(Routledge Research in Applied Ethics 16.) Fischer, Bob - The Ethics of Eating Animals - Usually Bad, Sometimes Wrong, Often Permissible (2020, Routledge)

The Ethics of Eating Animals
“This is one of the most honest books I’ve ever read. Rather than grinding
an axe, Fischer follows the reasons to the conclusions they support—
conclusions at odds with what he had hoped to establish.”
—Donald Bruckner, Penn State University,
New Kensington, USA
Intensive animal agriculture wrongs many, many animals. Philosophers

have argued, on this basis, that most people in wealthy Western contexts
are morally obligated to avoid animal products. This book explains why
the author thinks that’s mistaken. He reaches this negative conclusion
by contending that the major arguments for veganism fail: they don’t
establish the right sort of connection between producing and eating
animal-based foods. Moreover, if they didn’t have this problem, then
they would have other ones: we wouldn’t be obliged to abstain from
all animal products, but to eat strange things instead—e.g., roadkill,
insects, and things left in dumpsters. On his view, although we have a
collective obligation not to farm animals, there is no specific diet that
most individuals ought to have. Nevertheless, he does think that some
people are obligated to be vegans, but that’s because they’ve joined a
movement, or formed a practical identity, that requires that sacrifice.
This book argues that there are good reasons to make such a move, albeit
not ones strong enough to show that everyone must do likewise.
Bob Fischer teaches philosophy at Texas State University. He’s the

author of Animal Ethics—A Contemporary Introduction (Routledge,
forthcoming) and the editor of The Routledge Handbook of Animal
Ethics (Routledge, forthcoming).
Routledge Research in Applied Ethics
The Injustice of Punishment

Bruce N. Waller
Friendship, Robots, and Social Media

False Friends and Second Selves
Alexis M. Elder
The Capability Approach in Practice

A New Ethics for Setting Development Agendas
Morten Fibieger Byskov
The Ethics of Counterterrorism

Isaac Taylor
Disability with Dignity

Justice, Human Rights and Equal Status
Linda Barclay
Media Ethics, Free Speech, and the Requirements of Democracy

Edited by Carl Fox and Joe Saunders
Ethics and Chronic Illness

Tom Walker
The Future of Work, Technology, and Basic Income

Edited by Michael Chobli and Michael Weber

Usually Bad, Sometimes Wrong, Often Permissible
Bob Fischer
For more information about this series, please visit: www.routledge.com/

Routledge-Research-in-Applied-Ethics/book-series/RRAES
Usually Bad, Sometimes Wrong,
Often Permissible
Bob Fischer
First published 2020
by Routledge
52 Vanderbilt Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an
informa business
© 2020 Taylor & Francis
The right of Bob Fischer to be identified as author of this work
has been asserted by him in accordance with sections 77 and 78
of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted
or reproduced or utilised in any form or by any electronic,
mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information
storage or retrieval system, without permission in writing from
the publishers.
Trademark notice: Product or corporate names may be
trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book has been requested
ISBN: 978-0-367-23004-3 (hbk)
ISBN: 978-0-429-27788-7 (ebk)
Typeset in Sabon
by Apex CoVantage, LLC
For Jennifer, who helped me see animals.
Contents
Acknowledgmentsx
1 Introduction 1
2 Contemporary Animal Agriculture 8
2.1 Low-Level Stress, Acute Pain 9
2.1.1 Cows 9
2.1.2 Pigs 11
2.1.3 Chickens 12
2.1.4 Incentives 13
2.1.5 Taking Stock 16
2.2 What Follows? 17
3 Bad Arguments for Eating Animals 20

3.1 Contractualism 21
3.2 Human Exceptionalism 24
3.3 Environmentalism and Agrarianism 27
3.4 Taste 31
3.5 Health 33
3.6 Ageism, Classism, Sexism, Racism 35
3.6.1 Racism Without Racists 37
3.6.2 Limited Access 40
3.6.3 The Neocolonial Economic System 44
3.7 Conclusion 46
4 Utilitarianism and the Causal Inefficacy Problem 50

4.1 The Simple Act Utilitarian Argument for
Veganism 50
4.1.1 Causal Inefficacy, a First Pass 53
4.1.2 The Order Threshold Solution 55
viii Contents
4.1.3 The Main Problem With the Order Threshold
Solution 57
4.1.4 An Objection 61
4.2 Causal Inefficacy and Welfare Thresholds 64
4.2.1 Welfare Thresholds: The Basic Idea 64
4.2.2 Welfare Thresholds: The Argument 66
4.2.3 Objections and Replies 68
4.3 Conclusion 71
5 Causal Inefficacy Aside, Utilitarianism Requires

Eating Unusually 75
5.1 Animals and Plant Agriculture 76
5.2 Roadkill and Freeganism 77
5.3 Bivalves 80
5.4 Insects 81
5.4.1 Are Insects Conscious? 81
5.4.2 The Utilitarian Argument for Entomophagy 84
5.5 Conclusion 86
6 The Rights View and the Production/Consumption Gap 89

6.1 Regan 89
6.2 Benefiting 92
6.3 McPherson 94
6.4 Driver 95
6.5 Hooley and Nobis 98
6.6 Korsgaard 99
6.7 Conclusion 101
7 Eating Animals the Rights Way 104

7.1 An Exception for Family Farms? 105
7.2 Backyard Chickens 108
7.3 Roadkill and Freeganism Revisited 110
7.4 Insects Revisited 112
7.5 Hunting 117
7.6 The Doctrine of Double Effect 123
7.7 Conclusion 126
8 Beyond Utilitarianism and the Rights View 128

8.1 Virtue Ethics 128
8.2 Ecofeminism 132
8.3 The Argument by Analogy 135
Contents ix
8.4 The Holocaust Argument 138
8.5 An Epistemic Argument? 140
8.5.1 A Stacked Deck 140
8.5.2 The Badness of Eating Animal Products 144
8.5.3 The Wrongness of Eating Animal
Products? 148
8.6 Eating Unusually? 150
9 Activist Ethics 158

9.1 Relationships, Identities, and Participation in
Movements 158
9.2 The Duty to Be Vegan 160
9.3 Reasons to Opt In 163
9.4 Conclusion 165
10 Taking Stock 167
References172
Index189
Acknowledgments
I started writing about animals and food ethics in 2012. At that time,
I had no idea that this topic would take over my professional life. Seven
years later, my debts are many.
To begin, publishers kindly gave me permission to rework the follow-
ing essays for this book: “Arguments for Consuming Animal Products”
in The Oxford Handbook of Food Ethics, edited by Anne Barnhill, Mark
Budolfson, and Tyler Doggett, pp. 241–266 (New York: Oxford Univer-
sity Press, 2018); “Bugging the Strict Vegan,” Journal of Agricultural &
Environmental Ethics 29.2 (2016): 255–263; and “Is Abolitionism Guilty
of Racism? A Reply to Cordeiro-Rodrigues,” Journal of Agricultural and
Environmental Ethics 31.3 (2018): 295–306.
My editor, Andrew Weckenmann, has been supportive from the begin-
ning, and the staff at Routledge has been wonderful. Alyse Spiehler, who
compiled the index, has been invaluable.
I owe a special thanks to the fine students in my Spring 2019 course
on food ethics (as well as Colleen Myles, a colleague who sat in on the
course) for reading the entire manuscript and offering lots of wonderfully
critical commentary.
I received a great feedback on a draft of this book at the annual food
ethics workshop at the University of Vermont.
Several years ago, Jimmy McWilliams agreed to teach a course with
me called “Eating Animals in America.” Since then, he’s been a constant
conversation partner, and many of the ideas in this book emerged during
our exchanges. Thanks for taking a chance on me.
When I write, I usually have a specific interlocutor in mind. Though
he never asked for this role, David Killoren is that person for me, at
least when I write about the ethics of eating. This book is my attempt to
convince you that veganism isn’t morally required, and by that standard,
I think it’s safe to say that the book is a complete failure. Guess I’ll have
to get another hobby.
Dan Demetriou and I wrote the paper on hunting that I summarize in
Chapter 7, and Josh Milburn and I wrote the paper on backyard chickens
that I summarize in the same place. It was great fun to work with them
Acknowledgments xi
both: they’re simply wonderful people with whom to write and discuss
philosophy. In addition, Josh was gracious enough to read and comment
on a complete draft of the manuscript, going far beyond the call of duty.
Thank you.
Andy Lamey also took the time to read and comment on a complete
draft of the manuscript, as well as send a number of helpful challenges
via email. Of the many people with whom I’ve discussed this material,
you’ve had to be the most patient. I’m grateful.
Finally, without Mark Budolfson’s encouragement, I wouldn’t have
written this book. And without Mark’s work, I couldn’t have. The best
ideas here are yours.
1 Introduction
This is not the book I set out to write. That book—which I began some
years ago, and eventually had to scrap—tried to show that whatever your
commitments in moral theory, you would be forced to conclude that
you shouldn’t eat products that trace back to factory farms. The basic
thought was that it was easy to establish that you should be vegan if you
assume utilitarianism, some sort of rights theory, or a standard form of
virtue theory. The challenge would be to show that the same conclusion
follows from approaches that are (typically thought to be) less friendly
to animals, such as contractualism. When I started, though, it seemed
plausible to me that it didn’t make much difference whether animals have
direct moral standing: even if we set animals aside, contemporary farm-
ing has so many problems—its environmental costs, the public health
costs, the harms to workers, and so on—that we could make the case
from them alone.
I was wrong. It’s indeed possible to argue that everyone has moral rea-
sons to abstain from animal products, even granting their starting points;
likewise, it’s probably possible to argue, given the same assumptions,
that everyone has reasons to be critical of intensive animal agriculture.
However, you aren’t going to get moral obligations. The moral reasons
just aren’t strong enough.
That much might be unsurprising. In retrospect, I suppose I should’ve
seen it. But it’s a bit harder to fault me for thinking that veganism is
required if act utilitarianism is true, or that anyone committed to animal
rights ought to be vegan. These are standard positions. It seems to me
now, however, that they’re mistaken. And that’s how I came to write the
book that’s now before you. My goal here is to explain how I came to
change my mind. In what follows, I argue that regardless of where you
start, it doesn’t look like most people are morally obligated to be vegan.
I contend that the moral reasons to which philosophers have appealed
are dramatically weaker than those philosophers have taken them to be.
(Nevertheless, I do think that animals are often wronged in intensive
systems and that some people are obligated to be vegan. More on these
points in Chapters 2 and 9.)
2 Introduction
When I talk about “veganism,” I’m talking about the strict abstinence
from the consumption of animal products. I think that most people have
something like this in mind when they talk about veganism, whether or
not they’re vegan themselves. Admittedly, though, “being vegan” can be
heard in at least two ways. On the one hand, you can hear it as a success
term: you only count as being vegan if you do, in fact, strictly abstain
from the consumption of animal products. (This is why vegans can—
playfully or unkindly—police one another by saying that someone no
longer counts as vegan because she unknowingly ate a granola bar with
some honey in it, or some bread with an egg wash.) On the other, you can
hear “being vegan” as a goal or identity: you count as being vegan if you
try to strictly abstain from the consumption of animal products. How-
ever, the identity isn’t totally flexible; it needs to come along with some
measure of success. Granted, you can be a “bad vegan” who occasionally
eats a chicken wing after drinking too much. But if you drink too much
most nights, and so eat chicken a few times a week, then your status as
a vegan is in serious doubt. For present purposes, though, I’m not going
to worry about these nuances. My view is that most people aren’t obli-
gated to strictly abstain from the consumption of animal products, and
so I don’t think that they’re obligated to try to strictly abstain.
This way of interpreting “veganism” isn’t universal. Some people in
the animal movement define veganism as the attempt to minimize your
role in harming animals, or your involvement in animal exploitation, or
something to that effect. They are, of course, welcome to their definitions,
but I take it that they are proposing revisions to the ordinary under-
standing because of their objectives as animal advocates—not because
they take such definitions to best represent common usage. For what it’s
worth, I don’t think that we’re obligated to do those things either: given
all the other causes in the world, and the minimal impact that any par-
ticular individual can make, I find it implausible that we have an obliga-
tion to optimize our behavior for animals. However, I won’t have much
to say about the general optimization question here. Instead, I’ll try to
answer questions like this: if you ought to minimize your role in harm to
animals, should you strictly abstain from consuming animal bodies and
byproducts? For my purposes, I take that question to be equivalent to:
if you ought to minimize your role in harm to animals, should you be
vegan, as that term is ordinarily understood? By answering these kinds
of questions, I hope to shed light on the central one, namely, are most
people morally obligated to be vegan?
As I’ve said, I don’t think so. And as I’ve said, this is perfectly com-
patible with saying that people have reasons to go vegan; even good
reasons. But we have good reasons to do lots of things: volunteer in
after school programs, donate a large percentage of their income to
charity, phone bank for the least-awful political candidate, serve meals
at the local soup kitchen, minimize our carbon footprint, and avoid
Introduction 3
objectionably-produced textiles. And presumably, some people are obli-
gated to do some of them. But few are obligated to do all, and crucially
for present purposes, there is no general obligation to do any of them.
That is, most people—even the relatively well off—aren’t morally obli-
gated to do any of the things I just listed. They certainly ought to do
something with their position of privilege; I’m not arguing that anything
goes. Still, I take it that most people have a fair amount of discretion in
their moral projects—a claim that’s practically a truism outside of certain
philosophical circles.
But you don’t have to agree with me about the discretion claim or the
truism claim. And I grant that it isn’t easy to establish the negative claim
I want to establish—there is no general obligation to be vegan (though
you may well have reasons to go vegan!)—especially if you concede that
we can have obligations of which we’re blamelessly ignorant. However,
we navigate this problem easily enough in other circumstances. When we
conclude, for instance, that it’s morally OK to break certain laws—such
as the ordinances in many cities that forbid feeding the homeless—we
just assess the various arguments against that conclusion. If none of them
works, and we can’t think of any better arguments for always following
the law, then it seems reasonable to believe that it’s sometimes permissi-
ble to engage in civil disobedience. This is a kind of “noseeum inference,”
as discussed by philosophers of religion: we don’t see a good argument;
so, probably, one ain’t there.
Admittedly, such inferences aren’t perfectly reliable. And this book
isn’t exhaustive, which is a further limitation: I don’t respond to every-
thing that anyone has ever said on behalf of a general obligation to be
vegan. Maybe the arguments I’ve omitted are the good ones, even though
I think I respond to the most promising options. Alternatively, it’s pos-
sible that we’re just on the wrong track: we could have good arguments
for veganism, but we’re working with second-rate moral frameworks; we
just need to develop better ones.
Let’s consider the “I’ve ignored the good arguments” challenge. I admit
that I worry about this. Still, it isn’t only possible to miss good argu-
ments: we can also overestimate the quality of arguments. Indeed, I’ve
come to think that this is a peculiar liability of doing applied ethics as
a professional philosopher.1 No one wants to write or publish an article
that claims that we have some moral reasons to eat fewer animals, or
give more money to charity, or have slightly more permissive immigra-
tion policies. Instead, people want to publish articles that insist we have
demanding obligations in these domains, ones that we’ve largely been
neglecting. A good paper defends the most ambitious conclusion that
you can possibly justify given the premises available to you. It shouldn’t
be surprising, then, that people present reasons as obligation-generating
even when the arguments for the conclusion aren’t cogent: the “go big or
go home” incentive in academia is real. And if you’re invested in there
4 Introduction
being a general obligation to be vegan, then it’s easy to be insufficiently
critical of these arguments.
Of course, the possibility of this error doesn’t show that we’re com-
mitting it; it’s also possible that we’re engaging in motivated reasoning
to avoid making uncomfortable changes to our lives. But it does serve as
a check against the thought that I’m somehow stacking the deck against
concluding that we ought to abstain from animal products. Moreover,
given that I’ve tried to make different sorts of arguments for veganism
(see, e.g., my 2014, 2016, 2018, and forthcoming), and given my own
dietary choices, I don’t think that I can fairly be accused of bias. The
upshot? Yes: I could be ignoring the good arguments for a general duty
to be vegan. But it’s hardly obvious that this is the case, and I don’t think
that the mere possibility is a serious challenge to the noseeum inference.
The other worry is that there is something wrong with our moral
frameworks. So, rather than concluding that it’s permissible to consume
animal products, we should conclude that we need better moral tools.
(Maybe the problem is that our various ethical frameworks are individu-
alistic or speciesist, or that they were built for a time when individual
action made much more of a difference than it does in a large consumer
society.) The first thing to say is that this move places the burden of
proof on the critic. If someone wants to propose a moral framework that
implies that we ought to be vegans, I’m happy to consider it. But until it’s
proposed, I can’t be faulted for ignoring it.
The second thing to say is that this sort of worry, though legitimate,
makes some assumptions about what we should want from our moral-
ity. It may indeed be reasonable to think that personal morality ought
to provide solutions to the most significant problems we face. But it also
seems reasonable to maintain that there are limits on what we can ask
of individuals apart from larger institutional changes, and that in many
circumstances we need political action, not personal obligations that are
supposed to work independently of the social and political structures that
support compliance. Someone might complain that this puts us in tragic
circumstances, where morality’s demands are inadequate to address
major social, economic, and environmental problems. But why think that
our circumstances aren’t tragic? I, for one, am inclined to think that we’re
living—and eating—in a broken world.
The third thing to say concerns moral epistemology, and it’s also rel-
evant to the “I’ve ignored the good arguments” challenge. The basic
thought is this: we shouldn’t put too much stock in a single argument
for a particular obligation, especially if that argument relies on a moral
framework in which we aren’t highly confident. And if you think, as
I do, that we shouldn’t be terribly confident in revisionary moral
frameworks—based a principle of epistemic conservativism, if nothing
else—then this thought implies that we shouldn’t place too much stock
in a single argument that relies on a revisionary moral framework. So
Introduction 5
while I don’t doubt that people will continue to generate provocative
arguments for a general duty to be vegan, I’m not inclined to give them
much epistemic weight.
I take all this to be independently plausible, and so not in need of much
defense here. It doesn’t take long for philosophers to develop a healthy
caution about arguments. After all, everyone has an argument, lots of
those arguments seem sound, and yet because the conclusions conflict,
we know they can’t all be sound. Moreover, philosophers are tremen-
dously creative people; we all know that if the literature doesn’t already
contain the principle that implies our preferred conclusion, we’ll be sure
to “discover” it. The demand for consensus provides a check on all this.2
And it also helps fend off the threat that, in not considering every argu-
ment out there, the noseeum inference is unwarranted. Granted, there are
probably some plausible arguments for a general duty to be vegan. But if
we can’t reach the same conclusion from a range of moral assumptions,
then we should be suspicious. When most of our moral theories point
toward a particular duty, that’s a good reason to think we’ve got the duty
in question. When most don’t, and only a handful do, that strikes me
as evidence that we’re in the realm of how to do better relative to some
important values, rather than what we’re obligated to do.
With these challenges behind us, let’s turn to the plan for the book.
I begin with a quick introduction to contemporary animal agriculture.
The main goal there is to get on the same page about the industry that
generates this debate in the first place, but also to make it plausible that
animals are being wronged. (This is the foundation of the “Usually Bad”
part of the subtitle.) Then, I consider and reject what I take to be some
bad arguments for the permissibility of consuming animals. After the
deck-clearing is complete, I turn to the arguments that I—and most
others—have long taken to be more promising. I contend that those argu-
ments fail. What’s more, I make the case that they show the opposite:
instead of having obligations to abstain from consuming animal prod-
ucts, these arguments imply that there are circumstances in which we’re
morally obligated to consume them. The most straightforward cases
involve leftovers and other products that would go to waste, but they
aren’t the only ones.
I find this conclusion highly implausible: surely people aren’t obligated
to eat animal products. Now, if there weren’t reasons to doubt that the
standard arguments for veganism succeed, then perhaps we ought to
believe what’s highly implausible. However, since there are reasons to
doubt that the standard arguments succeed, I take the highly implausible
conclusion to be further evidence that there is no general obligation to
be vegan. It’s a confirmation that we’ve had the wrong approach to the
ethics of eating.
Put more concretely, it’s tough to defend a conclusion about the wrong-
ness of eating animal products based on the wrongness of producing those
6 Introduction
foods. That isn’t a terrible problem in its own right: philosophy is hard,
and it does seem plausible that there’s a link; so, we just need to work a
bit harder. But it turns out that if there were a link, it wouldn’t generate
the conclusion we thought we’d get: namely, that there’s a general obliga-
tion to be vegan. Instead, if there were a link, we should eat unusually:
oysters and insects and garbage and so on. That isn’t plausible. So, given
the antecedent difficulty of linking the ethics of consumption and produc-
tion, we should give up the idea that there’s a link to be found.
In my view, then, we just don’t have obligation-generating agent neu-
tral reasons to eat in one way or another. (This is the “Often Permissible”
part of the subtitle.) Yet again, though, I think that becoming a vegan is
a good thing to do, and that some people ought not to consume animal
products. So in the penultimate chapter, I try to explain how we can end
up with duties to be vegans—which, in my view, also explains why it’s
so easy to think that everyone has dietary obligations that, in fact, they
lack. The basic story is that it’s easy—and a mistake—to collapse the
distinction between activist ethics and ethics for the rest of us. Many
animal activists do indeed have obligations to avoid eating animals, but
this is because of a ratcheting up effect, where relatively weak agent neu-
tral reasons are converted into quite strong agent relative reasons in the
context of a life lived in service to animals. But of course, those people
who make commitments to animals—by taking on particular practical
identities, by entering into a certain sort of relationship with animals, by
joining activist movements—do indeed have obligations as a result. (This
is the “Sometimes Wrong” part of the subtitle.)
As you’ll see, I ignore many criticisms of intensive animal agriculture.
I say nothing about the environmental costs of animal agriculture, the
public health costs, or the burdens that animal agriculture imposes on
workers in CAFOs and slaughterhouses. This isn’t because I take these
issues to be unimportant. Rather, it’s because I don’t take them to be cen-
tral to the case against producing and consuming animal products. In my
view, if it’s wrong to consume animal products, it’s because of something
we’re doing to the animals themselves. In short, the argument for this
is that there is nothing unique about animal agriculture otherwise, and
most people don’t think that it’s wrong to, say, fly on airplanes (which
have a large carbon footprint), or enjoy sugary drinks (which have pub-
lic health implications), or wear some “fast fashion” articles of clothing
(behind which there are well-known worker safety issues).
I’m not writing this book as someone who denies direct moral stand-
ing to animals, or who rejects the principle of equal consideration of
interests, or who doesn’t think that nonhuman beings deserve respect.
This is a book about the ethics of eating, not about whether animals mat-
ter. Moreover, it’s a book that I’ve written with some trepidation. Many
years ago now, I stopped eating animals because my wife wept as we
drove past a truck full of pigs, no doubt destined for slaughter. And since
Introduction 7
then, I’ve spent a lot of time giving others arguments to the effect that
they ought to do the same. I don’t think I was wrong to encourage peo-
ple to take animal seriously, nor to encourage them to show their com-
mitment to animals through their food choices. What’s more, I haven’t
stopped giving people reasons to change their diets. And I regret, quite
deeply, that some people are bound to use this book to discourage people
from going vegan. But I’m trying to be honest about the arguments I’ve
long offered. As far as I can see, they don’t establish as much as I once
thought they did.
Notes
. I owe this line of thought to David Alexander, though I’m not sure he believes it.
1
2. I have additional reasons for insisting on consensus that are far more contro-
versial. I think that a moral framework gives you a glimpse of the way moral-
ity might work if we were to ignore lots of the things we value. Thinking like
a consequentialist or a contractualist is a bit like looking at the world through
tinted glass, where only some of the moral features are visible. So if you’re like
me, an irredeemable value pluralist (as opposed to someone showing some
measure of epistemic humility in the face of moral uncertainty, which is all
I need in the main text), then it would be awfully strange to argue for particu-
lar obligations from a single moral framework. Instead, you should be looking
through each lens, and if you see the same thing almost all the time, then—
and only then—does it seem reasonable to say that we’ve got an obligation
simpliciter, as opposed to saying that we’ve got an obligation according to (a
given version of) utilitarianism or according to (a given version of) Kantian-
ism. Consensus is what could convince us that our values collectively require
a course of action, and so it’s obligatory, as opposed to some weaker conclu-
sion. Granted, there are wrinkles here. Couldn’t we find it plausible that while
our values don’t collectively require a particular conclusion, one of our values
trumps the others? Sure. But I tend to think that such intuitions should usually
support the view that it’s particularly morally good to act in a certain way, not
that it’s obligatory to act in that way, as moral theories are typically designed
to accommodate the powerful intuitions that favor their competitors. So, it
would be surprising if, say, a virtue theory didn’t acknowledge that a given
case is one where the consequences are especially important. Moreover, I don’t
think that we’re in a “one value clearly trumps the others” situation when it
comes to the ethics of eating, since I don’t think it’s easy to find any value that
squarely supports strict veganism.
2 Contemporary Animal
Agriculture
My goal in this chapter is to avoid looking away. It’s very easy to get
caught up in the nuances of various debates and, in so doing, lose sight
of the animals who are supposed to motivate it. If we are going to say,
as I do, that most people aren’t obligated to be vegans, then we shouldn’t
say it as a result of willful blindness about the realities of intensive animal
agriculture.
Some are, I think, guilty of just this. Consider, for instance, Loren
Lomasky (2013, 192):
[I]t is plausible to suppose that the animals that end up on our tables
enjoy net gains in virtue of our culinary proclivities. My hunch is that
these gains are considerable, but hunches carry limited weight and
mine, perhaps, more limited than most. To be cautious, let’s call it a
toss-up whether animals on balance win or lose out. If so, then the
case for eating animals [based on its contribution to human flourish-
ing] is overwhelmingly positive.
Frankly, I’m not sure that animals don’t “enjoy net gains in virtue of
our culinary proclivities.” I grant that this is difficult to assess.1 How-
ever, even if we accept that animals in intensive systems generally have
lives that are good-on-balance, it doesn’t follow that those systems
are morally unproblematic, nor that “the case for eating animals is
overwhelmingly positive.” Animals could be given lives that are good-
on-balance and yet not good enough. After all, good-on-balance is a
low bar. Presumably, many rape victims live lives that are good-on-
balance; their lives are worth living despite including a truly horrific
event. But their lives haven’t been good enough: something terribly
wrong has been done to them. So the question we need to consider
is not whether animals have lives that are good-on-balance. Instead,
it’s whether producers harm animals in ways that make their lives
insufficiently good. If they do, then we can fairly criticize the practice
of creating animals who have such lives, which is to say that we can
criticize intensive farming.
Contemporary Animal Agriculture 9
It seems to me that most animals in intensive systems have lives char-
acterized by low-level stress that’s punctuated by acute pain. There are
others who have argued for this conclusion in great detail, so I won’t
provide an exhaustive catalog of the ways in which animals are harmed
in modern farms. Instead, I’ll just draw attention to a few representative
cases. My aim in this book is to defend the thesis that it’s morally permis-
sible to eat the bodies and byproducts of animals who have had lives like
this. I’m not claiming that it’s good to do it. Indeed, I think it’s usually
bad. But if you’re going to defend bad behavior, then you should be frank
about the costs that such behavior imposes. That’s the project here.
2.1 Low-Level Stress, Acute Pain

