
7 The Saint-Making Theodicy II: Positive Phase 118

7.1 The fine-tuning argument for theism from evil 118
7.2 Application to the Bayesian argument from evil 122
7.3 Too much suffering? What’s the evidence? 127
7.4 A narrative approach 131
8 Animal Saints 134

8.1 Doubt about animal soul-making from within the ranks 136
8.1.1 A cause of the lacuna in previous soul-making
theodicies 137
8.1.2 Some consequences of the lacuna 139
8.2 Deification: human and non-human 143
8.3 The transcendental argument for animal deification 145
8.4 Identity issues: objections to the animal deification 148
9 Animal Afterlife 154
9.1 Animal souls 155
9.1.1 Biblical support for animal souls 155
9.1.2 The traditional Christian view about animal souls 158
9.1.3 Philosophical support for animal souls 162
9.1.3.1 The knowledge argument 163
Contents ix
9.1.3.2 Qualia: inverted and absent 164

9.1.4 Conclusion 166
9.2 Gappy existence 166
9.2.1 Introduction 166
9.2.2 Natural skepticism about gappy existence 166
9.2.3 A description of the model 167
9.2.4 Split brains: the terrible tale of twins Ted and Todd 170
9.2.5 Objections to the present view of resurrection 173
Concluding Summary 179
Bibliography 183
Index 195
Series Editors’ Preface
The philosophy of religion has experienced a welcome revitalization

over the last fifty years or so and is now thriving. Our hope with the
Palgrave Frontiers in Philosophy of Religion series is to contribute to the
continued vitality of the philosophy of religion by producing works that
truly break new ground in the field.
Accordingly, each book in this series advances some debate in the
philosophy of religion by offering a novel argument to establish a strik-
ingly original thesis or approaching an ongoing dispute from a radically
new point of view. Each book accomplishes this by utilizing recent devel-
opments in empirical sciences or cutting-edge research in foundational
areas of philosophy, or by adopting historically neglected approaches.
We expect the series to enrich debates within the philosophy of reli-
gion both by expanding the range of positions and arguments on offer
and establishing important links between the philosophy of religion
and other fields, including not only other areas of philosophy but the
empirical sciences as well.
Our ultimate aim, then, is to produce a series of exciting books that
explore and expand the frontiers of the philosophy of religion and
connect it with other areas of inquiry. We are grateful to Palgrave
Macmillan for taking on this project as well as to the authors of the
books in the series.
Yujin Nagasawa
Erik J. Wielenberg
x
Acknowledgments
I wish to express my gratitude to editors Yujin Nagasawa and Erik

Wielenberg for encouraging me to pursue this project. Since the ideas
in this book are primarily inspired by both John Hick (who, God rest
his soul, died while I was writing it) and C.S. Lewis (whose 50th memo-
rial passed while I was writing it), it is a special blessing that it is being
published by a publishing tradition – Palgrave Macmillan – in which
their major works appeared. I would also like to express thanks to Alex
Pruss – an Anselmian colleague – for bringing the series to my attention
and bringing my work to the attention of Yujin and Erik. We may have
made the connection anyway, but it sweetens the deal that Alex was the
bridge. I owe a great debt of gratitude to the Center for Philosophy of
Religion at the University of Notre Dame where major parts of the manu-
script were written, to Mike Rea as its exemplary director, and to the
other fellows – especially Ty Goldschmidt for many great conversations,
and to Beth Seacord and Dave Anderson whose work tied in with my
own. Beth gave me a steady stream of excellent feedback on my writing
and ideas and provided guidance to the literature on animal suffering.
She is writing her dissertation (at Colorado with Wes Morriston and
Michael Tooley) on the problem of animal pain, and her work will be
much more detailed than my own in many ways. Also writing a disser-
tation (at Baylor, with Jon Kvanvig) on the problem of animal pain is
Matt Douglas, who was helpful at various stages of writing. Faith Pawl’s
dissertation (at Saint Louis University with Eleonore Stump) also relates
to animal pain especially as it relates to primates. These three students
will likely usher in a new wave of philosophical and theological consid-
eration of animal suffering. Eleonore Stump has been an inspiration in
the writing of this book in an interesting way. Not only has her work in
philosophical theology been an inspiration generally, but she shares a
concern for those who face certain cognitive challenges in life, such as
those animals and people with special mental challenges face (Temple
Grandin’s work indicates the tie between the two). More specifically,
her Wandering in the Darkness came out during the writing of this book,
and confirmed and clarified many of my own speculative wanderings
as I strove for a better understanding of the nature of the problem of
evil. Richard Swinburne has been a huge support to me for the whole
xi
xii Acknowledgments
of my career, and it is obvious to see the ways he’s influenced me. Alvin
Plantinga also gave me very helpful feedback on portions of the MS.
Some of my ideas that led me to the literature on pain and emotions
started with conversations with Linda Zagzebski. Almost every idea I’ve
ever had has been discussed with either Alex Pruss or Jon Kvanvig, and it
is impossible to footnote all the places where their influence is at work.
My stay at Notre Dame would have been impossible without the
generous support of the Baylor Institute for Studies of Religion and its
director Byron Johnson, and the tireless advocacy of my chair, Mike
Beaty. The proofs were completed while at the University if St. Andrew’s
and I thank the Faculty of Philosophy for an office at the beautiful and
aptly-named Edgecliffe and for St. Mary’s Divinity School for an office
in their historic 16th century offices.
I have been thinking about these matters since childhood and quite
seriously since college. I was reinvigorated to this inquiry by a class I
have taught at Baylor University since the Fall of 2010 on the thought
of C.S. Lewis as well as an independent study with Lyndsay McReynolds
and Josh Jeffrey (whose term paper was on certain problems with Lewis’s
view of animal resurrection with which I have yet to find a satisfac-
tory solution). I benefited from the paper and discussion of Collin Huse
who did a similar independent study with me. My superb colleague Bob
Roberts was an invaluable help in conversation and in his excellent work
on emotions. His influence on this work is inestimable and is too perva-
sive to footnote adequately, even though it may appear I have failed to
take his excellent advice in places. Kate Zeigerer (DVM, Tufts, Animal
Care Specialist/Veterinary Medical Officer, USDA) was a constant source
of encouragement, enthusiasm, and information throughout. Her influ-
ence is also pervasive on this book (though government regulations
require me to say that she never acted in any official capacity as advisor
to the book).
I received wonderful feedback from all of the students who read
the penultimate draft of the manuscript for a graduate seminar: Nick
Colgrove, Brandon Dahm, Blake McAllister, Dan Padgett, Rebecca
Phillips, and Allison Thornton. Each one made a valuable contribution
to the book. Nick must be singled out, however, for having the extreme
misfortune of being my RA during the final stages of the book. It is not
mere acknowledgments boilerplate to say the whole thing may have
fallen apart at the end if not for his and his wife Bethany’s support.
Chris Tweedt was my previous RA and also worked extensively on the
manuscript.
Acknowledgments xiii
Finally, I wish to thank my beloved wife Sarah, and my kids Fiona,

Annabelle, Jeep, and Sam who are such great animal lovers and such
great animal care-takers (indeed, they have a knack for discerning and
promoting the conditions of flourishing of every living thing).
1
The Plan of This Book
In this chapter, I provide some background that will aid the reader in
following the argument of this book. This book is philosophical, yet
very personal. I will state here some of the personal and professional
background for the writing of the book (§1.1) as well as some of the
epistemological assumptions (§1.2). I describe a core notion of the book:
“expanded ‘bare’ theism” (§1.3, §1.4), the main hypothesis under scru-
tiny in this book.
1.1 Background
I taught this manuscript in a graduate seminar in philosophy of religion

in the fall of 2013 (the names of the students are in the acknowledg-
ments). I skipped this prologue. I wrote it primarily for myself as a sort of
“backstory.” After having them read it at the end of the seminar, they all
said that it was very helpful and should be included in the book. They
are all very good and smart people, so I have followed their advice. And
upon reflection, it now seems clear to me that they were right. Having
just re-read the entire manuscript from beginning to end, it is clear to
me that though the ideas to follow in this chapter were guiding all I
wrote, I was not as explicit there as I could have been. One option was to
skip this introduction and explicitly mention these guiding ideas at each
point in subsequent chapters, but to be so meta so much of the time
would severely clutter the first-order material. Thus, I highly advise that
you do not skip this chapter, but, on the contrary, read it very carefully if
you are interested in the theme and argument of this book, for keeping
it in mind as you go will help avoid misunderstanding and frustration.
When I saw the call for proposals for this series, I interpreted it as, in
effect, a call to write down and defend crazy stuff one would otherwise not
1
2 The Problem of Animal Pain
dare to write. That’s not what it was, of course, but it was close enough for
me to believe I had just the right project. I do not deny for a moment that
many of the ideas I will put forward in this book will strike the average
reader as a bit far-fetched. However, I wish to mitigate that in at least two
ways. First, I will try to situate those ideas in a context in which they won’t
seem so strange after all.1 I will even try to affect a gestalt in which what at
first seemed crazy now seems obvious when looked at from the right angle.
This context will include neglected ideas from the Judeo-Christian tradi-
tion, especially in the Eastern Church. Documenting these sources would
be a monumental task, but my thinking on these matters has been most
deeply influenced by Eastern Orthodox thought. Some of these ancient
and Eastern ideas will be somewhat familiar to readers of this book, such
as the “soul-making” theodicy pioneered by Saint Irenaeus, the 2nd
century Bishop of Lyons who led his church during the persecutions of
Marcus Aurelius. Thus sometimes the soul-making theodicy is called the
“Irenaean Theodicy.” Irenaeus’ ideas were promoted most prominently in
the 20th century by John Hick’s (1966) watershed Evil and the God of Love.
Other ideas, such as deification/theosis and the redemption of animals
and, indeed, creation as a whole, are less well-known. Some ideas that are
a standard part of Eastern Christianity are long-forgotten or neglected in
the West. I hope to recover those ideas and present them in a compelling
way and to extend them in ways that, though truly radical, are also in a
way quite natural. I only wish I could do full justice to this tradition by
documenting the hundreds of sources that influenced me. A volume on
the history of these ideas would be easy to produce and quite beneficial.
Their inclusion in this volume, however, was not an option.
As Hick points out, the Irenaean picture is very different from the
Augustinian one. Though Augustine may have considered the Fall a
“felix culpa” it still didn’t play anything like the role attributed to it by
Irenaeus. For Irenaeus, the struggle is the point of creation, or, rather,
the first stage of creation. (The idea that we should expect creation to
proceed in stages is of immeasurable importance.) One could be forgiven
for thinking that the ideal creation according to Augustine consists in
stasis after initial perfection. For Irenaeus, nothing could be further from
the truth. We need trials and tribulations to become what God wants us
to be. What we become at the end of tribulation is what we were meant
to be all along: Saints and Martyrs.
1
In this regard my project is somewhat like that of Eleonore Stump’s use of
Aquinas in her masterful Wandering in Darkness: Narrative and the Problem of
Suffering (2010), only I am going further back and farther east than 13th century
Europe.
The Plan of This Book 3
Another father important to the East is the 6th/7th century theolo-

gian Maximus the Confessor. One of his key ideas – and he is far from
the only eastern father to hold such a view – was called “theosis” or
“deification.” The basic idea is that because God the Son took upon
human nature fully (the hypostatic union: Jesus Christ is “fully God
and fully man”) the human and the divine are inextricably bound
together, and the eschatological, soteriological destiny of humans is to
be taken up into the divine nature. But theosis doesn’t end there. It ends
up expanding to encompass the redemption of all of nature. I want to
bring these themes from Irenaeus and Maximus together and expand
them. I will defend the thesis that a class of animals (to be discussed
later) will not only be resurrected at the eschaton, but will be deified
in much the same way that humans will be. That they will become, in
the language of Narnia, “talking animals.” Language is the characteristic
mark of high intelligence. So I am suggesting that they will become
full-fledged persons (rational substances) who can look back on their
lives – both pre- and post-personal – and form attitudes about what has
happened to them and how they fit into God’s plan. If God is just and
loving, and if they are rational and of good will, then they will accept,
though with no loss of the sense of the gravity of their suffering, that
they were an important part of something infinitely valuable, and that
in addition to being justly, lavishly rewarded for it, they will embrace
their role in creation. In this embrace, evil is defeated.
A movie scene that nicely captures part of this picture is in Peter
Jackson’s adaptation of The Lord of the Rings. In the last movie, The Return
of the King, toward the end, the coronation of Aragorn is depicted. After
his coronation, he walks through the ranks, all bowing to him. Then he
directs attention to the principle hobbit protagonists. He praises their
courage, and all bow to them in respect, including the King. At that time,
without for a moment downplaying the gravity of what they have been
through (one is exhausted from their exhaustion by that point), it seems
that they would not change what they have been through.2 They embrace
their role in the story, and in so embracing it, the evil of Sauron is defeated
in a way greater than the military destruction of him and his forces.
Few have explored some of the territory I cover here, fewer in analytic
philosophy of religion.3 As the series name implies, this book is a report
2
And they did not exactly know what they were getting into!
3
Stephen Webb raises the problem of animal pain in Webb (1998), in a section
of Chapter 2 and raises the question of animal afterlife in a section of Chapter 7.
Jerry Walls mentions the problem and raises relevant questions in Walls (2002:
64); and gives a hint and a hope on p. 91.
from the frontier. All I can do is send back a report and hope others
get interested and move in to explore the territory further (as indeed
some of the people mentioned in the Acknowledgements are doing even
now). One of the most exciting things I discovered is that many people
are doing research relevant to the problem I treat. It’s just that there is
not a strong network connecting them.4 I hope this book contributes to
greater networking among these individuals in general and especially
between analytic philosophers on the one hand and applied ethicists,
theologians, and scientists on the other. I think all three of the latter can
benefit greatly from the tools of the analytic philosopher, and I think
analytic philosophers can learn much about what matters and how to
look at the big picture from them.
Of course, I didn’t start with a wholly untilled field. Just a few years
ago, Mike Murray, sitting not very far from where I sit now in the Notre
Dame Center for Philosophy of Religion, perhaps in the same office,
perhaps in the same chair,5 wrote Nature Red in Tooth and Claw (2008).
Most of the ideas for the present book were formed well before that book
was written, and I have quite a different take on the issue and quite a
different aim and style in this book. However, he is due much credit
for his pioneering work on this subject for analytic philosophers. Both
because I am tilling parts of the same field (some of which overlap parts
he plowed, others farther out in the field) and because the literature on
this topic is quite small, I will be interacting with Murray quite a bit,
both positively and negatively.
The greatest difference between my project and Murray’s is that
Murray dismisses the possibility of a soul-making theodicy for animals
with almost no consideration (125). I expect he is not alone in this.
But why might this be? Are there really conclusive reasons against this
possibility? Is there some set of necessary conditions for a soul-making
theodicy to apply, which are clearly not met? One might be forgiven for
thinking so. I will do my best to decrease the reader’s tendency to think
so. My central project is to develop a new species of more radical soul-
making theodicy I call the “saint-making” theodicy, since it focuses not
just on the (very important) second-order moral virtues and character
4
Though I have great hope for the Oxford Center for Animal Ethics, whose
director, Andrew Linzey, has just been appointed the first ever Professor of Animal
Theology. For more details, see Oxford Centre for Animal Ethics (2013). I was not
able to read as much of Linzey’s work as I would have liked, but I especially found
confirmation of my ideas in Linzey and Regan (1990).
5
We both wrote significant parts of our books while fellows at the Center for
Philosophy of Religion, at Notre Dame.
traits but, specifically, on the characters of a saint as conceived especially

in Catholicism but is present in other forms of Christianity, certainly in
Judaism, and perhaps other religions both theistic and atheistic.
The above constitutes generalities that help situate the reader, so that
they can get the most from this book. In the next section, I discuss the
general structure of my project. Analytic philosophy is characterized by
prizing explicit logical structure and explicit appeal to logical systems.
The good reasons for this are clear: It makes for clear and precise expres-
sion. There is another virtue in it: It makes it easy for the reader to iden-
tify where they disagree. It exposes the flaws, puts them front-and-center,
rather than burying them beneath turgid prose. I confess that I cannot in
this book meet many of the standards standard analytic philosophers will
hope for. That is a necessary feature, for this poor author, of the pioneering,
interdisciplinary work in a short book. Nevertheless, there is a formal
aspect to the book, indeed, at work throughout it. It is a work of applied
Bayesianism. My fundamental approach is one of probabilistic reasoning.
This approach differs significantly from approaches that focus only on
deductive reasoning. The next section of this introduction provides a
brief but explicit treatment of the formal structure of my reasoning.
1.2 Epistemology
Throughout the book, I will be using some probabilistic vocabulary just

for economy of expression and to make more (but not perfectly!) precise
certain claims and relations. Thus, I want to introduce this vocabulary
right away. I will use “Pr(A)” to express the probability of whatever is
in the parentheses, in this case a proposition A. I will use “Pr(A | B)”
to express the probability of A given B. This is called conditional proba-
bility. For example, Pr(A heart is drawn | A card is drawn from a shuffled,
complete deck of cards) is ¼. The meaning of “given” is not perfectly
clear. Relating conditional probabilities to the probability of conditionals
is hard, but we needn’t go into that here.6 The intuitive notion gener-
ated by the card example will take us far. But not all conditional prob-
abilities are so simple. Consider Pr(My wife will be disappointed | I forget
to bring home salad). This probability is high but it is not obtained just
from counting cases. Now consider Pr(I will be moved to tears | I hear
Bach’s St. Matthew’s Passion). This is also high, but involves presently
ineffable or barely-effable knowledge. If you are to assign a value, you
6
See Hájek (2003a, b) for an overview of the issues.
must have some second-personal knowledge of me.7 This affects but

does not negate our ability to assess the probability in question.
Also, there are many different interpretations of probability.8 I think
most of what I say here will work on almost any theory of probability. I
do not, however, favor a “mathematical” interpretation of probability.
I take myself to be employing a broadly “epistemic” and “evidential”
notion of probability. Roughly, the evidential, epistemic probability of
a proposition p on your total evidence E is a measure of how much
“believability” E confers upon p. This is not an analysis, but merely a
heuristic for understanding what I am saying. Because I am a staunch
fallibilist (see Dougherty 2011b), I cannot accept any epistemic prob-
ability of either 1 or 0, thus I cannot accept a mathematical account of
epistemic probability. Another reason is that I think that everything that
an agent considers a metaphysical possibility should be given by that
agent a probability greater than zero (infinitesimal probabilities are only
a red herring here), which is not the case for mathematical probability
(think of the probability of a real-valued random variable). Commonly,
people give counterfactuals to explain epistemic probability, something
like “The epistemic probability of p given E is how much a rational or
properly functioning or otherwise ideal agent would give p if all the
information they had was E.” There are well-known problems with this
as an analysis, but, again, it is a helpful heuristic, and I have nothing
more helpful (and short) to say.
Now I’ll discuss two notions about evidence and its influence.9 We
will assume the positive relevance theory of evidence. Some people think
there are fancy counterexamples, but they aren’t relevant to us even if
they are exceptions, which I doubt. The idea is that some data D count
as evidence for a hypothesis H when acquiring D makes H more prob-
able than it was before. “E<X,Y>” says that X bears the is evidence for
relation, E, to Y (without saying to what degree).
Positive Relevance E<D,H> = df Pr(H | D) > Pr(H)
When we have evidence in this sense, we have incremental or relative

confirmation. That is, D has given H some boost over where it started,
7
One of Eleonore Stump’s great and recent achievements is to draw proper
attention to second-personal knowledge.
8
Two great introductions to interpretations of probability are Galavotti (2005)
and Gillies (2000).
9
For more, see Dougherty (2010).
but this is consistent with this boost being very little. So, for example, if
there are one hundred numbered doors in front of me and I don’t know
which one I’m supposed to go through, the odds that it is a randomly
chosen door are 1/100. But if I find out that the door I need is labeled by
a non-prime, then I get some evidence that it is door #12. It was 1/100
and now it is 1/75, so the new information gave the hypothesis that it is
door #12 a boost relative to its starting point, some increment of confir-
mation. However, we do not have absolute confirmation, that is, the
hypothesis has not been shown to be probably true. On the contrary, it
remains very probably false.10 How to measure the degree of confirma-
tion is a very difficult subject, and there are a number of sophisticated
treatments. However, I will be making no attempt to precisely quantify
the degree to which I think my arguments provide confirmation for my
theory. I simply want to move the reader as far as I can in the belief that
any significant change makes a difference. In this book, I hope to show
that my story is probably true, but I will be happy if what I say gives
it a non-trivial boost for the reader. In my view, the immediate goal of
philosophy is to move people’s probability in the right direction. As
with everywhere else, success in philosophy is incremental.11
1.3 Not-so-bare bare theism
Since this book is addressed not just to theists struggling with the
intellectual12 problem of evil but also to atheists and agnostics open-
mindedly interested in what strategies theists can fruitfully pursue to
mitigate the problem of animal suffering, I need to say something about
a natural objection by those latter parties. A natural thought is “I’m not
a Christian, so your proposal can’t get an epistemic boost from the fact
that it is more consonant with the Christian tradition than I previously
thought it was.” A similar issue arises due to differences in overlap of
Christian belief. The coherence of my story with Christian tradition can
affect atheists’ and agnostics’ credences indirectly by blocking some or
all of the disconfirmatory power of the data of evil, as we shall pres-
ently see. However, I am not merely adding some other theses – taken
10
See Dougherty (2010) for more discussion.
11
Note how different this perspective is than that of van Inwagen (2006), where
success is convincing a panel of ideal agnostics. It is also markedly different from
Murray, where success is something like standing one’s ground.
12
The existential problem is better treated by those with pastoral gifts, which
I wholly lack, though I have as a goal to produce a theodicy that has definite
pastoral applications.
from Eastern Christianity – to Western Christianity or a sparse or bare

philosophical theism. Rather, inspired by the teachings of Eastern
Christianity, I show that a bare philosophical theism logically entails
those doctrines. So there is no real addition at all, but rather a teasing
out of consequences. What we learn, I claim, is that bare theism isn’t so
bare after all. There is another way to bridge bare theism and Christian
theism, at least what we might call “mere” Christian theism after C.S.
Lewis’s idea of “mere” Christianity, though what I have in mind is a
little weaker. Assuming fairly uncontroversial historical facts about the
life of Jesus in our background information, then I think that Pr(Mere
Christian Theism | Bare Theism) = approx. 1. That is, I think that if
bare theism is in one’s background evidence, then if one appreciates the
historical evidence for Jesus’s divinity, they should believe the core of
the Christian Gospel.13
The issue just discussed will somewhat complicate exposition, for
there are at least three different “movements” ranging from purely
“defensive” to “offensive.” In the most “defensive” movement, I am
treating the very legitimate question “What should the committed
Christian say when confronted with the problem of animal pain?” (I
exclude consideration of natural theology, though I am very pro natural
theology.) Because Christian commitment differs quite a bit it will be
generally useful to consider the very reasonable question “What should
the ‘mere Christian’ say when confronted with the problem of animal
pain?” Many readers will be doctrinally committed Christians and many
more will share the core content of mere Christianity. But then there
will be readers who don’t share any Christian beliefs. Addressing them
involves the same issues as considering the previous two questions. For
when the atheist, agnostic, or deist asks the question “How do the data
of animal suffering bear on Christianity?” the question will be meaning-
less without a determinate notion of which hypothesis “Christianity”
names in that question. Therefore, much of what I say will involve
arguing for a particular view about which hypotheses constitute the
reasonable target. And it is very important for a probabilistic approach
that the core theory is not weighted down with auxiliary hypotheses.
So my key move is to show that what are often considered auxiliaries
are either not auxiliaries at all or are so probable on theism that they
are “nearly entailed” and so add almost no additional baggage to bare
theism or mere Christian theism.
13
Swinburne (2003) essentially defends this thesis.
Here is an analogy that might help. Anselm defined God as “that than
which no greater can be conceived.” In short, the absolutely perfect
being. God has what Plantinga calls “Maximal Greatness.” Making this
a central notion of philosophical theology has come to be known as
“perfect being theology” (see Morris 1987, 1991; Rogers 2000). It is
not perfectly transparent what all will be included in the property of
being perfect. But once one comes to a reasonable judgment about what
perfection entails, then for many great-making features, once one real-
izes it is a great-making feature one will have thereby seen that God has
that property as well. And when this happens it is not that one has added
to one’s concept of God, one has merely discovered what one was already
implicitly committed to. So, again, to be clear, I am not seeking to add
the notion of animal resurrection and deification to bare theism, I aim
to show that it is implicitly included in theism (conditionally). That is,
every possible world in which God exists and animals suffer is a world
in which they are resurrected and deified by God.
Here is one way that these ideas get applied. If we let E represent the
data of animal suffering and T represent theism, it appears that Pr(T | E) <
Pr(T) which makes E disconfirmatory of T. Now let S represent my story
about animal resurrection and deification. If Pr(T | E & S) = Pr(T), we say
that S has “screened off” E’s disconfirmatory power with respect to T just
mentioned. In the language of defeaters, E was a defeater for T, but S is a
defeater-defeater, and T goes back to its original probability. And it could
be, of course, that S doesn’t screen off all of E’s disconfirmatory power,
but rather only a portion of it. However, as will become clear, theism
needs more than a defeater-defeater, and the greater work of S will be
something more striking, to show that Pr(E | T & S) is actually surpris-
ingly high, higher, in fact that Pr(E | ~T) by a large margin. That is, not
only does E not disconfirm T, it positively confirms it, relative to naturalism
(the relevant alternative to T14).
Of course, unless T entails S, Pr(T & S) will be at least somewhat less
than Pr(T) and so even if S “saves” T from E, S might “sink” T with its
own weight. So the question of the value of Pr(S | T) is a very important
one. And here there is revealed an interesting way in which the data of
evil can be used to plumb the content of theism.
So suppose that we are certain about E. A very natural way for the
atheist to argue against T on the basis of E will be to point to some
14
I take Deism to be broadly logically impossible: The attributes in bare theism
entail that if there are beings of worth in the universe, God finds worth in them
and responds appropriately.
alleged consequence of E – which itself merely states the empirical facts

about animal suffering, and is, therefore, not prima facie incompatible
with T – X, such that X is incompatible with T. In the case of the argu-
ment from evil, the “X factor” is typically unjustified suffering. The facts
described in E are prima facie cases of unjustified evils. And unjustified
evils are logically incompatible with T. However, note that one can’t just
go directly from a description of the suffering to its being unjustified.
There must be something about it that is unjustified, something that
makes it unjustified. So if we can state the condition, Y, which is what
it takes to get from E to X, then we shall have a potential formula for a
problem, getting us from X to ~T. The “Y factor” in this case seems to be
something like what is contained in the following proposition.
Y Animals receive no compensation for their suffering; their evil is

not defeated.15
But even the lemma Y does not get from E to unjustified animal suffering
(X) all by itself. There are, in fact, other assumptions, part of a series
of “Z factors” that are very basic assumptions required to get from the
facts about evil to the conclusion that there exists unjustified animal
suffering. Here are some plausible examples.
Z1 The goods that allow permission of an animal’s suffering must

come to the individual animals who undergo the suffering.
Z2 Animals do not receive compensating goods in this life.
Z3 A Creator has moral duties toward his creation.
I will not spend much time addressing such Z-factors. Such basic assump-
tions are safe with me. In this book, I am happy to assume that all the
necessary Z factors are true, though I think each are surprisingly non-
obvious upon reflection. Thus I will often bracket them in discussion.
Doing so allows me to say that reflection on the data of animal suffering
described in E allows us to discover this.
(*1) (E & Y) → ~T
But this is logically equivalent to
(*2) E → (T → ~Y)
15
The notion of the defeat of evil will be treated below.
And if T → ~Y, then T is logically equivalent to T & ~Y, so we have
(*3) E → (T ≡ (T & ~Y))
In this way, learning E can teach us about what it really means to believe
T. Given E, we must embrace T & ~Y, so the believer in T needs to tell a
story S that entails both T and ~Y. The more probable the story is, the
better. My main endeavor, then, is to tell a story S, such that (a) S entails
T & ~Y, (b) we do not have direct evidence against S, and (c) Pr(T & S) is
not much lower than Pr(T). One important part of this is to show that,
modulo E, T entails S. The strategy will be to show this by showing that
S is the only possible way that ~Y. Assigning my theses as the values of the
variables, my project crucially involves arguing that the only possible
way animal suffering is not unjustified is if they are resurrected and
deified.
Here’s another way to think about the way in which the facts about
animal suffering can teach us about theism. There are two ways for T to
be true, for T is logically equivalent to
(T & S) v (T & ~S).
E can rule out (T & ~S) and teach us that believing T rationally commits
one to believing (T & S). Thus, if (T & S) is not implausible – that is,
it’s not the case that the two conjunctions “don’t get along,” and S is
not intrinsically improbable, and S is not improbable on our evidence,
and so on – T will have been discovered to have really been (T & S)
all along. E is no objection to (T & S), given the story I will tell. In
short, E does not even partially disconfirm (T & S) and therefore does
not even partly disconfirm T (since (T & ~S) is logically impossible, it has
no measure in the model16). Indeed, as I hope to show, (T & S) makes E
likely, and therefore, with almost trivial assumptions, E confirms (T & S).
This is necessary to show because a Bayesian version of the problem of
evil – such as the neo-Humean one Draper (see Draper and Dougherty
2013) offers – can gain some epistemological purchase from E even if E
does not directly disconfirm T. For, letting N stand for naturalism, these
comparative results are consistent.
(CR1) Pr(E | T & S) > P(E)
16
This is consistent with my fallibilist scruples: this is a mathematical model,
not a case of confirmation by evidence.
(CR2) Pr(E | N) > Pr(E | T & S)
And so even though E doesn’t directly disconfirm T in this hypothetical

case, it still favors some alternative to T, allowing the scales of evidence
to point to N rather than T. I will argue that (CR1) is true and (CR2) is
false.
1.4 Two ways of learning what theism entails
In some cases learning that one way a proposition can be true is closed
off decreases the probability of that proposition. This is very common in
games of chance and other purely mathematical contexts. For example,
if you have a lottery ticket that wins if either 923846 or 453789 are
drawn, and you find out that it is not the former number, your chances
of winning are cut in half. In confirmation theory, disjunctive theo-
ries are commonly used to illustrate problems in updating. Paul Draper
(2014) gives a good example of when learning that one way of p being
true has been eliminated reduces the probability of p.
Suppose three balls are randomly drawn from one of four urns. Perhaps
a card is randomly drawn from a standard 52-card deck and the suit
of that card determines from which of the four urns the ball is drawn.
According to theory-12, the balls all came from either the first urn or
the second urn. According to theory-34, the balls all came from either
the third urn or the fourth urn. Now suppose that the first, third, and
fourth urns each contain 900 yellow balls and 100 purple balls, while
the second urn contains 3 yellow balls and 997 purple balls. Further,
it is observed that all three of the balls drawn are yellow. Notice
that there are two versions of each of the two theories. According
to first version of theory-12, all of the balls came from the first urn.
According to the second version, they all came from the second urn.
Similarly, theory-34 can also be divided into two versions. Obviously,
the fact that the first ball drawn is yellow “favors” theory-34 over
theory-12 (in the sense of raising the ratio of the probability of the
one theory to the probability of the other) because that ball is almost
twice as likely to be yellow if theory-34 is true than if theory-12 is
true. Of course, once a defender of theory-12 sees that first yellow
ball, he will have good reason to believe that the first version of his
theory is more probable than the second version and so, if he clings
to his theory, he will base his predictions about the color of the next
two balls on that version of his theory. Of course, when the next two
balls turn out, as predicted, to be yellow, the defender of theory-12

should not conclude that the first ball’s being yellow did not favor
theory-34 over theory-12 ... . It would be no less mistaken, however,
for a defender of theory-12 to think she has no evidential problem on
the grounds that she now knows that the first version of her theory
is the most likely to be true and that version can predict the color
of all three balls, including the first, just as well as theory-34. (2014:
175–176)
In this context, Draper is absolutely right. Finding out that one way
for T12 to be true has been effectively eliminated makes T12 less prob-
able relative to T34. However, it is fairly well known that not all contexts
are like this. Buchak (2014) offers an example (with relevant references)
like the following, which is an adaptation.
Consider two hypotheses. IN says that Trent is in town, and OUT says
that Trent is out of town. We’ll assume that IN and OUT are equiprob-
able for you. You know that if I’m in town I’ll be at a bike shop. There
are two bike shops in town: Townie Bikes, a locally owned and oper-
ated bike shop, and Big Bart’s Bike Barn, a large chain store. As far as
you know, I’m equally likely to be at the one as the other.
At this point, your partition looks like this:
IN & at Townie: .25

OUT: .5
IN & at Bart’s: .25
But then you learn I’m a total hipster snob and I’d never go to the box
store. That eliminates the lower right cell. Standard updating is going
to result in the OUT column being larger than the IN column.17 One
of the ways of IN being true has been effectively eliminated. So if we
were to treat this case like the balls-in-urns kind of case, we should take
ourselves to have obtained a reason to think that IN is false, we should
decrease our credence in IN. But of course that would be crazy. This is
fairly obvious and generally agreed upon among those who have written
on the topic (see Buchak for references). What is not agreed upon is
a general diagnosis of what separates such cases from the mundane
17
One standard way of updating would give 2/3 of IN & at Bart’s “credence
mass” to OUT and 1/3 to IN & at Townie, since there is a 2:1 ratio of credence
mass between the two.
mathematical cases. I think that there are two kinds of situations that
are sufficient to preserve the probability of the general hypothesis when
a species has been eliminated.
One such situation is when the reasons for believing the hypothesis
pertain to the generic features of the hypothesis independently from the
species. For example, suppose, as is quite plausible, that the evidence for
common descent of humans and chimps (that they share a common
ancestor) is consistent with both Darwinian gradualism and punctua-
tionism. Now suppose biologists discovered it was simply impossible for
the mutations postulated by punctuationism to occur, so that punctu-
ationism was effectively ruled out. This should not decrease our confi-
dence in common descent at all. Another example: Suppose Liz becomes
convinced from historical studies that Christianity is true. Suppose
further that the only three ways to be a Christian are to be Catholic,
Eastern Orthodox, or Protestant. In the course of investigating each of
these species of Christianity, she learns that sola scriptura is incoherent,
since Scripture affirms authority outside Scripture, effectively eliminating
Protestantism. In no way should Liz’s credence in Christianity go down.
These cases suggest that when one’s support for a theory pertains directly
to its general features, eliminating a species will not decrease the prob-
ability of the main theory. Note that in Draper’s case there is no genus
theory. The theories in question are mere disjunctions, not species of some
natural kind.
Another kind of situation where it seems the genus does not lose prob-
ability from losing a species is when one learns (or takes oneself to learn)
that one of the species is impossible. So suppose Dorothy (mistakenly,
in my view) thinks she has refuted dualism and so come to know that
materialism is true. She then reflects on the fact that there are reductive
and non-reductive materialisms. She initially assigns equal probability
to these two species. Then she comes to learn (as I suspect) that non-
reductive physicalism is conceptually incoherent and so metaphysically
impossible. Should Dorothy reduce her credence in materialism? It seems
to me that she should not. In this case, not only do the previous consid-
erations apply, but because the hypothesis has been discovered to have
been impossible all along, what is happening is essentially are reparti-
tioning of the outcome space: You are going back to the beginning and
taking out one of the possible outcomes in the list because, as you now
know, it never should have been there in the first place. In the literature
on probabilistic updating, how to update on conditionals and how to
update on information about the outcome space are matters where there
is currently no consensus. What there is general consensus on, however,

is that there are cases where standard updating is inappropriate.18
I argue in this book that what the data of evil do is to rule out various
species of theism in ways that are more analogous to the bike shop,
evolution, and Christianity cases than to any game of chance case. I
think the data of evil carve out a form of theism which, once revealed,
is not only consistent with the data, but which is consonant with them.
I will spell out this application more in later chapters, but it is crucial
to understand what I am trying to do up front, since I will apply these
ideas several times throughout the book. Now it is high time to get on
with the project.
18
This is true both for the reasons listed above but also for many other reasons
including problems with commutativity (see Weisberg 2009 and McGrew
2014) regularity (see Easwaren 2014), and rigidity (Leitgeb and Pettigrew 2010,
not to mention anyone who has ever had a paradigm shift in what makes what
probable).
2
The work of this chapter is to state the problem under consideration with
some precision (§2.1), and to gain some reasonable clarity in the terms
at play (§2.2). I will also address the issue of just how much suffering I
take there to be (§2.3).
2.1 Stating the problem
The first half of this chapter treats the nature and structure of the problem
along with some of the main conceptual components involved in that
structure. Two crucial notions arise as a core component of framing
the problem: pain and suffering. Discussion of these ideas in the litera-
ture is particularly vexing. The second half therefore focuses on these
central notions. The challenge of this chapter is to adequately balance
presenting the “big picture” with providing a sufficient level of detail to
make the problem clear. This will not be easy to do. My method, befit-
ting a Texan, will be a sort of “two-step” where I focus in, then pan out,
then focus in again. We begin, of course, with the broadest question.
2.1.1 What is the problem?

What is the problem of animal pain? And for whom is it a problem? It is
a problem for theists. At least, it is a problem for those Christian1 theists
1
I will typically write from the perspective of a Christian theist, since that is the
tradition I know best, and the one most under attack in the previous literature.
However, much of what I say is likely to apply to Judaism or Islam. I agree with
Adams (1999: 3) that there are advantages to be had by working within a specific
tradition, though, as will become clear later, my view of the relationship between
“bare” theism and the theism of Christianity differs importantly from hers.
16
The Problem of Animal Pain 17
who hold that God is a part of the moral order. (That God is a part of the
moral order is denied both by radical Calvinists and radical Catholics.
The former is common; for the latter see Davies 2006.2) Henceforth, by
“theists” I shall refer to only those theists who take God to be in the
moral order. Animal pain – more particularly, its profusion3 and inten-
sity – is a problem for these theists because, prima facie at least, it is not
at all what we would expect from an all-powerful, all-loving God. Yet it
seems, prima facie at least, to be just what we’d expect (or at least not
too surprising), given naturalism. Thus, until other data and explana-
tory virtues – especially simplicity – are taken into account, it looks like
naturalism is a much better explanation of the data of animal suffering
than is theism. That is surely a problem for theists.
In other words, the problem of animal pain for theists consists in a
certain kind of epistemic threat. For when one theory explains some
data, that is, some evidence, better than a rival, then, ceteris paribus, the
evidence favors that better explanation. As I hinted at just above, it may
well be that there are many other data that favor theism and so tip the
scales of evidence on the whole in favor of theism. And it may be that
theism is so much simpler than naturalism that even an explanatory
benefit for naturalism will not, in the end, tip the scales of evidence
in favor of naturalism.4 Though it will be necessary to treat these other
issues in cursory fashion below in order to see the broader picture into
which the work of this book fits, full treatments of them remain subjects
for other books. This book is only aimed at assessing the evidential force
the data of animal suffering have on theism.
Now, according to one kind of view, sometimes called fideism (from
Latin fide, “faith”), being evidentially challenged is not necessarily a
problem for religious belief. On a strong fideism, religious beliefs are
2
If I understand such views, I find them monstrous and something like blas-
phemous (libelous?) accounts of God. I hope I don’t understand them correctly.
3
I use the term “profusion” quite intentionally. It is an allusion to Hume (1980:
ch. XI). The term is picked up in Rowe’s original, influential formulation of the
problem (Rowe 1996 (orig. 1979): 2, 5, 10). Howard-Snyder (1996b: 289) points
to “so much horrific evil rather than a lot less” and Langtry (2008: 162) to “a lot
of suffering or dysfunction.” I agree with what all these accounts of the starting
point of the problem of evil point to: that the quantity of evil in the world matters
to how strong the argument from evil is. There are also issues of its intensity and
particularly cruel nature in what have been called “horrendous” evils (e.g. Adams
1999). I will not always use that term as specifically as Adams does.
4
Swinburne (2004: ch. 5) gives an argument for the quite striking simplicity of
(“bare”) theism.
not to be evaluated for evidential fit at all. Rather, religious beliefs are
simply exempted from the normal process of evidential evaluation. This
view has few, if any, sophisticated defenders today.5 (There is a view
with the same consequence sometimes called “antirealism” about reli-
gious beliefs, which is a form of noncognitivism about religious beliefs
or discourse. A toy theory that illustrates the basic idea is the thesis that
“God” refers to something like the best intentions of mankind or some
other abstraction of the human mind like that. As far as I can tell, this
kind of antirealism seems to be fairly popular in religious studies depart-
ments and many theology departments as well, but it is almost nonex-
istent among philosophers, thankfully.)
However, more modest fideisms exist. For example, Bishop (2007a)
requires that evidence be essentially counterbalanced and that for religious
belief to be appropriate (in the absence of compelling evidence) that there
be strong moral reasons for it. I will not argue for it, but I will assume a
form of evidentialism in this book. A precise formulation is not necessary
(see Dougherty 2011a: 6ff), but I will take evidentialism to entail that it is
always epistemically unjustified to hold a belief that has a lower epistemic
probability for one than an alternative. So I will take it that the evidential
threat posed by the apparent explanatory advantage of naturalism over
theism requires that the theist address the issue squarely.6
I will attempt to address that problem by giving a theodicy (broadly
construed). That is, I will be offering a plausible story the theist can tell
about why the facts are as we find them. This story will appeal to more
than bare theism. It will add to bare theism some other theses. It will
thus, to some extent, reduce the simplicity of the theism defended here.
I will be arguing two things about these additions, however. First, I will
be arguing that some of these additions are not, in fact, additions at all,
but somewhat hidden consequences of theism, teased out as such by our
line of inquiry. I take it to be a common phenomenon in philosophy
that one only comes later to realize some of the logical entailments of a
thesis. Second, I will argue that the auxiliary hypotheses included in the
story do not significantly reduce the probability of theism. To represent
that argument, I need to say a few things about how I will understand
simplicity and expectation.
5
It is not always easy to tell how radical the fideisms of Phillips (1993) and
Malcolm (1993) are.
6
I shan’t have room to say much about epistemic probability in this book,
but more on it shortly. For the most part, I am happy to officially hew closely to
Swinburne (2001).
The comparative simplicity of theism and naturalism is a conten-

tious matter. For that matter, the nature and relevance of simplicity
considerations to the epistemic status of a theory is itself a contentious
matter.7 (And I shall assume that both theism and naturalism are theo-
ries in a relevant sense of “theory”: a narrative that attempts to tell a
coherent and compelling story about the relevant data in such a way
as to portray its probable truth.) One mark of theoretical simplicity is
a sparse ontology: postulating the least number of (and kinds of) basic
entities sufficient to cover the data. That this is a good-making feature
of a theory is enshrined in such norms as the Principle of Parsimony: do
not explain with many entities that which can be explained with a few.
Before this principle picked up the name “Ockham’s razor,” Aquinas
used it – along with the problem of evil – as one of his two objections to
the existence of God (Summa Theologica I.2.iii). Though it is controver-
sial to do so, I will endorse this long-standing norm.
2.1.2 Getting more precise

Now let us take a first stab at making the problem more precise. I want to
add a few caveats about belief, probability, and knowledge. Also, I want
to say a bit more about some aspects of the datum of suffering from
which I will begin. Here is a succinct version of the problem.
Given (the programmatic assumption) that theism and naturalism

are roughly on par with respect to simplicity, since the profusion and
intensity of animal suffering is more to be expected on naturalism
than theism, naturalism will be the better explanation of all our data,
unless some other data support theism more than the data of evil
detract from it, or it is discovered that the data of evil are either not
as expected on naturalism as originally thought or not so unexpected
on theism (or all of the above).8
Here is a schematic of the problem.
Assumption: The simplicity of bare theism and naturalism is roughly

on par.
Appearance: The data of suffering are considerably less surprising on
naturalism than theism.
7
See Swinburne (1997b) for a defense.
8
Draper (1989) defends an argument which may be equivalent. In Draper and
Dougherty (2013, and forthcoming) I present the argument this way.
Options: (i) Other data (for example, arguments from natural theology)
support theism over naturalism at a greater ratio than the data of evil
favour naturalism over theism; (ii) The data are more surprising on natu-
ralism that at first appeared; (iii) The data of evil are less surprising on
theism than at first appearance; (iv) any combination of (i)-(iii).
Even though my statement of the problem leaves open several possible

routes out for the theist, it is still a problem. A theist may well think that
even if the facts about animal suffering disconfirm theism there are other
facts that stack up more strongly in favor of theism. However, there are
reasons why even in the presence of such facts the theist may wish to argue
against the disconfirmatory power of the facts about animal suffering.
First, many religious theists (and of course theism is neither necessary nor
sufficient for religious belief) hold that it is important to believe that theism
is true (but see Pojman 1986, Dougherty and Poston 2008 and Audi 2008
for nuances). Stating precisely what the connection is between probability
and belief is a difficult matter, but it is not necessary to know the details to
realize that belief entails considerable confidence in a proposition, so that
there is something incoherent about believing a proposition one judges to
be improbable (though, see Kaplan 1996: ch. 4, sec. IX). It is only a bit less
obvious that there is tension in believing a proposition and judging it to
have only moderately high probability on one’s evidence. Still, it is also not
obvious that one should outright believe p when p is only moderately prob-
able. (See Swinburne 2001: 34–38, esp. n. 8, and Plantinga 1993: 168 for
careful discussions – coming from two different angles – of the relationship
between believing p and believing p to be moderately probable.) But most
Christians don’t want to merely believe that Christianity is true, they want
to confidently believe it, and even claim to know it.
So if one’s plan is to counterbalance disconfirmatory facts with
confirmatory facts, one would need quite a margin in favor of confir-
mation in order to have a final probability sufficient for robust belief or
knowledge. Just as an example, suppose that it starts to become vague
whether a probability is high enough to support rational belief as values
slip below around .8 (note the higher-order vagueness). Expressed as
a ratio, .8 is 8:2 or 4:1. Thus, the pro-reasons would have to have four
times the weight of the con-reasons to allow for a final probability of .8.
And some think belief requires a higher probability than that. Even those
quite sanguine about theistic arguments may not think they satisfy that
ratio (for the record, I do). Thus, it is very much in the theist’s interest
to seek to minimize the amount of counterbalancing that needs to be
done, and so theists should welcome a perspective on animal suffering –
if it is plausible and consonant with theism – that shows that the facts
about animal suffering do not count against theism. (For more on this,
see Dougherty and Walls 2013.)
Second, if auxiliary hypotheses can allow theism to predict, at least to a
modest degree, the sort of situation we find in nature and to do so at least
as well as alternatives to theism, then theism can even pick up a moderate
boost in probability, which may help if there are other data that turn out to
disconfirm theism. Are there such data? Well, even though Aquinas tried
to list at least three objections to any thesis he defended in the Summa
Theologica, on the existence of God he only raised two objections. One
was the problem of evil (a version of the so-called “logical” problem of
evil: that the mere existence of evil is incompatible with the existence of
an all-good and infinite God), and the other was, essentially, the idea that
God was an “unnecessary hypothesis” (to use Laplace’s famous phrase)
because natural science could explain everything in need of explanation.
But if that were true it would not be a fact that disconfirmed theism, it
would merely show that theism did not have empirical evidence on its
behalf.9 But what of the data about human suffering, isn’t it a separate
matter from the suffering of non-human animals? Yes, however, if the
explanation for animal suffering I will propose works, it will also work
for humans. Thus, if I am successful in this endeavor, there should be no
epistemic remainder to the problem of evil. That is, there should be no
residual disconfirmation to theism from the problem of evil.
Two notes on the datum: First, the datum is not merely the existence
of some animal suffering. The datum concerns intense animal suffering,
and it concerns the profusion of it. That is, of the profusion of intense
suffering. Profusion of trivial suffering would not be strong evidence
against theism. And perhaps a few isolated cases of intense suffering
also would not be strong evidence against theism. A case could be made
that they did, but I will focus on the clear case of the profusion of very
serious suffering. The profusion of intense suffering constitutes strong
prima facie evidence to favor naturalism over theism. On the surface,
that there is widespread intense suffering fits much better with natu-
ralism than theism.
Also, the datum from which I start is not pointless suffering or the
appearance of pointless suffering as in the work of William Rowe.10
9
Actually, that is true only on certain ways of construing the alternatives to
theism, but that is another matter.
10
Rowe (1979, 1996, and 2001) provide a sampling. See Trakakis (2007) for a
detailed study of Rowe’s arguments.
Pointless suffering is logically incompatible with God’s existence.11

Wouldn’t asserting the existence of pointless evil prove a stronger
starting point? Not necessarily, for to show that an instance of
suffering is in fact pointless is not a trivial matter. The history of
moving from “apparently pointless” to “actually pointless” is fraught
with difficulties. This so-called “noseeum” inference has been subject
to many serious objections, some of which are devastating (see
Howard-Snyder 1996a, McBrayer 2010, and Dougherty 2011c and
2013). A better strategy is to adopt a form of common sense episte-
mology, as Swinburne does, and then the appearance of pointlessness
(which is more than the non-appearance of a point) automatically
provides defeasible immediate (non-inferential) justification (or, as I
prefer, a pro tanto reason) for the proposition that there is pointless
suffering (see Swinburne 1998: 20ff, as well as Dougherty 2008, 2011c,
2014, section 6.2, and Dougherty and Walls 2013, 372). Though much
better than the inferential strategy, this strategy also faces challenges
it would be an advantage to avoid.
Though the so-called “skeptical theist” approaches to the problem
of evil are (in their present form) side-stepped by the common sense
approach to the problem of evil, they can be re-tooled to serve as
defeaters for the defeasible justification acquired immediately in accord-
ance with common sense epistemic principles.12 Most of the same
objections apply to the skeptical theist move, and one goes to the core.
Skeptical theists advocate a form of agnosticism about God’s reasons
based on an analogy with a parent13: A child cannot be expected to
understand the (often good) reasons for which her parents allow her
to experience preventable pain. But the epistemic difference between
a human child and a human parent is dwarfed by the epistemic differ-
ence between humans and God. The analogy is an intuitive one, but
the skeptical theist use of it neglects the fact that a loving parent wishes
they could explain, and an omnipotent God would have the ability to
11
One might think that Peter van Inwagen (1996a, 1996b, and 2006) has shown
that, due to vagueness concerns, there could be morally justifiable pointless evils.
For example, for any (or nearly any) instance of suffering in the world, it could
have been very slightly less without making the world significantly worse off.
But this is only to say that the residual had no particular point. It will still have a
global point: being a consequence of a natural law for which there is a point or of
a chance process for which there is a point.
12
For a detailed treatment of skeptical theism, see Dougherty (2014).
13
See Wykstra (1984) for a very early use of this analogy and Bergmann (2001
and 2009) for a more recent use of this strategy.
either augment our cognitive capacities or make sure that the majority
of suffering he permits is for reasons we understand14 (or, alternatively,
he could be clearly present to us during such times when he has suffi-
cient reason to keep us in the dark about the reasons for which he
permits the suffering15).
An even more devastating problem for skeptical theism is that skepti-
cism (or agnosticism) about what to expect from God does not make the
version of the problem I have focused on go away. The problem as I have
formulated it is one of favoring. The data, I claim, seem, at least at first,
to favor naturalism over theism. And if theory A says nothing about
datum D but theory B predicts it, then D gives B an advantage over A. So
even if we cannot sensibly assign any probability at all to how likely it is
that we would find what we find given theism, if it is predicted by natu-
ralism, then the evidence favors naturalism. Think of it like a game. You
try to predict some event. If you get it right, you get a point, if you get
it wrong you lose a point. Since there is a penalty for getting it wrong,
you are allowed to pass. So, according to skeptical theism, theists pass
when it comes to predicting the data. So theism doesn’t lose a point,
on the assumption that skeptical theism is true. That is, theism is not
in the least directly disconfirmed by the evidence. But that does not
mean that theism isn’t indirectly disconfirmed. For the naturalist does
not pass. The naturalist predicts a world indifferent to animal suffering.16
Since rational belief encompasses alternatives, even giving the skeptical
theist everything she wants, the evidence still favors naturalism. Or so
it seems.
At any rate, I am not skeptical of our ability to form reasonable expec-
tations concerning theism. So I shall be arguing for a certain theodicy
that I think is at least not unexpected on theism and, perhaps, quite
expected.
14
See Dougherty (2012) for an extended consideration of the parent analogy.
15
And thus the problem of evil leads directly to the problem of divine hidden-
ness. See Howard-Snyder and Moser (2001) and Dougherty and Parker (2013). See
Dougherty (2012) for an argument against the parent analogy.
16
But doesn’t it fail to predict a world at all or a world complex enough to have
suffering animals or a world where there is any consciousness at all or a world
where there are human-like beings who care? Well, no, I think, and though I think
natural theology is quite successful, I am here concerned only with the question
of how successful natural atheology is. Many theists have argued against the argu-
ment from evil, but none have defended theism well against the neo-Humean
argument (van Inwagen 1996c comes close, but gets no cigar). Bergmann (2009)
touches on it briefly. See Draper and Dougherty (2013) for a direct engagement.
2.1.3 Theism as “hypothesis”