In the US, roughly ten billion land animals are raised and slaughtered for
food. How do they fare during their short lives?
2.1.1 Cows
The USDA (2018d) reports that 32.5 million head of beef cattle were
alive in the United States in July of 2018. Nearly all of them began their
lives as calves who were quickly separated from their mothers. About
half of beef producers wean and ship calves the same day (USDA, 2007b).
Rapid weaning and separation are in themselves sources of stress for
both calves and their mothers (Weary, Jasper, and Hötzel, 2008; Boland
et al., 2008), but they also place the calves at greater risk for developing
illnesses—especially bovine respiratory disease, which reduces appetite,
causes depression, and results in difficulty breathing—partially because
most calves aren’t vaccinated before sale (Peterson et al., 1989b; Schip-
per, Church, and Harris, 1989). This accounts for more than half of mor-
talities in feedlots (Loneragan and Brashears, 2005).
Beef cattle are routinely branded or tagged, disbudded or dehorned,
and castrated. None of these experiences is pleasant. During branding,
for instance, cattle often try to escape, bellow, and sometimes collapse
(Schwartzkopf-Genswein, Stookey, and Welford, 1997; Schwartzkopf-
Genswein et al., 1998; Lay et al., 1992). The wounds are painful and take
about eight weeks to heal, and while medication may be given imme-
diately before or after branding, it isn’t given throughout the healing
process (Tucker et al., 2014a, 2014b).2 Essentially the same is true of
dehorning and castrating, both of which cause significant discomfort and
heal slowly (Stafford and Mellor, 2011; Stock et al., 2013; White et al.,
2008; Currah, Hendrick, and Stookey, 2009; González et al., 2010).
We’ve got good reason, then, to think that the first two months of their
lives are shaped, at least in part, by these injuries.
Life on a feedlot can be relatively benign, at least when the stocking
densities are low and the climate is favorable, but this isn’t the case for
10 Contemporary Animal Agriculture
most cattle. For instance, exposure to mud has been found to reduce
growth and increase the incidence of lameness (Borderas et al., 2004),
and one study found that 74% of cattle have manure or mud on their
bodies at the time of slaughter (Garcia et al., 2008). Beef cattle have been
found to avoid wet, cold environments when given the choice (Fregonesi
et al., 2007; Fisher et al., 2003), yet many beef cattle are raised out-
doors in environments with severe winters. Exposure to heat outdoors
can also adversely affect cattle, resulting in significant—and sometimes
mortal—physiological stress (Brown-Brandl et al., 2005; Nienaber and
Hahn, 2007); nevertheless, few producers provide adequate shade or
water-cooling systems (UNDP, 2000). Overcrowding indoors contributes
to increased stress and aggression, but economic pressures encourage
producers not to build more barn than is necessary.
Even when the physical environment doesn’t pose problems, there
remain social stressors. Cattle get stressed in large groups, especially
when they are placed with unfamiliar animals (Patison et al., 2010). This
makes them more aggressive, which becomes a welfare problem for other
cattle (Warren, Mandell, and Bateman, 2010). This can take dark forms,
where a single steer is repeatedly mounted by a group of steers, causing
fear, injury, and exhaustion in the victimized animal (Taylor et al., 1997).
Not all the threats are external. Beef cattle often suffer from sub-acute
ruminal acidosis (SARA) and acidosis, which in turn cause liver abscesses
and laminitis (painful inflammation in the hoof), which are common
digestive conditions that are known to be caused by the high-concentrate
diet that most US cattle are fed ( Schwartzkopf-Genswein et al., 2003;
Nagaraja and Lechtenberg, 2007a). Estimates of the incidence of liver
abscesses vary widely, with reports ranging from 12 to 56% (Nagaraja
and Lechtenberg, 2007a; Fox et al., 2009). Feeding cows antibiotics
can decrease the incidence of liver abscesses (Nagaraja and Lechten-
berg, 2007b; Wileman et al., 2009), but this creates now-familiar wor-
ries about the development of antibiotic-resistant bacteria. This means
that for many animals, what may seem like a straightforward pleasure—
namely, feeding—comes at a cost.
The animals’ final days have their own sources of pain. Over half of
cattle in the US are administered beta-agonists in the final weeks of their
lives to promote efficient weight gain before slaughter (Lean, Thompson,
and Dunshea, 2014). These drugs seem to increase aggressive behavior,
decrease food intake (Reinhardt et al., 2014), and increase the likeli-
hood of death prior to slaughter (Loneragan, Thomson, and Scott, 2014;
Stackhouse-Lawson et al., 2015).3 Then, transport between production
facilities presents a host of welfare concerns. Cattle can be in transport
for 28 hours (USDA, 1999), can be packed at uncomfortable densities,
and aren’t provided with food or water. They can be exposed to excessive
heat and cold, contributing to lameness and premature death, because
temperature isn’t regulated inside trailers and railcars (González et al.,
2012; Schwartzkopf-Genswein et al., 2012). Slaughter itself is usually
quick, though there are predictable errors when enormous animals are
being killed at a rate of one every twelve seconds. It’s hard to know the
specific error rate, but it’s clear that some cattle are still conscious when
their throats are cut and they are hung upside down to bleed out.
2.1.2 Pigs
Roughly 121 million pigs were slaughtered in 2017 (USDA, 2018a), and
some 73 million are alive at any given time (USDA, 2018c). Unlike cat-
tle, most pigs live on large commercial farms where they’re confined to
indoor living spaces. A pig in a growing and finishing facility has a space
allowance that isn’t enough to accommodate their natural social and
foraging behaviors. Moreover, the close quarters predispose some pigs
to tail-biting—a problem often solved by tail-docking, itself a painful
procedure.
Pigs’ digestive systems are biologically suited to small amounts of high
fiber feed, but they’re generally fed low fiber diets in unlimited quantities,
leaving them prone to potentially fatal gastric ulcers (Nielsen et al., 2006;
Eisemann and Argenzio, 1999). And due to high stocking densities, the
air quality in pigs’ living spaces is poor. Animal waste, feed particles, dan-
der, and noxious gases contribute to reduced activity levels, reduced abil-
ity to resist bacterial infections, influenza, pneumonia, pleuritis, porcine
reproductive and respiratory syndrome (PRRS), and increased neonatal
mortality (Gregory, 2007; Donham, 2000; Von Borell et al., 2007). In
fact, respiratory problems are the main cause of death for growing pigs
in intensive settings (USDA, 2007c).
Having spent most of their lives confined indoors, being transported
to a slaughterhouse is a novel—and so stressful—experience. Many fac-
tors conspire here: crowding; aggression between unfamiliar individuals
(Bradshaw et al., 1996); extreme temperatures ( Grandin, 2001), sudden
braking and changes in acceleration; vibration due to motion (Perremans,
et al 2001); and dehydration and hunger, since they often aren’t fed or
given water both prior to and during transport (Lambooij, 2007). Under
these conditions, pigs pant and their skin becomes discolored, which are
signs of anxiety (Ritter et al., 2008). These difficulties are exacerbated for
some pigs by a genetic defect. Selective breeding has increased feed con-
version efficiency, but it’s also led to the emergence of porcine stress syn-
drome, a condition characterized by difficulty breathing, cyanosis, and
increased body temperature in response to hardship (Grandin, 1998).
Federal law requires that pigs must be rendered insensible to pain prior
to slaughter (U.S.C., 1958), and this is usually accomplished through the
use of a captive bolt gun, an electric current, or carbon dioxide gassing.
These can all be ineffective if equipment isn’t well maintained or person-
nel aren’t adequately trained ( EFSA, 2004). But even if they’re effective,
the process can be brutal. Using CO2 to stun animals can cause them
to experience severe respiratory distress before death ( Rodríguez et al.,
2008; Raj, 2008), and extreme line speeds—where 1200 pigs are killed
per hour; one every three seconds—create welfare problems akin to the
ones we discussed earlier.
2.1.3 Chickens
But chickens probably have things the worst. A total of 8.91 billion
chickens were produced for their meat in the US in 2017—a two percent
increase from 2016 (USDA, 2018b). Meat production has generally been
increasing for at least the past 50 years, and chicken production has seen
by far the greatest relative increase. In 1961, chicken accounted for about
20% of all meat produced; in 2013, it accounted for 48% ( FAO, 2018).
Most broiler chickens (chickens raised for meat) are raised indoors in
crowded, warehouse-like facilities (Watt Poultry USA, 2005). For broiler
chickens nearing market weight, the average floor-space allowance for
each bird is an area slightly larger than a sheet of letter-sized paper (Este-
vez, 2007). Generally, increasing the number of chickens in a given indoor
space increases dust and ammonia concentrations, which adversely affects
chickens’ respiratory health (Banhazi et al., 2008; Al Homidan and Rob-
ertson, 2003). Crowding is also correlated with a higher incidence of
injury—for example, sores, scabs, and scratches caused by other birds,
as well as skin irritation, lesions, or deep ulcers on the backs of the legs
and feet caused by walking and lying in wet, infrequently changed litter.
Chickens raised in crowded conditions are unable to spend adequate time
resting, as they frequently walk over each other (Duncan, 2004; Hall,
2001; Febrer et al., 2006; Buijs et al., 2010). Chronic interruptions to
rest have been associated with bone quality problems (Buijs et al., 2012).
The indoor spaces in which these chickens live tend to be lit artificially
on a schedule that promotes feeding behavior and body growth at the cost
of sleep, rest, and normal eye and bone development. Nearly continuous
lighting is common, though the lights tend to be dim (Duncan, 2004;
Schwean-Lardner et al., 2013), again to promote growth (low lighting
inhibits physical activity; Lacy, 2002). Dim lighting has been associated
with abnormal eye development and ulcerative footpad lesions (Prescott,
Kristensen, and Wathes, 2004; Deep et al., 2010).
Selective breeding for larger birds and faster growth presents perhaps
the most significant welfare problem. Average daily growth rates of chick-
ens have quadrupled in the past 50 years (Knowles et al., 2008). Extreme
rapid growth contributes to poor bone health (Shim et al., 2012), leg
deformities, ruptured tendons, pinching of the spinal cord (Julian, 2005),
accumulation of fluid in the abdominal cavity, enlarged heart, shrunken
liver, (Duncan, 2001; Boersma, 2001), and sudden death syndrome
(SDS), among other ailments (Bessei, 2006; Bradshaw, Kirkden, and
Broom, 2002). Heavier chickens are more likely to be unable to walk nor-
mally or at all (Kristensen et al., 2006), and research suggests that non-
ambulatory birds are probably in pain (Danbury et al., 2000; Nääs et al.,
2009). Faster growing birds are more prone to developing tibial dyschon-
droplasia, where an abnormal mass of cartilage at the tibial growth plate
causes the bone to develop abnormally—bending backward, in some
cases—resulting in bone fracture, necrosis of the cartilage, and inability
to walk or stand (Julian, 2004, 2005). Chickens who can no longer walk
or stand may be forced to lie in their own excrement and that of previous
flocks Dozier, Lacy, and Vest, 2001).
Broilers are often harmed by human handlers as they are caught and
crated for transport. Six- to seven-week-old chickens are grabbed by the
legs and held upside down, six to eight birds at a time, and placed in
crates (Scanes, Brant, and Ensminger, 2004). The chickens experience
fear, pain, and sometimes severe injury (e.g., bone dislocation, hemor-
rhage, broken bones) during the process (Gregory and Wilkins, 1992;
Weeks, 2007). As handlers get tired after lifting so many birds over the
course of a shift, things get worse. At some point, preventing injuries
tends not to be their primary concern (Nijdam, et al 2004; Lacy and
Czarick, 1998).
As with cattle and pigs, the ride to the slaughterhouse is a difficult
one: the trucks are crowded; the birds aren’t given food or water; the
temperatures can be extreme; and there is the constant noise, vibration,
and motion of the vehicle itself. Chickens sometimes die from injuries
sustained during catching and crating, infectious diseases, and heart dis-
orders on the way—between 10 and 39 million per year (Nijdam et al.,
2006).
Chickens are dumped onto conveyor belts when they arrive at the
slaughter plant. Soon thereafter, they’re hung upside-down, by their legs,
in shackles. This is disorienting and painful for all, but it’s particularly
difficult for individuals suffering from broken bones or dislocated joints
(Gentle and Tilston, 2000; Raj, 2004). Many birds injure themselves fur-
ther as they struggle. Shortly thereafter, they are dunked into an elec-
trified water bath to render them unconscious, though high line speeds
mean that this doesn’t always work (and even when it does, it can be
painful; Rao, Knowles, and Wotton, 2013). So, some chickens get to
experience having their throats cut (EFSA, 2004), though that doesn’t
always kill them. Chickens go to a scald tank to remove their feathers,
and those who survived the stunning and cutting find themselves boiled
alive (Raj, 2004).
2.1.4 Incentives
Why is this normal? Why is it the status quo? It’s easy to read all this
and demonize those who are responsible for the care of these animals.
Perhaps some demonization is appropriate. But it’s worth recognizing
the ruthless logic that drives producers and handlers to treat cattle in
these ways. Timothy Pachirat (2011) describes his experiences as an
undercover worker in a slaughterhouse, where he volunteered to work
the “shoot,” which is the ramp that cattle travel from a staging area to
the “knocker,” who puts a large steel bolt between the animal’s eyes.
He’s being trained by two senior workers, Gilberto and Camilo, who are
supposed to use plastic paddles to keep the cattle moving, but have the
option of using electric prods instead. Pachirat writes:
After a few hours in the shoots, it becomes clear to me that both Gil-
berto and Camilo used electric cattle prods extensively, sometimes
taking them under the animals’ tails and into their anuses. The cat-
tle jump and kick when shocked in this way, and many also bellow
sharply. Gilberto uses the prod in almost wrote fashion, shocking
practically every animal, especially as they near the whole in the
slaughterhouse wall that leads into the knocking box [where the ani-
mal is killed]. Even when the cattle are tightly packed, with the nose
of one animal pushed up against the rear of the animal in front of
it—sometimes even with its head squished between the hind legs of
the animal in front of it—Gilberto still delivers electrical shock, often
causing the cow to mount the animal in front of it.
(144–145)
Once, when the line is moving too slowly for Fernando’s liking,
he sprints up the walkway from the squeeze pen, grabs the plastic
paddle out of my hand, and shows the electric brought into it. “You
motherfucking pussy!” he yells. “Do your job and use the fucking
hotshot!”
“Why?” I yell back. “What’s the point of shocking them? They’re
all moving through the line anyway.” . . .
“The point is pain and torture,” Fernando retorts, laughing. . .
Furious, I repeat the question. “Okay,” he finally shouts back;
“you wanna know why I use this?” He shoves the tip of the electric
prod across the shoot in my direction. “I use this because I like to
have my work. And if we don’t keep these cows moving through,
they’re gonna call us up to the office and were going to get fired.
That’s why.”
. . . By my third day in the shoots, after several warnings from [my
supervisor], I too increasingly rely on the electric prod. The point
of using the prod is not “pain and torture,” in Fernando’s mocking
words, but rather avoiding conflict with coworkers and supervisors;
in addition, once the abstract goal of keeping the line tight takes
precedence over the individuality of the animals, it really does make
sense to apply the electric shock regularly. Rather than electrocuting
an individual animal, the prod keeps a steady stream of raw mate-
rial entering the plant, satisfies coworkers and supervisors, and saves
[workers] from having to expend the energy it takes to move the
animals with plastic paddles.
(148–149)
I suspect that many people in the industry are just like Fernando: they’re
motivated by perfectly ordinary considerations—paying bills, keeping
jobs, avoiding interpersonal conflict—that, due to their line of work and
the world they’ve inherited, lead to routine harms to animals. And they
inherited it because people before them were also motivated by perfectly
ordinary considerations that, when the animals themselves aren’t in view,
have disastrous results in the aggregate.
If you were to visit an American farm 150 years ago, you would see a
diversified farm. The farmer would be growing a number of crops, partly
to provide a range of important goods to his family, but also to maximize
productivity: if he doesn’t rotate what he plants in any particular field,
pests would become a much larger problem, and the soil would soon be
sapped of nutrients. Moreover, he would have a mix of animals, raised
for meat, milk, fertilizer, and trade.
The vast majority of modern farms are nothing like this. Pesticides
and chemical fertilizers have made it possible to plant the same thing
in the same fields, year after year, with remarkable yields. As yields go
up, feed becomes cheaper. Long-distance transportation and cold storage
have also become reliable and relatively inexpensive. These changes have
made it possible for a single farm to raise, slaughter, and sell an increas-
ing number of animals so instead of the diversified farm, we have the hog
farm, the chicken farm, the feedlot.
The economic incentives here are dramatic. In 2007, farms raising
fewer than 1,000 pigs spent about $45 to add 100 pounds of weight to
their pigs. For farms raising more than 10,000 pigs, they could add the
same weight for about $25 (Key and McBride, 2007). Confinement is
bad for welfare in many respects, but it’s very good for ensuring adequate
feed, preventing “unnecessary” movements that burn calories, and pre-
venting “early” mortality from the disease and injury. Moreover, it keeps
labor costs down: a handful of people can ensure food, water, and mini-
mal medical care to thousands of animals, which would require dozens
of people in an extensive system.
Moreover, economies of scale make an enormous difference. Obvi-
ously, it’s cheaper to buy feed and medicine in bulk. On a per-hog basis,
it’s cheaper to heat and cool a large barn than a small one. And crucially,
it’s cheaper to transport, slaughter, and process large numbers of animals.
Small producers can minimize stress to their animals by using a mobile
slaughter unit, which slaughters and processes on the farm and offers
what is, arguably, the best option from the perspective of maximizing
animal welfare. But it can only handle about 24 hogs per day. By con-
trast, and as I’ve mentioned, large slaughterhouses can slaughter and pro-
cess 1,200 hogs per hour.
With these kind of incentives in mind, it should be obvious that farm-
ers have strong reasons to specialize and scale up quite independently of
consumer demand. Even if consumers were to consume far fewer animal
products, the result wouldn’t necessarily be higher welfare farms. Rather,
it could be fewer farmers in the marketplace, with all the surviving ones
using intensive production systems. I certainly don’t mean to suggest that
it would be a trivial thing to have fewer animals in intensive production
systems. To my mind, that would be an obvious good. My point here is
just that we shouldn’t think of the problem of farmed animal welfare
as exclusively about demand, as though not purchasing animal prod-
ucts changes how it’s rational for farmers to treat their animals. Farm-
ers will continue treat their animals poorly unless consumers actually
demand higher welfare or regulators step in and require it. And until
then, we should expect to find what the evidence suggests: namely, sys-
tems designed to convert feed into meat, dairy products, and eggs in the
most efficient way possible. The welfare problems in these farms stem
from their structure; they’re the everyday consequences of trying to be
profitable. In an important sense, the animals themselves aren’t in view,
making stress and pain the result.
2.1.5 Taking Stock

I haven’t said anything about dairy cows or veal cows or laying hens; I’ve
also ignored farmed fish in aquaculture. This isn’t because their lives are
marvelous. Instead, it’s because telling their stories would be more of the
same, and the above is enough to make the point: low-level stress, acute
pain.
Of course, even if I were to discuss these animals, I wouldn’t be dis-
cussing all the ones that people eat. Wild caught fish are perhaps the most
common and complex case. They certainly suffer when caught: they make
every effort to escape when hooked or netted; they can be crushed by the
weight of other fish when being hauled out of the water or when dropped
into the boat; they usually die by asphyxiation. However, their lives prior
to being caught are either as good—or as bad—as those of other wild
aquatic animals.4 So, the total amount of human-inflicted harm is less,
and the “in principle” moral question turns largely on whether we’re jus-
tified in shortening these lives for our purposes. The upshot is that while
there are obvious connections between the ethics of fishing and the ethics
of farming, they’re importantly distinct.
But let’s not get distracted by wild fish, though I do think that they’re
an interesting and complicated case. By and large, people eat land ani-
mals and farmed fish. So that’s where the real action is, morally speaking.
And when we focus on those cases, we should note that the above helps
us dismiss a kind of skepticism about reports of welfare problems in
intensive systems. Consider Lomasky again:
Is some factory-farming unduly brutal and morally corrupt?

Undoubtedly. Is much consistent with good-on-balance animal lives?
In a word, yes. Why then, do so many people categorically reject ill-
defined factory-farming? I can think of two explanations. First, one
may imaginatively put oneself in the position of the cooped-up bird
or the steer waiting to be stunned and slaughtered and respond with
a shudder. Such emotional responses are understandable, but I do
not believe that they carry much epistemic weight. Second, disturb-
ing videos of meat-packing operations are typically secreted out and
made public by groups that are stridently anti-meat. It can, I believe,
be safely assumed that the items they release for general viewing are
those that most strongly support their own agenda and are not rep-
resentative of the full range of industry practices.
(2013, 192 n. 20)
It should be clear why Lomasky’s debunking story fails. You don’t need
to imagine yourself as the farmed animal to be troubled by their plight:
there is excellent evidence that they are living in circumstances that impose
serious welfare costs. Moreover, even if we don’t reach their conclusions
based on their testimony, we can’t easily dismiss the claims that activists
make about factory farming. I’ve tried to be measured in my presentation
of the welfare problems in contemporary animal agriculture, drawing
largely on industry-friendly sources; still, the picture isn’t pretty. Again,
I take no stance on whether these are lives that are good-on-balance, but
that issue seems irrelevant. These are lives that aren’t good enough, and
any conversation about the ethics of eating animals needs to begin by
acknowledging this.
2.2 What Follows?

Still, it isn’t clear what follows. Here’s a sampling of the things we might
believe based on the above:
1. It’s wrong for producers to bring these animals into existence.

2. Producers are wronging (or at least acting wrongly with respect to)
the animals they farm.
3. It’s wrong for regulators to allow these practices to continue.
4. We—consumers (in general), consumers (of animal products), relatively
privileged consumers (in general or of animal products), Americans,
Westerners, voters, the politically connected, or what have you—have a
collective obligation to prevent animals from living such lives.5
5. Each of us, individually, has an obligation to respond in some way to
the realities of intensive animal agriculture.
6. Each of us, individually, has an obligation not to consume animal
products.
I find versions of (1)–(5) plausible, but I won’t argue for any of them. My
only claim is that (6) is false. A corollary is that even if (1)–(5) are indeed
true, then I don’t think (6) follows from them.
I wouldn’t have thought that the corollary would be that controversial.
Consider, for instance, the claim that “we”—whoever that is—have a
collective obligation to prevent animals from living such lives. It’s ante-
cedently plausible that this doesn’t imply anything about any particular
consumer’s specific obligations. After all, it’s just generally true that our
collective obligations don’t specify our individual obligations. Americans
have a collective obligation to respond to human rights violations, both
at home and abroad, but nothing immediately follows about the respon-
sibilities of the average American citizen. So I can readily grant that we
have a collective obligation not to raise animals and slaughter animals as
we do. From that premise, however, nothing immediately follows about
what the average consumer should do.
One natural conclusion is that we have an imperfect duty to respond to
our collective obligation. On such a view, it’s wrong not to do anything in
response to the realities of intensive animal agriculture, but there won’t
necessarily be a single appropriate response—there may be considerable
variation. (In other words, (5) is true.) As it turns out, that’s my own
view. But it isn’t the standard one among those who work in animal
ethics. The standard view is that veganism is the appropriate response,
though there is plenty of disagreement about why that is.
In any case, before considering arguments to the effect that we ought
to be vegans, I want to consider and set aside a number of arguments for
thinking that we’ve gotten off on the wrong foot. Perhaps it’s the case
that, despite appearances, intensive animal agriculture isn’t objection-
able. Or perhaps there is some relatively straightforward problem with
the idea that veganism is morally required: for instance, it would require
us to sacrifice our health. I don’t think that these arguments work, but
there are valuable lessons to be learned from them—lessons that will
prove useful when assessing the pro-vegan arguments in the subsequent
chapters. That’s the project for the next chapter.
Notes
1. First, there are questions about what makes lives go well. Positive mental
states? Bodily health? Natural behaviors? An adequate life narrative? Some
combination of these factors? Second, unless we offer a theory in which there
is only one thing that contributes to well-being, we have to develop a theory
about the relative weights of the contributing factors. It’s worth noting that
this is important even for a hedonistic theory, since we can’t assume that one
“unit” of happiness makes a positive contribution to welfare that’s equivalent
to the negative contribution made by a “unit” of pain (unless, of course, we
define the terms in a way that makes them equivalent, which simply pushes
the problem back). Third, once we settle these philosophical questions, we
still have to collect the relevant empirical data. We immediately find more
problems. For instance, there are plenty of questions about how to operation-
alize the philosophical account that we find compelling. Relatedly, there are
difficulties associated with interpreting the existing empirical literature in light
of our preferred account of well-being and its operationalization, since it isn’t
practical (or possible) to redo countless person-hours of research.
2. In the case of castration, for instance, the best estimate is that roughly 30%
are given pain relievers. See www.avma.org/KB/Resources/LiteratureReviews/
Pages/castration-cattle-bgnd.aspx.
3. These studies investigated effects of zilpaterol, which was taken off the mar-
ket, but other beta-agonists are still used. One of them did examine ractopa-
mine, however, which is still on the market.
4. We tend to operate with what Oscar Horta (2010) calls “the idyllic view of
nature.” According to this view, nonhuman animals generally have lives worth
living when left to themselves; most of them experience more pleasure than
pain. But many have argued that this is false: see, for instance, Faria (2016),
Faria and Paez (2015), Horta (2010, 2017), Ng (1995), and Tomasik (2015).
On this alternative view, most animals live “net negative lives,” that is, they
experience more pain than pleasure for reasons having little to do with human
beings. Animals regularly starve, suffer from debilitating injuries and diseases,
and find themselves the victims of predation. And when not facing these acute
sources of pain, they are often stressed by their possibility: many animals live
their lives on high alert, constantly having to be wary of deadly threats. More-
over, it’s plausible that most animals don’t live long enough to reach more
secure positions in the food web where they are less vulnerable to predation.
Consider that elephants might have 6 offspring; mice, 40; some species of
salmon, 17,000; and ocean sunfish, 300 million per year. Obviously, most of
these animals die young.
5. If we do have such an obligation, then we might wonder what we should be
aiming for. According to Torbjörn Tännsjö (2017), we should be total hedon-
istic utilitarians, that is, we ought to maximize the total amount of happi-
ness. As a result, we ought to bring animals into existence, provide them with
excellent lives, kill them painlessly, and enjoy their bodies and byproducts.
According to Christine Korsgaard (2018), by contrast, we should treat ani-
mals as ends in themselves, which she takes to be incompatible even with most
“humane” forms of animal agriculture. Of course, there’s always disagree-
ment in philosophy, and I’m not suggesting that we need to settle all debates in
moral theory before making a judgment call. Maybe moral caution is appro-
priate here, and we should err in favor of Korsgaard’s position (or something
in that neighborhood). My point here is just that there is indeed debate, and
that debate is relevant to the shape of any purported collective obligation. If
the goal is to reform an industry, that’s one thing; if it’s to abolish it, that’s
another.
3 Bad Arguments for Eating
Animals
In this chapter, I want to do some ground clearing. My goal here is to

introduce and evaluate arguments for the permissibility of eating ani-
mals that I don’t take to be compelling—and, frankly, with which I don’t
want to be associated. I do think that it’s permissible for most people to
consume animal products, but I don’t think that it’s permissible because
animals lack moral standing, or because I’m in the grip of some agrarian
vision of the world, or because I think that a general prohibition on the
consumption of animal products is sexist or racist. And once we appreci-
ate the problems with these arguments, it becomes easier to see the con-
straints on the case that I build in subsequent chapters.
To be clear, in this chapter, I’m simply assuming that if there is a prob-
lem with standard arguments for veganism, it’s one of these problems. So,
I’m assuming that the problem might be something like this: the standard
arguments presuppose the principle of equal considerations of interests,
according to which like interests deserve like consideration, and we have
good reason to endorse a theory of moral status according to which this
principle is false. Obviously, if this principle is true, then animals should
be getting far more consideration than we give them. They have an inter-
est in being free from pain, and that’s an interest we generally neglect. If
their interest in being free from pain is as important as our own, animal
agriculture should change dramatically, if not end entirely.
Alternately, the problem might be something like this: the standard
arguments for veganism presuppose that gustatory pleasure is a trivial
good, whereas it’s in fact hugely valuable. In fact, it’s so valuable that the
pleasure we derive from eating animals outweighs any costs we impose
on them. Alternately again, the problem might be something like this: the
standard arguments for veganism presuppose that everyone has equal
access to vegan fare, though that’s false.
I’m assuming, then, that each of these objections is supposed to work
on its own. In other words, the value of taste is itself enough to show
this, or that the point about access is itself enough to show this. I’m
assuming that these arguments aren’t smuggling in premises about indi-
vidual choices not making a difference, or not having any obligation to
Bad Arguments for Eating Animals 21
disassociate from injustice, or what have you. Indeed, that’s why I think
they’re bad arguments. With those additional premises, some of them
may not be so bad—a point to which I’ll return at the end.
3.1 Contractualism
Let’s begin with most radical challenge to arguments against eating ani-
mals. The basic argument, which we owe to Peter Carruthers (1992),
goes like this: contractualism is a more plausible moral theory than the
alternatives, and if contractualism is true, then it’s permissible to eat ani-
mals; so, it’s permissible to eat animals.
Stated this simplistically, the argument is plainly invalid. To fix the argu-
ment, you need to add this premise: if contractualism is more plausible
than the alternatives, then it’s permissible to act; however, it would be per-
missible to act if contractualism were true. But that premise is false: con-
tractualism could be more plausible than the alternatives without being
plausible enough to warrant such a blanket permission. After all, “more
plausible than the alternatives” is compatible with “probably not true.”
However, let’s set this worry aside, focusing instead on whether we
ought to accept the sort of contractualism that implies that it’s permissi-
ble to eat animals. To answer this question, we need to get a better handle
on the view.
In much of the animal ethics literature, there’s an assumption that the
correct account of moral status will (1) agree with our considered judg-
ments about those beings that do and don’t deserve at least some moral
consideration and (2) provide a direct rather than indirect account of
the moral importance of various beings. To clarify the second point, the
thought is that the correct account of moral status isn’t simply going to
be extensionally adequate, where the account can be coupled with plau-
sible assumptions to say that an action is morally required or forbidden
if and only if our considered intuitions say that it’s morally required or
forbidden. Additionally, the correct account will explain a being’s deserv-
ing moral consideration in terms of properties intrinsic to it.
Carruthers (1992, 2011) rejects the demand for direct explanation.
That demand is designed to preclude indirect duty views of our obliga-
tions to animals, perhaps most famously associated with Kant. Accord-
ing to Kant (Lectures on Ethics, 212),
we have no immediate duties to animals; our duties towards them

are indirect duties to humanity. Since animals are an analogue of
humanity, we observe duties to mankind when we observe them as
analogues to this, and thus cultivate our duties to humanity. If a
dog, for example, has served his master long and faithfully, that is
an analogue of merit; hence I must reward it, and once the dog can
serve no longer, must look after him to the end, for I thereby cultivate
22 Bad Arguments for Eating Animals
my duty to humanity, as I am called upon to do; so if the acts of
animals arise out of the same principium from which human actions
spring, and the animal actions are analogues of this, we have duties
to animals, in that we thereby promote the cause of humanity. So if a
man has his dog shot, because it can no longer earn a living for him,
he is by no means in breach of any duty to the dog, since the latter
is incapable of judgement, but he thereby damages the kindly and
humane qualities in himself, which he ought to exercise in virtue of
his duties to mankind.
According to such views, when it’s wrong to harm animals (and it often
isn’t), this is not, fundamentally, because of some property they have.
Instead, it’s because in so doing we either violate the rights of those who
own those animals, or deform our characters, or otherwise negatively
affect someone who does in fact have moral standing. By contrast, Peter
Singer (2009) and Tom Regan (2004)—the most famous “animal libera-
tionists” among philosophers—agree that insofar as it’s wrong to harm
an animal, it’s primarily because of what it does to the animal—not for
one of these other reasons. In any case, without that demand, Carruthers
can develop a form of contractualism that denies direct moral standing
to animals.1 And if we assume that no animal is a rational agent, then it’s
easy to see why animals wouldn’t have moral standing, since contractual-
ists take morality to be something like the set of rules that self-interested
and rational agents would accept when reasoning together under ideal-
ized conditions. No agent is going to agree to a set of rules on which she
has no right to moral consideration, so every agent will have standing.
The question then is whether agents have something to gain by grant-
ing the same right to animals. Since they probably don’t, they probably
wouldn’t.
The trick is to explain why agents would agree that every human
should have moral standing, even when the human in question isn’t an
agent. Carruthers has several things to say about this. First, the rules
have to be psychologically supportable: that is, agents have to be able to
endorse them without coercion. And agents probably won’t be able to
endorse rules that don’t grant moral status to infants or those with severe
cognitive disabilities, even if self-interest might be served by excluding
such individuals. Our sympathies for these beings are too strong. Second,
self-interest will lead agents to protect the senile and comatose and brain
damaged, given that this may well become their fate. And third, contrac-
tors have reason to endorse rules that promote virtue in themselves and
others, at least insofar as virtue serves the end of the contract process:
namely, establishing rules that lead to a stable society. So, there will be a
strong presumption in favor of including beings like us, since we’re most
likely to become desensitized to harms to one group of humans if we
tolerate harms to another group. Carruthers grants that agents might not
agree to rules that grant standing to absolutely every human being, but
he thinks that the rare exceptions will be tolerable, such as anencephalic
infants. The upshot is that (nearly all) humans deserve moral considera-
tion, no animal deserves it, and our obligations to animals are severely
limited.2 In fact, he goes so far as to argue that even factory farming is
permissible, since it can’t be ruled out by considerations of character.
He claims that “almost any legitimate, non-trivial motive is sufficient to
make [an] action separable from a generally cruel or insensitive disposi-
tion,” which means that the desire to make a living can excuse factory
farm workers for many of their ostensibly cruel actions (1992, 159). Pre-
sumably, the same point applies to consumers’ desires to be sated by tasty
and nutritious animal products, to preserve familial and cultural tradi-
tions, and to do so conveniently and inexpensively.
Let’s start with Carruthers’s defense of factory farming. It’s worth
noticing that it isn’t clear that it holds up on empirical grounds, as
there does seem to be a link between cruelty to animals and cruelty to
humans (Linzey, 2009). Perhaps the most disturbing evidence of this is
that rates of domestic violence are higher in communities with slaugh-
terhouses, even after controlling for other variables (Fitzgerald, Kalof,
and Dietz, 2009).3 Moreover, contractors are bound to consider the envi-
ronmental and public health costs of intensive animal agriculture, and
while such considerations certainly wouldn’t lead them to grant animals
rights, it isn’t at all clear that they would sanction factory farms.4 If we
were fully informed, fully rational, and purely self-interested, would we
agree to a contract that sanctions a system that threatens public health
in many ways, sets back our environmental goals, and is objectionably
inefficient—especially when alternatives without these problems are
available? It seems unlikely.
However, there are more fundamental problems. Jennifer Swanson
(2011) points out that Carruthers’s contractualism has some deeply
counterintuitive implications. For instance, if Carruthers is correct, then
the last person on Earth has no reason not to torture animals for fun, as
there’s no chance that he will harm any human beings. Carruthers antici-
pates this kind of objection, replying to it by saying that the problem isn’t
in the risk of harm per se, but rather in the deformation of character; you
aren’t supposed to become a cruel person, and you would have to be a
cruel person to torture animals for fun. However, this reply won’t work.
If you know, as the last person on Earth does, that in performing a certain
action you can’t possibly harm anyone who matters morally, then it’s
very hard to see how your behavior would count as being cruel.5
Perhaps Carruthers might be willing to bite the bullet here. If so, then
no matter, as there’s another issue that is harder to ignore. The crucial
move in Carruthers’s story is that we can use features of our psychology
to draw a line between human and nonhuman animals, where nearly all
instances of the former end up with direct moral standing, and nearly
all instances of the latter end up without it. Alastair Norcross (2000,
140–141) points out that even if we grant all the psychological claims
that Carruthers makes, this method implies that truly trivial differences
between beings can generate enormous moral differences:
To see this, imagine that a new kind of birth defect (perhaps asso-
ciated with beef from cows treated with bovine growth hormone)
produces severe mental retardation, green skin, and a complete lack
of emotional bond between parents and child. Furthermore, suppose
that the mental retardation is of the same kind and severity as that
caused by other birth defects that don’t have the other two effects.
It seems likely that denying moral status to such defective humans
would not run the same risks of abuse and destruction of social sta-
bility as would the denial of moral status to other, less easily distin-
guished and more loved defective humans. Would these contingent
empirical differences between our reactions to different sources of
mental retardation justify us in ascribing different direct moral sta-
tus to their subjects? The only difference between them is skin color
and whether they are loved by others. Any theory that could ascribe
moral relevance to differences such as these doesn’t deserve to be
taken seriously.
This sort of problem is going to affect any form of contractualism that

relies on psychology to draw the line between human and nonhuman ani-
mals, which is essentially the only means available. I take it to be a good
reason to set aside Carruthers’s proposal.
3.2 Human Exceptionalism

Many philosophers haven’t argued directly for eating animals, but have
simply tried to defend the idea that human beings matter more, morally
speaking, than animals. Broadly speaking, we can refer to views of this
kind as “human exceptionalism.”6 These views vary considerably. Some
reject the assumption that moral reasons are agent neutral, insisting that
privileging species membership is akin to privileging family or friends;
just as we have no reason to take up perspectives that would undermine
the partiality we show the special people in our lives, we have no reason
to take up perspectives that would undermine the partiality we show
other humans (Williams, 2006; MacLean, 2010). Others posit proper-
ties that all and only humans have, arguing that they ground our special
moral status (Liao, 2010; Hsiao, 2015). Neither position is speciesist per
se. That notion goes back to Richard Ryder, but is best known from
Singer’s Animal Liberation. As Ryder and Singer use the term, speciesism
is privileging human interests in a morally arbitrary way, in the same way
that racists privilege the interests of members of one race in a morally
arbitrary way. The human exceptionalists aren’t defending that position,
but rather one that gets some of the same first-order normative results. In
any case, will human exceptionalism get us an argument for the permis-
sibility of eating animals?
There are two kinds of challenges for human exceptionalists. The first
is familiar from the problem of species overlap (or “marginal cases,”
as it was once called; see Horta [2014] for arguments against this
phrase): either it isn’t plausible that all and only humans have the status-
conferring property in question, or it isn’t plausible that the property
grounds a special moral status. However it goes, privileging human
beings begins to look unjustified. But the second challenge is the more
serious, at least insofar as defenses of preferential treatment for humans
are supposed to fend off arguments for either the reform or abolition of
animal use. The problem is that it isn’t clear what, exactly, the human
exceptionalism is supposed to imply.
In Singer’s case, by contrast, it’s clear what the rejection of speciesism
is supposed to imply—namely, the failure of one challenge (“But they’re
just animals!”) to his use of the principle of equal consideration of inter-
ests. However, Singer doesn’t need anything so strong to get abstinence
from animal products: animals could matter much less than humans, but
still enough that we can’t justify how poorly we treat them. The upshot
is that absent additional moral principles, human exceptionalism doesn’t
tell us anything about whether it’s permissible to kill, purchase, or con-
sume animals. It merely tells us that we get to give extra weight—though
no one ever says how much—to human interests.
Timothy Hsaio is one of the few who tries to fill in the story here. On
his view, any human “moral” interest trumps any interest that animals
have; the former are lexically prior to the latter. “Moral” interests, he
says, “are welfare interests of members of the moral community. They
refer to things that members of the moral community need in order to
flourish. Non-moral interests are welfare interests of non-moral entities,”
namely, beings who aren’t members of the moral community (Hsaio,
2015, 279). Given this language, he argues as follows:
1. Moral welfare interests trump non-moral welfare interests.

2. Human consumption of meat for the sake of nutrition is a moral
welfare interest.
3. The interests of nonhuman animals in not feeling pain is a non-moral
welfare interest.
4. Therefore, human consumption of meat for the sake of nutrition
trumps the interests of nonhuman animals.
(2015, 280)
Let’s ignore the first and second premises, focusing on the third. The
argument for it is straightforward. The moral community is composed
of those beings with “the root capacity for rational agency” (2015, 286).
Animals lack this root capacity, so they aren’t included in the moral
community. Harms to those outside the moral community are bad for
them, but not morally bad. So, while animals have an interest in not
being harmed, theirs is a non-moral interest. The payoff? Hsaio’s human
exceptionalism implies that any human nutritional interest outweighs all
interests that nonhuman animals have in not feeling pain. Factory farm-
ing is back on the table, and you can eat what you like—as long as it’s
for your health.
There are a number of unattractive features of this view. First, in argu-
ing this way Hsaio complicates the project of looking for a property
that we could be rationally required to acknowledge as grounding moral
status. It may well be the case that his metaphysical theory is true and
that the permissibility of animal consumption falls out of it. But it seems
plausible that disagreement only becomes more reasonable as you com-
plicate the relevant philosophical thesis. It’s hard enough to secure agree-
ment in ethics without adding in our metaphysical differences, so there
should be little hope of expecting agreement once we do. Methodologi-
cally, then, this is a strike against his view: a moral view is better if it’s
plausible that agents ought to believe it, and the more controversial the
assumptions that compose the view, the less plausible it is that anyone
ought to believe it.
Second, Hsaio’s view seems to make too much of trivial benefits. Sup-
pose that people would be ever so slightly better off by eating a diet con-
taining meat rather than a vegan diet by various objective standards (e.g.,
an extremely small difference in lean muscle mass). Suppose that they
would feel the same in the short run either way; moreover, the difference
wouldn’t appreciably affect their long-term health, perhaps because it’s
swamped by other factors. Still, in such circumstances they would have
a nutritional interest in eating meat. Could they justify causing exten-
sive animal suffering for such trivial gains? Presumably not. And yet on
Hsaio’s view, they can.
Third, it isn’t even clear that Hsiao’s view delivers the general permis-
sion for which he means to be arguing. After all, lots of animal products
aren’t consumed for health reasons; they are consumed purely for the
sake of pleasure. If our interest in that pleasure is non-moral, then the
interest that animals have in avoiding pain may well trump the human
interest in gustatory satisfaction. (Hsiao could insist that all pleasure con-
tributes to flourishing, though this wouldn’t sit well with his commitment
to natural law and his views in sexual ethics; see his 2017. But if he
were to make this move, we could run a parallel version of the “making
too much of trivial benefits” argument.) Moreover, though the nutrition
literature is notoriously difficult to interpret, there certainly are some
nutritional considerations that seem to favor well-planned vegan diets. If
that’s so, Hsaio’s conclusion follows in even fewer cases.
The deepest problem, however, is this. Hsaio maintains that all and
only humans have the root capacity for rational agency, and it isn’t sup-
posed to be a potentiality account. Instead, it rests on a metaphysical
assumption—a variety of essentialism. But it’s implausible that a dif-
ference in essence will do the work that Hsiao needs it to do. Imagine
beings who look exactly like human infants, have all and only the cog-
nitive capacities of normal human infants, and yet will never develop
the cognitive capacities that we associate with normal adults. Moreover,
let’s imagine that these beings aren’t human, due to some crucial genetic
differences between us (assuming that we individuate species based on
genetic factors; if we use some other criterion, we can vary the case
accordingly). On his view, these beings are going to lack the root capac-
ity for rational agency—they aren’t the kind of beings that, in normal
circumstances, acquire the relevant cognitive capacities—and so won’t
have moral standing. But of course, if you have one of these beings side-
by-side with a human infant, it would be hard to think that such a dif-
ference is a good reason to think that one deserves protection, while the
other could be killed and eaten permissibly.
3.3 Environmentalism and Agrarianism

Let’s set human exceptionalism aside. We get a very different defense
of animal consumption from environmentalists and agrarians. For the
former, this is often based on rejecting the view that moral standing is
an intrinsic property, offering extrinsic, relational accounts instead. Aldo
Leopold (1949, 224), for example, famously claimed that a “thing is
right when it tends to preserve the integrity, stability, and beauty of the
biotic community.” Here, the biotic community is what matters first and
foremost, so it’s no surprise that killing and eating animals can be mor-
ally permissible—even morally good—since it’s often in the interest of the
biotic community that those animals die. Whatever a biotic community
is supposed to be, it’s plain that predation sustains a good deal of it.
(Indeed, Ned Hettinger [2004] argues that bans on animal consumption
are incompatible with the environmentalist outlook, since they’re incom-
patible with regarding predation as good. This probably isn’t true, as
Jennifer Everett [2004] shows, but it’s indicative of what matters by the
environmentalist’s lights.)
Note that such moral frameworks aren’t going to deliver a defense
of much ordinary animal product consumption. Contemporary animal
agriculture is not winning any environmental awards; it cannot plausibly
be characterized as tending to preserve the integrity, stability, and beauty
of the biotic community. Of course, hunting is going to be permissible,
and probably some small-scale agriculture as well. So you will definitely
get the permissibility of consuming some animal-based foods. But most
of us aren’t choosing between vegetables and game meat; instead, we’re
choosing between vegetables and the wares of factory farms. The upshot
is that although such views generate some permissions, they aren’t blan-
ket ones; nor, for many of us, are they particularly relevant ones.
But let’s ignore this issue, focusing on ecocentrism itself. The main
problem for naïve versions of ecocentrism is that they run the risk of
ecofascism. From the perspective of the biotic community, it may well
be the case that humans are a scourge. Does it follow that most humans
ought to be killed?
J. Baird Callicott solves this problem by subsuming ecocentrism within
a larger, communitarian ethic. This view is
paradigmatically monistic (duties and obligations are generated by

community membership) and practically pluralistic (we are simul-
taneously members of multiple communities—familial, municipal,
national, global, mixed, biotic—and so are importuned by multiple
and often conflicting duties and obligations, which we are obliged to
prioritize for purposes of coherent moral action).
(2015, 61)
According to Callicott, the communities in which we’re most deeply

embedded usually deserve our loyalties first, and more distant spheres
of moral obligation trump the more immediate only when the stakes are
high. So while there may be a general obligation not to support factory
farms, there may also be circumstances in which honoring a host’s hospi-
tality requires eating the factory-farmed meat that he prepared.7
It’s easy to generate stock philosophical objections to views of this
kind. Imagine a very bad family and a generally good community. When
will your obligations to family trump those to your community? If Cal-
licott isn’t careful, he will end up offering a defense of the mob. Or, to go
at it the other way, imagine a good family within a deeply corrupt com-
munity. When are you going to be required to betray a loved one because
the community regards the stakes as being high—as, for instance, in cases
where virgins were selected to be living sacrifices for angry gods? But let’s
ignore these kinds of issues and focus on the ethics of eating in particular.
I suspect that the main trouble with this theory is the difficulty in seeing
what follows from it. True enough, there may be a general obligation not
to support, and a particular obligation to eat—but we could equally well
appeal to community membership to defend the regular consumption of
animal products (“Eating at Cracker Barrel is a family tradition”) and
the rejection of hospitality (“We all care about animals and the environ-
ment here, and I could start a conversation by abstaining”). Moreover,
Callicott says almost nothing about the complex interplay between our
spheres of obligation. It seems plausible that I have duties to my children
not just to look after their well-being, but also to provide them with
moral leadership. Might that often mean privileging the interests of those
far away and vulnerable, as an example to my kids of what really mat-
ters? If so, then the implications of the view are even harder to assess.
Perhaps this line of criticism isn’t totally fair. Philosophers like Leopold
and Callicott are trying to reorient us, not simply develop moral theories.
And their work has some affinities with the agrarian tradition, which also
encourages us to think differently about our lives in relation to the land.
Perhaps the charitable thing to say is that the members of these traditions
are trying to get us to rethink our place in the natural world, and the
kinds of virtues that are called for as members of a larger biotic commu-
nity. Consider, for instance, the work of Benjamin Lipscomb (2015), who
tries to draw together the threads of Wendell Berry’s agrarianism. Berry’s
agrarianism is a “back to the land” philosophy, an outlook that stresses
the character traits required for coaxing sustenance from the earth, living
in small communities, making your own goods, and giving yourself over
to the rhythm of a particular place. Moreover, Lipscomb notes, this view
offers an approach to raising and killing animals on which killing and
eating animals can be virtuous:
[W]e can only live by taking life, and [we might] try to reorient our
thinking about this. . . . [Some] believe, or seem to believe, that the
order by which coyotes prey upon prairie dogs and rabbits and such
is a horrific one, one we should abstain from and perhaps even inter-
rupt. It is a temptation . . . to regard the death by which the world
lives with mere horror—as not the way things are supposed to be.
But to think thus is to be alienated in one’s thinking from the order
Darwin uncovered—the order in and by which we live. [We] might
try to learn to see our condition, not as a merely fallen, but as one we
can inhabit with gratitude.
(2015, 70)
According to Berry, we do have obligations to animals—lots of them,

in fact. But those obligations don’t preclude living in intimate life-
giving and life-taking relationships with them. Indeed, animal husbandry
emerges as a kind of spiritual practice, a way of embracing the natural
order of things.8
Again, there is no defense of intensive animal agriculture here, nor of
consuming the products that come from it. But the more important point
is that even if we restrict ourselves to the kinds of cases that they have in
mind—the slaughter of the sheep that you yourself raised, followed by
eating it around the table with your family—I don’t see why the virtues
so dear to agrarians wouldn’t be enhanced by having less violent relation-
ships with animals. Granted, there may be a lot to be said for living in
small communities that are largely self-governed with a close relation-
ship to the land. But none of that gets you animal death on a human
timetable. You can live with the animals that fertilize your fields, enjoy
their excess milk and eggs, and let them live out their lives as honorary
members of the community. When death is in their best interest, then
eating their bodies is fine, but you certainly won’t be getting much meat
that way.9 The idea of living in a way that embraces the cycle of life and
death is one thing. However, it turns out that for most small farmers, that
cycle includes a bunch of commercial interests, personal preferences, and
cultural norms about the frequency with which consumers would like to
be eating certain products. So I doubt that the cycle is “natural,” in any
robust sense of that word. It looks to me like the sense here is determined
by human interests that, in this case, aren’t in favor of life.
The reply, I suspect, will be that my reservations here amount to the
kind of alienation from which the agrarian is trying to bring us back.
Now it seems to me that we are left with judgments about which aspects
of the world are the ones for which to be grateful and which are the
ones that are to be regretted and fought. Consider, for instance, find-
ing out that your parent has cancer. I understand—and in some respects
admire—the impulse to see even these kinds of tragedies as gifts, to focus
on the way that we can be made more virtuous through them. But there
are, at the same time, clear defeaters for thinking that this is a good fea-
ture of the world. This disease destroys relationships; it causes pain; it so
often damages what’s unique and precious about the person who suffers
from it. If that’s right, then the question just becomes whether there are
similar defeaters when it comes to animals. And it seems plain to me that,
on an individual basis, there are. In other words, unless you stop focusing
on individuals, and start focusing on species, the vast majority of what
we do to animals, and even what they experience in the wild, is flatly bad
for them. If calling this out amounts to alienation, then that strikes me as
a point in favor of alienation.
Moreover, I don’t think that we should so readily concede that it’s
appropriate to describe this aversion to death as alienation. To take that
view seems to concede that there are certain virtues that can’t be devel-
oped without embracing (early) slaughter. This seems implausible on its
face. But if there are any such virtues, it’s worth considering whether this
is because some virtues are bound up with particular identities. Consider
Christopher Ciocchetti’s (2012) fascinating examination of identity-
based defenses of meat consumption in which he considers the possibility
that our identities provide us with special reasons to act. Ciocchetti con-
cludes that our identities are often more flexible than we think, and we
can breathe new life into traditions when we bring them in line with our
moral convictions, so identity-based defenses aren’t successful.10 (Jona-
than Safran Foer’s Eating Animals is an excellent example of this: on
one level, the entire book is an attempt to explain why he won’t eat his
grandmother’s chicken and carrots, which is at the heart of his family’s
gatherings, and then to refashion an identity that’s compatible with opt-
ing out.11) In the face of Ciocchetti’s work, the task for agrarians is to be
more precise about why slaughter is essential to the virtue(s) they value,
and why those identities can’t be reformed in a way that treads a bit more
lightly on the earth.
3.4 Taste
Perhaps all these arguments have just conceded too much to the propo-
nent of veganism. Have we understated the costs of giving up meat?
This is Loren Lomasky’s (2013, 178) concern: he insists that “eating
meat contributes to a very great good for human beings without imper-
missibly impinging on animal well-being.” The “very great good” here is
the aesthetic-cum-gustatory pleasure associated with meat consumption
and its associated traditions. And he maintains that this can be had with-
out impermissibly impinging on animal well-being because (1) he thinks
that we couldn’t get this very great good any other way, (2) he denies that
animals have rights, so we aren’t infringing them, and (3) he thinks we
just don’t know whether animals suffer very much, so welfare considera-
tions don’t trump our interests.
Lomasky thinks that this justifies even eating products from factory
farms, but as discussed in the last chapter, the claim about our knowl-
edge of animal suffering is false. And while there are indeed pleasures
distinct to meat-consumption, there is room for debate about whether
they’re qualitatively superior, which is what Lomasky needs. Lomasky
tries to address the latter problem by appealing to Mill’s competent judge
test, insisting that, in general, those who’ve tried both vegetarian and
omnivorous diets acknowledge the superiority of the latter: “All across
the globe the same phenomenon is observed: as incomes increase so does
the amount of meat in people’s diets” (2013, 185). Michael Gill (2013)
objects that we shouldn’t take members of the general population to be
competent judges, since they haven’t experienced the best that vegetar-
ian cooking can offer. However, I’m inclined to concede this point to
Lomasky. There are gustatory costs to giving up animal products12—
which, of course, leaves open whether they’re costs we should bear.
Jean Kazez (2018) offers a much better—but also far more modest—
version of the taste-based challenge. There are two key moves. First, she
notes that people value food much more than proponents of veganism
sometimes recognize. For instance, people are willing to make enormous
sacrifices for food, quite literally giving up years of their lives to consume
the products that they enjoy the most. That is, many people judge certain
pleasures to be, in the aggregate, more valuable than entire human life
years. Second, she questions how we should interpret the “necessity” in the
claim that only necessary harm to animals can be justified. She writes this:
Imagine a sequence of vegan diets that are a person’s only alternative

to an omnivorous, harmful-to-animals diet. Make the diets gradually
less and less enjoyable. The flavour decreases. The smell gets unpleas-
ant. The texture of food becomes slimy. Imagine that that unpleas-
ant diet is the only alternative to an omnivorous, harmful-to-animals
diet. It doesn’t take too much of a decrease in palatability before just
about anyone will do what they must to avoid it. . . . If they apply the
Necessity Principle, they’re likely to say that avoiding the unpleasant
food is important enough to be worth causing harm to animals.
In fact, taste seems significant even before we get to the outright
unpleasant point in a series of possible diets, from delicious to dis-
gusting. Suppose a perfectly ethical food were available—PEF, for
short. PEF causes no harm at all to animals, no harm to the environ-
ment, no harm to people, etc. Imagine that it looks and tastes like
oatmeal but has all the nutrients a person needs. So it’s pleasant tast-
ing, nutritious, and not disgusting. Ethicists capture you and put you
on PEF for a month. They then release you with a life-time supply.
Once again, just about anyone armed with the Necessity Principle
will say harming animals to avoid PEF is necessary.
(670)
Obviously, these aren’t the choices that we face, as Kazez well recognizes.
But I tend to agree with her intuitions about what most people would say
in these circumstances. What follows from this, morally speaking?
According to Kazez, not much. She grants that many people could
do far better when it comes to eating in more animal-friendly ways; she
grants that they probably should. But she also imagines someone who is
willing to make significant sacrifices for animals, to the point of eating
PEF instead of risking any harm to them. She concludes that
there’s something a bit dubious about [the person willing to give up

so much for animals]. She has become a ‘moral saint’ in Susan Wolf’s
negative sense. You might say that she enjoys the pleasures of food
too little and moralizes too much. She’s too willing to wear a hair-
shirt for the good of others.
(673)
I sympathize with Kazez’s assessment here, and I also think that she is right
to be very cautious about the practical upshot. It takes a case of this kind
to make it plausible that it’s a mistake to rely on PEF. When we consider
more ordinary ones, the reason that taste generates is weaker, perhaps only
rarely overriding the reason that we have to abstain from eating animals.
However, I discuss Kazez here not because she has a compelling case for
the moral permissibility of consuming animal products in ordinary circum-
stances, but because she makes it clear that taste—among other human
goods—isn’t completely trivial, and when we assess the merits of an argu-
ment for veganism, we will need to consider the strength of the reasons
that it generates. If the reasons are strong, then they will likely overwhelm
the reasons of taste. But if they’re weak, then perhaps not.
3.5 Health
If the significance of taste is controversial, the significance of health gen-
erally isn’t. Many people—defenders of veganism among them—have
thought that if your health would suffer as a vegan, then you aren’t
obliged to be one. The dispute, then, tends to be about whether your
health would indeed suffer. Proponents of omnivorism tout the health
benefits of animal products; defenders of veganism cite the health benefits
of a plant-based diet. The conversation quickly devolves into my expert
versus your expert, and progress is rarely made.
However, even if we grant that there is a strong presumption in favor
of either protecting or promoting your health—which are different stand-
ards, but let’s ignore that here—we should note that the presumption can
be overridden. Consider a case where you need to sacrifice your health
to care for a sick family member. You can still be morally obligated to
make the sacrifice, despite the personal cost. Moreover, this can be the
case even if the cost is long-term. Consider someone who’s in the position
of needing to care for a chronically ill child or partner or parent. This
might require losing sleep, foregoing opportunities to exercise, eating sig-
nificantly less healthily than you might otherwise, and exposing yourself
to greater risk of infection—just to name a few significant and all-too-
common burdens. But that you would bear such burdens, even over an
extended period of time, seems perfectly compatible with there being a
responsibility to provide the relevant care. The upshot of this is that we
face a line-drawing problem. How much can be required of you?
We might try to answer this question by analogical arguments, looking
at cases not involving animals or health to avoid any biases that we might
have in either domain. During World War II, for instance, there were
drives that attempted to collect various products—for instance, rubber
and scrap metal—that could be recycled for the war effort. There was a
strong sense at the time that people ought to donate to those drives, even
at some personal cost. Might we make an analogy here, and say that if
people were morally required to give in that context, then some conclu-
sion follows about the consumption of animal products?
Perhaps, but I’m not optimistic. I don’t think that there’s any way to
dodge questions about the difference you can make by not consuming
animal products, the relevant health risks, the moral importance of ani-
mals, the seriousness of the wrongs perpetrated against them, and so on.
This isn’t the place to address many of those questions, as they’ll come up
later. So let’s just focus on the relevant health risks.
At this juncture, proponents of veganism tend to cite evidence that
plant-based diets lower risk of all-cause and cause-specific mortality
(Bouvard et al., 2015; Song et al., 2016; Springmann et al., 2016), and
point out that, as a result, a number of major health organizations now
either support or actively recommend them (American Dietetic Asso-
ciation, 2009; Mayo Clinic Staff, 2017; American Heart Association,
2016). Indeed, one major insurance provider is encouraging physicians
to suggest plant-based diets to patients, arguing that it offers a cost-
effective and low-risk intervention to lower BMI, blood pressure, and
cholesterol levels, in addition to reducing dependence on medications to
treat various chronic diseases (Tuso et al., 2013). So, rather than there
being a health-based objection to veganism, health-based considerations
support it.
But this is too quick. What, exactly, do the health studies show? Con-
sider a hypothetical study that reports the following result: those on a
plant-based diet are at 50% less risk of heart attack than are omnivores.
What’s the basis for that claim going to be? Very likely, it’s going to be
something like this: the researchers took a look at a large data set that
was collected for some other purpose; they searched for the individuals
who report that they follow a plant-based diet; they compared the health
outcomes for that group with a randomly selected group of omnivorous
individuals from the data set; they found that half as many vegans expe-
rienced heart attacks; they ran some regressions; they found that dietary
difference explains the majority of the effect.13
Studies like this tell you about average effects, and those averages
obscure individual variation. Still, average effects can do a lot of work.
Studies on the benefits of a plant-based diet can provide you with a rea-
son to abandon your prior conviction that animal products are essential
parts of a healthy diet. They can provide you with a reason to try out a
plant-based diet. They can give you a reason to discount any initial dif-
ficulties that you may have on a plant-based diet. This alone is enough
to make it plausible that when people offer health-based considerations
against veganism, they are sometimes offered out of ignorance or bad
faith.
However, studies about the benefits of a plant-based diet can’t estab-
lish that you will be better off (or at least won’t be worse off) on a vegan
diet whatever the details of your physiological makeup, medical history,
environment, and so on. Nor are they designed to establish that con-
clusion. However, the point at issue is whether you have an obligation
to be vegan, and insofar as health is relevant, it’s this last thesis that’s
the important one. So if you make a good-faith effort at maintaining a
healthy vegan diet and continue to experience unacceptable health con-
sequences that are plausibly traced to diet, then you may not have an
obligation to be vegan. But we need to emphasize “may” in that last
sentence. Until we assess the kinds of costs that you ought to take on
for the sake of animals, we can’t say anything about whether it’s mor-
ally permissible for you to abandon your veganism. All we can say is
this: pro-vegan nutrition studies become increasingly irrelevant to your
judgments about what’s in your best interests healthwise. (It’s easy to
imagine people like this. Consider, for instance, someone with allergies
to gluten and soy; or with a disease that requires a high intake of calories
generally and fat specifically, such as cystic fibrosis. Plainly, these people
don’t make up the majority of the population, and so I don’t mention
them to suggest that most people aren’t obliged to be vegan. Instead, they
serve as reminders that the health studies only show so much.)
The upshot is this. The health-based argument for eating animal
products isn’t particularly good, for two reasons. First, even if there are
health-based costs to plant-based diets, they don’t show anything until
we say something reasonably precise about the kinds of costs that you
ought to take on for the sake of animals. At the same time, we should
acknowledge that this point cuts both ways. If there are costs, then the
moral relevance of those costs depends a great deal on the strength of the
reasons we have to abstain from consuming animal products. If those
reasons are weak, then health considerations may well be decisive; if not,
then not. Second, nutrition studies do generally seem to cut in favor of
plant-based diets, which is a good reason for people to try them quite
apart from any costs that we ought to take on for the sake of animals.
3.6 Ageism, Classism, Sexism, Racism

I’ll consider one last argument for the permissibility of eating animals. At
first blush, it looks like another version of the health-based argument, but
there is much more to it than that. Kathryn Paxton George (1994, 21)
summarizes the worry in the following way:
In arguing for vegetarian or vegan diets on traditional moral grounds,

philosophers suppose that all humans have the same bodies and that
the burdens are relatively evenly distributed—we are all pretty much
the same physiologically and differences among us may safely be
ignored as morally irrelevant. Both assumptions are false. Women
and men are not the same physiologically. Children and adults are
not the same. The aged and the young are not the same. Indeed, in
at least a few respects relevant to food practice, races and ethnicities
are not the same. The empirical evidence for this has been a recent
topic of debate. . . . What all my opponents center on is risk. But
risk is not the primary issue. What is at issue is this: From whose
perspective shall we assess risk? From the perspective of predomi-
nantly white, adult Western males? Or from the perspective of Third
World women or their children? Shall we ask these questions from
the perspective of the powerful or the vulnerable? The answer to
such questions depends upon whom you think ethics is meant to
serve. If ethics is meant to protect the rich and the powerful, their
rights and privileges, then perhaps you will not agree with my argu-
ments. . . . But I believe that ethics must consider the concerns of the
vulnerable and that some attempt must be made to draft an ethics
which accounts for and concerns itself with the perspectives of the
very differently situated persons and animals with whom we share
the earth. Only in that way will ethics truly be for everyone.
Obviously enough, George is claiming that it’s permissible for mem-

bers of vulnerable populations to eat animal products because vegan-
ism places excessive burdens on them. But the claim of special interest is
that it’s a mistake to think that there is a general obligation to be vegan
because bodies and contexts are different; a general requirement to be
vegan doesn’t take all this variation into account. Of course, there’s a
standard reply here, which is that that those individuals are excused from
the general obligation because of their specific circumstances. George
regards this as unacceptable:
When nearly everyone on Earth, except the most privileged class of

humans, is excused from living the most virtuous life on the grounds
that their natures or their cultures bar them from it, we should sus-
pect that the moral tradition itself has been set up to serve that class
and that class alone. . . . As it is now, the vegan ideal would discrimi-
nate against other socioeconomic classes: there is ethnocentric bias
in the vegan ideal; it presupposes power; it entwines power with a
false virtue. Why? Because the moral command to a vegan diet pre-
supposes a wealthy society and an industrialized culture with open
access to information and a wide variety of food, education, and
medical care. These privileges form a web within which a vegan life-
style is a healthy alternative. Vegans must supplement their diets with
certain vitamins, especially B12 and D, and infants, children, women
and others may need supplementation for certain minerals, especially
calcium and iron.
(1994, 23)
The upshot is as follows. Typically, people argue for veganism by claim-

ing that, relative to the enormous harms associated with animal agri-
culture, animal products provide relatively trivial benefits to us; so, we
ought to abstain. But it matters a great deal who “we” are in this argu-
ment. George doubts that the argument goes through even for fairly well-
off white males in the Western world. However, she certainly doesn’t
believe that it goes through for more vulnerable persons. And crucially,
she thinks that promoting veganism as an ideal creates a kind of moral
underclass, full of people with suboptimal bodies that can’t support them
in fulfilling their ostensible moral obligations. This, she claimed, is dis-
criminatory, so we should reject veganism as a general moral obligation.
This line of argument deserves serious consideration, but it’s difficult
to assess. The main problem is that it isn’t obvious how to distinguish
cases of discriminatory ideals from non-discriminatory ones. After all, we
might think that it’s often going to be the case that relatively well-off peo-
ple in the Western world have obligations that most people in the world
don’t have, based on the idea that privileges come with corresponding
responsibilities. Why should this case be any different?
The reply might be that there is an important difference between
explicitly framing an obligation in terms of a corresponding privilege,
on the one hand, and framing an obligation as a general one that can be
excused, on the other. I grant that the rhetorical difference is obvious,
but I’m a lot less clear on the normative difference. Presumably, there
aren’t species of moral oughts. There are just the moral oughts there are,
which are always sensitive to the capacities of the persons to whom they
apply. (We can express those oughts differently, and I grant that some of
those expressions are morally better and worse. However, our assessment
of the expression can—and should—be distinct from our assessment of
the purported obligation that’s being expressed.) Moreover, given that
ought implies can, we should expect very significant differences between
the oughts that apply to the privileged and the oughts that apply to the
disadvantaged. And where people can’t comply with those oughts, there
are no oughts at all—it isn’t that they are excused; it’s that there’s no
obligation in the first place.
3.6.1 Racism Without Racists

At this juncture, it may be helpful to introduce a different version of
this objection. Luis Cordeiro-Rodrigues (2017) levels a similar charge
against a general duty to be vegan, providing some useful tools for think-
ing about the precise nature of the critique.14 His stalking horse is Gary
Francione’s abolitionism, according to which we ought to abolish the
institutions and practices that support the exploitation of animals. This
view has two components: the first, which is moral, insists that animals
are right holders who deserve equal consideration to human beings; the
second, which is empirical, insists that the best way to move toward the
rights of animals being respected is not to push for welfare reforms and
the gradual reduction of animal use, but to advocate consistently and
exclusively for the end of that use. In particular, this means advocating
for veganism, which Francione regards as the “moral baseline”—that is,
he thinks it’s morally required of nearly everyone in the developed world
(and many beyond it).15 Cordeiro-Rodrigues argues that Francione’s view
is guilty of “an unintentional and subtle form of racism” (2017, 745). So,
Cordeiro-Rodrigues is criticizing exactly that to which George objects—
namely, a vegan ideal. But his criticism is based on “the new dynamic
of racism,” a phrase that Eduardo Bonilla-Silva (2015) uses to describe
the form that racism takes after explicit prejudice becomes socially unac-
ceptable, though many people still desire (perhaps without realizing it)
the world that explicit prejudice created. As Bonilla-Silva understands it,
racism is above anything, about practices and behaviors that produce

a racial structure—a network of social relations at social, political,
economic, and ideological levels that shapes the life chances of the
various races. This structure is responsible for the production and
reproduction of systemic racial advantages for some (the dominant
racial group) and disadvantages for others (the subordinated races).
(2015, 1360)
So the new racism isn’t about explicit negative attitudes and beliefs about
racialized groups. Instead, it uses more subtle methods to disempower
historically disadvantaged peoples. In particular, it uses the structure just
mentioned to unfairly blocks and/or burden their ability to access the
means necessary to meet their fundamental interests, as well as to con-
strain the options of individuals with respect to fundamental aspects of
their lives. These include the familiar kinds of cultural, economic, and
institutional means by which racialized groups have had their basic inter-
ests set back: cultural marginalization, the preservation of economic ine-
quality, and the ways in which institutions cater to white individuals to
the detriment of others.
If we employ this framework, we may be able to get a better sense of
why we should regard a vegan ideal as discriminatory, just as George
claims. The thought will be that a vegan ideal is a norm that contrib-
utes to the production and reproduction of systematic racial advantages
and disadvantages—and, presumably, systematic gender advantages and
disadvantages, age-based advantages and disadvantages, and so on. So
instead of focusing on the justification of the norm, and the question of
whether a particular person is excused for failing to comply with it, we
should instead focus on the effect, at the macro level, of there being such
a norm.
We are now in a position to appreciate the way that, like George,
Cordeiro-Rodrigues sees veganism as discriminatory. On his view, a
vegan ideal forces racialized peoples to make a difficult choice. On the
one hand, they can go vegan and suffer various negative health conse-
quences (because they don’t have access to inexpensive and nutrition-
ally adequate vegan fare. On the other hand, they can fail to go vegan
and thereby become members of a “moral underclass,” that is, “being
negatively classified and/or socially perceived as people [who], because
of personal failings and an inferior nature, are incapable of being fully
moral” (751).
This objection rests on some empirical claims, some of which Cordeiro-
Rodrigues states explicitly and some of which are implicit. One of the
implicit claims is that people who can’t go vegan are at risk of facing
certain reputational consequences: they will judged by others to be
morally substandard. Of course, it seems plain that proponents of the
vegan ideal don’t intend to relegate certain people to membership in
a moral underclass. How, then, should we assess whether that’s what
they’re doing, or whether that’s what the position does independently
of what any advocate does? Cordeiro-Rodrigues doesn’t say, and it isn’t
entirely obvious how to answer these questions. I would expect, though,
that we would have good evidence for guilt in one respect or another if
there were reason to believe that a vegan ideal affects how members of
racialized groups are perceived. That is, suppose that when individuals
become convinced that people should go vegan, they come to see mem-
bers of racialized groups as part of a moral underclass, and lose respect
for them as a result. That would be bad in itself, and may well result in
further harms.
However, the fact remains that Cordeiro-Rodrigues offers no evidence
that it has this effect. Instead, he argues that certain racialized groups
have been the victims of negative stereotypes in the past, and that this
has had, and continues to have, negative consequences for those groups.
Unfortunately, that’s irrelevant to the question of whether this particu-
lar ideology would have such consequences. And if we look at relevant
parallels, such as the environmental movement, we find reason to think
that perceptions wouldn’t generally develop in that way. For instance,
there isn’t a general consensus that members of poorer communities are
part of a moral underclass in virtue of not buying carbon offsets, or that
they are worse people for not buying electric cars, or that they are some-
how inferior when they recycle at lower rates. Instead, there seems to
be wide agreement that the burdens of poverty excuse these suboptimal
behaviors, and moral pressure tends to be directed toward the relatively
wealthy. We might expect exactly the same thing when it comes to a
vegan ideal. Where people genuinely can’t switch to a plant-based diet,
their behavior would be regarded as excusable. And where they could but
don’t, their behavior would rightly be regarded as morally problematic,
which is exactly as it should be—at least on the assumption that there’s a
general obligation to be vegan. So, it seems to me that we shouldn’t allow
concerns about creating a moral underclass to guide our assessment of
a vegan ideal, and, even if we do, it isn’t obvious that we have enough
evidence to accuse it of having this consequence.
Additionally, we need to qualify some of the explicit empirical claims.
For instance, Cordeiro-Rodrigues claims that certain racialized groups
have more difficulty, physiologically, transitioning to a vegan diet. Let’s
grant that this is true. Even if so, though, nothing follows until we take
into account the various health benefits of a vegan diet, discussed earlier.
After all, even if there are notable risks for certain populations, those
risks could still be outweighed by other health gains. For instance, the
very diseases that pose the greatest threats to certain racialized commu-
nities in the US—namely, heart disease and diabetes—are the ones that
plant-based diets are best equipped to address (CDC, 2017). So instead
of setting up a racial disadvantage, it may turn out that vegan diets can
be part of a strategy for resisting food environments where powerful cor-
porate interests flood the market with inexpensive, high-fat, high-sugar
products. Indeed, that’s precisely what a number of black vegans have
argued (e.g., Danielle, 2009).16
I take no stance on whether there are net benefits or costs of a vegan
diet for the relevant racialized communities; that’s a matter I’m not in a
position to assess. However, the point here is just this: if we should assess
whether a vegan ideal is discriminatory based on its consequences, then
we shouldn’t say that it’s discriminatory based on the arguments consid-
ered so far.
That said, Cordeiro-Rodrigues has two other arguments for thinking
that a vegan ideal is an instance of the new dynamic of racism. One
of them is that a vegan ideal requires that some racialized communi-
ties increase their food insecurity: some blacks and Latinos live in food
deserts, where access to vegan staples is limited, and they lack the eco-
nomic and temporal liberty to shop in places with better access.17 The
other argument is that abolitionism would create a kind of “racist power
hierarchy,” since the switch to vegan diets would bolster a neocolonial
economic system that already disadvantages people in Africa and Latin
America. I’ll argue that these arguments don’t fare any better.
3.6.2 Limited Access