In following this method, I am obviously treating theism as a hypothesis
in a way relevantly similar to a scientific hypothesis. I will not defend
this assumption in much detail, but it is unavoidable in this context.17
This problem concerning forming expectations about God’s choices and
treating theism like a scientific hypothesis raises deep issues in the epis-
temology of religion. It is worth considering at some length a sketch of
a way of thinking about the matter that is sensitive to the issues at hand.
This way of thinking bridges a rightful respect for the fact that God is a
personal being of infinite wisdom with a humane understanding of how
science – indeed rational investigation generally – proceeds. Recall how
I used the term “theory” above.
Theory = df a narrative which attempts to tell a coherent and compel-

ling story about the relevant data in such a way as to portray its prob-
able truth.
Notice that this kind of theorizing is common to history, literature,

crime-solving, and medical diagnosis. I don’t think there is anything
exclusively “scientific” about it. Rather, I think having theories of things
is nothing other than a rational creature’s attempt to make cognitive
contact with the world. There may be forms of non-cognitive contact
with God, and they may well be more important than any kind of cogni-
tive contact with him. There is nothing in the least bit inconsistent with
holding this thesis and believing that God’s relationship to the world
is at times intelligible. When you consider the case for, say, common
descent of all living organisms, there is no single line of argument that is
at all compelling. This is part of the reason why in debates with young-
earth creationists, scientists don’t come off looking very good and why
creationist books and websites came to be so compelling. It is not that,
as some would have it, creationists are (to put it ironically enough)
knuckle-dragging Neanderthals who can’t read except for their Bible.
On the contrary, creationist literature has been, for the most part, fairly
sophisticated (and only occasionally sophistical) from the start. The case
for common descent is subtle; it weaves many, many strands of inquiry
together into a coherent whole.18 Managing the data is no mean feat.
It is somewhat like reading The Lord of the Rings trilogy (or maybe even
17
A full issue of Topoi (1995) is dedicated to the subject of whether or not
theism is a theory in the relevant sense.
18
See Sober (2008) and Kitcher (1989) for information pursuant to this point.
The Silmarillion). There is a lot of action, many characters of unfamiliar

kinds, and separate but interweaving plot lines. Assessing the evidence
for a complex theory is much like reading an intricate story. In the next
section I will expand on this notion a bit, for I wish to forestall as much
as possible some objections I often run into when presenting material
like this, which requires assigning probabilities to divine acts.
I have just briefly clarified the central notions of the problem of evil
as I am treating it. It is a problem about belief and probability, in a
broadly Bayesian explanationist framework, based on the profusion of
intense animal suffering, which can’t be addressed by skeptical theism.
I will next treat each of these core notions in significantly more detail.
The rest of this chapter provides more detail on the notions of pain and
suffering. The probabilistic parts will be presented in more detail in the
next chapter.
2.2 How I shall understand “pain” and “suffering”
Let us look again at the datum.
E Since near the beginning of sentient life on Earth, there has been a
profusion of intense animal suffering. That is, wherever there has been
sentient life very significant levels of suffering have been quite common,
and there has been a lot of sentient life.
The careful reader will have noticed that I have been wavering between
the terms “pain” and “suffering.” In many contexts this is innocent
enough. However, it will become very important that we are clear about
these notions for what is to follow. Therefore, I shall now explicate the
notions of pain, suffering, and the relationship between them.
The International Association for the Study of Pain defines human
pain (though there is nothing specifically human about it, as I will
argue) as “an unpleasant sensory and emotional experience associated
with actual or potential tissue damage, or described in terms of such
damage.”19 This is too narrow a definition of pain as such, obviously,
because some pain has no physical cause at all, even if it has physical
realizers, pathways, etc. One thing that is important about this defini-
tion, which I shall accept as generally correct and quite important, is
that it defines pain in emotional rather than cognitive terms (I will not
spend time considering the meaning of “sensory” in this context, but I
19
International Association for the Study of Pain (2012).
flag it because the term is used quite differently in the pain literature).
This will be important for Chapters 3 and 4 because a prominent objec-
tion to the notion that animals are the subjects of pain is that to have
pain, an organism must have a certain kind of consciousness and that
they have that only if they have certain cognitive capabilities. The above
definition of pain does not seem to be encumbered with such views.
Its primarily affective character has the potential to circumvent the
described objection completely. The details of this dialectic will become
clear in Chapters 3 and 4, but for now I will simply introduce the bare-
bones idea behind the objection to make clear why this notion of pain
is important.
There is a view called neo-Cartesianism according to which for an
organism to experience pain, it must be capable of an advanced form
of cognition: higher-order thought. That is, an organism must have the
ability to think about its own mental content. Unsurprisingly, it is quite
questionable whether or not non-human animals can satisfy the require-
ments of this highly intellectualized theory of pain. Some theorists try
to argue that non-human animals do satisfy this condition. I, however,
will be arguing, essentially, that neo-Cartesianism is simply confused or
irrelevant given current trends in pain theory as revealed by the above,
notable account of pain.
Though I like the broad notion of pain above, the term “pain” tends
to have a narrower usage in the literature on the problem of evil as well
as in ordinary language, so I will for this and other reasons prefer the
term “suffering” as the name of the central phenomenon in my datum.
Suffering will include both physical pain and mental pain as well as
other forms of distress and failure to flourish. I have in mind something
very close to what Rollin calls “negative mattering” (2011: 425) or what
van den Bos, Houx, and Spruijt call “negative experiences” (2002). As
Rollin points out “physical pain is by no means the only morally rele-
vant mattering – fear, anxiety, loneliness, grief, certainly do not equate
to varieties of physical pain, but are surely forms of ‘mattering’ ” (2011:
425). This is in perfect accord with the National Association for Science
directives that focus not just on pain but stress and distress (National
Research Council 2008). Furthermore, animal welfare advocates routinely
express their concern for animal well-being in terms broader than phys-
ical pain. Consider this member of the “Animal Welfare Principles” from
the American College of Animal Welfare (this is a group aimed primarily
at providing specialized training for veterinarians).20
20
American College of Animal Welfare (2010).
“Animals should be cared for in ways that minimize fear, pain, stress,
and suffering.”
It is unclear to me what American College of Animal Welfare mean by

“suffering” here, but it is significant that the principle differentiates
between pain – presumably, physical pain – on the one hand and fear
and stress on the other, which are psychological (and so, presumably,
“suffering” is meant as a catch-all term for other forms of physical and
psychological negative experiences, which is close to how I understand
the term). This broad notion of suffering is significant for three reasons.
First, because it is less clear how neo-Cartesian gambits apply to it. Also,
it provides a useful framework for fully appreciating the scope of the
problem of animal suffering, which is worse when we consider its multi-
dimensional nature. Finally, it is significant for the theodical strategy I
will pursue below. For animals that can feel fear and stress can also feel
the positive reciprocals of those, the blessings of which might be made
sweeter for having gone through a “vale of tears.”
This broader emphasis on animal well-being is not at all idiosyncratic.
The American College of Animal Welfare (2010) goes on to list the
characterizations of animal welfare from a number of different authors
(emphases often added, see the ACAW document for specific references)
from which they draw.
1. “The welfare of an individual is its state as regards its attempts to cope

with its environment.” (Broom 1986)
2. “Welfare is a broad term which includes the many elements that
contribute to an animal’s quality of life, including those referred to
in the ‘five freedoms’ (freedom from hunger, thirst and malnutrition;
freedom from fear and distress; freedom from physical and thermal
discomfort; freedom from pain, injury and disease; and freedom
to express normal patterns of behaviour).” (OIE Animal Welfare
Guidelines 2005)
3. “Welfare is a broad term that describes how well animals cope with
their environment. Assessment of animal welfare should include
evaluation of an animal’s physiological and psychological well-being.”
(Illinois State Veterinary Medical Association 2004)
4. “Animal welfare relates to both the physical and behavioural health of
the animal or species; therefore, good animal welfare is the providing
of the correct environment and care to satisfy the ‘five freedoms’ for
the animal or species concerned.” (Simon Adams 2005)
5. “Animal welfare is where over time an animal experiences positive

emotions that it values and is reasonably free from negative emotions
that it would avoid.” (David Main 2005)
6. “Animal welfare relates to how an animal is feeling, and this is
determined by the factors included in the ‘five freedoms.’ ” (Sean
Wensley 2005)
7. “Animal welfare is the state in which an animal is in reasonable
harmony with its environment, has adequate fulfillment of phys-
ical, health and behavioural needs and is not subjected to unneces-
sary or unreasonable pain or distress.” (Murray Gibb 2005)
8. “Animal welfare is the acceptance of responsibility of all people
to fulfill their calling as custodians of creation towards all animals
with compassion, in order to relieve hunger, prevent maltreatment
and alleviate pain and fear.” (South African Veterinary Association
2005)
9. “Animal welfare is a human responsibility that encompasses all
aspects of animal well-being, including proper housing, manage-
ment, nutrition, disease prevention, responsible care, humane
handling, and, where necessary, humane euthanasia.” (Canadian
Veterinary Medical Association 2005 Directory)
10. “Animal welfare is the general physical and psychological well-being
of an animal or group of animals.” (Bonnie Beaver 1994)
11. “Animal welfare (well-being) is a state of harmony between the
animal and its environment, characterized by optimal physical and
psychological functioning and high quality of animal’s life.” (Hurnik,
Webster, Siegel 1985)
12. “Animal welfare is the physical and mental ‘well being’ of a
nonhuman animal while it is alive.” (Barrows 1995)
Notice how consistent the emphasis is on psychological and emotional

well-being as well as other more general types of well-being that outstrip
mere physical pleasure and pain: normal behavior, nutrition, and espe-
cially contentment. Adopting a focus on well-being, if we use the term
“harm” to describe the result of deprivation of natural well-being, then
“suffering” will be an apt term for the quite broad sense of “pain” I have
in mind in the title, for harm is something suffered.
As much as I hate to give up a broad usage of “pain” to describe both
physical and mental harm or dysteleology, much of the literature – both
philosophical and scientific with which I will be dealing – uses the term
more narrowly, so I will tend to use the term “suffering” in this book
to describe the states of affairs that constitute the basis for the problem
theists face from facts about how the lives of animals go badly in various
respects.
In considering suffering to be the basic phenomenon, I am in agree-
ment with Eleonore Stump. She points out (2010: 5ff) that the standard
concept of pain is not broad enough to constitute the problem that goes
by the name “the problem of pain.”21 I think it is fitting that Stump’s
account – the most recent book-length treatment at time of writing – is
in agreement with the language from Rowe that kicked off the contem-
porary discussion of the “evidential” problem of evil. In the very first
lines of his first section of his first major article on the subject he says:
In developing the argument for atheism based on the existence of

evil, it will be useful to focus on some particular evil that our world
contains in considerable abundance. Intense human and animal
suffering, for example, occurs daily and in great plenitude in our
world. Such intense suffering is a clear case of evil. (Rowe 1996 (orig.
1979): 2)
This only commits Rowe to the thesis that what he means by “suffering”
is a paradigm of evil, but the fact that he mentions suffering often (and
its profusion and intensity) and never evil in general is some evidence
that in fact it is the suffering itself that constitutes the problem rather
than the existence of evil as such. And his particular examples of human
and animal suffering are not limited to physical pain. In fact, they exem-
plify part of what Stump (2010: 6) calls “the variegated awfulness of
human evils” which also applies to non-human animals.
Caveat: Technically, the way Stump puts it, suffering is an instance of
a more general category of bad states of affairs. She asks the propaedeutic
questions, “What is wrong with the evil human persons suffer? Why is it
to be lamented and avoided? What is bad about it?” (2010: 5, emphasis in
original). And later, after concluding, rightly in my view, that it isn’t just
pain itself, she asks “What is bad about human suffering if it is not just a
matter of pain?” (2010: 6, emphasis added). Thus, badness seems to be the
conceptual anchor in Stump. Swinburne makes this his official position.
I shall henceforward normally contrast the “good” with the “bad”

rather than with the “evil,” both when talking about the actions of
21
For a fascinating set of reflections on the nature of pain and suffering, see
the section “When Is Pain Not Suffering and Suffering Not Pain” in Coakley and
Shelemay (2007).
agents and their characters, and also when talking more generally
about states of affairs – though an agent of very bad actions and char-
acter is appropriately called evil [sic] ... .Pains and other suffering are
bad states of affairs, but it is odd to call them evil (as I did in the
Introduction), even if some agent causing them or allowing them to
occur would be doing an evil act. Although the problem with which
we are concerned is called “the problem of evil” (and so those words
form part of the title of this book) it is really the problem of the
existence of bad states of affairs, such that (it is claimed) it would be
bad for an agent who could prevent them to allow them to occur.
(Swinburne 1998: 4)
Swinburne goes on to make the problem of evil precise in an argument

the key premise of which features the claim that “There is at least one
morally bad state e which is such that ... God does not have the right
to allow it” (1998: 13). He then dedicates half his main theodical effort
defending that it is permissible for God to bring about (“permit” might
be better) the bad states of affairs there are for the sake of certain good
ones (1998: ch. 12). Some object to the notion that God has any duties
to humans (Davies 2011), but I think Swinburne is correct that God
voluntarily enters into certain obligation-bearing relationships (1998:
10). The focus of the problem as I formulate it does not focus on the
issue of rights and permissible action. The principle focus of the problem
as I present it is that certain axiologically negative states are a better fit
with naturalism than with theism. However, of course, if it could be
demonstrated that it is impermissible for God to allow it to be the case
that E, then Pr(E | T) would be 0 and the debate would be ended. This is
not a topic for which there is time to discuss in this book.
Stump goes on to flesh out suffering in terms of failure to flourish
(2010: 8). She explicitly connects this to well-being. This ties in well with
what I’ve said above about suffering and harm, and note that numbers
3, 9, 10, 11, and 12 in the American College of Animal Welfare accounts
of animal welfare all define animal suffering in terms of well-being. This
is a happy convergence of the sensitively philosophical and intelligent
advocacy. So my broadest account of the problem of animal pain, then,
will be the problem of the ways in which animals suffer harm in their
failure to flourish. These are the bad states of affairs that, to many, do
not seem to fit with an account of a God aiming at good states of affairs.
My task, then, will be to provide a defeater-defeater in the form of a story
the plausibility of which I will defend. Just how plausible it needs to be
to count as a success will be treated at the end of the next chapter.
2.3 The amount of suffering
Finally, in this chapter, I will treat a vexing issue. Here is a moderately

precise version of that datum.
E Since nearly the beginning of sentient life on Earth, there has been
a profusion of intense animal suffering. That is, almost everywhere
there has been sentient life very significant levels of suffering have
been quite common (and there has been a lot of sentient life).
A few clarifications are in order. I am thinking of “profusion” in terms

of frequency.22 “Quite common” is vague, but plausibly rules out values
like 10–15% of sentient organisms. So the vagueness does not rule out
substantive implications. I should add that by “sentient” here I mean
“merely sentient,” since I am not considering human pain. Furthermore,
I am using “sentient” in a fairly traditional sense for creatures that have
the power of sensation (more will be said of this in Chapters 3 and 4), in
the sense of the ability to feel pain.
There is a marginal issue that needs mentioning before being dismissed.
Suppose that worms are sentient, which I don’t find wholly implausible.
Nevertheless, I find it implausible that they are capable of very much
pain at all and not enough to describe them as suffering. So I wish E
to be understood in such a way as not to entail that for every kind of
sentient creature, that suffering has been profuse and intense within
that kind. So perhaps it should say “since the beginning of significantly
sentient life.” I believe the charitable reader can tell what I am getting
at, so I will not further attempt to explicate the datum for now. Some
charity is always required from both author and reader when consid-
ering difficult issues. I am trying to exercise authorial charity in trying
to be fairly precise and flag possible issues, but it is impossible to get
it exactly right, so I will need to depend, not viciously, I hope, on the
charity of the reader here and elsewhere.
Of course, profusion and intensity come in degrees. And the more
profuse and intense suffering becomes, the less likely it is that God would
allow it. Thus, the great profusion and great intensity we find (and inter-
section of the two) would seem to be powerful evidence against theism.
This thesis can be challenged. So long as there is the possibility of a great
22
And, somewhat, distribution. If all suffering were sequestered in a relatively
small portion of space-time, the problem might not seem as bad. I’m not sure
about that though.
enough good – and surely the greater the good the greater the prob-
ability of it eluding our grasp – how can the quantity of the evil make an
evidential difference? The suggestion is quite simple: If, for any quantity
of evil, there is some good that outweighs it (I bracket deontological
issues for now), and the possibility that that enormous good logically
requires the permission of a horrendous evil of arbitrary size, then it
may well be for that great a good that this great an evil is permitted. This
line of thought has been pursued by Michael Tooley.23
[I]f evil can be necessary for a greater good that outweighs it, the
idea of worlds that are larger and larger ... strongly suggests that one
cannot place any upper limit on the quantity of evil that might be
found in a world created by God. (Tooley 1991: 92)
This suggests the following thesis, the No Quantity Too Large (NQTL)
thesis.
NQTL Given the (epistemic) possibility of arbitrarily great goods,

greater quantities of evil do not disconfirm theism more than lesser
quantities.24
Of course it is true that there is no upper limit on the size of a world

or the amount of good a world contains, but this does not obviously
have the consequence of supporting the “no greater evidence” thesis,
which would undercut my thesis that the great profusion and inten-
sity of suffering that are part of the animal kingdom potentially provide
stronger evidence against theism than worlds with lesser evils. Here is a
reason to think NQTL is false.
Imagine that for the last 300,000,000,000 years every sentient creature
(let there be 300,000,000,000 of them to keep it simple) has been in
excruciating pain at every moment of its existence (and none of them
ever sleep). They despise their past and look with horror upon their
future. Furthermore, none of them ever experience pleasure of any kind.
In fact, every sentient creature, assume, currently lacks the ability to feel
pleasure. No creature has been happy or had any positive emotion at
any time. You find yourself in this world and the question occurs to
23
And is, I think, implicit in Bergmann’s skeptical theses (most recently,
Bergmann 2009 and somewhat explicitly in Bergmann 2012).
24
Tooley’s point may well be about logical compatibility, but I think similar
reasoning applies to confirmatory concerns.
you: Is there an all-powerful, all-knowing, perfectly loving being who

designed all this? It is hard to take the question seriously in this context.
And won’t it still seem unlikely after you’ve also had the thought that
there are some goods and bads and entailment relations between them
of which you are unaware? Won’t a properly functioning mind find
incredible the idea that this world was made by such a being? I have a
great deal of sympathy with such thought. But my official thesis is much
weaker. My official thesis is that observing that you are in the world
just described gives you more evidence against the existence of an all-
powerful, all-loving God than the observation that one is in the current
universe. Likewise, then, the observation that we are in the current situ-
ation – assuming the current situation is as grim as it often seems – gives
one more piece of evidence against theism than a much more benign
world.
Less extremely, compare two worlds as alike as possible except one of
them is the actual world, with its 10^13 turps of evil, and the other is
two orders of magnitude greater volume (and density) and with 10^14
turps of evil. The sum total of evil has increased, but the “quality” (in
some sense) of the world as measured in proportional terms, has gone
up. We could consider a series of worlds ever larger and with ever more
suffering but with ever less a percentage of evil. Question: Are the
worlds with much, much more evil that constitute a much, much lower
percentage of the whole world better worlds or worse worlds than those
smaller worlds with less evil in higher concentration? Which worlds
form a better basis for the problem of evil?
It is natural to think that in the larger worlds, even though there is
much more evil, this is offset to a greater degree than in the smaller
world by the greater proportion of good. However, there’s so much more
suffering in that larger world! To take an extreme case, consider a world
with three people, one of whom suffers and the other two lead pretty
easy lives. Compare this with a world with 1,000,000,000,000 people
where “only” 1,000,000,000 people suffer. All else being equal, is the
smaller world worse because the per capita suffering is so much higher?
That doesn’t seem right to me.
Another natural thought is that the bigger world is better because in
it, for every suffering person, there are 1000 people with good lives,
whereas there is only a 2:1 ratio in the smaller world. I get that, I do, but
here’s a reason to resist. Suppose you accept Stump’s requirement that
the goods that justify the permission of evils must accrue to the very
individuals who suffer. Then it doesn’t seem that those 1000 people’s
great lives make a difference to the degree of evil in their world, if it
doesn’t somehow translate into a benefit for the suffering. This suggests
that quantity matters more than proportion when considering how
strong the basis for the problem of evil is in a world. However, here is a
consideration that pushes me in the opposite direction.
Here’s something I think is true. If you and most of the people you
consider to be properly functioning, intelligent, sensitive people found
yourselves in a world as much like ours as possible except that the
number of people suffering horrendous evils was doubled, you and
most of them would have considerably more doubt about theism than
you do. And I think that one of the ways to assess the expectibility of
states of affairs for theism is to observe the expectation of (properly func-
tioning) theists.
Earlier, we looked at Tooley’s No Quantity Too Large thesis. We’ve
basically been considering the following cousin, the No Proportion Too
Large Thesis (NPTL).
NPTL Given the (epistemic) possibility of arbitrarily great goods,

greater proportions of evil in worlds do not disconfirm theism more
than lesser proportions.25
The considerations just above are sufficiently strong for me to conclude

that the NPTL thesis is wrong. In the end, I am far from certain about
these things, but the cases I have considered that pull me toward the
conclusion that both magnitude and proportion matter pull me much
harder than any of the other cases or principles.26
Whether considered in terms of quantity or proportion, it is clear that
there is a morally significant quantity of animal suffering in our cosmos.
That should not be much in dispute. We have no reason at all to think
that there are untold billions of blissful beings bandying about beyond
the borders of our best telescopy. So in at least one clear sense, there
25
The proportion is over observed portions of total lifespans. Also, there is
an interesting epicycle here that is worth mentioning but not pursuing here.
Suppose there are five beings with moral standing. Four of them experience, for
the duration of their finite lives, constant, low-grade pleasure. The fifth indi-
vidual suffers excruciating pain for her duration on earth. I want this world to
come out as having a high proportion of suffering. It is intuitively a world that
raises a problem of pain. But to spell this out would require classifying moments
not just as good or bad but weighting them by a scalar quantity. It is not hard
to see how this should go, but the details would be tedious and take us too far
afield.
26
And it is not insignificant to me that the tradition clearly takes the profusion
of evil to be important, not just its mere existence.
is a significant proportion of suffering in this universe. And since the

vast majority of this suffering is constituted by non-human animals (or
at least after we subtract out for a successful soul-making theodicy for
humans), we seem to have a very serious problem of animal pain: A
significant portion of the sentient beings in our universe suffer signifi-
cant pain in significant numbers a significant amount of the time.
I will continue to work with the datum as I have described it, but I
think it is clear that if my theodicy works for the datum as stated it will
work as necessarily modified by the Tooley-esque theses.
3
The Bayesian Argument from
Animal Pain
3.1 Formalizing the problem: a Bayesian approach
We have been considering the nature of the datum that is the foun-
dation of the problem of animal pain. The first chapter identified the
problem as one of a potential failing in comparative confirmation rela-
tive to animal suffering. The second chapter defended the possibility
and sensibleness of making the kinds of judgments necessary for that
kind of problem to exist and clarified important features of the datum.
Now that we have clarified our starting point adequately, we can discuss
the path from there to the problem for theism. I will give the problem
a probabilistic analysis. This chapter will cover three things. First, I will
briefly describe the formal structure of the argument in Bayesian terms.
Then I will consider, for each of the two components of the equation,
what a plausible assessment of their values might be. The first step is of
obvious importance. The second (two-part) step is important for situ-
ating the importance of the outcome of considering the problem of
animal pain and situating it within a broader context in the philosophy
of religion.
3.1.1 The formal structure of the argument from

animal suffering
Consider the odds form of Bayes’s Theorem.
(
Pr (Hp | Evidence ) Pr Evidence | H p
= ×
)
Pr H p ( )
Pr (Hd | Evidence ) Pr (Evidence | H d ) Pr (H d )

Posterior Odds Likelihood Ratio Prior Odds
36
The Bayesian Argument from Animal Pain 37
In this equation, there are three kinds of expressions, each exemplified

twice: once for each of two competing hypotheses – Hp and Hd. We will
consider them from right to left. The three kinds of expressions are, first,
the prior probabilities of the hypotheses under consideration. This is the
measure of the intrinsic plausibility of the hypotheses before we consider
how a body of evidence bears upon them. Next in the right-to-left
direction are the likelihoods. They measure the predictive power of the
hypotheses with respect to the evidence in question. They express how
expected the evidence is given the hypothesis in question. Since we can
ask how expected something that has already happened ought to have
been given what evidence was had at some previous point, predictive
power needn’t be – in fact usually isn’t – a matter of literally predicting
the future. Finally come the posterior probabilities, which are our final
target. They express the probability of the hypotheses after the influence
of the evidence has been assessed (via the likelihoods) and weighted1
by the prior probabilities. The equation is a theorem of the standard
probability calculus,2 but hopefully the relations are all intuitive. The
equation says essentially that there are two ways things can go well (or
poorly) for a hypothesis: It can start out with a measure of intrinsic
plausibility and it can get a boost from its ability to render intelligible
the evidence we have.
Since this equation compares one hypothesis to another in terms
of the components enumerated above, it is composed of three ratios.
The posterior odds – our final target – express the ratio of the posterior
probability of one hypothesis – posterior to considering the evidence
in question – to the posterior probability of its competitor. If the odds
are, say, 3:1 in favor of Hp – that is, if the quotient equals three – then
1
Perhaps you will remember weighting from grammar school. Suppose the
teacher informs you that you will be given two exams – a midterm exam, which
counts for 40% of your grade, and a final exam, which counts for 60% of the
final grade. And suppose you get 93 points on the first and 88 points on the
second and you wonder if you’ve made the cut at an average of 90. You can’t just
average the grades, for that would ignore the weighting. So you use the following
equation .4(93) + .6(88) = your grade. The weightings are multiplied with the
exam scores to give you your grade, which is the “weighted average.” Similarly,
you can’t just assess how well hypotheses do on the evidence; you also have to
consider how they ought to be weighted by their prior probability. Thus, the prior
probabilities are multiplied with the likelihoods to give the final probability of
the hypothesis.
2
I do not assume (or think) that the Kolmogorov axioms are the only (or best)
way to represent epistemic probability, but we needn’t get into that here.
Hp is a substantially better explanation.3 When the prior plausibility of

two theories is equal – as we’ve programmatically assumed here – then
the posterior odds are equal to the likelihood ratio. This quotient is also
called the “Bayes Factor” and is a standard way to compare hypotheses.
Here is the odds form of Bayes’s Theorem with the following issue-spe-
cific variables: “E” for the evidence of animal suffering, “T” for theism,
and “N” for naturalism.
Pr(T | E) Pr(E | T Pr(T)

= ×
Pr(N | E) Pr(E | N ) Pr(N)
Programmatically assuming that theism and naturalism have approxi-

mately the same intrinsic probability is represented by setting Pr(T)
approximately equal to Pr(N), which makes the final term on the far
right of the equation approximately 1. Thus it will have approximately
no effect on the final outcome. That is, the posterior odds will be
approximately equal to the likelihood ratio. It is worth briefly treating
the nature of simplicity in order to establish that it is worth considering
how theism fares with respect to the evidence as opposed to naturalism.
For if one had a massive advantage in prior probability over the other,
then it might not matter much whether or not the evidence of animal
suffering decreased the probability of the hypotheses.4 Following that,
the central focus will be on the likelihood ratio.
3.1.2 The simplicity of the hypotheses

Suppose we are judging the relative simplicity of two model universes,
our own universe and an alternative universe, AU. When we go to
assess the simplicity of a universe, how should we do so? One method
would be to count up all the things each universe contains: diamonds,
daffodils, dogs, etc. and see how many there are. We then compare
the two numbers and judge that the universe with the lesser number
of things is simpler. Even if we counted only types rather than tokens
(which would be misguided in my view) this would in a way exaggerate
3
Jeffreys (1961: 432) lays out a hierarchy of favoring, but I think it is far too
conservative.
4
The fact is, I think the prior probability of theism is many, many, many times
more than naturalism, given that there is anything at all. Thus, I think evil would
have to favor naturalism many, many, many times more than theism for agnosti-
cism to be justified. I do not expect most atheists will think this though, so I will
argue that it is not at any rate much less.
Elementary
Particles
u c t γ
Quarks up charm top photon
Force Carriers
d s b g
down strange bottom gluon
νe νμ ντ
electron muon tau
Z
Z boson
Leptons
neutrino neutrino neutrino
e μ τ W
electron muon tau W boson
I II III
Three Families of Matter
Figure 3.1 The “Standard Model” of Fundamental Particles
the complexity of our universe. Consider the Standard Model (SM) in

particle physics.5
SM posits 16 fundamental particles that fit into three categories, and
in two categories there is an obvious symmetry. Let us assume for the
sake of argument that two kinds of reductionistic theses are false. First,
let’s rule out levels reductionism: the thesis that biological and chemical
facts are nothing but complex redescriptions of the physical facts. Next,
let’s rule out a kind of parts reductionism: the thesis that composite objects
are nothing but the sum of the simples of which they are composed.
Even with these kinds of “nothing but-ery” set aside, there is something
clearly right in the notion that the right way to assess the complexity
of our cosmos takes into account not just how many types and tokens
there are but also how they are related to one another, their structure as
represented in SM (both in terms of the fundamentality of the particles it
quantifies over and the relations and symmetries it posits among them).
For SM tells you, in some sense, what is fundamental to the universe.
And this is so not just in terms of fundamentality about parts, but
also a kind of explanatory fundamentality. Knowing SM allows you to
5
The Higgs boson now stands awkwardly off to the side, but I’m going to
ignore it for illustrative purposes. Also, I assume that there are gravitons, though
this has not been confirmed yet.
understand – given initial conditions of the relevant sort of a universe –

what kind of universe you will end up with and why it has the features it
has. You don’t need to know anything more than what SM can tell you
to understand how the universe works.
Some theists think that the simplicity of a theistic universe should be
assessed in a way not wholly unlike the way just discussed. (Swinburne
2004: ch. 5, is one such proposal, which is similar to what follows.
He makes a few changes in Swinburne 2010a.) Theism postulates one
brute fact6 and the rest flows from that in conjunction with necessary
truths about value. Bare theism’s brute fact is the existence of a person
with two properties – knowledge and power7 – held in the simplest
possible way – zero limitation. The explanation of every contingent
truth (other than his own existence, if that is taken to be contingent,
an issue too big for the present discussion) is a function of the good-
ness of the corresponding state of affairs. Since there is no best world,
an arbitrary choice must be made as to which initial world segment
to actualize among sufficiently good initial world segments. (This
wording allows for the universe to unfold in ways perhaps unforeseen
even to God, if it contains beings with free will or if it contains certain
kinds of chance processes. If the best kinds of worlds have a good deal
of chance, then a world created by God will be expected to have a good
deal of chance.8)
So theism, like SM, postulates as the base of its system a simple struc-
ture. It posits one fundamental entity of a familiar kind (a person9) and
6
I will not here address the issue of in what sense God is a necessary being.
Plantinga, and a large number of analytic theists, think that God’s existence
is “broadly logically necessary” (Plantinga 1979: 2), but others who are less
Platonistic think of God’s necessity in other ways. See specifically Swinburne
(1994). If Plantinga is right, then theism postulates no brute facts at all.
7
Unlimited power plausibly entails perfect freedom and Swinburne argues that
God’s perfect knowledge, power, and freedom entail his perfect moral goodness
(2004: 99ff).
8
For more on the place of chance in a world sustained by God, see van Inwagen
(1995: 42–65). See also Kraay (2008a and 2008b).
9
But isn’t this postulated person of a very unfamiliar kind? I don’t think so,
since it is a being with beliefs and desires, an agent that acts for ends. But what
about the fact that this person is conceived as disembodied? In answer, first, I am
not willing to concede to materialism about minds whether human or animal.
Second, I don’t see why the dissimilarity of disembodiment is particularly rele-
vant considering the much more significant similarity in agency. To think other-
wise seems to fly in the face of a fairly obvious functionalism. I recommend
Taliaferro (1994) on this.
a principle of creativity (in terms of goodness)10 to explain the existence

of all contingent reality. So in terms of explanatory axioms theism turns
out to be a very simple hypothesis indeed.
Naturalism lacks this kind of explanatory simplicity and systematicity.
On naturalism, there will be quite a number of brute facts: the existence
of contingent being, the existence of a number of laws, the many partic-
ular parameters of those laws, etc.11 Counting up the number of brute
facts in naturalism by the same method used above will be difficult, but
it seems that inevitably it postulates more than one brute existent. And
if it did, unless it only had one property (with a non-arbitrary measure)
then it would be at most equally simple. I don’t buy for a second Lewis’s
notion that it is only types that matter. There are more than 1080 funda-
mental particles, and it is hard to see how each particle’s existence isn’t
a brute fact on naturalism.
Of course, there are other methods of assessing simplicity, and it
could be that there is some plausible one that would favor naturalism
over theism. I am not aware of one, however. The usual “theists believe
one extra thing”12 response is, however, completely inadequate. I have
pointed out in this section that there is at least one reasonable way of
thinking about simplicity on which theism turns out to be much more
simple than naturalism.13
10
Might this framework better support some kind of axiarchism (Leslie 2001
and Rescher 2000), rendering a personal God needless? I think not. For I cannot
find intelligible the notion of a non-personal force or principle with agency, and
unless one holds something along the lines of the thesis that every possible good
state of affairs is actualized in some world, one will need an explanation for why
some goods obtain rather than others.
11
In fact, given some kinds of Humeanism about laws, there might be trillions
upon trillions of brute facts.
12
This is what I hear the most in conversation. The expanded version is some-
thing like this. “We both agree that there exists the cosmos and all the stuff in it
organized the way it is. The difference is that you postulate one extra entity, so
clearly theism is at least a little bit less simple.” This is clearly too crude and will
not do.
13
Here is a brief presentation of and reply to an argument I heard from Paul
Draper that naturalism is simpler than theism, especially Christian (even mere
Christian) theism. I will put it in my own terms, speaking of partitioning prob-
ability space. First, partition the probability space into naturalism and supernatu-
ralism and assume for the sake of argument that the partitions are of nearly equal
size. There will be many different kinds of supernaturalisms – monotheism, finite
godism, polytheisms of various kinds, et al. – so the area occupied by theism will
have to be some fraction of the area occupied by supernaturalism, mere Christian
theism a sub-area of that, and full doctrinal Christianity a sub-area of that. I like
this argument. But I think it fails for a number of reasons. First, it is not clear that
Having sketched the outlines of a defense of the thesis that theism is

much simpler than naturalism, I feel comfortable putting the issue aside
with the programmatic assumption of their approximate equivalence of
simplicity.
That leaves us concerned with just the Bayes Factor, to which we now
turn.
3.1.3 The Bayes Factor

The Bayes Factor, you will recall, is another name for the likelihood ratio.
Here is the likelihood ratio with our issue-specific variables introduced.
Pr(E | T)
Pr(E | N)
Having for the sake of argument taken the priors of T and N to be

approximately equal, the Bayesian problem of pain then comes down
to this ratio. The problem is that it looks like it is very bottom-heavy,
indicating a considerable advantage to naturalism.
It is worth considering whether there is any obvious imbalance here
that would make careful consideration of Pr(E | T) otiose. First, recall the
starting point.
E Since near the beginning of sentient life on Earth, there has been a
profusion of intense animal suffering. That is, almost wherever there
has been sentient life very significant levels of suffering have been
quite common (and there has been a lot of sentient life).
Here is one way to see it is plausible that Pr(E | N) is not very low. N
entails that the universe is indifferent to the well-being of sentient crea-
tures on Earth. It is plausible that left to chance the distribution of pain
and pleasure among sentient beings would be utterly indiscriminant
of any discernable connection to moral properties.14 In a naturalistic
universe we don’t expect there to be any rhyme or reason to suffering.
naturalism and supernaturalism start out with the same area. Second, if they do,
it is not clear that the naturalism you get when you take it to be the negation
of supernaturalism will have much by way of explanatory power. Also, the argu-
ment works only to the extent that the non-monotheistic region of supernatu-
ralism has much measure. I think there are good arguments that it has very small
measure, so that the probability of traditional infinitary monotheism takes up
most of the space of supernaturalism. For more see Draper and Dougherty (2013),
and Draper and Dougherty (forthcoming).
Or rather, the fact that the only rhyme or reason seems to be tied to
reproductive success is something not surprising on naturalism. E seems
to be a natural consequence of an indifferent universe.
Of course, what is presupposed in the preceding paragraph is a cosmos
that exemplifies a number of properties the exemplification of which
plausibly constitute good evidence for theism. The existence of animal
pain entails the following items: the existence of contingent beings, an
orderly cosmos, biological complexity, and consciousness. And in some
way we must include in our total evidence the higher-level fact that
investigating the issue entails some moral knowledge. And it is not at all
clear how naturalism can explain the existence of moral knowledge. We
might as well not leave out evidence provided by the fact that we can
make the kinds of judgments required to think through the issue. And of
course this entails the existence of rationality as well. But even leaving
aside the evidence we get from our consideration of the evidence, these
other first-order items are all things that are often considered good
evidence for theism. And since if A entails B, A is logically equivalent to
A&B, a consideration of the evidential impact of A is in a way incom-
plete without a consideration of the evidential impact of B. So theists
may well have a rebutting defeater in the wider evidence for the theism-
defeater from E. However, it would be better to have an undercutting
defeater for evil, which is essentially what a theodicy provides. E seems
so natural on the assumption of N that the theist may need both some
rebutting defeater and some undercutting defeater to appeal to. Natural
theology considers the positive evidence for theism, including what is
entailed by the existence of animal pain. We will set this aside, since this
is a work of theodicy.
3.1.4 Taking stock thus far

Let us take stock. I have given a formal characterization of the structure
of the problem of animal pain. I sketched the outlines of an argument
that theism is sufficiently simple to have the initial plausibility to be on
the playing field. And I have given a brief indication of why naturalism
poses a serious threat (while at the same time gesturing at the kind of
considerations on the basis of which theism may be expected to gain
some empirical confirmation, including, potentially, from the problem
of animal pain itself). All this points to a situation in which the inves-
tigation of Pr(E | T) – such as we shall make in this book – should be of
interest to the theist and atheist alike: If it is very low, then theism will
not have sufficient positive epistemic status for rational belief; If it is not
very low, then the best argument the atheist thought she had is largely
ineffective. These theses explicate the intuitive characterization of the

problem that I set out earlier. Here it is again.
Given (the programmatic assumption) that theism and naturalism are
roughly on par with respect to simplicity, since the profusion and inten-
sity of animal suffering is more expectable on naturalism than theism,
naturalism will be the better explanation of all our data, unless other
data support theism more than the data of evil detract from it or it is
discovered that it is either not so expected on naturalism or not so unex-
pected on theism (or both).
I am largely assuming that the Irenaean (or “soul-making”) theodicy
works for humans. In Chapters 6 and 7 I lay out in detail a new version
of the soul-making theodicy. Principally I am principally opposing the
following kind of thesis, which is commonly enough expressed.15
The Irenaean theodicy may well work for humans, but it is of no use
when considering the problem of animal pain, so we should set it
aside for dealing with that problem.
I will be arguing against this thesis, and that will be work aplenty.
Here are some of the more substantive assumptions I’ve made above.
1. Moderate rationalism: Bare theism (and mere Christian theism) can right-
fully and usefully be treated relevantly similarly to scientific hypotheses.
That is, they can be suitably judged for simplicity, coherence, and fit
with the evidence in ways that bear on their probable truth.
2. Non-fideism: Evidential considerations are importantly relevant to
religious belief. Thus, it is a serious problem for a religious perspective
if it is at odds with the evidence or if the evidence favors a competing
hypothesis. (The moderate fideism in Bishop 2007b may be compat-
ible with this, though I reject it for other reasons.)
3. Non-skepticism about science: Most (or at least most of the core (or
at least much of the core)) of standard scientific doctrine is at least
approximately true (to some appropriate degree of probability).
4. Realism: The entities postulated by scientific and religious theories
that are probably true, probably exist. All the entities postulated by
completely true theories exist.
14
Paul Draper develops at length this Humean argument against theism (but
not necessarily for atheism, since it is consistent with deism). See Draper (1996).
15
See, for two prominent examples among many, Murray (2008: 6, 125) and
Betty (1992).
5. Simplicity is an alethic explanatory virtue: Simplicity considerations

are a guide to truth.
6. Simplicity is principally judged by ontological commitment: Theories
that cover the data with fewer entities postulated are more likely to
be true.
7. Ontological commitment is to be defined over basic entities or prin-
ciples of origin.
Here is a question that might have arisen for readers not unfamiliar
with the literature on the problem of evil. Addressing the question
will prevent some possible confusion. Some may think that the Bayes
Factor version of the argument I have given constitutes a version of
the “evidential” problem of evil rather than the “logical” problem of
evil. My position is that there is no such useful distinction to be had.
Mackie called his version of the argument a “logical” problem because
it consisted of the problem for theists that it appeared that a certain set
of propositions was logically incompatible (that God exists, that God
is omnipotent and perfectly good, and that evil exists). But of course
the existence of pointless evils (in a strong sense of “pointless,” not
Hasker’s weak sense16) that form the basis of Rowe’s so-called “eviden-
tial” argument are also logically incompatible with God’s existence.
Appeals to knowledge and certainty will not help, for those are person-
relative notions and one would only obtain a sociological categoriza-
tion of arguments.17 It is nearly certain that there is widespread animal
suffering (the opposition will be treated in Chapters 4 and 5), and if
it were to turn out that this was simply irreconcilable with theism,
as is plausible in the absence of the kind of theodicy I will advocate,
then the problem of animal pain would be like paradigm instances
of the so-called “logical” problem of evil. All that matters in consid-
ering a challenge to a hypothesis is how an item of evidence affects
the resulting probability of that hypothesis or how it fairs with respect
to competing hypotheses. All that matters in the problem of pain is
whether it reduces the probability of theism and if so to what extent
and how it affects the evidential balance between theism and relevant
competitors. That will be the focus of this book.
16
Hasker (1992).
17
For an account of the distinction that has these faults, see Howard-Snyder
(1996b: xii–xv). For a particularly baroquely Boolean version see Murray (2008:
11).
3.2 A Narrative approach to assigning probabilities
I will be treating theism and naturalism as hypotheses and assessing

their relative epistemic merit with respect to the evidence via assigning
probabilities. In presenting this kind of material in classrooms, confer-
ences, and colloquia I have frequently been met with skepticism
concerning judgments about what is how probable relative to what,
given that God exists. I have some sympathy for this skepticism, but
the fact is, I think our ability to assign such probabilities reliably comes
from our general ability to comprehend agency and, on a larger scale,
stories. When I cover this material in classes, I point out to my students
that they are assigning probabilities to the acts of agents based on their
character all the time. Their surprise at someone’s behavior, I avow, is
proof of this.
If an agent is portrayed in a story as repeatedly behaving in a surprising
way, the story will not be a good one unless there is something to defeat
this surprise, unless there is some subtext revealed that makes sense of it
all, in the light of which, it is no longer surprising. This happens when
we find out that someone has been struggling with alcoholism for a
long time while we never suspected a thing. In reading fiction, espe-
cially fantasy and sci-fi, even in the midst of good readers’ “suspension
of disbelief” about the details of the imaginative world – elves, Vulcans,
what have you – a work can be criticized for being “unrealistic.” This
criticism does not refer to features of the fictional world that are not
like our world. Rather, they refer to the way things that are the same
in our world don’t go how they should, such as people acting out of
character.18
Whatever the analysis of these probabilistic constraints, they will be
hard to map directly to displayable conditional probabilities. For the
relata will be very complicated since the plausibility of the story will
depend on how many, many details hang together. We just don’t easily
track in fine detail the discrete conjunctions that abstractly constitute
the theories we entertain. Instead, we often have some more direct
acquaintance with them, which is phenomenologically holistic. It is not
unlike the way we enjoy a multi-modal experience of, say, a dog barking
as it splashes in the water around us. We do not think of the discrete
sensory modalities, the feel of the spray of the water as distinct from the
smell, etc. Rather, it presents itself to us as a unified whole. Something
similar to this is the case, I think, for the theories we entertain.
18
Thanks to Blake McAllister for jogging this point.
The question we want to ask, proximately, may not be “What’s the

probability of this proposition on this one?” but rather “How would we
expect this story to unfold?”19 Story evokes a frame of mind in which
inquiry into support relations might be more effective. There is ample
evidence from the relevant cognitive science literature that priming
different frames of mind can significantly impact assessment. We really
do want to know, in the end, what the probability of one proposition
is on another. It’s just that asking about it directly may not be the most
effective means of obtaining that value. Among the reasons are how
complex the proposition that expresses the evidence is. I suggest that
a narrative frame of mind will lead to a better assessment of the prob-
abilities than, for lack of a better term, an “atomistic” state of mind. A
holistic frame of mind, which is more synthetic, can be more effective
than an analytical frame of mind because in the latter we try to section
off discrete conjuncts, which is in theory fine, but we don’t have the
mental horsepower to consider enough of them to avoid a misleading
judgment about what we are really attending to when we think about
two complex theses in relation to one another.
It can therefore be misleading to just consider the probability of a
certain rather narrow state of affairs conditional on theism. So when I
talk about what we would “expect” from God, while I ultimately aim
at a probabilistic notion of expectation, I wish it to be understood that
I believe such expectations can be judged “narratively.” We might call
this “narrative expectation.” This is a matter of how well the actions of
an agent in a story fit that agent’s character in light of the setting of the
story, the other characters, and myriad other events in the story. Thus
how the action of an agent is embedded in a larger story will determine
what is expectable, that is, what makes sense in terms of the parts of the
story we know.
So we have a model for assigning probabilities to God’s acts from
the fact that God is an agent who acts for reasons. Cognitive psycholo-
gists assert that we evolved a “theory of mind” – including, essentially,
the ability to reconstruct the thoughts of other agents – to successfully
navigate a social world.20 Because God is a person, a rational agent, his
actions fall under our theory of mind. Our paradigm application is the
minds of other humans, and there are no doubt perils in too extensive
19
I’m talking about the heuristic level here. That is, that asking a question
directly might not be the best way to get that question answered.
20
Though I already had material on the evolution of theories of mind in
animals in Chapters 4 and 5, Todd Buras pointed out its relevance here.
extensions of this to non-human persons, but it remains the case that

rational action is rational action, love is love, whether in dog, doctor, or
deity. There exists a core notion to support analogical thinking.21 I will
now consider a few “pious” objections to this line of thought.
The first objection comes wrapped in a cloak of humility. The idea of
narrative expectation was premised on how events fit into the rest of
the story. But of course when the story is the story of the whole world,
indeed, of the whole history of all that exists, in a way, we clearly can’t
claim to know the whole story. How, then, can we form reasonable
expectations about what God might do? Doesn’t this suggestion of mine
ultimately tell against rather than for the idea of discerning probabilities
by attention to narrative aspects?
I said in Chapter 2 that skeptical theism doesn’t eliminate the argu-
ment from evil. It can at best weaken the force of it. Where E = the data
of evil, T = theism, and N = naturalism, if we could rightfully claim that
Pr(E | T) were low and Pr(E | N) was high, then T would take a knock
while N would advance. But if the skeptical theist is right and we cannot
sensibly assign any value or (non-trivial) range of values to Pr(E | T),
then E won’t be a direct setback for T, but it will still be an advance for
N. And if T and N begin at the same starting line, E’s advancing while T
“stands still” will result in N becoming the better explanation. So skep-
tical theism can’t help the theist much, but it can hurt the theist a lot.
For if the skeptical theists are right, then we are “in the dark”22 about
what to expect concerning theism, and my project of arguing that my
theodical story is sufficiently highly probable on theism is moribund.
Theism’s sole axiom is that there exists a person23 of a certain kind: a
rational agent – a being with mind and will – whose mind comprehends
all truths and whose will is perfectly effective and unhindered by any
outside force. Here are some further assumptions to which I will help
myself without defense. First, I assume that facts about what constitute
right action and right reasoning are necessary truths. Second, I assume
that when an object is conceived of as good, it automatically gives the
agent who grasps it as such a reason to pursue it (also that a perfectly
21
There is a lot of confusion over how close to Scotus Aquinas is on theological
predication, but only the most extreme interpreters of Saint Thomas should have
a problem with what I am saying here.
22
Bergmann (2001: 289, 291) and (2009).
23
Though in orthodox Christianity, God is tri-personal, it is still sensible to
speak of God as “a” person due to the unity of the Godhead. I wish to be explicit,
though, that this manner of speech is not meant to deny a robust doctrine of the
Trinity. Indeed, I am a social Trinitarian.
free will will in fact pursue it when there is sufficient reason). Thus,
states of affairs are, at the most general level, expected to the extent
that they are good states of affairs, that they realize some kind of value,
and the higher the value, the higher the expectation. Of course there
will be a large infinity of ways of realizing the highest goods (there are
likely to be very high goods the co-exemplification of which in a world
is not compossible), so no particular way of realizing them will have
an antecedently high mathematical probability. But we are not trying to
predict in advance which goods will be realized in which ways. We are
“retrodicting,” asking ourselves what makes sense of what we take to be
the case.24 When doing this, what matters is not so much which tokens
occur but rather which types are tokened. Given that God is a rational
agent, we expect the best kinds of values to be tokened (and, of course,
for means to serve ends well: We want the goods to be brought about in
the most fitting way).
Of course, this explicit description of the basis for expectations from
God makes it easy to state the skeptical theist objection. For will there
not be goods greater than we are capable of conceiving? And if so, isn’t
it as likely as not that animals are allowed to suffer for some good we
cannot conceive and one that doesn’t require anything like the kind of
theories you will propose in this book? And, anyway, with an unknown
number of unknown goods in God’s purview, how can I argue that we
should expect just sorts of states of affairs as I will advert to to explain
why animals suffer so, instead of one of the vast number of unknown
goods? In other words, since explaining some aspect of creation is
just explaining for what reasons God acted – what good he sought to
achieve – if we have so little access to the range of reasons for which
God might act, how can any such explanation be offered with any confi-
dence that it – and not another – is the right one?
Some authors grant this point and are content to assume a defen-
sive posture wherein the mere possibility of God’s having reasons allows
them to dig in and remain unmoved by the data of evil. Perhaps there
is something to this, but it is clearly sub-optimal. I will have more to
say about this below. For now, however, I just want to point out that
there is no clear reason given by skeptical theists for the proposition that
there are unfathomable goods in the relevant sense. Yes, even a good
24
There is a view according to which forward-looking predictions generate
stronger epistemic reasons than do backward-looking ones. This view is called
“predictivism.” For an argument against it, see Swinburne (2001: 221–232,
appendix on Predictivism).
Hollywood director can make you say “I didn’t see that coming!” but
it’s not at all clear that that our inability to anticipate in advance what
God might do is relevant. The reason is that I will not offer a great deal
of detail in my theodicy, and I will only be suggesting what I take to be
logically necessary given that God’s character entails that he will pursue
the highest kinds of goods.
I see no clear reason to think there are kinds of goods of which we
are unaware, in any relevant way. This is because we have reason to
believe we have a sufficient grasp on human nature. We know what
is best for humans – love, friendship, knowledge, virtue, health, etc.
But we also have a grasp, then, on the general goods for sub-human
creatures insofar as they approximate the functions of human beings.
So for animals that can experience friendship, friendship is a good for
them. For animals that can have knowledge, knowledge is a good for
them. And of course health is a good for animals. So then if there were
some grand unknown good for the sake of which animals were allowed
to suffer, it would have to be “built out of” these goods of which we are
aware. And since a reason is just a grasped good, in knowing that these
are the kinds of goods that would have to be realized with the appro-
priate relation to current suffering, we know the kinds of reasons God
would have to have in mind.
There is another kind of objection to this picture that also stems from
a sense of piety. Doesn’t this image of God-as-rational-agent result in
something like deism? I don’t think so, but if it were so, that should not
be regarded as a huge downfall. Would it really be surprising if the gap
between the “God of the philosophers” and the “God of the covenant”
needed to appeal to revelation to be bridged? For my own part, I am
assuming, without defense25, that if we let H state some fairly standard
historical facts about Jesus and the Apostles, and let M state a kind of
mere Christianity, then Pr(M | T & H) is high. Thus, the picture I take
myself to be defending is a kind of mere Christianity, but I am happy
to defend a more deflated form of theism and work to bridge the gap
elsewhere. The move from atheism to deism is a step in the right direc-
tion, and it makes the prior probability of revelation go up. As Philo says
in the closing remarks of Hume’s Dialogues Concerning Natural Religion,
after concluding that something like deism was the result of natural
religion, “the most natural sentiment which a well-disposed mind will
feel on this occasion, is a longing desire and expectation that Heaven
would be pleased to dissipate, at least alleviate, this profound ignorance,
25
But see Swinburne (2010b).
by affording some particular revelation to mankind, and making discov-