Cordeiro-Rodrigues insists that various racialized communities have lim-
ited access to the foods necessary to flourish on a vegan diet and lack the
economic means to do much about it. As a result, they would need to
accept either serious health or financial burdens to switch to a vegan diet.
Since that burden falls disproportionately on racialized communities,
we have the kind of systematic disadvantage that’s relevant to whether
there’s an instance of racism without racists.
This sort of objection rests on empirical claims to which I’ll return
in a moment. But before I say anything about them, there is a more
fundamental issue that needs to be addressed: namely, that Cordeiro-
Rodrigues’s argument confuses a larger burden with an unjustified burden.
For the sake of argument, let’s grant that certain racialized communi-
ties would need to accept either serious health or financial burdens to
switch to a vegan diet. Still, the question shouldn’t be whether others
would bear less serious burdens; instead, the question should be whether
the burdens on members of certain racialized communities are so seri-
ous that they outweigh the reasons favoring the switch. If they are, then
there is no obligation to change, and that should undermine the charge
of racism. (If a vegan ideal imposes no obligations on you, then assuming
the arguments of the last section, it also imposes no burdens on you, and
so can’t be accused of unjustly disadvantaging you.) If, on the other hand,
those burdens aren’t so serious that they outweigh the reasons favoring
the switch to a plant-based diet, then there is such an obligation, onerous
though it may be for some people.
It’s important to see that this is an instance of a general phenomenon.
It’s tragic that morality effectively demands more of some people than it
does of others. However, the tragedy is not the moral demand, but the
circumstances that create differential impacts. To appreciate this, note
that wealthy people are affected much less severely by the moral wrong-
ness of stealing than are the poor—that is, it’s much less costly for the
wealthy to obey the anti-theft norm than it is for the poor to obey it—and
the poor are disproportionately black and Latino. After all, the wealthy
can just buy what they want, whereas the poor have to go without if they
are morally prohibited from taking it. Plainly, though, it doesn’t follow
from this that the norm against stealing is racist. It is, of course, bad that
the poor are poor, especially when this affects their fundamental inter-
ests, and even worse when poverty tracks arbitrary racialized groupings.
However, all this badness either justifies stealing or it doesn’t. If it does,
then those who steal in such circumstances act permissibly; if it doesn’t,
then they act wrongly, even if morality asks more of them in this respect.
Of course, morality doesn’t impose all the bad circumstances faced
by members of many racialized groups. Differences in wealth, scarce
employment opportunities, and limited social mobility stem from a leg-
acy of injustice, subtle forms of discrimination (both past and present),
and the like. It isn’t as though white people, as a class, have some special
claim to better-than-average paying jobs, to which they can appeal to
justify their privileged position. As a result, the structures that support
these systematic benefits can rightly be condemned as the new racism;
they have no moral justification, and members of racialized groups have
a claim to a better circumstances than the ones in which they find them-
selves. The upshot: even if it’s always bad that some groups bear greater
burdens than others, it doesn’t immediately follow that those differences
amount to an unjust or otherwise immoral systematic advantage. It’s rac-
ism without racists if members of certain groups have a moral claim not
to bear those particular burdens; it isn’t if they don’t.
When we appreciate this point, we can see the basic problem with
extending Bonilla-Silva’s framework to the vegan ideal. If the harms
associated with intensive animal agriculture show that ordinary consum-
ers have an obligation to be vegan—which I’ll take for granted for the
moment, just for the sake of argument—then the critic should also con-
cede that people who could change, but opt not to, are acting in a way
that’s morally wrong.18 And if that’s right, then the critic should concede
that morality imposes a burden on us all—though, admittedly, one that
weighs more heavily on some. But the burdens imposed by veganism
are motivated by weighty moral considerations—namely, serious harms
to billions upon billions of sentient beings. So although the vegan ideal
creates a structural imbalance that tracks racialized groupings, we don’t
for that reason have racism without racists. The greater burdens on some
remain tragic, but it seems strange to make that tragedy the fault of the
obligation not to eat animals. As with the obligation not to steal, the
tragedy stems from the material conditions of the relatively disadvan-
taged group and the various factors that created and preserve them.
So any criticism of the vegan ideal needs to be clear about whether it
grants the moral claims that (ostensibly) justify the ideal. If so, then it will
be hard to make the charge of racism stick. And if not, then there isn’t
much of a criticism: it’s no surprise that the vegan ideal looks extreme
given, say, the assumption that animals don’t matter nearly as much as,
or in the way that, many vegans think they do. That is, if animals were
minimally morally considerable—where almost any human interests, no
matter how trivial, could outweigh a violation of even animals’ weightiest
interests—then members of racialized communities (among others) wouldn’t
be acting wrongly in consuming animal bodies and byproducts. A fair point,
but not one that any proponent of the vegan ideal needs to deny.
But even if we set all this aside, we should notice that Cordeiro-
Rodrigues is mischaracterizing the problem here. Let’s return to the choice
that Cordeiro-Rodrigues take to confront racialized communities—
namely, between either health burdens or financial burdens. I submit that
we should be skeptical about the associated empirical claims, at least
when focusing on the US and other developed countries. (I’ll turn to the
Global South shortly.) These claims are, first, that many members of
racialized communities lack local access to the foods necessary for flour-
ishing on a vegan diet, and, second, that it would be expensive for them
to secure them. Cordeiro-Rodrigues cites the American Dietetics Associa-
tion, which states that vegans need to have (1) a range of fruits and veg-
etables every day, (2) legumes and grains, (3) vitamin B12 supplements,
and (4) sunlight for the sake of producing vitamin D. Note, however, that
doesn’t say that people need to have access to fresh fruits and vegetables
every day; canned and frozen are just fine. (Indeed, frozen fruits and veg-
etables often have a superior nutritional profile, since unlike fresh ones,
they don’t lose certain nutrients in the period between being harvested
and being consumed.) Access to canned and frozen fruits and vegetables
is quite good in the US, and is certainly good enough to meet people’s
nutritional needs. (We shouldn’t confuse limited access to ideal options—
which is real—with limited access to adequate options—which generally
isn’t.) What’s more, it’s relatively inexpensive to have B12 supplements
sent to a PO Box (if the mail is unreliable at home); or, it’s always pos-
sible to take multivitamins that include B12, which are available at every
pharmacy and grocery store. Finally, the vast majority of people in the
US have no trouble finding legumes, grains, and sunlight. So, the access
claim is false. Likewise for the financial claim: first, because omnivores
need many of the same things; second, because even after factoring in
supplements, all these options are generally cheaper than animal-based
alternatives, troubling any argument from economic hardship.19
The immediate objection will be that my claims ignore the realities of
“food deserts.” This term is operationalized in different ways, but the
USDA gives this as one interpretation of a “low access” census tract:
Low-income census tracts20 where a significant number (at least 500

people) or share (at least 33 percent) of the population is greater than
½ mile from the nearest supermarket, supercenter, or large grocery
store for an urban area or greater than 10 miles for a rural area.21
And if we accept this usage, then roughly 54.4 million people (17.7% of

the population) live in food deserts. But we need to get perspective here.
Yes, many people live more than a half mile from the nearest grocery store
(or more than ten miles away in rural contexts). This means that some
people will find it difficult—due to mobility issues, or caretaking respon-
sibilities, or time constraints, or what have you—to go to a grocery store
over a closer option. However, “difficult” doesn’t mean “impossible,” at
least for the great majority of those 54.4 million people.22 So again, the
question isn’t about access per se. It’s about access relative to (1) certain
expectations about the burdens that food acquisition should involve and/
or (2) the strength of the reasons that favor individual dietary change.
On that latter point, let’s observe that if food deserts excuse the peo-
ple in them from a duty to be vegan, then that’s because the moral rea-
sons favoring individual dietary change aren’t very strong. In subsequent
chapters, I defend something like this. But Cordeiro-Rodrigues isn’t sup-
posed to be assuming that, for instance, individual purchasing choices
make no difference, so those in food deserts can do what’s most conveni-
ent (which isn’t my view). Instead, he’s supposed to be granting that the
case for veganism is strong enough to generate individual obligations
even for people in food deserts, which is what allows him to criticize a
general duty to be vegan as imposing an excessive burden on racialized
communities. However, if it’s true that the case for veganism is so strong,
then it’s also true that we ought to take on lots of costs for nonhuman
animals. Presumably, those costs might include traveling longer distances
and spending more time and money to find animal-free foods.23
To be clear, I don’t want to minimize the claim that there are burdens
associated with vegan diets. It’s just that Cordeiro-Rodrigues overstates
some burdens while neglecting others. Access and price are exaggerated
problems, whereas foodways are ignored—namely, the social and eco-
nomic infrastructures that make it easy to eat in some ways and difficult
to eat in others.
After all, the things you need to have a nutritious, plant-based diet
aren’t particularly difficult to find if you know what to look for, and
they aren’t terribly hard to prepare if you know how to prepare them.
But most people don’t know what to look for, nor how to prepare the
relevant dishes. Moreover, they generally have to know these things to
eat well. Animal products can obviously be part of a healthy diet, sup-
plying a wide range of macro- and micronutrients that, while generally
available in plant-based diets, may not be available in the relevant quan-
tities when people simply replace chicken with pinto beans or fish with
lentils. Moreover, most people can’t go to inexpensive restaurants and get
nutritionally adequate vegan options: they will get options that are both
protein and B12 deficient, since if inexpensive places offer vegan options
at all, they typically do so by removing animal products, not by plan-
ning entirely new meals that are designed for people who never consume
animal products. The result is that as the US food system is currently
configured, it’s certainly easier to get your calories and nutrients through
animal products than it is to get them through plant-based alternatives.
And depending upon the pace of life, and the other burdens you bear, a
difference in ease can be a significant difference. It takes time and a great
deal of effort to learn how to flourish as a vegan, and we shouldn’t dis-
count that.
Moreover, the transition will be more frustrating for members of some
groups based on the food they prefer. There is no good vegan substitute
for chitlins; tempeh “ribs” are not ribs. We shouldn’t ignore the loss of
foods that are filled with personal and cultural significance. Nor should
we idolize them. As many anti-sexism and anti-racism educators have
observed, celebrating the positive aspects of the status quo, and nostalgia
for the world that once was, can be manifestations of resistance to moral
change.24
Do any of these considerations show that a vegan ideal is discrimina-
tory, or that it’s permissible to be an omnivore? I don’t think so. Granted,
we are sometimes worse off when we internalize the costs that we previ-
ously externalized onto others. But if there are good moral reasons in
favor of internalizing those costs, then our loss is unlikely to override
them. That said, as those reasons become weaker, the calculus changes.
So yet again—as we saw with the discussions of taste and health—it mat-
ters a great deal that we get clear on the strength of the reasons that sup-
port there being an obligation to be vegan.
3.6.3 The Neocolonial Economic System

Cordeiro-Rodrigues’s final objection to abolitionism is that he thinks it
would bolster a neocolonial economic system that already disadvantages
people in Africa and Latin America. These countries may not produce the
kinds of supplements and plant-based foods that vegans require, and so
they would be dependent on imports from wealthy Western trading part-
ners that, historically, have taken advantage of their powerful economic
positions.
Before saying anything else, we should be clear that poverty in the
developing world is dramatically different from poverty in the developed
world. In South Sudan, for instance, at least 80% of the population lives
on less than one dollar per day, meaning that people have virtually no
choice about what to eat (UNDP, 2017). In the US, by contrast, essen-
tially no one is that poor, as nearly everyone has access to social services
that provide resources far above that level (which isn’t to suggest that
there aren’t serious problems with the social services available).25 As a
result, it really is plausible that many people in parts of Africa and Latin
America lack the means necessary to eat vegan; for those people, the
choice is between animal products and starvation. Ought implies can,
and they can’t. However, nothing similar can be said for most people in
the developed world, including most members of racialized communities.
So, it’s a mistake to run together the plight of people in the developed and
developing world, as Cordeiro-Rodrigues does, when trying to assess the
burdens on each group that are relevant to their ostensible obligations to
eat a plant-based diet.
But now to the objection itself. The main problem, as with the health-
related objection discussed earlier, is that Cordeiro-Rodrigues fails to
consider the broader context in which any risks are taken. With the
health-related objection, the error was that Cordeiro-Rodrigues didn’t
consider the various health benefits of eating a plant-based diet. With
the neocolonial objection, the mistake is not considering the economic
benefits that come along with global trade. Deirdre McCloskey (1991, 1)
quips that we can sum up economic history as follows: “Once upon a time
we were all poor, then capitalism flourished, and now as a result we’re
rich.” And the point doesn’t just apply to whites in the developed world.
In 1800, over 80% of the world lived on less than two dollars per day.
Now, over 80% of people live on more than two dollars per day, with the
average being a bit over six dollars per day, and global income inequality
has actually declined (Roser, 2016). Obviously, there is a sense in which
an average of just over six dollars per day is not something to celebrate:
it’s far less than it ought to be. However, we shouldn’t lose sight of the
fact that free markets have been responsible for an enormous improve-
ment in the global standard of living. To maintain anything else would
simply be to ignore the consensus among economists. And with that in
mind, we should think quite differently about the economic dependence
that Cordeiro-Rodrigues decries. Framed as “dependence,” it sounds
objectionable. But framed as the means by which most human beings are
being pulled out of poverty—and, in this case, spared from supporting
the exploitation of animals—the economic relationship doesn’t seem so
pernicious. To be clear: I am not suggesting that all trade deals are fair,
that food sovereignty isn’t valuable, or that we have any reason to praise
colonialism. My point is just that the economic relationships that make
diverse foods available to many people are part of a larger economic
system that, all things considered, benefits members of the Global South.
The upshot: it may indeed be true that global trade will be necessary to
make veganism possible across the board. But since global trade is, on
balance, a good thing for most people in developing countries, this isn’t
an obvious problem.
George argued that a vegan ideal is discriminatory. I don’t think her
argument for that conclusion works, and so I’ve examined the way that
Cordeiro-Rodrigues tries to reach a similar conclusion—namely, that
when we focus on its effects, we see that it perpetuates systematic racial
advantages. As far as I can see, his arguments are no better. Although
moral obligations place different burdens on different groups, that alone
isn’t a reason to think that those obligations aren’t real.
3.7 Conclusion
It does seem to me, though, that there are some important lessons that
emerge from these unsuccessful attempts to defend the permissibility
of eating animals. The first is that there are serious hurdles facing any
view that attempts to deny moral standing to animals. It’s no small feat
to explain why humans matter morally while animals don’t (or even to
explain why, if humans matter more, they matter at least enough more
to make eating animals morally permissible). Good arguments for the
permissibility of eating animals won’t begin by denying moral standing
to animals.
Second, we should be wary of romanticizing food, nature, or anything
else in this conversation—including, of course, animals.
Third, there are real costs associated with giving up animal products,
and some people bear heavier burdens than others. If the strength of
the reasons to be vegan are very strong, it will be rare that those costs
justify consuming animal products. If the strength of the reasons to be
vegan are relatively weak, the analysis gets more complicated. But even
as I say that, it’s worth stressing that we should be fully aware of our own
cognitive biases, including the speciesist ones that stack the deck against
animals. Our intuitions about when it’s worth sacrificing animal interests
for human ones are almost certainly going to be skewed in our favor, and
it’s no virtue of an argument for the permissibility of eating animals if it
relies primarily on self-serving intuitions.
With these observations behind us, let’s turn to some better arguments
for the permissibility of eating animals.
Notes
1. Not every version of contractualism does this; see, for instance, Rowlands
2009.
2. Carruthers is also known for denying that animals are phenomenally con-
scious, which interacts in interesting ways with his contractualism. For an
overview, see Carruthers 2011; for the details, see Carruthers 2005. How-
ever, I have nothing to say about this here. Also, for a different form of
contractualism that allows for some animal consumption, see Zeis 2013.
According to Zeis’s view, contractors produce three sets of moral rules
depending on what’s bracketed behind the veil of ignorance: one for rational
individuals, one for sentient individuals, and one for living individuals. The
rules are binding in that order, so that your obligations to beings qua sentient
beings can’t trump your obligations to beings qua rational beings.
3. Someone might say: “Cruelty is about your disposition of character, and
cruel dispositions aren’t inherent to factory farming.” However, if it turns
out that factory farming creates such dispositions, that’s a strike against it.
Compare: if it turns out that a certain kinds of policing strategies created
violent dispositions in officers, we wouldn’t excuse them by saying that the
strategies themselves aren’t problematic, as the violent dispositions aren’t
inherent to the strategies. Instead, we’d judge the strategies (partially) by
their effects on the people who implement them.
4. For details, see Kemmerer 2014. The worst environmental consequences are
due to factory farming, and the impacts of industrial operations aren’t con-
troversial. However, the debate about small-scale operations is very much
alive. Kemmerer is critical based on concerns about methane production as
well as land and water use. However, there are those who maintain that there
are environmental benefits to raising animals, such as reversing desertifica-
tion. For an overview, see Schwartz 2013.
5. True, you might have a duty to yourself not to become a cruel person. But
while a not-cruel (and so, presumably, compassionate) person wouldn’t nor-
mally torture animals, what’s cruel changes in a world with no other people.
Whatever she does to an animal, she isn’t being cruel in an environment
where she can’t harm any being who matters morally.
6. Proponents of this view include Welchman (2003), Williams (2006), Gun-
narsson (2008), Liao (2010), Maclean (2010), Chappell (2011), and Hsiao
(2015), among others.
7. Julie Rubio (2014, 52) develops another version of this view: “The tragedy
of human existence does not allow for a clean conscience. Instead, we have
to be content with our always partial efforts to do less evil and more good.
In the case of meat-eating, though the choice seems simple when considered
from an individual perspective, when placed in the context of family and
community it is far more complex. . . . While very few people “need” to eat
meat, families and community are fundamental human goods. Our realiza-
tion of basic human goods is always partial because they so often conflict,
and inevitably we will have to choose: For whom will we have the most
compassion today? Sometimes it will be animals but other times it may be the
teenager who would enjoy some time with the family if it included chicken
pot pie.”
8. For similar sorts of views, see Scruton 2006 and Cerulli 2012.
9. In a sense, this is the basic idea we find in the work of Alastair Cochrane
(2012) and Tzachi Zamir (2004). Both argue that there’s a difference between
use and exploitation, and that we should only ban the latter. This means that
if we can find non-exploitative methods of animal husbandry, we may hap-
pily consume the products derived from it.
10. This is compatible with there being some loss. Of course, morality asks peo-
ple to give up various racist traditions, such as flying the Confederate flag
and telling racist jokes at family gatherings. But surely we have no particular
reason to mourn that loss. That said, I tend to think that philosophers seri-
ously understate the value of traditions, perhaps in part because they are
often people who measure fairly high on openness to experience. Social trust
is extremely valuable for any society, as it greatly enhances our ability to live
flourishing lives. And traditions are extremely important for creating and
preserving social trust, as we tend to be much more inclined to trust people
who are like us. Of course, the problem is that this means that we tend to
be much more willing to trust men if we are men, straight people if we are
straight, and white people if we are white. So we have to get past that, and
find behavioral and ideological cues instead of physiological and psychologi-
cal ones. The upshot is that if you want people to get past things like race,
sex, and gender identity as relevant to trustworthiness, then you had better
not take away the things that people use as alternative forms of similarity,
like beloved sports, norms of etiquette, and national cuisines. This means
that continuing traditions like animal consumption is a weightier thing than
we sometimes realize, and it’s why fracturing relationships over food choices
isn’t a trivial matter. There are limits here, and I don’t want to claim that any
tradition whatever should be preserved. I’m only claiming that it’s a mistake
to give short shrift to seemingly arbitrary traditions.
11. Ultimately, it isn’t clear that he succeeds, as he changes the subject from
Grandma’s chicken and carrots to the Thanksgiving turkey. Moreover, he
never says how, exactly, he navigates his relationships with the people that
his abstinence offends.
12. Not everyone agrees. I once had a conversation with Carol Adams during
which she insisted that there are no costs whatever to giving up animal prod-
ucts. I don’t know if that’s her considered view, but she seemed quite ada-
mant about it at the time.
13. There are some methodological problems that can be either corrected or
minimized by examining evidence that either corroborates or disconfirms the
relevant hypothesis. For instance, self-report is notoriously unreliable, and
we have independent reasons to think that the majority of people who claim
to follow a plant-based diet do not, in fact, follow it strictly. So, although the
methodology should increase our confidence that eating fewer animal prod-
ucts has various health benefits, we should be wary of the claim that those
benefits can be attributed to a strict plant-based diet, simply on the basis of
the above. However, this kind of problem is fixable, and I don’t want to focus
on it here.
14. For similar lines of argument, see Johnston 2008, Harper 2010, Van Dyke
2015, and Muller 2017.
15. Francione’s view faces some serious challenges. Frameworks invoking rights
for anyone—animal or human—are controversial, and even if humans have
rights, it isn’t clear that they should be extended to animals (Ost, 1986).
What’s more, even if animals do have rights, there is an ongoing debate
about whether they have the kinds of rights that Francione ascribes to them
(Cochrane, 2012). What’s more, there are reasons to doubt the empirical
claim on which abolitionism depends: recent work on the impact of welfare
reforms suggests that they really do reduce animal use overall, contrary to
speculations that such reforms entrench marginally more “humane” animal
agriculture (Mullally and Lusk forthcoming). Like every view, then, it has its
problems.
16. This also complicates the health-based argument discussed earlier. When we
consider the risks of a vegan diet, George tells us that we ought to think
about the person from whose perspective the risks are being assessed. That’s
entirely fair. But we also have to think about that person’s default diet. There
are indeed risks to which a vegan diet exposes that person to which they were
not previously exposed. But there are risks to their current diet to which a
vegan diet doesn’t expose them. Without some global assessment of the risks,
the mere presence of some risks tells us very little.
17. The worry about unfair burdens on the poor needn’t be connected to race:
consider, for instance, the case of rural whites.
18. What’s more, it may even be appropriate to regard the change as “easy,”
because in light of the wrong being committed, the costs associated with
veganism look trivial.
19. A quick search on Amazon.com reveals a range of B12 supplements that
cost less than $0.10 per day, and multivitamins containing B12 for less than
$0.20 per day. Even if we assume that these supplements would cost twice as
much at local stores, they are within the financial reach of most Americans.
What’s more, it isn’t entirely fair to blame that entire cost on the obliga-
tion to eat vegan, given that omnivores have independent health-related rea-
sons to take these supplements. Finally, and as Josh Milburn pointed out to
me, even if vegan foods are more expensive now (because they are specialty
items), their cost will drop as they’re more widely adopted.
20. A tract counts as low-income if “the tract’s poverty rate is 20 percent or
greater; or the tract’s median family income is less than or equal to 80 per-
cent of the State-wide median family income; or the tract is in a metropoli-
tan area and has a median family income less than or equal to 80 percent
of the metropolitan area’s median family income.” See www.ers.usda.gov/
data-products/food-access-research-atlas/documentation/.
21. www.ers.usda.gov/data-products/food-access-research-atlas/documentation/
22. My guess—and it’s only a guess—is that age and disability status, rather than
membership in a racialized community, are better predictors of true lack of
access to a wide range of healthy foods.
23. Note too that this is a problem with respect to all food obligations for those
in food deserts—not just ones connected to animals. If people morally ought
to feed their children healthier options than, say, what’s available at gas sta-
tions and fast food establishments, then they ought to go to grocery stores
despite the costs associated with doing so. If they have obligations to them-
selves or to others to eat equally healthily (perhaps to model good practices,
or to stay healthy to provide care for someone), then they ought to go to
grocery stores despite the costs associated with doing so. In other words, if
food deserts excuse the people in them from going vegan, then they excuse
the people in them from most obligations regarding food. And I’m not sure
how plausible that it is.
24. Carol Adams makes this point in many of her lectures. For a thoughtful dis-
cussion of the way nostalgia functions as cover for racism and sexism in US
culture, see McPherson (2003).
25. It’s easy to find articles claiming that 1.5 million people in the US live on less
than two dollars per day, but the claim is misleading: they are reporting that
1.5 million people spend no more than two dollars per day at some point
during the calendar year. However, that figure doesn’t represent their total
buying power, both because of social services and because it doesn’t repre-
sent their annual income. By contrast, the one dollar per day figure in South
Sudan represents their annual income and their entire buying power.
4 Utilitarianism and the Causal
Inefficacy Problem
Granted, we get pleasure from eating animals, but not nearly as much as
they suffer in the process. They live in stressful environments, die more
slowly and painfully than we might like to think, and—even if we limit
our attention to land animals—outnumber us roughly 33 to 1. These
facts seem to provide the fodder for a powerful act utilitarian case against
eating animals—the theory that says that an action is morally required
just in case it maximizes expected utility. “Utility,” in this context, refers
to a unit of well-being, and there are tricky questions about how to spec-
ify what that means. But we’ll ignore those questions here. The important
point is that if this is a numbers game, it looks like we should be eating
veggies.1
In what follows, I introduce the simple act utilitarian argument for
veganism before turning to the most notable problem it faces: namely,
the issue of individual causal inefficacy. I then explain the standard solu-
tion to this problem, which appeals to “order thresholds,” and detail
Mark Budolfson’s criticism of it. Finally, I argue that even if Budolfson’s
criticism fails, the solution faces an entirely different problem. Either
way, we don’t get a straightforward argument from act utilitarianism to
veganism.2
4.1 The Simple Act Utilitarian Argument for Veganism

Here’s the simple act utilitarian argument for veganism:
1. You ought to act so as to maximize expected utility.

2. By abstaining from consuming animal products, you maximize
expected utility.
3. So, you ought to abstain from consuming animal products.3
The crucial premise is the second one, and it’s false for obvious reasons:
there are plenty of cases where consuming animal products sends abso-
lutely no signal to producers to the effect that they ought to reduce pro-
duction levels, and yet you would clearly benefit from consuming animal
The Causal Inefficacy Problem 51
products. For instance: your roommate goes out of town, leaving his
leftover Kung Pao chicken in the fridge. It will go bad before he returns,
and he will just expect you to throw it out long before then. No one
will know if you eat it, and you really like Kung Pao chicken. So, if the
goal is to maximize expected utility, you should eat it. This act will have
no negative effects (for animals, for the environment, for you, etc.) and
some decidedly positive ones (e.g., gustatory pleasure, satiation of hun-
ger, nutrition).4 So, the simple act utilitarian argument for veganism isn’t
sound.
The problem here is created by a gap between the ethics of purchas-
ing animal products—where it seems plausible that you might thereby
send some signal that could, at least in principle, affect production
levels—and the ethics of consuming them, which needn’t send any signal
at all. Whenever you can consume without buying, and your consump-
tion doesn’t encourage anyone else to make a purchase, the challenge
looms. And there are many such cases: eating leftovers in cases like the
above, pulling food out of the trash in full freegan style, collecting road-
kill, and so on. I suspect that the same problem looms in much more ordi-
nary cases, such as a university function at which food is served. In that
context, only so much food is available, and what isn’t consumed will be
tossed. If Abe eats the last chicken wing, the one that Bill would other-
wise have eaten, Bill just has some carrot sticks and hummus instead; he
doesn’t ask for and receive more chicken, as there isn’t more to be had.
So it doesn’t matter whether Abe or Bill eats the chicken wing: if there
is a negative expected utility in the neighborhood, it attaches to the pur-
chase, not to the consumption choices, and that purchase is independent
of their behavior.5
There are number of moves for the utilitarian to make at this juncture,
but I think the most promising is a mixture of concession and doubling
down. Here’s what I imagine the utilitarian saying:
Granted, you aren’t obligated to be a strict vegan. However, it’s very

rare that you’re in cases like the one described above, so it’s a bit
pedantic to focus on them. Moreover, although it’s permissible to eat
the leftovers in such situations, there are strong reasons not to. For
instance, in realistic cases, it’s often difficult to determine whether
your eating the leftovers won’t make a difference—it’s always pos-
sible that it might weaken your resolve to maintain a vegan diet
in cases where your choosing animal products wouldn’t maximize
expected utility, or that others might learn about your behavior and
have their resolve weakened. So, it’s morally prudent to adopt a rule
of abstaining in all circumstances, even if indulging can be justified
on occasion.
There are several points to note here.

52 The Causal Inefficacy Problem
First, my guess is that such cases aren’t nearly as infrequent as this
reply suggests. In my department, there is some event or other that cre-
ates non-vegan and frequently tossed leftovers nearly every week. And
I am married to someone who isn’t vegan, which explains why my chil-
dren aren’t vegan. Those small people waste food with wild abandon, so
my opportunities to eat non-vegan food abound. Granted, my situation
may be exceptional in certain ways, but just as we shouldn’t overestimate
the frequency of circumstances where behavior seems not to make a dif-
ference, we shouldn’t underestimate them either.
Second, notice that there is going to be individual variation here.
Admittedly, for many people it is morally prudent to follow a strict
No meat rule. As Rozin, Markwith, and Stoess (1997) and Rothgerber
(2014, 2015) demonstrate, we’re better at following simple rules than we
are at following more complex ones, and exceptionless rules are certainly
simpler than ones that require us to assess the likelihood that a particular
act will make a difference. But some people seem to have more willpower
than others, following rules consistently even when they become com-
plex. If you are someone with a strong will, this sort of reply doesn’t get
much purchase.6
Third, we are dealing with probabilities that are increasingly difficult
to assess. As we’ll see below, it’s very hard to say anything precise about
the expected utility of an individual purchase, much less an act of con-
sumption. But now we aren’t simply being asked to assess those prob-
abilities, which are crucial to establishing negative expected utilities for
the baseline cases, so that caution can be recommended in the others—
which, I take it, is the line of argument in the reply that I outlined earlier.
In addition, we are being asked to assess the probability (1) that someone
notices what I do, (2) that their behavior is somehow affected by what
I do, and (3) that their behavior makes a difference of the relevant kind.
No one ever provides any empirical evidence that’s relevant to assessing
these probabilities; people simply raise the possibility, and then recom-
mend following a strict rule. However, there are all sorts of possibilities
that, due to their improbability or inscrutability, we safely ignore. Why
should animal products be any different?
Fourth, I worry that this kind of reply indicates that the conclusion of
the simple act utilitarian argument is not, in fact, simply that you ought
to behave like an animal activist. That is, you ought to behave as though
it’s an important part of your identity that you promote the cause of
animals in all sorts of circumstances—that you optimize your behavior
for animals. This might mean, for example, advocating for vegan options
at social functions, and then (1) conspicuously consuming them while
(2) conspicuously avoiding the non-vegan alternatives. If you do this, of
course, people will become much more attentive to your behavior. And
if that happens, then there’s a greater risk that if you slip up, people will
level the charge of hypocrisy against you, and people are altogether too
happy to use the hypocrisy of others as an excuse to behave as they’d like.
Moreover, such behavior makes the cost of responding appropriately to
the plight of animals much greater. We aren’t simply talking about choos-
ing lentils over chicken, but about engaging in a range of minor con-
frontations about food that can have personal, social, and professional
costs. Of course, for most people most of the time, none of these costs
may be particularly significant, either individually or in the aggregate. So,
depending on the probability of making a difference, and the size of that
difference, it may be the case that most people, most of the time, are obli-
gated to act like animal activists. But until we know those probabilities,
we aren’t in a position to make that assessment.
Fifth, and finally, I’ll emphasize a point that I’ve already made. The
central reason to care about your resolve—or the resolve of others, or
any other indirect impact that you might have—is because the utilitar-
ian thinks that you usually maximize expected utility by abstaining from
animal products. That isn’t obviously so, because there isn’t only a gap
between the ethics of purchasing and the ethics of consuming: there is
also one between the ethics of production and the ethics of purchasing.
4.1.1 Causal Inefficacy, a First Pass

Admittedly, this is counterintuitive. And it seems like there is good rea-
son to deny it. Prominent voices in the animal advocacy movement have
argued that we can reliably estimate the number of animals saved by an
individual’s decision to go vegan. Consider, for instance, the methodol-
ogy that Harish Sethu recommends:
To determine the number of animals saved by a vegetarian [in the

US], we need [to know] the total number of animals killed for food
consumed in the US in a given year[,] the size of the US popula-
tion during that year[, and] we have to divide the number killed by
the size of the meat-eating population, as expressed in the following
formula:
Total number of animals killed

Number saved by a vegetarian = 
(
Population size × 1.0 −  v )
where v is the fraction of the population that is vegetarian or
vegan.7
He then provides some estimates of those numbers—which is surpris-

ingly difficult to do, largely because fish aren’t counted by the head, but
by the tonne. With those estimates in hand, we can turn the above into
an argument for veganism:
1. If you switch to a vegan diet, you can save between 371 and 582
animals per year from being killed for food.8
2. If you can save between 371 and 582 animals from being killed for
food, you should.
3. So, you should switch to a vegan diet.
Similarly, consider the work of two agricultural economists, Bailey Nor-

wood and Jayson Lusk (2011, 221–223), who insist that your purchases
do indeed make a difference, and then go on to quantify the impact you
can make:
It is tempting to think that your food choices have no impact on the

number of animals raised for food. After all, there are over 6.5 bil-
lion people in the world. However, to think that one consumer’s food
choices have no impact on livestock is illogical. . . . It may be hard to
see the consequences of our decisions, but let there be no doubt, each
purchase decision matters. To deny this fact is to contend that every
human becoming a vegan would have no impact on the number of
livestock raised. . . .
Suppose you decided to eat 5 fewer pounds of chicken next month
and every month thereafter. Your choice means that the grocery store
now has 5 more lbs of chicken meat left to sell. How will the store
convince other people to buy extra chicken? In the near term, they
will likely drop the price to ensure that they do not have to throw
away stock they have already purchased. . . . [So, if] you abstain from
eating 5 lbs of chicken each month, the overall amount of chicken
that continues to be produced and consumed will likely fall by some-
thing less than 5 lbs each month, because lower chicken prices induce
other consumers to buy more chicken. . . . It is the magnitude of
the demand and supply elasticities—the degree of which consumers
and producers are sensitive to price changes—that dictate the exact
change that will occur. . .
Using estimates of the elasticities of supply and demand for dif-
ferent animal products we can determine how total production of
a food item will respond to changes in a person’s consumption pat-
terns. Table 4.1 shows the impact of a decision to reduce consump-
tion for six animal food products. . . . [Again, differences have] to do
with the differences in the elasticities of supply and demand.
Those opening remarks from Norwood and Lusk bring out the main
problem with this line of argument: from the fact that a group makes a
difference, it doesn’t follow that a given member of that group makes a
Table 4.1
Long Run Effects of Reducing Consumption of Six Animal Food
Products
If you give up. . . Total production Per capita consumption of

eventually falls by food item
One pound of beef 0.68 lbs 65.20 lbs

One pound of chicken 0.76 lbs 85.10 lbs
One pound of milk 0.56 lbs 600.00 lbs
One pound of veal 0.69 lbs 0.50 lbs
One pound of pork 0.74 lbs 50.80 lbs
One egg 0.91 lbs 250.00 eggs
difference. In other words, we can’t immediately infer either the actual

or the expected impact of an individual from the average effect that each
individual has.
The most obvious reason for this is that you can’t draw a straight line
between each purchase and some percentage of those 371 to 582 animals
per year, or between not buying a pound of pork and the life of any par-
ticular pig. Even if Sethu and Norwood and Lusk are correct about the
average effect of not buying animal products, there are bound to be cases
where, for whatever reason, the grocery store doesn’t change its ordering
at all in response to your change in behavior, and as a result, no signal
is sent to producers that affects the number of animals harmed. After
all, grocery stores don’t order individual chicken wings: the wings are
shipped in cases from the regional distributor, and the regional distribu-
tor orders the cases by the pallet from the national distributor, and the
national distributor orders pallets by the truckload from the processor,
and so on. At each of those stages, your purchasing signal can be blocked
simply because the total number of purchases doesn’t cross either the
upper or lower tipping points, where either more or fewer animals would
be raised for slaughter. Moreover, we should expect this to be the normal
state of affairs, rather than the exception: you usually aren’t going to be
the one shopper whose purchase changes how much gets ordered. This
is the causal inefficacy problem.9 And given this problem, it looks like
the simple act utilitarian argument doesn’t just admit a few exceptions,
but fails generally. Buying and consuming animal products is a highly
reliable way of producing particularly satisfying pleasures, and the odds
that your purchase causes any suffering is essentially nil. So if you want
to maximize expected utility, you should buy animal products.10
4.1.2 The Order Threshold Solution