eries of the nature, attributes, and operations of the Divine.”
3.3 Criteria for success: between theodicy and defense
The literature on the problem of evil has suffered from two unfortunate
distinctions. The first is the misleading distinction between the so-called
“logical” and “evidential” problems of evil. The second was introduced
by Alvin Plantinga in 1974. In God, Freedom, and Evil (1974a), Plantinga
defines a theodicy as an attempt “to specify God’s reason for permitting
evil or for creating a world that contained evil” (10, see also 27, 28, and
58). He distinguishes this from a defense. “Here,” he says, “the aim is not
to say what God’s reason is, but at most what God’s reason might possibly
be” (28). The chief aim of a defense, as Plantinga has it, is to demonstrate
that God and evil are broadly logically compatible.26
Both the defender and the theodicist, Plantinga says, seek a story that
includes both that God’s creating the world and that it contains evil,
demonstrating their consistency. On the weak interpretation of defense
in Plantinga, “whether [the story] is true is quite beside the point” (1974a:
28). On this interpretation, it doesn’t seem like the defender even needs
to consider the story epistemically possible. Of course, if, like me, you
think that everything metaphysically possible is epistemically possible,
then you will get epistemic possibility for free.
This notion of a defense is very weak indeed, but not the weakest
notion in the literature. David O’Connor’s notion of a defense, for
example, is weaker.
[A] defense tries to ensure that [theism] survives the charge or attack
against which [it] is being defended. [It does not] have to establish
that the evils cited in a given antitheistic argument are logically
consistent with God ... .Thus, against a prosecuting argument whose
aim is to show that certain facts of evil are inconsistent with God, a
defense succeeds if it shows that the prosecuting argument fails to
establish its conclusion. (O’Connor 1998: 2)
26
“Might possibly” is ambiguous. One reading, which seems consonant with
what else Plantinga says is that the first modal is epistemic and the second is
alethic. That is, a defense is an assertion of the following kind. “It is epistemically
possible that it is metaphysically possible that X” where X entails the existence
of some state of affairs which includes both God’s existing and the evils of this
world.
Just as his notion of a defense is weaker than Plantinga’s, his notion

of a theodicy is even stronger than Plantinga’s notion of theodicy. He
considers a theodicy to be “an attempt to answer, in a systematic and
comprehensive way, the question, ‘what is the source of the evil we find,
and why does God permit it?’ ” (2). Swinburne (1998: 15) takes a weaker
line, saying that the theodicist proposes “reasons God has” but stops
short of saying they are his reasons, that is, that he actually acts on these
particular reasons. I think in God the distinction between reasons there
are for one to act and reasons one acts on collapses, since I think God
acts on all the reasons there are for him to act.
Recently, Eleonore Stump (2010) has given an enlightening take on
the relationship between theodicy and defense. I will briefly present her
usage and attempt to generalize it in a way that blurs the lines between
theodicy and defense for an understanding of replies to evil that tran-
scend this unfortunate distinction. Often, those replying to the problem
of evil choose to give what they consider a “defense” rather than a
“theodicy” – usually using Plantinga’s terminology– because they take a
dim view of our ability to discern God’s reasons. But this is not Stump’s
motivation (14), and I take it she does believe the medieval theodicy she
defends. Rather, she is applying an important observation she makes
about theodicies. “A theodicy and a rejection of theodicy thus in effect
each offer or presuppose a view about what the world is like” (18). One
way of rejecting a theodicy, then, is to reject the basic worldview in
which it is embedded. She then says there are two ways of approaching
this rejection. One is to “argue through each claim” that constitutes the
division. Another is to “describe,” for the interlocutor, “in great detail
our own worldview” and then let them reconsider (18). In her book, she
pursues the second strategy. That, she says is, strictly speaking, to offer a
defense (19). Thus, she “will explicate the worldview within which the
typical medieval theodicy is defended” (21). Her final language is then
somewhat weakened from saying that her method is “strictly speaking”
a defense to saying that it is “in effect” a defense (there is seemingly
a further hedge at the end of the chapter (22)). I wish to weaken the
distinction further.
Stump says that in explicating the worldview behind our theodicy “we
would not be arguing for the truth of our own claims” (18) and further
that her “aim will not be to argue for the truth of the medieval account”
(21). What I want to suggest, though, is that describing is a way of defending.
That is, that the distinctions between the two methods – though picking
out something importantly different – do not differ much epistemically,
in the end. For coherence can increase the epistemic probability of an
explanation. Though it would take too much space to defend, I think

there are ample examples in the history of science – as well as in literature,
history, and the forensic disciplines in general – where the coherence of
a story is taken as evidence, and rightly so, of its truth. In terms of the
“explanatory virtues” the virtue of explanatory power is plausibly a kind of
coherence.27 Explanations aim at truth (or at least verisimilitude), and so
good explanations are likely to be true (or nearly true – I shall henceforth
drop the qualifier). Thus, I will not primarily use the terms “theodicy”
or “defense” and simply defend the probability of my story as best I can,
realizing I will have variable success. I will tend to use the term “explana-
tion” for my theodicy and discuss whether the explanation I offer is plau-
sible or probable and to what degree. But because the term was originally
broader – and, as we have seen, has a diversity of usages today – I will also
use the word “theodicy” and its cognates. But as I understand it, theodicy
encompasses a broad range of plausibility in explanations for suffering.
The range covered from the weak notion of defense to the strong notion
of defense to the weak notion of theodicy to the strong notion of theodicy
indicates that, in reality, we are dealing with a continuum. So I propose to
dispense with the binary and ambiguous notion and simply defend as best
I can the probability of the story I tell. But, like Stump, I will do so mostly by
means of an explication of the idea rather than via direct argument.
And the scalar nature of epistemic support is key for my project. For
I will simply be defending, as best I can, the plausibility of a certain
combination of stories (that is, I will be defending their plausibility both
jointly and severally). Plausibility at a certain level gives over to prob-
able truth. Results will vary by user. Indeed, my own credence in these
stories varies significantly over time. For regarding the central claim –
that animals will be resurrected and given the capacity to reflect upon
their life and so may see and appreciate, indeed be glad for, the ways
in which they were of service to others (even when no benefit came
to them at the time), and so will be able to benefit from their suffering
as well as have it defeated by their acceptance – my own credence has
varied greatly from inception to completion. At times it seems nearly as
crazy to me as it will to most readers at first, and at times it seems that it
could be no other way. In the end, I seem to have convinced myself.
Even partial belief can have considerable significance as a part of a more
complete explanation of the suffering of animals, what we might call a
“cumulative case theodicy.” There may well be other reasons for animal
27
For a very interesting foray into the relation between probability and expla-
nation, see McGrew (2003).
suffering, and if one finds them somewhat plausible as well, then the
theist’s overall burden is lifted. Let’s follow Daniel Howard-Snyder (1999) in
denoting the sum of all the reasons God might have for allowing the evils
we take there to have been and call it “The Reason” (100). Since if there is a
God, he acts on all the reasons there are, The Reason would be God’s reason
for permitting the evils that occur. We would like to know The Reason. That
is, we would like to know all the goods for the sake of which God permits
the suffering of animals. There is no way to know in advance the number
of such goods (and of course many of them would be expected to lie in the
future, which complicates matters). So the only reasonable thing to do is to
go around collecting them where we find them and see how they add up.
If we only label the reasons we collect “possible” or “impossible”
(whether metaphysical or epistemic), we will end up with no useful way
to sum their significance. But if we label them, as best we can, with some
weight, the evidential weight they have in our epistemic scales, then we
will have some way to estimate their collective strength. Knowing this
magnitude is important for another reason that requires that I make
clear two different ways in which reasons can add up for a person.
Take a number of reasons R1–Rn. Reasons can be fully sufficient for an
action or only partly sufficient for an action. By a “sufficient reason” here
I mean a good (grasped as such) that is sufficiently good to provide an
all-things-considered reason to perform some action. Though there are
intermediate and mixed cases, I think a simple illustration will suffice.
Suppose you want to know if S had a good reason to A. I assert that
there are three goods – G1–G3 – that will result from A. And you agree
that if any one of G1–G3 result from my A-ing (and all else remains
the same) then A-ing was justified. Still, you must consider for each Gi
how convinced you are that it will in fact obtain. Suppose that for each
of them you think there is, say, about a 20% chance that it will occur.
Suppose further – as is very unlikely in the real world – that the each Gi’s
occurrence is probabilistically independent of the others. Then there is a
60% chance that one of the good outcomes will occur. And since each of
the good outcomes provides a sufficient reason for the action, you could
be committed to the rationality of A-ing even though for each individual
reason for A-ing you are pretty sure it will not occur (the details matter
for a final determination). Humans are prone to ignore this kind of
fact and instead exhibit a bias toward “big reasons” ignoring cumula-
tive effects.28 The fallacious thought goes something like this: “None of
28
This is a type of availability bias most like the subadditivity effect. See Tversky
and Koehler (1994) for discussion.
these things is at all likely to occur, so there’s insufficient reason to act.”

Likewise, when considering reasons proffered for divine acts, even if one
is far less than convinced, one should take seriously each additional
gain in probability, for in a cumulative case theodicy, even relatively
small probabilities can sum to significance.29 And of course, a similar
situation has the same effects: where one is more certain that there are
goods that are not individually sufficient but which might be jointly
sufficient or jointly have considerable weight – for example, suppose
you are 80% sure that there are eight goods, each of which is about
1/10th the weight necessary for a sufficient reason. You will have gone
a considerable distance in showing that there is sufficient reason for the
act. And this indicates another way in which partial reasons can help
with the problem of evil.
Above I mentioned reasons for thinking that theism does a good job in
general of explaining many features of reality. The goodness of contin-
gent being, cosmological complexity, beauty, organic life, conscious-
ness, and moral agency all fit the theistic story very well and are all quite
surprising given naturalism. Though I will be arguing for the significant
probability of a good that would by itself be sufficient to justify God in
allowing animals to suffer as they do, that suffering is at least prima facie
an anomaly within the theistic narrative. But as common practices in
the sciences indicate, it can be completely rational to believe a generally
well-confirmed theory in the face of anomalies.30 However, doing so is
much more clearly permissible when one has “something to go on” in
moving toward an understanding of the recalcitrant data. For example,
the many “just so” stories told by evolutionary biologists about vestigial
organs, though usually wholly without independent evidence, do fit
well within the evolutionary narrative and so can contribute to the
rationality of belief in evolution even while keeping in mind the parts
that it doesn’t predict. My position is that, likewise, even an imperfect
account of the possible reasons for a divine action can play an impor-
tant role in the overall rationality (depending, of course, on the details)
of theistic belief. For this reason and the reason mentioned just above
even imperfectly defended theodicies can be important.
29
Michael Murray (2008) comes close to saying something along these lines
(195–197) but never gives the suggestion much shape and then (198–199) trails
off into another line of thought. I was worried I’d been “scooped” but came out
with the sense that I hadn’t been.
30
See Dougherty and Pruss (2014) for an extended defense of thinking of evils
as anomalies.
4
Is There Really a Problem?: The
Challenge of Neo-Cartesianism
The problem of animal pain, as I have formulated it, entails that there
are a significant number of animals that suffer in a morally significant
way. Though that suffering was defined quite broadly, physical pain has
for most constituted the crux of animal suffering. But, of course, purely
physical pain is very rare. When you are in “physical” pain, it is how
much you don’t like it that is the real problem, and that is a psycho-
logical state. The two may come apart in theory, but rarely in practice.
Some theorists would deny the very existence of a significant portion of
the data the problem of animal pain stems from. If there is no problem
worth considering, then there is no sense in strenuously seeking a solu-
tion. If there are not data crying out for an explanation, it is a waste of
time to carefully craft one. In this chapter, after some epistemological
prolegomena, I consider a prominent attempt to deny the data.
Here is an overview of the main dialectic of this chapter. Section 4.1
argues for the proposition that most animals (at least at the level of
mammals and above) feel morally significant pain is a dictate of common
sense (§4.1.1). The kinds of reasons we attribute pain to animals are
exactly the same as those for which we attribute pain to other humans.
It is also the common consensus of the scientific community (§4.1.2).
Therefore, unless we are given a defeater for this thesis, the most reason-
able attitude to have toward it is acceptance. Section 4.2 presents the
thought of those who do attempt to defeat this default starting point
mount both a positive and negative assault. The positive assault has two
prongs: philosophical and neuroanatomical. The philosophical assault is
to advance a theory of conscious experience that requires some kind of
cognition that animals are supposedly unable to have. Though surpris-
ingly popular, this is the weaker facet of the attack.
The neuroanatomical front is to argue that at least non-primate
mammals lack key neuroanatomical features necessary for morally
56
Is There Really a Problem?: The Challenge of Neo-Cartesianism 57
significant suffering. This two-pronged frontal attack from philosophy

and neuroanatomy is then combined with a negative assault stemming
from hard cases that cast doubt upon our common sense beliefs about
animal suffering. This latter I term the “negative” strategy. Because the
negative strategy is also neuroanatomical, I will divide the presentation
of neo-Cartesian arguments into philosophical (§4.2.3.1) and scientific
(§4.2.3.2) rather than positive and negative. In the next chapter, I will
answer the objections raised here. So the movement of this chapter
and the next is in three parts: First, I establish that affirming animal
suffering is the defeasible, default starting point; next, I consider an
attempt to defeat this presumption, then (in the next chapter), I show
that the attempt to defeat the presumption fails, and so the presump-
tion stands. Thus, I will conclude that there is indeed a problem of
animal pain.
The question in this chapter and the next is whether and to what
extent non-human animals in which the neural process of nociception
occurs also suffer as a result. The alternative is that nociception is merely
an informational process. A key route to addressing this question will
be the more specific question whether and to what extent non-human
animals are consciously aware of the noxious stimuli that are neurally
encoded. In short, do they feel pain (as opposed to merely reacting to
information about the threat of damage)?
4.1 A starting point: common sense and

scientific consensus
4.1.1 Common sense epistemology

My position is that philosophers should start their theorizing from
common sense. There is no absolute requirement to end there, for para-
doxes or empirical science may show common sense to be incomplete or
even incoherent (an example of the latter, plausibly, is counting artifacts
as objects). Thus, I reject the apparent faith of certain ordinary language
philosophers that ordinary discourse is sacrosanct. Nevertheless, I think
it is harder than some think to rightfully overturn common sense judg-
ments. I propose the moderate methodology of a common sense reflec-
tive equilibrium.1
1
I wish to note that reflective equilibrium epistemologies do not have to be
examples of coherentism. A foundationalist reflective equilibrium epistemology
is perfectly coherent; it only requires corrigible foundations. Plausibly, this is the
method Chisholm endorses.
My position is somewhat inspired by that of Aristotle’s notion of the

“endoxa”: the “reputable opinions” or “accepted opinions” or “cred-
ible beliefs.” Aristotle says this about the sources of endoxa. “Endoxa
are those opinions accepted by everyone, or by the majority, or by the
wise – and among the wise, by all or most of them, or by those who are
the most notable and having the highest reputation” (Top. 100b21–23).
Admittedly, as stated this is a bit of a grab bag. But it contains insight
into several degrees of credibility when beginning to theorize. And the
fact is, our theorizing must start somewhere. In particular, as Shields
points out, the endoxa are a species of “phainomena” or appearances
(Shields cites EN1154b3–8). As Shields puts it, for Aristotle, “we begin
with how things appear to us” (Shields 2007: 26). So the method blends
reliance on one’s own intuition or common sense with those of one’s
community.2
There are two ways appearances can gain credibility: through quantity
or through quality (or both). The more widespread assent there is to an
appearance, the more credible it is. And the more support the appearances
have from the wise, the more credible they become. It is important to
screen for appearances that can be seen through after extended reflection.
So for Aristotle, there is no more natural place to start than the endoxa
and an attempt to try to “save the appearances.” They can, however, be
overturned if philosophical or scientific reflection demands it (Shields
cites Meta 1073b36, 1074b6 and PA 644b5 and EN 1145b2–30).
A similar epistemology was defended in the 18th century by Thomas
Reid; a Scottish contemporary of the more skeptical David Hume. Reid
is considered one of the founders of the modern “common sense” tradi-
tion. Like Aristotle, he begins from and is generally hedged by common
sense and ordinary language, but he does allow that it is not always
correct (Yaffe and Nichols 2009). (For latter-day defenses of credulism
distinct from Reid’s but in his spirit see Chisholm 1977 and Swinburne
2001.) My position is that this epistemological tradition is largely
correct and that the burden of proof is on the individual who contra-
venes common sense. In the present case, as I shall argue, this means
that those who deny the existence of morally significant animal pain
bear the burden of providing powerful enough arguments to overturn
the weight of the endoxa. The remainder of this chapter will state the
case that the common sense view of animal pain is indeed an illusion.
The next chapter will argue that it is mistaken.
2
For interesting takes on the relationship between self-trust and trust in others,
see Foley (2007) and Zagzebski (2012).
Dennett (1978), Nelkin (1986), and Nelkin (1994) agree that the common
sense notion of pain may have to be compromised to some degree or
another. Dennett practically gives away the shop (see Conee 1984 for a
reply), but Nelkin offers a more measured approach, though he does give
up the notion that to hurt is essential to being in pain. In fact, he goes on
to say, after considering the variety of pain experiences, “There is nothing
obviously alike in all pain phenomena, if we consider only the phenomenal
states in themselves. They all hurt, of course; but the question is whether
the hurt of these states lies in their phenomenal ‘feeling’ ” (Nelkin 1994:
329–330). I confess I simply lose my grip on the subject when people start
talking this way. There is a way to hurt that is independent of how some-
thing feels? Even if there is some sensible interpretation of this language,
things have gone awry when you have to start speaking that way.
We will look at this in detail below, but the idea here is to get before
our minds how much of a departure from common sense we are
willing to tolerate. Neo-Cartesianism, though in no way logically tied
to Cartesian dualism about mind, has provoked strong statements by
philosophers of mind. Dretske asks “[W]hen a dog scratches, are we to
believe that the itch is not conscious, or that the dog’s experience is
totally different from ours ... ?”3 Alastair Norcross (2012) even goes so
far as to refer to “neo-Cartesian Silliness.” Clark (2013: 531) does him
one better: “Anyone acquainted with non-human animals, whether in
a domestic, a laboratory or an agricultural setting, and still more in the
wild, is likely – unless they are behaviourists or psychopaths – to act on
the assumption that animals have feelings and a point of view.”
When we step on a dog’s or cat’s tail, and they give their characteristic
shrill yelp or shriek, common sense issues a clear verdict: The animal is
in conscious pain.
And I must say I find the necessary phrase “conscious pain” redun-
dant. Consider this strong statement by the International Association
for the Study of Pain (IASP).
Note: Pain is always subjective. Each individual learns the appli-

cation of the word through experiences related to injury in early
life. ... Experiences which resemble pain, e.g., pricking, but are not
unpleasant, should not be called pain. (IASP 1986: 250, as quoted in
Aydede 2009)
3
Dretske (1995: 110–111). Thanks to Beth Seacord for bringing this quote to my
attention in her excellent and detailed “Higher-Order Theories of Consciousness
and Animal Experience” presented at the Center for Philosophy of Religion,
University of Notre Dame, Spring 2012.
I think Kripke had it just right.
To be in the same epistemic situation that would obtain if one had

a pain is to have a pain; to be in the same epistemic situation that
would obtain in the absence of pain is not to have a pain ... . Pain ... is
not picked out by one of its accidental properties; rather it is picked
out by its immediate phenomenological quality ... . If any phenom-
enon is picked out in exactly the same way that we pick out pain,
then that phenomenon is pain. (Kripke 1980: 152–153, as quoted in
Aydede 2009)
The concept pain, it seems to me, includes the property being unpleasant.
I simply cannot make sense of a notion of a pain that doesn’t hurt.
There are no doubt some painful sensations that are very nearly indis-
tinguishable from pleasant sensations (there may even be indeterminate
cases) which accounts for my love of very spicy food. It could be that
I’m rewarded with endorphins for the experience, which would give a
certain positive association to the experience even if the core experi-
ence were intrinsically unpleasant. The reader is referred to Dennett
(1988) for an interesting discussion of these issues. And there are no
doubt complex sensations that involve some painful sensation as essen-
tial to the complex and the complex is on balance pleasant. This latter
phenomenon is apt to be misdescribed as a “pleasant pain” especially
when the pain and pleasure are of quite different types as when a “phys-
ical” pain (a pain clearly associated with a physical state) gives rise to a
mental pleasure: that familiar ache of the legs during or after a strenuous
hike or of the arms during or after a hard climb.
Ultimately we must distinguish a number of issues, given the IASP
definition of pain (even when expanded). I think it wise to work with a
philosophically informed version of their definition rather than to turn
to the philosophers as a primary source for the notion of pain. Much
research is currently underway on the nature of pain,4 but philosophers
have previously been fixated almost exclusively on the epistemology of
pain and its relation to the mind–body problem. It is astonishing that
4
At the time of writing, I have just presented at the Nature of Pain: Hedonic
Tone, Motivation, and Non-Human Animals project at the University of Glasgow,
part of the Pain and the Nature of Minds project run by the Center for Philosophy
of Religion at the University of Notre Dame. Millions of dollars were allocated to
study the nature of pain, with considerable attention to non-human animals.
a philosopher would have to say the following sort of thing. “[T]here

is an increasing recognition of the fact that pain has an affective and
motivational aspect which seems at least prima facie distinct from its
sensory or perceptual aspect” (Aydede 2009). With only slight exaggera-
tion, the idea is that philosophers are beginning to realize that feeling
bad and being unpleasant are important aspects of the notion of pain
distinct from nociception.
So if we are going to work with something like the IASP definition,
we need to think about how to understand “an unpleasant sensory and
emotional experience.” This implies that the experience must be sensory
and emotional and unpleasant. I think the best way to understand this is
as a sensory experience with an unpleasant affect. (I will bracket entirely
as irrelevant the old puzzle about masochism.) This implies a state of
conscious awareness or, equivalently as I will use the phrase, phenom-
enal consciousness. That non-human animals have such states has been
challenged. That challenge is the subject of this chapter. But as I said
above, that challenge is to something that begins with a good deal of
warrant. Thus, the challenge has a considerable burden to bear. This
metaepistemological framework for my argument is controversial, so I
shall next give a brief defense of it.
As I said, the common sense approach does not rule out revision a
priori, but in this case, certain kinds of revision do in fact seem conceptu-
ally bankrupt. But that animals have conscious pain is not just a dictate
of common sense, it is confirmed by scientific consensus.
4.1.2 Scientific consensus

Given the common sense method just defended, if the endoxa include
that animals feel pain – especially if it is the opinion of most of those
in a position to know and with the best judgment and good will – then
that will establish the opinion as the epistemically proper starting point.
And in fact, as we shall see, this is precisely what is the case.
In 2009, the National Academies, the U.S. umbrella organiza-
tion that includes the National Academy of Science, the National
Academy of Engineering, the Institute of Medicine, and the National
Research Council, appointed 13 esteemed scholars to the Committee
on Recognition and Alleviation of Pain in Laboratory Animals. The
Committee produced a report aptly titled Recognition and Alleviation of
Pain in Laboratory Animals. The “report” is in fact a 181-page book. The
report was produced in concert with the National Research Council’s
Institute for Laboratory Animal Research, a specially selected group of 20
scientists and researchers. What makes the testimony of this committee
particularly valuable is the extent to which it is likely to be free from

biases. Given the amount of time and effort they put into reaching their
conclusion – all pro bono as the National Academies require – they clearly
have a concern for animal welfare. And given the fact that they do not
advise a wholesale rejection of the laboratory use of animals, they are
not mere ideologues. Furthermore, the committee is fairly subtle and
careful in their pronouncements. They conclude, “Although definitive
evidence is often unavailable, this report does not treat the absence of
evidence as evidence of absence. Instead, the consensus of the committee
is that all vertebrates should be considered capable of experiencing
pain” (National Research Council 2009: 23). This is based on their most
basic conclusion that “Current scientific evidence strongly suggests
that mammals, including rodents (the most commonly used laboratory
animals), are able to experience pain” (National Research Council 2009:
4). These carefully arrived-at, measured conclusions are signs of trust-
worthiness among people with no axe to grind.
The committee even makes a distinction between the evidential aspect
and the practical reasoning aspect of the problem. They note the nega-
tive consequences of being mistaken about animal pain and carrying
out experiments that might produce pain. This would justify the practice
of treating laboratory animals as if they experienced pain even if the
evidential probability were low that this was the case. However, they
conclude on the epistemic front that there is a “strong likelihood that
this is correct” and in particular that there is “compelling evidence for
rats” (and therefore all mammals above rats) feeling pain (National
Research Council 2009: 23).
They note that there are two main ways to assess whether a kind of
animal experiences pain. “The first is to demonstrate the presence of the
anatomy and physiology that appear to be a requirement5 for pain in
humans. The second is to investigate which species show responses to
noxious stimuli suggestive of pain”6 (National Research Council 2009:
22). Thus there will be two ways to challenge the professional judgment
that animals feel pain: 1. to point to a lack of the required anatomy
and physiology or 2. to provide a better explanation of the behavioral
evidence. As we will see, both these strategies are pursued by those who
deny that animals feel pain or wish to cast doubt on the proposition
that they do.
5
It’s clear from context they also consider it sufficient.
6
Especially how it interacts with endogenous painkillers and administered
analgesics.
I will not here investigate the neuroanatomical case in much detail,

since the details are many and all parties admit that our present state
of understanding is very limited.7 But one thing bears keeping in mind.
Issues in neuroplasticity make clear that though the brain has areas that
are specialized for certain functions, other parts of the brain can often
play those roles when need be. And indeed there is some evidence that
the areas of specialization for certain functions migrated over evolu-
tionary history. While non-human primates and non-primate mammals
often share versions of important neuroanatomy – though in less-highly
evolved forms – even reptiles and birds often have areas that plausibly
are homologous to important areas in the human brain, and so their
brains might perform the functions related to pain and the perception
of pain in different areas than in humans. In short, what we humans
do with, say, our prefrontal cortex, rats may well do with other parts of
their brain.
There are many other professional organizations that share the judg-
ment of the National Association of Science’s Committee Report. For
example, the Institute for Laboratory Animal Research produced a Guide
for the Care and Use of Laboratory Animals (National Research Council
1996) that indicates that the ability to experience pain is widespread
among animals. The American College of Laboratory Animal Medicine
(2001) issued a Position Statement on Pain and Distress in Laboratory
Animals that directs that “procedures expected to cause more than slight
or momentary pain (e.g. pain in excess of a needle prick or injection)
require the appropriate use of pain-relieving measures.”
There are countless professional organizations (not advocacy groups)
and government agencies (such as the United States Department of
Agriculture, the U.S. Department of Health and Human Services, the
National Institutes for Health, the Centers for Disease Control, and Food
and Drug Administration) that have either adopted the Committee
report’s guidelines or endorsed relevantly equivalent guidelines, proce-
dures, and conclusions (National Research Council 2009: Appendix B).
It seems clear, then, that the proposition that non-human animals
of at least the mammalian level and above experience pain relevantly
analogous enough to humans to present a legitimate welfare concern
7
As of this writing, Gary Varner’s (2012) book Personhood, Ethics, and Animal
Cognition is at the stage of galley proofs, and I have set up a “debate” between
him and Colin Allen for the 2013 American Philosophical Association Both the
state of the investigation and the complexity of the issues make my consensus
approach most prudent, given my defense of the general methodology.
is a part of the endoxa. Given the endoxic method I have endorsed

together with the seeming obviousness of the conclusion from daily life
with animals, this should be our default conclusion unless the default
is defeated. In the next section, I will consider defeaters that cast doubt
upon the foundation of our common sense judgments about animal
pain as well as threaten to undercut the scientific consensus.
4.2 A statement of the objection
4.2.1 Introduction
Descartes is often thought to have considered non-human animals to be
unconscious automata, like extremely complex wind-up toys. Probably
this was not his mature view, though some of those he influenced seem
to have thought this. Leibniz attributes to “divers Cartesians” (probably
Arnauld and Malebranche) the view that animals “are only machines.”
For his own part Leibniz thinks this thesis unnecessary since “properly
speaking, perception is not sufficient to cause misery if it is not accompa-
nied by reflexion. It is the same with happiness: without reflexion there
is none.” So though he grants, “One cannot reasonably doubt the exist-
ence of pain among animals” he nevertheless concludes, “animals, since
they do not reflect, are susceptible neither to the grief that accompanies
pain, nor to the joy that accompanies pleasure” (Theodicy, III.2508). He
thus does not consider animal pain of moral significance and so does not
take there to be a problem of animal pain component of the problem of
evil. Reid also considered this same distinction.
The power of the understanding to make its own operations its

object, to attend to them, ... is the power of reflection, ... . This reflec-
tion ought to be distinguished from consciousness, with which it is
too often confounded, even by Mr. Locke. I conceive, this is sufficient
to shew the difference between consciousness of the operations of
our minds, and reflection upon them; and to show that we may have
the former without any degree of the latter. The difference between
consciousness and reflection, is like to the difference between a super-
ficial view of an object which presents itself to the eye, while we are
engaged about something else, and that attentive examination which
8
Thanks to Sam Newlands for bringing this passage to my attention when I
presented this chapter at the Center for Philosophy of Religion, University of
Notre Dame, Spring 2012.
we give to an object when we are wholly employed in surveying it.

(2002: 58–59)
It will be important for the discussion below that this distinction

upon which Leibniz and Reid seem to agree is legitimate. There are
modern defenders of the kind of view that Leibniz defends. They are
most frequently called “neo-Cartesians” because they represent a more
moderate version of the thesis attributed to Descartes. It appears they
would have better been named “Leibnizians” but it is too late for that
now. Peter Carruthers is most closely associated with defending a version
of the thesis Leibniz asserts. Michael Murray has written the most
recent defense of the case for neo-Cartesianism9 about animal pain. I
will examine their arguments below. Their view clearly runs contrary
to common sense. As Clark says, “Almost no one in the mainstream
European tradition would have doubted that animals were sentient crea-
tures until the eighteenth century ... . So until the eighteenth century
the more usual commonsensical claim was that animals had feelings
and emotions much like ours” (2013: 533).
4.2.2 Kinds of neo-Cartesianism and how they function

epistemically
Murray considers views that progressively allow for more mental life on
the part of animals, yet each of which is sufficient, if true, to “solve” the
problem of animal pain by making it go away. Here are the three kinds
of views in successive order of permissiveness about the mental life of
animals (my taxonomy of the views is much different from his, but the
mapping is easy).
Type I: Animals lack any phenomenal consciousness at all.

Type II: Animals have phenomenal consciousness, but are not aware
that they have it.
Type III: Animals have phenomenal consciousness, and they are
aware of it, but they don’t have positive or negative affect/evalua-
tions of it.
Below, I will positively assess scientific evidence against these philo-

sophical theses and negatively assess Murray’s replies. Phenomenal
consciousness is consciousness with a certain phenomenal character or
9
Murray does not claim to be a neo-Cartesian, but he does defend it.
“feel.” If there is something it is likely to be in a mental state, then

that state has phenomenal consciousness. An entity has phenomenal
consciousness when it has experiences that exemplify phenomenal
concepts. Familiar phenomenal concepts are red, sour, shrill, bright, hard,
and painful. To have a phenomenal concept one has to have the right
kind of experience. For example, you can’t have the phenomenal concept
green without having a greenish experience. You can have an ordinary
concept green, of course, for even a colorblind person can have the ordi-
nary concept. Indeed, someone colorblind might know more about the
color green and colors generally than someone who is not colorblind.
Yet that colorblind individual does not know what it is like to see green
and thus lacks the phenomenal concept green.
According to Type I neo-Cartesianism – which I will call the “phenom-
enal blindness” theory or just “the blind view” for short – an animal’s
informational access to its environment is wholly unmediated by
conscious states that have phenomenal character. The blind view has
little going for it when it comes to creatures that most clearly display
sentience, so I will focus my attention below on the latter two kinds
of neo-Cartesianism. Nevertheless, since the view is represented in the
literature, I will give it cursory treatment.
Functionalism in the philosophy of mind is a fairly widespread view,
and according to (at least one version of) it, whether a state exempli-
fies phenomenal consciousness depends on whether the organism has
a further mental state that takes the first as its object. So according to
the Higher-Order Thought (HOT) theory of phenomenal consciousness,
if an organism can’t have such a second-order thought, then it can’t
have phenomenal consciousness. Thus if one denies that animals have
HOT, then one will have a nice modus tollens argument against animals
feeling pain, given that pain should be a state of phenomenal conscious-
ness. This view seems to radically over intellectualize the phenomenon
of phenomenal consciousness, but we will get the other views on the
table before drawing a conclusion.
According to Type II neo-Cartesianism, though animals have conscious
mental states with phenomenal character, such as pain or pleasure, they
aren’t aware that they have them. They have experiences of the relevant
sort, but they are oblivious to that fact. I’ll call this the “obliviousness
thesis” or just “the oblivious view” for short. What this thesis asserts
will sound paradoxical to many a reader, but the phenomenon is wholly
implausible.
How could one be in a state of pain and not realize it? Plausibly the
following is metaphysically possible. You are busy looking at a map,
intensely trying to figure out where you are. Suddenly, you realize that
you are being bitten by a mosquito and it hurts! Not only that, but you
realize that the pain was mounting during the seconds preceding the
crossing of a threshold; you just weren’t paying attention to it. Similarly,
you might suddenly realize that for some time you’ve been hearing an
alarm go off or that a light on your office phone has been blinking.
Prior to turning your focal consciousness on these experiences, you were
still having them. You hosted phenomenal red, heard the sound of the
alarm, and felt the pain before you realized you were in pain. You were
oblivious to it. It is at least plausible that this occurs.
According to Type III neo-Cartesianism, animals have phenomenal
consciousness, and they are aware of it, but they don’t care. If they have
“pain” it doesn’t bother them. This is because they lack the ability to
form pro- or con-attitudes toward these feelings. I will call this “the numb
view.” What all these kinds of neo-Cartesianism have in common – the
blind view, the oblivious view, and the numb view – is the thesis that
non-human animals lack some key mental capacity, and that this lack
prevents them from experiencing morally significant pain.
I will argue against neo-Cartesianism as follows. Neo-Cartesians have
offered philosophical and neurological arguments for neo-Cartesianism.
Each of the three types of neo-Cartesianism is supported by arguments
for retentive theories of consciousness. The numb view depends espe-
cially on neuroanatomical arguments that we will discuss below. My
arguments for more liberal theories of consciousness tell against all three
types of neo-Cartesianism. I will cast doubt upon the alleged neuroana-
tomical differences between human and non-human animals, which
are used in support of the numb view. HOT theories of consciousness
offer a rebutting defeater for our endoxic starting point by providing the
basis for a simple argument against animal consciousness of the relevant
sort: They do not meet a necessary condition for having it. The neuro-
anatomical arguments offer both rebutting defeaters of an analogous
kind – sub-human animals lack neuroanatomical features required for
the relevant kind of consciousness – as well as undercutting defeaters –
hard cases that show that common sense could be misleading and that
the observational bases for the scientific consensus could be mistaken.
Murray (2008) does not argue for any neo-Cartesian view. Rather, he
simply says “As one’s commitments in the philosophy of mind go, so
will go one’s stance with respect to the possibility of the [neo-Cartesian]
explanations” (2008: 58). Yet he just defends the view as at least plau-
sible, and it seems to play an important role in his theorizing. Similarly,
Goetz (2009) asks, “Does any one of us have adequate knowledge of a
fawn’s psychology?” (494). Since the answer to this question is “no,”
Goetz argues that we cannot decisively conclude that there is a problem
of animal suffering. The thinking in both of these cases seems to be

something like this.
1. Here is a possibility you cannot rule out: neo-Cartesianism.

2. If neo-Cartesianism is true, then there is little or no morally signifi-
cant animal pain.
3. Thus, here is a possibility you cannot rule out: that there is little or no
morally significant animal pain.
This method of argument leaves a lot to be desired. On the weak view of

epistemic possibility I favor,10 all the conclusion entails probabilistically
is that the probability that there is little or no morally significant animal
suffering is greater than zero. That is consistent with it being less than
1% likely. On the standard view of epistemic possibility, all that follows
from the conclusion is that we don’t know that there is any morally
significant animal pain. If we use, say, .95 as a standard for knowledge,
the argument is of very limited value, because it is consistent with it
being 90% likely that there is morally significant animal suffering. And
if there is, and there is no theodicy forthcoming or no support from
natural theology or religious experience, then that spells trouble for
theism. For if it is 90% likely that a state of affairs obtains that is incon-
sistent with theism, then, ceteris paribus, it is 90% likely that theism is
false. This would make it very reasonable to affirm atheism, even if it
does not allow the atheist to know atheism is true. (And it is cold comfort
even if this entails that one doesn’t know there is no God.11) And of
course it puts agnosticism in a good position.
Here is how I suggest the argument would need to go to make things
interesting. It would be a simple modus ponens beginning with Premise
2 above.
Neo-Cartesianism entails that animals don’t feel morally significant

pain.
Neo-Cartesianism is true.
Thus animals don’t feel morally significant pain.
I am told Chisholm used to say that the probability of the conclusion is

“in the line you draw between the premises and conclusion.” In other
words, in my interpretation anyway, you present a deductive argument
10
Dougherty (2011b).
11
Most of van Inwagen (1996a) seems to play at this level.
and then you assign probabilities to the premises and “turn the crank.”
Since entailment preserves probability without loss, the probability of
the conclusion will just be the probability of the affirmed antecedent
(we can assume we are certain of the conditional). That is, in an entail-
ment, the probability of the consequent is always at least as high as the
antecedent. Thus, the probability you “get out” of an entailment via a
modus ponens is the same as what you “put in” with the affirmed ante-
cedent. Below, I’ll consider reasons in favor of the second premise of the
argument just above.
In what follows, I don’t intend to “rule out” neo-Cartesianism. I intend
to drive its probability down for the reader as low as I can make it go. My
efforts will be met with mixed success. Some readers will not find the
higher-order theory of conscious awareness very plausible to begin with.
For such readers my goal is to drive that probability down farther or to
help keep it low. Some readers might start out thinking it plausible or
highly probable. For such readers my goal is the same. The same goes for
the hard cases Murray uses to raise doubts about the existence of animal
pain. I want to drive down, as much as I can, the probability that those
cases present a problem. Binary concepts like “proving” or “ruling out”
are not part of my methodology.
4.2.3 Some motivation for neo-Cartesianism

We should begin by deepening our previous understanding of some basic
terminology in the neurology of pain. As mentioned previously, I prefer
to use the term “pain” more broadly than to refer to “physical” pain, but
will frequently set this preference aside and use “suffering” and “harm”
for the broader notions. Thus, I will usually stick to using “pain” for the
physically based pains (which, I claim, are ultimately psychological).
But even when focusing on physical pain, the IASP definition of pain
is too narrow. Here it is again: “An unpleasant sensory and emotional
experience associated with actual or potential tissue damage, or described
in terms of such damage.”12,13 I have a colleague who has a crushing
handshake. His hand is rock hard and like a vice! It hurts when he shakes
my hand. Nevertheless, he is a basically benevolent fellow and there’s
12
I initially thought this ruled out phantom pain, but the last clause plausibly
captures it.
13
Why have I called this a definition of “physical pain” if it involves emotional
suffering? Much more will be said of this below, but the taxonomy is essentially
etiological. The salient fact about the IASP definition in this regard is that the
root notion is tissue damage. Emotional suffering can also be caused in ways not
relevantly related to tissue damage.
never any real threat of tissue damage to my hand. (This could possibly
be accommodated by changing “potential” to “threatened” for it seems
my body interprets his grip as a threat.)
More significantly, consider being very cold or extremely fatigued
with a headache, or having abdominal cramps from diarrhea. These are
all physically bad states that don’t necessarily involve tissue damage or
the threat of tissue damage. You might want to put them in the category
of “discomfort” rather than “pain,” but if they get bad enough, that will
seem like a mistake. The reason I think this is significant is that I suspect
that such bad states of affairs – fatigue, hunger, cold/heat – constitute
a significant portion of animal suffering that is physically based (the
contrast class here is the possibly worse, more paradigmatically psycho-
logical suffering, such as the loss of an offspring or mate). However, I
have no intention of trying to Chisholm the definition of pain until
a true and explanatory set of necessary and sufficient conditions are
obtained. Rather, I will rely upon the extension to guide the intension.
That is, the examples of suffering themselves should be taken as primary,
not the attempts to state what they all have in common (and, indeed,
there may be nothing they all have in common, for it might be that they
are related by a web of family resemblance, which, not being transitive,
does not guarantee that there is at least one thing each member shares
in common). Nevertheless, jargon is efficient, so I need to use it, and to
do so without misleading, I need to say where I think artificial precision
is dangerous.
Another important term is “nociception.” The IASP definition is “the
neural process of encoding noxious stimuli” where a “noxious stimulus”
is “a stimulus that is damaging or threatens14 damage to normal tissues.”
We see reflected here some of the same narrowness as in the IASP defini-
tion of “pain.” However, as long as we keep this in mind, we can harm-
lessly adapt. We must be a bit careful, however, because some of the
terminology is somewhat more importantly specialized. For example, a
“nociceptor” is “a high-threshold sensory receptor of the peripheral soma-
tosensory nervous system that is capable of transducing and encoding
noxious stimuli.” This makes reference to a particular set of neurons.
This leads to a legitimate distinction between “nociceptive pain” and
“neuropathic pain.” The former, naturally, is pain arising from the trig-
gering of nociceptive neurons, the latter arises from damage or disease
to the somatosensory nervous system. That is, it ultimately reports what
14
Note that here the contrast with the actual is the threatened, which supports
my proposed modification above.
is going on inside the body, not what is happening to the body from
outside of it. It is not, strictly speaking, part of the sensory apparatus.
Thus, we should keep in mind both a broad sense of nociception – the
neural process of encoding any realization of negative physical affect –
and the narrow sense of nociception provided in the IASP definition.
4.2.3.1 From philosophy: HOT of consciousness

Just above, I used the term “higher-order theory,” but I should have
spoken of higher-order theories, for there is a family of such theories. To
begin with, higher-order thought, or HOT theories of consciousness are
but one kind of higher-order theory of conscious mental states. There
are other higher-order requirements one might impose as a criterion
for conscious experience. A higher-order or HO theory might require a
specific kind of thought – belief, say – or a specific kind of other mental
state – a perception, say – but most generally, a higher-order theory of
phenomenal consciousness will be one that imposes a condition that a
first-order experience be the object of some form of mental state of the
organism.
Though I am not wholly convinced that any HO requirement is neces-
sary, I do recognize the intuitive pull. Thus, though I will defend an
anti-HO stance in the sleepwalker case in the next chapter, my primary
task in what follows will be twofold. First, I will raise some doubts about
strong HO theories like the HOT. In general, I don’t find strong HOT theo-
ries very plausible, so I won’t spend much time rebutting them. However,
since, as I say, I do feel the general intuitive pull of HO theory, I will
consistently attempt throughout the next chapter to argue that much
weaker HO states are sufficient to satisfy the intuition. Now we’ll look
more closely at the HOT theory of consciousness and how it is applied.
Imagine two people hiking an old logging trail in the forest, one in
each old tire track, engaged in an intense conversation about the nature
of consciousness. As they debate the merits of opposing views, they step
over downed sticks, over larger rocks, and around puddles and muddy
patches. It seems that they are aware, in some sense, of these obstacles as
they are walking, even though they are “paying no attention” to them.
One interpretation is that they have no phenomenal consciousness of
these items at all. That is, though light enters the retina and precipitates
signals that activate the visual cortex, there is never any “image” of the
rocks and sticks on the “screen” in the mind. If this happens, it seems to
be a case of unconscious experience, where “experience” is, perhaps, used
loosely, since there is no sensuous experience of them by the individuals.
Yet it is in some way a case of sense perception.
If a reader balks at this point – on the grounds that an “unconscious

perception” is a priori impossible – then the rest of the discussion might
make little sense, for the whole point of HO theories of consciousness is
to say when certain mental states count as conscious. I am not wholly
without sympathy for this position. For “unconscious perception” is
ambiguous between two things. It can mean “sensory experience to
which one paid no attention.” Or it can mean “taking in information
from the visual sensory organs and processing it in the visual cortex of
the brain without any phenomenal consciousness.” The former kind
of case already presupposes some phenomenal consciousness. I think a
good example of this is that which lies on the very edge of our periph-
eral vision. The latter stretches the meaning of “perception.”
I think an ordinary reader is apt to think, and I cannot blame them,
that mere visual or auditory neuroprocessing does not count as a form
of perception or awareness. Let us then take “mental state” as a neutral
term. If one thinks there can be unconscious mental states, then we
can distinguish two kinds of considerations relevant to philosophi-
cally motivated neo-Cartesianism. If you are a functionalist about
mental states as most naturalistic philosophers of mind are, then let
the following discussion be about when mental states – such as cortical
processing of signals from distal sense organs – count as conscious. If
you think that for a state to count as a perceptual state or a state of
awareness it must have phenomenal consciousness – the perceiving
individual must token some phenomenal concepts – then consider this
discussion to be about what makes a state of phenomenal conscious-
ness conscious enough to make it morally relevant. Just for convenience,
let us refer to states of awareness with phenomenal content that we
don’t pay any attention to as “semi-conscious.” I don’t mean to be
ideological about it, so the “semi-” should be harmless. For balance, I
will use the term “perceptual mental state” rather than “perception”
when no phenomenal consciousness is present. I think it is an appro-
priate name because there is stuff going on in the mind as a result
of activation of the perceptual system. So even the individual who
thinks that conscious entails phenomenally conscious will have interest
in our discussion, except that they will be focused not on the ques-
tion, “What makes a mental state (in the information processing
sense) phenomenally conscious?” But rather, on the question, “When
is a state of phenomenal consciousness morally relevant?” It is not
implausible that much of the discussion on lobotomy cases discussed
below is subject to this ambiguity, so the reader will do well to keep it
in mind.
Explaining why some perceptual mental states have phenomenal

content and some do not is a central problem in the philosophy of
mind. HOT theories provide a tidy, naturalistic explanation in terms of a
widely held view: functionalism. The kind of mental state the literature
often focuses on to illustrate “perception” – in the relevant sense that
is unconscious – is blindsight (Carruthers 1989 and 1996). Armstrong’s
early case (1968) of absent-minded driving can be interpreted either
along the lines of my example of ignored peripheral vision or might be
called “zombie driving” (the situation I described in the hiking in the
woods case above could be called “zombie walking”).
So we have two kinds of cases to consider. One is “blind pain” that
consists of mere nociception, activation of pain receptors, communica-
tion to the central nervous system, and appropriate motor response (e.g.
aversive response) – sometimes this requires a signal from the brain,
sometimes it is activated from somewhere prior to the basal ganglia –
without any phenomenal consciousness. This kind of thing has no
obvious moral relevance. This sub-personal state must be the object
of some kind of attentional state in order to be conscious in a sense
warranting moral importance. This seems quite plausible.
The second kind of case is one where there is an experience with
phenomenal consciousness, but one pays no attention to it or is unaware
of it for some reason (assuming this is possible). It is harder to discern
whether pain (actually registered phenomenally, not just nociceptively
and cortically) of which one is unaware is morally relevant. I am gener-
ally inclined to think that here too, there probably needs to be some kind
of higher-order mental state that takes this state as its object. However, I
think it needn’t be a very high level of awareness. Furthermore, it is not
obvious that there needs to be any higher level of awareness for moral
significance, which is a point I’ll press in the sleepwalker case below.
Below, I will be making a persistent case that if a higher-order require-
ment is imposed, it ought only be required to be the object of a less
“fancy” mental state than thought or belief and, rather, only some other
kind of perception or awareness (de re at that). Presently, therefore, I
only want to focus on one direct critique of a strong HO theory, such
as HOT. Consider the immense complexity of our sensuous experience.
It is mind-boggling to contemplate. I think of this weekend when I was
watching a group of waterfowl consisting of two kinds of ducks, some
geese, and some swans vying for breadcrumbs on a rocky shore near
twilight. The sun reflected off the water at innumerable angles, off the
heads and wings of the ducks with innumerable colors; the waterfowl in
their various shapes in constant motion on the constantly moving water.
To think that for each conscious experience there would have to be a

corresponding thought attributed to a sentient being having this experi-
ence is to suppose an incalculable number of thoughts. In part due to a
problem like this, Carruthers moves to a dispositional HOT theory. But it
is mysterious – extremely mysterious I would say – how mere potential to
be observed could account for the subjective character of phenomenal
consciousness (See his 2011 for his attempt at a semantic reply). This is
not, of course, an objection for which advocates have nothing to say.
But it is one reason to hold it suspect and thus to seriously consider the
options I present below.
4.2.3.2 From neuroanatomy: the lobotomy argument

There are some strange cases that provide some reason to doubt
the dictates of common sense regarding non-human animal pain.
Philosophers, of all people, should be prepared to revise common sense
if it cannot be reconciled with hard cases. There are two such cases – both
used to raise neo-Cartesian doubts about animal pain by Murray – which
are particularly problematic. Consider, first, the case of some patients
who have received frontal lobotomies.
Some patients given frontal lobotomies are said to report that they still
feel their pain, it just no longer bothers them and they do not express
a desire to take measures to “relieve” their pain. The phenomenon in
question is essentially what can be called “blind pain.” We will consider
it more below in conjunction with Murray’s sleepwalker example, but
it is a good way to introduce the current suggestion. A similar phenom-
enon also reportedly exists in cases of “morphine pain.” This is where
patients with serious injuries are given a certain dose of morphine, and
as a result – much like the lobotomy patients – say that they still have
their pain, but they no longer mind it. We will face interpretive issues
when we look at the details more closely below, but notice that, in
conjunction with some empirical claims about non-human brains, there
are the makings of an argument for neo-Cartesianism.
The lobotomy argument
1. Lobotomized patients have pain but form no con-attitudes about it.

2. Animals are relevantly like lobotomized patients in that they lack a
rich frontal lobe.
3. Therefore, animals, even if they feel pain, form no con-attitudes
about it.
One can then use 3 in an argument like this.

4. Even if animals feel pain, they form no con-attitudes about it.

5. If pain does not give rise to con-attitudes, then it is not morally
significant.
6. Therefore, even if animals feel pain, it is not morally significant.
This argument, if successful, would establish a Type-III neo-Cartesianism.

More ambitious arguments seem to me not much worth pursuing. The
evidence for Premise 2 comes from the fact that, indeed, the frontal
lobe of the mammalian brain, especially the most highly functioning
portion, the prefrontal cortex, is a very late evolutionary development,
almost the exclusive domain of latter-day hominids. Premise 4 can be
doubted, but I shall have very little to say about it.
I will challenge Premise 2 first by examining some of the relevant neuro-
science. Then I will challenge Premise 3 – or perhaps the inference from
1 and 2 to 3, depending on how we read “relevantly” – by exploiting an
ambiguity in “con-attitude.” My challenge to Premise 2 is not to deny that
non-human animals, especially non-primate mammals and below, lack
the rich frontal lobe of humans and higher primates (there is a significant
step in complexity even from higher primates to humans), but rather to
deny or at least cast doubt on the relevance of this difference (though I
will also be pointing to important brain similarities and underscoring the
uncertainty of the claims about the brain involved).
I have taken a stand on the side of common sense epistemology,
granting it the benefit of the doubt. Nelkin notes, however, that “cases
of prefrontal lobotomy patients and of patients given morphine prior to
the onset of pain, however, call our common-sense intuitions into ques-
tion” (Nelkin 1994: 326). Again the two kinds of cases are very different
in their nature but almost exactly alike in their results: “The manifes-
tations of lobotomy and morphine are similar enough to lead some
researchers to describe the action of morphine (and some barbiturates)
as ‘reversible pharmacological leucotomy [lobotomy]’ ” (Dennett 1978:
432). Helm agrees that such cases “threaten to undermine our sense that
we have any clear concept of pain” (Helm 2002: 27, n.2).
These phenomena, then, might be thought to present a very serious
challenge to my argument, since I have foregone a presentation of the
traditional analogy argument for non-human animal pain.15 However,
15
As with the argument for the existence of minds in other humans, the
analogy argument – or, more broadly, any discursive inference – is subject to
many criticisms. For treatment of this problem see Allen (2004) and Allen et al.
(2005). For a reply, see Varner (2012).
these cases are actually quite problematic: They are not very well under-
stood and there are multiple interpretations of their meaning.
The only basis I have been able to find for affirming that lobotomized
and opiated individuals really do feel pain is that they use some pain
terms like “sharp,” “throbbing,” and the like to describe their experi-
ence. While under normal circumstances we invest considerable trust
in people’s reports of their experiences (indeed, some strong founda-
tionalisms hold such introspections to be infallible), these cases are far
from normal. Why take the word of someone lobotomized or opiated
on this matter? For lobotomized individuals especially, we know that
their emotional and affective abilities are dulled. This could lead to
misleading answers on the standardized questionnaire used to gather
the results (Melzack 1975: 277–299).
Also, anyone who has had a significant medical or dental procedure
under a local anesthetic can recall that one can feel many different kinds
of pressure during such procedures that lend themselves to the same
terminology used by those in pain: “sharp,” “dull,” “throbbing.” When
I had a severely broken arm/wrist, the paramedics immediately gave me
a huge dose of morphine. When they cut my biking glove off, I could
easily have described my experiences along several dimensions with
pain language. But it wasn’t pain, it was just a form of pressure with
pain-like properties amenable to description in pain language.
Though Nelkin (1994: 333) reverses his 1986 hypothesis that the
language practitioners take to be descriptive of phenomenal states are
in fact of affective states of the patient, he admits it is ambiguous, and I
think the reversal is a mistake. I suggest that until these phenomena are
better understood, what seem like clear conceptual connections should
be retained rather than rejected because of some odd cases.
Let me put this suggestion in the language of the epistemolog-
ical foundations I laid out in the second section. I suggested that the
endoxa – in this case the common sense attribution of pain to animals
plus the consensus of core (fairly non-ideological) animal welfare
researchers – receive a benefit of the doubt and that skeptics – in this
case, neo-Cartesians (whether philosophically or scientifically based) –
bear a burden of proof. I was clear that philosophical reflection or scien-
tific discoveries were most definitely capable of overturning the endoxic
view. However, for that to occur the anomalous phenomena would have
to be well understood. This is far from the case with morphine pain,
lobotomy cases, and blindsight. On the contrary, there is considerable
uncertainty about the underlying mechanisms at work and much uncer-
tainty about how consciousness is realized in the brain.
5
There Is a Problem: The Defeat of
Neo-Cartesianism
In the last chapter, I stated my methodology of endoxa: starting from

the assumption that common sense and learned opinion get the benefit
of the doubt. Because this is defeasible epistemic support, the support
could be lost if there were some reason to think that the endoxa were
misguided. The methodology was applied thusly: (i) it is simply a datum
of common sense that (a significant number of) animals experience
morally significant pain and (ii) that they do so is also the consensus
of concerned scientists. Neo-Cartesianism serves as a potential defeater
for this defeasible justification. I then delineated some forms of neo-
Cartesianism and canvassed some reasons to accept neo-Cartesianism
from both philosophy and science.
In this chapter, I remove the threat of defeat in two ways. First, in Section
5.1 I will provide what are essentially undercutting defeater-defeaters. That
is, I will form considerations in light of which the potential defeater of neo-
Cartesianism is no longer a threat, even if its fundamental assumptions
are correct. I will also argue that some fairly foundational assumptions
of neo-Cartesianism are false. In Section 5.2, I will promote a version of
the problem of animal pain that is focused on primates in particular, thus
making an end-run around much debate about animal consciousness.
5.1 Initial reply to neo-Cartesianism: moving away from

rationalism
Neo-Cartesianism tends to rely on rationalistic pictures of self-knowledge.