However, many consequentialists have thought that there’s a way to
dodge this problem. You don’t know when your purchases do and don’t
make a difference, but perhaps you don’t need to know. Norcross (2004),
Singer (2009), and Kagan (2011) argue that because there are thresholds
where changes in ordering do happen, and hitting those thresholds makes
a large difference, the expected utility of purchasing animal products is
still negative. The probability of making a difference may be very low,
but the size the difference you can make if you do hit such a threshold is
dramatic. And since you don’t know if or when you will hit those thresh-
olds, you should just assign the same negative expected utility to each
purchase. Here’s the way Norcross explains it:
Suppose that the industry is sensitive to a reduction in demand for

chicken equivalent to 10,000 people becoming vegetarians. For each
group of 10,000 who give up chicken, a quarter of a million fewer
chickens are bred per year. It appears, then, that if you give up eat-
ing chicken, you have only a one in 10,000 chance of making any
difference to the lives of chickens, unless it is certain that fewer than
10,000 people will ever give up eating chicken, in which case you
have no chance. Isn’t a one in 10,000 chance small enough to render
your continued consumption of chicken blameless? Not at all. While
the chance that your behavior is harmful may be small, the harm that
is risked is enormous. The larger the numbers needed to make a dif-
ference to chicken production, the larger the difference such numbers
would make. A one in ten thousand chance of saving 250,000 chick-
ens per year from excruciating lives is morally and mathematically
equivalent to the certainty of saving 25 chickens per year.
(2004, 232–233)
In other words:
1. If 10,000 people stop buying chicken, 250,000 chickens will be

saved.
2. So, you have a one in 10,000 chance of saving 250,000 chickens.
3. A one in 10,000 chance of saving 250,000 chickens is equivalent to
(the expected effect is) saving 25 chickens.11
4. If you can save 25 chickens, you ought to stop buying chicken.
5. So, you ought to stop buying chicken.
Call this the Order Threshold Solution. If the Order Threshold Solution
works, then we can set aside one of the most significant hurdles for the
utilitarian argument against purchasing animal products.
Unfortunately, I don’t think it works. But before I explain why, I want
to point out some neglected implications of this solution. For instance, if
this argument is sound, then there will be various cases in which it also
condemns consuming plant-based food. Consider a fried chicken place
that happens to have one or two vegan options. Suppose that you go out
with friends and order the vegan fare. In so doing, you make it slightly
less likely that the fried chicken place will go out of business. And if it
were to go out of business, it would probably be replaced by a place
that serves less meat, since prospective restauranteurs will be sensitive to
what’s failed and succeeded in that location in the past (or, at least, per
Lamey and Sharpless [forthcoming], it’s likely to be a place where the
meals are associated with fewer deaths; after all, even if it ends up being
a barbecue place, it’s likely to be a place that serves larger animals, which
means fewer deaths per serving). So, given the restaurant-closing thresh-
old that your purchase might hit, and the incredible difference that would
make for animals, it’s likely that the expected utility of buying the vegan
food is negative—or, at least, no one is entitled to say that the expected
utility of buying chicken at the grocery store is negative unless they are
also willing to conclude the same about the expected utility of buying the
vegan food at the fried chicken place.12
We can run this basic sort of argument in any situation where purchas-
ing plant-based foods supports the purchase of animal-based foods, as
long as the alternative to the status quo is substantially fewer animal-
based foods being for sale. You would have to run the numbers, but a
number of other cases immediately come to mind. Consider, for instance,
the “finer foods” sections in some grocery stores, where vegans can hap-
pily buy cured olives and omnivores can purchase cured meats and exotic
cheeses. These products stand or fall together: there won’t be an olive bar
unless prosciutto is available, and vice versa. And if the finer foods sec-
tion proves not to be profitable, then we have to consider what might be
sold instead. If you live in an area with a large Latino community, as I do,
then it could be something like a bulk dried bean case, where you can buy
dried pinto beans by the pound. (I actually saw this substitution occur
at our local Walmart.) In such cases, the expected utility of purchasing
cured olives may well be negative.
Some utilitarians will simply accept these results. I don’t think it’s obvi-
ous that they shouldn’t. Still, it’s worth recognizing that threshold-style
reasoning generalizes—a point to which I’ll return later on.
4.1.3 The Main Problem With the Order

Threshold Solution
In any case, let’s set aside such issues and evaluate the Order Threshold
Solution more directly. Does it work?
No: it calculates expected utility in the wrong way. The Order Thresh-
old Solution assumes that we should calculate expected utility by looking
at the average effects of such behavior. So, for instance, if the impact
of 10,000 consumers changing their behavior would be x, then we say
that the impact of your behavior is x/10,000, plus or minus any impact
that your particular change in behavior has, directly or indirectly, on
the other 9,999 consumers (e.g., by raising or lowering the price of the
relevant good, or affecting the availability of the good). But as Budolfson
(2019) shows, that’s a mistake: we can sometimes tell that the expected
effect of your behavior is virtually zero, even if the average effect of all the
aggregate instances of that behavior is some appreciable amount. This is
because we know that there is waste in the system (or market distortions
created by government incentives, or whatever) that makes the purchase
of any particular individual undetectable by suppliers, even if changes in
purchasing patterns are detectable when similar behavior is aggregated.
To see what he has in mind, consider this case, which is analogous in the
crucial respects to the way things work in the food industry:
Richard makes paper T-shirts in his basement that say ‘HOORAY

FOR CONSEQUENTIALISM!’, which he then sells online. The
T-shirts are incredibly cheap to produce and very profitable to sell
and Richard doesn’t care about waste per se, and so he produces far
more T-shirts than he is likely to need each month, and then sells the
excess at a nearly break-even amount at the end of each month to
his hippie neighbor, who burns them in his woodburning stove. For
many years Richard has always sold between 14,000 and 16,000
T-shirts each month, and he’s always printed 20,000 T-shirts at the
beginning of each month. Nonetheless, there is a conceivable increase
in sales that would cause him to produce more T-shirts—in particu-
lar, if he sells over 18,000 this month, he’ll produce 25,000 T-shirts
at the beginning of next month; otherwise he’ll produce 20,000 like
he always does. So, the system is genuinely sensitive to a precise tip-
ping point—in particular, the difference between 18,000 purchases
and the ‘magic number’ of 18,001.
(2019, 1716)
In this case, it’s very clear that the expected effect (i.e., the sum of the
number of T-shirts that Richard will make depending on whether you do
or don’t purchase an additional T-shirt, where each number is discounted
by the probability of that possibility being realized) doesn’t equal the
average effect (i.e., the number of T-shirts we get by dividing the total
effect of T-shirt purchasing by the number of T-shirt purchasers). Moreo-
ver, it’s clear that the expected effect is dramatically lower than the aver-
age effect. What establishes the low expected effect is the fact that, for
many years, Richard has always sold between 14,000 and 16,000 T-shirts
each month. As that period increases, the probability that Richard will
suddenly sell 16,001 shirts—much less 18,001 shirts—steadily falls.
Someone might object that nothing similar can be true of grocery
stores. Given the razor thin margins, they have a very strong incentive to
minimize waste. Moreover, nearly all major chains have adopted auto-
mated replenishment systems: when someone buys chicken wings, that
information is fed into a system that can, at least in principle, order more
of them without any human input. These systems are so attractive to
retailers because they automate the difficult work of forecasting future
demand: they employ algorithms that factor in the demand over the
course of the last relevant period, the demand in the same period during
the previous year, any upcoming holidays or other special events, and so
on.13 Given all this, we shouldn’t expect the kind of waste that we see in
Richard’s case.
But we do find it. Consider, for instance, that grocery stores have other
incentives as well. For instance, stores want to ensure that their meat
cases are always full, as studies have shown that consumers buy less if
they aren’t. Likewise, stores order products that they know have high
waste levels because they would lose customers to rivals otherwise: if
you want to buy lamb, and lamb isn’t available at your local store, but
only at the one across town, you will go across town to get it—and then
probably do much of your other shopping there as well. The upshot
here is that it’s often better to have some waste and not miss a sale than
the alternative; after all, the goal is to maximize profit, not to minimize
waste per se. And the empirical evidence bears this out. For these and
other reasons, Buzby et al. (2009) found that grocery stores throw away
approximately 4.5% of fresh meat. That report was based on data from
2006. Buzby et al. (2016) revisited the issue, finding that in 2012, stores
threw out 12.1% of fresh meat. The data was collected differently, which
means that it’s inappropriate to make any straightforward comparisons
between these results. Essentially, in 2012 they simply used UPC data,
which is available for purchase to anyone who wants to study it. The
2006 study, by contrast, was based on a blend of UPC data and inter-
views with supermarket executives. Those individuals probably factored
in ways that stores try to minimize their losses, for instance, taking nearly
expired meat and converting it into prepared foods. So, if we have to bet
on a number, we should probably go with the lower one offered in Buzby
et al. 2009. But whichever number we use, there is good reason to think
that there is some non-trivial amount of waste in the system.
As McMullen and Halteman (2019) point out, though, the mere pres-
ence of waste in the system doesn’t show that a particular purchase is
extremely unlikely to trigger another order that, ultimately, results in a
change in production:
The mere knowledge that there is waste in the system overall, or

even that there is waste at one’s particular supermarket, does not
give the customer any information about whether their purchase will
influence the retailer to order more or fewer products. Similarly, the
knowledge that buffers exist does not tell a consumer anything about
their location relative to those buffers. It is quite possible that wasted
meat in your supermarket gives the consumer information that their
producer is near the threshold, since their normal purchasing pattern
could have been excessive. The normal case, however, is that con-
sumers know little to nothing about the distance to a threshold at
their particular retailer, and even less further down the supply chain.
(7–8)
If McMullen and Halteman are right, then consumers should continue to

use the average effect as a proxy for the expected effect.
However, McMullen and Halteman aren’t right. To see this, let’s
suppose that Buzby et al. are correct, and grocery stores throw away
approximately 4.5% of fresh meat on average. Let’s also suppose, just
for simplicity’s sake, that there is no other waste in the system, and that
over the course of the last five years, the percentage of discarded meat in
a given month has varied between 1.5% and 7.5%. (Recall: the grocery
store isn’t trying to minimize waste; it’s trying to maximize profits. And
if it tries to eliminate the “excess” waste—namely, the 1.5% in months
where demand for meat happens to be higher—then it wouldn’t be able
to keep the cases full consistently, or wouldn’t have the right selection,
and would therefore be less profitable overall.) Given all this, it seems
highly likely that your buying a chicken breast won’t trigger a new order,
since the grocery store has an incentive not to be extremely sensitive to
individual purchases: it should tolerate normal variation within that
1.5% buffer, since if it tries to fine tune further, it will run the (unaccep-
table) risk of having empty cases and, therefore, losing sales overall. The
probability isn’t actually zero, of course, but it’s zero for all practical pur-
poses. This isn’t simply because there’s waste in the system, but because
there is waste and an incentive to tolerate a certain amount of variation
in consumer demand.
This should be clear on reflection. If you were developing an ordering
system, you wouldn’t set it up simply to replace every purchased prod-
uct: that would generate ridiculous results at every holiday. Instead, you
would have an algorithm that factors in recent purchases and histori-
cal sales data, with the goal of minimizing waste while ensuring that
the meat case is always full. So for a particular purchase to change the
ordering, it can’t simply be that you had one more order than you had
last month. It would need to be enough of a difference to indicate that
there is a new trend in purchasing. And since that’s very unlikely, given
the information fed into the ordering algorithm, the store is effectively
insensitive to any small difference in purchasing, even if it’s extremely
sensitive to purchasing trends.
In any case, if we focus on the average effect, we miss all this entirely:
we just see total number of orders that the grocery store places, we divide
that by the number of transactions, and we get some low-but-non-trivial
probability that any particular transaction is the trigger. If we confuse the
average effect with the expected effect, then we think that your behavior
is much more significant than it is. So again, you don’t maximize utility
by abstaining from animal products; if you enjoy them (or would receive
some health benefit from consuming them, or what have you), you maxi-
mize expected utility by purchasing them.
4.1.4 An Objection
All that said, someone might object that the above only shows that the
probability is lower than we originally thought; it doesn’t show that the
probability is so low as to generate a positive expected utility for an
individual purchase of an animal product. Moreover, the critic might say
that it doesn’t really matter if the probability is extremely low, given the
enormous harm at stake.
To see what the critic might have in mind, let’s return to the numbers
that Norcross uses. As he set things up, hitting an order threshold would
result in 250,000 chickens being saved, and he supposes that you have a
one in 10,000 chance of saving them. Let’s suppose that the probability
is much lower: say, one in one million. Now, it looks like you’ve got a
one in four chance of saving a single chicken on any given purchase. Or,
if we prefer, we can think about the expected utility in terms of a portion
of a broiler chicken’s life. Since broiler chickens are brought to slaughter
weight in about 42 days, we can think of the expected utility as roughly
equivalent to ten days of a chicken’s existence. Someone might think that,
given how much chicken suffer in the context of intensive systems, this is
going to swamp any benefit to the individual purchaser, especially when
you recall that the relevant comparison is not to the total benefit, but
only to the difference in benefit between purchasing chicken and purchas-
ing some plant-based alternative.
The main problem with this line is that we just don’t know the value
of the relevant probability. It could, of course, be much lower still: say,
one in one billion, rather than one in one million. In that case, it becomes
implausible that the expected utility calculation cuts the way that Nor-
cross suggests. In principle, of course, it’s possible to offer an empirically
informed estimate of the probability of hitting a threshold. You would
need to collect a lot of proprietary information from grocery stores and
their suppliers about sales, ordering, waste levels, and the like; and you
would need to examine the sensitivity of producers to previous changes
in demand. As it happens, I’ve approached several grocery stores about
collecting some of the relevant information, and as you might guess, they
have no interest in sharing it. I also wrote to Bailey Norwood about
this, one of the economists responsible for the estimates of the differences
you can make by refraining from purchasing certain animal products.
Norwood agreed that he and Lusk were focused on average rather than
expected effects, and he said that for all practical purposes, expected effects
are impossible to determine. So it seems to me that the safe thing to say is
that, at present, the value of the relevant probability is inscrutable, which
means that we aren’t in a position to assess whether the expected utility
of an individual purchase is positive or negative. That’s bad, because the
goal of the simple act utilitarian argument is to convince the purchaser
of animal products that she shouldn’t purchase them. In this context,
the act utilitarian has the burden of proof. So if the value of the relevant
probability is inscrutable, then the act utilitarian can’t meet the burden.
There are three ways that someone might resist this conclusion. First,
someone might object to the idea that the act utilitarian has the burden
of proof here. After all, we might think that harm needs to be justified,
so the purchaser of animal products actually has the burden: she needs
to show that the expected utility is positive. If the relevant probability is
inscrutable, then she should abstain.
However, this assumes exactly what needs to be shown: namely, that
the purchaser of animal products is on the hook for harming animals.
But the upshot of the causal inefficacy objection is that she probably isn’t
on the hook for harming animals.
The second way that someone might resist this conclusion is to insist
that if we have to err on one side or the other, we should assume a nega-
tive expected utility rather than a positive one. After all, there is no doubt
that animals are being harmed, and it seems plausible that consumer
action is part of the answer.
But I don’t know how far this reply goes toward meeting the original
burden. The goal was to establish that if act utilitarianism is true, then we
ought not purchase animal products. Instead, we are now being told that
we ought to make a plausible but ultimately unsubstantiated assumption,
and if both this assumption and act utilitarianism are true, then we ought
not to purchase animal products. But so far, we don’t have an act utilitar-
ian argument for making that plausible but unsubstantiated assumption,
and I’m not sure how to provide one.
The third move is to say that it doesn’t matter how low the probability
gets. Instead, what matters is whether the probability is inversely related
to the possible effect. If so, then the expected disutility of purchasing may
be stable; however low the probability of making a difference, the costs
of hitting a threshold scale up accordingly. In that case, whatever the
benefits of purchasing animal products, they’ll still be swamped by the
costs. Let’s call this the “inverse relationship hypothesis.”
I don’t think we should believe the inverse relationship hypothesis.
As far as I can see, it’s based on a simplistic model of the relationship
between consumer choices and production levels, where demand always
drives supply. But we know that the opposite can be true. For instance,
Jayson Lusk has pointed out that when cattle inventories were at a ten-
year low in 2014, we might have been tempted to think that demand had
fallen off, as per-capita beef consumption was also down. To the contrary,
though, it looks like the inventories were low due to high feed prices, and
beef consumption dropped because the price was high. (Consumers were
just buying cheaper meats.)14 Of course, this doesn’t show that demand
never drives supply, or even that it doesn’t usually drive it. The point
is just that the relationship between consumer behavior and producer
behavior isn’t straightforward.
The more damning point is this: when producers aren’t selling enough
product to their primary consumers, it isn’t as though they automatically
reduce their production levels accordingly. Instead, they often look for
other markets—abroad, perhaps, or in a different domestic industry, such
as pet food—and then accept whatever prices they get. (Indeed, this can
sometimes be an incentive to produce even more, as smaller margins mean
that greater numbers of animals have to be sold in order to cover the same
costs. For evidence of this, it’s worth reading Leonard (2014), who offers
a truly depressing window into the plight of chicken farmers. Even when
losing money on each flock, they still raise the largest flocks they can, hop-
ing, often quixotically, to dig themselves out of debt in the future.) In other
words, the buyers aren’t fixed in the contemporary meat market. So while
it’s possible that a threshold purchase changes the rate of production, it’s
also possible that the threshold purchase just redirects the product.
The reply here is this too can be factored into the expected utility cal-
culation. The utilitarian grants that buyers aren’t fixed, but then insists
that there must be some point at which a reduction in demand drives a
reduction in production levels—that point where the change in demand
can’t be accommodated by redirecting product—and that’s the order
threshold that matters.
But what happens at that point? Presumably, the answer is that the
reduction in demand drives the marginal producer out of the market-
place. However, that doesn’t get you a drop in production that’s equiva-
lent to what he would have produced. The marginal producer wasn’t able
to sell enough to make his business viable, but may still have been able
to sell a lot. All else equal, then, other producers will pick up as much of
that volume as they can, since increasing their production makes them
slightly more profitable. So instead of a large drop that corresponds to
the extremely low probability of difference-making, you get a more mod-
est drop. This means that you can’t assume that lower probabilities of
difference-making correspond to higher potential impacts. Lower prob-
abilities can simply be lower probabilities.
Granted, even if everything I’ve said is correct, it doesn’t show that
the expected utility of purchasing is positive. Instead, it shows that there
are some all-but-sure benefits for the consumer that are associated of
purchasing, the magnitudes of which I haven’t specified, which have to
be weighed against the costs and their probabilities. It may turn out that
the potential positives are sufficiently small that the costs swamp them.
But it may turn out the other way as well. Thus, without further empiri-
cal work, the simple act utilitarian argument doesn’t establish what it’s
supposed to establish.
4.2 Causal Inefficacy and Welfare Thresholds
I think we can see this point in an even more vivid way. Recall: we are
investigating a solution to the causal inefficacy problem according to
which there are order thresholds, so that while the probability of making
a difference—crossing an order threshold—is very low, the difference you
make if you hit a threshold is dramatic. However, I think that there are,
in addition, what I’ll call “welfare thresholds” that cut in the opposite
direction. That is, they count for purchasing rather than against it. I con-
tend that for all that’s been said by the defenders of the Order Threshold
Solution, it’s at least as plausible that you maximize utility by buying
animal products.
4.2.1 Welfare Thresholds: The Basic Idea

The intensification of animal agriculture has involved welfare
thresholds—points where economic pressures led producers to trim costs
by lowering the welfare of animals. Of course, the relevant practices are
never framed as cost-cutting measures that are taken at animals’ expense.
Instead, they’re framed as ways to improve welfare. Consider, for exam-
ple, this 1930 circular about egg production, which is the first-known
published discussion of debeaking. The author begins by acknowledging
the link between chicken “vices” and intensification, as well as the link
between intensification and novel management practices:
[Although] feather picking and cannibalism become a more serious

problem as poultry keeping becomes more intensified, they are not
new vices among chickens. The Complete Poultry Book published
some fifty years ago [i.e., circa 1880] refers to feather picking as a
“pernicious habit” and suggests that “the chopping block is the sur-
est remedy, but for valuable fowls a wire bit passed thru the mouth
like a horse’s bit and held in place by being passed thru the comb,
the wire being just large enough to prevent shutting the beak firmly
together, will render the bird unable to grasp feathers, and it will
soon abandon the habit.” The chopping block or the bit is hardly
practicable under present methods of poultry management [that is,
large numbers of animals in relatively small spaces]; hence special
precautions for prevention and control have become a necessity.
(Kennard, 1930, 20)
Then, he makes clear the value of debeaking:
Removal of the tip of the upper beak often becomes necessary as a

control and defensive measure to save the flock. The tip of the beak
is removed to the quick, leaving it tender and in such shape that it
is impossible for a bird to grasp firmly either [the] feathers or flesh
[of other birds, causing] the bird [to be] rendered harmless for about
three weeks. During this time the birds usually forget their past vices
and no further trouble results. . . . Tipping the beaks need not hinder
the birds from eating mash, nor affect egg production any more than
handling for any other purpose.
(1930, 27, emphasis added)
The upshot: if you debeak chickens, they will still eat enough that they
can lay eggs at an acceptable rate, and you’ll lose fewer birds to exces-
sive pecking by others. Welfare goes down relative to a more extensive
system, but “up” relative to the more intensive one without debeaking.
However, the aim is to protect the flock: the welfare “improvement” is
designed to improve productivity, and so profit. The economics forced
intensification, and intensification forced methods that would protect the
relevant commodities—namely, the birds that produce eggs, who will
ultimately be sold for meat after their laying careers are over; and, of
course, the eggs themselves.
This is, of course, an old example, but I mention it for that reason:
each objectionable practice has been around long enough that we tend to
forget that it had an origin. Grinding male chicks, battery cages, forced
molting, defeathering tanks—all these practices are responses to eco-
nomic pressures, whatever else they might be. And these are just a few of
the ones in egg production: there are countless others related to broiler
chickens, pigs, dairy cows, and beef cattle.
What’s more, we shouldn’t think that welfare thresholds are a thing of
the past, with things having been relatively stable for farmed animals in
more recent history. There are newer examples. For instance, mastitis can
significantly reduce milk production in dairy cows, and cows sometimes
need to be culled as a result. So, farmers have plenty of motivation to pre-
vent such infections from occurring. One way to improve udder health
is to use trimmers to remove hair from around the teats, which make
them easier to clean. However, trimming udder hair is time-consuming,
and in the 1980s, some American farmers began using “flame-clippers”
instead.15 These are, essentially, gas-fueled torches that singe the hair off
the animal, akin to the ones used to burn weeds. It’s easy to find people
in the industry insisting that it’s no more painful than using clippers,
though that isn’t obviously true, and in any case, certainly doesn’t mean
that the process causes no pain—in industry videos, the cows react in
ways that clearly indicate discomfort.16 Moreover, the risks are greater. In
one online forum for dairy farmers, a producer from Lebanon, Pennsyl-
vania encouraged others to “[use] an oven mitt or welders glove to wipe
the udder after [singeing] to stop any flare ups,” and you find similar
advice from the University of Wisconsin Extension Service.17 This sug-
gests that such burns are relatively common. Indeed, it seems likely that
cows quickly learn what this means for them. Since flame-clipping has to
be done regularly, Gamroth, Downing, and Ruddell (2000) warn farm-
ers to keep the flame out of sight of the cows, as they “react” otherwise.
Now that the practice has become standard, farmers are finding new
ways to deploy the same technology. In 2017, in an effort to improve
the rate at which cows shed their winter coats, the University of Mis-
souri Extension encouraged farmers to use these torches to singe hair
off livestock each spring.18 Again, when done properly, this may not be
any more painful than shearing, and since cows are susceptible to over-
heating, they benefit from accelerating the rate at which they shed their
coats. However, as with “torching udders,” the risk of burns isn’t incon-
sequential. Why did “flame-clipping” or “torching” become common?
Such practices, like debeaking, make sense in the increasingly competi-
tive dairy market: financial pressures discourage farmers from hiring the
additional staff that would be needed to maintain herd health without
them.
Moreover, we should expect similar creativity in the future. The USDA
runs the US Meat Animal Research Center in Clay Center, Nebraska,
under the auspices of the Agricultural Research Service, the mission of
which is to “[conduct] research to develop and transfer solutions to agri-
cultural problems of high national priority,” with the aim of, among
other things, “[sustaining] a competitive agricultural economy” and
“[providing] economic opportunities for rural citizens, communities, and
society as a whole.”19 In 2015, the New York Times published an exposé
on the center, finding that its “drive to make livestock bigger, leaner, more
prolific and more profitable can be punishing, creating harmful compli-
cations that require more intensive experiments to solve.”20 Examples
include breeding experiments that nearly doubled the number of piglets
in a single litter, resulting in many more piglets being crushed by their
mothers when they roll over; cows being bred to bear twins and triplets,
leading to complicated and high-risk deliveries, weak and deformed
calves, and high mortality rates; and a long-standing attempt to create
“easy care” sheep—namely, ones that can survive without human over-
sight or man-made shelters—the cost of which is many more deaths as a
result of predation, inclement weather, and starvation.21 These particu-
lar breeds may never become mainstream in the industry, but given the
many ways that the center’s research has already been influential, and
the obvious economic appeal of increasing productivity in these ways, it
would be surprising if related projects were never to make their way onto
American farms—and from there, to farms the world over.
4.2.2 Welfare Thresholds: The Argument

All that said, let’s return to chickens to make the argument about welfare
thresholds precise. Granted, debeaking became “necessary” in the 1930s
because demand for eggs was growing. But a drop in demand can have
the same effect. A shrinking market puts additional economic pressure
on everyone in the supply chain, forcing more aggressive competition
for fewer customers. It would, of course, be better in such circumstances
if some producers were to drop out, leaving more profit on the table for
those who remain. And, eventually, that will happen. But in the mean-
time, it’s individually rational for each producer to cut costs however
he can so as to stay in business, regardless of whether those cost-cutting
measures involve welfare compromises. However, once welfare compro-
mises have been embraced for financial reasons, they aren’t abandoned
when failing producers drop out of the market: instead, they become the
new normal.
The existence of welfare thresholds allows us to run an argument that
mimics the Order Threshold Solution. Let’s consider someone’s decision
to purchase a pack of chicken breasts from the grocery store. At some
point, another sale triggers an order to the distribution center. At some
point, another order to the distribution center triggers another order to
the supplier. And at some point, another order to the supplier triggers the
raising and slaughtering of 10,000 chickens. For simplicity’s sake, let’s
suppose that you’ve got a one in ten chance of being at any one of these
thresholds. Then:
1. The probability that your purchase is a threshold purchase is one in

1,000.
2. If your choice is a threshold purchase, then 10,000 chickens will be
killed.
3. So, your purchase has (roughly) the same expected impact as
(directly) killing ten chickens.
Granted, it’s dramatically less likely that you’re on a welfare threshold

than an order threshold. However, the impact is likely to be much greater,
as the practice will rapidly spread across the industry (as debeaking did
throughout the 1930s, quickly becoming standard):
1. The probability that your purchase is a welfare threshold purchase is

one in one million.
2. If your choice is a welfare threshold purchase, then one billion chick-
ens will suffer more intensely than they would otherwise.
(directly) causing 1,000 chickens to suffer more intensely than they
would otherwise.
In terms of total utility, it’s bad to kill ten chickens, but it’s worse to cause
1,000 chickens to suffer more intensely than they would otherwise. To
make this more precise, let’s focus on broilers rather than layers, just to
keep the numbers simple. The average broiler reaches slaughter weight in
about six weeks, and let’s stipulate that, out of 100 possible welfare points,
broilers currently average 80. Suppose your purchase hits a welfare thresh-
old that institutes some new practice that causes a 5% drop in broiler
well-being—say, by the introduction of a new breed of broilers that reach
slaughter weight slightly faster, and so have more musculoskeletal and res-
piratory issues than the breeds that are currently in production. Then, we
can compare the quality-adjusted life years (QUALYs) as follows:
Order Threshold: Loss of 92 QUALYs (36/365 ·10 ·80)

Welfare Threshold: Loss of 460.3 QUALYs ((36/365 ·1000 · 80)
- (42/365 ·1000 ·76))
Granted, I made up these numbers, and none of them is remotely plausible

as an estimate of the number of chickens could, at least in principle, be
traced back to a single purchase, or of the number of chickens that have
been affected by, among other things, the introduction of debeaking. How-
ever, they don’t need to be plausible estimates of such things. What mat-
ters is just that the relationships between the numbers are reasonable. In
other words, is it reasonable to suppose that, while the odds of hitting an
order threshold are relatively high compared to the odds of hitting a welfare
threshold, the number of animals affected would be dramatically higher?
Indeed it is. A large chicken shed might house 60,000 birds, and let’s sup-
pose that that’s the number that might be associated with a single order. But
a welfare threshold will affect some 9 billion chickens per year for an indefi-
nite period of time. If anything, the numbers I chose are highly conservative.
The upshot is this. True enough, you might hit an order threshold. But
you also might hit a welfare threshold. And although the odds of the
latter are much lower—indeed, dramatically lower—the consequences
are much worse. They are so much worse, in fact, that the existence
of welfare thresholds counts in favor of purchasing chicken rather than
abstaining. Again, I don’t actually want to defend the claim that you
ought to purchase chicken. Rather, what I’m arguing is that the burden of
proof is on those who want to run utilitarian arguments for veganism, as
it looks like the expected utility calculation doesn’t favor the conclusion
they thought it would.
4.2.3 Objections and Replies

I can imagine five objections at this juncture. The first is based on a plausi-
ble economic assumption, namely, that producers will use the lowest-cost
legal production method for a given quantity of eggs. If that’s right, then
there aren’t any welfare thresholds, as producers have already removed
any available waste in the system.
This assumption is indeed plausible, but inertia limits its applica-
tion: sometimes, there’s a lower-cost method available, but that move
would require extra time or money to make the change; so, the status
quo is preserved. (This can happen even when the cost would be recov-
ered relatively quickly: lots of businesses tolerate all sorts of inefficien-
cies largely to avoid having to think about how to engineer them away.)
Alternatively, there can be unconceived lower-cost methods—ones that
are available in principle, but haven’t yet been imagined. And, of course,
a certain amount of economic pressure could make the time or money
worth spending, or could inspire the dark imagination that results in
decreased animal welfare. What’s more, and as I indicated above, the his-
tory of animal agriculture supports the existence of welfare thresholds.
Before 1930, people tolerated the losses that resulted from not debeaking
chickens. But then those losses became intolerable, and debeaking began.
Granted, as farms are consolidated and management centralizes,
agribusinesses hire professionals with exactly one job: optimizing for
maximum profits. So we might think that they’d eliminate the economic
irrationality in the system (i.e., they wouldn’t leave money on the table
by not lowering welfare to the profit-maximizing point). But these pro-
fessionals aren’t wizards. They’re constrained by the facilities and hus-
bandry technologies that are already in place, by industry norms, by
noncompliance with their guidance (perhaps intentionally, but more
likely due to poor training and workforce turnover), and by the bounda-
ries of their own creativity. Obviously, these professionals will do a good
job of improving the efficiency of the systems they oversee. However, I’m
not claiming that these kinds of oversights are common. I readily concede
that the probability of hitting a welfare threshold is very low. I’m only
claiming that the industry probably hasn’t pursued every way of reduc-
ing animal welfare, and that they might explore a new way given enough
financial pressure. That shouldn’t be terribly controversial.
The second objection is that my argument makes an implausible
assumption, namely, that as more animal products are purchased, animal
welfare tends to increase. The thought here is that if there are economic
thresholds that can drive welfare down, there must be thresholds going
in the other direction as well.
My argument makes no such assumption. I only need the following
claims: first, that in response to increasing economic pressure, produc-
ers will look for ways to cut costs; second, some of those methods will
result in lower overall welfare for animals relative to earlier production
methods; third, one source of increasing economic pressure is decreased
demand while the number of producers remains constant; and, fourth,
that while the odds of hitting an order threshold are much higher than
the odds of hitting a welfare threshold, the expected disutility of hitting
a welfare threshold is much greater than the expected disutility of hit-
ting an order threshold, simply because of the incredible difference in the
number of affected animals. The first and second claims are borne out by
history (or so I’ve suggested), the third is an economic truism, and the
fourth is made plausible by the calculations set out above. So, my argu-
ment does depend on a modest counterfactual—namely, that if we hit a
welfare threshold, animals will be worse off than they would have been
otherwise—but not the hypothesis on which the objection relies (which
is indeed implausible).
The third objection is that even if it’s possible that there are welfare
thresholds, their relevance to assessing the ethics of purchasing decisions
is swamped by the wealth of evidence that producers respond to reduced
demand by reducing production levels. That is, we should discount the
welfare threshold problem based on our uncertainty about it and con-
tinue to offer, on act utilitarian grounds, a negative moral evaluation of
purchasing animal products.
For reasons that Budolfson (2019) makes clear, I doubt that the evi-
dence provided by agricultural economists about demand and produc-
tion levels is relevant to the ethics of individual purchases, as I suspect
that they are confusing average effects with expected effects. However,
let’s set that point aside and grant the objection’s premise. Now recall the
QUALY estimates that I proposed above:
Order Threshold: Loss of 92 QUALYs (36/365 ·10 ·80)