In the main two subsections of this section, I present two models of knowl-
edge of pain that avoid the rationalistic assumptions of neo-Cartesianism.
In the first, Section 5.1.1, I present recent research on pain that presents it
as more emotion-like. In the second, Section 5.1.2, I explore the application
77
of certain ideas from linguistics for how they can yield a model of knowl-
edge of pain that avoids neo-Cartesian rationalistic assumptions.
5.1.1 Pain as emotion-like

The lobotomy argument as I have presented it is similar to the HOT
attack from the philosophical flank of neo-Cartesianism, and thus much
of what follows will apply to the HOT strategy as well. It depends upon
an over intellectualizing of the pro and con attitudes associated with
pain. Without the more “cerebral” parts of the brain – the highly evolved
neofrontal cortex – animals can’t, the story goes, make the kinds of eval-
uative judgments required to consider their pain bad and thus suffer.
Both philosophically motivated neo-Cartesianism and neuroanatomi-
cally motivated neo-Cartesianism share a common structure.
1. If animals lack X, then even if animals have “pain” states, they cannot
form pro/con judgments about them.
2. If animals cannot form pro/con judgments about their “pain” states,
then those states have no moral significance.
3. Animals lack X.
4. Therefore, animals’ “pain” states have no moral significance.
Whether the X-factor be higher-order thought or the prefrontal cortex

(or both), the result is the same: Animal pain is, essentially, “zombie”
pain. That is, from a moral point of view, no pain at all.
There are many moves – both philosophical and neuroanatomical –
for and against neo-Cartesianism. I propose to short-cut this debate
by making neo-Cartesianism irrelevant in two ways, both concerning
how we think of animal suffering. First, I wish to change the focus of
animal suffering from the physical-cognitive to the emotional. I want to
do this in two ways. First, I endorse and defend the thesis that pain is
“emotionlike” (as Gustafson 2005: 238 puts it) and apply it in the case of
animals. Second, I want to focus on emotional pain like fear, stress, etc.,
as it occurs in animals. Suffering conceived in both these ways does not
require the same kind of cognitive or neuroanatomical endowments as
physical pain on the cognitive-perceptual model.
The second way I want to change the way we think about animal
suffering is to broaden the notion to include animals’ lack of flour-
ishing. Think of a fulfilled non-human animal1 life. It includes many of
1
This applies to at least mammals and above, but plausibly to some birds as
well, and possibly to some other non-mammal vertebrates.
There Is a Problem: The Defeat of Neo-Cartesianism 79
the same features as the good life for a human animal: shelter from the
cold and heat, companionship, plenty of food and water, leisure time to
play, etc. Probably, a very low percentage of non-human animals thrive
in this way; few achieve their natural telos. Importantly, this kind of
suffering doesn’t depend on one’s reflective or perceptive abilities. But
this is puzzling if the world is created by an all-powerful (and so all-
knowing and all-loving) Being. Even when we move from bare theism to
Christian theism, one traditional explanation – the Fall – seems muted.
There is no corresponding puzzle on Naturalism.
The basic idea behind the strategy I will describe, endorse, and expand
is that the plausible thesis that neo-Cartesianism takes too far – that
pain states must be the object of some kind of awareness – can be satis-
fied in ways more plausibly fulfillable by non-human animals. I will
suggest that this is plausible due to the availability of more direct and
non-conceptual modes of knowledge.
Nelkin, a leading pain theorist, opines that “There must also be a
cognitive state which involves an evaluation of the phenomenon as
something like, ‘Harm to the body,’ ” and thus agrees with a basic intui-
tive tenet of neo-Cartesianism. However, he goes on to lower the bar:
“The evaluation is a kind of de re belief, regarding the phenomenon”
(Nelkin 1994: 325). This kind of evaluation is said to be a “sponta-
neous, non-inferential evaluation of that state as representing a harm”
(332). He goes on, when considering the consequences of his evaluative
theory of pain, to suggest that only a very low-grade awareness of self is
required for the “evaluations” that form the core of his theory of pain.
Basically, he thinks any animal capable of being aware of the external
world as external implies a basic concept of self, enough, in his view,
to satisfy the higher-order awareness requirement of his own theory
of pain (Nelkin forthcoming). “Seeing as” is an immediate, intuitive
mode of knowledge not requiring the explicit application of concepts.
Research on seeing-as petered out in the 1970s for the most part as an
object of independent study (Polanyi’s (1967) “tacit knowledge” has
not, sadly, found much of an audience in contemporary Anglophone
philosophy) but has survived somewhat in the literature on emotion,
to which we will return below.
With respect to the consequences of his theory for animal conscious-
ness, Nelkin considers this objection to his theory:
A second apparent difficulty for the evaluative theory goes as follows:

pains are said to consist of a CS state and a C2 state. But surely C2 is
an evolutionarily sophisticated state that we human beings possess
only because of our large neocortex. But many non-human animals

possess very little, if any, neocortex, yet feel pain. So the analysis
cannot be right. (1994: 337)
He makes a fascinating proposal: “The mistake, then, must be in

thinking that such reflective states are highly sophisticated. I would
predict, instead, that introspection must be, in itself, a fairly primitive
state, appearing at least as early as the first creature that felt pain” (1994:
338). I will pursue a strategy much like this.2
Human introspective ability might be more sophisticated than that of
creatures located lower on the phylogenetic scale, but the mere ability
to focus attention on inner states, perceiving them as inner rather than
outer, does not clearly require neurological complexity beyond that of the
mammalian brain. Nelkin makes a convincing case that we are thrown
off by thinking non-human introspection must be as rich as human
introspection. He points out that this would be as misguided as thinking
that non-human vision would have to be as rich and sophisticated as
human vision. The latter we know not to be the case, and it is plau-
sible that the ability to attend to one’s inner states developed gradually
after a simple appearance. Alternatively, the nature of the “perceptual”
aspect of pain could be of even less sophistication. Fernandez and Turk
(1992) suggest, “pain may be less like vision and audition but more like
hunger and thirst, in which a certain unpleasantness/emotional valence
is confounded with the drive state” (Quoted in Gustafson 2005: 235).
I mentioned before that Nelkin (1994) points out that pain, as an
evaluative notion, needs an evaluative theory. One might be forgiven
for thinking that this would put pain in a group with cognitive elements
rather than conative elements, since evaluation could be considered a
kind of judgment (as indeed seems right), but it is not a cognitive judg-
ment standardly conceived. Aydede exhibits the unity-in-diversity of
“mixed theories” of pain:
The basic idea is that the nature of pain is complex consisting of at

least two mental elements. Generally one element is characterized by
using one or more of the following group of terms: sensory, percep-
tual, representational, discriminating, descriptive, or informational.
The other is characterized by one or more of the following: affective,
emotional, motivational, evaluative, or imperative. (Aydede 2009)
2
It is worth noting that it apparently doesn’t even occur to Nelkin to deny that
animals feel pain.
An evaluative account that more clearly emphasizes the affective side

is Helm (2002).3 He argues that we should think of pain as emotion-
like. He describes a category – “felt evaluations” – in which he places
desires, emotions, and pain (and pleasure). Though he calls them a type
of evaluation, he is explicit that they cannot be reduced to beliefs or
desires. He “rejects this approach and instead sees pleasure and pain as
characteristic of a distinctive kind of intrinsically motivating evaluation
that is shared in common among emotions, desires, and (some) sensa-
tions” (13). Aydede summarizes nicely:
These evaluations aren’t judgments understood ordinarily.

Nevertheless, they have intentional contents rationally responsive
to a broader range of background conative and cognitive states of
the experiencer. Thus as felt evaluations they intrinsically motivate
and rationalize behavior typically associated with pain and pleasure.
(Aydede 2009)
Helm appreciates part of Nelkin’s project but notes it will be hard to

apply to animals (2002: 14) and even says it fails to do justice to the
motivational nature of pain, which he rightly sees as crucial. Helm’s
theory is that emotions are a species of pleasure and pain. “This means
that emotional pleasures and pains, namely what one feels in having the
emotion, are essentially intentional and evaluative, a sense of how things
are going – whether well or poorly. ... In short, particular emotions are
feelings of things going well or poorly” (16, 19). This evaluative account
requires less cognitively than any other. Because the seeings as good
and bad here are emotive, they are pure aversion or attraction, not the
kind of considered judgments dependent upon the neocortex and not
requiring higher-order thought that animals plausibly don’t have. On
Helm’s view “in feeling bodily pleasure or pain, we do not make the rele-
vant evaluation by having an attitude in response to what we perceive;
rather, the goodness or badness of what is happening in particular parts
of our bodies impresses itself on us directly in our feeling” (24–25). On
his view, knowledge of pain is direct acquaintance with normative prop-
erties, which is not equivalent to making a normative judgment: “To
feel emotional pleasure or pain is to have one’s attention gripped by the
goodness or badness of something in such a way that one thereby feels
3
See also Solomon (2007: ch. 18), and Roberts (1988 and 2003) for background
on the connection between emotion and evaluation.
the pull to act appropriately” (20). These exciting new emotionesque

theories of pain are all much richer than the older accounts upon which
neo-Cartesianism is based. Aydede’s (2009) conclusion is quite telling.
When we look at the science of pain, especially at what has happened

since the publications of Melzack and Wall’s (1965) and Melzack and
Casey’s (1968), which revolutionized the scientific research on pain,
we see that the science of pain has increasingly conceived of pain as
less like perception of an objective reality and more like emotions by
first drawing the sensory/affective distinction and then emphasizing
more and more its affective aspect. (Aydede 2009)
Prior to Aydede, after a survey of several recent theorists moving in

this direction – most not mentioned in Aydede (2009) – Gustafson said
along the same lines “I could continue to cite sources from recent pain
research, both experimental and clinical, showing the move away from
pain as sensation toward categorizing pain as a complex emotional and
motivational condition” (2005: 237). His conclusion: “we need to think
of pain as an emotionlike condition” (238).
The appeal to seeing as is not an appeal to mystery, though much more
analysis of the notion needs to be done. Direct acquaintance theories are
not new, though, and the appeal to it in this context seems no more prob-
lematic than in other philosophical lines of inquiry. Whether the final
truth on pain is more like Helm’s very emotion-centered theory or one
of the more moderate theories, the neo-Cartesian would have to argue
that non-human animals do not have the neurological wherewithal to
embody the right kind of emotional experiences. This would seem to be
an up-hill battle, since the pain center (including what once was called
the “limbic system”) is a much phylogenetically older portion of the
brain than the neocortex. The Lobotomy Argument has no force against
a theory of pain that is sufficiently emotion-based. In the next section,
I will point the way to further research in immediate self-knowledge of
one’s own condition drawing from semantics and epistemology.
5.1.2 Non-conceptual knowledge of pain

In the last section, I affirmed the position that pain was emotion-like
and that knowledge of one’s affective states was importantly immediate
in a way that is over intellectualized by HOT theories. However, I said
little by way of a positive account of the epistemology of such states.
Here, I sketch the outlines of a research project for such a theory so that
the above is at least a bit of a down payment on the promissory note.
There are many unexplored paths the theorist about animal conscious-
ness might take through the philosophy of language. If we grant to the
creature, a chimp, Noam, a thought relevantly similar to That hurts!
Then we might apply a semantics similar to Evans’s “Understanding
demonstratives” (1981) to arrive at the conclusion that Noam does
have a thought about his own state of mind. The demonstrative will
clearly denote Noam’s pain. Then on a structured theory of propositions
with relata as Evans describes, Noam and his thought will literally be
a constituent of the content of the thought That hurts. Stanley gives a
modified account of an Evans-like view using relations one can only
stand in to oneself or at least in virtue of which we think of ourselves as
ourselves. He concludes, “The fact that certain thoughts contain these
first-person ways of thinking as constituents is what makes them first-
person thoughts” (Stanley 2011: 85–86). There is not time to explore
this lead much further, but it is indicative of the many ways in which
future discussion could turn. For now, I will just sketch the outlines of
this kind of strategy.
In the philosophy of language, it is almost impossible to make sense of
various puzzles and linguistic phenomena without being at least semi-
Fregean. It is extremely difficult to write down a satisfactory account
of modes of presentation, but the basic idea is not hard to understand.
When Lois Lane sees Clark Kent, she sees Superman. However, she
doesn’t see him as Superman. He is not presented to her in a way that
her mind processes under that description. And of course the problem is
not that she lacks the relevant concepts, for she clearly has the concepts
Clark Kent and Superman, it is just that she lacks the appropriate mental
connection between them in the “files” in her mind. Though the Clark
Kent/Superman example is the textbook way to introduce the sort of
puzzle that gets one thinking about modes of presentation, there are
more helpful examples for our purposes.
There is a famous argument against physicalism called the Knowledge
Argument. Here is a brutally oversimplified version. Consider Mary. Mary
is the world’s foremost authority on color. She has performed experi-
ments doing cutting-edge research in color theory. She has read every-
thing ever written about color and has deduced everything else. Oh, and
one more thing: Mary is colorblind. Now consider this argument.
1. Mary knows all the physical facts.

2. Mary does not know what it is like to see red.
3. Therefore, what it is like to see red is not one of the physical facts.
I think that theists need to reject this argument, for God, being omnis-
cient, will know all the facts (there are puzzles about that, of course,
especially de se facts indexed to others), but he will not know what it is
like to be envious or gluttonous or to be weak in other ways. One way to
resist the knowledge argument against physicalism is to say that Mary
doesn’t learn any new facts, but that, rather, she grasps an old fact under
a new mode of presentation. This mode of presentation consists in part
of the application of phenomenal concepts. A phenomenal concept is,
very roughly, a concept one can have only by having a certain expe-
rience, by hosting a certain phenomenal content, or exemplifying a
certain phenomenal property (which property may well be the denota-
tion of the concept; the phenomenal property is at least in part constitu-
tive of the phenomenal concept).
The suggestion, in brief, is that to require animals to cognize their
pain via a reflective higher-order thought seems to over intellectualize
the process, but at the same time hits on an important idea: A crea-
ture utterly blind to its own pain could not truly be said to suffer. A
middle ground is that many animals – at least mammals, but plausibly
some birds and perhaps reptiles – grasp the fact that they are in pain via
a mode of presentation peculiar to first-personal experience and thus
don’t need the concept of self. Since many philosophers take modes of
presentation to be narrow functional roles (i.e. internal causal roles), it
is in fact quite plausible that a semi-Fregean proposition structured of an
individual, a property, and a mode of presentation – or on a neo-Millian
view, a “guise” (Braun 2004, Cullison 2007) – could be the content of a
thought for an animal who is otherwise reflectively ungifted.
The technicalities can be left to further research, but the funda-
mental idea is that one can host a phenomenal content, in the words
of Stanley’s translation of Frege, in that “special and original way
in which each of us is presented to himself, and not to any other”
(Stanley 2011: 83). There is something it is like to host something first-
personally, and this is phenomenologically both obvious and so close
to us as to be impossible to pin down. I see no reason, however, to
think this mode of knowledge takes much by way of cognition. Clearly
it takes more neuroanatomy than an earthworm has, but I suspect that
even in humans pain isn’t (ordinarily) made conscious by any kind
of cognition, especially cases of extreme pain that seem to wipe the
mind blank of cognition. One is simply directly acquainted with the
raw Owwwww!!! of the pain in a way where the directness itself is so
palpable that it is itself a way of knowing the pain as one’s own. There
are two components here. One applies Russell’s distinction between
“knowledge by acquaintance” and “knowledge by description” (see

Fumerton 2008) considering knowledge of pain qualia as a form of
knowledge by acquaintance rather than knowledge by description. The
second involves direct acquaintance with a property of the pain, it’s
“mine-ness.”
I have been drawing on considerations taken primarily from the
philosophy of language, but there are considerations in epistemology
that could put some flesh on the bones of the kind of gambit just
mentioned or provide a similar but alternative route. Epistemological
foundationalists from Descartes to Fumerton (1996, 2005) have posited
a form of direct acquaintance with one’s mental states. On Fumerton’s
model, one is directly acquainted with the truthmaker of propositions
about one’s own mental states. Gertler (2001) and Chalmers (2003) have
similar views that apply specifically to knowledge of one’s phenomenal
states where one’s belief about one’s inner states is partly constituted
by the phenomenal property instantiated. Whether focused on truth-
makers or properties, some theory of direct acquaintance seems plau-
sible for some mental states. (They needn’t be luminous to be useful,
so the anti-luminosity arguments in Williamson 2000 needn’t be taken
into account here.)
The suggestion, then, is that animals that lack any kind of self-concep-
tion but which nevertheless exhibit all the natural signs of experiencing
pain are in a position to bear the relation of direct acquaintance to the
truthmaker for the fact that they are in pain and/or with the phenom-
enal property that constitutes their pain. Here is what I have not been
trying to do: to mount a convincing argument that animals have the
kind of direct awareness of their “ownership” of their experiences, much
less to offer a detailed theory of how that works. Rather, I have simply
been trying to suggest plausible routes forward for understanding what
we antecedently have reason to believe – that even mid-level mammals
and possibly birds and reptiles experience morally significant pain –
while trying to do justice to the intuition that a creature must be in
some way aware of their pain for it to be morally significant.
These ruminations provide a possibility for non-human primates and
non-primate mammals (even the lesser ones, and, perhaps, even birds
and reptiles) satisfying an appropriate kind of higher-order awareness
constraint on conscious experience. This possibility represents a poten-
tially fruitful area of further research and would allow one to grant to
HOT theories of conscious experience that there is something intuitive
about them without over intellectualizing the process of consciousness,
as they seem to.
5.1.3 Recap
In Chapter 4, Sections 4.2.3.1 and 4.2.3.2, I considered two ways to moti-
vate neo-Cartesianism: philosophical – the HOT theory of phenomenal
consciousness – and empirical – the Lobotomy Argument. In Section
5.1.1, I argued that the Lobotomy Argument can be circumvented by
a move away from the rationalism of neo-Cartesian views and toward
more contemporary emotion-based theories of pain. And then in
Section 5.1.2 I argued for a mode of knowledge of pain that also rejects
aspects of Cartesian rationalism. These moves serve as defeater-defeaters
defending the default view of common sense and scientific consensus:
non-human animals feel morally significant pain.
5.2 The problem of primate pain or seeing through the

mirror test
5.2.1 A statement of the problem

In this section, I grant a great deal to the neo-Cartesian. I grant that
there is no morally significant pain experienced by non-primates. Some
arguments for neo-Cartesianism are somewhat plausible even at the
mammalian level given what has been shown about rats in particular
(Murray 2008). In this section, though, I point out that even if only
primates, indeed only the most advanced of the primates – great apes, a
category into which humans fall (Homonidae) – feel pain, there is still
plenty to worry about. Great apes might present a particular problem,
for they seem to be at a crucial stratum of mental capacities: just high
enough to suffer terribly and yet just low enough that a traditional soul-
making theodicy does not apply. In this section, I will examine current
thought about the relevant data concerning great apes.
A standard way to assess self-awareness in non-human animals is the
“mirror test” developed by Gordon Gallup Jr.4 In experiments, observers
place some mark on the face of an ape – usually a chimp, but sometimes
an orangutan (gorillas are much harder to work with, as they are more
endangered, and consider eye contact a form of aggression) – and put
them in front of a mirror to see how they react. When they exhibit
such behavior as noticing the mark, looking curiously at it, touching
it, looking at the hand that touched it, etc., they are deemed to have
passed the test and are said to be self-aware. Even people who are rela-
tive skeptics about animal self-consciousness – like mirror test pioneer
4
Gallup (1970).
Gallup himself – attribute self-awareness to great apes. Lesser apes (e.g.

gibbons), monkeys, and almost all mammals, birds, and reptiles fail the
test. Thus, if we accept both a form of HOT and the validity of the mirror
test, we are forced to accept that the vast majority of apparent animal
suffering is illusory.
Gallup himself, referring to the phenomenon of “infant amnesia” –
almost no human memories are retained before two years of age –
suggests that “Prior to the emergence of self-awareness, we may all
have been unconscious” (Gallup 1985: 638). This does indeed seem
to be an entailment of his thesis. This presents a dilemma for Gallup.
Either being conscious in the way specified is a necessary condition for
moral standing, or it is not. If it is not, then failing a mirror test is not
a strike against moral standing. If it is, then Gallup is committed to
the proposition that children much under two have no standing in the
moral realm. This once would have been considered a reductio on the
view. However, now this view is gaining popularity in academia. (Peter
Singer was once notable for his advocacy of this view, but it seems to be
catching on in the academy. The most recent advocacy at the time of
writing is Giubilini and Minerva 2012). This consequence puts into stark
relief the consequences of one’s position in this arena for ethics gener-
ally. It needs to be faced squarely.
As I say, I will focus on great apes, but it is worth noting that there
is reason to believe the experiments discussed have been thought to
produce false negatives (see Koerth-Baker 2010). The tests often make
assumptions that prevent them from capturing the reality. For example,
in some iterations, gorillas, alone of adult great apes, failed the mirror
test. This was naturally quite shocking. As it turns out, it is thought, the
problem was because in gorilla culture eye contact is considered very
threatening, so the gorillas were not looking at the foreheads of their
reflection. In an opposite sort of situation, elephants, who are accus-
tomed to putting mud on their heads, probably didn’t react because
it was nothing new to them to have marks on their heads. There are
even variations in passing the mirror test along human cultural lines.
Non-western kids – such as those in Fiji and Kenya – don’t pass the
mirror test until well past the time by which very self-aware western kids
do. Thus, self-consciousness might be present in animals farther down
the phylogenetic scale than the mirror test suggests.
I mentioned that mirror test inventor Gallup is quite retentive in attrib-
uting second-order mental states, but he admits that other humanoid
apes have self-awareness. So let’s assume for a moment, just for the sake
of argument, that Gallup is right and only humans and (the other) great
apes – gorillas, orangutans, and chimpanzees (including bonobos) –

have second-order mental states and, specifically, a concept of self (or
at least some broader form of relevant self-awareness). There would be
no reason not to attribute second-order mental states to extinct homi-
nidae, including the australopithecines and perhaps a bit farther back.
This gives us at least three to four million years of a population of crea-
tures capable of great suffering and perhaps as much as seven to thir-
teen million years. Over this time it would be difficult to estimate the
amount of suffering encountered. If this were all the data there were on
non-human animal suffering and I had to change the title of the book
to The Problem of Primate Pain, this would still constitute a strong basis
for an objection to theism, for the millions of years of suffering of these
closest of human ancestors seems to fit the naturalist narrative better
than the theistic story. The theist must offer some account of how this
suffering is a sub-plot of a wider saga.
Murray is even more retentive than Gallup. He claims that the mirror
test “shows at most merely that some animals have the capacity to
recognize certain bodies as their own” (2008: 61). This is in contrast to
what the neo-Cartesian says is required for consciousness of pain: recog-
nizing certain mental states as their own. In reply, first, I want to consider
Murray’s use of “shows” in that quotation. As I said in Chapter 1, one
of the ways in which I think Murray and many other theorists go wrong
(in general but especially) in the literature on the problem of evil (van
Inwagen is the greatest offender here) is to operate – sometimes explic-
itly but usually tacitly – with an “all-or-nothing” epistemology that is
insufficiently sensitive to the degrees in which justification comes and
that is insufficiently in tune with the kind of explanatory reasoning I
think most appropriate in both philosophy and science.
If Murray means by “shows” anything like “proves,” then I can
certainly agree. Perhaps the experiments only prove that when certain
apes are placed in certain circumstances they display certain behaviors.
Yet surely the studies lend credence to, indeed, lend considerable support
to the thesis that they have an appropriate form of self-awareness. The
data fits nicely into the theory that they are recognizing themselves in
the relevant way. This has been the nearly uniform reaction of scientists
working directly with them and though that does not strictly guarantee
that the interpretation is correct, it surely lends support to it, making it
the credible conclusion.
Furthermore, consider what is supported even by what Murray seems
to admit: “some animals have the capacity to recognize certain bodies
as their own” (2008: 60). This breaks down into at least two parts. 1. To
recognize that something is a body. 2. To recognize that thing as one’s

own. Recognitional abilities are closely tied to concept of possession
(indeed, some see the tie as identity or near identity (Dummett 1993)).
It is difficult to see how one could recognize something as one’s own
without a substantive concept of oneself, that is, one’s self. Murray offers
no explanation as to how a creature could distinguish themselves from
others nor how they could recognize something as belonging to them as
opposed to another without possessing the concept self. (Remember, we
are bracketing my arguments for non-conceptual recognition of states
as one’s own.)
We must distinguish between the conceptualization of self that under-
writes discrimination of oneself from others and the ability to attribute
things to oneself from a generalized concept of selves or the ego. Though
the mirror tests provide a bit of evidence that the animals that pass it
have such a rarified concept, it does not seem on balance to support
affirming that animals have that abstract concept. But not even neo-
Cartesians with their intellectualized theory of consciousness think that
concept is required for consciousness. So the view I am suggesting is a
golden mean. On the one side we have the extreme view that would
attribute to animals a generalized notion of selfhood (I’m not aware
of anyone who proposes this). On the other side we have the extreme
view that they merely represent something about their bodies, which
seems to be what Murray intends to suggest (his actual words might
belie something more). The view I’m suggesting is that they have some
kind of important reflexive concept whereby they represent themselves
to themselves, not just something about their body. They can have the
concept mine without necessarily having the concept I.
Consider this. Here is a very strange speech. “Koko knows that that
body is hers and not another’s, but she doesn’t know that she is not
another.” Can any sense be made of such a speech? I can’t make any
sense of it. But it seems to be just the sort of thing that Murray is
committed to having a sensible interpretation.
Now if higher primates and perhaps the highest-functioning mammals
(like dolphins and elephants, which have also passed the mirror test)
have the kind of self-awareness of their bodies they seem to have and
which Murray admits they have, which includes the concept mine, then
there is in principle no barrier to them also being aware that certain
mental states are theirs, applying their mine concept to their own mental
states. In fact, what could be more natural? Mental states have a kind of
immediacy that makes them perfect candidates for recognition as one’s
own. If Reid is correct, then animals, like humans, might in fact first
form causal concepts by observing their own internal agency. In fact,

Povinelli and Vonk. (2004) allow for representation of agency as a plau-
sible interpretation of the results of the mirror tests (24).
The form of my argument is this.
1. Murray admits that apes can recognize bodies as their own.

2. But recognizing bodies as their own requires that they possess the
ability to self-attribute.
3. Therefore, Murray’s admission entails that apes possess the ability to
self-attribute.
4. But if they have the ability to self-attribute at all, then they should be
able to self-attribute mental states as well.
One might object that it is precisely in virtue of their immediacy that

mental states are apt to be overlooked as objects of consideration and
attribution. Evidence of this is the rarity with which ordinary folk refer
to their mental states, training their attention, rather, on the external
world objects they are perceiving. In reply, I suggest that people attend
to external world objects via their attention to their experiences. But a
great example of when people attend to their experience apart from an
external world object is their experience of pain!
I have found no study more sympathetic with Murray’s objection than
Povinelli and Vonk (2004). They suggest that representation of body is
more fundamental to what the mirror test shows than representation
of mental states. Like Murray, they seem to be a bit shortsighted about
the possibilities they open up by admitting this, although they seem to
have a more hedged conclusion, something like this: If the mirror tests
reveal self-awareness, they reveal it via some more basic revelation, such
as awareness of one’s own body, bodily states, or agency. My position
is that whether mirror tests reveal self-awareness directly or indirectly
doesn’t matter so long as it reveals it.
Yet even they affirm “chimpanzees and orangutans may possess an
explicit or integrated enough representation of their bodies to have a
concept (my body)” (Povinelli and Vonk 2004: 24). The concept my body
could only be applied to a subject like that body. It would attribute two
features to what it is applied to: (i) that it is a body and (ii) that it is mine.
And as I argued just above, given that a creature is in possession of this
reflexive concept, there is no reason to think that it would not or could
not apply it to its own mental states. At least Murray has provided no
such reason. I tentatively conclude, therefore, that even on the most
retentive reading of the mirror tests, there is good inductive support for
a substantive enough concept of self-awareness to underwrite second-

order awareness of their own mental states. Thus, even if a HOT theory
of consciousness is correct – which I have argued, along with a broad
consensus of philosophers of mind, does not have much going for it –
there remains a considerable quantity of animals over time who meet
the requirements for consciousness and for whom there is therefore no
reason to deny that they suffer pain.
5.2.2 Objection from Carruthers

Peter Carruthers – the chief defender of a HOT theory of consciousness –
when considering directly the implications of his views for animal
welfare considers a line of thought similar to mine above. He admits that
at least great apes, especially chimps, have second-order beliefs (1992a:
138–139) based on “an impressive body of evidence” concerning decep-
tive behavior (he cites Byrne and Whiten 1988). Chimps are routinely
observed engaging in deceptive behavior. For example, they pretend that
they have food stored in one place to throw other members of the band
off and keep them from stealing the food. This is particularly common
among females who tend to get their food taken by males.
This supports the thesis that chimps have a specific form of high-
er-order thought: second-order beliefs. That is, they have beliefs that
take as their objects the beliefs of other chimps. This seems required by
a natural understanding of what goes on in deception. The deceiving
chimp must form a plan that involves getting a fellow chimp to believe
that there is food in a spot where the deceiving chimp is digging. Likely
this also involves believing that the other chimp desires food and that
they believe digging outside camp is a sign of a food stash. (Note that
this plausibly involves grasping the notion of evidence; indeed it is hard
to make sense of the idea of deception apart from evidence, for one
represents oneself as doing something indicative of one thing when it is
not so: It involves providing the victim with misleading evidence.) So
the deceiving chimp has beliefs about beliefs and beliefs about desire,
which are two kinds of second-order beliefs, which is a kind of higher-
order thought.
Carruthers then asks if the admission of second-order beliefs makes it
plausible to ascribe to chimps thoughts about their own thoughts (which
is what the HOT-theorist requires for consciousness to be present). After
all, this seems natural. If they can have thoughts about the thoughts
of others, why not about their own thoughts? Carruthers’s response is
terse: “There is not a shred of evidence to suggest it” (185). This bravado
is repeated later in his chapter summary: “[T]here is no reason to
believe that any animals are capable of thinking about their own think-
ings ... [so] none of their mental states will be conscious ... ” (193). This
conclusion seems utterly incongruent with the admission that chimps
can have beliefs about the beliefs of other chimps. One would think that
a chimp’s ability to have beliefs about the beliefs and desires of another
chimp would at least inductively support the proposition that chimps
can have beliefs about their own beliefs and desires. Let’s examine the
reasoning that leads Carruthers to this bold conclusion.
He asks the reader to consider the greatest display of intelligence by a
chimp that he is aware of. This consists in watching a bunch of fruit and
vegetables (18 pieces or so) being hidden in a one-acre field. Then, entering
by a different point, it can collect two-thirds of the food on average. It also
collects first the fruit, which is preferable to chimps. His argument that
great apes cannot think about their own thoughts consists entirely in an
analysis of this case. He says that if they could, we would expect them to
do much better. Specifically, we would expect them to figure out that they
would save time by collecting some vegetables along the way with the
fruit rather than getting all the fruit first and then getting the vegetables.
I find this argument weak for a number of reasons.
First, the attribution of reflective second-order thought in no wise
implicates that one is a genius. In fact, it doesn’t even imply that one
will be very smart. I have a five-year-old and a three-year-old right
now. I have no doubt that either of them is capable of second-order
reflective thought. Yet they are, not to put too fine a point on it, pretty
dumb when it comes to efficiency planning. I can easily see them being
far less successful than the chimps in the study. I’ll wager that even
if you told them “Guys, think about your plan here, aren’t their ways
you could save yourself some work?” they would not improve measur-
ably. From past experience, I could even design the plan and give it to
them and they still couldn’t do a lot better. I’ll bet most readers have
an adult uncle or brother-in-law who is only marginally more capable.
So it seems to me there is just no considerable expectation that the
bare existence of second-order reflective thought would be manifest in
“dramatic improvements in performance” (186). For that expectation, a
certain amount of additional “IQ” would be required. According to Paul
Cooijmans, chimps’ IQs average in the “Moderately Retarded” category
(between mild and severe) (see Cooijmans 2012). On the other hand,
bare second-order reflective thought does seem sufficient to allow an
animal to think about its own pain. You don’t need much by way of
intelligence to know that you are in pain. (Furthermore, that one has
the ability for reflexive second-order thought does not imply that it will
be on-line for all processes. It may be that the chimp’s thoughts about
food collection are not poised for uptake into the reflective module,
but pain is. This would not be too surprising, since the former are quite
complicated and the latter quite simple.)
One wonders just how dramatic an improvement Carruthers expects.
For all I know, it might be a considerably complex coordination problem
to try to weigh efficiency against desired outcome. After all, for all
the chimps know, they are in a risky game. Whether in the wild or in
captivity, food is typically available only for a limited time. So the most
“efficient” path in terms of least investment of walking distance might
not be likely to result in the best payoff in the form of a fruit-to-vege-
table ratio.
In fact, unless we know the degree of preference of fruit over vegeta-
bles, we don’t know that the gather-all-the-fruit-first gambit isn’t the
decision theoretically most efficient. For if they value fruit more than
vegetables by a wide enough margin, then relegating vegetables to any
“extra” time beyond the estimated allowance makes perfect sense. If
you put me on a one-acre farm that had dark-chocolate peanut clusters
and broccoli hidden all over it, I wouldn’t waste much time looking for
broccoli. Once I had satisfied myself that I had recovered all the peanut
clusters I was likely to gather, only then would I begin to think about
wasting time on the broccoli. Now it is probably less of a preference
imbalance to vegetables for chimps than there is between dark-chocolate
peanut clusters and broccoli for me. However, this exaggerated analogy
establishes an important point: Efficiency of plan is a function of rela-
tive preferences, and this is information that Carruthers lacks, except
that he knows they definitely prefer fruit. It is, therefore, unclear that
Carruthers has much by way of justification for his prediction.
In fact, even in fairly intelligent agents it is often a more efficient
use of limited resources to rely on heuristics. So even if the chimps
were capable of making the (probably quite) complicated calculations
concerning whether it is better to stop and gather a less-desired item
and risk losing a more-desired item or to proceed on to the more-desired
item, it might not be worth their time to do so. Instead, they might just
proceed by evolved heuristics (which may, in fact, have evolved during
less-advanced cognitive stages and simply stayed on board because they
were sufficiently reliable). One such heuristic is a class of “sure thing”
heuristics. The slogan for these heuristics is “Go with what you know.”
The chimp knows it loves fruit best, so it makes sense to just go get
that first. This hypothesis fits well with what we know about bounded
rationality from cognitive psychology.
Thus, it seems to me that Carruthers’s humbler conclusion at the end

of his book is a more accurate assessment of his ruminations here. In
the concluding section of his book when it is “time to pull together the
threads of my argument” he says, “I shall by-pass the position ... that
the mental states of animals are non-conscious ones” because it is “too
highly speculative to serve as a secure basis for moral practice” (194).
One would think the confidence shown above would at least allow the
thesis to feature into the discussion at the end, but apparently Carruthers
thinks not. I am inclined to agree with his latter assessment.
5.2.3 A final argument

In the previous section, I granted more to the neo-Cartesian than in
fact I think reasonable. Yet even granting a restrictive domain to animal
self-consciousness, the problem of primate pain is enough to ground a
serious objection to theism. In this section, I grant even more. I grant
at least Type II neo-Cartesianism, perhaps Type I, depending on how
certain key terms are understood. In this section I will defend the thesis
that even if an organism is not aware of its pain, the fact that it is in
pain is a bad state of affairs. And since a perfectly loving God would not
allow any bad state of affairs without a justifying reason, the existence
of these bad states constitutes prima facie evidence against theism. I can’t
say much positive to motivate this thesis, but I think its plausibility
actually comes out in the course of thinking through objections to it.
My method, then, will be to move directly to the objections and hope
that, like me, the reader finds that in my replies to those objections, the
thesis gains plausibility.
Gallup (1985: 631ff) – who, as we mentioned before, is quite skeptical
about animal self-awareness for all non-human animals but the great
apes – considers the suggestion that even “unconscious pain” is morally
significant and uses a thought experiment concerning sleepwalking to
attempt to rebut it.
Gallup first notes that sleepwalkers clearly have consciousness of their
surroundings, for they avoid walking into walls and such. He then offers
a dilemma about the sleepwalker. Either they are aware of being aware or
they are unaware of being aware, and “the latter is tantamount to being
unconscious” (638). This is similar to how Carruthers uses blindsight
cases. He uses them to frame the question “[A]re non-conscious pains an
appropriate object of sympathy and moral concern?” (Carruthers 1992a:
187). He seems to think not. I will offer two brief objections to this view.
First, there is an interesting regress worry for these cases. Gallup says
that being aware is of no moral significance unless one is aware that
one is aware. But why stop there? What if one is utterly unaware that
they are aware that they are in pain? Why is that irrelevant to the moral
significance of the first-order state but one order less isn’t? There is a
plausible dictum that “a thing cannot impart what it does not have” (a
rule of thumb, not a necessary truth). Carruthers and Gallup agree that
a state is conscious when it is the object of a higher-order state. But what
if that higher-order state is not itself the object of an even higher state?
What’s so special about it then? How does it come to have the ability to
bestow consciousness on a state when it itself is not conscious? It seems
utterly arbitrary to me.
Second, if pain of which one was unaware were of no moral signifi-
cance, then it would be permissible to torture the sleepwalker. But clearly
it’s not. One might think that this objection fails because the sleep-
walking human would be aware that they were in pain. But that needn’t
be the case. Suppose the sleepwalker is hooked up to a device that tells
us with certainty when someone is asleep. Out comes the sleepwalker
from his room, deftly stepping over a sleeping dog. All the while, the
meter reads “ASLEEP.” To test the resilience of sleep in a sleepwalker, a
scientist pricks him with a pin in the arm. The sleepwalker yells “Ouch!”
but the needle on the meter stays right in the middle of the “ASLEEP”
band. Curious, as scientists tend to be, she is tempted beyond her ability
to resist increasing the intensity of the jabs. Proceeding by small incre-
ments, the scientist continues, and before too long the sleepwalker is
screaming at the top of his lungs. It seems clear this is impermissible.5
But this would not be so if Carruthers and Gallup were correct, for the
sleepwalking subject, being asleep, is unaware that they are feeling such
excruciating pain. This is a reason to doubt that pain is rendered morally
irrelevant just because the subject of the pain is “unaware” that it is
occurring. But if we reject that restriction, then even on high-octane
neo-Cartesianism, morally significant pain is widespread through the
non-human population of the animal kingdom, and the problem of
animal pain returns with a vengeance.
5
Murray and Ross (2006) consider an objection that sounds like mine, saying,
“Access to these states, the critic might contend, is irrelevant to whether it is bad
to be in the state itself. Clearly, if a state is intrinsically bad, it is not made better
merely in virtue of the fact that the creature does not know about it” (176). But
they have in mind a distinction between “metaphysical evil” and “moral evil.”
This distinction does no work against the problem of pain as I have framed it. For
the hypothesis that there is an all-good, all-powerful agent who created the world
predicts that any evil of any kind will need an explanation.
6
The Saint-Making Theodicy I:
Negative Phase
In this chapter, I will begin to describe a species of soul-making theodicy

that I will – after the description is completed – apply to the case of
animal suffering. This new species owes much to Hick, Swinburne,
Adams, Hasker, and others but extends and elaborates upon the Irenaean
idea in a way which reflects my own spiritual journey and that, it seems
to me, significantly strengthens the theodicy (in ways, I might add, that
will make it seem particularly hard to apply to animals). First, however,
in Section 6.1, I wish to fend off an expected problem from the start.
There are certain caricatures of the soul-making theodicy that I wish to
avoid from the outset. Though it has had its defenders all along, it has
also had plenty of critics, who I think are not always fair. Then in Section
6.2, I will describe the basic value system I will assume, which is at the
heart of the project. In Section 6.3, I will show how the work of Marilyn
McCord Adams – primarily her adaptation of Chisholm’s notion of the
defeat of evil – can be combined with Hick’s. for a richer theodicy. This
chapter focuses on the value theory behind the saint-making theodicy,
the notion of defeat, and the way they fit together.
6.1 Some necessary ground clearing
The essence of the soul-making theodicy is that we find the kind of evil
we find in our world because it is precisely the kind of suffering required
to create great “souls,” that is, persons of great character, persons who
exercise the greatest virtues, especially various aspects of the selfless love
which in the Christian tradition is often referred to by the transliterated
Greek word agape. The theodicy is subject to caricature because it may
be seen as a species of the “greater good” gambit.
96
The Saint-Making Theodicy I: Negative Phase 97
Here is a caricature of the greater good idea.
GGC It’s fine for God to allow you to suffer, because He’s got a big pie
in the sky for you that will make up for it.
This is, of course, crassly put. But it accurately conveys the attitude
of some to a class of arguments that are historically important. Saint
Thomas Aquinas considers two objections to theism: naturalism and
evil. His reply reiterates Augustine’s reasoning.
As Augustine says (Enchiridion xi): “Since God is the highest good,

He would not allow any evil to exist in His works, unless His omnipo-
tence and goodness were such as to bring good even out of evil.” This
is part of the infinite goodness of God, that He should allow evil to
exist, and out of it produce good. (Summa Theologica I.2.iii, Reply to
Objection 2.)
Note that neither Augustine nor Aquinas say that God brings about indi-
vidual evils so that he can bring good out of them. So far as these quotes
are concerned, the idea is no more than that God will permit evil only
if he can bring good from it later, at some point in the future. The word
“greater” does not occur in either what Augustine or Aquinas says here,
but the tradition has generally interpreted the idea as being that God
does bring about from evils goods which outweigh them. We can put
this idea less crassly but still simplistically as follows:
GGS It is morally permissible for God to allow an evil so long as from

that evil he brings about some good whose goodness is greater in value
than the quantity of the disvalue of the evil (or possibly exactly equal).
I will not here address the issue of whether the greater good must
include an appropriate benefit for the very individual who suffers the
evil or whether a good accruing to someone else might justly balance off
an evil suffered by another.1 I’m not aware of any major contemporary
1
Stump (1983) makes a good case that the evils which befall individuals must
be compensated for in the life of that very individual, not just justified by goods
which their suffering brings to others. Adams (1999) adds that they must have
some knowledge of the fact that they are so benefitted. Swinburne (1998) depends
heavily on the good of being useful even when one doesn’t know it, especially
for animals.
figure who defends the flatfooted version of the greater good defense
represented in GGS. It is not too hard to see why. It seems to allow for
the following kind of scenario.
The Gatenator Bill Gates enters your house in the middle of the night
with his hulking minions. They start smashing everything in sight,
including family heirlooms. They burn all your family pictures and
then drag you all out on the lawn while they set fire to your house.
Then he gives you a billion dollars.
The idea behind such a caricature is that there is something deeply unsat-
isfying about merely “buying off” someone after bringing suffering to
them. It is particularly troubling that in the book of Job, God can come
off as appearing to act like the Gatenator. (However, see Stump 1996.)
We can assume that there is some dollar amount such that you would
consider yourself overall better off as a result of the payoff. But even
assuming this, it does not seem to make it permissible for Gates to do
this in the first place. One can’t just walk around punching people in the
face and throwing fistfuls of money at them as they lie on the ground
bewildered. Reparations, even if they are in some way sufficient for
damage done, do not buy one’s way out of the guilt of one’s actions.
Permissibility cannot merely be purchased like a box of cereal.
Furthermore, the hypothesis of theism is that God is perfectly good. So
even if a theodicy could show that creatures had, in the end, received
due compensation, it would not yet follow that the role of the creator in
that theodicy was adequate to the conception of God involved. And of
course, the merely Christian conception of God emphasizes God’s love
for mankind, so any theodicy that wishes to apply to mere Christianity
will be inadequate if its solution does not include a loving role for God.
So if there is anything to a “greater good” defense, it needs to be more
subtle than GGS. But there are crass and simplistic versions of the soul-
making theodicy that sound like instances of the worst kinds of the
greater good defense. Consider these.
SMDC It’s fine for God to allow you to be raped, that way you have
the opportunity to forgive your rapist.
SMDS It is morally permissible for God to allow an evil so long as from
that evil some individual’s virtue of sufficient goodness is displayed.
One problem with these two statements of soul-making is that they are
easily read with a focus on individual, particular events. But the soul-
making theodicy has a significant free-will component. It does not rest

on the value of the exercise of free will alone, nor on the morally signifi-
cant exercise of free will alone.2 Though these might be great goods, the
Irenaean does not suggest that their goodness is sufficient to explain
God’s permitting the evils he does. Still, the display of virtue does entail
the morally significant exercise of free will.
Recall the imaginative world of the Fall narrative. God creates Adam
and Eve and then turns them loose in the Garden. He comes back and
senses something is amiss and summons them to account. In that narra-
tive, God is not hovering over Adam and Eve deciding whether to permit
this or that. Rather, we can say God is portrayed as creating a milieu
and letting things take their course within it for a certain time. The
same idea is at work in Irenaeus. God creates the conditions for a world
that develops largely of its own accord. It must fit certain parameters; it
must be a world that provides opportunities for (and, surely, promotes)
the best kinds of goods to be realized. According to the Irenaean, this
consists primarily in the highest forms of love, and of things that
love entails such as practical and theoretical rationality, agency, and
autonomy. Having promoted (and, surely, secured) such general condi-
tions, God then “sets us free” to make of the world, and ourselves, what
we will, knowing that there are many futures made possible by having
creatures in such an environment. In the meantime, as Saint Paul says,
“he maketh his sun to rise on the evil and on the good, and sendeth rain
on the just and on the unjust” (Matt 5:45, AV. See also Jesus’s parable
of how the wheat and the weeds grow together until the harvest, Matt
13:24–30). For the “open theist” this will be a literal truth (See Hasker
2004 and 2008). For the more traditional Boethian, for whom the future
is eternally present to God, this will be an alternative creation myth, but
one that expresses important truths. For, I suggest,3 the way to think
about particular evils is as arising from a “world-ensemble” (to borrow a
term from the fine-tuning argument literature), and so the focus of both
the atheologian and the theodicist should be on whether it is permis-
sible and consistent with divine goodness and love for God to pursue
such a policy, a policy that leaves so much open-ended. So according to
the standard soul-making theodicy, God purposefully puts us in circum-
stances that challenge us. (This appears to be in stark contrast to the
Biblical account of the Fall, and Hick makes much of this contrast, but
2
Plantinga (1974b) focuses on the notion of morally significant use of free will,
for example. See, for example, p. 166ff.
3
Here I depart from Swinburne, I think. See Swinburne (1998: 218).
there is nothing in the Fall narrative that strongly supports that it was
God’s primary plan for our Parents to remain in a perpetual state of
paradise.)
As a concrete example of the kind of thing I have in mind, I share
this story. I wrote the last paragraph on the road to a mountain bike
race in which my oldest daughter, 14 at the time of writing, is racing. As
I write this, we are between the race and the awards ceremony. Fiona,
with appropriate pride, won the women’s A-level collegiate finals. This
is a major achievement for a 14-year-old. It is a hard-won achievement,
earned with, literally, blood, sweat, and tears. In fact, this morning I
had to stitch up one of her teammates, Kate, who badly cut herself, and
I myself have had many metal parts installed in my crushed wrist and
probably hundreds of stitches. So I know all too well what I’m permit-
ting her to risk in racing bikes. And it is a “promoting permitting” as
well. For though I didn’t push her to do it, and she did ask, I also encour-
aged her at various stages of the process to take on the risk. I believe she
knows the risks as well. So why does she take those risks? And why do I
permit my beloved daughter, my beautiful, first-born child to take these
risks?
It is hard to say, honestly, but the answer surely involves a number of
things. First, there is the simple fact that she just loves doing it. This is
not sufficient, of course, since one might love doing crack (I suppose).
But it would not be permissible to let your child do crack just because
they liked it. And the same is true for many activities less nefarious than
doing crack. But there is also the fact that it makes her a better person
in so many ways. It makes her physically and mentally stronger, which
is to say that it perfects her human nature. It teaches her that she can
do things she once thought impossible (this is pretty strictly speaking
true: She now rides both up and down rough, steep grades I myself once
thought not within the power of any human being). It teaches her the
value of teamwork. It has been an opportunity for her to exercise the
virtue of magnanimity in how she treats those she has beaten, especially
when she is, quite literally, put on a pedestal above them. I suppose,
now that I think of it, there are probably scores of essays about the value
of sport from which I could crib notes (without ever having to quote the
glib “No pain, no gain”). And though I think Americans make an idol
out of sports and sporting figures, I do think there is considerable merit
to the notion that it can be and often is – at least at the amateur level
once so typified by the Olympics – a prime source of personal and collec-
tive virtue. And with my own case of mountain bike racing as our family
activity, as with many sports, the risks and sacrifices are not trivial.
Another relevant example of purposefully courting danger for the sake

of virtue is the “challenge course.” A challenge course, sometimes called
a “ropes course” due to the number of high-off-the-ground events, is a
course for those, usually in an educational or corporate environment,
who wish to build group cohesiveness and develop confidence on the
part of individual members, via facing adversities together. They can
vary in length and intensity, but the essence is that people are put in
very scary situations (though not much risk of actual physical damage)
and asked to confront their fears with a special emphasis on trusting
their peers and the professionals in charge. I participated in such courses
in high school and worked at such a course as an undergraduate. Our
“alpine tower,” a five-story tall log timber tower, required people to climb
a series of independently free-swinging, vertically hanging logs to get to
the top. While supporting people with a rope on this obstacle, I have
had people cry and freeze, unable to proceed for long periods of time.
When they eventually reach the top, they have a different conception
of themselves along various dimensions. Post-session surveys showed
people cherished the transformative aspect of the adversity.
Similar to the challenge course but with greater risk of harm is the
“adventure trek.” In the adventure trek, one leaves behind the safety of
the artificial course and turns to the mountains, rivers, and seas to find
character-strengthening adversity. I have led such treks for over 20 years.
Looking through news files just now, I see that at one of the Boy Scout
ranches I have led treks on there was a bear attack at 5:30 this morning,
the second in a week. The scout – a 14-year-old boy, the same age as my
daughter – “suffered a bite to the hand and a deep scratch to the head,”
but was otherwise okay.4 Others have not been so lucky. This is a particu-
larly revealing case from my point of view. For I have a map of that ranch
on my wall as a reminder of what a formative experience my time there
was. A huge part of who I am – and now who my kids are – was forged
in that ranch. It is really hard to overstate the influence of my outdoor
adventures as a Boy Scout on who I am today. And though I, like the
vast majority of Scouts, came out relatively unscathed, I was a part of a
milieu or system or ensemble that was virtually guaranteed to result in
tragic loss of young life on a regular basis. All such accidents could be
4
Raton Range (2010, July 8). My very brief search also turned up a North
Carolina scout who died from falling off a cliff in a hiking/climbing area a few
months ago (WBTV 2012, September 23), a 9-year-old cub scout who was killed
earlier this year falling into a ravine in a cave area I have been to (Ruzich and
Sadovi 2012, May 21), and many similar stories.
prevented by restricting access to risk, or by hedging in our agency. But