Welfare Threshold: Loss of 460.3 QUALYs ((36/365 ·1000 · 80)
- (42/365 ·1000 ·76))
To make this objection work, it isn’t enough to say that we are rightly
more confident about the existence of order thresholds than we are of
welfare thresholds. Instead, the objection needs to insist that we are
rightly more than five times more confident that about the former than
the latter—and that’s assuming that my numbers are accurate, whereas
I think that I’ve been far more charitable to the Order Threshold Solution
than is warranted. Given that it’s very clear that there have been points in
agricultural history where economic pressures appeared to have played
the role that I need for my argument, and given that we see new welfare-
reducing animal handling and breeding practices being developed, such
confidence in the Order Threshold Solution seems excessive. Moreover,
recall that I am not arguing for purchasing animal products: I am arguing
that the Order Threshold Solution doesn’t save the act utilitarian argu-
ment against purchasing animal products from the causal inefficacy prob-
lem. So I don’t need the claim that we should be so confident about the
existence and significance of welfare thresholds that we should assess the
ethics of purchasing in light of those thresholds alone. Instead, I just need
the claim that in light of the arguments above, we aren’t entitled to claim
that by not purchasing animal products, you maximize expected utility.
A fourth objection is that we’ve ignored yet another threshold: an
industry-ending one. Suppose that there is such a threshold. Then:
1. The probability that your purchase is an industry-ending threshold

purchase is one in one billion.
2. If your choice is an industry-ending threshold purchase, then 100 tril-
lion chickens will not be killed that would have otherwise been killed.
(directly) saving 100,000 chickens from being killed that would have
otherwise been killed.
If we think that preventing 100,000 chickens from coming into existence

beats sparing 1,000 chickens from suffering, then the possibility of being
on an industry-ending threshold tips the scales back in favor of not pur-
chasing. So, we get the upshot of the Order Threshold Solution, though
by different means.
However, I think that we can reject the first premise of this argument.
I don’t know the probability that your purchase is an industry-ending
threshold purchase, but it’s effectively zero. We know that the market for
animal products is large, involving many buyers and sellers. We know
that the market for those products is competitive and responsive. So, no
one purchase has any chance of ending the industry as a whole; at best,
it has a chance of kicking one bad actor out of the marketplace. But the
expected impact of that change is relatively small, as others will scale up
production to fill in the gap. (That the market couldn’t support that actor
doesn’t mean that the market couldn’t support most of his production.)
So, we can safely ignore the possibility of an industry-ending threshold.
The fifth objection is that we’ve had too narrow an understanding
of the harms associated with animal agriculture. For instance, we’ve ignored
the field animals killed during feed production (e.g., mice crushed by com-
bines in cornfields), the challenges facing wild animals that can be traced
to climate change, which is itself linked to animal agriculture, and so forth.
Since purchasing doesn’t have a statistical effect just on farm animals, but
on many others besides, might this tip the scales in favor of not purchasing?
Indeed it might. But we don’t yet have an objection here; we just have
ideas that could be developed into one. That’s perfectly compatible with
my point about the burden of proof. My only claim here is that the Order
Threshold Solution doesn’t put the ordinary shopper on the defensive.
And until someone works out the details about field deaths or the loss of
animal life due to climate change, that claim remains untouched.
4.3 Conclusion
Where does this leave us? Again, according to the simple act utilitarian
argument for veganism, you maximize expected utility by not purchasing
animal products. I’ve argued that there are clear cases where this premise
is false. And in the cases where it isn’t clear that it’s false—such as ordi-
nary purchasing at grocery stores—it isn’t clearly true, and we ought to
suspend judgment about it. And if we should suspend judgment about it,
then the act utilitarian hasn’t met the burden of proof. The causal inef-
ficacy problem undermines the argument.
As we’ll see, though, the causal inefficacy problem isn’t only relevant
to the simple act utilitarian argument for veganism. Given causal inef-
ficacy, a number of other arguments fail as well, as the impact of your
behavior is relevant to assessing when you’ve violated a right, or when
you’re exhibiting a vice, or how we ought to balance various competing
goods, such as health or convenience, on the one hand, and symbolic
action, on the other.22 But we will come to those issues in time. In the next
chapter, though, I’ll ignore everything I just said, and assume causal effi-
cacy rather than inefficacy. As I’ll argue, this helps the utilitarian criticize
much animal product consumption, but certainly not all.
Notes
1. I’m assuming that plants aren’t sentient. Michael Marder (2013) rejects
this, arguing that research on plants suggests the opposite. On this basis, he
claims that we should rethink the ethics of eating generally as the process of
developing respectful eating practices. Here’s hoping he’s wrong.
2. I ignore the various forms of indirect consequentialism (virtue, rule, etc.) in
this chapter. They come up later on when I discuss anti-complicity principles.
3. For arguments along these lines, see Matheny 2002, Norcross 2004, and
Singer 2009.
4. This is, essentially, an inversion of the argument in Garrett (2007). Very
roughly, Garrett argues that despite the causal inefficacy problem, you can
be obliged to abstain from animal products thanks to the health benefits of
a vegan diet. He contends that the extra pleasures you’d have in a life made
healthier and longer by a vegan diet outweigh any additional gustatory pleas-
ures you might gain by eating animals and their byproducts. So, you should
eat a vegan diet. However, the evidence only shows that a predominantly
vegan diet is superior to a comparably balanced omnivorous diet in terms of
health and longevity. The studies don’t show that a strict vegan diet beats a
predominantly vegan one. Indeed, the occasional consumption of lean meats
is probably good for you, and many people clearly enjoy them. Given these
benefits, and assuming that your actions in “leftovers that will be thrown out”
cases make no difference to whether future animals suffer and are slaughtered,
the causal inefficacy problem doesn’t support just the permissibility of eating
animal-based foods, but an obligation to consume them.
5. Similar reasoning applies, with some additional wrinkles, when you go to a
buffet where you pay the same price regardless of what you choose to eat.
Given the remarkable amount of waste in these contexts, it seems highly
unlikely that the restaurant will notice the absence of an additional chicken
wing.
6. Moreover, this reply assumes that the expected utility of eating animal prod-
ucts is generally negative, and so where it’s inscrutable, you ought to take a
cautionary approach. It remains to be established, though, that the expected
utility of eating animal products is generally negative. More on this below.
7. www.countinganimals.com/how-many-animals-does-a-vegetarian-save/
8. This figure includes 25 land animals per year, but the bulk of it comes from
(1) wild sea animals captured to feed aquacultured animals (fish and shrimp)
plus (2) bycatch.
9. There is a large body of literature on the causal impotence problem: see,
for instance, Budolfson (2015, 2019), Chignell (2015), Frey (1983), Kagan
(2011), Harris and Galvin (2012), Martin (1976), Matheny (2002, 2006),
and Warfield (2015). There is a related problem concerning climate change:
see, for example, Sinnott-Armstrong (2005), Hourdequin (2010), and Garvey
(2011). However, I say nothing about it here.
10. This is too quick, but only slightly. I’m assuming that you’ll spend some fixed
amount of money on food either way, and the choice is just between animal-
and plant-based products. But if you might spend that money on things that
could provide even more utility than animal products—like whiskey, or less
selfishly, feeding starving children—then that would be required instead.
11. There’s an assumption here that whatever the probability is, the threshold
must have the value necessary to yield that the expected marginal effect
equals the average effect. So, if we know the average effect of a single pur-
chase based on aggregating all purchases, we should conclude that the mar-
ginal effect of one more one less individual purchase is the same.
12. Someone might object that the success of a fried chicken place with vegan
options could inspire other fried chicken places to offer vegan options, and
that could make it easier for people to be vegan—or at least to head in that
direction. True enough. But all I need for my point here is that the situa-
tion described in the main text could be realized, not that it always is real-
ized. And when it is, the expected utility argument condemns purchasing the
vegan option from the fried chicken place.
13. A system that said, “tomorrow’s demand will equal today’s demand, so order
accordingly” would lead to disastrous results for the store: it would lead to
a wild amount of turkey being ordered right after Thanksgiving, due to how
popular it was immediately before. Obviously, stores want systems that are
more sophisticated than this.
14. http://jaysonlusk.com/blog/2014/9/22/economics-of-meatless-monday
15. The origins of the practice aren’t entirely clear. The first reference I could
find is from a 1990 issue of Farm Show Magazine, available here: www.
farmshow.com/a_article.php?aid=9642.
16. www.youtube.com/watch?v=IXiElYojKYA
17. For the advice from the Pennsylvanian farmer, see: www.dairyforums.com/
forums/showthread.php?2496-flame-udders. For the perspective of the Uni
versity of Wisconsin Extension Service, see: http://milkquality.wisc.edu/wp-
content/uploads/2011/09/udder-hair-removal_flaming-udders.pdf.
18. https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/59616/
g02098ImprovingSummerHeatTolerance.pdf
19. www.ars.usda.gov/about-ars/
20. www.nytimes.com/2015/01/20/dining/animal-welfare-at-risk-in-experiments-
for-meat-industry.html
21. Strikingly, the USDA audited the center in response to the Times’s report,
and although it contests most of the statements in the article as being either
inaccurate or lacking in context, the corrections and added context largely
affirm the initial reporting. See Exhibit A, pp. 14–32, available here: www.
usda.gov/oig/webdocs/02007-0001-31.pdf.
22. It also undermines a way of trying to argue for consuming fewer animal
products. Mark Budolfson (2015) suggests—though doesn’t endorse—the
view that we should just try to keep our food choices “under budget.” That
is, we should use the harm footprints of various foods to decide what to eat,
and simply try not to have “too large” of a footprint. But if it’s indeed true
that we don’t make a difference, then it’s trivial to get under budget—you
already are. Budolfson could revise the view he suggests, and say that when
we think about what it means to be “under budget,” we should understand
that notion in terms of the average effects of our actions. If he were to go this
route—and I am not suggesting that he would—that would be tantamount
to saying that we should treat the average effects of our actions as proxies
for their moral significance. Why, though, should the average effect be the
proxy? Why not social function, or personal significance, or economic ben-
efit? Why not actual causal impact, whatever that may be? I don’t see good
answers to these questions.
5 Causal Inefficacy Aside,
Utilitarianism Requires Eating
Unusually
In the last chapter, I argued that individual purchases generally don’t

make a difference, and in any case, we shouldn’t think that the expected
utility of purchasing is negative. But let’s set all that aside. Suppose that
purchasing (or eating) does make a difference. Is that all the utilitarian
needs to establish that we ought to be vegans? In a word: no. Even if indi-
vidual consumers can make a difference, utilitarianism doesn’t require us
to be vegans. It requires us to eat unusually.
There are some easy ways to see this.1 Perhaps the most obvious, on
which I won’t dwell here, is that the simple utilitarian argument for vegan-
ism ignores the differences between various animal husbandry practices.
You might grant that when animals have radically net negative lives—
involving much more suffering than pleasure on net—we can’t justify
farming them based on the pleasure we gain from eating them. But it isn’t
clear that most farmed animals live radically net negative lives. Moreo-
ver, we should recall that some animals have the capacity to feed large
numbers of humans, which makes it easier to justify harms to them on
utilitarian grounds.2 So, the simple utilitarian argument might still work
against some farming, but it wouldn’t necessarily work against all—or
perhaps even most.
This is especially relevant to the thought that the problems with the
utilitarian argument stem from focusing on act rather than rule utilitari-
anism. According to rule utilitarianism, you ought to act in accord with
rules such that their wide adoption would maximize utility. And, some-
one might think, if a “No animal products” rule were widely adopted,
utility would indeed be maximized. However, once we appreciate the
above, and recognize that not farming animals involves forgoing a con-
siderable amount of human and animal utility, it’s plausible that a “No
animal products” rule isn’t best. Instead, it should be something like,
“Eat animal products from animals who have lived net positive lives”—
which, of course, isn’t a particular high bar. Rule utilitarianism doesn’t
imply veganism either.
Of course, all this presupposes that total utilitarianism is true, accord-
ing to which the best outcome is the one with the most utility overall,
76 Utilitarianism Requires Eating Unusually
as opposed to versions that try to restrict the assessment in some way.
It’s this version of utilitarianism that generates the Replaceability Argu-
ment, which Peter Singer (2009) first offered.3 In sum: total welfare isn’t
affected by one animal’s death as long as we bring another into existence
(given the kinds of beings they are, that is, beings with few psychological
connections to their own futures), and total welfare would be increased
insofar as meat-eating benefits us; so, we ought to eat happy animals.4
Some utilitarians argue for restrictions on their views that block the
Replaceability Argument, so I don’t want to dwell on it here.5 And there
are some practical problems for anyone who wants to eat in a way that
this argument sanctions.6 But it seems safe to say that there are defensible
versions of act utilitarianism that deliver this result in principle, and there
are likely to be some farmers—even if they are few and far between—
who treat their animals well enough, and kill them painlessly enough,
to deliver this result in practice. Of course, this isn’t the kind of animal
product consumption on which we tend to focus, as we assume that most
people aren’t choosing between plants and products from extremely high
welfare farms, but instead between plants and products from rather low
welfare farms. But if the relevant version of utilitarianism is true, then it’s
likely that some “animal-friendly” agriculture isn’t just morally permis-
sible, but morally required.
5.1 Animals and Plant Agriculture

However, let’s ignore the above and focus on problems that afflict any
version of act utilitarianism—assuming, again, that causal inefficacy wor-
ries can be addressed. All these implications begin with the observation
that some animals are harmed in plant agriculture.
Consider, for example, Steven Davis’s (2003) observation that we’ve
overlooked certain harms involved in a vegan diet: namely, the harms
to animals that are associated with growing plants for food. After all, a
great many animals make their homes in and around the fields where we
grow plants for human consumption. Some of these animals are killed
unintentionally by farm machinery; some are killed intentionally by poi-
sons, traps, and on-farm hunting; and a great many more are killed by
predators who only have access to those animals because we remove their
cover when we harvest. Davis goes on to argue that we ought not to be
vegans if we want to minimize harm. (This means that Davis isn’t offering
a squarely utilitarian argument, of course, but it could be reworked as
one.) He estimates that, if the US population were to go vegan, 1.8 billion
animals would still die each year as a result of our agricultural practices.
But if we were to supplement our diet with cows that forage on open
pastures (read: living nice lives), then that number might fall as low as
1.35 billion. Why? Because raising large ruminants on the pasture-forage
model is likely to result in fewer animal deaths per hectare.7
Utilitarianism Requires Eating Unusually 77
The math doesn’t work out. As Gaverick Matheny observes, Davis
assumes that you get the same amount of food from one hectare of graz-
ing land as from one hectare of cropland. But this isn’t true: in a years’
time, one hectare of cropland can produce ten times the protein as one
hectare of grazing land. Moreover, there are excellent reasons not to trust
Davis’s numbers in the first place. For instance, it isn’t at all clear whether
we can generalize from the very few studies on field deaths, which focus
on different species in different agricultural contexts at different times of
the year. One of the studies that’s crucial to Davis’s estimate is based on a
study of rats in sugarcane fields in Hawaii—where, for complex histori-
cal reasons, the harvesting methods in Hawaii were unique, using a mix
of relatively small machines and traditional hand tools. How relevant is
that data to, say, estimates of field mouse deaths on Iowa wheat farms,
where all harvesting is done via enormous combines? When you combine
the differences in species, location, time, and method, any conclusions
will be tentative indeed. (For more on these issues, see Lamey, 2019;
Fischer and Lamey, 2018.)
However, no one disputes that some wild animals are currently
harmed. And this opens the door for a different sort of anti-vegan argu-
ment, namely, one that only relies on an extremely low estimate of the
body count in plant agriculture. There are several such arguments.
5.2 Roadkill and Freeganism

Suppose we can find a source of meat that isn’t a direct or indirect prod-
uct of our agricultural practices, and suppose that meat will be wasted
if we don’t consume it. Now, might we be obligated to supplement our
diet with that meat, thereby reducing our dependence on plant agricul-
ture, and thus reducing the number of wild animals harmed in plant
production?
Donald Bruckner (2015) thinks so: he contends that we ought to eat
roadkill—specifically, large, intact, and recently killed animals. (Think
of the deer that wasn’t laid out by the side of the road when you went
to the grocery store, but is there now.) To reach this conclusion, he
appeals to a principle that David DeGrazia (2009, 159) defends, accord-
ing to which “[it] is wrong (knowingly) to cause, or support practices
that cause extensive, unnecessary harm to animals.”8 If there is indeed
an alternative, then maintaining a strict vegan diet involves supporting
practices that cause extensive, unnecessary harm to animals—namely,
those in plant agriculture. So, we shouldn’t be strict vegans. What’s the
alternative? Well, by scavenging, we cause no harm whatever: the claim
isn’t that we should try to hit animals with our cars; the claim is that
we shouldn’t let potential food sources go to waste. Scavenging is also
no riskier than eating meat from hunted game. As long as it’s fresh, it’s
lean, healthy meat. It’s also free, and you can learn how to prepare it by
watching a few YouTube videos. So, we should scavenge. (Obviously,
this isn’t a straightforwardly utilitarian argument, but it’s easy to see how
to reformulate it as one, and in any case, DeGrazia’s principle is supposed
to be one that the utilitarian can endorse.)
Of course, as Bruckner notes, there certainly are philosophers who
both defend veganism and have no objection to this practice. Peter Singer
has been asked a number of times about his views on consuming road-
kill, to which he replied, in 2011, that it “is perfectly ethical” (Kendall,
2011). Likewise, DeGrazia claims that his defense of veganism “does not
oppose, say, the consumption of a dead animal one finds in the woods”
(2009, 148, fn. 14). And Jordan Curnutt claims that his case for vegan-
ism “allows . . . eating animals who died due to accidents” (1997, 156).
But none of them take roadkill consumption to be obligatory. That’s the
respect in which Bruckner pushes the envelope.
But perhaps the conclusion can be avoided. Recall that we aren’t assum-
ing any particular body count per hectare—perhaps it’s as low as one per
hectare. In that case, the harm footprint of any particular plant-based
meal is going to be negligible, given the enormous amount of food that
a single hectare can produce. Then, we can appeal to the harms to wild
animals in virtue of having a food source removed. After all, Bruckner’s
argument is strongest when we focus on cases where you can be the most
confident that the meat available is safe to eat. Those are going to be cases
in the wintertime, when bacterial growth is inhibited by lower tempera-
tures. But those are also the circumstances in which many animals are
living on the margins, and the difference between finding a carcass and
not can be the difference between starvation and not. Moreover, a single
carcass might feed a large number of animals. Vultures, coyotes, squir-
rels, and even chickadees will feed on roadkill, which is no small boon for
them during lean months. Of course, municipal authorities often pick up
roadkill for composting, so if the carcass is in a spot that those authorities
frequent, then this point is moot: wild animals wouldn’t have had access
anyway. But if it’s likely that wild animals would have, perhaps because
the area is more remote, then roadkill may not pass the utilitarian test.
However, whatever the story regarding roadkill, Bruckner’s argument
more squarely supports freegan practices generally, such as scavenging
from dumpsters and department refrigerators, though he doesn’t make
that point. In these circumstances, there is no harm to wild animals to
worry about, as these discarded items were never going to be made avail-
able to them. Again, there are surely people who won’t object to this
being permissible: Stuart Rachels, for example, says that “if someone
else is about to throw food away, you might as well eat it” (2011, 883).
However, the argument supports a stronger conclusion: it’s required.
But I suspect that few would be willing to accept that it would be mor-
ally wrong to let animal products go to waste, simply because we could
slightly lower our harm footprint by being scavengers.
At this juncture, the utilitarian who dislikes this conclusion might
appeal to the signaling value of a strict vegan diet. We might imagine a
response along these lines:
When animals suffer at human hands, it’s usually because of specie-
sism. So, insofar as the goal is to minimize harm to animals, we should
be doing what we can to challenge speciesism wherever we find it. Admit-
tedly, if we focus narrowly on the harm footprint of a plant-based meal
and the harm footprint of a meal containing discarded animal products,
it’s likely to work out that the former is larger than the latter. But the dif-
ference isn’t likely to be enormous. Moreover, we can have a significant
impact on others by refusing to eat animals under any circumstances,
challenging their ideas about the relative importance of human and non-
human animals, and eliminating the possibility of their discounting our
moral witness based on charges of hypocrisy. So, although this way of
extending Bruckner’s argument initially seems to establish an obligation
to be freegans rather than vegans, the argument is too narrow, ignoring
the other impacts that your behavior can have.
There is certainly something to this reply. However, it seems to me that
it both overstates the signaling impact of our actions and understates the
value of explicit discussions, the opportunities for which are often cre-
ated by our food choices.
On the former point, we should begin by remembering how often
our behavior goes unnoticed. It’s easy to think that we are always being
watched, which is just false. And even when we are being watched, peo-
ple often don’t think all that hard about what we do. Unless utilitari-
anism requires us to behave like activists, with very public pro-animal
identities that attract significant attention, we can count on being ignored
most of the time. And if it does require us to behave like activists, then
we should recognize—as mentioned earlier—that the cost imposed on
the consumer is not simply the sacrifice of trivial pleasures. Instead, utili-
tarianism is demanding that animal advocacy become one of your life
projects, which is a much more burdensome requirement, and not one
that’s obviously justified.
On the latter point, about the value of explicit discussions, I’ll say this:
the motivation for the objection above seems to be that people aren’t par-
ticularly good at reasoning about what is and isn’t driven by speciesism.
But if that’s the problem, then it’s likely to be a symptom of a much larger
failure to reason well, which generally isn’t best addressed by letting your
actions speak for themselves. Rather than think that we should reject
speciesism, people are at least as likely to think you take animal bodies to
be sacred, or that you think that animals care what happens to their dead
bodies, or that veganism is about your own moral purity, or that you just
“love animals” so much that you’re disgusted by the prospect of eating
them. If we want—or ought—to convince people that they shouldn’t be
speciesists, we need to talk with them about speciesism, explaining what
is and isn’t an instance of it. If it turns out that making freegan exceptions
to an otherwise vegan diet creates opportunities for such conversations,
that seems like a mark in favor of those freegan exceptions.
5.3 Bivalves
If the utilitarian argument for a freegan diet works, then it can probably
be extended to cases where killing is required. The trick is simply to find
cases where the probability of sentience is sufficiently low. Many bivalves
fit the bill.
However, I’m not going to say much about bivalves here, as the case for
their not being conscious is so strong. With respect to oysters and mus-
sels, for instance, consider the arguments that Diana Fleischman (2013)
makes. First, they are unable to move, which means that they don’t need
consciousness to find food or to evade anything that might damage them.
Given that consciousness is a relatively resource-intensive adaptation, you
wouldn’t expect to find it in organisms that don’t need it.9 Second, oysters
and mussels have very simple nervous systems with no obvious cephaliza-
tion, that is, the gathering of nerve ganglia, sense organs, and a mouth
that indicates the presence of a head. Without this kind of concentration,
coordination becomes less likely, reducing the probability of conscious-
ness. Third, they don’t change their behavior in response to tissue damage,
nor do they appear to have a pain-management system (e.g., opiate recep-
tors). This strongly suggests that they don’t have the capacity to feel pain.
For all that, it’s possible that oysters and mussels are conscious. But the
probability seems to be negligible. And given that there are unintended
harms associated with plant-based foods, and virtually no unintended
harms (and, in fact, some environmental benefits10) associated with oys-
ter and mussel farming, the utilitarian case for eating them seems fairly
strong. Suppose that a dish contains ten mussels, and we set the odds of
mussels being sentient at one in 100,000 (which seems generous). Then,
the expected utility of eating the dish is equivalent to a one in 10,000
chance of harming a sentient being. Now suppose that, just as in animal
agriculture, there are order thresholds in plant agriculture. The odds of
being the one to trigger an additional hectare going into production are
extremely low. But if you do trigger that change, then we can be quite
confident that millions of insects will be killed by pesticides (more on
this below), and some number of clearly conscious animals will also be
harmed—some mix of birds, small mammals, reptiles, and so on. Sup-
pose we borrow Norcross’s number, assigning a one in 10,000 chance
of hitting the threshold. Even if one clearly conscious animal is killed on
that hectare, you have parity. Once you factor in any other mammalian,
avian, or reptilian deaths, plus the deaths of insects (even if we discount
those deaths based on the improbability of insects being conscious), the
scales will tip in favor of eating mussels.11
5.4 Insects
Insects, then, are the more interesting case, as it’s far more complex to
assess whether they have the relevant kinds of mental states. If they don’t,
then the case for eating them is strong. First, anyone can raise and pre-
pare them. Mealworms and crickets, for example, are readily available
from your local pet store, and with the help of a fish tank, some food
scraps, and a water source, both species will multiply like mad. Second,
they’re easy to process and prepare without any food safety concerns:
you just need boiling water, and then your culinary options are open.
Third, they’re very good for you: crickets, for example, are low in fat
and high in protein, B12, omega-3s, iron, and potassium. Fourth, they’re
environmentally friendly: they will take products that would otherwise
go to waste and convert them—very efficiently—into nutrient-rich food.12
If it turns out, however, that insects are conscious, then this won’t mat-
ter very much. If farming them harms them, then given the sheer number
of insects required to make a meal suitable for a human being, the case
against farming them will be powerful. And if the case against farming
them is powerful, and we assume causal efficacy, then the case against
purchasing and consuming them will be decent as well. So what should
we say about the cognitive capacities of insects?
5.4.1 Are Insects Conscious?

It’s difficult to be sure. Insects do demonstrate surprisingly complex behav-
ior. For instance, earwigs display maternal care (Kolliker and Vancassel,
2007), honeybees can find artificial food sources that have been moved
significant distances by experimenters (Griffin, 1984), and weaver ants
can recruit other ants to fight without themselves having been exposed
to the threat (Hölldobler and Wilson, 1978). Elsewhere, Bateson et al.
(2011) have argued that bees can come to interpret ambiguous stimuli
negatively, a phenomenon caused by negative emotions in humans. How-
ever, there remains a tendency to apply a particularly stringent version
of Morgan’s Canon, according to which we should deny that insects are
conscious as long as it’s possible to explain their behavior without pos-
tulating conscious states. There are serious problems with this idea, for
instance, it ignores the fact that the explanation in terms of conscious
states may be the better one (Fitzpatrick, 2017), but since it remains the
default approach, more pointed evidence is required.
Presumably, if insects can feel pain, then they’re conscious. But can
they feel pain? The evidence is difficult to interpret. Insects have neurons
that respond to negative stimuli, and some insects react to analgesics
in ways that suggest these negative signals can be blocked. Zabala and
Gomez (1991), for instance, found that cockroaches left a heated box
more slowly after being injected with morphine, and Balderrama et al.
(1987) showed that receiving morphine caused bees to sting less aggres-
sively after having a round of electrical shocks. What’s more, it doesn’t
look like morphine is simply decreasing behavioral responsiveness gen-
erally. For instance, fruit flies tend to move toward light, but if they’re
placed in a glass tube and the center of that tube is heated, they tend not
to pass through the hot portion. However, if given an analgesic, they’re
willing to pass through that otherwise-uncomfortable spot, again reach-
ing the light source (Dimitrijevic et al., 2005). Fruit flies also act in ways
that suggest that they can learn to avoid painful stimuli: they can come
to associate electric shock with an odor, and then will avoid that odor for
up to 24 hours afterward (Yarali et al., 2008).
Nevertheless, it is also the case that insects often fail to respond to
damage in the way that humans might expect were they to be suffering.
In the words of an influential review by Eisemann et al. (1984, 166):
[O]ur experience has been that insects will continue with normal
activities even after severe injury or removal of body parts. An insect
walking with a crushed tarsus, for example, will continue applying it
to the substrate with undiminished force. Among our other observa-
tions are those on a locust which continued to feed whilst itself being
eaten by a mantis; aphids continuing to feed whilst being eaten by
coccinellids; a tsetse fly which flew in to feed although half-dissected;
caterpillars which continue to feed whilst tachinid larvae bore into
them; many insects which go about their normal life whilst being
eaten by large internal parasitoids; and male mantids which continue
to mate as they are eaten by their partners. Insects show no immobi-
lisation equivalent to the mammalian reaction to painful body dam-
age, nor have our preliminary observations of the response of locusts
to bee stings revealed anything analogous to a mammalian response.
Of course, many animals have systems that block pain—in humans, as

well as some insects, endorphins play that role (Duve et al., 1979; Remy
et al., 1979)—and it is possible that insects have a similar system that
allows them to maintain relatively normal function despite severe dam-
age. Moreover, in circumstances where the odds of survival are sufficiently
low, it may be in the best interest of an insect to simply stay the course:
this might allow mating to take place when the opportunity is available
and the continuance of other behaviors (e.g., feeding) might simply be
a byproduct of that fitness-enhancing mechanism. Furthermore, chitin
does not regrow, which means that insects do not have the same incentive
to protect damaged parts of their bodies that, say, mammals do. A dog’s
leg can heal, so it makes sense to try to save a damaged limb; however, an
ant’s leg will not heal, so it may as well be abandoned.
Viewed in isolation, these competing considerations are difficult to
resolve. So there are also broader theoretical reasons that we might
consider. Against insect consciousness, we should note two points. First,
insects have relatively small, distributed nervous systems (generally not
exceeding one million neurons, whereas humans have around 100 bil-
lion). If we assume that cognitive sophistication varies with the number
and organization of neurons, and that consciousness is a fairly sophis-
ticated way of processing information, then we shouldn’t assign a high
probability to the hypothesis that insects are conscious. Second, if we
think that consciousness is an evolutionarily “expensive” way of pro-
cessing information, then we shouldn’t assign a high probability to the
hypothesis that insects are conscious unless they exhibit behavior that’s
difficult to explain with simpler cognitive mechanisms. And given that
no one has created conscious artificial intelligence, we should note that
current AI can produce many of the behaviors that insects can. For
instance, robots have been programmed to respond to “painful” stimuli
and become “fearful” of potential sources of pain (Lee-Johnson and Car-
negie, 2010).13
On the other hand, consider the work of Barron and Klein and (2016),
who draw on Merker’s (2005) theory of consciousness. Merker (2005)
argues that consciousness arises from the need to integrate information
from a variety of sources, and that this happens when organisms become
mobile and have to represent their location in space. Barron and Klein
(2016, 7–8) point out that
there are significant parallels between the functional organization

of the insect brain and that of the vertebrate midbrain behavioral
core control system. Both systems have specialized regions for pro-
cessing the position of the moving animal in space. In both systems
action selection is resolved by combining information on position
with information on the environment, the relevance of stimuli in the
environment to the animal, and the state of the animal. In vertebrates
the channels of information are sufficiently unified such that the sys-
tem as a whole creates a functional representation of the state of the
mobile animal in space as a solution for effective decision making.
As we have argued, processing of this kind supports the capac-
ity for a subjective experience of the environment. Processing in the
insect brain is unified to a similar degree, for similar reasons. Hence
we propose that the insect brain can also support a capacity for sub-
jective experience.
In other words, we find systems that seem to fulfill the same function in
vertebrates and invertebrates, and in the case of vertebrates, it’s plausible
that the relevant system explains the presence of conscious awareness in
the organism. This simulation is highly valuable to mobile organisms that
find themselves having to integrate various sources of information—visual,
auditory, and so forth—so that they can navigate their environment.
And, if it fulfills the same function in invertebrates, that is a reason to
think that they have some degree of consciousness as well. (For further
discussion of this conclusion, see the special issue of Animal Sentience
devoted to Klein and Barron’s essay.)
Merker’s theory is hardly the only one in the field. Some insist that con-
sciousness requires higher-order thought (Edelman, 2003); others main-
tain that it requires awareness of one’s self as a subject (Morin, 2006).
But it’s certainly a prominent theory, and Merker is one of many who
argue that certain integrated structures in the vertebrate midbrain sup-
port conscious experience (e.g., Damasio and Carvalho, 2013; Mashour
and Alkire, 2013). Moreover, Klein and Barron (2016) provide good
reasons for applying this approach to insects. Does this show that insects
are, in fact, conscious? Of course not. This is one line of argument based
on one theory of consciousness. And even if this argument works, it may
only support the conclusion that insects have something like access con-
sciousness rather than phenomenal consciousness—that is, there may not
be anything like the phenomenal experience of pain, but only the capac-
ity to integrate and act intelligently based on information. This line of
reasoning may raise the probability that insects can experience pain, but
it’s hard to say by how much.
5.4.2 The Utilitarian Argument for Entomophagy