not only would that result in the loss of the great good of the exercise
of that agency, it would make impossible great goods of personal being,
great goods that are integral to valuable human lives.
I cannot offer an analysis of what exactly is going on in the kinds of
cases I have described. All I can do is try to get the reader to see that our
practices support the permissibility and goodness of a certain kind of
endeavor, and to suggest that, from the right vantage point, the world
according to Irenaeus sufficiently resembles, in relevant respects, that
kind of endeavor.
But does it all add up? Is it worth it in the end to give up so many
weekends to racing, to trade so much potential pleasure for so much
guaranteed pain? There are six riders in my family, and it seems
unlikely to me that we will all escape serious injury. That is equivalent
to saying that I find it more probable than not that one of us – myself,
my wife, or one of my four beloved children – will be seriously injured
in a bike accident. We also rock climb, downhill ski, kayak, and do
whitewater rafting. Every year people die in each of these activities. I
ask again: Is the risk worth it? The only answer I have to offer is that
apparently it is, because we show no signs of quitting. Nor does it seem
irrational to continue. It is a calculated risk that I make for myself
and for my family. I firmly believe the probable benefits outweigh
the potential risks. But I can only do this from within the framework
of a value system that puts ease and comfort well below courage and
determination. The value system one operates in will largely deter-
mine how plausible they find a theodicy. This cannot be emphasized
enough.
A final thought: Such matters as risking serious injury in outdoor
adventure – bones being broken, people being killed, paralyzed,
drowning, dying of starvation, or being killed by a bear – bad though
they be, are far from the worst that can happen in a law-governed, solid
world of creatures with the powers and freedoms we have. In this world,
one risks being tortured physically or psychologically in ways that can,
at least for a time, shatter one’s consciousness. Is it worth the risk to live
in such a world? Was it permissible for God to allow this risk? If it is, it
is so for the same kinds of reasons it is worth it to Fiona to take the risks
she does and for me to allow her to do it and for the reasons it is worth
it to me to face challenges I do in the wilderness and to lead others
into those challenges. And of course, I believe there are goods much
greater than the goods to be obtained through the challenge of real-
life outdoor adventure. These, we all know, come with correspondingly
much greater risk. The picture will be dramatically altered if God has
a way of guaranteeing that in the end, all reasonable creatures think
that their ultimate fate was worth God’s risking the evils that actually
occurred to them.
What I have been trying to do in this section is some important ground
clearing and some important preparatory groundwork. To plant a tree in
weedy or unprepared soil is bad practice and drastically decreases the
probability of that tree growing to maturity.
6.2 Axiology and teleology
Having cleared some brush for the soul-making theodicy to be presented

and critiqued in a non-crass manner, I begin by noting Hick’s emphasis
that the Irenaean soul-making theodicy is teleological. That is, it looks
forward to the fulfillment of God’s purposes to explain present evil rather
than, like the Augustinian view he places it in opposition to, looking back
to the Fall for the explanation. This will give rise, in the next chapter, to
a teleological argument for God’s existence, from the fine-tuning of the
universe for sainthood. Given the structure of the Bayesian argument
from evil in Chapter 3, the theist must show not only that evil doesn’t
count against theism, but that it is at least as probable on theism as on
naturalism.5
Hick’s presentation of the soul-making theodicy begins from within
the Christian revelation. Specifically, it is bathed in quotations from the
Christian Scriptures about God’s purpose for humankind being to make
of humans “sons” or “heirs” of God (Romans 8:14, 17, Galatians 3:25–6,
4:6–7, Hebrews 2:10–11, John 3:1). As I said in Chapter 1, I am operating
in three different modes at once. I am considering the consequences
of the data of evil from the standpoint of “bare” theism; of “merely
Christian” theism, somewhere between a “robust” Christian theism
and “bare” theism; and of the non-theist. However, as I pointed out in
Section 1.3, I think there is risk of blurriness here, because “bare” theism
isn’t so bare when some of its a priori consequences are teased out, and
“bare” theism conditioned with historical evidence already in our back-
ground probabilifies mere Christian theism. Thus, the line between
“bare” theism and mere Christian theism is fuzzy. So I will continue to
work “elastically” with both notions in mind. It is not hard to adjust the
arguments for degree of strength within these elastic limits.
5
More precisely and holistically, it must show that any disadvantage is more
than off-set by the contribution of natural theology.
Nevertheless, I don’t think the soul-making theodicy needs anything

by way of special Christian assumptions to work. Though Hick shows
there is amply sufficient affirmation of it in the Christian Scriptures,
such affirmation is not necessary to motivate it. For it is a datum that a
world where the highest virtues are displayed is a world with great value
in it (that value could be offset, perhaps, by the existence of existences
that were on the whole not worth living, but it is impossible to have
any significant evidence at present that anyone will ever have such an
existence). And, given this, it follows that an omnipotent, omniscient
being will have strong reason to bring about such a world. So the only
question that remains is whether the present world-ensemble is a good
means to that end.6 Hick’s teleological approach is hand-in-glove with
the nature of personal explanation. God as conceived in bare theism
is a personal agent who always knows perfectly what the best kinds of
states of affairs are, desires that such states obtain, knows the best way to
bring it about that they do, and has the power to execute any logically
possible plan. So God can be expected to bring about effective morally
permissible means to the best ends. The question, in the confrontation
between theism and naturalism, is whether what we find in the world is
more amenable to personal explanation or “chance” explanation (i.e. no
explanation). And that question just is the question whether the world is
apt to realize the highest kinds of goods.
Thus it is crucial that we get a picture of the virtue-based axiology
in concrete terms. Within my own tradition, I think of the thousands
of canonized saints, and especially some particularly heroic ones. And
the power of even one saint should not be underestimated. A single
life lived in accordance with love can have a magnitude of ramifica-
tions. In our own time, Mother Teresa – known to Catholics as Blessed
Teresa of Calcutta – is the standard example, yet still not weakened
with use. She is not yet a canonized saint, but she will be, and there
are countless nameless unofficial saints like her. A saint I am espe-
cially fond of is Father Damien – Saint Damien of Moloka’i – a Belgian
priest canonized just before my birthday in my last year of graduate
school. He spent 16 years in service to lepers on a remote island in the
6
I’m talking about whether the making of a world with the exercise of morally
significant freedom is a good means, not whether particular instances of evil are.
We send our kids to camp as a means to the end of some kind of transformation,
but we don’t ordinarily cause particular events to happen as means to that end.
Likewise, God places us in a world he knows will involve significant suffering,
but he does not intend (at least routinely) any of the particular evils that occur. A
good research project is to apply the doctrine of double effect here.
Pacific, part of the Hawaiian island chain. Eventually, he succumbed

to the disease himself. Another saint who has captivated me is Saint
Maximilian Maria Kolbe. Beatified just 10 days before my birth, he
died in Auschwitz. He was sentenced there for sheltering thousands of
Jews. And when, in Auschwitz, a man was being taken to be starved to
death in a hole as an example, Saint Maximilian volunteered to go in
his stead. These examples of self-sacrifice are but three recent examples
of the countless beautiful acts by those who have given themselves to
God and mankind.
Likely you have never heard of these latter two saints, but there
are hundreds of such “martyrs of charity” recognized and count-
less more unrecognized like them. Their lives add to the value of the
world immensely. The best kinds of worlds are those that allow for
such individuals to exist. God did not create Auschwitz in order for there
to be saints. That is not the kind of teleology involved in the saint-
making theodicy at all, far from it. He created a world in which the
morally significant use of freedom did not rule out such things, not
just because the value of the exercise of morally significant freedom
is so valuable in and of itself (though it is very valuable), but because
it allowed for a world in which we could become heirs of God, saints.
Perhaps God got “unlucky” and ended up with a world on the rougher
edge of the spectrum within the range of permissible Saint-fostering
worlds. Yet our world contains far more saints than suicides.7 This is
a sort of barometer for whether the world is in the right range. But
the working out of where in this range the world will fall when all is
said and done must be by human acts, for it is by human acts that the
greatest (created) goods in a world will or will not be achieved.8 This
will rule out much intervention on God’s part (and note well that we
have no idea how much he does intervene). He has, for the most part,
handed over human wellbeing to us humans for the time being. Hick
expresses this so well he is worth quoting at considerable length, for
there is no more concise treatment of this issue than the following
three paragraphs.
7
It could be that many are kept from suicide only by fear or guilt, but I don’t
think that will affect the numbers much. And I can fit despair into theism much
more easily than I can fit deliverance into atheism. The theist can see despair as
authentic, whereas the atheist must “see through” deliverance. See also Swinburne
(1998: 246–247).
8
I bracket here Plantinga’s “O Felix Culpa” theodicy according to which the
greatest good is the Incarnation of God the Son in Jesus Christ (2004: 25). This is
a suggestion worthy of much consideration.
Suppose, contrary to fact, that this world were a paradise from which
all possibility of pain and suffering were excluded. The consequences
would be very far-reaching. For example, no one could ever injure
anyone else: the murderer’s knife would turn to paper or his bullets
to thin air; the bank safe, robbed of a million dollars, would mirac-
ulously become filled with another million dollars (without this
device, on however large a scale, proving inflationary); fraud, deceit,
conspiracy, and treason would somehow always leave the fabric of
society undamaged. Again, no one would ever be injured by acci-
dent: the mountain climber, steeplejack, or playing child falling from
a height would float unharmed to the ground; the reckless driver
would never meet the disaster. There would be no need to work, since
no harm could result from avoiding work; there would be no call to
be concerned for others in time of need or danger, for in such a world
there could be no real needs or dangers.
To make possible this continual series of individual adjustments,
nature would have to work by “special providences” instead of
running according to general laws that men must learn to respect
on penalty of pain or death. The laws of nature would have to be
extremely flexible: sometimes gravity would operate, sometimes not;
sometimes an object would be hard and solid, sometimes soft. There
could be no sciences, for there would be no enduring world struc-
ture to investigate. In eliminating the problems and hardships of an
objective environment, with its own laws, life would become like a
dream in which, delightfully but aimlessly, we would float and drift
at ease.
One can at least begin to imagine such a world. It is evident that
our present ethical concepts would have no meaning in it. If, for
example, the notion of harming someone is an essential element in
the concept of a wrong action, in our hedonistic paradise there could
be no wrong actions – nor any right action in distinction from wrong.
Courage and fortitude would have no point in an environment in
which there is, by definition no danger or difficulty. Generosity,
kindness and the agape aspect of love, prudence, unselfishness, and
all other ethical notions which presuppose life in an objective envi-
ronment could not even be formed. Consequently such a world,
however well it might promote pleasure, would be very ill adapted
for the development of the moral qualities of human personality. In
relation to this purpose it might be the worst of all possible worlds!
(Hick 1973: 41–42. See also Swinburne’s similar thought experiment
(Swinburne 1998: 249).)
It all comes down to what we take God’s purpose to be. Hick rightly
places a strong challenge on the one critiquing the setup of this world
to describe how things could have been significantly different without
making things worse (cf. van Inwagen 2006). Every world with significant
virtue is a world with significant pain. Every world without significant
pain is a world without significant virtue. Significant good and evil are
in this regard a “package deal.” The evil is not a means to the good, but it
is a cost of there being that good. It is simply not possible to manipulate
the world in such a way as to achieve some mythical “optimum balance.”
Consider Plantinga’s notion of universal transworld depravity (Plantinga
1974b: 184ff). Plantinga presents it as a thesis that is true for all we know.
I think it is quite plausible, given the nature of finite persons, that the
inclination to selfishness is inherent (and to degrees that differ less than
we would like to think) in all persons worth creating. Thus, the degree of
badness of worlds it is permissible for God to create does not vary radi-
cally. This suggests a kind of “butterfly effect” problem for those inclined
to think that God could intervene here and there to make a world with
just as much value over the long run with lots less evil over that same
period. The limits of possibility space affect the evidential formulation of
the problem of evil just as much as the logical formulation.
At the very least, it is obvious there are no hard boundaries: There is
no best world, nor is there any world with a least amount of suffering
necessary (van Inwagen 2006). So the question “Why not just one less
horrendous evil?” is confused from the start (see also Hasker 2004:
39). There are innumerable possible outcomes of a world-ensemble
consisting of free agents with significant power to harm and to heal in
a world of fixed natural laws. Theism predicts we would fall somewhere
in this range. It does not distinguish where in this range the world will
fall. The world-ensemble from which we have emerged has given rise to
wars and heroes and diseases and nurses and hunger and charity and
bullies and teachers, sinners and saints. In short, it gives rise to consider-
able opportunity for virtue, much of which has been taken, and, sadly,
much of which has not. And there is no good reason to think that the
removal of one evil here would not result in the addition of an approxi-
mately equal evil over there. This is a very special kind of world, one as
rare as prime numbers, perhaps, and that there should be such a world
is entailed by God creating anything at all.9 In the next chapter, I will
9
God’s creating is contingent. However, so is my eating a fresh, warm cookie
that has been offered to me. Something can be contingent and still be a “slam
dunk.”
take these thoughts and put them into a fine-tuning argument from evil
for theism. But for now there is more to say about sainthood and the
defeat of evil.
6.3 The defeat of evil
One striking feature of many canonized saints is that not only do they
not regret their suffering, they seem to desire it. Consider this rather
strong statement often attributed to St. Ignatius of Loyola.
If God gives you an abundant harvest of trials, it is a sign of great holi-

ness which He desires you to attain. Do you want to become a great
saint? Ask God to send you many sufferings. The flame of Divine
Love never rises higher than when fed with the wood of the Cross,
which the infinite charity of the Savior used to finish His sacrifice. All
the pleasures of the world are nothing compared with the sweetness
found in the gall and vinegar offered to Jesus Christ. That is, hard
and painful things endured for Jesus Christ and with Jesus Christ.
(Ignatius of Loyola 2009)
This is quite typical, indeed, even tame by comparison to the related

sentiments of many saints. In passage after passage, saints who truly
suffered seem to relish it, in a sense. I will not be defending this desire,
but I do wish to learn from it. The strand I wish to use from this cord is
that saints endorse their suffering as essential parts of the fabric of their
lives. Suffering is a sine qua non of who they have become in a way they
accept as integral to the persons they have become. This helps us see
that the saint-making theodicy is not a crude or crass theory that entails
that suffering is a mere means to some greater good. It is a means in a
sense, but only in that it is partly constitutive of a certain kind of value.
True love is not the reward for the many trials of a marriage; the trials
are partly constitutive of true love.
The greatest of saints, in my tradition, is Mary. The characteristic
Marian attitude is expressed in her “fiat” that echoes the primordial
“fiat.” She says to God “Be it unto me according to thy word” (Luke 1:38,
AV). Her Son expresses this same Marian virtue in Gethsemane. “Father,
if thou be willing, remove this cup from me: nevertheless not my will,
but thine” (Luke 22:42). This “Marian virtue” or “Gethsemane virtue” is
the essence of saintliness: a turning over of the will to God, and willing
embrace of what one wishes not to do, trusting that an all-knowing, all-
good God has their good and the good of all creation in mind.
Jesus’s words acknowledge the badness of what he was to face and even
a desire not to face it. The saint-making theodicy does not call evil good.
He both wants and doesn’t want to face his trial. And that is surely how
it is in retrospect for those who have suffered greatly: They both wish it
never happened and don’t wish it never happened. Both Mary and Jesus
knew they were part of a Divine Drama they didn’t fully understand.
Before their earthly stint was over, they both suffered terribly. Yet if the
Christian story is true, no doubt they both are glad to have played a
role, any role, in the drama of salvation. If the Christian story is true,
then they don’t regret their “Let it be” at all, even while recognizing the
intrinsic badness of their suffering. Neither, in the end, doubts that God
has been good to them.
I have suffered very little in my life, barely even worth mentioning.
But in the minute suffering I’ve encountered, I’ve caught a glimpse of
how it has been integral to what modicum of saintliness I have. On one
occasion, I spent a long day in jail. Supposedly innocent until proven
guilty, I was treated like scum. I was held in a six-by-six foot solitary
confinement cell (easy to measure as I am just a little over six feet tall)
for hours before I was processed. Then my belt, watch, and shoes were
taken and I was given “flip-flops” to wear instead. Without exception,
the officers seemed to assume I was a murderer or something given the
way I was dragged, pushed, barked at, and otherwise neglected (the food
wasn’t terrible though, I have to admit). Fortunately for me, I had a close
friend who was a lawyer and plenty of money to make bail before having
to spend the night. Most of my cellblock mates were not so lucky. It
was a miserable, humiliating experience with both mental and physical
suffering. Time deprivation, for me, was like psychological torture and
the handcuffs left marks on my wrists for weeks.
On another occasion, I broke my hand off, snapping the radius and ulna
right at the joint. After hours of reconstructive surgery, I had considerable
hardware sticking out of me, what they call an “external fixator” (google
it, if you dare). People stared at me and grimaced. Other’s quickly turned
away. Children could be heard asking their parents what happened. People
didn’t want to converse with me, as they always had before. I got a small
taste of what it was like to be an outcast or someone with a permanent
disability (I still suffer considerable pain and have reduced mobility, but I
can cycle again). After the metal was removed, I endured months of phys-
ical therapy where, day by day, I gradually tore lose the scar tissue a millim-
eter or so at a time on machines that were very much like torture devices.
Yet I am happy for this solidarity with those brothers and sisters that suffer
more deeply than I did, gruesome and costly as the experience was.
I would not undo suffering that injustice and the other attendant
sufferings. I am glad I went through them. They made me who I am
and, though I am clearly not done yet, I am a much better person for
these experiences. Do I hope that I suffer more that I might be a better
person yet? At one level, of course, no, for I fear suffering. But when I
can muster the moral courage, yes, I hope that I have further experi-
ences that make me more saintly, more a lover of all persons human and
divine. I am like Augustine saying, roughly, “Lord, make me pure, but
not yet” (2006: Confessions, Book 8). In the end, only God knows what
I can take.
Can I possibly expect someone who went through the Holocaust to be
able to say they were glad it happened? No, I cannot possibly do that. It
is psychologically impossible for me to fuse those two concepts in one
thought. But I do find Auschwitz survivor Viktor Frankl’s response to the
Holocaust inspiring and illustrative (Frankl 1985). His life, and many he
documents, are a testimony to the human ability to find meaning under
any conditions. It seems clear he would not have wished it away if it meant
losing the enlightenment he found there.10 Many, of course, did not find
10
And his own suffering cannot be underestimated. Perhaps the most famous
passage in his most famous book contains a testament to the love of his wife.
We stumbled on in the darkness, over big stones and through large puddles,
along the one road leading from the camp. The accompanying guards kept
shouting at us and driving us with the butts of their rifles. Anyone with very
sore feet supported himself on his neighbor’s arm. Hardly a word was spoken;
the icy wind did not encourage talk. Hiding his mouth behind his upturned
collar, the man marching next to me whispered suddenly: “If our wives could
see us now! I do hope they are better off in their camps and don’t know what
is happening to us.”
That brought thoughts of my own wife to mind. And as we stumbled on for
miles, slipping on icy spots, supporting each other time and again, dragging
one another up and onward, nothing was said, but we both knew: each of us
was thinking of his wife. Occasionally I looked at the sky, where the stars were
fading and the pink light of the morning was beginning to spread behind a
dark bank of clouds. But my mind clung to my wife’s image, imagining it with
an uncanny acuteness. I heard her answering me, saw her smile, her frank and
encouraging look. Real or not, her look was then more luminous than the sun
which was beginning to rise.
A thought transfixed me: for the first time in my life I saw the truth as it is
set into song by so many poets, proclaimed as the final wisdom by so many
thinkers. The truth – that love is the ultimate and the highest goal to which
man can aspire. Then I grasped the meaning of the greatest secret that human
poetry and human thought and belief have to impart: The salvation of man
is through love and in love. I understood how a man who has nothing left in
enlightenment there and did lose their faith (though some found it11).
But it is no part of theism that such enlightenment comes only or mostly
in this life. Rather, it is a part of theism that there will be an afterlife to
continue the process. And the afterlife is not an auxiliary hypothesis. For
there is no possible world in which creatures suffer and are not given full
opportunity for compensation and recovery. The afterlife is “built in,” a
logical concomitant of theism in worlds with suffering.
What my own little experiences illustrate in a little way and what
Frankl’s life and those he and others chronicle illustrate in a very large
way is the concept of the defeat of an evil. Chisholm introduced the term
though the general idea surely predated him.12 It is similar to his notion
of epistemic defeat. Chisholm’s formal definition of defeat in the realm
of value is very complicated, but we can get the gist of it by bridging from
his notion of epistemic defeat. Let h be some hypothesis and e and d be
this world still may know bliss, be it only for a brief moment, in the contem-
plation of his beloved. In a position of utter desolation, when man cannot
express himself in positive action, when his only achievement may consist in
enduring his sufferings in the right way – an honorable way – in such a posi-
tion man can, through loving contemplation of the image he carries of his
beloved, achieve fulfillment. For the first time in my life I was able to under-
stand the meaning of the words, “The angels are lost in perpetual contempla-
tion of an infinite glory.” (56–57)
“Listen, Otto, if I don’t get back home to my wife, and if you should see her
again, tell her that I talked of her daily, hourly. You remember. Secondly, I
have loved her more than anyone. Thirdly, the short time I have been married
to her outweighs everything, even all we have been through here.” (76)
She died in Bergsen-Belsen concentration camp. His mother was gassed, his
father died in camp, and his brother died in the mines.
11
Consider the following testimony, of which much more could be produced.
“Why hold fast to the God of the covenant? Former believers lost Him in the
Holocaust Kingdom. Former agnostics found Him. No judgment is possible.
All theological arguments vanish. Nothing remains but the fact that the bond
between Him and His people reached the breaking point but was not for all
wholly broken. He (a survivor) is a witness the like the world has not seen”
(Fackenheim 1973: 171).
“Loss of faith for some equaled discovery of God for others” (Wiesel 2004: 6).
“The truth is that among those who actually went through the experience of
Auschwitz, the number whose religious life was deepened – in spite, not to say
because, of this experience – by far exceeds the number of those who gave up
their belief” (Frankl 1975: 16).
All three quotes appear in Williams (2000).
12
It is reminiscent of certain holistic ideas of Leibniz.
different items of evidence. Let e be evidence that confirms h and let d be

some further evidence. Here is Chisholm’s definition of epistemic defeat.
Epistemic defeat d defeats e’s tendency to make h probable =Df e tends

to make h probable; and d&e does not tend to make h probable.
(1989: 55)
It is easy to see how this works with an example. That (e) your high
school teacher says that (h) the civil war was all about slavery is a good
reason to think that it is so. But when (d) your college history professor
says there were other factors, your new total evidence (e&d) no longer
supports h. The stronger testimony epistemically defeats the support of
the weaker testimony.
But note that there is a very different way that defeat can work. Begin
again with your high school teacher’s testimony. That is, let e and h
have the same values as above. But this time consider that you learn (d’)
that your high school teacher never even graduated from high school
and has been discovered to have been making up all kinds of things.
Again, this is a case of defeat, because though e confirmed h, e&d does
not. However, this time it is not because d’ is stronger than e. Rather, it
is because d’ shows that e was misleading in the first place. In a sense, in
light of d’, e no longer counts in favor of h.13 The former kind of defeat is
usually called rebutting defeat. The latter is usually called undercutting or
undermining defeat. Now let’s look at the parallel notion in value theory.
Let g be some good, e be some evil, and L be a particular life.
Axiological defeat g defeats e’s tendency to make L bad on the whole

=Df e tends to make L bad on the whole; and g&e does not tend to
make L bad on the whole.
So suppose (e) I wreck my car but (g) my insurance buys me a new, better
car without raising my rate (and I have no deductible) by selecting my
name at random in a drawing to receive this benefit. By itself, wrecking
my car is a bad-making feature of my life. But when taken together with
the new car, I’m on the whole better off. But just as there are two kinds
of epistemic defeat, there are two kinds of axiological defeat. Compare
the former case to a case in which due to the car wreck, I am paralyzed
13
And in a sense, it still does count against it, since relations of epistemic
support are quasi-logical relations that hold necessarily when they hold at all.
Thus, the “in light of” condition is needed.
from the waist down. As a result, I come to depend heavily on my family

from whom I had been previously estranged. Through the necessity of
relying on them I am humbled and drop my weakly grounded ill will
toward them and come to love them and even be shaped by them into a
person of great generosity myself. In this case, the defeat does not seem
to be adequately explained merely by saying the value in my healed
relationships and forged character is of greater value than the lost car
(though it surely is). In this case, the car crash is a crucial component in
the narrative of my identity. It is an essential part of the story of how I
became who I am. It is a dramatic component in radical transformation.
So that evil, while remaining an intrinsic bad, no longer counts against
the value of my life.
At work here is a form of partial incommensurability. For though we
can say that the value of a transformed life is greater than the value of
a car, there is no fact of the matter about how many times more it is
worth. A whole new kind and level of value has been reached. A ladybug
is great, but an angel is greater. But there is no amount of ladybugs that
would be of more value than one angel.14 Nor does the attempt to make
a measurement of their relative value make sense.
Marilyn McCord Adams (1999) adopts and adjusts Chisholm’s notion
of the defeat of evil and recognizes this incommensurability. She states
the contrast this way:
The balancing-off [essentially, compensation] relation is arithmetical

and additive: value parts are balanced off within a larger whole if other
parts of opposite value equal or outweigh them. Alternatively, value-
parts may be integrated into a whole to which they bear (in Moore’s
words) “no regular proportion” via relations of “organic unity.” (21)
This allows Adams to give an informal definition of the defeat of evil.
Defeat, informal E is defeated within the context of the individual’s

life if the individual’s life is a good whole to which e bears the rele-
vant organic unity ... by being relevantly integrated into x’s relation
to a great enough good. (1999: 28, 29)
Adams is clear that this meets two of her desiderata: 1. It ensures not just
that God is just but that he is good; 2. That evil is not just systemically
14
This is based on an example Todd Buras raised in a talk by Mark Murphy at
Baylor Universitiy.
defeated but in a way that (i) accrues to and (ii) is at least partially
recognized by the one who suffers the horrors (55, 156). Of course, the
individual need not recognize and appropriate the defeat right away.
Indeed, since the integration is almost always a gradual process, it is
impossible to notice all at once. It could dawn on one or one could
notice it happening, but full appreciation only happens in hindsight.
Since it does not appear to happen to most people in this life (though we
have little to no evidence about what happens at the moment of death),
we can assume that most of the work gets done in the next phase of
existence. As Hick (1997) says, we march under the banner “No theodicy
without eschatology” (48). What is important is that the agent herself
eventually comes to have a perspective on her life in which she endorses
the events that have constituted her path to virtue. This is an important
part of Adams’ picture from start to finish, and I endorse it.
[R]etrospectively, from the vantage point of the end of the journey,

the person one eventually becomes would be glad to have made the
sacrifice [of being a participant in the horror]. Participation in horrors
can thus be integrated into that overall development that gives posi-
tive meaning to his or her life, and so be defeated within the context
of the individual’s existence as a whole. (1999: 53)
Retrospectively, I believe, from the vantage point of heavenly beati-
tude, the victims of horrors will recognize those experiences as points
of identification with the crucified God, and not wish them away
from their life histories. (1999: 167)
A few clarifications are called for here. The first quote attributes to the
horror participant the property of being glad for their participation in
horror, while the second attributes the property not wishing away their
participation in horror. These are two different properties, though the
former plausibly entails the latter. Furthermore, as I noted above, one
can both wish away and not wish away a past event. I have a hard time
endorsing being glad for as universal to the defeat of evil, at least if it is
unequivocal gladness (though see Robert Adams 1987 and Hasker 1981).
Gladness is a paradigm case of defeat, and as I have said, I myself am glad
for some of the worst things I’ve suffered. But then I haven’t suffered
much. I think that one can exemplify the latter property without exem-
plifying the former property and that that latter property is sufficient for
the defeat of evil. It is said “Time heals all wounds” and perhaps even-
tually this is true. (What will Holocaust survivors think a billion years
of paradise hence? I really don’t know.) On the other hand, virtue is its
own reward, so perhaps there is nothing one can’t eventually become

glad for if it is integral to their affirmed identity, which is a real good.
This is not wholly implausible, but I will not assert or assume it. I will
assume instead that when, due to one’s affirmation of one’s identity, one
does not wish away one’s participation in horror – participation that was
integral to the formation of one’s identity as a truly virtuous person –
that the evil is defeated.15
A second difference between the two quotations is that the first says
that evils can be integrated into the goodness of one’s life while the
second says they will be. This raises a difficult question as to the criteria
of success for a theodicy. Is it sufficient that for each evil suffered God
ensures that that evil is capable of being defeated or is it necessary that
God guarantee that it will be defeated? The former seems too weak. If
I drop you in the midst of a maze with many deadly dangers, it is no
consolation that there is a possible route of escape. Possibilities must
be relevantly available. Yet the latter seems too strong. Since the defeat
of evil requires the use of the subject’s own agency, it seems that God
could not guarantee the defeat of evil in a way consistent with morally
significant free will.
I favor a middle way. Here is a rough sketch of the outlines of the kind
of view I favor. What the goodness of God entails is that he would only
permit suffering which is objectively “worth it” to any well-functioning
rational person from their perspective (and recall that God has some
control over what kinds of people come to exist, and also knows people
better than they know themselves). To illustrate the idea, consider that
we have no sympathy for people who turn down advantageous bargains
for no good reason (even if they are first-person rational according to
an expected utility function which accepts perverted utilities and wildly
inaccurate probabilities). We say things like “If they had a lick of sense,
they’d take the deal.” That is, we judge them dysfunctional in some way
when they cannot see that the deal is “worth it.” Consider this scenario.
Someone tells you that they will match – at a ratio of 1,000,000:1 –
donations you solicit for starving children in sub-Saharan Africa. That
is, if someone gives you $1 to benefit the children, the sponsor will add
15
Hasker’s (1981) treatment of the logic of regret is not wholly successful, but it
is powerfully suggestive. What seems clear to me is that future researches should
look for a plurality of pro and con attitudes we have toward our suffering. So long
as there is some sufficiently strong con attitude toward our suffering, the badness
of that suffering will receive its due; so long as there is some sufficiently strong
pro attitude to our lives, God will have been sufficiently good to us.
$1,000,000 to it. So you go to a wealthy friend and ask for a $100 dona-
tion. They refuse, saying that that would cause them to go without a
really nice meal. Even if we assume, contrary to what I suspect, that they
are under no obligation to give, we will issue a negative judgment on
them. We will judge them as miserly and uncaring, as vicious in various
ways. In like manner, if, after reviewing one’s earthly life one would
not join their will to harms that were objectively “worth it,” we would
judge them and not God. A full treatment of this issue would involve
issues of autonomy and consent. For example, does the Creator require
the consent of the creature in this particular kind of case? I think not. Is
there a kind of tacit consent involved in living in community with other
human beings? Perhaps. Further work needs to be done here, and I hope
some readers will do it.
It could even be that if Molinism is true that God uses his middle
knowledge to ensure that people are only permitted to suffer harms
he knows they will endorse when they understand their role. A similar
move is available to the Boethian.16 These and other important and
interesting issues pertain to details of the defeat-enriched soul-making
theodicy which we must leave for another occasion.17 If you don’t like
16
Eleonore Stump plausibly interprets Aquinas as holding that “a person of
faith has in effect given consent to living a life which includes suffering” (Stump
2012: 405). And, in a sensible way of looking at it, whether a person is a person
of faith is something that, from God’s perspective, does not change throughout
a person’s life. So if, from our perspective, they become a person of faith, in the
relevant sense, only long after death, then the whole continuant person will be
considered as a person of faith for God and their consent will be “retroactively”
applied to their whole life, including the portion of their career where, from our
perspective, they did not appear to be, perhaps even to themselves, a person of
faith. Swinburne, who is not a Boethian, says this: “But of course I cannot choose
[before I am born], and God has to make the choice for me. God sometimes pays
us the compliment of supposing that, if we had the choice, we would choose to
be heroes” (1998: 250). These matters deserve more investigation.
17
Aquinas seems to hold an interestingly related position. He thinks that greater
excellence of character (in at least a dispositional way) leads God to allow that
person to suffer the greater, much as a general sends his best men into the worst
battles. Stump glosses this passage saying “Presumably, part of what makes such
persons better is that they would be willing to accept greater suffering for the
sake of greater glory” (Stump 2003: 577, n. 52). On the view I sketched, people
are better in proportion to how much they are willing to suffer for others. Perhaps
this can reach pathological proportions. If so, then I take it there is still some
threshold such that unwillingness to sacrifice that amount for a specified benefit
for others is vicious (lacking in empathy, compassion, love) and that willingness
to suffer more than that amount is increasingly virtuous up to but not including
the point of pathology.
this line of thought, there is surely some reasonable constraint on the

agent that provides a middle way between merely requiring defeat to be
possible and guaranteeing it occur. In simple terms, what the triumph
over evil shows is that evil never needs to be a soul-destroyer. It may
temporarily – even if for an earthly lifetime – flatten someone, but it
need not have the final word, no matter how potent.
Before this chapter ends, I just want to briefly point out that this idea
is not just the provenance of philosophers. The phrase “It made me who
I am” is ubiquitous in pop culture. It is the central theme of R&B artist
Faith Evans’ hit song “Again” (she was abandoned as a child and her
husband – rapper Notorious B.I.G. – was killed in a drive-by shooting).
A Facebook page with nothing but the quote “I never regret the past
because it made me who I am today” has 17,716 “likes” at present
(https://www.facebook.com/noregrets270). In my sometime role as lay
counselor to the grieving, I run into this thought again and again. And,
importantly, from the mouths of those who have suffered the most.
These people believe, and I am in no position to say unreasonably, that
God has been good to them. I confess I used to be annoyed by this, but it
seems to be a nearly ubiquitous human response, and I think it was only
haughty high-mindedness that made me doubt it.
I myself never understood the human capacity for suffering until I
had children. I have a friend who lost a daughter to childhood leukemia
when she was five (the age of my darling youngest at present). We had
never talked about it, but I felt I must know how he processed this
(he clearly lived a joyful Christian life). So I told him I had something
important to talk about, something urgent. I had him drive us out into
the country, suspense and tension building. I can’t imagine what he was
imagining this was all about. Finally, I told him to pull over so we could
talk. After much hesitation, I finally asked him how he managed to go
on after the tragedy. His expression was blank for a few seconds, and I
worried I had asked an incredibly offensive question, then his face broke
into a huge grin, “Suffering is huge,” he said, “but God is bigger.” That
is not what I was prepared to hear. Had I heard it from someone else,
I might have found it trite. But I knew the intellectual integrity of his
man and I knew many of the details of what he had been through. And
like so many, their faith grows stronger through suffering. This is not
always the case, but it is proof that terrible evil can be defeated in one’s
life. And it is evidence that, given enough time and understanding, evil
can be defeated in the full postmortem lifetime of any individual.
7
The Saint-Making Theodicy II:
Positive Phase
The last chapter concerned the defeat of evil. Given theism’s capacity to
defeat evil from within its value system, evil will not count against God’s
existence. For theism is only in tension with indefeasible evil (which is
probably a better way to characterize what has been called “gratuitous”
evil). And we have no evidence whatsoever that there is any indefeasible
evil. However, though this is our theodical foundation, it is not enough
to fully answer the Bayesian argument from animal suffering. For, on the
face of it, the suffering is just what we’d expect given naturalism. And
failing to predict the opposite of the data is not on par with predicting
the data. So in this chapter, I take the considerations of the last chapter
and present in Section 7.1 the fine-tuning argument for theism from
evil. Its structure is much like the fine-tuning argument for theism from
the initial conditions of the universe. In Section 7.2, I make its appli-
cation to the Bayesian problem of evil from animal suffering perfectly
explicit. In Section 7.3, I consider whether there is any evidence that the
amount of suffering observed is definitely beyond the range of expecta-
tion. I finish, in Section 7.4, with some exploratory reflections on how
we can assess what worlds to expect based on what kind of story the
lives of the sufferers amount to.
7.1 The fine-tuning argument for theism from evil
On the one hand, it would be really surprising if chance turned up the

best possible world. But if there were a best possible world, God would
be obliged to create it. Thus, its existence would be excellent confirma-
tion for theism. But there is no best possible world. This makes gener-
ating expectations a bit harder. But it does not make it impossible. There
is no best to work with, but that still leaves us good and better.
118
The Saint-Making Theodicy II: Positive Phase 119
God would, at a minimum, have to do better than chance. Better than

chance would be pretty good. But we expect more than pretty good
from God. We expect the best where there is a best and a sufficiently
high level of goodness if there is not. Since there are ever better worlds,
we cannot sensibly discuss a range of worlds in the way we might if there
were a finite number of them. But there are reasonable ways to proceed.
If there is a best kind of world, we would expect that kind of world given
theism, but it would be quite surprising to find ourselves in the best
possible kind of world, given the range of possible outcomes, on the
assumption of a chance outcome.1
Even if the Cantorian cardinality of the primes is equal to that of the
non-primes, it is still sensible to say that the probability of choosing a
non-prime at random is greater than the probability of choosing a prime.2
And even though there is a bijection from good worlds to bad worlds,
it is sensible to say that the a priori probability of a bad world is greater
than that of a good world. “Cantor’s Paradise,” exquisite though it be,
has proved at best purgatorial for a semantics for logical probability. I
will take on board the assumption that a natural probability measure has
available a sensible notion of “more” according to which there are more
non-primes than primes, for I think the assumption is also necessary for a
confirmation theory which makes sense of our best science. That is, good
scientific inferences can only be rationally reconstructed on the assump-
tion that there is such a natural non-Cantorian probability measure (or
one which has the same results). And I take it that there are more bad
worlds than good worlds because there are more ways to go wrong than
there are to go right. At any rate, once we can identify a best kind of world,
it will follow deductively that God must create a token of that type. Thus,
the probability that God creates some member of that set is 1.
But isn’t it just as Panglossian to think that the world is the best possible
type of world as to think that it is the best possible token? I don’t think so.
1
This general line of theodicy began in an undergraduate paper at the University
of Missouri written for Peter Markie called “The Best Possible Kind of World: A
Quasi-Leibnizan Theodicy.” I thank him for his comments and for conversing with
me like a philosopher regardless of my rank. It was seeded by a careful reading of
Peter van Inwagen’s “The Place of Chance in a World Sustained by God” in van
Inwagen 1995. Kraay 2008a and 2008b were a fascinating follow-up.
2
I mean from all the natural numbers. The Prime Number Theorem gives the
probability for any finite sequence. It is not uncommon to hear mathematicians
say that there are more primes than squares (see, for example, Coppel 2009: 364
and Schumer 2004: 3). I take this to be a perfectly sensible way to speak about a
literal truth that is hard to systematize.
To see why not, we need to be careful not to make a certain kind of easy
mistake. Think of the best kind of vacation. Nothing ever goes wrong: the
drive to the airport is relaxed, the plane is on time, the room is perfect,
and all the events go just as planned. If we think of a good life as like
a good vacation and of a good world as one where all the lives are like
good vacations, then this world is not in the running for best possible
kind of world. But we must resist this kind of thinking. We must remember
to judge the value of a world by the kinds of values realized there. When we
think about the value of a world, that is, we must first have a scale of
value to place it on. We must be able to make some comparative judg-
ments about which kinds of states of affairs are better than others. We
must ask ourselves what kinds of goods are best. I have tried to flag unde-
fended assumptions as I have gone along. Every important intellectual
endeavor must make substantive assumptions along the way, and no one
book can defend them all. I have also tried to be as clear as possible, given
various limitations (space, my cognitive abilities, relative importance of
ends), providing definitions where I can with suitable efficiency (which
in complex labors can prove quite laborious). Above, I spoke often of “the
highest virtues” and mentioned some of them. I said that what unites
them is love. Here is a more complete specification of them. Undefended
assumption: courage, compassion, kindness, generosity, benevolence,
mercy, magnanimity, tolerance, honor, truthfulness, trustworthiness,
responsibility, friendship, cooperation, diligence, discipline, helpfulness,
gratitude, and, especially, empathy and forgiveness are among the highest
manifestations of love (which is the highest of the virtues).
The Irenaean theodicy begins with this value-driven approach. Its
fundamental assumption is that how one responds in life is more impor-
tant than what happens to one (cf. Swinburne 1998), that the best kind
of life is the one exemplifying the highest virtues. And the greatest of
these is love. Of course, one may reject this theory of the good life, but
that is separate from one’s rejection of God. That is a dispute in axiology.
So among the additional assumptions I am willing to make for my argu-
ment is that the right way to judge the moral value of a world is how it
exemplifies the highest virtues.
As Swinburne has so ably defended (2004: ch. 2 and ch. 6) the predic-
tive power of theism comes from the nature of the God posited. What
God can be expected to want is a world in which the highest virtues
are displayed.3 This involves two components: what can be expected is,
3
Undefended assumption: The display of virtue is significantly more valuable
than the mere presence of the disposition. Objection: So then what is so good
first, the display of the highest virtue and, also, the highest virtue. What
must God bring about if he wishes to create a situation in which these
virtues will be fostered? The logically necessary preconditions for the
display of the highest virtues are evils of sufficient intensity (but not
so intense as to destroy the psyche of most of those in the situation)
and of the sufficient frequency to provide multiple opportunities and to
form habits (but not too frequent so as to yield too high a probability
of complete demoralization of the souls of all who go through them:
There must be sufficient chance of success for a sufficient number of
individuals). This gives us a toy matrix of nine possible relevant kinds
of universe (illustrated by Figure 7.1). Only one of them successfully
occupies the “goldilocks” region with respect to the right amount of risk
in the world.4
1. Too frequent and too mild

2. Too frequent and right intensity
3. Too frequent and too intense
4. Right frequency and too mild
5. Right frequency and right intensity
6. Right frequency and too intense
7. Too infrequent and too mild
8. Too infrequent and right intensity
9. Too infrequent and too intense
Figure 7.1
We find ourselves in a type-5 universe: This world is a crucible for and

producer of much opportunity for (and considerable exercise of) virtue.
Yet it is relatively rare that evils of a stultifying degree occur (more of that
later). This is because it has a setup with considerable risk built in. Yet it is
about Heaven, where virtues will not be on display? Reply: Properly speaking,
the highest good is union with God (Stump 2010 and much of her work that has
preceded it has made a powerful Thomistic theodicy from this launching point),
and this requires, for reasons I shall not get into, knowledge gained by having
displayed the highest virtues. The details must await another occasion. And of
course there are other great goods that require bad states that give God additional
reasons to allow considerable suffering. See Swinburne (1998: Part II).
4
Of course there will be other parameters too, but appropriate sophistication
will only make the model stronger. For example, I assume a uniform distribution
here, but I expect that favors naturalism.
quite rare for people to lead lives5 they consider on the whole not worth
living (and, of course, it is more rare that they are right). This is much
more to be expected given theism than naturalism for the reasons given
above, so the facts about the distribution and intensity of evil favor theism
over naturalism. Naturalism leaves it to chance which kind of world there
should be. Theism entails that we will be in a type-5 world, since it is the
best possible kind of world, so Pr(Type-5 | T) = 1. Of course there are more
variables than just frequency and intensity (and there will be various
dimensions of intensity). So a mature version of our toy model would be
more complicated. But I think this would only amplify the result.
Let a “badness-ensemble” describe which types of bads are tokened
and to what degree and their distribution among the population of
beings with moral standing in a universe (and let it be exhaustive, so
that one could deduce the corresponding “goodness-ensemble” from
it). The range of kinds of badness-ensembles consistent with or promo-
tive of opportunity for the exercise of the highest virtues among the
total possibilities (properly understood and measured) is very narrow.
Theism predicts we would land in that range. Naturalism leaves it totally
at random. What I am suggesting here6 is that the Irenaean theodicy
nicely plays into a fine-tuning argument for theism from the nature of
suffering we presently find displayed in our world.
7.2 Application to the Bayesian argument from evil
I have told a story about how bad states of affairs fit into not only a
good world, but the best kind of world. My main goal was to explicate
an up-to-date and original version of the soul-making theodicy to apply
to animals, which for obvious reasons I have called the “saint-making”
theodicy. A central theme was the transformative power of triumph over
tragedy in the formation of virtuous, saintly character. I have assessed the
value of this largely in the only form possible: illustrating it from various
aspects of our lives. If my assessment of the value of those stories being
lived out, and the permissibility of allowing them to be lived, is secured
by their reasonable potential for defeat of any evil encountered in the
lives of those who live them, then the purpose and permissibility of
creating a world-ensemble like ours will have been established. I will now
direct these thoughts explicitly to the Bayesian argument from evil.
5
By “life” here, I mean to pick out the this-worldly segment of an individual’s
total existence.
6
I first suggested this in print in Dougherty and Walls (2013), 380–381.
In the Bayesian formulation of the argument from evil, having made

the (briefly defended) programmatic assumption that the hypotheses of
theism and naturalism are roughly on par with respect to prior proba-
bility, the issue comes down to the likelihood ratio of the two hypotheses
with respect to the facts pertaining to evil. Let S be the theodical story
about God wanting to bring about a world that offers ample opportunity
for the development of the highest virtues which requires significant
risk of significant abuse of morally significant agency (all three kinds of
significance are required). Let T and N, as before, represent theism and
naturalism. And let EH be the relevant data regarding human suffering.
EH Since very near the beginning of human existence, there has been
an abundance of intense human suffering. That is, wherever there
has been human life, very significant levels of human suffering have
been quite common.
This is one way of describing the situation. But the same facts can be
described in various ways depending on what concepts we apply and
from what angle we are viewing it. My wife suggested the following
example. Consider the following description of a state of affairs.
D- Sarah is in excruciating pain, experiencing intense spasms at regular

intervals, and she has been in this situation for several hours.
D- presents the facts as they are, without misrepresentation. However,

the very same facts can be given the following description in light of the
application of different concepts and at a different level of explanation.
D+ Sarah is about to have a baby!
The same facts could no doubt be described in terms of fundamental

physics, as well, though their significance would be lost on most humans
under that description. Moving between D–, D+, and the microphysical
account does not entail a shift of focus from one set of facts to another.
Rather, such movement represents focusing on the same facts with
different lenses.7
7
If one pounded their fist and insisted that there was shift in the facts under consid-
eration in this case, then I would stomp my foot and insist that either it wasn’t a
difference that made a difference or that the shift (or addition) was necessary to accu-
rately represent the problem, just as psychological explanations of human action
would be more enlightening from the human level even if reductionism were true.
I now wish to look at EH through the lens of God’s plan. That is,
having considered above what kind of values God can be expected to
have and, consequently, what kind of actions is he likely to take, we can
re-describe the facts described by EH in the following way.
EH’ Since very near the beginning of human existence, suffering has
come with a frequency and intensity which falls into the relatively
narrow band with enough magnitude to foster saints but not so much
as to widely overwhelm people or make struggle futile.
This data places our world in a certain class of worlds, one theism says
must exist, if God creates. It is also describes a class of worlds which it
is permissible to create, so far as we know, since there is no evidence of
any indefeasible evil. If you prefer to think of EH’ as a second-order fact
about EH or the facts described by EH, I will not press the matter. The
question is whether the following likelihood ratio is top-heavy (greater
than 1) or bottom-heavy (less than 1) (or dead even, but that’s terribly
unlikely).
Pr(EH’|T&S)
Pr(EH’|N)
The more top-heavy this ratio is, the better the theodicy; the more
bottom-heavy it is, the worse it is. Suppose that one thinks the theist
has some decent natural theological arguments for theism: cosmological,
ontological, fine-tuning from natural laws, what have you. Then even
if it is modestly bottom-heavy, the theodicy might be successful in the
scheme of a cumulative case. For even if the ratio is somewhat bottom-
heavy, theism will not have lost much probability from the data of evil. If
the value of the ratio is 1 (or approximately 1), the probability of theism
will not be affected (or not affected much) by data about evil. A ratio of 1
represents the theodicy as exactly neutralizing the argument from evil.
The Numerator So how high is Pr(EH’ | T&S)? I have suggested that it
is very high indeed, 1, in fact. For an EH’ world is a logically necessary
condition to the best kind of world, and God is morally required to
create the best kind of world, or in more basic language, his knowledge
of values motivates him to do so with no countervailing motives. Let
Kbe the class of best possible worlds. Every world k in K is a world with
a significant number of opportunities for sainthood and a not insignifi-
cant number of saints. All saints display some of the highest virtues. All
opportunities for sainthood involve very significant trials. Thus every
world k in K is a world in which there are significant trials. Let K be

the proposition that the actual world is a member of K. Let S be the
saint-seeking story according to which in creating humans God sought
saints and all that entails.
a. T → K (From that God must create the best kind of world ensemble)
b. K → S (From the value of sainthood)
c. S → EH’ (From that sainthood requires very significant trials)
d. T → EH’ (From a, b, and c)
e. If d, then Pr(EH’ | T) = 1 (Theorem)
f. Pr(EH’ | T) = 1 (From e and d)
Someone may object that adding the saint-seeking story S to T, “bare”

theism, lowers the prior probability of the theistic hypothesis so much
that the advantage goes to naturalism.
Notice, though, that a. and b. entail the lemma
g. T → S.
This, in turn, entails that
h. T ≡ (T&S).
Which by a simple theorem entails
i. Pr(T) = Pr(T&S).
So the worry is unfounded.8 Recall here my remarks from Chapter 1,

Sections 1.3 and 1.4. ((T & EH’)& ~S)) is logically impossible,9 so we
must renormalize. There is no loss of initial distribution across T.
Given motivational internalism, you can deduce S a priori from T.
The generic pattern of argument is:
A perfect being desires the best kinds of goods.