We do know, however, that eating insects—entomophagy—involves kill-
ing an extraordinarily large number of them. So, given even very low
odds, the expected utility of insect farming might be negative.
Some dismiss this worry. Meyers (2013, 124), for instance, writes:
Even if insects were capable of pain, the conditions that they would
be raised in are conditions that would not cause them to suffer.
Unlike cattle, pigs, or chickens—and unlike even crabs, lobsters, or
shrimp—most insects actually prefer to live in crowded, hot, and
filthy conditions.
In other words, what matters isn’t just the probability of insect con-
sciousness, but also the probability that insect farming is actually bad for
insects. As the latter probability drops, it may turn out that the number
of insect deaths becomes irrelevant. In that case, harm to even a very low
number of clearly sentient animals in plant production would be enough
to tip the expected utility calculation in favor of eating bugs.
However, I suspect that Meyers is a bit too quick here. Mortality rates
on insect farms are pretty high. Mealworms get diseases and die en masse,
crickets sometimes cannibalize one another, various species find ways to
crush one another, some are eaten by predators that manage to get into
enclosures, and many die for no known reason. In my conversations with
cricket farmers, I’ve been told you’re doing well if you get your overall loss
rate under 20%. If crickets can feel pain, then it’s likely that some of them
live uncomfortable lives followed by unpleasant deaths. So we shouldn’t
romanticize insect farming. Like any intensive form of farming, the welfare
of the “product” isn’t first and foremost. Instead, profit maximization is
first and foremost, and welfare matters insofar as it serves that end.
Moreover, there’s a plausible case for thinking that the expected utility
of insect farming is negative. For simplicity’s sake, though, we’ll focus on
the expected number of deaths. (Granted, an expected death estimate is
unlikely to be a perfect proxy for expected utility, but there’s no other
workable metric, so it will have to suffice.) The first step is to find a
way to compare the number of invertebrates affected by plant and insect
production. One way to do this is to focus on yield. According to recent
USDA numbers, soybean fields generate about 2,868 pounds of soybeans
per acre.14 A mealworm weighs about 100 mg, so it would take roughly
13 million mealworms to replace an acre of soybean production.15 Sup-
pose that we take the probability of mealworm sentience to be very
low—say, one percent. In that case, it looks like trading mealworms for
soybeans would give us an expected death rate of 130,000 individuals.
Plant production, therefore, seems preferable.16
However, this overlooks an important part of the calculus—namely,
pest management—and this detail tips the scales in favor of insect farm-
ing. Estimates of invertebrate abundance in agricultural contexts vary
considerably, but suppose that there are, on average, ten million inver-
tebrates per acre at any given time. It takes anywhere from 90 to 150
days for soybeans to mature, and they might be sprayed every ten days
or so. Assume that it takes an average of 100 days for soybeans to reach
maturity, that pest management strategies manage to kill even 13%
of the population at any given time, and that we should stick with the
standard low average probability of invertebrate sentience that we have
been employing (one percent). That’s enough to get us parity: 130,000
expected deaths.
Of course, that’s all very speculative, but it’s worth noting that it’s
speculative in ways that are friendly to the defender of plant production.
Ten million invertebrates per acre is a very modest estimate,17 and pest
management strategies probably cause far more deaths. For instance, one
study found that most insecticides killed over 80% of soybean aphids
after application (see Rice et al. 2007). And while we may need to
increase the expected death count for mealworm production, it’s going to
be a relatively modest increase. We certainly wouldn’t need to increase it
by a factor of five or ten (or more), which is what would be required to
get a more realistic population estimate in fields where plants are grown
for food, as well as more realistic mortality rates due to insecticides.
Granted, if insect farmers don’t use recycled feed—for instance,
spent grains from breweries, or the corn mash left over from ethanol
production—then the scales tip back in favor of plant production. In that
case, we’d have to count both the insects killed in the creation of the feed
and the insects being raised for food. But of course, many insect farmers
do use recycled feed, as it’s far cheaper. And when I went to an industry
conference, it was obvious that producers who don’t already use waste
products are extremely interested in moving in that direction. (This isn’t
because they care about the expected utility calculation; it’s because it’s
dramatically cheaper to use spent grains. One farmer told me that he
buys his for $147 per tonne.) In any case, if you raise insects at home, you
can guarantee that the expected utility calculation comes out the way you
want, since you can feed your colony using food waste. The upshot here
is that an expected death assessment doesn’t come out squarely in favor
of plant production, and probably favors insect farming.
5.5 Conclusion
Utilitarians need to be careful to separate the ethics of production, the
ethics of purchasing, and the ethics of consumption. And when they con-
sider the ethics of purchasing and consumption, they need to be careful
not to overestimate the probability of an individual making a difference.
But in this chapter, we’ve ignored all that, assuming that we can generate
a utilitarian ethics of eating by assessing the relative harm footprints of
various meals (where the harm footprint of a meal is determined by assess-
ing the average effect of producing it). If we make this assumption, then
it turns out that utilitarianism doesn’t require you to be a vegan: instead,
it requires you to eat unusually. You should be eating some animal prod-
ucts from the “animal-friendly” farm; you should be scavenging from
dumpsters and department fridges; you should be pulling mussels out of
their shells; and you should sample some crickets, termites, and their ilk.
In my experience, many utilitarians are simply willing to embrace these
results if they judge that I’m right about what maximizes expected utility.
So I don’t offer these arguments as objections to utilitarianism. Rather,
I simply want to make the case that utilitarianism clearly does not require
veganism—at least if we use “veganism” to refer to a strict plant-based
diet. You don’t make a difference, so unless animal products somehow
harm you, the expected utility of purchasing and consuming decisions
isn’t negative; if those products benefit you, then the expected utility is
probably positive. And even if you did make a difference, you shouldn’t
be vegan: you should eat unusually.
Notes
1. Joel MacClellan (2013) points out that it’s an empirical question as to
whether there is an animal such that utility is maximized by killing and eat-
ing it. And, he suggests, “[i]t is intuitively plausible . . . that a whale fits
[this profile]. Indeed, it would be rather surprising if the pleasure resulting
from eating whale meat did not yield higher overall utility than the suffer-
ing inflicted on the whale,” given the enormous number of people who can
be fed by it (61). MacClellan overlooks the impact of whaling on the many,
many animals that would otherwise feed on whale carcasses, which might tilt
the balance in favor of having humans abstain.
2. This is the payoff of welfare footprint arguments, discussed in most detail in
Saja 2013.
3. For extensive discussion of this argument, see Višak 2013.
4. An older argument of a similar stripe is “the Logic of the Larder”—so-named
by Henry Salt—according to which we do animals a favor by bringing them
into existence to be slaughtered for our purposes, since they wouldn’t exist
otherwise, and coming into existence is a benefit (at least as long we give
them good lives). The Replaceability Argument doesn’t assume that existence
is a benefit, and the Logic of the Larder doesn’t assume that sentient beings
are replaceable. So they’re clearly distinct arguments. Still, it’s often very
hard to know which of these arguments someone has in mind, as remarks
along these lines tend to be made rather quickly. You can find arguments
in this ballpark in Frey 1983, Crisp 1988, Hare 1993, Scruton 2006, and
Belshaw 2015—among many, many others.
5. The merits of these arguments depend, in part, on hard questions about
whether and how the welfare of merely possible beings always counts in the
utilitarian calculus, as well as whether merely possible beings have levels of
welfare at all. If these issues can be finessed in ways that favor the Replace-
ability Argument or the Logic of the Larder, then there remains the charge
of speciesism. Suppose, for example, that we were to apply the same line of
reasoning to humans. Those with severe cognitive disabilities may well be
replaceable in whatever sense a healthy pig is replaceable. What follows?
Likewise, may we bring infants into existence as organ donors—as long as
their short lives are pleasant—since they wouldn’t exist otherwise, and com-
ing into existence is a benefit?
6. First, those who run these arguments rarely factor in the environmental costs
of animal agriculture. Second, there are worries about the inefficiency of
animal agriculture, which ties up resources that could be devoted to other
projects (such as famine relief). Admittedly, not everyone is impressed by
the inefficiency argument for abstaining from animal products. For the best
critique, see Fairlie 2011. It’s worth noting that Fairlie still supports reducing
animal product consumption dramatically. Moreover, the consumption he
does defend is partly based on backyard agriculture—an unrealistic option
for most people. Finally, even if there are farms where animals do live good
lives, it’s unclear whether ordinary consumers are in a position to determine
as much. It should come as no surprise that products marketed as humane
often aren’t, even from reputable suppliers. (Whole Foods comes to mind.)
And if consumers aren’t in a position to make such determinations, it isn’t
clear whether these arguments justify their animal product consumption.
7. For a parallel argument in an Australian context, see Archer 2011.
8. Bruckner just argues for a conditional: if we accept DeGrazia’s principle, then
we ought to collect and consume roadkill. For ease of exposition, though,
I’m assuming the conditional’s antecedent, as Bruckner in fact does. And yes:
he follows through on the conditional’s consequent. Also, this isn’t a utilitar-
ian argument, both because of the support clause—which can but needn’t
be glossed in utilitarian terms—and because of the restriction to extensive
harm. But if there’s a solution to the causal inefficacy problem, then support
can be glossed in utilitarian terms and the argument will go through. If there
isn’t, then utilitarian considerations may still support roadkill consumption;
recall the inverted version of Garrett’s argument. In any case, DeGrazia’s
principle is supposed to be acceptable to those from different moral perspec-
tives, utilitarians included.
9. That is, unless consciousness is simply “left over” from some earlier point in
evolutionary time, when the ancestors to the relevant organism were indeed
conscious. But there is no evidence of that here, and in any case, selection
pressures would still favor removing the capacity if it served no purpose
and demanded significant resources. (Think of the way cave-dwelling species
eventually lose their eyesight.)
10. See Cox 2010. Strict vegans will probably balk at eating insects and oysters
based on some sort of precautionary principle. However, it’s likely that pre-
cautionary arguments actually support eating insects and oysters. More on
this below.
11. Mark Budolfson (2015) also argues that we’ve overlooked some of the

harms involved in plant agriculture. Suppose our concern is to have the diet
with the smallest welfare footprint—or, at least, the one that falls below
some threshold. Then, we should note that not every vegan meal beats every
meal that includes animal products, since some plant products—such as qui-
noa, avocados, blueberries—are associated with environmental harms, are
expensive in terms of land or water use, or are implicated in some sort of
economic injustice (e.g., they’re harvested in ways that tend to involve the
exploitation of migrant labor, or their sale in developed countries makes
them prohibitively expensive for people in the region where they’re grown).
Once we factor in these harms, the scales tip even more clearly in favor of the
mussel-including meal.
12. Some seem to embrace this line. David DeGrazia argues for veganism, and
yet says that “[h]ighly virtuous people may wish to give [invertebrates] the
benefit of the doubt and abstain from eating them. My view does not con-
demn eating these animals” (1996, 289).
13. See Adamo 2016 for more detailed theoretical arguments against insect

consciousness.
14. www.usda.gov/nass/PUBS/TODAYRPT/cropan15.pdf
15. At least if we assess by weight. Mealworms are actually more protein-dense
than soybeans, so if that’s the metric, the numbers will come out somewhat
differently. But the basic point will go through without this complication,
and in any case, ignoring it is a gift to the critic of entomophagy.
16. For an argument along these lines, see https://reducing-suffering.org/

vegans-should-not-eat-insects-a-reply-to-fischer-2016/.
17. Pearse (1946) estimates about 125 million invertebrates per acre in the Duke
Forest. Sabrosky (1953) reports that a team found nearly 10,000 arthropods
per square foot in a “scrub oak area” in Pennsylvania, which is over 400 mil-
lion per acre. Menhinick (1962) counted approximately 26,500 invertebrates
in 3.7 m³ test plots in Cornell’s apple orchards, which is about 107 million
per acre. The lowest estimate I could find is due to Barratt et al. (2009), who
found densities between 8 and 40 million invertebrates per acre in New Zea-
land grasslands, depending on when they sampled relative to a burn. Surely
there are important differences between soybean fields and these environ-
ments, but I’m not aware of any data on total invertebrate populations in
soybean fields, and so these numbers have to be used as proxies.
6 The Rights View and the
Production/Consumption Gap
Over the course of the last two chapters, I’ve argued that utilitarianism
doesn’t require most of us to be vegan. If it requires anything of you with
respect to your diet, it’s that you eat unusually. Proponents of animal
rights will say, “I told you so.” Here is Tom Regan, the foremost propo-
nent of animal rights, on utilitarian defenses of veganism:
Judged by its own standards . . . the utilitarian basis for [veganism] is far
from compelling. . . . [But the] rights view succeeds where utilitarianism
fails. My acting as duty requires does not depend on how many others
act similarly, and no [vegan] should be deterred from his or her course
because of the many who continue to support the animal industry or
because it is uncertain whether and, if so, when and how one individu-
al’s abstention makes a difference—for example, to how many animals
are thereby spared the abusive factory farming. The individual is right
not to purchase the products of an industry that violates the rights of
others, independently of how many others act similarly, and the case
against the animal industry does not stand or fall, according to the rights
view, on the individual’s knowing, or in any individual’s knowing, the
aggregate balance of goods over evil for those affected by allowing fac-
tory farming or by not allowing it. Since this industry routinely violates
the rights of these animals . . . it’s wrong to purchase its products. That
is why, on the rights view, [veganism] is morally obligatory.
(2004, 350–351)
Is Regan correct that the rights view succeeds where utilitarianism fails?
In what follows, I’ll argue that it does no such thing: it fails to establish
the appropriate connection between the rights violations of which pro-
ducers are guilty, on the one hand, and consumer behavior, on the other.
6.1 Regan
Regan summarizes the rights view as follows:
Some nonhuman animals resemble normal humans in morally

relevant ways. In particular, they bring the mystery of a unified
90 Rights and the Production/Consumption Gap
psychological presence to the world. Like us, they possess a variety of
sensory, cognitive, conative, and volitional capacities. They see and
hear, believe and desire, remember and anticipate, plan and intend.
Moreover, what happens to them matters to them. Physical pleasure
and pain—these they share with us. But also fear and contentment,
anger and loneliness, frustration and satisfaction, calming and impru-
dence. These and a host of other psychological states and dispositions
collectively help define the mental life and relative well-being of those
(in my terminology) subjects-of-a-life we know better as raccoons
and rabbits, beaver and bison, chipmunks and chimpanzees, you and
I. . . . [Both] human and nonhuman subjects-of-a-life, in my view, have
a basic moral right to respectful treatment. Of course, moral positions
can be advanced that either dispense with rights altogether or, while
affirming the rights of human beings, deny them in the case of nonhu-
man animals. But . . . such views appear to be deficient—for example,
because they are inconsistent or needlessly complicated, because they
lack precision or adequate scope, or because their implications clash
with a large body of are well considered moral beliefs, our moral
intuitions. The basic moral right to respectful treatment places strict
limits on how subjects-of-a-life may be treated. Individuals who pos-
sess this right are never to be treated as if they exist as resources for
others; in particular, harms intentionally done to anyone subject can-
not be justified by aggregating benefits arrived by others.
(2004, xvi–xvii)
In what follows, I’ll assume that this view is correct. What does it imply
about animal agriculture? According to Regan, the implication is that all
animal agriculture should be abolished, regardless of how “humane” it
may be. All such agriculture involves treating animals as our resources,
and even when they live the best lives imaginable, harms to them can’t be
justified by aggregating benefits to us.
But what does this mean for purchasing or consuming animal prod-
ucts? In the passage quoted earlier, Regan simply says that you shouldn’t
purchase the products of an industry that routinely violates the rights of
animals. Why do these rights violations generate obligations for individu-
als who are far removed from the process of raising and killing animals?
Here’s what he says elsewhere:
To treat farm animals as renewable resources is to fail to treat them

with the respect they are due as possessors of inherent value. . . .
Since . . . the current practice of raising farm animals for human
consumption fails to treat these animals with respect, those who
support this practice by buying meat exceed their rights. Their pur-
chase makes them a party to the perpetuation of an unjust practice.
[Veganism] is not supererogatory; it is obligatory.
(2004, 346)
Rights and the Production/Consumption Gap 91
On the face of it, then, Regan’s argument goes something like this:
1. Animals possess inherent value.

2. If a being possesses inherent value, then it’s unjust to treat it as a
renewable resource.
3. Raising animals for human consumption treats them as renewable
resources.
4. So, it’s unjust to raise animals for human consumption.
5. If a practice is unjust, then it’s wrong to be “a party to the perpetua-
tion” of it.
6. By purchasing animal products, you become party to the perpetua-
tion of an unjust practice.
7. So, it’s wrong to purchase animal products.
For present purposes, everything turns on (5) and (6). On one

interpretation—which certainly isn’t the only one—you become a party
to the perpetuation of an unjust practice if you contribute causally to
that practice. But if that’s the right reading, then the causal inefficacy
problem is of obvious relevance, and it’s unlikely that (6) is true: the
odds are very good that you don’t make a difference, as discussed ear-
lier. To make the argument work, Regan would need a much stronger
premise, namely, that if a practice is unjust, then it’s wrong to take even
a minuscule chance of being party to the perpetuation of it (i.e., cause
or help to cause it). This premise will get you the conclusion that Regan
wants, but it faces some pretty serious problems. First, it seems entirely
insensitive to the massively more-likely good consequences that actions
may have. If you are buying a meal for a homeless person, and the person
wants a hamburger, you can be quite confident that you’ll better respect
and satisfy that person’s desire—and provide more calories—if you buy
that person a hamburger rather than a black bean burger. If the odds
are extraordinarily low that you’re causing any harm by buying the beef
patty, why shouldn’t the likely good consequences be relevant?
Second, it threatens to imply that the vast majority of our actions are
morally wrong: since causal chains are so complex, any number of our
actions involve a very small chance of being party to the perpetuation of
an unjust practice, even if those actions have no obvious relationship to
the objectionable behavior. When I order black coffee at the coffee shop,
I support a business that sells animal products: many of the products
there aren’t vegan. So, I’m taking a minuscule chance of being party to
the perpetuation of various unjust practices: dairy production, egg pro-
duction (in their baked goods), and so forth. This seems too demanding
a standard.1
Third, this way of developing the argument doesn’t fit well with Regan’s
argument that the rights view is superior to utilitarianism in the way it
grounds the duty to be vegan, namely, that it isn’t supposed to depend
on empirical questions about the likelihood of your having an impact
(recall the passage at the beginning of this chapter). But if Regan runs
the argument using this premise, then he’s essentially conceding that such
considerations are indeed relevant, since if your purchase had no chance
whatever of causally contributing to the perpetuation of factory farming,
then the action wouldn’t be wrong on these grounds.
Clearly, then, Regan should offer a different interpretation of what’s
involved in being a party to the perpetuation of an unjust practice. What
might he say?
6.2 Benefiting
One possibility is to insist that it’s wrong to benefit from an unjust prac-
tice. This helps with the gap between purchasing and consuming, as well
as with preserving the uniqueness of the rights-based approach to the
issue. However, it leaves two of the problems discussed already: first,
the difficulty of insensitivity to the significance of the benefits relative to
the insignificance of the contribution to injustice, and second, the explo-
sion of moral responsibility, in the sense that we benefit in countless ways
from an extraordinary amount of injustice. (An example to drive home
this latter point: many of my students consume animal products, and
that food provides them with the energy that allows them to partici-
pate in good discussions in my courses; I benefit from those discussions
going well, as I enjoy them and find satisfaction in having run class ses-
sions in which the students are engaged. So, I’ve benefited from an unjust
practice—namely, animal agriculture. But surely I haven’t done anything
wrong in teaching those classes.) Moreover, even if we can bite these
bullets, it’s worth noting that the standard is independently implausible.
Consider a case from Tyler Doggett (2018):
A terrorist bomb grievously injures Bob and Cece. They attend a sup-
port group for victims, fall in love, and live happily ever after, leaving
them significantly better off than they were before the attack.
Plainly, Bob and Cece haven’t done anything wrong, but they have indeed
benefited from an unjust act.
Maybe it isn’t always wrong to benefit from wrongdoing, but it’s
plausible that it’s sometimes wrong. Consider, for instance, the “chef in
shackles” case:
Alma runs Chef in Shackles, a restaurant at which the chef is known

to be held against his will. It’s a vanity project, and Alma will run the
restaurant regardless of how many people come. In fact, Alma just
burns the money that comes in. The enslaved chef is superb; the food
is delicious.
(Doggett, 2018)
Presumably, many of us would think that it would be wrong to eat at
Chef in Shackles. But if you eat animal products, aren’t you doing essen-
tially the same thing?
No: the cases aren’t analogous. Here’s a case that is analogous, or at
least much more so:
Nearly all restaurants and grocery stores are run by slave labor.
Nearly everyone goes to these restaurants and grocery stores, and
even those who don’t still end up supporting them indirectly, as
there are economic links between all the suppliers in the food indus-
try. Since businesses that use slave labor can force their workers to
meet the highest culinary standards, they tend to produce foods that
people find far more satisfying than what you’ll find at the “free”
restaurants and grocery stores. And regardless of whether you go
to slave-labor-based restaurants and grocery stores, these businesses
will continue to take advantage of slave labor.
Is it wrong to give your money to one of these businesses? I grant that

it seems suboptimal. However, I suspect that if we actually lived in this
world, I think most people would say, “Well, you should make an effort
to support the ‘free’ businesses, but don’t get hung up on perfection.
There will be plenty of cases where it makes most sense to go to a slave-
labor business, and since you can’t make a difference, it’s fine to do
that.” In other words, if Chef in Shackles is a single establishment, it
certainly seems wrong to go; but when you live in a Chef in Shackles
world, it’s implausible that you have such a strict duty. I expect we’d
agree that it’s good to support the free establishments and thereby avoid
supporting the use of slave labor; we just wouldn’t think that it’s usually
obligatory.
If this is right, then it follows that people have a good reason to reduce
their animal product consumption. I agree. But it’s perfectly compatible
with this to say that the average person at the (non-vegan) burger stand
isn’t acting wrongly. We should certainly be willing to acknowledge that
we have a good reason for reducing, and the average person may fail in
this regard. But many don’t: they acknowledge the reason, and they just
think that it’s outweighed in the relevant case. If the reason is relatively
weak—as I think it is, given their inability to make a difference—they
may well be correct.
To make this a bit more plausible, consider the alternative. Someone
might insist that in the face of injustice, the only limit on the duty to
disassociate is supplied by the ought implies can principle. That is, you
ought to disassociate from as many injustices as you can (perhaps ranked
in order of severity, though that doesn’t matter much for present pur-
poses), and when you can’t disassociate from any others, you’ve reached
the limits of your obligations.
This view isn’t particularly plausible. First, it’s generally thought to be
a selling point of deontological moral theories that they draw a reason-
ably intuitive distinction between obligatory and supererogatory actions.
It’s within our power to disassociate much, much more extensively than
we do from known injustices, making our obligations highly demanding.
But with the exception of some philosophers, no one seems to believe
that our obligations are that extensive, and so it doesn’t seem that this
principle draws a plausible distinction between obligatory and supere-
rogatory actions.
Second, because of the sheer number of injustices in the world, this
approach will regularly give us conflicting advice. Lots of vegan meals
contain ingredients behind which there are injustices against both humans
and animals. (Consider, for instance, that even in US, some tomatoes are
picked by slaves, as well as the way in which field animals are harmed
in plant agriculture.) To avoid this problem, the natural move is to say
that you ought to minimize your associations with injustice. But it’s then
an empirical question whether being vegan minimizes your associations
with injustice. As I’ll argue in the next chapter, I don’t think that’s true.
6.3 McPherson
All that said, perhaps there is a better way to refine the proposal, “It’s
wrong to benefit from wrongdoing”. Tristram McPherson thinks so: he
argues that it “is typically wrong to aim to benefit by cooperating with
the wrongful elements of others’ plans” (2015, 83). Since factory farm-
ing is wrong, and grocery stores intend to profit by cooperating with the
wrongful plans of that industry, they act wrongly. And since they act
wrongly, consumers act wrongly when they cooperate with the wrong-
ful plans of grocery stores. But this account runs into problems as well.
First, although McPherson says that it’s only “typically” wrong, I doubt
that he’s hedging enough. I think that there are many aspects of higher
education that are unfair to students, and even exploitative. Neverthe-
less, I teach philosophy at a university, and thereby cooperate with the
wrongful plans of others. However, my contribution to that wrongdoing
is small, and I hope that I’m doing some good.2 As a result, it seems to me
that my behavior is permissible, despite its being in violation of McPher-
son’s principle.
Moreover, we should note that either the notion of cooperation is sen-
sitive to your intentions, or it isn’t. Suppose that it is. Then, consider
anyone who thinks, “If I thought that this purchase would make a dif-
ference, I wouldn’t do it; but since I’m convinced that it doesn’t, I will.”
It’s implausible that such a person is cooperating with the wrongful ele-
ments of others’ plans; the person certainly doesn’t intend to give any aid
to animal agribusinesses, as shown by the fact that this person wouldn’t
buy the products if she thought that there were any real risk of making
a difference. So, such a person isn’t doing anything wrong when buying
animal products: this person aims to benefit, but not by cooperating.
(This may be why, in the revised Chef in Shackles case, it seems implau-
sible that it’s always wrong to go to a grocery store that uses slave labor.
When one grocery store uses slave labor, your shopping there looks like
cooperation with the wrongful plans of others. When virtually all grocery
stores use it, then your shopping there doesn’t have the same symbolic
significance.) On the other hand, suppose that your intentions aren’t
related to whether you’re cooperating. Then, we need to know what it is
to cooperate. McPherson can’t say that that notion should be interpreted
causally, lest he face the causal inefficacy problem. Until McPherson fills
in the story, either we have an account that doesn’t seem to work, or we
have no account at all.
6.4 Driver
Julia Driver (2015) might be able to help out here, despite not being a
proponent of the rights view, as she takes up a similar sort of objection
in a discussion of complicity. She treats cooperation as unproblematic
and uses it to give an account of participation: she claims that we don’t
need to intend to participate in an activity in order to participate in it; it’s
sufficient to “[cooperate] with others who are intending to bring about
the outcome in question” (76). However, let’s not press her there. The
interesting suggestion is that even if you’re participating in a wrongful
activity, the participation itself may not be wrongful; on her view, we
should distinguish between complicity and wrongful complicity. The sug-
gestion is this:
It may very well be that complicity is like causation in that there are
causes everywhere—and yet, when we pick out or identify something
as a cause, or as the cause, we are guided by pragmatic considerations—
such as considerations, in the case of causation, that involve some
sort of norm violation. In the case of complicity there are all sorts of
things that people are involved in, knowingly, that have some con-
nection to wrongdoing. One thing that impacts complicity is how
tight the causal connection is between, let’s say, the purchase one is
making in the wrongdoing in question, particularly when there is a
great deal of intervening agency. Another, and more important for
the discussion here, is how many options one has. And yet another,
as is the case with causation, has to do with [the degree to which
one can avoid supporting the wrongdoing.] One can make better or
worse choices in this regard. Someone who is wrongfully complicit
will be someone who has not made [the effort to minimize support
for the wrongful activity.] In [the case of someone who wouldn’t eat
meat if she thought that abstaining would make a difference, but eats
it because she thinks her actions are inefficacious], eating the meat
is supporting the industry in a situation where there [are] plenty of
other, better, options. . . . What makes her complicit is that she is a
participant. What makes that participation morally problematic . . .
is that the eating of meat displays a willingness to cooperate with the
producers of a product that is produced via huge amounts of pain
and suffering.
(2015, 78–79)
On Driver’s view, then, we can give very thin accounts of notions like
“participation” and “cooperation,” as we aren’t trying to use those
notions to do all the moral work; the position doesn’t require us to distin-
guish permissible from impermissible actions (or relatively unproblematic
from relatively problematic actions) simply on the basis of whether the
action constitutes participating or cooperating. Instead, we can appeal to
other kinds of considerations. These include causal ones, but also expres-
sive ones, such as the degree to which your behavior “displays a willing-
ness to cooperate” with objectionable practices. If you have lots of other
options available to you, and it’s very easy to pursue them, then eating
meat would display that willingness to cooperate; if not, then not.
Driver wants to think about the “displaying” here in terms of a display
of character, not in terms of, say, what your behavior communicates to
others about your values. But I don’t understand why you display a poor
character if you act in a way that you think—justifiably—won’t hurt any-
one, and won’t make any other sort of difference in the world. Suppose
that I’m at a wedding where they’re serving chocolate cake. I know some-
thing about the chocolate industry, and so I know that this chocolate is
probably the product of child labor, slavery, and/or significant harms to
the environment. But the leftover cake will simply be thrown away, as the
happy couple has ordered far more cake than anyone will consume, so
it’s very clear that my action makes no difference. In such circumstances,
I don’t find it plausible that eating the cake displays willingness to coop-
erate with objectionable practices. Instead, my sensitivity to difference-
making displays the opposite. If I wouldn’t eat the cake if I thought that
abstaining would make a difference, then I’m clearly unwilling to cooper-
ate: I don’t want to help bring about those objectionable states of affairs.
Of course, Driver could revise her view, instead cashing out “display-
ing” in terms of what your behavior communicates to others. This won’t
help, however. Eating animal products communicates to vegans that you
are willing to cooperate with objectionable practices. But it would be
circular to appeal to that, since it’s plausible that vegans interpret the
significance of eating animal products based on their prior commitments
about that action being wrong, and the wrongness is what we’re trying
to establish. To non-vegans, by contrast, eating animal products doesn’t
clearly communicate much of anything. It’s just eating.3
Driver can cash out “displaying” in yet another way. She writes that
it would seem rule-consequentialist theories would have an advan-

tage here: what is problematic [about the person who wouldn’t eat
meat if she thought that abstaining would make a difference] is that
her eating of meat is inconsistent with a policy, such that if that
policy were adopted, it would have enormously good consequences
by her own lights. . . . Given her normative commitments, [she] must
agree that if enough other people stopped eating meat the effects
would be very good indeed. [She] is displaying an unwillingness to
engage in the cooperative enterprise of ending animal suffering.
(2015, 74)
But note that Driver has abandoned the original target, namely, showing
that the person is willing to cooperate with wrongdoing, and replaced it
with a new one: showing that the person is unwilling to cooperate in the
project of ending animal suffering. This move matters a great deal. First,
even if you grant the change, it doesn’t get you veganism: it gets you high-
welfare farming, since the objective is the end of animal suffering, not
ending animal death. Second, while it’s plausible that we shouldn’t coop-
erate with wrongdoing, it’s hardly obvious that we’re always obligated
to cooperate with morally progressive causes, even when any particular
action is relatively easy to perform. It’s indeed easy to be a bit better
about recycling, or to spend a bit more time volunteering, or to donate
just a bit more to famine relief. But it isn’t easy to do all those things all
the time, and partially for that reason, we generally don’t expect people
to make every cause their own. There’s a division of moral labor: people
have morally valuable projects that they pursue, neglecting other morally
valuable things in the meantime; it’s good that this happens, as projects
need champions. Of course, there’s something to the idea that people
should “do their part.” But Driver hasn’t shown that, of all the things we
might ask of people, only strict veganism counts as “doing your part.”
And that’s crucial here.
Could Driver use the rule-consequentialist story to say, of the person
who eats meat but wouldn’t if she thought that abstaining made a differ-
ence, that she’s willing to cooperate with wrongdoing? I don’t think so,
as there’s a good reason why we shouldn’t assess behavior by this stand-
ard. There are lots of policies such that if they were adopted, they would
have enormously good consequences. If everyone were willing to limit
themselves to a single outfit, we would massively reduce the human rights
violations and environmental problems associated with the textile indus-
try. But it seems ridiculous to say, of someone with two or more outfits,
that she’s willing to cooperate with the wrongdoing in that industry. Our
judgments about what actions symbolize are—and should be—sensitive
to what it’s reasonable to expect of people in a context. It’s unreasonable
to expect people to limit themselves to a single outfit. And if current rates
of veganism are any guide, it’s unreasonable to expect veganism too.
6.5 Hooley and Nobis

Let’s consider one more way of fixing Regan’s argument. Perhaps we
should abandon the idea that there is going to be a single notion that
forges a connection between the ethics of producing animal products
and the ethics of purchasing them. Instead, perhaps we need something
much more complex, a principle that’s sensitive to various dimensions of
the decision to abstain from animal products. Dan Hooley and Nathan
Nobis (2015, 99) offer just such a principle, which suggests a novel way
of completing Regan’s case for veganism:
If a product is such that (a) its production causes serious and unjusti-
fied harms and so is morally wrong, (b) the product can be avoided,
(c) avoiding the product would not seriously harm the boycotting
individual, (d) there are readily available alternatives to that product,
(e) the boycotting individual might benefit from boycotting, (f) the
probability that not purchasing or consuming the product will lessen
or eliminate the wrongdoing is equal to or greater than the probabil-
ity that purchasing or consuming the product will lessen or eliminate
the wrongdoing, and (g) boycotting will make that individual a mem-
ber of a morally progressive group that opposes a wrongdoing, then
individuals are obligated to not purchase or use that product.
Hooley and Nobis’s principle isn’t terribly plausible. For instance, this
principle makes it morally obligatory not to purchase or use a product
even when, based on condition (b), the product can be avoided only at
significant cost; based on condition (c), avoiding the product would harm
the boycotting individual; based on condition (e), the boycotting indi-
vidual in fact doesn’t benefit from boycotting; based on condition (f), the
probability of lessening or eliminating the wrongdoing is zero; and based
on condition (g), boycotting makes the individual a member of a less
effective morally progressive group than the one in which she could be a
part. Moreover, even if they patch up the principle to block these implica-
tions, it’s worth noting that we are unlikely to have strong, pre-theoretic
intuitions about a principle that’s so complex. As a result, those inclined
to accept the conclusion are likely to accept the principle; those not, not.
And finally, even if they patch up the principle to block these implica-
tions, those sympathetic to the principle ought to consider it alongside a
weaker one with the same antecedent and a different consequent: namely,
if a product is such that conditions (a) to (g) hold, then while it isn’t mor-
ally wrong to purchase or use that product, it would be morally good not
to purchase or use it. When we recall the many sources of complicity in
contemporary life, I doubt that Hooley and Nobis’s principle is the more
plausible of the two.
6.6 Korsgaard
At this juncture, someone might well argue that we’ve gotten off on the
wrong foot. We shouldn’t try to bridge the gap between the ethics of
production and the ethics of consumption via some theory of complicity:
perhaps we already have the resources to forge the link in a simple Kan-
tian argument. This is Christine Korsgaard’s (2018) view. She writes this:
The question is . . . about you and a particular animal, an individual