X is the best kind of good.
8
Swinburne misses this, I think, for the exact same reasoning applies to the
existence of Heaven. See Swinburne (2004: 264).
9
Letting C be the proposition that God creates, then I think that ((T & C) &
~EH’) is logically incompatible.
So a perfect being desires X-type things.10

One can argue with the moral theory or the moral judgment. But with
those two items in place, T entails S, since saint-fostering worlds are the
best kind.
The Denominator What about Pr(EH’ | N)? As stated above, it is very
unlikely that our world would fall in the suitable saint-fostering range
by chance. Suppose it is as high as 1%, which it surely is not. 1/.01 is
100. According to Jeffreys (1961: 432), a Bayes factor greater than 100 is
“decisive”. So, on this line of thought, not only does the data of evil not
disconfirm theism, it decisively confirms it.11 Suppose you the band of
creatable worlds is wider by an order of magnitude than I have said. That
still makes theism beat chance by an order of magnitude. In fact, if, as I
think pure folly, one were to say that on chance it was exactly as likely
as not that we would occupy the acceptable range, that would still put
theisms predictive power at 2:1 over naturalism.
Now, some readers are bound to feel at this point that they’ve been
tricked (or at least that I have attempted to trick them). I can sympathize
with that.12 It came as a bit of a shock to me as well. But a sort of gestalt
occurred after many repeated readings of Draper, Hick, Swinburne, and
Hasker (and Adams, and Stump, and van Inwagen). The more deeply I
became convinced of the value of virtue, especially saintly virtue, and
the more deeply I delved into the lives of the saints, the more I recog-
nized so many of the virtues as manifestations of agape love, the more
I realized a world which did not offer ample opportunity for sainthood
would have no appeal for a being motivated by the greatest goods. And
very great manifestations of virtue, such as the best kind of world would
need some of, what we might call “heroic” manifestations of virtue,
require nearly unbearable conditions; nearly unbearable. I am not saying
that God brought about such circumstances. I am saying that he had
sufficient reason to make a world in which they were significantly prob-
able and for permissibly refraining from intervening in many cases if
they did occur (we have no idea how often he has intervened). It is as if
10
There might be a many-way tie for best kind of state with incommensu-
rable goods and/or incompatible goods. We would then expect a good balance
between compatible kinds of goods, and none of the greatest goods would be at
all surprising.
11
I actually think Jeffreys’ scale is extremely conservative, as I suspect he was
looking for a kind of robustness in measure that ignored prior probabilities.
12
Indeed, I agree with Swinburne (1998: 245) that there is something wrong
with one if one isn’t sympathetic with some initial incredulity concerning
theodicy.
God says “I am not going to intervene in this suffering, terrible though

it is, because I know you can make it through it, and I want you to have
experiential knowledge of this fact as well.” At first these reflections
manifested themselves in a conviction that a world of trials ranging
from the mundane to the monstrous was not improbable on theism. But
it gradually dawned on me that there is simply no other kind of world that
will have the kind of goodness such that creating it is consistent with
God’s character.13 But then, of course, it follows that such a world is quite
probable. Indeed, if God creates at all, it is the only kind of world he will
create. So that God creates a world entails that God creates this kind of
world; a world with the significant probability of hagiographic stories
and the horrible suffering that is so often their cost.
7.3 Too much suffering? What’s the evidence?
Now of course it is a common objection that there is too much suffering

in the world; too many instances of too intense suffering. But it is hard
to substantiate this claim given our understanding that God desires a world
with many opportunities for heroic virtue. “Too much” in the context of
the above reasoning would have to mean “too much for a finite being
to handle, so much that it could never be defeated (in addition to being
counterbalanced) over the course of their total existence.” But we have
almost no evidence at all that any being has ever suffered such an evil.
All we know is that some people were losing at the time of their depar-
ture from this world. But that tells us very little about their eventual
state of mind with respect to that suffering. There may be traumas from
which no finite mind could ever recover. But there is almost no evidence
at all that any such evil has ever befallen any creature. We know there
are people who are bitter or broken or both until the end of their earthly
lives. But this tells us little to nothing about the permanence of their
13
Strictly speaking, all goods have a prima facie allure for God. But only the
best kind of goods can have an ultima facie allure. And, again, we must allow for
best compossible aggregates and for the possibility of incommensurability. The
main idea, though, is that all the best worlds where God creates have saints, and
all saints suffer; the greater the saint, the greater the suffering, within the limits
of what a creature can bear and what it is permissible for God to allow. Just how
much a creature can ultimately bear is in itself an interesting question. I am
aware of no sustained attempt to show of any particular evil that it is intrinsi-
cally impermissible for God to allow. But see Dougherty (2008), (2011c), (2014),
and Dougherty (2014, Section 6.2), and Dougherty and Walls (2013, 372) for a
problem concerning intrinsic impermissibility.
perspective. It is like inferring that your birthdate does not appear in the
decimal expansion of pi because it doesn’t occur in the first thousand
digits (you have to go nearly half a million digits out to get to mine).
Contrast our ignorance with respect to evil’s defeating humans with our
knowledge of evil being defeated in the lives of humans. What we do
know is that for every type of horrendous evil that has befallen a human
on earth, some of the individuals who have undergone them have gone
on to flourish as human beings in ways to which their horror was inte-
gral. Frankl (1985) is a prominent example, but there are many more.14
When an evil is defeated, it is integrated into a good whole; it does not
itself become a good. The Holocaust is an evil that is so vile I can scarcely
write or say the word. It is almost like the “N-word” in that respect. Yet
I am constantly amazed by the human spirit as I discover more and
more books by Holocaust survivors like Mike Jacobs, who lives about an
hour north of me, whose book Holocaust Survivor: Mike Jacobs’ Triumph
over Tragedy tells how “he kept hope alive in his heart by helping others
and by undermining Nazi efforts, no matter the risk to himself. A story
without hatred or bitterness, Holocaust Survivor teaches us that when we
recognize that freedom comes from within, we are never completely
powerless” (from http://www.dallasholocaustsurvivor.com/book.html).
There are more books like this than I could read in preparation for this
one.15 One that stands out is The Triumph of Wounded Souls (Lerner
2004) in which a child of Holocaust survivors tells the story of seven
children who survived the Holocaust. Pulitzer Prize winner Robert Coles
(1999) says this in his endorsement of the book: “As a child psychia-
trist, I was completely immersed in every page of this well written and
compelling manuscript, which poses enormous questions about survival
and meaning – how the young endure (and even prevail) under the
most awful of circumstances.” I picked these two books because they
are two of my favorite Holocaust survival accounts, but I now realize
they also both have “triumph” in the title. It is a testament to the fact
that not even the worst evils humans encounter are indefeasible. Many
will not triumph in the very midst of such horrors or even shortly after
14
Not long ago a book called The Shack (Young 2011) was very popular in
Christian circles. While theologically very light, it dramatically portrayed the
ability of horrendous evil to be defeated in someone’s life.
15
My reading of Holocaust literature is primarily to “hold my feet to the fire”
and not cut corners or lose touch with the severity of the problem or its human
dimension. Yet it is virtually impossible to delve into Holocaust literature and not
encounter a number of stories like this whether one is looking for them or not.
them. But these stories show that the human spirit – which is the imago
dei16 from receiving the breath of God – is equal to anything thus far
observed on earth. It provides reasonable hope that, in the end, there
are two kinds of people: those who have already triumphed over their
suffering, and those that will.17
Do such stories mean that it was a good thing the Holocaust happened?
No, it was a bad thing that it happened, a terrible thing. Does the fact
that some people triumphed over this great evil show that there is some
character flaw in those who did not triumph? No, the most we can say
about those whose earthly stories did not end in triumph is that they
had not yet triumphed. Whether they eventually will is something about
which we have very little evidence.
From these reflections I conclude that there is not enough evidence
for the proposition that the degree of suffering in this world exceeds
the range predicted by theism. That range is set by a desire for a world-
ensemble that will likely result in a considerable number of saints. That
range would be exceeded in a particular life only if evils occurred to that
person which were such that either it was impossible for that person’s
existence to be on balance good for them or it was impossible for that
evil to be defeated within the context of a life that individual is glad for
and self-endorses. There is no good reason to believe evils that satisfy
either of these conditions have occurred. Thus, we must conclude that
the evils of this world fall within a permissible and predictable range,
given theism.
It is my hope that the reader who has not had the kind of gestalt I
had can at least see that such a world is, upon reflection, unsurprising
given theism, or at least not very surprising. But on naturalism, it is
quite surprising. For there are ever so many ways it could have been
unsuitable for virtue and saintliness, the greatest of created goods. It
seems to me that the range of reasonable values for Pr(EH’ | N) will
not reach as high as .1. If the considerations above convince you that
such a world as ours is at least as likely as not, then that gives theism a
Bayes factor of 5, which according to the Jeffreys scale (which I think
is too conservative) results in “substantial” confirmation. But suppose
you have some modicum of sympathy for the picture I painted but are
16
For more on animals and the imago dei, see Clark (2013: 534ff).
17
I bracket here the issue of whether anyone could consciously wrongfully
choose to “hold on to” their bitterness over suffering forever. There is no better
treatment of this than C.S. Lewis’s The Great Divorce (2001a). For treatments of
the problem of Hell, see Kvanvig (1993) and Walls (1992).
still pretty convinced it’s false: Suppose you would take 2:1 odds against
God’s creating a world-ensemble likely to end up approximately like our
world. Still, that results in a Bayes factor of over three, and this is spotting
the naturalist a value that is much larger than what I think reasonable
for a maximum value. The numbers, of course, are somewhat artificial,
but they do represent the dynamics of confirmation. And the dynamics
here seem to indicate that it is going to be a hard sell for the naturalist
to claim an advantage on the Bayes factor, and thus the saint-making
theodicy achieves the goal for which it is deployed. It shows that the
Bayesian argument from evil provides no advantage to naturalism.
I end with a summary of the argument.
Likelihood Argument for Theism from Evil
1. God is all-knowing and all-powerful.

2. If God is all-knowing, then he will always accurately perceive the
degree of goodness of states of affairs.
3. Every good that God recognizes motivates God to bring it about in
proportion to its goodness.18
4. God is most motivated to bring about some of the best kinds of
goods. From 1, 2, and 3
5. If God is all-powerful, nothing could prevent God from acting on
his motives and (a la 4) bringing about some of the best goods.
6. Nothing could prevent God from acting on his motives and bringing
about some of the best goods. From 1 and 5
7. The best goods are the authentic display of agape love-manifesting
virtues.
8. God will bring about a world-ensemble fostering the authentic
display of agape love-manifesting virtues.19 From 4, 6, and 7
9. The authentic display of the highest virtues logically entails the
occurrence of very significantly bad states of affairs (but not so bad
as to cause widespread, permanent despair).
10. God will bring about a world-ensemble described by EH’. From 8, 9,
and the definition of EH’.
18
There is no best world. Perhaps God creates an infinite number of persons or
even an infinite number of worlds (see O’Connor 1998). Even a finite number of
beings existing for an infinite amount of time creates an infinite amount of value,
though there are puzzles here (see Lauwers and Vallentyne 2004). These issues are
too far afield to pursue here.
19
He will, that is, if he brings about any world at all. The state of affairs of God’s
existing is sufficiently good that satisficing allows God to justly choose not to
create anything at all.
11. If 10, then T strongly predicts EH’.

12. T strongly predicts EH’. From 10 and 11.
13. N does not strongly predict EH’.
14. EH’ favors T over N. From 12 and 13.20
Some will deny Premise 3, a form of motivational internalism. I suspect

that some of the objections to motivational internalism as a theory
for humans are no good as objections to the thesis that internalism is
true for God. But I have taken that assumption on board as an unde-
fended assumption. It is an example of how theses in ethical theory play
into philosophy of religion. To deny Premise 7 is to reject not only the
Christian moral axiology but also one that shares widespread secular
support. Some will deny Premise 11 on the grounds that some horren-
dous evils fall outside a permissible range. That was addressed above. I
conclude that the saint-making theodicy is a success.
7.4 A narrative approach
One way to express thoughts effectively is to supplement ordinary

language with a more precise artificial language that better reveals the
underlying, often hidden, logical structure that is so frequently obscured
by ordinary language. I have used various forms of logic to this end in
this book and will do so more below. At the same time, however, these
ideal languages have certain limitations. Above – in my personal stories
of outdoor adventure and in the lives of the saints – I have had recourse
to a more familiar form of expression which permeates human thinking
both in its ordinary form and in some of the best philosophical modes:
narrative. When I look at the above list of virtues, I immediately and
spontaneously think of The Lord of the Rings. All of those virtues are on
display there being formed and exhibited in the midst of great trials. It is
a story that weaves together tragedy and triumph in a supremely mean-
ingful way. I think also of the Narnia books – some people forget the
hardships suffered there – The Odyssey, The Brothers Karamazov (Alyosha
is one of my favorite fictional saints), Anna Karenina, One Day in the Life
of Ivan Denisovich, and many others (some of which are not Russian).
The great narratives that move us and teach us about meaningful lives
are stories of suffering, struggle, and virtue. And we seem to want, in
20
And recall we have made the provisional assumption that the prior prob-
ability of theism and naturalism are about equal, though I think theism is very
much simpler than naturalism.
our deeper selves, to live such stories, even if we are often afraid to
seek them. Great literature seems to depend heavily on these themes.
Shakespeare himself has said,
Sweet are the uses of adversity,

Which, like the toad, ugly and venomous,
Wears yet a precious jewel in his head;21
There are, of course, great literary comedies (though is any comedy

really to be classed with the great tragedies? I love A Midsummer Night’s
Dream, but it is no Macbeth), as well as great stories without tragedy,
and tragedies without much virtue (though they tend to be didactic,
cautionary tales which inculcate virtue, like Macbeth and Othello). What
I am suggesting is an affinity between the larger, longer, deeper narra-
tives and adversity as a crucible of character for the characters to develop
virtue or vice. The story of our world is such a narrative. That is how I
read it, anyway. On the theistic telling of the tale, the story is very far
from finished. Through analogy with such narratives as well as with
stories from the lives of the saints and from my own life, I have tried
and will try to paint a suggestive picture in which our world has the
makings for a great story, a story of millions of saints of all varieties
whose lives exhibit the highest virtues and give great value to the worlds
they inhabit.22 And note that to be the right kind of story, not every
character has to be a saint or at least to the same degree. However, there
must be saints featured in the story.
Another way to put the saint-making idea in terms of narrative is that
each of our biographies is potentially on the way to being a hagiographies.
The canonized saints are special examples of saintliness (the apotheosis
of Christian virtue) found more commonly scattered through the lives
of more ordinary people. (I wish to make it clear that I think saintly
moral heroism is displayed on the part of poor working mothers who
sacrifice so much for the good of their children.) Just as Marilyn McCord
Adams has suggested using resources unique to Christianity – in her case
Christianity’s central doctrine of the Incarnation – so I have tried to do
in a somewhat different way by incorporating the lives of the saints.
Whereas adding the doctrine of the Incarnation seems to add much by
21
As You Like It, Act 2, scene 1, 12–17. The Duke has been betrayed and lost all
he has when he says this.
22
In fact, Aquinas says that “among the best of all the parts of the world are
God’s saints” (quoted in Stump 2012: 402).
way of content to “bare” theism, which I am primarily interested in

defending, and thus appears to lower its prior probability23, I am not
adding any content to bare theism. Rather, I am simply appealing to
stories that happen24 to be from within the Christian tradition. I could
easily have given examples of saintly virtue from outside the Christian
tradition. It is for the most part my greater familiarity with my own
tradition that explains why I use Christian examples. But I also happen
to believe that being in more direct union with the Holy Spirit is an
advantage in the path to saintliness.
23
I actually think this appearance is misleading. I think the probability that
Jesus is the Incarnate Word conditioning on the conjunction of bare theism and
historical facts about Jesus is very high. So letting “I” name the proposition that
Jesus was the Incarnate Word and “J” name the proposition containing all the
historical facts about Jesus, I think that Pr(I | T&J) is so close to 1 that Pr(T&I)
is only very slightly less that Pr(T). For arguments to this effect, see Swinburne
(2003) and Wright (1992).
24
The truth is I do not think it just “happens” to be within the Christian tradi-
tion. I don’t think it is a coincidence at all. The point is that because I am not
appealing to any legendary material about saints but rather matters of historical
record, I am appealing to background evidence and so not adding anything to the
content of the theism which I am defending.
8
Animal Saints
In the previous two chapters, I presented what I take to be a stronger

theodicy than has previously appeared. The key was combining – in
a natural and organic way, not just through amalgam – (i) a Hick/
Swinburne(/Hasker) Irenaean theodicy with (ii) a version of Adams’s
adaptation of Chisholm’s notion of defeat, (iii) refracted through the
lens of sainthood. Through attention to the “Fathers” of the church as
well as Saints through the ages, I upped the ante from soul-making to
saint-making by noting that (i) the evils of this world make available
not just generally good moral character, valuable as that is, but true
saintliness, the value of which is almost inestimable, and (ii) most of the
highest virtues may be seen as revelations of agape love. Then I noted
that one paradigmatic way evil may be defeated is through martyrdom,
a love-generated (love for both God and for those made in the image
of God) willingness to suffer for the sake of others and to glorify God,
to give oneself wholly to God and abandon oneself to his plan. It is my
position that rightly oriented souls will, in retrospect, look upon their
role in the drama of salvation, embrace their role, and see in their role
God’s goodness to them. This further assumes that God only acts justly
toward them, for they must be recognizing and appreciating the fact
that they were being both justly and lovingly included in important
ways by God in ways that were of benefit not only to others, but to
them. This dispositional fact (the disposition to endorse this role on
the part of a soul) in itself grounds, I claimed, a kind of tacit consent
to such inclusion. On many theories of divine foreknowledge, God will
have access to facts about who will have this disposition and who won’t.
On Molinism, this picture is very natural indeed, but I take it that other
views can accommodate it as well. This disposition itself is a good state
of affairs, but the authentic exercise and display of this disposition is,
134
Animal Saints 135
I claimed, the greatest created good.1 So we have in one picture saintly

virtue, consent to inclusion, and defeat of evil. That’s the good news.
The other side of the coin is not so shiny. For to the extent that I have
presented a stronger species of soul-making theodicy, it appears all the
harder for it to be applied to animals. Mixing the Chisholm–Adams
defeat material with Hick’s Irenaean theodicy and placing saintliness at
the center greatly strengthens it. However, it also seems at first glance
to make it seemingly impossible to apply to the animal kingdom since
most animals seem to lack the requisite cognitive capacities it calls for.
Since all extant theodicies for animal suffering fail, it would be both
surprising and quite a blessing for theists if the theodicy I have devel-
oped applied to animals.
In this chapter, I will attempt to bring together in an original way the
work of John Hick, C.S. Lewis, and Marilyn McCord Adams, extending it
into new territory as I go. As I will show, though I am weaving a tapestry
out of cloth they dyed, they seem not to anticipate the way in which
their work could be extended, mirabile visu, to animals. Yet I do believe
their work points in the direction I will take it, a direction that was first
pointed out to me not in their works, but in a sort of mystical experi-
ence. Yet the source of the idea is not important. I believe that having
once seen the possibility, as well as that there is no real barrier to it, save
unfamiliarity, a reader can come to see the previously odd as the perfectly
obvious (and in an entirely non-dysfunctional way!). It is a sort of revela-
tion that can dawn on one when contemplating the possibilities as well as
a recovery to the West of some lost ideas from within the eastern portion
of the Christian tradition. So, surprising as it may be, I will be giving a
soul-making (even saint-making!) theodicy for non-human animals. I will
with the objection that animals do not have souls in the next chapter.
In Section 8.1, I consider doubt about animal soul-making from those
who otherwise favor soul-making for humans. After demonstrating this
doubt from their works, I turn in Section 8.1.1 to a cause for leaving
animals out of the theodical picture: Essentially, a failure to appreciate
not only their current meaning-making powers, but the ways in which
their capacities for meaning might be enhanced. In Section 8.1.2, I turn
to a consequence of the marginalization of animals: essentially, making
1
Some, including several students and Marilyn McCord Adams, have asked
whether participating in the beatific vision isn’t a greater created good. All I can
say to this now is that the goods of being are prior to and partly constitutive of
the goods of relation. One must first be the right kind of being before being able to
enjoy the beatific vision.
the worth and spiritual capacities of animals wholly dependent on

humans. I argue in these two sections that both the cause and the conse-
quence of animal spiritual marginalization are avoidable.
In Section 8.2, I very briefly introduce the notion of deification. I
then describe a picture of both human and animal enhanced cognition
required for enhanced spiritual capacities. In Section 8.3, I give an argu-
ment for the conditional If God exists and animals suffer, then animals are
resurrected and deified. In Section 8.4, I consider objections to this thesis
pertaining to the concerns about numerical identity and psychological
fragmentation.
8.1 Doubt about animal soul-making from

within the ranks
As we saw previously, one reason the problem of animal pain seems so

powerful is that it does not appear to be subject to resolution by tradi-
tional ways of addressing the problem of pain for humans. This is often
thought true for approaches emphasizing free will and especially the
soul-making theodicy. We saw statements of this earlier. More strikingly,
this sentiment is shared by the soul-making theodicy’s most prominent
proponent, John Hick.
To some, the pain suffered in the animal kingdom beneath the human
level has constituted the most baffling aspect of the problem of evil.
For the considerations that may lighten the problem as it affects
mankind – the positive value of moral freedom despite its risks; and
the necessity that a world which is to be the scene of soul-making
should contain real challenges, hardships, defeats, and mysteries – do
not apply in the case of the lower animals. (Hick 2007/1966: 309)2
As we will see, the reason Hick thinks his Irenaean strategy does not apply
to non-human animals is that they do not possess the requisite cognitive
powers or the power of free will. Adams expresses a similar sentiment.
[M]y analysis of both problem and solutions will pertain to animals

only to the extent that their cognitive and affective capacities consti-
tute something like meaning-making powers. (1999: 28)
2
A contemporary counterpart: Goetz (2009), saves the problem of animal pain
until the last section of his very, very long entry on the problem of evil. And after
having developed a genuinely interesting theodicy, all of a sudden goes skeptical
when it comes to animals. See Goetz (2009: 492ff).
Animal Saints 137
It might be that “meaning-making” powers go beyond free will plus

certain cognitive powers and affective powers. Those are plausible neces-
sary conditions, but I will adopt her phrase “meaning-making powers”
to refer to a contextually relevant set of necessary conditions for effec-
tively falling under my saint-making theodicy. So the dictum is “no
soul-making possibilities without meaning-making powers.” We will
come back to this after considering both a cause of and a consequence of
leaving animals out of the standard soul-making theodicy.
8.1.1 A cause of the lacuna in previous soul-making theodicies

The cause, or at least a cause, for their omission of animals is that Hick
and Adams seem to subscribe to a version of C.S. Lewis’s neo-Carte-
sianism. His position was not examined in the chapters on neo-Carte-
sianism (Chapters 4 and 5) because I take it to have been thoroughly
treated by Bassham (2005) and Murray (2008) and otherwise refuted or
side-stepped by the material of Chapter 5, which was focused on more
recent and scientifically sophisticated neo-Cartesianisms. It is, however,
worth seeing how deeply his argument or similar thoughts seem to have
affected both Hick and Adams.
Lewis does not deny that animals are sentient in the traditional sense
of being able to have sensations. However, he does deny higher forms of
sentience that depend on the ability to perceive the flow of time.
Suppose that three sensations follow one another – first A, then B,
then C. When this happens to you, you have the experience of passing
through the process ABC. But note what this implies. It implies that
there is something in us which stands sufficiently outside A to notice
A passing away, and sufficiently outside B to notice B now beginning
and coming to fill the place which A vacated [and so on] ... so it can
say ‘I have had experience ABC.’ Now this is something which I call
Consciousness or Soul. (2002: 630)
It is this kind of consciousness Lewis takes animals to lack. Accepting
this picture, Hick draws a natural conclusion.
The animal’s goods and evils are exclusively those of the present
moment, and in general it lives from instant to instant either in
healthy and presumably pleasurable activity, or in a pleasant state of
torpor. The picture, then, of animal life as a dark ocean of agonizing
fear and pain is quite gratuitous, and arises from the mistake of
projecting our distinctly human quality of experience into creatures
of a much lower and simpler order. (2007/1966: 314)
On Hick’s reading, the Lewis view replaces gratuitous suffering with

gratuitous accounts of suffering resulting from sloppy thinking. How this
squares with Hick’s agreement that “[T]here is evidence that some of the
higher animals not only experience physical pain but also a degree of
non-physical suffering, in forms analogous to loneliness, fear, jealousy,
and even bereavement” (2007/1966: 311) I admit I do not know.
Adams’s view is more nuanced but still has a whiff of the same line of
thought.
Folk psycho-biology would place worms and clams well below horrors’
reach but might equivocate about chimpanzees or dolphins, perhaps
even dogs or cats. If all mammals and perhaps most kinds of birds,
reptiles, and fish suffer pain, many naturally lack self-consciousness
and the sort of transtemporal psychic unity required to participate in
horrors. (1999: 28, emphasis added)
Even if folk psycho-biology equivocates about chimpanzees, I trust that

the judgments involved in the Problem of Primate Pain I presented
earlier do not. But the convergence of opinion here is striking. The
lack of “transtemporal psychic unity” (hereafter TTPU) is treated as an
axiom, though there is little evidence to support it – and none is offered
by any of the three authors considered. Granted, Adams is only denying
that animals can possess the ability to suffer horrendous evils, not that
they cannot suffer pain at all. However, I doubt there is a sharp cut-off
for horrendousness, and some animals seem to share some of human
life’s greatest burdens, such as mourning the death of a mate. The neo-
Cartesian denial of TTPU to non-human animals is a root cause of the
three authors’ belief that traditional approaches to the problem of evil
cannot apply to animals.
Hick sums up well the problem of the lack of meaning-making powers
of animals.
The more fruitful question for theodicy is not why animals are liable
to pain as well as pleasure – for this follows from their nature as living
creatures – but rather why these lower forms of life should exist at all.
Christian theology enables us to understand, within its own presup-
positions, why the human creature exists: he is a rational and moral
being who may freely respond to his Maker’s love and so become a
“child of God” and “heir of eternal life.” But this explanation cannot
cover the lower animals, lacking as they do a rational and moral
nature. Their existence remains a problem. (2007/1966: 314)
Animal Saints 139
Again, it is the lack of meaning-making powers – rational and moral

capabilities, including free will – which seem to cut the Irenaean theodi-
cist off from explaining the facts of animal pain by reference to their
becoming fully actualized (or at least asymptotically full) representa-
tions of the imago dei.
8.1.2 Some consequences of the lacuna

We began this section (8.1) by noting the way in which our key authors
admit that their theorizing does not, by their lights, extend to non-
human animals. We have seen that the cause of this is their denial of
the claim that animals have TTPU (the ability to exemplify meaning-
making properties). But our authors (at least Lewis and Hick) seem to
think, wrongly I have argued, that this very denial alleviates (at least
to a significant degree) the problem of animal pain. Alas, if only it were
that easy. Having looked squarely at the cause of their denial, we shall
now look at an interesting consequence of their leaving animals out of
the theodicy picture.
After considering the matter, Lewis’s (2002) conclusion about animals
is that “The beasts are to be understood only in their relation to man ... ”
(634). Hick agrees: “The problem of animal pain is thus subordinate
to that of human sin and suffering ... ” (2007/1966: 316). While it is
true that according to the Adamic narrative in the Hebrew Scriptures
mankind is given dominion over the animal kingdom, extra premises
would be needed to move from this to the Lewis–Hick thesis. Without
straining at an exact explication of Lewis’s or Hick’s words, I propose a
view contrary to the prima facie import of their utterances. I propose the
following Animal Independence Thesis (AIT).
AIT Non-human animals have a direct relationship to God apart from

their relationship to humans.
I take it that AIT is consistent with taking the Hebrew Scriptures seri-
ously in their statements about the relationship between man and beast,
but I will not press that here. I note only that AIT entails an independent
relationship. I do not mean a “personal” relationship requiring animals
to think about God (though animals do desire the good, which at least
on a Thomistic account is tacitly desiring God de re). I just mean that
in virtue of being beings with life and sentience, they thereby resemble
God to some degree and are therefore to some degree made in his image
and thus bear intrinsic value.
Objection: But so are plants insofar as they live and rocks, insofar as
they have existence. Reply: I’m totally OK with that. Furthermore, as we
shall see in the next chapter, non-human animals share the very same
“breath of God” as humans. Thus, they bear the imago dei in a special
way. As such, they have moral standing (which is an independent issue
from whether they have rights). As beings with moral standing, God
will have concern for them as such and not merely in relation to other
beings.
And if God has duties to his created beings in virtue of bringing them
into being – as Swinburne (1993: 185) affirms and McCann (2012:
135) denies (I’m with Swinburne on this) – then God will have duties
to them that do not depend on other created beings. But even if we
deny God has duties or that he has duties to non-human animals, God
will surely have concern for all beings with moral standing, which count
morally. Thus, to grant animals any moral standing is to place them
in the circle of God’s concern. Though (in many cases) lacking TTPU
and some other meaning-making properties might entail that they are
not capable of being the subjects of horrendous evils, they are neverthe-
less capable of suffering repeated synchronic pains, which are intrinsi-
cally bad states of affairs. And though this might not be as bad as true
diachronic suffering, it is still a very bad situation and so calls for expla-
nation, given that they are creatures about whom a good God would
be concerned. They have conditions of flourishing, and many do not
flourish. This is bad.
There is a very serious further consequence of denying TTPU and some
other meaning-making properties to non-human animals. It is a more
specific version of the Lewis–Hick line, a soteriological version. Lewis
(2002) says “Certain animals may have an immortality, not in them-
selves, but in the immortality of their masters” (635). He is well aware
this leaves out animals in a state of nature – “it is upon the tame animal
that we must base all our doctrine of the beasts” (634) – for it is part of
his view that due to the subordination doctrine in the Book of Genesis,
the tame animal is in fact in a more “natural” state.
There may be something to this line of reasoning. It may be that,
on Earth at least, animals are meant to be subordinate to humans. If
so, then wild animals are not in their “natural” state. This would be
an instance of “fallenness” in nature. The problem is that this way of
looking at the problem conspicuously leaves aside Rowe’s fawn (Rowe
1979), and that is unacceptable. The fact of such suffering demands an
explanation of the theist.
Animal Saints 141
Lewis gives at least one specific argument for why TTPU is necessary
for Heavenly recompense.
The real difficulty about supposing most animals to be immortal

is that immortality has almost no meaning for a creature which is
not “conscious” in the sense explained above. If the life of a newt is
merely a succession of sensations, what should we mean by saying
that God may recall to life the newt that died today? It would not
recognize itself as the same Newt ... (2002: 633)
This is an interesting point. Lewis goes on to suggest that in light of

this failure of self-recognition the bliss of the resurrected newt would
be of no more relevance to the suffering newt than the bliss of any
later existing newt. “There is, therefore,” he concludes “no question of
immortality for creatures that are merely sentient” (634).
Again, Hick follows Lewis.
[I]t does not seem that an eschatological new heaven and new earth
with a new animal creation, could relieve the problem of earthly
animal pain. For, unless we postulate a heaven for the millions of
millions of individual animals that have perished since sentient life
first appeared, no future state of the universe will be relevant to the
pains that these creatures have undergone. (2007/1966: 316)
Of course, it is no problem for God to resurrect millions of millions of

individuals. It’s not as if, for God, that is any harder than resurrecting a
single individual. Nevertheless, says Hick,
[I]t is extremely doubtful whether even a zoological paradise, filled

with pleasure and devoid of pain, could have any compensatory value
in relation to the momentary pangs of creatures who cannot carry
their past experience with them in conscious memory. (2007/1966:
316)
So he winds up in exactly the same place as Lewis for exactly the same
reason. Not only is the value and proper understanding of animals tied
to that of humans, so is their possible salvation. However, there does not
seem to be much by way of salvation on offer to animals, and none at all
for those animals in a state of (“fallen”) nature.
But pay close attention to Lewis’s conclusion and spot the non sequitur.
After noting the uselessness of immortality to a creature lacking TTPU,
he concludes, “There is, therefore ... no question of immortality for crea-

tures that are merely sentient.” But from the fact that a creature is merely
sentient it does not follow that it will remain merely sentient. The same
goes for Hick: his conclusion is about “creatures who cannot carry their
past experience with them in conscious memory” (316). But from the
fact that a creature cannot now carry its past into its future it does not
follow that it cannot gain that power in the future.
In fact, the whole edifice of exclusion of animals from the Irenaean
understanding is founded upon the current lack of meaning-making
properties. But of course what import this has in the long run depends
on what happens in the long run. Let’s assume that baby humans also
lack the meaning-making powers discussed. Now consider the objection
that the soul-making theodicy does not for that reason apply to babies.
That would be absurd, for the baby’s lack of meaning-making properties
is only a temporary phase of their existence. They will go on to acquire
these properties, and these properties can go on to “seep back into” their
infant life and redeem suffering endured during the infant stage.
For example, a male infant can undergo the pain of circumcision as
part of a religious ritual. At that time he has no ability to appropriate this
into the narrative of a good life. Yet, later, after acquiring normal adult
cognitive capabilities, he might endorse this act as part of his inclusion
in his religious tradition. He could not only forgive his parents in light
of a benevolent understanding of their purposes for including him in
the ritual, but be glad they did so.
Objection: But suppose we know that the infant will die shortly there-
after and there is no possibility of such future redemption of the suffering
via an appropriation into a life of virtue? Reply: On the Christian concep-
tion of the world, that infant (some Augustinian objections notwith-
standing) will get a chance to develop further in the afterlife. And it does
not matter in which mode of life the individual forms their soul. And,
again, since theism entails that there is no possible world containing
fallen creatures that lacks an afterlife, heaven is an entailment (modulo
suffering) of bare theism. And the same, I suggest, goes for animals. They
will have ample opportunity in the afterlife to develop in the requisite
ways, to form a perspective on their earthly careers.
What I am suggesting is that we think about non-human animals in
essentially the same way we think about infants. The key feature is that
we see animals in terms of a developmental spectrum. They are now in an
extended infant-like state (or perhaps child-like in the higher phyla). But
this is a stage in their development just as it is for the human infant.
And on the Christian view, the level of mature human development
at average age of death is immeasurably below what one can expect in a
Animal Saints 143
favorable afterlife. And, of course, the whole earthly stage of life is infini-
tesimal in size in comparison with the measure of their whole existence.
Different organisms, the fruit fly comes to mind, have different rates of
development, and it is in no way an objection, in a Christian context,
that animals would spend their whole earthly existence in an intellec-
tually immature state, leaving further intellectual (and other) develop-
ment to the next stage of existence.
8.2 Deification: human and non-human
The early church used the word “theosis” for a view whereby humans
were, especially in the afterlife, transformed into a state ever more like
that of God. Sometimes called “deification” in the West, this view has
support by both Eastern and Western theologians, though Irenaeus is
an early and ardent proponent. More will be said of this later, but the
Irenaean dictum is “[T]he Word of God, our Lord Jesus Christ, who did,
through His transcendent love, become what we are, that He might
bring us to be even what He is Himself,” (2010: 554) or as more pithily
put by Athanasius, The Word of God “assumed humanity, that we might
become God” (2007: 93).
“God” here expresses the property of divinity but not in the same sense
as God exemplifies divinity (whether God exemplifies the same property
in a different way or a distinct but analogous property is not a debate
we need enter into here). God’s attributes are traditionally divided into
the “communicable” and “incommunicable” attributes. A communicable
attribute is one that can be shared with a created being such as being alive,
conscious, or rational. An incommunicable attribute is one that no created
being can exemplify, such as, preeminently, uncreatedness. The same
naturally goes for aseity, necessity, and the omnis. However, though no
created being can be (or become) infinite, there is no limit to the growth
that can occur along the dimensions of any communicable attribute. So,
though no human can ever become omniscient, there is no finite upper
limit to how much a human can know. And though no human can ever be
omnipotent, there is no upper limit on how much finite power a human
can have. According to one version of the theosis doctrine, humans, at
least some humans, will never cease growing in knowledge, power, and
love, and thus, in a sense, never cease to approximate God’s nature.
If this is the case, then the gap between what humans currently are
and what they will someday be is truly astounding, beyond the limits of
our imagination. To quote Lewis (2001b), “It is a serious thing to live in a
society of possible gods and goddesses, to remember that the dullest and
most uninteresting person you talk to may one day be a creature which,
if you saw it now, you would be strongly tempted to worship” (45). In

the next chapter, I will give a brief but vigorous defense of the proposi-
tion that animals bear the imago dei. And so it hardly seems surprising
that non-human animals would undergo at least enough deification
to reach the level of clear personhood and be able to appropriate their
experiences from their non-personal stages into a life of virtue just like
humans.
Objection: But isn’t this just helping oneself to what one needs to
meet the objection? There can’t be a shred of independent evidence
for animal immortality or deification. Reply: More will be said of this
shortly, but the notion of “independent” evidence is fraught with prob-
lems. If justice entails something, then theism entails that thing, for
theism entails that justice will be done. What’s more, if being a good
person entails something, then theism entails it, for bare theism entails
that God is good to all creatures. Since one can do one’s duty without
doing well by someone, God’s goodness entails much more than justice
demands. I emphasize that “goodness” here is intended to mean omnibe-
nevolent, i.e. perfectly loving. The law of love is much more severe than
the law of justice in its requirements on persons.
It is interesting that even Lewis seems to have an inkling of the
sort of animal theosis proposal I’m making. Consider this statement.
“Supposing, as I do, that the personality of the tame animals is largely
the gift of man ... their sentience is reborn to soulhood in us as our mere
soulhood is reborn to spirituality in Christ” (Lewis 2002: 636). If we
reword this in accordance with my AIT and make the analogical form
explicit, it would go like this.
(CA) Just as humans are reborn from mere soulhood (“personality”)
to spirituality, so animals are reborn from sentience to soulhood
(“personality”).
Now we need to interpret “soulhood” in accordance with the passage
above from Lewis (2002), and we need to interpret “spirituality” in
accordance with Lewis (2001b). Doing so, we get the following.
(CA*) Just as humans are reborn from mere personhood to some-
thing that will be godlike, so animals will be reborn from sentience to
personhood.3
3
There is a passage in Butler where he seems not to be defending as true but
as at least possible a similar thesis: “that they [the ‘brutes’] must arrive at great
attainments and become rational and moral agents” this, he says, “would be
no difficulty, since we know not what latent powers and capacities they may be
endued with” (Linzey and Regan 1990, 90). Butler seems to have in mind what
Plantinga calls a “defense” rather than a “theodicy,” but it’s not clear.
Animal Saints 145
I have simply stripped Lewis’s analogy of the dependency thesis and

filled out the terms in accordance with what he has said elsewhere. I will
regiment the thinking of this section in the next.
8.3 The transcendental argument for animal deification
In this section, I will pull together many strands of thought from above
into an explicit argument for the conclusion that in the face of animal
suffering, theism entails animal deification.
The Transcendental Argument for Animal Deification (TAAD):
1. Animals have sentience. (From Chapter 5.)

2. Animals are made in the image of God. (From 1 and Genesis (see
next chapter).)
3. Animals have moral standing. (From 2 and from 1, independently.)
4. God is all-powerful and perfectly loving, overflowing with love and
concern for everything with moral standing. (Assumption.)
5. God will do justly and lovingly by animals. That is, he will not allow
harm to come to them that is not somehow compensated for, he
will see to it that their existences are on the whole quite good (more
than just better than on balance good) and that any suffering can be
defeated within the context of their lives. (From 3 and 4 and what
it means to be loving.)
6. The only way God could do justly and lovingly by animals is to
enfold their suffering in a greater good that organically defeats their
evil. (Established above.)
7. The only way God could enfold animal suffering into some greater
good that organically defeats it is either (i) via their relation to
cosmic order, (ii) this-worldly soul-making, or (iii) other-worldly
soul-making. (Provisional assumption.)
8. The argument from cosmic order is almost completely unsuccessful
as it stands. (From the fact that no aesthetic good can justify horren-
dous evil.)
9. This-worldly soul-making cannot occur to a significant degree due to
current lack of TTPU and other cognitive capacities. (Assumption.)
10. The only way God could enfold animal suffering into some greater
good is via future soul-making. (From 7, 8, and 9.)
11. The only way God could do justly and lovingly by animals is via
future soul-making. (From 6 and 10.)
12. Future soul-making requires both animal resurrection and deifica-
tion. (Seemingly obvious assumption.)
13. The only way God could do justly and lovingly by animals involves
both animal resurrection and deification. (From 11 and 12.)
14. If God exists, then animals will be resurrected and deified. (From 5
and 13.)
This argument is a regimentation of the material that preceded it. I was

delighted to find that Keith Ward hints at something like this when he
says, after considering a kind of defeat of evil,
This must be true for the whole of creation, insofar as it has sentience at
all. If there is any sentient being which suffers pain, that being – what-
ever it is and however it is manifested – must find that pain transfig-
ured by greater joy. I am quite agnostic as to how this is to happen; but
that it must be asserted to be true follows from the doctrine that God is
love, and would not therefore create any being whose sole destiny was
to suffer pain. (From a passage in Linzey and Regan 1990, 105).
I think he is exactly right, but I am not agnostic about how it happens. I

think it must happen via deification, because I think the only way God
can be just and good is to defeat all evil, and for animals that requires
deification.
We need to be clear about how the conclusion of TAAD – let’s call 14
the Key Conditional (KC) – constitutes a resolution of the puzzle with
which the chapter began. That puzzle was constituted by the fact that
on the one hand, in the previous chapter I argued that the only way
one could plausibly argue that the evils of this world were not radi-
cally improbable on theism was if one were able to incorporate a version
of the Chisholm–Adams notion of defeat into the soul-making picture;
but, on the other hand, such a route seems utterly blocked for animal
suffering due to a lack of TTPU. And note that defending animal resur-
rection isn’t enough to solve the problem. As Lewis points out,
A future happiness connected with the beast’s present life simply as a

compensation for suffering – so many millenniums in the happy pastures
paid down as “damages” for so many years of pulling carts – seems a
clumsy assertion of Divine goodness ... . In such a botched adjustment
I cannot recognize the master-touch. (2002: 635–6, emphasis added)
I completely concur (and affirm Lewis’s focus on divine goodness

rather than justice), as should be clear from my Gatenator case above.
Compensation is not even enough for justice, which is the low bar of
theodicy. Or, if you prefer, justice is not enough for a loving God (it is to
Animal Saints 147
their credit that Hick and Adams also keep this before the reader’s mind).
What is needed is the defeat of animal suffering. The paradigm case of
the defeat of evil is when the individual endorses their role in the drama
of creation and salvation and is glad to have played it (which might be
different than being glad for it4). I argued in the last two chapters that
the possibility of this kind of self-endorsement of one’s role in horrors
makes possible an effective theodicy. What the Key Conditional does
is establish two things. First, it establishes that the defeat-augmented
saint-making theodicy is available for animals. This establishes the
Conditional Thesis I promised to argue for: That if the defeat-augmented
saint-making theodicy is effective for humans, then it is just as effec-
tive for non-human animals as well. Whether the antecedent can be
discharged was given only limited defense. Full defense awaits another
volume. But the Conditional Thesis greatly ramifies the implications of
a successful defense of the defeat-enhanced saint-making theodicy. The
Key Conditional establishes the Conditional Thesis.
The second important thing the Key Conditional does is this: it not
only opens the theodicy of the previous chapter up to animals, it does
so without any loss in prior probability to theism. The logic of this move
is exactly parallel to that in the last two chapters. Let the thesis that
animals will be resurrected and deified be called “The Package,” P. Let
“T” stand for theism. KC is then represented as follows:
T → P, modulo empirical facts5
But this entails that
T ≡ (T&P).
The only metaphysically possible cell of the partition of T into ((T&P)

v(T&~P)) is the first disjunct. So the logical space of T just is the area of
(T&P). They have exactly the same measure and so the same probability.6
When one cell is logically impossible, we renormalize.7 So, given the
4
Again, see Hasker (1981). Nick Colgrove has written a paper for my graduate
seminar on this topic called “Gladness, Regret, and the Problem of Evil: A New
Haskerian Argument” which is a considerable extension of Hasker.
5
That is, all possible worlds in which God creates animals that suffer like they
do in our world are worlds where he grants them the package.
6
It is important to keep in mind that this is not a case where one rules out cells
empirically..
7
In this context, we don’t need to worry about zero-measure non-empty sets.
facts about animal suffering, the TAAD puts the problem of animal pain
on the same plane as the problem of human pain. As I have repeatedly
said, that might not be so great (though I firmly believe the theodicy of
the last chapter shows otherwise), but given the fact that the problem
of animal pain at first appeared to be a significantly harder problem,
putting them on par is progress for the theodicist.
8.4 Identity issues: objections to animal deification
A natural worry is whether resurrection and deification of certain

sentient creatures is even metaphysically possible. The concern8 is that
numerical identity cannot be tracked across a career spanning both a
stage as, for example, a newt and a stage as a very high functioning
person. In virtue of what would we say that this latter being with super-
human cognitive capacities was the same individual as the lowly crea-
ture? I will offer two answers.9
Continuity of soul A standard way to track personal identity for people
who believe in a soul is by following the soul. Since I will defend the tradi-
tional view that animals have souls in the next chapter, I am inclined to
take that route for animals, too. However, a common Aristotelian view
(whether Aristotle thought it or not) is that the animal soul is essentially
limited to the standard characteristic functions of sentience and local
motion so that there is no possible world where any possible newt is a
person, and it is impossible for Socrates to have been an alligator.
I simply deny this thesis, which as far as I can tell is undefended. I
am happy to take a view more like Scotus. According to Scotus, roughly,
every soul has the same capacities, but each soul’s functioning is limited
by the matter to which it is wed. An apt illustration involves electricity.
The same current entering a house can “animate” a toaster or a televi-
sion. The latter has much more complex functions (though I prefer toast
to TV) because of the material structure and circuitry into which it flows.
I will provide support in the next chapter that this is what the Hebrew
Scriptures teach about the nature of soul. Thus this notion of soul should
already be in the background evidence of the mere Christian. If I am
right, then the worry is unfounded.
8
This was pressed especially hard by Blake McAllister in class and by Marilyn
McCord Adams in her comments on a version of a portion of this chapter at the
Pacific APA, 2013 in San Francisco.
9
Both suggested in Adams (2013), though it never occurred to me to doubt
either one of them.
Animal Saints 149
Primitive thisness But there is another way to track identity that can
stand up to the changes under consideration, which is a philosophical
necessity anyway. Consider the following thought experiment based
on one by Chisholm.10 Consider Smalls and Talls. Talls is six foot seven
inches tall, weighs 285 pounds, and has a very pale complexion. Smalls is
barely over five foot tall, weighs barely over 100 pounds, and is very dark
complected. Furthermore, they look nothing alike, and have very different
beliefs, desires, memories, etc. For example, Talls is a vegan and Smalls
likes nothing more than pit barbeque. Talls drinks mostly Oregonian
Pinot Noir, but Smalls wouldn’t be caught dead with the stuff. In short,
Smalls and Talls have almost nothing in common, inner or outer. Yet we
can imagine them becoming gradually more like one another. Smalls can
become taller and Talls can become smaller. We can transfer pounds from
Talls to Smalls. We can even imagine their features coming to resemble
one another more and more. As they begin to converge on qualitative
duplicates, we begin to notice that they will “meet in the middle” and will
at some time be perfect qualitative copies of one another. Then they will
“pass” one another and Smalls will begin to resemble more the original
Talls, and vice versa. Not only can we do this for their physical appear-
ance, mind you, but also for their mental life and even for their external
surroundings. That is, we can gradually switch their beliefs, desires, and
apparent memories, until, at the end of the process, everything that
Smalls once believed, desired, and hoped for is now a belief, desire, and
hope of Talls instead. Also, we can place them in a symmetric world where
every statement like “Three meters to the left of _____ there is a statue
of Morgan Freeman” that is true of one of them, is true of the other. At
the end of the process, Smalls is an exact qualitative, mental, and relational
match for the original Talls, and vice versa. Yet Smalls is still Smalls and
Talls is still Talls. The reason is that each has what Scotus calls a haecceitas
or “thisness.” This thisness is a non-qualitative property that individuates
individuals. No matter what changes an individual were to undergo, their
thisness “follows” them. So Scotus offers us two ways to account for the
numerical identity of an individual across stages with radically different
functionality. On either of them, the worry goes away.
Identifying with one’s past But even if personal identity can be estab-
lished between the earthly, animal stage of the continuant and its heav-
enly, deified stage, perhaps the latter creature will not be able to “identify
with” their earlier suffering in the requisite way. Marilyn McCord Adams
makes this objection.
10
Chisholm (1967). See also Adams (1979) and Swinburne (1994: ch. 2).
Even if a newt feels pain at T1 when its tail is bitten off by another
animal at T1, the newt qua newt cannot turn this into “a meaningful
growth experience” at T2, because the newt has no self-consciousness
and so does not function as a meaning-maker during its newt career.
Even afterwards, it is difficult to believe that anything about the way
it responded qua newt to its newt-pain would “carry-over” to shape
that soul’s character-development in later lives.11
This seems to be a worry about particular goods gotten from particular

suffering. That is, it is a concern that no particular good is likely to come
to the newt at any time in virtue of anything about the way the newt
responds in her pre-personal newt-stage.
I wish to make it clear that I do not endorse the doctrine of meticu-
lous providence as it is usually described. I do not think it a criterion of
success for a theodicy that each evil have some particular good – bearing
a unique causal or logical relation to it – to justify it. I am painting a
global picture. Those pains were incurred as part of their playing a role
God needed played for global reasons pertaining to the whole commu-
nity of which the organism is a part. The global good God aims at did
not require that particular evil to have occurred, but it entailed or made
likely that something relevantly similar would occur.
Still, that very pain of that very creature – assuming we have personal
identity established – will be considered, on my view, by the creature who
suffered it as to whether it – and the whole of which it is a constituent –
is something they ultimately accept as an integral part of a very good
life. If when suitably engaged – I think of something like C.S. Lewis’s The
Great Divorce (2001a) here – the creature comes to see their relationship
to God as the saints do, then they will not regret having played their
role in the drama of salvation and they will not doubt that God has been
good to them. And their gladness is (and/or is a manifestation of) a great
virtue, and like the virtues of self-sacrifice, the authentic display of this
Marian/Gethsemane virtue is the highest created good. Their benefits
include their being of use to God and fellow creatures, their being in a
position of making a monumental decision of self-determination, their
glad awareness of the good they have done, and the other virtues this
leads to. Their earlier suffering was an occasion for their later, enhanced
selves to make a judgment call about whether they are glad they were of
use to God and fellow creatures or whether to be bitter about it. It is a
good thing that they have this opportunity. And it is unreasonable and
11
Adams (2013).
Animal Saints 151
vicious to regret it on the assumption that God is in fact just and worthy
of love and total dedication and is good to all of his creatures.
So I don’t think there is anything to worry about here. Nevertheless,
pace Adams, I do think that even if there needed to be some particular
connection between the newt’s newtly pain and the newt’s saintliness,
that connection can be accommodated. Imagine that you are reading
accounts of horrendous childhood suffering. It is hard to read, so you
can only read a little bit each day. On one particular day, you are reading
a story about the suffering of a child with the exact same name as you.
As the story goes on, more and more details of the child’s life match
what you know of your own childhood. Intrigued by this, you perform
some investigation only to discover that it was you, and you blocked it
out. Suppose further that, even after repeated attempts to recover these
memories, they remained blocked to you. Nevertheless, all the relevant
details of your childhood are recorded in the book.
So take one particular detail, a day upon which some terrible thing
occurred to you. You are horrified that such a thing should happen to
a child. And though you don’t remember it, you are well aware that it
is you who suffered it. Here’s what I didn’t tell you. You are reading this
book in a cottage in Purgatory. You’ve been given the cottage, which is
on the edge of an area of transition between a dense wood and a plain,
for as long as you want it to review your life. Occasionally, you will be
paid a visit by a person, a human person, but very tall, who asks if you
need anything. One of the things your aide can do is provide you with
information about your life and its connection to enough of the rest
of creation that it is unlikely that your questions would outstrip their
knowledge.
Your aide visits on the day that you have read this passage, and you
have been wondering how in the world this instance of suffering, which
seems to you so utterly pointless, could have led to any good. The aide
begins unwinding the story. At first, they seem like meaningless, unre-
lated details. The aide’s story goes on well into the night. At times, you
think you see connections between what they are saying now and what
they said at some time earlier. The eastern sky begins to lighten and your
aide pauses their story only to make you a tea made from some herbs
they brought in a pouch. With you refreshed, the story continues. After
seven cups of tea, your aide finally enters silence with the words “And
that is how it happened.” For what seems a very long time, you both sit
in silence. With a nod, your aide takes leave. You remain seated, silent for
some time more. “Does this story make sense?” (It doesn’t occur to you to
doubt whether it is true.) “Was it worth it?” You return to the book. You
re-read the daily portion that you had finished before your aide arrived.
You still can’t recall this happening to you, but re-reading the story with
the aide’s context in mind triggers a gestalt. “Very well then.” you say, “I
did my part.” You then go and release a dove from a cage, which is the
signal that you are ready to depart your cottage for the next destination.
This story illustrates how even details of one’s life one can’t recall can be
incorporated into one’s first-personal account of one’s life.
C.S. Lewis pictures the first stages of this process for an animal. In The
Magician’s Nephew (1994), which is chronologically first in the timeline
of The Chronicles of Narnia collection, the protagonists, young Polly and
Digory, end up, via some bad choices, bringing a witch into their own
world from one they were illicitly visiting. This begins to cause serious
trouble for Earth, so the children realize they need to get the witch out
of their world’s realm and back to the Wood between the Worlds. The
witch has commandeered a horse-drawn cab and made off with it, astride
the top, driving it like a chariot. Polly and Digory finally get their hands
on her and slip on the yellow ring (the means by which they can return
to the Wood between the Worlds), but she is on the horse and the cabby
is holding the horse, trying to calm it. As a result, they all (including
some shards of the lamp post they crashed into) end up in a different
world – Narnia – just as it is being created. The old cab horse, Strawberry,
gets caught up in the magic of creation as Aslan is bestowing upon some
animals the gift of speech. The Cabby seems nervous about this.
All this time the Cabby had been trying to catch Strawberry’s eye.
Now he did. “Now Strawberry, old boy,” he said. “You know me. You
ain’t going to stand there and say as you don’t know me.”
“Well,” said Strawberry very slowly, “I don’t exactly know ... I’ve
a feeling I lived somewhere else – or was something else – before
Aslan woke us all up a few minutes ago. It’s all very muddled. Like a
dream.”
“What?” said the Cabby. “Not know me? Me what used to bring
you a hot mash of an evening when you was out of sorts? Me what
rubbed you down proper? Me what never forgot to put your cloths
on you if you was standing in the cold? I wouldn’t ’ave thought it of
you, Strawberry.”
“It does begin to come back,” said the Horse thoughtfully. “Yes. Let
me think now, let me think. Yes, you used to tie a horrid black thing
behind me and then hit me to make me run, and however far I ran
this black thing would always be coming rattle-rattle behind me.”
“We ’ad our living to earn, see.” Said the Cabby. “Yours the same as
mine. And if there ’adn’t been no work and no whip there’d ’ave been
Animal Saints 153
no stable, no hay, no mash, and no oats. For you did get a taste of oats
when I could afford ’em, which no one can deny.”
“Oats?” said the Horse, pricking up his ears. “Yes, I remember some-
thing about that. Yes. I remember more and more. You were always
sitting up somewhere behind, and I was always running in front,
pulling you and the black thing. I know I did all the work.”
“Summer, I grant you,” said the Cabby. “‘Ot work for you and a cool
seat for me. But what about winter, old boy, when you was keeping
yourself warm and I was sitting up there with my feet like ice and my
nose fair pinched off me with the wind, and my ’ands that numb I
couldn’t ’ardly ’old the reigns?”
“It was a hard, cruel country,” said Strawberry. “There was no grass.
All hard stones.”
“Too true, mate, too true!” said the Cabby. “A ’ard world it was. I
always did say those paving-stones weren’t fair on any ’oss. That’s
Lunn’on, that is. I didn’t like it no more than what you did. You were
a country ’oss, and I was a country man. Used to sing in the choir,
I did, down at ’ome. But there wasn’t a living for me there.” (1994:
144–6)
Their discussion is interrupted, so we don’t get to hear its end, but the
important thing to note here is the way the human needs to give an
account to the horse for his behavior to the horse during their earthly
careers. The Cabby is able to say some things that justify his behavior
but also shows regret. One gets the sense that they will “work it out”
and come to peace. Heaven is bound to include this pairwise process for
every pair of individuals who had any impact on one another and then,
ultimately, with God. In the process of reconciliation, if God has been
good to his creatures and acted justly by them, then if they are virtuous,
they will find their peace with God. In this peace will be the defeat of
the evil they suffered.
If the future of each sentient creature is as I have envisioned it, then
even though, on my view, there is no need for each individual instance
of suffering to have some particular impact on the psyche of the person
who suffered it, there will nevertheless be an opportunity for it to do so.
And even if an individual cannot “identify with” their past in a certain
psychological way, they can still – knowing how it fits in with the rest
of their life and the lives of others – embrace their role in the drama of
creation.
9
Animal Afterlife
For animals to survive death, one of two things must be true. Either
animals have non-material souls, as most people, especially religious
people, think (though Evangelical Protestantism is a notable exception),
or, it seems, it would have to be possible for the existence of a creature
to be “gappy” in some way. That is, it would have to be possible for a
creature to exist for a time, cease to exist for a time, and then come back
into existence at a yet later time. In fact, I think both of these things are
true. But all I need is for one or the other to be true. By arguing independ-
ently for each disjunct I will be providing two independent arguments
for the needed disjunction. I will be providing arguments that are brief
and related to a large literature, but, I trust, demonstrate the path I stake
out in these issues.
In Section 9.1, I argue for the existence of animal souls. The argu-
ment for animal souls is broken into three parts corresponding to the
Anglican idea of the “three-legged stool” upon which theology sits (the
idea is often attributed to Richard Hooker, but it only has a modicum of
support in his work): Scripture, Tradition, and Reason. I will treat them
in that order in Section 9.1.1, Section 9.1.2, and Section 9.1.3. The argu-
ments will be quite condensed, but, I think, powerful. The purpose is
not to end any debate in philosophy, but, rather, to demonstrate what
a reasonable person can have (and many readers do have) in their back-
ground evidence prior to coming to the problem of evil in general and
the problem of animal pain in particular. In Section 9.2, I offer a model
for gappy existence of a sort. The principle resources are four-dimension-
alism and wormholes. Heaven is conceived of as a contiguous region of
spacetime or a parallel spacetime.
154
Animal Afterlife 155
9.1 Animal souls
9.1.1 Biblical support for animal souls

In the beginning of the Hebrew Scriptures,1 it seems clear that animals
and humans share the same animate nature (thus “animals”). First let
us display the passage that dramatically tells the story of the creation of
the first human.
Then the Lord God formed man from the dust of the ground, and
breathed into his nostrils the breath of life; and the man became a
living being. (Genesis 2:7)
By the time many readers get to this passage, they have already forgotten
the easy-to-gloss-over verse not far away.
And to every beast of the earth, and to every bird of the air, and to
everything that creeps on the earth, everything that has the breath of
life, I have given every green plant for food. (Genesis 1:30)
It is true that in Genesis 2, “breath of life” renders the Hebrew neshamah

and “living being” renders the Hebrew nefesh, but apart from an intruding
mysticism, it seems clear that becoming a living being just is having the
breath of life. So even as different a translation as the New International
Version (NIV) renders nefesh as “breath of life” in Genesis 1:30. There is
no reason at all from within the text to think that neshamah and nefesh
mean anything significantly different. The NRSV and the NIV (as well
as Young’s literal translation and the New English Translation) no doubt
translate nefesh in Gen 1:30 as equivalent to neshamah in Gen 2:7 since
in Gen 1:30 nefesh is the object of the preposition ásher-Bô, “wherein”
(there is). The more flat-footed New American Standard translation has it
“and to every beast of the earth and to every bird of the sky and to every
thing that moves on the earth that has life,” then includes the foot-
note regarding the last phrase “Lit in which is a living soul.” Throughout
Genesis 1 (verses 20, 21, and 24) nefesh is translated as “creature,” “living
thing,” “living creature,” “creature having life,” “soul,” and “living
soul” by various translations. And this makes sense, since Gen 2:7 has
two important phrases that share parts the “breath of life” transliter-
ates as nish’mat chaYiym, the first term being the cognate of neshamah,
1
All quotations will be from the New Revised Standard Version (NRSV) of the
Bible, unless otherwise stated.
“breath,” and the second being the cognate of chay, “life.” The second
is what the thing that receives the breath of life becomes: nefesh chaYäh,
“a living soul.” You become a living thing when you receive life; you are
alive when you get life. Life comes in the form of breath “neshamah,”
which like the Greek word pneuma means “breath,” “spirit,” and “soul,”
as it is in Isaiah 57:16, Job 26:4, and Proverbs 20:27 at least.
The upshot is that there is no reason at all in the text (or from context)
to drive a wedge between the life that humans get in Genesis 2 and the
life that non-human animals get in Genesis 1. In the beginning, animals
and humans are presented as having the same divine stuff. One might
wonder why the breathing in is only mentioned in Genesis 2, which
tells of the enlivening of humans and not in Genesis 1, which treats the
enlivening of non-human animals. Isn’t this a sign of humans’ special
status? The answer is that it is indeed such a sign, since humans are not
only ensouled but have capacities on earth that allow them to enter into
a special kind of covenant with God. But this is no reason to think that
only humans and not non-human animals have a non-material compo-
nent. That non-human animals also have this, and that it is of the same
essence as humans, is made clear above. It is generally accepted that the
author(s) of Genesis 1 is different from the author(s) of Genesis 2 (tech-
nically the transition is at 2:4), and the author(s) clearly have different
purposes. Genesis 1 is about the creation of the world, and Genesis 2 is
about the beginning of the human family. Yet the author(s) of Genesis
2 clearly picks up the language of the author(s) of Genesis 1 and sees
humans as metaphysically continuous with other animals.
This covers the most important passage of the Scriptures relevant to
our purposes here, but there are others that merit brief treatment. First,
consider this passage from Ecclesiastes 3, where the author is fretting the
lot of mankind.
I said in my heart with regard to human beings that God is testing

them to show that they are but animals. 19 For the fate of humans
and the fate of animals is the same; as one dies, so dies the other.
They all have the same breath, and humans have no advantage over
the animals; for all is vanity. 20 All go to one place; all are from the
dust, and all turn to dust again. 21 Who knows whether the human
spirit goes upward and the spirit of animals goes downward to the
earth?
One must, of course, be careful in interpreting passages that are plau-

sibly either non-literal or written from a perspective not endorsed by
God. Scripture records speeches of people who are speaking evil and
speeches of people who are speaking nonsense. In this passage, we

would clearly be wrong in taking it as a scriptural teaching that “all
is vanity” or even that humans are “but animals.” The literal content
is clear though: Animals and humans are on par in this regard: They
both die, and the fate of each is uncertain. Nevertheless, there is a clear
presupposition here that both humans and non-human animals have
souls. In fact, the author picks up the very same language of Genesis,
saying that they “have the same breath,” that is, the breath of God
that makes them more than mere matter, makes them animate beings,
animals.
This pattern is also seen in the Psalms. Consider these verses of Psalm
104.
24 O Lord, how manifold are your works!

In wisdom you have made them all;
the earth is full of your creatures.
25 Yonder is the sea, great and wide,
creeping things innumerable are there,
living things both small and great.
26 There go the ships,
and Leviathan that you formed to sport in it.
27 These all look to you
to give them their food in due season;
28 when you give to them, they gather it up;
when you open your hand, they are filled with good things.
29 When you hide your face, they are dismayed;
when you take away their breath, they die
and return to their dust.
30 When you send forth your spirit, they are created;
and you renew the face of the ground.
This language clearly evokes Genesis 1. In the NRSV, there is a footnote

after “your spirit” reading “or your breath.” This is the breath that God
breathes into matter to make living beings. When he takes it back, the
creature dies. We have in these three passages a clear pattern: a laying out
of the view in Genesis and an affirmation of it in Ecclesiastes and Psalms.
Furthermore, there are no counter-trends in the Hebrew Scriptures, no
passages in support of the opposite picture. Thus, for the theist already
committed to any significant authority for Scripture, their background
evidence prior to considering the problem of animal suffering will
already contain an important component of the saint-making theodicy.
We next turn to the broader Christian tradition.
9.1.2 The traditional Christian view about animal souls

There is little said explicitly about animal souls (in the relevant sense) in
the Christian tradition (medieval theologians even talked about “vege-
tative souls” in plants, but discussing this would be too big a digression).
So much of what we glean from the tradition has to be inferential and
holistic. However, given both that the Hebrew/Christian Scriptures seem
clear on the matter and that the major theologians of the church bathed
their thought in Scripture, their words should be read in consonance
with what they are taught in Scripture.2 The purpose of this section is
merely to sample three important figures from the three main branches
of Christendom – Eastern Orthodox, Roman Catholic, and Protestant –
and argue that their teachings are not only consistent with the thesis
that animals have souls (and to that extent corroborate it) but favor
that hypothesis over its negation. This further supports that the average
mere Christian should have in their background evidence for that
hypothesis or material that favors that hypothesis. Thus, bare theism
plus fairly minimal background historical information make very prob-
able mere Christian theism. So, without a significant decline in prior
probability, the bare theist has the expanded explanatory power of mere
Christian theism. And we are gaining evidence to believe this includes
animal souls. These arguments are in addition to the a priori arguments
provided earlier.
Basil the Great Basil of Caesarea, or St. Basil the Great, was a crucial
supporter of the Nicene Creed, and thus a pivotal figure for the pre-
schism Church. He is one of the three most important theologians of
the Eastern Orthodox Church (where he is called “revealer of heavenly
mysteries”), and he is recognized as a Doctor of the Church by Roman
Catholics. He is responsible for a prayer in the Russian liturgy that
implies that animals are more than automata. We must remember that
not only does neo-Cartesianism deny that animals are sentient, it also
denies that they have any dignity and communion with mankind. So
affirmations of the dignity of non-human animals and their brother-
hood with us are denials of neo-Cartesianism as much as is evidence
that animals are sentient and have emotions. So consider the words of
St. Basil’s prayer.
O God, enlarge within us the sense of fellowship with all living

things, our brothers the animals ... . May we realize that they live, not
2
Much of the raw material for this section is gleaned from Jones’s magisterial
(2009).
for us alone, but for themselves and for thee, and that they have the
sweetness of life. (Jones 2009: 61)
This deep fellowship and kinship does not mesh well with the thesis
that animals are mere hunks of matter. He goes on,
We pray thee, O Lord, for the humble beasts ... and for the wild
animals ... for thou has promised to save both man and beast. (Jones
2009: 61)
The probability that he would say these things if he thought that animals
have souls is greater than if he did not believe that. His saying these
things fits better into a larger version of the story according to which he
believes in animals having souls than a story according to which he does
not. And, I think, this is so to a considerable degree: his saying these
things is almost inexplicable.
One might object that he goes on to include not only non-sentient
beings but even inanimate objects.
Remember O Lord, the air of heaven and the fruits of earth, bless them.
Remember, O Lord, the water of the rivers, bless them, raise them to
their measure according to your grace. Remember, O Lord, the seeds,
the herbs, and the plants of the field, bless them. (Jones 2009: 62)
The immediate response is to note that after all these have been grouped
together, he sections off humans and animals with a final “Remember,
O Lord, the safety of the people and the beasts” (Jones 2009: 62).3 The
combining of people and beasts in the same section, apart from all the
rest, is to be expected more on the hypothesis that Basil takes animals
to have souls than on the hypothesis that he does not. And it is relevant
to note that the eastern tradition, which Basil inhabits, includes the
following strong statement in the Armenian liturgy.
Every creature which was created by you will be renewed at the resur-
rection, that day which is the last day of earthly existence and the
beginning of our heavenly life. (Jones 2009: 64)
St. Thomas Aquinas The central figure of the Medieval Church and the
most revered Doctor in Roman Catholicism is St. Thomas. Rather than
engage Thomas’s text directly – which would be a daunting task – I will
3
Jones notes this as well.
present a standard account according to traditional scholastic manuals.

These works, even if they don’t get Aquinas exactly right, themselves
represent an important tradition within the church. Aquinas borrowed
from Aristotle the notion of three distinct kinds of soul: “vegetative,”
“animal,” and “rational.” Every living thing is said to be “ensouled.”4
Each kind of soul is distinguished by its characteristic functions. Given
the etymological definition of “psychology” (Gk psyche, soul) and that
“every living body (plant, animal, man) has a vital principle, or soul”
it follows that “all living bodies, including plants, are studied in philo-
sophical psychology” (Donceel 1955: 3). However, the scholastics are
clear that the souls of plants and animals are “material” in a sense:
A thing is said to be material either (1) when it can be seen, touched,

weighted, measured, when it is perceptible to the senses; or (2) when
in all its operations, and for its very existence, it is intrinsically
dependent on matter. (Donceel 1955: 35)
Thus, though they admit that animals have consciousness (37), they
deny that an animal’s soul can have post-mortem survival. After pointing
out that the characteristic functions of animals are sense knowledge and
sensory appetite, which are essentially dependent on matter, Donceel
concludes,
[T]he animal soul, from which these material operations derive, is

also intrinsically dependent on matter, is material. Therefore the
animal soul cannot exist without a body, and when bodily life ceases,
that soul disappears, as roundness disappears in a flattened ball. (64)
At first, it appears that the scholastic tradition is contrary to the survival

of animal souls. However, this appearance is removed along with the
bad thinking involved. It is clear from context that the scholastics do
not take the animal soul to be a material object (material in Donceel’s
first sense). And the inference
The functions the soul is the necessary condition for5 have ceased.
Therefore,
The soul has ceased to exist.
4
Brennan (1941: 8) and Klubertanz (1953: 50).
5
Brennan (1941: 7).
is transparently bad. A thing can cease to function without ceasing to exist.

Furthermore, the scholastics note, with some concern for man’s immor-
tality, that even human abstract reasoning as human abstract reasoning
depends on sense knowledge as well. However, they say that “Even after
its departure from the body, the soul of man retains its natural inclina-
tions to be united once more with matter” (Brennan 1941: 284). This
seems a bit of a cheat, and there is no reason at all why the same move
can’t be made for animals or dropped for both humans and animals in
favor of the simple persistence of an existent non-physical entity. So even
though, via their overly scrupulous concern to establish the superiority
of human over non-human animals (and this comes out clearly in not
only these three books but Thomist literature in general), it initially looks
like the scholastic Thomist tradition is against the existence of persistent
animal souls, their actual (reasonable) teachings support it.
John Wesley John Wesley makes a good representative for
Protestantism. He was a committed Anglican clergyman yet gave rise to
one of the most American brands of Protestantism. In his sermon “The
General Deliverence,”6 which begins from Romans 8:19–22, considers
the problem of evil. In considering the problem of animal suffering,
though like most he denies the Animal Independence Thesis discussed
previously, he affirms their resurrection. In Wesley’s terms, it seems one
needs a soul for that.7 Indeed, he even almost goes as far as the deifica-
tion thesis I have defended.
The whole brute creation will then, undoubtedly, be restored, not

only to the vigour, strength, and swiftness which they had at their
creation, but to a far higher degree of each than they ever enjoyed.
They will be restored, not only to that measure of understanding
which they had in paradise, but to a degree of it as much higher than
that, as the understanding of an elephant is beyond that of a worm.
(Sermon 60, Wesley 1872a: III.3)
But he does not go on to affirm anything like a soul-making theodicy. He

rests with the fact that God “will make them large amends for all they
suffer while under their present bondage” (III.5). But he does “conjec-
ture” the very thing I have defended.
6
Wesley (1872a).
7
See Sermon 109 “What Is Man?” (Wesley 1872b). Available at http://www.
umcmission.org/Find-Resources/John-Wesley-Sermons/Sermon-109-What-
Is-Man
May I be permitted to mention here a conjecture concerning the brute

creation? What, if it should then please the all-wise, the all-gracious
Creator to raise them higher in the scale of beings? What, if it should
please him, when he makes us “equal to angels,” to make them what
we are now, – creatures capable of God; capable of knowing and
loving and enjoying the Author of their being? (III.6)
This is one of the theologically strongest affirmations of the animal

soul I have come across. There is strong precedent for animal souls in
Orthodox, Catholic, and Protestant thought. The above three individ-
uals are just samples of the kind of thing detailed research unearths.
The considerations adduced thus far provide some good evidence that
animals’ souls are already included in the mere Christian view.
9.1.3 Philosophical support for animal souls

From Aristotle to Aquinas, the most common argument for a non-
material soul in humans was the human ability to perform abstract
reasoning. Though there are studies that suggest that certain higher
mammals can perform abstract reasoning, picking through the studies
and distinguishing using logic from acting in accordance with logic would
require painstaking detail.8 In the modern era, however, the issue
that has most motivated arguments against materialism has been the
“problem of consciousness.”9 The problem is that it is very hard to see
how consciousness – in this case, the hosting of “qualia,” mental proper-
ties with qualitative feel, like pain or visualizing bright yellow, or hearing
the sound of a minor chord – could be reducible in some way to mere
matter. How could a conscious mental state be a physical thing or fully
realized in a physical thing? Some go further and say that not only do
they fail to see that it could be the case, but, more strongly, they see that
it could not be the case.10 I will consider the two most prominent argu-
ments for the thesis that humans have non-physical parts founded upon
the possession of consciousness. I will suggest that theists have reason
to reject the first one (the “knowledge argument”), but I will defend the
second (the “zombie argument”). What is very surprising is how little it
has been noted that if these arguments are successful for humans, they
are equally successful (or very nearly so) for any sentient being.11
8
Some of the basic issues are treated in Andrews (2010).
9
It is really a cluster of problems. See Van Gulick (2012).
10
For this kind of claim see Plantinga (2007).
11
Swinburne (1997a: 180) is a notable exception.
9.1.3.1 The knowledge argument

Consider Mary.12 Mirabile visu, Mary has lived all her life in a black-
and-white house. That is, everything in her entire experience has been
either black or white (including, somehow, herself and all her food). At
some point, she does an intense study of perception. She reads every-
thing there is to know about the neurophysiology of vision, and she
studies fundamental physics, statistical mechanics, and anything you
think underlies the neurophysiology of vision. She watches many exper-
iments involving vision on her black-and-white television. She learns all
the physical facts about perception. But then, she is let out of her house.
She comes outside and, for the first time, sees a cardinal on a power line
against a blue sky. She has just learned something about perception. But,
since she already knew all the physical facts, it follows that there is more
to perception than the physical facts.
While this argument is highly suggestive, the theist has a reason to
reject it. The simplest form of theism entails that God knows all facts.
The use of the knowledge argument to defend anti-physicalism causes
problems for this simple view. Here’s how. Consider what kind of knowl-
edge Mary gains when she leaves the black-and-white house and sees
the bird on the wire. She learns what it is like to see in color. This is
obviously a very different kind of fact than that a certain wavelength
of light entering the retina causes a certain set of neurons to fire. Now
consider this. There are all kinds of things such that God does not know
what it is like to do those things. He does not know what it is like to be
tempted to evil. He does not know what it is like to fail in an attempt.
He does not know what it is like to have a toothache (the Incarnation
may complicate this one, but there are plenty of experiences Jesus didn’t
have). So if knowing what it’s like to do something is a fact in the ordi-
nary sense – the sense over which omniscience is defined – then God
will not turn out to be omniscient. But this seems like a weird reason
to deny God’s omniscience. It seems more natural to think of knowing
what something is like as not a fact in the relevant sense. This blocks the
argument against God’s omniscience, but it also blocks the argument
against physicalism.
The theist could modify the definition of omniscience to say that
God knows all the facts except facts of kind K, where K will have to be
spelled out in such a way to capture the relevant class of “what it’s like”
facts. This, however, would come at a considerable cost to elegance and
12
This argument is due to Jackson (1982). This telling is my own.
simplicity in the theistic hypothesis.13 I turn now to the second of the

two most prominent contemporary arguments against physicalism: the
problem of consciousness.14 I will give versions of the two most promi-
nent arguments concerning consciousness: inverted spectrum, and
“zombies.”
9.1.3.2 Qualia: inverted and absent

“Qualia,” you may recall from above, are features of what it is like to
undergo certain experiences. For example, the redness of a reddish expe-
rience of seeing a cardinal and the sourness of the experience of biting a
lemon put you in touch with a red quale and a sour quale, respectively.
In these experiences you host “phenomenal red” and “phenomenal
sour” as distinct from any related and similarly named properties in the
bird or the lemon. Our mental lives are rich in qualia.
But recall Mary, the homebound color expert. She and you – if you are
not color blind – look at one and the same object and yet have different
conscious experiences. Furthermore, these differences in your private
experience will show up in your public behavior. There is no simple map
from her black-and-white states to your multi-colored states. However,
we can imagine a case similar to that described by John Locke in which
someone’s color spectrum was simply “reversed” or otherwise mapped
onto the ordinary person’s color spectrum in such a way that this private
difference makes no public difference at all.
Probably, many readers thought about this in grade school. “What if
I saw apples as green and grass as red but because I grew up in the same
culture I called things that looked red to me ‘green’ and things that
looked green to me ‘red’ so we never knew about out the difference?”
There is nothing at all to suggest that this is not a genuine possibility.
It seems clearly possible, and should be presumed to be possible until
there is some reason to think it isn’t. But its possibility is hard to square
with physicalism. On a physicalism worthy of the materialist heritage,
fixing the physical facts will always fix the mental facts. But if spectrum
inversion is possible, there will be cases where two physical duplicates
will have different mental states.
13
Though, perhaps, such knowledge is logically impossible and so the theist
can still just say that God knows everything it is logically possible for a being like
him to know. There are well-known problems for this kind of move, however. For
a discussion of these issues, see Wierenga (2013).
14
Sometimes the problem of consciousness is divided into a “hard” problem
and an “easy” problem, but that is not relevant to my discussion.
In fact, it seems clearly possible that there could be physical dupli-

cates of us with no mental states at all. These beings are usually called
“zombies.”15 It seems clearly possible that particles just like those of
which we are composed could come together to form a “copy” of us,
but in such a way that the matter was unconnected to any conscious
states. Aside from a prior commitment to physicalism akin to fideism
(what Perry calls “antecedent physicalism”16), there is no good reason to
doubt the possibility. The average person, who is an instinctive dualist,
is well within her epistemic rights for believing humans to have a non-
physical aspect. However, she has no good reason to have significantly
less confidence that non-human animals also have this non-physical
aspect, since there is no good reason to doubt that they are sentient as
well, and it is consciousness as such that is the main issue here.
If one adopts a stance of modal skepticism according to which the
ways of the mind simply exceed the power of human understanding,17
then it is hard to see how such an individual can raise the problem of
evil in the first place. For the argument seems to require one to have
significant knowledge about the possibility space in naturalism.
But apart from the direct insight into the possibility of zombies,
perhaps looking at it from this angle will help one to see the possi-
bility of zombies. According to physicalism, any two states that, at some
time t, are physical duplicates should also be mental duplicates at t. So
imagine two very different worlds that begin in different ways. They
both begin with a big bang, but one expands much more rapidly than
the other. In one, the quantity of matter and energy remains constant,
while in the other, matter can be destroyed and energy lost forever. For
eons, they evolve quite differently, but then they start to converge on
the same state. For a moment, they “cross” and then diverge again. At
the moment at which they are physical duplicates, I think it is easy to
see, in this scenario, that even if the other world is, for that moment, an
exact physical match for ours, there is no reason to suppose it would be
a mental match or even have any mental states at all.
As a last-ditch effort, the physicalist could borrow a page from the
semantic externalist book and say that two physical states are not phys-
ical duplicates at t unless they had a type-identical causal history. What
could motivate such a move beyond the desire to preserve physicalism
from this objection, I simply don’t know.
15
The popularity of this term stems from Chalmers (1996).
16
Perry (2003).
17
See, for example, McGinn (1989 and 1999).
These arguments provide powerful reasons to reject physicalism. They

provide reason to think that any sentient being has a non-physical
aspect. I think people grasp the basic ideas explicated in these argu-
ments even if they never formulate them in these ways. So I think the
proposition that animals have souls is very probable on people’s back-
ground evidence.
9.1.4 Conclusion
In conclusion, it is a prior commitment of most mere Christians that
animals have non-physical souls, and – for reasons pertaining to the
argument from consciousness – it should be in the background evidence
of anyone who is a dualist about humans.18 Thus, when the theist appeals
to animal souls in her theodicy, she is not adding in extra content. This
is so whether we are considering the “baseline” to be where the theist
finds herself when she is confronted with the problem of animal pain,
or even with a “lower” baseline of a secular or religiously neutral dualist
about human persons (which is what the vast majority of people are).
9.2 Gappy existence
9.2.1 Introduction
I have just argued that animals do have souls, non-physical parts that
can track their identity after death and to the resurrection. However, I
understand that some readers just won’t be able to swallow that. But if
animals do not have souls, then it seems that the only way they could
be raised from the dead is if it were possible for things to have gaps in
their existence. In this half of the chapter, I will defend that, in a certain
sense, gappy existence is possible.
9.2.2 Natural skepticism about gappy existence

When a thing goes completely out of existence – that is, when it doesn’t
just go dormant in some way, but literally ceases to exist in any way –
the natural thought is “that’s it.” God (or a very clever scientist) could
make an exact copy of the organism, but that would be a different affair
than bringing the original organism back to life. You could make an
18
There is also an argument to be made from data pertaining to abstract
reasoning, for some studies suggest that some higher mammals are capable of
using simple syllogisms. Doing so was the basis of a standard scholastic argument
for dualism. There is not room here to do the argument justice, however, since it
would involve detailed scrutiny of the studies.
exact copy of me now. If you did, you wouldn’t have made another me.
You’d only have made something as like me as a thing could be without
being me. Since it is an exact qualitative copy, we would differ, my clone
and I, only in our location and in our primitive “thisness.”19 But that’s
enough to make it not me.
In light of these considerations, it might just seem hopeless to make
coherent a notion of gappy existence. As we shall see, however, this worry
depends on an impoverished conception of the world: a conception that
limits the world to what is present. I will introduce the thicker alterna-
tive first via fiction, then gradually introduce details from physics and
metaphysics. Some readers will say “Of course!” but others will be at first
incredulous on a number of aspects. Of those readers I ask patience and
sympathy. The goal throughout is to argue that what I propose is already
probable on readers’ background evidence. At the very least, I will describe
a kind of person – certainly a person like me before ever considering these
matters – on whose background evidence my thesis is antecedently likely.
9.2.3 A description of the model

Hopefully, many readers have read the delightful and enchanting book
A Wrinkle in Time (L’Engle 2007). In the book, the protagonists (a teenage
couple Meg and Calvin), are transported in their adventure to a planet
in Messier 101, a galaxy 21 million light years away from Earth. When
told this, Calvin is incredulous. “Even traveling at the speed of light it
would take us years and years to get here” (70). Their mysterious guide
replies “Oh, we don’t travel at the speed of anything ... . We tesser [from
the term “tesseract” for a four-dimensional hypercube]. Or you might
say, we wrinkle” (70). This notion gives the book its title. The idea is
simple, the background is a bit less so.
Einstein is most famous for his special and general theories of relativity.
The special theory came first, in 1905. The main upshot for our purposes
is that space and time are interdefined as different dimensions of a single
manifold or continuum or “fabric.” The general theory (so-called because
it generalized or brought together the special theory and Newton’s
theory) came a decade later. Its main upshot for our purposes is the idea
of spacetime bending around massive objects. This explains why light,
though lacking mass, appears to bend around massive objects. The light is
traveling “straight,” it’s just that the medium is itself bent. By nature, the
bending of a region of spacetime is proportional to the mass of the object
“occupying” it. But of course God or some other great power could bend
19
For more on this notion, sometimes called a “haecceity,” see Adams (1979).
or wrinkle it at will. L’Engle’s wrinkle in time is essentially what scientists

call an Einstein–Rosen Bridge, or, more popularly, a “wormhole.”
General relativity implies the possibility of wormholes, but whether or
not there are any is unclear, and proposed theories according to which
they are common at the very micro level only postulate (very, very)
tiny, temporary, and unstable ones. However, there seems to be cred-
ible research into their artificial stabilization with the introduction of
matter with negative mass to counteract gravity. However it goes on the
empirical and technological sides, the theory is quite simple. Given the
appropriateness of thinking of spacetime as a pliable substance capable of
bending and stretching, we can imagine the universe lying flat like a map
on a table. Let there be two planets on the “map” at a great distance. Label
one “A” and the other “B” and draw a line between them, with midpoint
M. If you fold the map at M along a line perpendicular to line segment
AB, you will eventually bring A immediately adjacent to B. So if one were
able to take advantage of “wrinkling” or otherwise warping spacetime,
whether in a wormhole or some other manner, one would travel to the
future “skipping” a segment of Earth’s timeline. If you google “wormhole
time travel” you will get some great visual illustrations of this. If you
google “tesseract” you will most likely get a hypercube, which, though
very cool, is not really what a wrinkle in time is.20
Note two things. First, the personal timeline of each individual would
be entirely non-gappy. What is “skipped” is the Earth’s timeline. Note,
second, that there is nothing particularly special about Earth’s timeline.
Given that most people there have died, Earth’s timeline is a minority
timeline. And presumably the Earth itself has no awareness of time, so
the important thing about Earth’s timeline is that it parallels the personal
timelines of individuals who have not yet died. But there is no (human)
person who lives on Earth from its inception to its destruction, so what
we are really talking about here is a timeline formed by the overlapping
timelines of the people on Earth who have not yet entered the timeline.
But for every person who enters that timeline, as many (or very nearly
as many) will eventually leave it, on this view, to enter the afterlife. So
there is no gappiness to be worried about.
Of course, all this assumes that the future is real in a way that is ontologi-
cally on par with the present – a thesis called “eternalism.” I confess I have
never been able to think in any other way and was greatly surprised that
20
A comment by Nick Colgrove made me realize that if one folds spacetime
about an axis “perpendicular” to the time line, one will only travel in space,
not time, though even here issues pertaining to simultaneity would complicate
matters a bit.
there were people who thought that only the present was real (I get claus-
trophobic just thinking about it!) – a thesis called “presentism.” Eternalism
also assumes some further theses about time, such as the reducibility of
tense (most importantly, the elimination of “now,” “was,” and “will,”
in favor of dated and comparative “before,” “after,” and “simultaneous
with” language). The eternalist view of time and of past, present, and
future seem to me to have been the mainstay of the Christian tradition,
and so they should be very probable on the background evidence of any
mere Christian. I think, in fact, it is entailed by bare theism, but to defend
that would take us far afield. To increase the probability of the view for
non-traditional Christians and non-Christians, I will briefly describe in
the next section what I think is a good reason to believe four-dimension-
alism. First, however, here is a sketch of the final details of the view.
From the dying individual’s perspective, they just see a swirl before
them as though the world were melting. They are then drawn into the
vortex, things go black with flashes of light, then they begin to see a new
scene emerging. The next thing they know, they are in another place-
time, a “new world” to them. And what of those standing by? Would the
wormholes of science fiction, which represent the technological hopes
and aspirations of many scientists, be visible to anyone in the imme-
diate vicinity of the dying individual? What would they see?
And what of the body that is left behind? Is it a fake? The body left
behind is indeed a sort of cast off “shell,” but it is really one’s body. But the
body one takes with one is also one’s body. There is really no other way it
could be. Heaven is a physical place. “We believe in the resurrection of the
body and the life in the world to come.” The life in the world to come is
a life with a resurrected body.21 The very matter constituting us is trans-
formed into a new heavenly body, but the body that is buried is just our last
connection with Earth’s worldline. The idea of a dead body coming back to
life is resuscitation, like what happened to Lazarus. Resurrection is radical
transformation. This may constitute a revision of the traditional Christian
view. If so, it might be a mere “development” and not be truly revisionary.
But of course, any Christian dualist will not need to avail themselves of
this option. Christian materialists have presumably already made their
peace with departures from the mainline of Christian tradition.
21
Throughout the middle ages, Paul’s mention of the “third Heaven” (2
Corinthians 12:2) is taken to refer to the “empyrean” Heaven which is beyond
even the firmament but is a physical place and the at least temporary repose of
those who have left Earth. For St. Thomas’s view see Summa Theologica I Q 66, a.
3, Q 102, a. 2; Supplement, Q 69, a. 1.
One clearly needs to carry a body into the afterlife on the mere Christian
view, but one also needs to leave one behind. How weird would it be if
people simply ... disappeared? We need some token that the person has
left this world entirely, not just wandered off. It might seem odd and even
a bit misleading, but the costs of not leaving a token behind are greater
than the costs of doing so. Compare here Jesus’s choice of mode of exit
from this world after his resurrection, the Ascension. It is a bit weird
and potentially misleading for Jesus to ascend into the clouds when his
mission on Earth was complete. However, consider his options: Go up,
go down, go laterally, just disappear. Any other direction but up has
worse implications, so the choice is between simply not showing up for
dinner one night or ascending. If he had simply disappeared, imagine
the cults we would have founded upon people saying they were Jesus
come back out of hiding! While the method he chose might have been
unusual, it appears to be the best available option. Likewise, I think for
leaving behind a bodily token in this world. Not to ascend would invite
all kinds of false impressions, at least, that is, until such time as fission
and wormholes would become common topics of discussion.
9.2.4 Split brains: the terrible tale of twins Ted and Todd
One of the chief virtues of a theory (ultimately the chief virtue of any
non-revealed theory) is that its being true would make sense out of an
otherwise very puzzling situation. I will present a puzzling situation for
which four-dimensionalism provides an elegant and satisfying solution.
First, however, I must cause trouble.
Consider the following picture of me.
Cute, I know. Now, that’s me. But what makes it me? Perhaps the most
obvious answer is that I remember being him, and if I was ever him, I am
always him. Another obvious answer has to do with the fact that that
body grew into my body. It would be really nice if these obvious answers
bore out, but we’ll examine them and see.
What I remember is looking in the mirror and seeing that freckled face.
I remember getting my fancy Pooh shirt out of the drawer. I remember
getting my picture taken in it. In short, there is a clear psychological
continuity with that little dude. Unfortunately, as is all too often the
case in metaphysics, the obvious answer just won’t work. For if you were
to implant all my memories, beliefs, hopes, desires, and all other mental
states into another individual, that individual would not be me. This
would be so even if at the exact moment you created this psychological
duplicate I got total and irreversible amnesia.
So the psychological continuity account of persistence fails. It’s a bit
surprising because it seems to be the most promising at first. Why is
that? Well, while psychological continuity is neither logically necessary
nor logically sufficient for persistence, in ordinary cases, it is a constant
concomitant. The cases used to show that the psychological continuity
account of persistence is false only need to be logically possible to do
their job, since a definition needs to hold in every possible circumstance.
But some logical possibilities are actually very remote, very strange, and
extraordinary. In the ordinary case, the obvious sign of persistence is
psychological continuity.
Another obvious answer has to do with the fact that that kid grew up to
be me. That is, there is a clear and natural path through spacetime from
that kid to me. There is a hunk of matter with fairly natural edges that
can be traced from then to now. It would be really nice if this obvious
answer fared better than the last. There is trouble for this view, but four-
dimensionalism will come to the rescue, saving obviousness. This is a
virtue of the theory and a reason to believe it is true.
Consider twins Ted and Todd. They are not identical twins, but they
are very much alike. One loves to drive boats in the sea, the other loves
to fly planes in the sky. One terrible day, by coincidence (or undetectable
malevolence), Ted is in a terrible boat crash and Todd is in a terrible plane
crash. They are both in a coma in beds in the same hospital room.
Now consider Rod. Rod’s fast car is struck by a train, and his body is
nearly destroyed. His brain is left in perfect condition with the exception
that his corpus callosum is severed, leaving his left and right cerebral
hemispheres out of touch. Some people are born with this condition,
however, and though they tend to be high on the autism spectrum, they
live happy lives. He lies awake but on life-support in the room across
the hall from Ted and Todd. The doctor monitoring all of these cases
notices, as the night goes on, that each individual is deteriorating in
certain ways. Ted and Todd are drawing closer to brain death, and Rod’s
body is growing ever more incapable of supporting his mental functions.
Finally, Ted and Todd are declared brain dead. Very shortly after that,
Rod’s body goes into total irreversible failure. The doctor thinks it’s going
to be his worst shift ever, but then he realizes there is a silver lining:
He can transplant Rod’s nearly perfectly functioning brain into one of
Ted or Todd’s bodies. He has done several successful hemispherectomies
wherein the patients led fairly normal lives (the science fiction has not
begun yet, hemispherectomies, though rare, do occur and are usually
successful). He sees how he can do the transplant. He’s ready to go. Then
he realizes that he will have the best chance of at least partial success if he
puts half of Rod’s brain in Ted’s body and half of it in Todd’s body. This is
what he decides to do, and he does it with aplomb. Both operations are a
complete success. Rod’s left hemisphere is successfully transplanted into
Ted’s body, and his right hemisphere is successfully transplanted into
Todd’s body. Congratulations, Doc! Now what did he do?
If presentism were true, this would be a dilemma indeed. The outcome
of each operation (which we’ll call Rod–Ted and Rod–Todd) has an equal
claim on being Rod. But there is absolutely nothing connecting Rod–Ted
and Rod–Todd on presentism. They are discrete organisms. They do each
bear an important relation of psychological continuity to Rod, but as
we’ve seen, that is not enough to establish personal identity. But if four-
dimensionalism is true, the reality is strange, but there is no problem
tracking Rod. Rod simply branches out, like a tree trunk that splits in
two. Maybe his branches will merge again sometime in the future, when
both hemispheres come into a single body. When we see Rod as spread
out over both space and time, a four-dimensional being, we see that
the appearance at present that Rod–Ted and Rod–Todd are disconnected
is merely a matter of perspective. We can’t see in enough dimensions.
We are like a two-dimensional being scanning a human from top to
bottom as they pass through our two-dimensional perceptual system.
First they appear as a dot, then it grows to an oblong circle and their
head passes through our perceptual plane. Then the circle shrinks at the
neck then branches out into a rough ellipse. Then things get interesting.
The ellipse undergoes a fission and separates into two shrinking circles.
For the two-dimensional being, there is no sign at all that the things
he perceives as two are in reality two parts of one thing, connected in a
dimension he cannot perceive.
We suffer a similar perceptual poverty with respect to Rod’s true

extended spacetime self. If we could see Rod in all his four-dimensional
glory, the “spacetime worm” stretching from his conception, winding
through the various stages of his life, we would see that this worm has
a fissure in it, at least for a while, and perhaps it splits permanently
into two “strands.” If we saw this, we would have no problem identi-
fying the things we named “Rod–Ted” and “Rod–Todd” as parts of Rod’s
four-dimensional spacetime worm. It’s a very uncommon shape for a
spacetime worm of a conscious being to take, but it is perfectly coherent,
and by far the most satisfying account of what happens in brain fission
cases.22 This is, of course, just a sample of the kind of arguments that
can be put forth on behalf of four-dimensionalism. It is an explanatorily
very powerful view (see Sider 2001 for the most detailed defense and see
Markosian 2004 for the other side).
9.2.5 Objections to the present view of resurrection

I have attempted to describe a picture of the resurrection with several
features: (i) it is consistent with materialism about conscious beings both
human and non-human; (ii) it is sufficiently probable on the conjunc-
tion of theism and materialism; (iii) it is relatively orthodox;23 (iv) we
have good reason to believe it is possible; and (v) the metaphysics it
presupposes is already needed for other problems. I believe I have given
effective short defenses that the view I have presented has these features.
I will now very briefly reply to three natural concerns. As with many
other areas in this book, there is much left unexplored which would
make worthy research projects for future scholars.
22
Strangely, many four-dimensionalists hold the unusual view (described
in Sider 2001: 153 and defended in Lewis 1976 and Noonan 2003) that what
happens in brain fission cases is that there are two people after the operation
(two distinct people, not just two distinct parts of one person) and that, mirabile
dictu, those two people had previously occupied the same body! It was only due
to the fortunate destruction of the body they shared that Rod–Ted and Rod–Todd
(which now name persons, not parts of persons) enjoy liberation in their own
bodies, according to this view. One wonders whether such coincident existence
is common or whether they were just extraordinarily lucky. I think my account
is far better.
23
The Catechism of the Catholic Church says that “Heaven” in the phrase in
the Lord’s prayer “who art in Heaven” “does not mean a place (‘space’)” (para-
graphs 2794 and 2802). However, the point here is that God in his inner nature
is not essentially tied to any one place so as to be distant from the Earth, he is
omnipresent; more related to this in the objections section.
Objection 1: Onlookers One natural question is this. I have described what

I surmise it would be like for someone to go through a wormhole. But
what about the other people in close proximity to those who die, perhaps
even holding their hands or clutching them? If the picture I described were
true, wouldn’t those people see that? Wouldn’t they in fact get “caught”
in the wormhole themselves? It is true that I presented traveling through
spacetime in its more science-fictiony form, but my theory works just fine
with more realistic wormholes (i.e. with the kind of wormholes that are
often thought actually to exist: very, very tiny ones). God could cause
there to be extremely small wormholes of just the right size to inhabit
the massive spaces between the particles that, on the present view, both
compose and constitute the conscious beings under consideration. Then
we could be transported to the New Jerusalem, a distant region of space-
time (“future” to us here now), particle by particle (that is, each particle
undergoes fission and one of each pair is transported. If the Heavenward
“gates” of the tiny wormholes are arranged in parallel to the this-worldly
gates, then the organism will appear on the other side in exactly the same
form as they were on the other side, as occurs when someone is transported
in Star Trek.24 Or by rearranging the “leeward” ends of the wormholes the
organism could appear in any posture (and mental state) God chooses.
Objection 2: Giving death its due? You might wonder whether, according
to the picture I’ve painted, anyone really dies. I myself have little
sympathy with this objection, but in the literature on materialist resur-
rection, it has been a topic of discussion.25 I don’t think Scripture,
Tradition, or Reason require much more of death than that we leave
this world behind. In particular, I don’t think it requires that death be
thought of as bad or as anything analogous to sleep. I expect the thought
of death as sleep is simply an aboriginal response to its physical simi-
larity to sleep combined with a hope that there is life after death. But
isn’t death an enemy in the Christian scheme of salvation? The Apostle
Paul, in his most extensive discussion of death, says “The last enemy to
be destroyed is death” (1 Corinthians 15:26).26 And since an enemy is
bad, death is bad.
Here is the context, however. Paul is addressing people in the Church at
Corinth who doubt that there will be a resurrection, perhaps influenced
by the Sadducees. Now, if there were no resurrection, as the Sadducees
24
Presumably, philosophy nerds have discussed at length what is going on there.
Though not a “Trekkie,” I humbly submit this possibility for their consideration.
25
See, for example, Hudson (2001) and Baker (2007).
26
Thanks to Nick Colgrove for pressing me on this.
thought (that’s why they were sad, you see?), then death would indeed
be the enemy. And before the advent of Christ, Paul might think, one
had wan hope of a resurrection. But since Jesus’s resurrection is “the
first fruits of those who have died,” fear of death, and the hypothetical
finality of death, is no more. They are vanquished. The enemy, then, is
not death itself, but death as it would be, were there no resurrection.
Note that mere cessation of organic activity is not intrinsically bad if
it doesn’t last. Creatures like us whose sleep was a state of suspended
animation would not suffer. Not even decay is bad if it is not permanent.
It depends on what kind of creature one is envisioning. I don’t see an
argument for decay being intrinsically bad that doesn’t have the conse-
quence that finitude is intrinsically bad. Each creature grows at a finite
rate, some faster than others.
One might worry that on my view, still, one doesn’t die. But what can
this mean? On any Christian view, the person persists. What dies is their
body (thus the dilemma for the materialist Christian). And on the view I
have presented, the entire old body stays here and is no longer animated,
the new body – formed from the fission of the particles of the old body –
enters a spacetime with different laws governing those particles and can
be quite different in its nature and powers. As on any view of death, my
view has people leaving behind a dead body. I see no reason whatsoever
to think it necessary to postulate an interval of unconsciousness for the
individual who walks no more upon the earth. However, my view is
clearly consistent with this. As I mentioned above, God could transfer
the individual into the new heavenly spacetime in any form he wanted,
including in some kind of tomb or grave, or in some kind of deep sleep
or total latency, awaiting the arrival of everyone else. Then God could
call them all forth at once.27
Still, it seems that there is one thing missing. It seems that on the
standard Christian view, it is the very body that dies that is raised. After
all, Paul says “What is sown is perishable, what is raised is imperish-
able” (1 Cor. 15:42). Paul also says that to be absent from the body is
to be at home with the Lord (2 Cor. 5:8, paraphrase based on the verses
leading up to it). At least prima facie, that suggests a different picture,
27
This was pointed out to me by Allison Thornton. She also offered me the
general writing advice that anytime one’s view offers flexibility, that flexibility
might as well be put to use to satisfy people’s desired outcomes, even if one doesn’t
share them. Apart from being generally good advice, it is worth mentioning in
the context of this book, since so much of what I say leaves so much open, that it
is worth offering the explicit invitation for people to finish the story in whatever
way works for them.
which suggests not taking the former statement too literally. Though
of course the two statements are consistent with one another, they are
very hard to fit literally together in a materialist metaphysics. I leave
further biblical exegesis to professionals. But it seems we already knew
that it couldn’t be the exact same body that is raised, due to cannibalism
(2 Cor. 5:8) and general atomic dispersal. So whatever is raised it isn’t
the same body, and in some cases, perhaps no part of the same body.
However, if one wants such a view, I can give it to them. For, as per
above, God could take one’s new body – which divides off from the old
body and takes consciousness with it – and have it interred in a suitable
grave in a suspended state, awaiting being called forth from the grave.28
In fact, the individual’s body could be literally dead with the cessation
of all metabolic processes so that it was neither alive nor decaying. The
view I have presented might have more resources to accommodate theo-
ries of resurrection than any on offer. This recommends it as true.
Objection 3: Heaven in outer space? Perhaps some will find objection-
able the idea that Heaven is not just a physical place, but either a space-
time region contiguous with that which we occupy or a parallel universe
of the sort that most physicists talk about. The mere Christian should
have no problem with the idea of heaven being a physical place. That
is core creedal content. However, there is a burden on me to explain
the “new” in “new heaven and earth” as it relates to the afterlife. The
phrase is borrowed from the prophet Isaiah by Peter (2 Peter 3) and John
(Revelation 21). In Revelation 21:1, where most people learn the phrase,
we are told “the first heaven and the first earth had passed away” and
then the New Jerusalem comes down from Heaven (Rev. 21:9ff). Earlier
in Revelation (6:12ff), at the opening of the sixth seal, we learn of cata-
clysmic events: earthquakes, black sun, red moon, black sky, mountains
and islands being tossed about. Peter offers more detail: “the heavens
will pass away with a loud noise, and the elements will be dissolved with
fire” (3:10) and “the heavens will be set ablaze and dissolved, and the
elements will melt with fire” (3:12). He says, in fact, that all along “the
present heavens and earth have been reserved for fire” (6:7). For now,
though, “we wait for new heavens and a new earth” (3:13).
None of the events listed in the Christian Scriptures requires more
than the destruction of our solar system, leaving everything else in the
galaxy untouched. In fact, all the data can be covered by an asteroid
(or comet or whatever) smashing into Earth, while leaving the rest of
28
For example, you might think that in Jesus’s speech, recorded in John 5:25,
he is speaking of the general resurrection instead of his own resurrection.
the solar system relatively untouched. It is consistent with all we know

about the formation of our planet that after being destroyed by a celes-
tial object, another planet could coalesce and take orbit around the sun
in Earth’s place. If it encountered enough stray matter in its formative
stages, it could be big enough to house the New Jerusalem and all its
inhabitants (its biblical proportions (roughly, a cube with a base about
the surface area of the continental United States) make it plenty large
enough to be the primary home of all the people who have lived thus
far and plenty more besides).
So it is entirely consistent with all we know that the vast majority of
the universe will remain in existence and that the New Jerusalem – the
center of “Heaven” in the relevant sense: The abode of the blessed – will
be in roughly the same spot the Earth is now in, but on a larger scale.
That is, there is a continuous path from the region of spacetime we
occupy now to the region the blessed will occupy. But not only is this
consistent with both scientific and biblical data, there is some reason to
think it is so.
Aquinas notes that there is some goodness to all being, even the being
of bad things (the following is an analogy: Even Hitler’s cleverness can
be recognized to be a good trait gone bad, i.e. put to bad use). Because of
this, to annihilate anything would be to reduce the amount of goodness
in existence. This gives God a reason, for every existing thing, to keep
it in existence. Conceivably, this reason could be outweighed. However,
since it “costs” God nothing to keep things in existence, it requires an
act but no effort, only in very unusual circumstances will God have over-
riding reason to annihilate something. In fact, St. Thomas finds it neces-
sary to defend the thesis that God even can annihilate (ST I.I Q 104, a.
3). He concludes that in fact God can, but goes on to consider whether
anything is annihilated. Through various considerations pertaining to
God’s grace and goodness, he concludes “Wherefore we must conclude
by denying absolutely that anything at all will be annihilated” (ST I.I Q
104, a. 4).
So if the matter from this spacetime region is “dissolved” but not anni-
hilated, what happens to it? It seems that either it gets thrown on some
kind of “scrap heap” or it gets “repurposed.” Given all the biblical and
theological support for the redemption of all nature in Christ, we ought
to expect it to be repurposed and fashioned into an incorruptible version
of what it always, in a way, aimed at being. This leads us to expect much
more continuity between the present spacetime and the spacetime of
the New Jerusalem. If the matter in question were truly dissolved into,
say, “quantum foam” – a plasma (or energy-charged particulate gas) of
disentangled quarks (elementary particles that compose, for example,

protons and neutrons) and gluons (fundamental particles that ordinarily
bind quarks together) – then they could be reconstituted to form almost
anything (and of course God can change the dispositions of these parti-
cles within limits that do not change their nature). So I not only see no
reason to think God will not form the new heaven (some level of the
firmament ranging from solar system to galaxy) and earth from the old
heaven and earth, but I think it is entailed or nearly entailed by theism.
Thus, there is little or no probabilistic cost for the materialist theist to
posit the metaphysical preconditions for and accept the metaphysical
implications of my wormhole account of the resurrection of conscious
beings within a materialist framework.
Concluding Summary
For some, this book will have been a long, strange trip. A central, over-
arching theme has been that so-called “bare theism” isn’t so bare. That is,
that a being who is omnipotent and omniscient will also be many other
things. That is because the properties of omnipotence and omniscience
entail a number of other properties, such as moral goodness and perfect
rationality. And given a fixed realm of value, which I have assumed,
these properties eliminate vast swaths of possibilia. More possibilities
about the future are eliminated when we learn facts about what has
happened in the past. There is no possible theistic world with a past like
ours the future of which lacks Heaven. And what Heaven must be like is
also constrained by what we know about this world.
Another overarching theme is that of probability or incremental
support (“confirmation” roughly as Carnap thought of it). The issue
here is that talk of what we “know” or don’t know is too coarse-grained
to be of much use in complicated issues. There is a bit of an epistemolog-
ical fad right now marching under the vague banner “Knowledge First.”
However, whatever knowledge might be valuable for (I can’t think of
much), knowledge is not the guide of life. Such a guide, as Bishop Butler
pointed out, is probability. If the conclusion of some piece of theodicy is
You can’t know that theism is false then that theodicy is relatively boring.
For that conclusion is consistent with having justified belief that theism
is false, its being an even toss whether theism is false, one’s being right-
fully inclined to regard theism as false, and a host of other attitudes that
would make the practice of religion quite difficult and even, perhaps,
nonsensical. A minimal condition of epistemic adequacy for a theodicy
is that it addresses the question: Just how bad is it for theism, that there
is this suffering? I address that question in this book. And I do so on
a “sliding scale,” realizing that readers will include firmly believing
179
Christians, Christians subject to considerable doubt, the religious non-

Christian, the half-convinced, the agnostic, and the atheist. I am also
well aware that some parts of the book will be more appealing than
others to certain readers. Thus I have tried to offer options and scalable
outcomes (e.g. “insofar as you are inclined to accept A, then that far
you should be inclined to accept B”), for a sort of choose-your-own-
adventure theodicy.
The problem of evil, as I have characterized it, is that the hypothesis
that there is a being who knows what is good, how good it is, is intrinsi-
cally motivated to bring about good states of affairs, and who has the
power to bring about any logically possible state of affairs is prima facie at
odds with the existence of bad states of affairs. Every bad state of affairs
requires explanation in terms of some good. The possibility of certain
goods is itself a good, as is the probability of good. The bad states of
affairs with which this book is concerned consists in the suffering of
sentient non-human animals. This suffering can be mental or physical.
Though I have, for the most part, focused on physical suffering, I think
my expanded notion of pain covers mental suffering as well. I defended
against two kinds of skepticism – philosophically grounded and scientifi-
cally grounded – that non-human animals do in fact experience morally
significant suffering. The defeat of these kinds of skepticism is sufficient
to defend the existence of a problem of animal pain, since it is both a
matter of common sense and of the consensus of concerned scientists
that a significant range of non-human animals do feel pain. I defeat this
skepticism in part by moving away from rationalist accounts of pain and
adopting an emotion-like theory of pain. I also gestured towards other
anti-rationalist accounts of pain that admit of direct knowledge of pain.
I build on work by Hick, Swinburne, Adams, and Hasker, and extend
Hick’s Irenaean theodicy from soul-making in two directions. First,
I extend it from exhibition of cardinal-like virtues (such as courage,
patience, generosity, perseverance) to exhibition of the theological
virtues of Faith, Hope, and Love, especially self-transcending love (which
I take to be manifested in, for example, sympathy, empathy, forgiveness,
humility, compassion). While this adopts material from a more fully
Christian theism rather than “bare” theism, there is no loss in prior
probability, for Love is the greatest good and an omniscient being would
know this and would be drawn to create a world in which the highest
forms of love were fostered.
The second way in which I extend the soul-making theodicy is by
adverting to a particular manifestation of self-transcending love mani-
fested most clearly in the Saints of the Christian tradition. Though it is
Concluding Summary 181
knowledge of the tradition that brings the relevant properties to mind, it

is upon reflection a priori accessible that this is a great good and, thus, to
be sought by an omniscient being. As Aquinas says, God shows us many
things we could have found out but which would have been occluded for
various reasons. But its being available a priori lets us put it in our back-
ground evidence. Tying together threads from Adams and Hasker with
this, I note that that saints have a disposition to want to be included in
God’s plan of salvation regardless of cost, and a disposition to see God’s
goodness to them and the goodness of their lives when they find out
that their suffering was a cost of their participation in God’s plan of
salvation. By gaining such a perspective on the suffering of their lives,
persons can come to see that God was not only just but good to them in
their life as a whole. In this glance, evil is defeated forever. Since every
possible theistic world containing suffering contains the redemption of
suffering, which we may call “Heaven,” there is absolutely no evidence
that anyone’s life is on the whole bad, nor could there be.
This transformation of the soul-making theodicy into a saint-making
theodicy makes it very hard to see how it could possibly apply to non-
human animals, especially if morally significant pain and suffering
extends as far down the phylogenetic scale as I suspect. For the higher
virtues are plausibly even harder to acquire than the cardinal virtues,
and, more to the point, the requirement that one gain a perspective on
one’s life and take the appropriate pro-attitude toward one’s life requires
a considerable degree of understanding. A hint is that human persons
are surely not capable of doing this sufficiently either. Their cognitive
resources will surely have to be augmented, and significantly so. And,
in fact, this is the traditional doctrine of theosis or deification. And if
this is so for human animals, why not then for non-human animals? I
suggest that their cognitive capacities will be enhanced to the point of
being able to see their evil defeated in the same way humans will. The
requirement that those who suffer must both themselves benefit from
the goods that impose the cost and be aware that this is so require that
non-human animals become persons. Later work will explore the ethical
implications for how to appropriately treat animals that have stages as
persons during stages they are not persons.
Animal afterlife, of course, presents certain problems. For it carries
with it the necessity of some kind of sufficient condition for identity over
time covering death. The traditional view in Western thought – both in
and outside Christianity, until the 17th century – is that animals have
souls. I defended this thesis philosophically so that it could be included
among the background beliefs of anyone who accepts dualism based on
certain problems associated with consciousness. For Christian readers,

I argued that due to considerations from Scripture and Tradition, they
ought to have animal souls in their background evidence. I considered
the fact that Scholastic philosophers admitted an animal soul but denied
animals’ immortality, rejecting their reasons for the differing treatment
of human and animal souls.
For those who steadfastly refuse to acknowledge animal souls, I offered
a model for resurrection without souls that avoids the presentist problem
of gappy existence. My model appeals only to four-dimensionalism and
the concept of wormholes in spacetime. I gave a brief defense of four-
dimensionalism by appeal to its ability to provide a satisfying solution
to a tough puzzle concerning split brains. Finally, I addressed various
objections to the thesis that one and the same animal can have human
and non-human stages.
A centerpiece of the book is the “Fine-tuning” argument for theism
from evil. According to this argument, when we consider the data from
evil to be the pattern of suffering thus far in history, and we pay attention
to a special feature of those data: They favor theism over naturalism by
a large factor. For the pattern of suffering in the world has in fact largely
been the cost of certain great goods (goods that are to be expected, given
theism). There is a finite band in the vast possible range of degrees of
hardship a universe might exemplify that will allow for or foster these
goods. Naturalism has virtually nothing to offer to narrow the range of
expected outcomes. Thus the data concerning evil, conceived as some
sort of fact about a pattern, favors theism over naturalism by at least an
order of magnitude. This is expressed by a decisive Bayes factor advan-
tage for theism. So, perhaps surprisingly at first, not only do the data of
evil not disconfirm theism at all, the data actually confirm it.
This is an interesting philosophical result, one of great practical signif-
icance. However, they say the greatest distance is the roughly 18 inches
from the brain to the heart. So it is no philosophical shortcoming that
a theodicy doesn’t heal the wounds that all sane people suffer from
beholding the suffering of animals (human and non-human). We often
continue to feel unloved or snubbed by a friend, for example, even when
we believe the evidence does not support that feeling. This puzzling
disconnect is itself a sad fact of life. Yet I hope that in some small way
the ideas in this book do help you bear the burden we collectively bear
in this world.
Bibliography
Adams, Marilyn McCord (1999). Horrendous Evils and the Goodness of God. Ithaca,
NY: Cornell University Press.
Adams, Marilyn McCord (2013). “Deification and Animal Pain: A Comment on
Trent Dougherty’s Extended Saint-Making Theodicy.” Presented at the American
Philosophical Association’s Pacific Division, March 27, 2013.
Adams, Robert M. (1979). “Primitive Thisness and Primitive Identity.” The Journal
of Philosophy 76 (1): 5–26.
Adams, Robert M. (1987). The Virtue of Faith and Other Essays in Philosophical
Theology. New York: Oxford University Press.
Allen, Colin (2004). “Animal Pain.” Noûs 38 (4): 617–643.
Allen, Colin, Fuchs, Perry N., Shriver, Adam, and Wilson, Hilary D. (2005).
“Deciphering Animal Pain.” In Murat Aydede (ed.), Pain: New Essays on Its
Nature and the Methodology of Its Study. Cambridge, MA: Bradford Book/MIT
Press.
American College of Animal Welfare (2010, March 5). “Animal Welfare Principles.”
Retrieved March 11, 2012, from <http://acaw.org/animal_welfare_principles.
html>.
American College of Laboratory Animal Medicine (2001). “Pain and Distress in
Laboratory Animals.” Retrieved March 11, 2012, from <http://www.aclam.org/
Content/files/files/Public/Active/position_painanddistress.pdf>.
Andrews, Kristin (Fall 2010). “Animal Cognition.” In Edward N. Zalta (ed.), The
Stanford Encyclopedia of Philosophy. Retrieved May 6, 2011, from <http://plato.
stanford.edu/archives/fall2010/entries/cognition-animal/>.
Armstrong, David M. (1968). A Materialist Theory of the Mind. London and New
York: Routledge.
Athanasius (2007). On the Incarnation, Cliff Lee (ed.). LaVergne, TN: Paradox
Media.
Audi, Robert (2008). “Belief, Faith, and Acceptance.” International Journal for
Philosophy of Religion 63 (1/3): 87–102.
Augustine (2006). Confessions, 2nd Ed. Indianapolis: Hackett Pub. Co.
Aydede, Murat (Summer 2009). “Pain.” In Edward N. Zalta (ed.), The Stanford
Encyclopedia of Philosophy. Retrieved May 6, 2011, from <http://plato.stanford.
edu/archives/sum2009/entries/pain/>.
Baker, Lynne Rudder (2007). “Persons and the Metaphysics of Resurrection.”
Religious Studies 43 (3): 333–348.
Bassham, Gregory (2005). “Some Dogs Go to Heaven: Lewis on Animal Salvation.”
In Bassham and Walls (eds.), The Chronicles of Narnia and Philosophy: The Lion,
The Witch, and the Worldview. Chicago, IL: Open Court.
Bergmann, Michael (2001). “Skeptical Theism and Rowe’s New Evidential
Argument from Evil.” Noûs 35 (2): 278–296.
Bergmann, Michael (2009). “Skeptical Theism and the Problem of Evil.” In
Thomas P. Flint and Michael Rea (eds.), The Oxford Handbook of Philosophical
Theology. Oxford: Oxford University Press, 374–399.
183
184 Bibliography
Bergmann, Michael (2012). “Commonsense Skeptical Theism.” In Kelly James

Clark and Michael Rea (eds.), Science, Religion, and Metaphysics: New Essays on
the Philosophy of Alvin Plantinga. Oxford: Oxford University Press, 9–30.
Betty, L. Stafford (1992). “Making Sense of Animal Pain.” Faith and Philosophy 9
(1): 65–82.
Bishop, John (2007a). Believing by Faith: An Essay in the Epistemology and Ethics of
Religious Belief. Oxford: Oxford University Press.
Bishop, John (2007b). “How a Modest Fideism May Constrain Theistic
Commitments: Exploring an Alternative to Classical Theism.” Philosophia 35
(3–4): 387–402.
Braun, David (2004). “Consciousness and Cognition.” Philosophy and
Phenomenological Research 69 (2): 484–491.
Brennan, Robert Edward (1941). Thomistic Psychology: A Philosophical Analysis of
the Nature of Man. New York: Macmillan.
Brown, Warran, Nancy Murphy, and H. Newton Malony (eds.) (1998). Whatever
Happened to the Soul? Scientific and Theological Portraits of Human Nature.
Minneapolis: Fortress Press.
Buchak, Lara (2014). “Learning Not to Be Naïve: A Comment on the Exchange
Between Perrine/Wykstra & Draper.” In Trent Dougherty and Justin McBrayer
(eds.), Skeptical Theism: New Essays. Oxford: Oxford University Press.
Byrne, R.W. and Whiten, Andrew (1988). Machiavellian Intelligence: Social Expertise
and the Evolution of Intellect in Monkeys, Apes, and Humans. Oxford: Oxford
University Press.
Carruthers, Peter (1989). “Brute Experience.” Journal of Philosophy 86 (May):
258–269.
Carruthers, Peter (1992a). The Animals Issue: Moral Theory in Practice. New York:
Cambridge University Press.
Carruthers, Peter (1992b). Human Knowledge and Human Nature: A New Introduction
to an Ancient Debate. Oxford: Oxford University Press.
Carruthers, Peter (1996). Language, Thought, and Consciousness: An Essay in
Philosophical Psychology. New York: Cambridge University Press.
Carruthers, Peter (2011). “Higher-Order Theories of Consciousness.” In Edward
N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Retrieved May 6, 2011,
from <http://plato.stanford.edu/entries/consciousness-higher/>.
Chalmers, David J. (1996). The Conscious Mind. Oxford: Oxford University Press.
Chalmers, David J. (2003). “The Content and Epistemology of Phenomenal
Belief.” In Quentin Smith and Aleksandar Jokic (eds.), Consciousness: New
Philosophical Perspectives. Oxford: Oxford University Press.
Chisholm, Roderick M. (1967). “Identity Through Possible Worlds: Some
Questions.” Noûs 1: 1–8.
Chisholm, Roderick M. (1977). Theory of Knowledge, 2nd Ed. Englewood Cliffs,
NJ: Prentice-Hall.
Chisholm, Roderick M. (1989). Theory of Knowledge, 3rd Ed. Englewood Cliffs, NJ:
Prentice-Hall.
Clark, Stephen R.L. (2013). “Animals.” In Charles Taliaferro, Victoria S. Harrison,
and Stewart Goetz (eds.), The Routledge Companion to Theism. New York:
Routledge.
Coakley, Sarah and Shelemay, Kay Kaufman (eds.) (2007). Pain and Its
Transformations: The Interface of Biology and Culture. Cambridge, MA: Harvard
University Press.
Bibliography 185
Coles, Robert (1999). “Endorsement of The Triumph of Wounded Souls: Seven

Holocaust Survivors’ Lives.” Retrieved December 26, 2012, from <http://undpress.
nd.edu/book/P00910>.
Colgrove, Nick (MS). “Gladness, Regret, and the Problem of Evil: A New Haskerian
Argument.”
Committee on Recognition and Alleviation of Pain in Laboratory Animals (2009).
Recognition and Alleviation of Pain in Laboratory Animals. Washington, D.C.:
National Academies Press.
Conee, Earl (1984). “A Defense of Pain.” Philosophical Studies 46 (September):
239–248.
Cooijmans, Paul (2012). “I.Q. and Real-Life Functioning.” Retrieved March 11,
2013, from <http://www.paulcooijmans.com/intelligence/iq_ranges.html>.
Coppel, William A. (2009). Number Theory: An Introduction to Mathematics. New
York: Springer.
Corcoran, Kevin, (ed.) (2001). Soul, Body, and Survival. Ithaca, NY: Cornell
University Press.
Cullison, Andrew (2007). “Privileged Access, Externalism, and Ways of Believing.”
Philosophical Studies 136 (3): 305–318.
Davies, Brian (2006). The Reality of God and the Problem of Evil. London and New
York: Continuum.
Davies, Brian (2011). Thomas Aquinas on God and Evil. Oxford: Oxford University
Press.
Dennett, Daniel C. (1978). “Why You Can’t Make a Computer That Feels Pain.”
Synthese 38 (July): 415–449.
Dennett, Daniel C. (1988). “Quining Qualia.” In Anthony J. Marcel and E. Bisiach
(eds.), Consciousness in Contemporary Science. Oxford: Oxford University Press.
Donceel, J.F. (1955). Philosophical Psychology. New York: Sheed and Ward.
Dougherty, Trent (2008). “Epistemological Considerations Concerning Skeptical
Theism.” Faith and Philosophy 25 (2): 172–176.
Dougherty, Trent (2010). “Evidence.” In Duncan Pritchard (ed.), Oxford
Bibliographies Online. Retrieved from <http://www.oxfordbibliographies.
com/view/document/obo-9780195396577/obo-9780195396577–0132.xml>.
Accessed Oct 27, 2012.
Dougherty, Trent (ed.) (2011a). Evidentialism and Its Discontents. Oxford: Oxford
University Press.
Dougherty, Trent (2011b). “Fallibilism.” In Sven Bernecker and Duncan Pritchard
(eds.), The Routledge Companion to Epistemology. New York: Routledge.
Dougherty, Trent (2011c). “Further Epistemological Considerations Concerning
Skeptical Theism.” Faith and Philosophy 28 (3): 332–340.
Dougherty, Trent (2011d). “Recent Work on the Problem of Evil.” Analysis
Supplement: 625–632.
Dougherty, Trent (2012). “Reconsidering the Parent Analogy: Unfinished
Business for Skeptical Theists.” International Journal for Philosophy of Religion 72
(1): 17–25.
Dougherty, Trent (2014), “Skeptical Theism”, The Stanford Encyclopedia of
Philosophy (Spring 2014 Edition), Edward N. Zalta (ed.), URL = <http://plato.
stanford.edu/archives/spr2014/entries/skeptical-theism/>
Dougherty, Trent and McBrayer, Justin (eds.) (2014). Skeptical Theism: New Essays.
Oxford: Oxford University Press.
186 Bibliography
Dougherty, Trent and Parker, Ross (2013). “Divine Hiddenness.” Routledge

Encyclopedia of Philosophy.
Dougherty, Trent and Poston, Ted (2007). “Divine Hiddenness and the Nature of
Belief.” Religious Studies 43 (2):183–198.
Dougherty, Trent and Poston, Ted (2008). “Hell, Vagueness, and Justice: A Reply
to Sider.” Faith and Philosophy 25 (3): 322–328.
Dougherty, Trent and Pruss, Alexander (2014). “Evil and the Problem of Anomaly.”
In Jon Kvanvig (ed.), Oxford Studies in Philosophy of Religion, Volume 5. Oxford:
Oxford University Press.
Dougherty, Trent and Walls, Jerry (2013). “Theistic and Anti-Theistic Arguments
from Evil.” In Charles Taliaferro, Victoria S. Harrison, and Stewart Goetz (eds.),
Routledge Companion to Theism. New York: Routledge.
Draper, Paul (1989). “Pain and Pleasure: An Evidential Problem for Theists.” Noûs
23 (3): 331–350.
Draper, Paul (1996). “Pain and Pleasure: An Evidential Problem for Theists.” In
Daniel Howard-Snyder (ed.), The Evidential Argument from Evil. Bloomington:
Indiana University Press.
Draper, Paul (2014). “Meet the New Skeptical Theism, Same as the Old Skeptical
Theism.” In Trent Dougherty and Justin McBrayer (eds.), Skeptical Theism: New
Essays. Oxford: Oxford University Press.
Draper, Paul and Dougherty, Trent (2013). “The Explanatory Argument from
Evil.” In Justin P. McBrayer and Daniel Howard-Snyder (eds.), The Blackwell
Companion to the Problem of Evil. Malden, MA: John Wiley & Sons.
Draper, Paul and Dougherty, Trent (forthcoming). God and Evil: Point/Counterpoint.
Dretske, Fred (1995). Naturalizing the Mind. Cambridge, MA: MIT Press.
Dummett, Michael A.E. (1993). Origins of Analytical Philosophy. Cambridge, MA:
Harvard University Press.
Evans, Gareth (1981). “Understanding Demonstratives.” In Herman Parret (ed.),
Meaning and Understanding. Oxford: Clarendon Press.
Fackenheim, Emil L. (1973). “The Human Condition After Auschwitz.” In
Jacob Neusner (ed.), Understanding Jewish Theology: Classical Issues and Modern
Perspectives. New York: Ktav Pub. House.
Fernandez, Ephrem and Turk, Dennis C. (1992). “Sensory and Affective
Components of Pain: Separation and Synthesis.” Psychological Bulletin 112 (2):
205–217.
Foley, Richard (2007). Intellectual Trust in Oneself and Others. Cambridge and New
York: Cambridge University Press.
Frankl, Viktor (1975). The Unconscious God: Psychotherapy and Theology. New York:
Simon and Schuster.
Frankl, Viktor (1985). Man’s Search for Meaning. New York: Washington Square
Press/Pocket Books.
Fumerton, Richard (1996). Metaepistemology and Skepticism. Boston: Rowman and
Littlefield.
Fumerton, Richard (2005). “The Challenge of Refuting Skepticism.” In Matthias
Steup and Ernest Sosa (eds.), Contemporary Debates in Epistemology. Oxford:
Blackwell, 85–97.
Fumerton, Richard (Fall 2008). “Knowledge by Acquaintance vs. Description.”
In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Retrieved
Bibliography 187
from <http://plato.stanford.edu/archives/fall2008/entries/knowledge-
acquaindescrip/>.
Galavotti, Maria (2005). Philosophical Introduction to Probability. Stanford, CA:
CSLI Publications.
Gallup, Gordon G., Jr. (1970). “Chimpanzees: Self Recognition.” Science 167
(3914): 86–87.
Gallup, Gordon G., Jr. (1985). “Do Minds Exist in Species Other Than Our Own?”
Neuroscience & Biobehavioral Reviews 9: 631–641.
Gallup (1995). – See ch. 5 p. 15.
Gertler, Brie (2001). “Introspecting Phenomenal States.” Philosophy and
Phenomenological Research 63 (2): 305–328.
Gillies, Donald (2000). Philosophical Theories of Probability. London and New York:
Routledge.
Giubilini, Alberto and Minerva, Franscesca (2011). See ch. 5 p. 15
Giubilini, Alberto and Minerva, Franscesca (2012). “After-Birth Abortion: Why
Should the Baby Live?” Journal of Medical Ethics. Published online February
23, 2012, doi: 10.1136/medethics-2011–100411. Retrieved from <http://jme.
bmj.com/content/early/2012/03/01/medethics-2011–100411.full.pdf+html>.
Accessed March 1, 2012.
Goetz, Stewart (2009). “The Argument from Evil.” In William Craig Lane and
J.P. Moreland (eds.), The Blackwell Companion to Natural Theology. Malden, MA:
Wiley-Blackwell, 449–497.
Gustafson, Don (2005. “Categorizing Pain.” In Murat Aydede (ed.), Pain: New
Essays on Its Nature and the Methodology of Its Study. Cambridge, MA: Bradford
Book/MIT Press.
Gustafson, Don (2006). – See ch. 5 pp. 3, 5.
Hájek, Alan (2003a). “What Conditional Probability Could Not Be.” Synthese 137
(3): 273–323.
Hájek, Alan (2003b). “Conditional Probability Is the Very Guide of Life.” In Henry
E. Kyburg, Jr. and Mariam Thalos (eds.), Probability Is the Very Guide of Life: The
Philosophical Uses of Chance. Chicago: Open Court.
Hasker, William (1974). “The Souls of Beasts and Men.” Religious Studies 10 (3):
265–277.
Hasker, William (1981). “On Regretting the Evils of This World.” Southern Journal
of Philosophy 19 (4): 425–437.
Hasker, William (1992). “The Necessity of Gratuitous Evil.” Faith and Philosophy
9 (1): 23–44.
Hasker, William (2004). Providence, Evil, and the Openness of God. London and New
York: Routledge.
Hasker, William (2008). The Triumph of God over Evil: Theodicy for a World of
Suffering. Downers Grove, IL: IVP Academic.
Hasker, William (2011). “Materialism and the Resurrection: Are the Prospects
Improving?” European Journal for Philosophy of Religion 3: 83–103.
Helm, Bennett W. (2002). “Felt Evaluations: A Theory of Pleasure and Pain.”
American Philosophical Quarterly 39 (1): 13–30.
Hick, John (1966). Evil and the God of Love. New York: Palgrave Macmillan.
Hick, John (1973). Philosophy of Religion. Englewood Cliffs, NJ: Prentice-Hall.
Hick, John (1997). “Transcendence and Truth.” In D.Z. Phillips and Timothy
Tessin (eds.), Religion without Transcendence? London: Macmillian.
188 Bibliography
Hick, John (2007). Evil and the God of Love, Rev. Ed. Basingstoke: Palgrave
Macmillan.
Hick, John (2010). “God and Christianity According to Swinburne.” European
Journal for Philosophy of Religion 2 (1): 25–37.
Howard-Snyder, Daniel (1996a). “The Argument from Divine Hiddenness.”
Canadian Journal of Philosophy 26 (3): 433–453.
Howard-Snyder, Daniel (ed.) (1996b). The Evidential Argument from Evil.
Bloomington: Indiana University Press.
Howard-Snyder, Daniel (1999). “God, Evil, and Suffering.” In Michael Murray
(ed.), Reason for the Hope Within. Grand Rapids, MI: Eerdmans.
Howard-Snyder, Daniel and Moser, Paul (eds) (2001). The Hiddenness of God.
Hudson, Hud (2001). A Materialist Metaphysics of the Human Person. Ithaca, NY:
Cornell University Press.
Hume, David (1980). Dialogues Concerning Natural Religion, Richard Popkin (ed.).
Indianapolis, IN: Hackett Publishing Co.
Ignatius of Loyola (2009). “Suffering.” From White Lily of Trinity. Retrieved
from <http://whitelilyoftrinity.com/saints_quotes_suffering.html>. Accessed
February 19, 2011.
International Association for the Study of Pain (1986). “Pain Terms: A List with
Definitions and Notes on Pain.” Pain 3 (Suppl): 216–221.
International Association for the Study of Pain (2012, May 22). “IASP Taxonomy.”
Retrieved from <http://www.iasp-pain.org/Content/NavigationMenu/
GeneralResourceLinks/PainDefinitions/default.htm>. Accessed February 19,
2012.
Irenaeus of Lyons (2010). Saint Irenaeus of Lyons: Against Heresies, Alexander
Roberts, James Donaldson and Cleveland A. Coxe (eds.). S.L.: Ex Fontibus.
Jackson, Frank (1982). “Epiphenomenal Qualia.” The Philosophical Quarterly 32
(127): 127–136.
Jacobs, Mike (2001). Holocaust Survivor: Mike Jacobs’ Triumph over Tragedy: A
Memoir, Ginger Jacobs (ed.). Austin, TX: Eakin Press.
Jeffreys, Harold (1961). Theory of Probability. 3rd Ed. Oxford Classic Texts in the
Physical Sciences. Oxford: Oxford University Press.
Jones, Deborah M. (2009). The School of Compassion: A Roman Catholic Theology of
Animals. Leominster, England: Gracewing Ltd.
Kaplan, Mark (1996). Decision Theory as Philosophy. Cambridge and New York:
Kitcher, Philip (1989). “Some Puzzles About Species.” In Michael Ruse (ed.), What
the Philosophy of Biology Is: Essays Dedicated to David Hull. Dordrecht: Kluwer
Academic Publishers.
Klubertanz, George (1953). The Philosophy of Human Nature. New York: Appleton-
Century-Crofts.
Koerth-Baker, Maggie (2010, November 29). “Kids (and Animals) Who Fail Classic
Mirror Tests May Still Have Sense of Self.” Scientific American. Retrieved March
11, 2012, from <http://www.scientificamerican.com/article.cfm?id=kids-and-
animals-who-fail-classic-mirror>.
Kraay, Klaas J. (2008a). “Can God Choose a World at Random?” In Yujin Nagasawa
and Erik J. Wielenberg (eds.), New Waves in Philosophy of Religion. New York:
Palgrave Macmillan, 22–35.
Bibliography 189
Kraay, Klaas J. (2008b). “Creation, World-Actualization, and God’s Choice Among

Possible Worlds.” Philosophy Compass 3: 854–872.
Kripke, Saul A. (1980). Naming and Necessity. Cambridge, MA: Harvard University
Press.
Kvanvig, Jonathan (1993). The Problem of Hell. New York: Oxford University
Press.
L’Engle, Madeleine (2007). A Wrinkle in Time. New York: Square Fish.
Langtry, Bruce (2008). God, the Best, and Evil. Oxford: Oxford University Press.
Lauwers, Luc and Vallentyne, Peter (2004). “Infinite Utilitarianism: More Is
Always Better.” Economics and Philosophy 20 (2): 307–330.
Leitgeb, Hannes and Pettigrew, Richard (2010). “An Objective Justification of
Bayesianism II: The Consequences of Minimizing Inaccuracy.” Philosophy of
Science 77 (2): 236–272.
Lerner, Bernice (2004). The Triumph of Wounded Souls: Seven Holocaust Survivors’
Lives. Notre Dame, IN: University of Notre Dame Press.
Leslie, John (2001). Infinite Minds: A Philosophical Cosmology. Oxford: Oxford
University Press.
Lewis, C.S. (1994). The Magician’s Nephew. New York: HarperCollins.
Lewis, C.S. (2001a). The Great Divorce: A Dream. San Francisco: Harper.
Lewis, C.S. (2001b). The Weight of Glory and Other Addresses. San Francisco:
Harper.
Lewis, C.S. (2002). The Problem of Pain. In The Complete C.S. Lewis Signature
Classics. New York: HarperOne, 543–646.
Lewis, David (1976). “Survival and Identity.” In Amélie Rorty (ed.), The Identities
of Persons. Berkeley, CA: University of California Press.
Linzey, Andrew and Tom Regan (1990). Animals and Christianity: A Book of
Readings. Eugene, OR: Wipf & Stock.
Malcolm, Norman (1993). Wittgenstein: A Religious Point of View? Ithaca, NY:
Cornell University Press.
Markosian, Ned (2004). “A Defense of Presentism.” In Dean Zimmerman (ed.),
Oxford Studies in Metaphysics, Volume 1. Oxford: Oxford University Press.
McBrayer, Justin P. (2010). “Skeptical Theism.” Philosophy Compass 5 (7):
611–623.
McCann, Hugh (2012). Creation and the Sovereignty of God. Bloomington, IN:
Indiana University Press.
McGinn, Colin (1989). “Can We Solve the Mind-Body Problem?” Mind 98 (391):
349–366.
McGinn, Colin (1999). The Mysterious Flame. Cambridge, MA: MIT Press.
McGrew, Lydia (2014). “Jeffrey Conditioning, Rigidity, and The Defeasible Red
Jelly Bean.” Philosophical Studies 168 (2): 569–582.
McGrew, Timothy (2003). “Confirmation, Heuristics, and Explanatory Reasoning.”
British Journal for the Philosophy of Science 54 (4): 553–567.
Melzack, Ronald (1975). “The McGill Pain Questionnaire: Major Properties and
Scoring Methods.” Pain 1: 277–299.
Melzack, Ronald and Casey, K.L. (1968). “Sensory, Motivational and Central
Control Determinants of Chronic Pain: A New Conceptual Model.” In Dan R.
Kenshalo (ed.), The Skin Senses: Proceedings of the First International Symposium
on the Skin Senses, Held at the Florida State University in Tallahassee, Florida.
Springfield, IL: Charles C. Thomas Publisher, 423–439.
190 Bibliography
Melzack, Ronald and Wall, Patrick D. (1965). “Pain Mechanisms: A New Theory.”
Science 150 (3699): 971–979.
Morris, Thomas V. (1987). Anselmian Explorations: Essays in Philosophical Theology.
Notre Dame, IN: University of Notre Dame Press.
Morris, Thomas V. (1991). Our Idea of God: An Introduction to Philosophical Theology.
Downers Grove, IL: InterVarsity Press.
Murray, Michael J. (2008). Nature Red in Tooth and Claw: Theism and the Problem of
Animal Suffering. Oxford: Oxford University Press.
Murray, Michael and Ross, Glenn (2006). “Neo-Cartesianism and the Problem of
Animal Suffering.” Faith and Philosophy 23 (2): 169–190.
National Research Council (1996). Guide for the Care and Use of Laboratory Animals.
Washington, D.C.: National Academy Press.
National Research Council (2008). Recognition and Alleviation of Distress in
Laboratory Animals. Washington, D.C.: National Academies Press.
National Research Council (2009). Recognition and Alleviation of Pain in Laboratory
Animals. Washington, D.C.: National Academies Press.
Nelkin, Norton (1986). “Pains and Pain Sensations.” Journal of Philosophy 83
(March): 129–148.
Nelkin, Norton (1994). “Reconsidering Pain.” Philosophical Psychology 7 (3):
325–343.
Nelkin, Norton (forthcoming). “The Belief in Other Minds.” In D.F. Gottlieb
and S.J. Odell (eds.), Wittgenstein and the Cognitive Sciences. Ithaca, NY: Cornell
University Press.
Noonan, Harold (2003). Personal Identity, 2nd Ed. London: Routledge.
Norcross, Alastair (2012). “The Significance of Death for Animals.” In Ben Bradley,
Fred Feldman, and Jens Johansson (eds.), The Oxford Handbook of Philosophy of
Death. Oxford: Oxford University Press, 465–474.
O’Connor, David (1998). God and Inscrutable Evil: In Defense of Theism and Atheism.
Lanham, MD: Rowman & Littlefield.
O’Connor, Timothy (2008). Theism and Ultimate Explanation: The Necessary Shape
of Contingency. Malden, MA: Blackwell Pub.
Oxford Centre for Animal Ethics (2013). “Director Appointed First Professor of
Animal Theology.” Retrieved December 3, 2013, from <http://www.oxfordan-
imalethics.com/2013/12/director-appointed-first-professor-of-animal-
theology/>.
Perry, John (2003). Knowledge, Possibility, and Consciousness. Cambridge, MA: MIT
Press.
Phillips, D.Z. (1993). Wittgenstein and Religion. New York: St. Martin’s Press.
Plantinga, Alvin (1974a). God, Freedom, and Evil. Grand Rapids, MI: Eerdmans.
Plantinga, Alvin (1974b). The Nature of Necessity. Oxford: Clarendon Press.
Plantinga, Alvin (1979). “The Probabilistic Argument from Evil.” Philosophical
Studies 35 (1): 1–53.
Plantinga, Alvin (1993). Warrant and Proper Function. New York: Oxford University
Press.
Plantinga, Alvin (2004). “Supralapsarianism or ‘O Felix Culpa.’ ” In Peter van
Inwagen (ed.), Christian Faith and the Problem of Evil. Grand Rapids, MI:
Eerdmans.
Bibliography 191
Plantinga, Alvin (2007). “Materialism and Christian Belief.” In Peter van Inwagen
and Dean Zimmerman (eds.), Persons: Human and Divine. Oxford: Oxford
University Press.
Plantinga, Alvin and Tooley, Michael (2008). Knowledge of God. Malden, MA:
Blackwell.
Pojman, Louis P. (1986). Religious Belief and the Will. London: Routledge & K.
Paul.
Polanyi, Michael (1967). The Tacit Dimension. London: Routledge & K. Paul.
Poston, Ted and Dougherty, Trent (2007). “Divine Hiddenness and the Nature of
Belief.” Religious Studies 43 (2): 183–198.
Povinelli, Daniel J. and Vonk, Jennifer (2004). “We Don’t Need a Microscope to
Explore the Chimpanzee’s Mind.” Mind and Language 19 (1): 1–28.
Raton Range, Staff Reporter (2010, July 8). “Scout Injured in Another Bear Attack
at Philmont.” Retrieved December 26, 2012, from <http://ratonrange.com/
scout-injured-in-another-bear-attack-at-philmont-p414–1.htm>.
Reid, Thomas (2002). Essays on the Intellectual Powers of Man: A Critical Edition.
Edinburgh: Edinburgh University Press.
Rescher, Nicholas (2000). “God’s Place in Philosophy (Non in Philosophia
Recurrere est Ad Deum).” Philosophy and Theology 12 (1): 95–105.
Roberts, Robert C. (1988). “What an Emotion Is: A Sketch.” Philosophical Review
97 (April): 183–209.
Roberts, Robert C. (2003). Emotions: An Essay in Aid of Moral Psychology. Cambridge,
UK and New York: Cambridge University Press.
Rogers, Katherin A. (2000). Perfect Being Theology. Edinburgh: Edinburgh
University Press.
Rollin, Bernard E. (2011). “Animal Pain: What It Is and Why It Matters.” Journal
of Ethics 15 (4): 425–437.
Rowe, William (1979). “The Problem of Evil and Some Varieties of Atheism.”
American Philosophical Quarterly 16 (4): 335–341.
Rowe, William (1996). “The Evidential Argument from Evil: A Second Look.” In
Daniel Howard-Snyder (ed.), The Evidential Argument from Evil. Bloomington,
IN: Indiana University Press.
Rowe, William (2001). “Skeptical Theism: A Response to Bergmann.” Noûs 35 (2):
297–303.
Rowe, William (2004). Can God Be Free? Oxford: Oxford University Press.
Ruzich, Joseph and Sadovi, Carlos (2012, May 21). “Lockport Scout Dies
After Falling in Ravine During Camping Trip.” Chicago Tribune. Retrieved
December 26, 2012, from <http://articles.chicagotribune.com/2012–05–21/
news/chi-lockport-boy-scout-dies-after-falling-in-ravine-during-camping-trip-
20120521_1_boy-scout-troop-scout-leader-cub-scout>.
Schumer, Peter D. (2004). Mathematical Journeys. Hoboken, NJ: Wiley-
Interscience.
Seacord, Beth (2012). “Higher-Order Theories of Consciousness and Animal
Experience.” Presented at the Center for Philosophy of Religion, University of
Notre Dame.
Shields, Christopher (2007). Aristotle. London and New York: Routledge.
Sider, Theodore (2001). Four Dimensionalism: An Ontology of Persistence and Time.
192 Bibliography
Sober, Elliott (2008). Evidence and Evolution: The Logic Behind the Science. Cambridge,
NY: Cambridge University Press.
Solomon, Robert C. (2007). True to Our Feelings: What Our Emotions Are Really
Telling Us. Oxford: Oxford University Press.
Stanley, Jason (2011). Know How. Oxford: Oxford University Press.
Stump, Eleonore (1983). “Knowledge, Freedom and the Problem of Evil.”
International Journal for Philosophy of Religion 14 (1): 49–58.
Stump, Eleonore (1996). “Aquinas on the Sufferings of Job.” In Daniel Howard-
Snyder (ed.), The Evidential Argument from Evil. Bloomington, IN: Indiana
University Press.
Stump, Eleonore (2003). Aquinas. New York: Routledge.
Stump, Eleonore (2010). Wandering in Darkness: Narrative and the Problem of
Suffering. Oxford: Oxford University Press.
Stump, Eleonore (2012). “Providence and the Problem of Evil.” In Brian Davies
and Eleonore Stump (eds.), The Oxford Handbook of Aquinas. Oxford: Oxford
University Press, 401–417.
Swinburne, Richard (1993). The Coherence of Theism. Oxford: Oxford University
Press.
Swinburne, Richard (1994). The Christian God. Oxford: Oxford University Press.
Swinburne, Richard (1997a (Orig. 1986)). The Evolution of the Soul. Oxford: Oxford
University Press.
Swinburne, Richard (1997b). Simplicity as Evidence of Truth. Milwaukee, WI:
Marquette University Press.
Swinburne, Richard (1998). Providence and the Problem of Evil. Oxford: Oxford
University Press.
Swinburne, Richard (2001). Epistemic Justification. Oxford: Oxford University
Press.
Swinburne, Richard (2003). The Resurrection of God Incarnate. Oxford: Oxford
University Press.
Swinburne, Richard (2004). The Existence of God. Oxford: Oxford University
Press.
Swinburne, Richard (2010a). “God as the Simplest Explanation of the Universe.”
European Journal for Philosophy of Religion 2: 1–24.
Swinburne, Richard (2010b). “Was Jesus God?” Religious Studies 46 (2): 265–269.
Taliaferro, Charles (1994). Consciousness and the Mind of God. Cambridge:
Tooley, Michael (1991). “The Argument from Evil.” Philosophical Perspectives 5:
89–134.
Topoi 14.2 (September 1995). Dordrecht: Kluwer Academic Publishers, 83–160.
Trakakis, Nick (2007). The God Beyond Belief: In Defence of William Rowe’s Evidential
Argument from Evil. Dordrecht, Netherlands: Springer.
Tversky, Amos, and Koehler, Derek J. (1994). “Support Theory: A Nonextentional
Representation of Subjective Probability.” Psychological Review 101: 547–567.
van den Bos, R., Houx, B.B., and Spruijt, B.M. (2002). “Cognition and Emotion
in Concert in Human and Non-Human Animals.” In Bekoff, M., Allen, C., and
Burghardt, G. (eds.), The Cognitive Animal: Empirical and Theoretical Perspectives
on Animal Cognition. Cambridge, MA: MIT Press.
Bibliography 193
Van Gulick, Robert (20012). “Consciousness.” In Edward N. Zalta (ed.), The

Stanford Encyclopedia of Philosophy. Retrieved May 6, 2011, from <http://plato.
stanford.edu/archives/win2012/entries/consciousness/>.
van Inwagen, Peter (1995). God, Knowledge and Mystery: Essays in Philosophical
Theology. Ithaca, NY: Cornell University Press.
van Inwagen, Peter (1996a). “The Problem of Evil, the Problem of Air, and the
Problem of Silence.” In Peter van Inwagen (ed.), God, Knowledge and Mystery:
Essays in Philosophical Theology. Ithaca, NY: Cornell University Press, 66–95.
van Inwagen, Peter (1996b). “The Magnitude, Duration, and Distribution of Evil:
A Theodicy.” In Peter van Inwagen (ed.), God, Knowledge and Mystery: Essays in
Philosophical Theology. Ithaca, NY: Cornell University Press.
van Inwagen, Peter (1996c). “Reflections on the Chapters by Draper, Russell,
and Gale.” In Daniel Howard-Snyder (ed.), The Evidential Argument from Evil.
Bloomington, IN: Indiana University Press.
van Inwagen, Peter (2006). The Problem of Evil: The Gifford Lectures Delivered in the
University of St. Andrews in 2003. Oxford: Oxford University Press.
van Inwagen, Peter and Dean Zimmerman (2007). Persons: Human and Divine.
Varner, Gary (2012). Personhood, Ethics, and Animal Cognition: Situating Animals in
Hare’s Two Level Utilitarianism. New York: Oxford University Press.
Walls, Jerry (1992). Hell: The Logic of Damnation. Notre Dame, IN: University of
Notre Dame Press.
Walls, Jerry (2002). Heaven: The Logic of Eternal Joy. Oxford: Oxford University
Press.
WBTV (2012, September 23). “School Mourns After Teen Boy Scout Falls to Death
During Weekend Hike.” Retrieved December 26, 2012, from <http://www.
wistv.com/story/19616193/school-mourns-after-teen-boy-scout-falls-to-death-
during-weekend-hike>.
Webb, Stephen (1998). On God and Dogs. Oxford: Oxford University Press.
Weisberg, Jonathan (2009). “Commutativity or Holism? A Dilemma for
Conditionalizers.” British Journal for the Philosophy of Science 60 (4): 793–812.
Wesley, John (1872a). “The General Deliverence” Sermon 60. Retrieved from
<http://new.gbgm-umc.org/umhistory/wesley/sermons/60/>. Accessed
November 29, 2012.
Wesley, John (1872b). “What Is Man” Sermon 109. Retrieved from <http://www.
umcmission.org/Find-Resources/John-Wesley-Sermons/Sermon-109-What-
Is-Man>. Accessed November 29, 2012.
Wierenga, Edward (Winter 2013). “Omniscience.” In Edward N. Zalta (ed.), The
Stanford Encyclopedia of Philosophy. Retrieved from <http://plato.stanford.edu/
archives/win2013/entries/omniscience/>.
Wiesel, Elie (2004). Legends of Our Time. New York: Schocken Books.
Williams, Jimmy (2000). “God, Evil and the Holocaust.” Probe Ministries
International, Retrieved from <http://www.leaderu.com/orgs/probe/docs/evil-
holo.html#text10>. June 12, 2012.
Williamson, Timothy (2000). Knowledge and Its Limits. Oxford and New York:
Oxford University Press.
Wright, N.T. (1992). The Resurrection of the Son of God (Christian Origins and the
Question of God). Minneapolis, MN: Fortress Press.
194 Bibliography
Wykstra, Stephen J. (1984). “The Humean Obstacle to Evidential Arguments

from Suffering: On Avoiding the Evils of ‘Appearance.’ ” International Journal
for Philosophy of Religion 16: 73–93.
Yaffe, Gideon and Nichols, Ryan (2009, Winter). “Thomas Reid.” In Edward N.
Zalta (ed.), The Stanford Encyclopedia of Philosophy. Retrieved from <http://plato.
stanford.edu/archives/win2009/entries/reid/>.
Young, William Paul (2011). The Shack: Where Tragedy Confronts Eternity: A Novel.
Newbury Park, CA: Windblown Media.
Zagzebski, Linda (2012). Epistemic Authority: A Theory of Trust, Authority, and
Autonomy in Belief. Oxford: Oxford University Press.
Index
Adams, Marilyn McCord, 16n. 1, Coakley, Sarah, n29

17n. 3, 96, 97n. 1, 113, 114, 126, Coles, Robert, 128
132, 134, 135, 135n. 1, 136, 137, Colgrove, Nick, n147, n168, n174
138, 146, 147, 148nn. 8–9, 149, Conee, Earl, 59
150n. 1, 151, 180 Cooijmans, Paul, 92
Adams, Robert, 114, n149, n167, 181 Coppel, William, n119
Allen, Colin, n63, n75 Cullison, Andrew, 84
Andrews, Kristin, n162
Anselm of Canterbury, 9 Davies, Brian, 17, 30
Aquinas, Thomas, n2, 19, 21, n48, Dennett, Daniel, 59, 60, 75
97, n116, n132, 159, 160, 161, 162, Descartes, René, 64, 65, 85
n169, 177, 181 Donceel, J. F., 160
Aristotle, 58, 148, 160, 162 Dougherty, Trent, 6, n6, n7, 11, 18,
Armstrong, David, 73 n19, 20, 21, 22, n23, n42, n55,
Arnauld, Antoine, 64 n68, n122, n127
Athanasius, 143 Draper, Paul, 11, 12, 13, 14, n19, n23,
Audi, Robert, 20 n41, n42, n44, 126
Augustine, 2, 97, 110, 142 Dretske, Fred, 59, n59
Aydede, Murat, 59, 60, 61, 80, 81, 82 Dummett, Michael, 89
Baker, Lynne Rudder, n174 Easwaren, Kenny, n15

Basil the Great, 158, 159 Einstein, Albert, 167, 168
Bassham, Gregory, 137 Evans, Gareth, 83
Bergmann, Michael, n22, n23, 32, n48
Betty, L. Stafford, n44 Fackenheim, Emil, n111
Bishop, John, 18, 44 Fernandez, Ephrem, 80
Braun, David, 84 Foley, Richard, n58
Brennan, Robert, n160, 161 Frankl, Viktor, 110, 111, n111, 128
Buchak, Lara, 13 Frege, Gottlob, 84
Buras, Todd, n47, n113 Fumerton, Richard, 85
Butler, Joseph, 144, 179
Byrne, R. W., 91 Galavotti, Maria, n6
Gallup, Gordon Jr., 86, 87, 88, 94, 95
Cantor, Georg, 119 Gertler, Brie, 85
Carnap, Rudolf, 179 Gillies, Donald, n6
Carruthers, Peter, 65, 73, 74, 91, 92, Giubilini, Alberto, 87
93, 94, 95 Goetz, Stewart, 67, n136
Casey, K. L., 82 Gustafson, Don, 78, 80, 82
Chalmers, David, 85, n165
Chisholm, Roderick, n57, 58, 68, 70, Hájek, Alan, n5
96, 111, 112, 113, 134, 135, 146, Hasker, William, 45, n45, 96, 99, 107,
149, n149 114, n115, 126, 134, n147, 180,
Clark, Stephen, 59, 65, 129 181
195
196 Index
Helm, Bennett, 75, 81, 82 Mackie, John, 45

Hick, John, 2, 96, 99, 103, 104, 105, Malcolm, Norman, n18
106, 107, 114, 126, 134, 135, 136, Malebranche, Nicolas, 64
137, 138, 139, 140, 141, 142, 147, Marcus Aurelius, 2
180 Markie, Peter, n119
Hooker, Richard, 154 Markosian, Ned, 173
Houx, B. B., 26 Maximus the Confessor, 3
Howard-Snyder, Daniel, n17, 22, n23, Melzack, Ronald, 76, 82
n45, 54 Minerva, Franscesca, 87
Hudson, Hud, n174 Moore, G. E., 113
Hume, David, n17, 50, 58 Morris, Thomas, 9
Moser, Paul, n23
Ignatius of Loyola, 108 Murphy, Mark, n113
Irenaeus of Lyons, 2, 3, 99, 102, 143 Murray, Michael, 4, n7, n44, n45,
n55, 65, n65, 67, 69, 74, 86, 88, 89,
Jackson, Frank, n163 90, n95, 137
Jacobs, Mike, 128
Jeffreys, Harold, n38, 126, n126, 129 Nelkin, Norton, 59, 75, 76, 79, 80,
Jones, Deborah, n158, 159, n159 n80, 81
Newlands, Sam, n64
Kaplan, Mark, 20 Newton, Isaac, 167
Kitcher, Philip, n24 Nichols, Ryan, 58
Klubertanz, George, n160 Noonan, Harold, n173
Koehler, Derek, n54 Norcross, Alastair, 59
Koerth-Baker, Maggie, 87
Kraay, Klaas, n40, n119 O’Connor, Timothy, 51, n130
Kripke, Saul, 60
Kvanvig, Jonathan, n129 Parker, Ross, n23
Perry, John, 165, n165
Langtry, Bruce, n17 Pettigrew, Richard, n15
Laplace, Pierre-Simon, 21 Phillips, D. Z., n18
Lauwers, Luc, n130 Plantinga, Alvin, 9, 20, n40, 51, n51,
Leibniz, Gottfried, 64, 65, n111 52, n99, n105, 107, n144, n162
Leitgeb, Hannes, n15 Pojman, Louis, 20
L’Engle, Madeleine, 167, 168 Polanyi, Michael, 79
Lerner, Bernice, 128 Poston, Ted, 20
Leslie, John, n41 Povinelli, Daniel, 90
Lewis, C. S., 8, n129, 135, 137, 138, Pruss, Alexander, n55
139, 140, 141, 143, 144, 145, 146,
150, 152 Regan, Tom, n4, n144, 146
Lewis, David, 41, n173 Reid, Thomas, 58, 64, 65, 89
Linzey, Andrew, n4, n144, 146 Rescher, Nicholas, n41
Locke, John, 64, 164 Roberts, Robert, n81
Rogers, Katherin, 9
McAllister, Blake, n46, n148 Rollin, Bernard, 26
McBrayer, Justin, 22 Rosen, Nathan, 168
McCann, Hugh, 140 Ross, Glenn, 95
McGinn, Colin, n165 Rowe, William, n17, 21, n21, 29, 45,
McGrew, Timothy, n15, n53 140
Index 197
Russell, Bertrand, 84 Tversky, Amos, n54
Schumer, Peter, n119 Vallentyne, Peter, n130

Scotus, John Duns, n48, 148 van denBos, R., 26
Seacord, Beth, n59 Van Gulick, Robert, n162
Shakespeare, William, 132 vanInwagen, Peter, n7, n22, n23,
Shelemay, Kay Kaufman, n29 n40, n68, 88, 107, n119, 126
Shields, Christopher, 58 Varner, Gary, n63, n75
Sider, Theodore, 173, n173 Vonk, Jennifer, 90
Singer, Peter, 87
Sober, Elliott, n24 Wall, Patrick, 82
Socrates, 148 Walls, Jerry, n3, 21, 22, n122, n127,
Solomon, Robert, n81 n129
Spruijt, B. M., 26 Ward, Keith, 146
Stanley, Jason, 83, 84 Webb, Stephen, n3
Stump, Eleonore, n2, n6, 29, 30, 33, Weisberg, Jonathan, n15
52, 53, n97, 98, n116, n121, 126, Wesley, John, 161, n161, 162
n132 Whiten, Andrew, 91
Swinburne, Richard, n8, n17, n18, Wierenga, Edward, n164
n19, 20, 22, 29, 30, 40, n40, n49, Wiesel, Elie, n111
n50, 52, 58, 96, n97, n99, n105, William of Ockham, 19
106, n116, 120, n121, n125, 126, Williams, Jimmy, n111
n126, n133, 134, 140, n149, n162, Williamson, Timothy, 85
180 Wright, N. T., n133
Wykstra, Stephen, n22
Taliaferro, Charles, n40
Thornton, Allison, n175 Yaffe, Gideon, 58
Tooley, Michael, 32, 34, 35 Young, William, n128, 155
Trakakis, Nick, n21
Turk, Dennis, 80 Zagzebski, Linda, n58

The Problem of Animal Pain: Trent Dougherty

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Problem of Animal Pain: Trent Dougherty

Uploaded by

Copyright:

Available Formats

PALGRAVE FRONTIERS IN PHILOSOPHY OF RELIGION

The problem of evil constitutes the greatest challenge to rational belief

The Problem of Animal Pain

Trent Dougherty is Assistant Professor of Philosophy at Baylor University,

Content Type: Black & White

Palgrave Frontiers in Philosophy of Religion

Transferred to Digital Printing in 2015

Series Editors’ Preface x

1 The Plan of This Book 1

2 The Problem of Animal Pain 16

3 The Bayesian Argument from Animal Pain 36

4 Is There Really a Problem?: The Challenge of

4.2.3 Some motivation for neo-Cartesianism 69

5 There Is a Problem: The Defeat of Neo-Cartesianism 77

6 The Saint-Making Theodicy I: Negative Phase 96

7 The Saint-Making Theodicy II: Positive Phase 118

8 Animal Saints 134

9.1.3.2 Qualia: inverted and absent 164

Concluding Summary 179

The philosophy of religion has experienced a welcome revitalization

I wish to express my gratitude to editors Yujin Nagasawa and Erik

Finally, I wish to thank my beloved wife Sarah, and my kids Fiona,

I taught this manuscript in a graduate seminar in philosophy of religion

Another father important to the East is the 6th/7th century theolo-

traits but, specifically, on the characters of a saint as conceived especially

Throughout the book, I will be using some probabilistic vocabulary just

must have some second-personal knowledge of me.7 This affects but

Positive Relevance E<D,H> = df Pr(H | D) > Pr(H)

When we have evidence in this sense, we have incremental or relative

1.3 Not-so-bare bare theism

from Eastern Christianity – to Western Christianity or a sparse or bare

alleged consequence of E – which itself merely states the empirical facts

Y Animals receive no compensation for their suffering; their evil is

Z1 The goods that allow permission of an animal’s suffering must

But this is logically equivalent to

And if T → ~Y, then T is logically equivalent to T & ~Y, so we have

(*3) E → (T ≡ (T & ~Y))

(T & S) v (T & ~S).

(CR1) Pr(E | T & S) > P(E)

(CR2) Pr(E | N) > Pr(E | T & S)

And so even though E doesn’t directly disconfirm T in this hypothetical

1.4 Two ways of learning what theism entails

balls turn out, as predicted, to be yellow, the defender of theory-12

At this point, your partition looks like this:

IN & at Townie: .25

is currently no consensus. What there is general consensus on, however,

2.1 Stating the problem

2.1.1 What is the problem?

The comparative simplicity of theism and naturalism is a conten-

2.1.2 Getting more precise

Given (the programmatic assumption) that theism and naturalism

Here is a schematic of the problem.

Assumption: The simplicity of bare theism and naturalism is roughly

Even though my statement of the problem leaves open several possible

Pointless suffering is logically incompatible with God’s existence.11

2.1.3 Theism as “hypothesis”

Theory = df a narrative which attempts to tell a coherent and compel-

Notice that this kind of theorizing is common to history, literature,

The Silmarillion). There is a lot of action, many characters of unfamiliar

2.2 How I shall understand “pain” and “suffering”

Let us look again at the datum.

It is unclear to me what American College of Animal Welfare mean by

1. “The welfare of an individual is its state as regards its attempts to cope