creature with a life of her own, a creature for whom things can be
good or bad. It is about how you are related to that particular crea-
ture when you eat her, or use products that have been extracted from
her in ways that are incompatible with her good. You are treating her
as a mere means to your own ends, and that is wrong.
(2018, 12.3.3)
There are several things to say here. First, I think we should be careful
about the claim that seems to be implicit: namely, that using an ani-
mal product is tantamount to using an animal. In the case of using dead
human bodies, that’s a reasonably plausible claim. For instance, if some-
one uses a person’s skin to make a lampshade, he’s used that person in an
objectionable way. But presumably, that’s because we have expectations
and desires about how our bodies will be treated upon our demise, and
by ignoring those expectations and desires, we convey that they weren’t
worthy of our respect, that their wishes don’t constrain us in the normal
way. However, animals don’t have any desires about how their bodies
are treated after their deaths—or, at least, there is no reason to think
that they have any such desires—and so whatever link there may be
between the use of a dead animal body and the use of the animal, it isn’t
so straightforward. Put another way, it’s very clear that we shouldn’t use
living animals as mere means; however, because animals don’t have the
kind of interests in the fates of their dead bodies that we have, it isn’t
clear why we shouldn’t use dead animal bodies as mere resources. So,
it isn’t clear why, by eating part of a dead animal’s body, you treat her
as a mere means. (Not incidentally, this is why it’s going to be difficult
for proponents of animal rights to reject the consumption of roadkill or
found animal products. More on this in the next chapter.)
Korsgaard might reply that there’s an obvious sense in which you are
using her as a mere means: you paid someone to kill her. But that isn’t
obvious. If anything, you paid someone to kill some future animal, not
the one on your plate. The one on your plate was dead long before you
made a decision at the grocery store. And given the insensitivity of the
market to the behavior of any particular consumer, there is nothing you
did that explains why she was harmed, nor anything you could have done
to prevent her from being harmed. Indeed, many consumers recognize
this situation when they reflect on their choices. They say: “What’s hap-
pening to animals is terrible, and if I could do anything about it, I would.
But I can’t, and so other considerations win out when it comes to my
food choices.” Perhaps people who say such things are being disingenu-
ous: maybe they wouldn’t do anything even if they could. But until we
establish that, it isn’t clear why we should say that they are using a par-
ticular animal as a mere means.
Korsgaard might run a different objection. If the earlier hypothesis
about the wrongness of disrespecting dead human adults is correct, then
it should be fine to eat dead babies. After all, they also don’t care what
we do with their bodies. But this simply doesn’t follow. Our norms about
respectful treatment of dead bodies aren’t so individualistic; they depend
on common desires and expectations, buttressed by their place within a
range of longstanding cultural practices. Granted, this is a modest kind
of cultural relativism, but it isn’t relativism about the value of respect;
instead, it’s relativism about how respect is manifested, and so provides
exactly the right amount of cultural variability as to what does and
doesn’t constitute use as a mere means.
This reply invites a third objection. If “our” culture shows respect for
people by not consuming their dead bodies, then doesn’t respect require us to
abstain from eating dead animal bodies? Not obviously, as respect norms are
sensitive to the boundaries of the community. Imagine that we come across a
community of human beings who burn rather than bury their dead. One of
those people dies and we end up being responsible for the body. Would it be
wrong to burn the body? Obviously not. Indeed, we ought to burn the body,
even if that isn’t how we show respect for our own.
Of course, this example involves deference to another respect norm.
What should happen in the case of dead animal bodies, where there is no
other norm to which to defer? Shouldn’t we follow our own norms there?
Again, not obviously. It depends on whether there’s a presumption in
favor of extending the norm. There is such a presumption in cases where
we don’t know enough about the values of the individuals in question.
Imagine finding an isolated group of human beings who, tragically, died
as result of some mysterious illness. We don’t know how they handle
their dead, so we have to decide how to proceed. In such a case, bury-
ing seems like the right thing to do. However, we aren’t in this kind of
situation with nonhuman animals. We’ve got ample evidence that many
of them don’t care for their dead the way we do, that the dead bodies of
their kith and kin don’t hold anything like the significance for them that
they do for us. In such circumstances, I don’t see why there would be
a presumption in favor of extending our norms about the treatment of
dead bodies.
Moreover, we should note that our society, like many others, already
has norms about the respectful treatment of dead animal bodies: namely,
using them completely as a way of showing respect for their lives. On this
view, we shouldn’t let parts of animals “go to waste” and we should be
careful to “use the whole animal,” as anything else wouldn’t show the
right reverence for the animal’s life. Korsgaard might condemn this as
speciesist, but I don’t think she could make that charge stick. Granted, it’s
often wrong to create a dead animal body—as that’s done by killing a liv-
ing animal. And many of the people who say that you shouldn’t let parts
of animals “go to waste” are themselves speciesist. But this doesn’t show
that the norm itself is speciesist. To the contrary, it seems like a perfectly
reasonable way of showing respect for beings who are quite unlike us in
important ways, namely, in the degree to which they have views about
the treatment of their own dead bodies, as well as in the degree to which
they’re concerned about the treatment of other dead bodies.
So Korsgaard still owes us an account of why eating meat (for instance)
counts as using the animal from which the meat was taken. If she starts
to appeal to something like the causal connection, or the benefit that you
derive, or what have you, then she will be left with the sorts of prob-
lems already surveyed. She might say, instead, that it counts as using the
animal as a mere means because the animal was treated wrongly in the
first place. But it isn’t generally true that if I use a product that some-
how derives from wrongdoing, I thereby use the person who was initially
wronged. If that were true, then we would all be guilty of using every
victim of historical injustice whenever we use something that exists only
because of their exploitation. Granted, what was done to those individu-
als was wrong. And we may have various collective responsibilities now,
as beneficiaries of all those injustices, to respond in one way or another
(e.g., by offering reparations to descendants of victims). The point is just
that you aren’t now guilty of using the past victim. Or, if we are guilty
of using them, we aren’t guilty of it in a way that makes our current
actions wrong. We’re the beneficiaries of wrongdoing, but not ourselves
the wrongdoers. And on use-as-mere-means-based grounds, the same is
true for individual consumers of animal products.
6.7 Conclusion
The upshot is simple. We’re no closer to closing the gap between the
ethics of production and consumption; what’s more, the presence of this
gap amounts to a serious problem for rights-based arguments for vegan-
ism. Still, these kinds of problems tend to be written off as puzzles to be
solved. And that’s understandable. It’s hard to come up with principles
that don’t admit exceptions; it’s hard to provide an account of benefitting
or complicity or use that doesn’t generalize in objectionable ways. So, if
the basic rights-based approach seems plausible, it can seem reasonable
to think that after sufficiently many iterations, we’ll find the right version
of the argument to get the conclusion. These problems, someone might
think, don’t show that the rights view fails to underwrite an obligation to
be vegan; they just show that philosophy is hard.
This line of thought is appealing, but I think it’s mistaken. By and
large, this is because proponents of the rights view—and many others—
have tried hard to address these problems. However, they haven’t had
much luck, and eventually, the absence of a solution is evidence that
there isn’t one.
Someone might insist that we have good reason to keep searching, as
the position I’m defending has unacceptable consequences elsewhere in
consumer ethics. Wouldn’t my arguments show, if they’re sound, that
even those who are able have no obligation to buy Fair Trade coffee (if
they’re going to buy coffee), or clothing from companies that don’t use
sweatshop labor (if they’re going to buy new clothing)?
There are two things to say here. First, I find it independently plausible
that people don’t have such obligations, so if these conclusions follow,
then so be it. Obviously, intuitions diverge here, but since this isn’t a
book about consumer ethics generally, I won’t try to adjudicate the dis-
pute. Second, it’s important to see that not everything I’ve said general-
izes to all consumer cases. Consider my reply to Korsgaard: what I say
there only applies to consumer actions pertaining to nonhuman animals.
If, for instance, sweatshop laborers want us not to buy the textiles they
produce, then I can grant that buying them is disrespectful and, as a
result, wrong. (I’m not sure they want this, so I’m not sure that buying
those products is either disrespectful or wrong. Still, if that’s what they
want, I can grant the rest.) This is because other human beings have a
say in the symbolic significance of consumer actions, whereas nonhuman
animals don’t.
Plainly, then, we should be thoughtful about the symbolic significance
that we assign to consumer choices that pertain to animals. But “should
be thoughtful” doesn’t entail “should assign a symbolic significance that
implies that it’s wrong to purchase animal products.” One alternative
is to say that it’s usually bad to purchase animal products, given what
befalls the animals we farm. This alternative might seem sensible if you
think, as I do, that individual consumer choices make no difference and
that inefficacy mitigates moral responsibility. It also might seem sensible
if you worry about the “individualization” of responsibility generally,
where the focus on individuals actually distracts us from “ponder[ing]
institutions, the nature and exercise of political power, or ways of col-
lectively changing the distribution of power and influence in society—
[from], in other words, ‘think[ing] institutionally’ ” (Maniates, 2001,
33). In any case, the point here is just that it isn’t enough to say that an
individual uses a particular animal when purchasing animal products.
Much more work needs to be done.4
Some won’t find any of this compelling. They’ll be convinced either
that one of my arguments fails or that there’s some other way to close
the production/consumption gap. Even if so, however, it’s important to
see that there are other reasons why the rights view doesn’t imply that we
ought to be vegans—reasons to which I now turn.
Notes
1. It’s no good to reply, as some might, by saying something to the effect that
morality just is really demanding, and the tragedy here is just that we live in
such an awful world. Part of the point of describing actions as morally wrong
is to distinguish the actions we shouldn’t perform from those we may perform.
We can, of course, say that they are all wrong; that’s a perfectly legitimate
theoretical position. But the upshot of taking it is that the category of moral
wrongdoing is of little interest, and we now need to focus all our attention
on the category that we are going to use to guide the behavior of individuals
who can’t produce the massive, systemwide changes required for their actions
to be morally permissible. So, for instance, the real debate now will be about
whether you are morally responsible for wrongdoing, or whether you are
blameworthy for it, or whatever. Suppose that blameworthiness becomes the
key notion. Then, if Regan ends up having to say that it’s morally wrong to
purchase animal products, but you aren’t blameworthy for it, that’s tanta-
mount to having his argument fail.
2. Notice that it isn’t obvious that I’m doing much good. How much do students
really take away from my courses?
3. Not incidentally, this is relevant to another deontic principle that we might
invoke here, namely, that it’s wrong to condone unjust practices. Someone
might say that purchasing animal products condones their creation, but
I don’t see why we should accept that, given the point in the main text.
4. See Chapter 9, on activist ethics, for one voluntarist proposal.
7 Eating Animals the Rights Way
My aim in this chapter is to argue that quite apart from the preceding
challenges, the rights view doesn’t imply an obligation to be vegan. Many
ordinary consumers are in positions that would allow them to obtain ani-
mal products that don’t violate any of its requirements. In other words,
even if someone manages to address the production/consumption gap,
the rights view doesn’t require a strict vegan diet. Instead, it permits—
and perhaps requires—eating unusually, just like utilitarianism.
I’m hardly the first to argue that the rights view doesn’t require vegan-
ism. Some have thought, for instance, that Regan’s view doesn’t imply
that we ought to be vegans because of his commitment to the so-called
“liberty principle,” which is designed, inter alia, to handle lifeboat cases:
Provided that all those involved are treated with respect, and assum-
ing that no special considerations obtain, any innocent individual has
the right to act to avoid being made worse-off even if doing so harms
other innocents.
(2004, 331)1
Hugh Lehman (1988) claims that many of us are in a lifeboat (proverbially,

if not literally), and so we can raise and slaughter animals on this basis:
Humans need a diet which includes a certain range of proteins. For

many people, these proteins are obtained by killing and eating ani-
mals. Equipping current food production systems to produce veg-
etarian foods in sufficient quantities would be a massive undertaking
as would educating all human beings about alternative sources of
nutrients. Until the alternative foods were produced and the educa-
tion was provided, people would have to continue to eat meat or face
death or illness resulting from malnutrition.
(161)
This isn’t the sort of argument that I’m going to defend. Lehman’s
argument only works on the basis of the liberty principle if raising and
Eating Animals the Rights Way 105
slaughtering animals is compatible with treating them with respect. It
takes some work to show that slaughtering animals is consistent with
respecting them. But that aside, those who can reconcile these two acts
should concede that his argument doesn’t license factory farming, nor
many practices that are standard even in small-scale agriculture, such as
purchasing layer chicks from a source that grinds the males. Whatever
respect involves, much animal agriculture doesn’t display it. However,
I readily concede that if the choice is between life and death, or even a
flourishing life and a malnourished one, then the case for the permissibil-
ity of eating animals is far better. But the question in general isn’t whether
this is so, but just to whom it applies—especially in the Western world.
7.1 An Exception for Family Farms?

Terence Cuneo (2015) offers a more promising way of reconciling the
rights view with animal consumption. He’s prepared to grant that ani-
mals have rights, but argues that they don’t clearly have the right “not
to be killed for the purpose of providing nourishing food, which pro-
vides gustatory pleasure, sustains valued social practices, and provides
a viable alternative to factory-farming, assuming that those animals are
given excellent lives” (2015, 34).2 In part, this is because he denies that
we can infer that animals have this right from the more basic right not
to be killed just for the pleasure of eating them. The suggestion, I take
it, is that there may be limits on the burdens that your rights can make
others bear, and demanding that people sacrifice nourishment, gusta-
tory pleasure, valued social practices, and a viable alternative to fac-
tory farming may be to demand too much. So Cuneo is not defending
all animal agriculture—far from it. Instead, he is simply saying that
although animals have rights, and even rights not to be used as renew-
able resources when that causes extensive suffering for relatively trivial
ends, they don’t necessarily have rights not to be used as renewable
resources when they are treated excellently, and the cost of not so using
them would be more significant.
In support of this view, he offers an historical thought experiment:
Imagine [that the Native Americans who lived in the US one hundred
fifty years ago] were offered the following choice (perhaps by others
of their tribe): You may either continue your way of life or stop kill-
ing animals and become farmers or merchants. . . . If these people
were to take the former option, I take it that their justification for
doing so would be very similar to the one offered by conscientious
omnivores when asked to justify their position. By killing animals,
the native Americans would say, they thereby provide their people
with nourishing and delicious food—these activities being at the
center of a deeply entrenched and valued way of life. The question to
ask is whether they would be wronging the animals they kill if they
were to take the first option. It is not apparent that they would.
(2015, 35)
Of course, it’s hard to imagine similar statements being made about the
rights of human beings. Those who favor rights-based ethics are unlikely
to think that as long as (1) slaves live excellent lives and (2) having slaves
makes possible a deeply entrenched and valued way of life, it’s permis-
sible to own other human beings. So we need some story about why the
rights of animals function differently than the rights of humans.
What are the prospects for closing the gap? To my mind, Cuneo’s best
bet is to draw on Cochrane’s (2012) interest-based theory of rights. On
Cochrane’s view, rights are grounded in certain basic interests. And if
it turns out that continued existence isn’t one of those basic interests—
perhaps because, as Christopher Belshaw (2015) argues, the animals we
tend to raise for food don’t have the kind of future-directed mental states
that are required for death to be bad for the one who dies—then they
won’t have a right not to be killed prematurely. And if they don’t have
that right, then it becomes much more plausible that their use can be
permissible, assuming, as Cuneo does, that they live excellent lives until
their premature death.
Cochrane’s view is that animals do indeed have an interest in contin-
ued existence. He writes that
to ascertain whether sentient animals have an interest in continued

life, we must ask whether continued life makes their lives better. The
most obvious way of arguing that it does is to point to the opportu-
nities that continued life affords for pleasant experiences. After all,
it seems only reasonable to claim that if suffering is bad for animals,
then pleasant experiences are good for them. Consequently we might
say that an animal has more well-being overall in her life the more
pleasurable experiences she has in that life. Clearly when an animal
dies or is killed, the amount of possible pleasure in her life is ended.
We can thus conclude that ordinarily animals have an interest in
continued life so that they may have more pleasant experiences and
greater overall well-being in their lives.
(2012, 65)
Cochrane argues that this interest isn’t as strong as others because of

important psychological differences between human and nonhuman ani-
mals, including the kind of future-directed planning in which humans
engage, as well as the degree of psychological continuity that most
humans have with their future selves. Of course, that’s compatible with
the interest being strong in absolute terms, but if we think that death isn’t
very bad for humans with limited cognitive capacities, then the interest
probably won’t be strong in absolute terms. (Obviously, this commits
us to explaining why we ought to have a norm against killing human
beings that doesn’t track their interest in continued existence.) In that
case, Cochrane’s position ends up being quite friendly to Cuneo’s. After
all, according to Cochrane, the claim the animals make on us for contin-
ued existence can be overridden by sufficiently strong human interests.
So if Cuneo is right that we have a very strong interest in producing food
that’s nourishing, provides gustatory pleasure, sustains valuable social
practices, and provides a viable alternative to factory farming, then that
interest may well trump the weak interest that animals have in continu-
ing to exist.3
How strong is our interest in producing such food? This is a diffi-
cult question to answer. The obvious worries are these. First, plant-based
foods can be quite nourishing. Second, although I’m certainly prepared
to concede that animal products provide greater gustatory pleasures than
do plant-based foods, there’s a difficult matter of assessing just how much
greater those pleasures are, as well as how much they matter. Third, it’s
difficult to assess the value of the social practices that are associated with
raising and slaughtering animals for food. When we look at them in
isolation, ignoring their impact on animals, I find myself drawn to the
thought that there are many admirable agricultural virtues. But these vir-
tues are tainted—or, at least, made far more difficult to assess—when we
keep animals in view. Consider, for instance, there are surely all sorts of
admirable virtues that patriarchs possess in strongly patriarchal societies.
And when we ignore the impact that those men have on women, it’s easy
to romanticize their characters. But when we realize the way that those
men limit the opportunities for women to flourish and act autonomously,
we have to reevaluate the ostensible male virtues. Finally, when we con-
sider whether small family farms provide a realistic alternative to factory
farming, it seems plain that the alternative is largely symbolic. The trend
in agriculture is not toward maximally humane family farms, but toward
consolidation and sustainable intensification—that is, getting more out-
puts while minimizing inputs, largely with an eye toward the environ-
mental consequences of animal agriculture. It may indeed be valuable to
have reminders that there are kinder ways to raise animals for food, but
there is no reason to think that this is a serious economic alternative.
If I were Cuneo, I would shrug off the first and second concerns. Animal
products clearly offer benefits that plant-based foods don’t, for instance,
high protein density (which can matter considerably depending on your
other dietary restrictions), B12 and omega-3s, and gustatory experiences
that many people plainly value. Also, the fourth concern can be mitigated
when you take the long view: the suggestion was never that small family
farms are going to replace intensive animal agriculture in the near future.
Instead, the claim is that these farms preserve practical knowledge, skills,
and traditions that allow even the possibility of an alternative, both now
and down the line. The third concern is the weightiest, but perhaps the
best reply is as follows. If we already know that animals have the right
not to be killed for the relevant reasons, even after being given an excel-
lent life, then it’s appropriate to reassess the value of traditions and pur-
ported virtues. That’s the way we approach a patriarchal society: we start
off with our commitment to the view that women are being mistreated,
and then we criticize the social institutions surrounding that mistreat-
ment on that basis. But we aren’t in that situation here. Instead, we are
attempting to determine the scope of animals’ rights, which means that
it’s inappropriate to discount the traditions and virtues of family farming
at the outset. They ought to be weighed on their own.
Where does this leave us? It’s hard to say. My own inclination is to say
that we simply need to hear much more about the value of the relevant
kind of food production. Many agrarians have, of course, tried to provide
just this. Unfortunately, I suspect that urbanites like me are poorly posi-
tioned to appreciate what they have to say. In any case, I think Cuneo’s
argument does make it plausible that the rights view may sanction small-
scale humane family farming. We can’t yet rule it out.
7.2 Backyard Chickens

However, even if it turns out that it’s difficult to defend the kind of ani-
mal agriculture on which Cuneo focuses, something much more modest
still seems to be permissible. As Josh Milburn and I (2019) have argued,
the rights view seems to permit raising backyard chickens and eating their
eggs. The argument for this is fairly straightforward. From the perspective
of the rights view, the chickens at your local farm store are enslaved. They
are beings with inherent value who shouldn’t be property, and yet they are
property. In such circumstances, though you may not have a duty to liber-
ate those animals, it does seem to be permissible to do so. However, “liber-
ating” chickens by simply setting them free wouldn’t be in their interests: if
set free, domesticated chickens would quickly become food for foxes. The
goal should be to spare them from being treated badly by human beings
who wouldn’t be sufficiently concerned with their welfare (manifested in,
say, letting chickens go far too long without medical care, in hopes that the
issue will clear up on its own) and who would kill them at the end of their
egg-laying years (sometimes in excruciating ways; see McWilliams, 2015).4
So, liberating them looks something like creating a miniature sanctuary,
where they can live out their lives in comfort and safety. And if you do
that, you’ll find yourself left with quite a few eggs, as chickens don’t seem
to care about them after laying them. (Indeed, people who have backyard
chickens tend to be quick to collect the eggs, as chickens otherwise break
them and scratch them into the dust.) Since, you don’t violate someone’s
rights by collecting and consuming what she abandons, it seems permis-
sible to collect and consume those eggs.
There are various objections that someone might raise to this argument.
The first, and perhaps most obvious, is that you shouldn’t buy chickens
because, in so doing, you support the sale of chickens. But I doubt that
your purchase makes a difference, so I can’t see that as a good reason not
to do what almost certainly helps a number of individual hens, namely,
buying and then giving them the best life available. Of course, if someone
insists that individual purchases do make a difference, then there are still
moves to make. On the one hand, you could say that this just shows that
if chickens are going to be kept, they should be stolen rather than paid
for. Purchasing may support the sale of chickens, but it seems doubtful
that theft does. On the other hand, and perhaps more plausibly, you
could stress that if chickens have a right to be spared from various harms,
including an abbreviated life, then we have a case where duties to one
animal conflict with your duties to others. It seems quite plain, though,
that whatever your contribution to the sale of more chickens, it’s trivial
compared to the difference that you make in the life of the one chicken
you purchase. And when weighing duties, these differences in the conse-
quences are important.5
A second objection is that this isn’t really “liberation.” Instead, you’re
simply taking advantage of the wrongdoing of others (namely, breed-
ing chickens for this purpose) for the sake of having a product that you
enjoy. There are two things to say here. One of them is that intentions
do matter. If you’re happy that chickens are for sale, or if you aren’t
prepared to promote their longevity (e.g., by paying for visits to the vet),
then there is a good chance that you are merely using them, and from the
perspective of the rights view, that’s wrong. But people need to have that
mindset, and many don’t. In my hippie town, for instance, there are a
number of people who describe themselves as “veggans”—vegans apart
from the eggs that they receive from their own chickens. As far as I can
see, they care about their chickens’ well-being as much they care about
the well-being of any companion animal (which is very much indeed).
The other thing to say is that if the rights view were to require total
purity regarding intentions, then that would be a mark against the rights
view, rather than evidence against the permissibility of having backyard
chickens. Any person who is deciding whether to take in an animal will
think about the various costs and benefits. If a woman feels safer going
for her evening runs with a dog by her side and adopts a dog from a
shelter partly for that reason, her reasons are not entirely pure: she isn’t
solely concerned with the well-being of the animal; her own interests
are in view. But surely that’s fine. The constraint is just that her interests
can’t be the only things she thinks about, nor always the winners when
her interests and the dog’s conflict—there has to be a balance. So, for the
person who genuinely objects to the way that people treat chickens and
wants to provide a better life for them, no moral problem is created by
the fact that she regards having fresh eggs as a pleasant side effect.
Additionally, it’s worth noting that we could run a now-familiar argu-
ment for consuming these eggs: to minimize our support for plant agricul-
ture, we should eat foods that would otherwise go to waste, including the
eggs of the chickens we liberate. Someone with this motivation wouldn’t
have herself in view when using eggs from backyard chickens, but rather
the animals harmed in plant agriculture. And surely that’s a permissible
motivation.
Finally, someone might object that if this argument works, it sanctions
too much: in addition to their eggs, you’ll be entitled to their bodies.
What’s more, it will be permissible to extend the argument to other spe-
cies, saying that we can “liberate” a cow for her milk and meat.
To be clear, nothing in the argument above is based on the thought
that we are entitled—in a sense of having some claim to—the bodies or
byproducts of animals. So that part of the objection is simply mistaken.
But there is a sense in which the argument can be used to defend very
limited dairy production and consumption—as Milburn (forthcoming)
argues—and perhaps Cochrane’s (2012) thought that it could be per-
missible to farm animals for their corpses (i.e., give animals full lives,
only eating their bodies after their natural deaths). Granted, the kind of
dairy products that Milburn describes, and the meat that Cochrane has in
mind, aren’t widely available. But it’s the rare person who’s in a position
to keep a small herd of bovines in the backyard, and so these in princi-
ple permissions aren’t terribly interesting. By contrast, there are plenty
of people who could have backyard chickens if they so desired, making
them much more interesting as an exception to the alleged restrictions
imposed by the rights view. Proponents of the rights view have simply
overstated what follows from it; it doesn’t seem to underwrite an obliga-
tion to be vegan.
7.3 Roadkill and Freeganism Revisited

Recall the basic argument for eating roadkill, leftovers, and food that’s
been thrown away: thereby, we can minimize our support for practices
that cause extensive, unnecessary harm to animals. Cheryl Abbate (forth-
coming), however, argues that this is really a utilitarian line of reasoning
that doesn’t follow on the rights view:
Viewing animals (humans and nonhumans) with respect requires that

we view them as things that are not to be eaten. This explains our
strong intuition that there is something morally problematic about
eating human corpses. Eating human corpses is wrong precisely
because such an act indicates a failure to recognize that humans,
as beings with inherent value, are not things to be eaten. To con-
sume human corpses is to express that humans are consumables or
resources, and this is a failure to view humans with the respect they
are due, even if this act of consumption does not cause experiential
harm to a particular human.
. . .
It’s likely that Bruckner didn’t argue that those who promote
human rights and welfare ought to consume human corpses because
he acknowledges that this somehow is disrespectful to human beings.
Yet to eat roadkill, while refusing to eat the corpses of humans, is
to express the sentiment that because humans have, while animals
lack, serious moral worth, human bodies should be venerated, while
animal bodies can be consumed. And it is precisely because we are
willingly to eat animal corpses, but at the same time, we refuse to eat
human corpses, that makes eating roadkill wrong. Indeed, it would
be terrible to suggest that humans should eat the corpses of black
humans, while remaining silent about the ethics of consuming the
corpses of white humans. It is terrible because this would express
the disrespectful view that black people, but not white people, are
mere consumables, and thus that black people have less moral worth
than white people. Likewise, to suggest that humans should eat the
corpses of animals, while remaining silent about the ethics of con-
suming human corpses, is problematic because it expresses the dis-
respectful view that animals, but not humans, are things to be eaten.
There are two things to say about this argument. First, for reasons that
I’ve already discussed in the previous chapter, I think that this argument
turns on a mistaken assumption about the implications of the rights view.
It isn’t part of the rights view that we’re supposed to view (all) animals as
things that aren’t to be eaten. Instead, it requires us to view (all) animals
as beings with inherent value. It’s then an open question: how should
you treat beings with inherent value? I’m assuming that you can’t answer
that question apart from considerations about their interests, as I find it
implausible that it’s simply a brute fact that we shouldn’t eat beings with
inherent value. (The brute fact view leaves unexplained exactly what the
rights view is trying to explain, namely, why we should think that par-
ticular kinds of utility-maximizing actions are morally impermissible.)
But many nonhuman animals have no interest in not having their dead
bodies consumed, which means that you don’t fail to respect any of their
interests when you consume them without killing them. By contrast,
many human beings don’t want people consuming their dead bodies, and
out of concern for their wishes, we accept a cultural norm according to
which consuming them would be disrespectful. However, I see no basis
for the norm other than these very common desires, and so where they’re
absent—as in the case of nonhuman animals—they aren’t binding.
As discussed earlier, the immediate objection will be that it must then be
fine to eat the bodies of dead human beings who don’t have the relevant
desires. But this doesn’t follow. Our norms about respectful treatment of
dead bodies are tied to membership in a culture. Nearly all human beings
are in that culture and nearly all nonhuman animals aren’t. (There may well
be distant people groups with different cultural norms, and different treat-
ment would be appropriate as a result; likewise, we may count companion
animals as members of our communities, in which case eating a dead dog
is different from eating a dead deer.) This isn’t speciesism. Instead, it’s sim-
ply the upshot of recognizing that while the duty to show respect may be
universal, the manifestation of respect depends on the details of a culture.
Second, let’s suppose that I’m mistaken about the above, and that
it’s inherently disrespectful to consume the dead bodies of human and
nonhuman animals. Even if that were true, it wouldn’t follow that we
shouldn’t eat roadkill. After all, by eating roadkill, we’re trying to respect
the interests of living animals, and it would take further argument to
show that we’re in a situation where the moral imperative to respect the
dead outweighs the moral imperative to respect the living. What’s more,
it seems implausible that were going to be able to do this: while we might
be able to give reasons to respect the rights of dead human beings even
when they run counter to the rights of living human beings, that task is
much harder when it comes to nonhuman animals. In the case of human
beings, we can say something like this: the living want their own wishes
to be respected after their deaths, and so they have an interest in main-
taining a practice of respecting the interests of the deceased. But in the
case of nonhuman animals, they have neither such wishes nor practices.
As a result, even if consuming their dead bodies is disrespectful, it may
be the lesser of two evils, at least if the alternative is showing insufficient
respect for the interests of living animals—a move we can understand
as a friendly amendment to Bruckner’s argument for consuming road-
kill. It seems to me, therefore, that the rights view doesn’t condemn the
consumption of roadkill and, for the same reasons, doesn’t condemn the
consumption of leftovers and food that’s been thrown out.
7.4 Insects Revisited

Like utilitarianism, the rights view sanctions the consumption of insects.
Someone might think that this is obvious, since on Regan’s view, you
aren’t a rights holder unless you’re a subject of a life. That standard is
much more demanding than mere sentience:
[Individuals] are subjects-of-a-life if they have beliefs and desires;

perception, memory, and a sense of the future, including their own
future; an emotional life together with feelings of pleasure and pain;
preference- and welfare-interests; the ability to initiate action in pur-
suit of their desires and goals; a psychophysical identity over time;
and an individual welfare in the sense that their experiential life fares
well or ill for them.
The evidence for insects’ being phenomenally conscious is sketchy
enough; it’s much more complicated to establish that they have a sense of
their own future, an emotional life (with the exception of that interesting
study on bees), or a psychophysical identity over time. Still, proponents
of the rights view have their own ways of thinking about moral risk-
taking.6 The argument might go like this:
Admittedly, we don’t have clear evidence for insect sentience. How-

ever, there is suggestive evidence concerning some species, such as the
honey bee, and there are a few factors that make it unlikely that we’d
detect consciousness in insects even if it’s there. First, we can’t meas-
ure consciousness directly. We can only look for whatever gives rise
to it, which may be different in very different forms of life. Second,
we should recognize that a “favor the simpler hypothesis” policy
amounts to a bias in favor of behaviorist explanations, which is to
say that the deck is stacked against attributing mentality to insects.
(For more on these two points, see Bradshaw, 1998.) Third, research
on attributing mentality to farm animals suggests that those attribu-
tions are influenced by an impulse to reduce cognitive dissonance
related to how we treat those beings (Loughnan, Haslam, and Bas-
tian, 2010). And as Hal Herzog (2010) notes, our sympathies can
be influenced by factors that are irrelevant to whether a being is
sentient, such as whether it has four or six legs. So, it wouldn’t be
surprising if our judgments of insect sentience are also affected by
such factors. Most importantly, though, it would be a tragedy if we
were wrong about insect sentience. If they are conscious, then to kill
them for food would be to cause significant unnecessary harm. Given
these considerations, we shouldn’t eat insects: the risk of significant
unnecessary harm is too great relative to the good that might come
from entomophagy.
Let’s call this the precautionary argument. Obviously enough, it doesn’t

merely apply to insects. You could make similar points about other
simple invertebrates: some mollusks (such as oysters), jellyfish, various
crustaceans (shrimp, crabs, lobsters), and so forth. So if successful, the
precautionary argument will deliver many of the results that proponents
of the rights view want. Are there any problems with it? I think so. Again,
no one contests that some animals are harmed in plant production. And
this point is enough to throw a wrench in the precautionary argument.
As vegans have long appreciated, our choice is not between a host of
diets that are complicit in harm and an alternative—veganism—that’s
harm free. Vegans hope to minimize the harm associated with their food
choices, but sensible ones are under no illusion that they’ve eliminated it
entirely. However, the precautionary argument seems to assume that there
is a harm-free alternative. So, either vegans need to run a precautionary
argument that doesn’t make that assumption, or they need to stop relying
on this line of reasoning.
To see the problem, consider two diets: the strict vegan diet and the
“plants and bugs” diet—a diet that’s mostly plant-based but includes
some insects. The proponent of the precautionary argument says that we
should prefer the strict vegan diet because insects might be sentient, and
we shouldn’t risk harming beings that might be sentient. So far, so good.
But in light of what we know about the animals harmed in plant agri-
culture, we can point out that the strict vegan diet harms beings that we
know to be sentient—namely, those critters that get crushed by combines
(among other fates). Granted, if we were to offset some of our plant con-
sumption with insects, the insects killed would vastly outnumber the field
animals saved. However, if our choice is between, on the one hand, harm-
ing beings that we know to be sentient and, on the other, harming beings
that we don’t know to be sentient, we should go with the latter option.7
We can represent the choice as a trolley problem. On the main track,
there is a person who will be killed by an out-of-control trolley. You
can divert the trolley—which appears to be empty—onto a side track.
Unfortunately, the side track leads to a bridge that was never finished,
and if the trolley goes that way, it will hurtle into a canyon. What should
you do? Well, you could be wrong about whether the trolley is empty.
But since it seems to be, you ought to divert it onto the side track. Again,
when the choice is between saving something clearly morally valuable
(the one person) and saving something that might be morally valuable
(the trolley—which is serving as a proxy for the possible passengers), the
scales tip in favor of saving the one we know to matter.
The proponent of the precautionary argument might balk at this. Per-
haps he’s not sure that the apparently empty trolley is a good analogue
for insects, since he’s not prepared to concede that insects don’t seem to
be sentient. To be clear, his worry is not that insects seem to be sentient.
If that were the worry, then we’d no longer be discussing a precaution-
ary argument—we’d be discussing a run-of-the-mill argument from the
moral value of conscious experiences. Instead, the claim is that insects
neither seem to be nor seem not to be sentient. It’s supposed to be an
open question.
If that’s his reaction, let’s tweak the scenario. Again, there is a person
on the main track who will be killed by an out-of-control trolley. You
can divert the trolley onto a side track. Unfortunately, the side track
leads to a bridge that was never finished, and if the trolley goes that way,
it will hurtle into a canyon. But this time, you just don’t know whether
anyone’s on the trolley; you can’t tell one way or another. (Suppose you
know that empty trolleys have been uncoupled accidentally in the past,
careening wildly down the tracks as a result. Hence, it really is an open
question to you whether an out-of-control trolley has passengers.) What
should you do?
Well, there could be someone on the trolley—or several people. How-
ever, there might not be anyone on board at all. Plainly, the consequences
of diverting the trolley could be massively worse than not diverting it,
and I suspect that intuitions will diverge as a result. And if intuitions
diverge, we probably don’t have a decisive argument.
However, we should revisit the worry that led us here. The proponent
of the precautionary argument wasn’t willing to concede that insects
don’t seem to be sentient, and we should question the epistemic stand-
ards that led him to that position. If there is a harm-free diet available
to you, it might be fair to set a high bar for evidence concerning insect
sentience, taking it as an open question absent decisive considerations.
But once you know that a strict vegan diet is complicit in harming some
animals that are clearly conscious, you shouldn’t set the bar so high,
since doing so may lead you to discount harms to clearly conscious
beings.
Let’s alter the scenario one more time. The change is that the out-of-
control trolley is now without a roof. You can see that the seats aren’t
occupied. In such circumstances, it would be wrong to hold out and say
that, since there still could be people hiding under the seats, you should
let the trolley kill the one. What you know about the person on the track
affects the epistemic standards that you should employ. If there’s a harm-
free option, then you don’t need to trust your eyes. But if there isn’t, you
should.
At this juncture, there are two moves that the strict vegan might make.
The first is to employ a stronger precautionary principle. The one to
which we’ve been appealing seems to be something like this:
Weak Precautionary Principle. In cases where we’re uncertain whether

a particular individual is sentient, and treating that individual as
sentient wouldn’t prevent us from fulfilling our obligations to any
being that clearly is sentient, we ought to treat that individual as
though it’s sentient.
Perhaps the proponent of the precautionary argument will want to drop

the second clause:
Strong Precautionary Principle. In cases where we’re uncertain whether

a particular individual is sentient, we ought to treat that individual
as though it’s sentient.
However, the Strong Precautionary Principle is implausible. It entails that,

unless you’re absolutely certain that panpsychism is false—according
to which everything is conscious, including tables and chairs—you ought
to treat everything as though it’s conscious. Surely we don’t have such
obligations.
The strict vegan might try to find a middling principle, such as the
following:
Middling Precautionary Principle. In cases where the probability that

a particular individual is sentient is over x, we ought to treat that
individual as though it’s sentient.
Sebo (2018) offers an interesting objection to the Middling Precautionary

Principle. Suppose, for example, that you set the relevant probability at
.5. Then, you don’t need to distinguish, morally, between beings at .49
and .01. But that seems wrong: given much greater odds of conscious-
ness, you deserve more moral consideration.8
The defender of the precautionary argument might be able to respond
to this objection by denying that we can accurately quantify the odds
of sentience. Instead, he could insist, we’re going to be stuck using very
rough estimates—say, definitely conscious, probably conscious, perhaps
conscious, and definitely not conscious. He might maintain that we’re in
the first category; minnows are in the second; insects in the third; tables,
fourth. And once our choice is between these rough categories, it seems
more plausible that we may apply different moral standards to beings
that fall into different ones.
That said, there’s no hope for a principle according to which we ought
to treat perhaps conscious beings as we ought to treat definitely conscious
ones. Such a principle would imply that in the choice between an infant
and a bee, it’s a moral toss-up. So we need to combine the probabilis-
tic dimension of the Middling Precautionary Principle, tempered by the
observation that we can’t accurately quantify the odds of sentience, with
the qualification that characterized the Weak Precautionary Principle:
The “Rough Estimates” Precautionary Principle. (1) In cases where a

particular individual is probably conscious, and treating that indi-
vidual as conscious wouldn’t prevent us from fulfilling our obliga-
tions to any being that is definitely conscious, we ought to treat that
individual as though it’s definitely conscious; (2) in cases where a
particular individual is perhaps conscious, and treating that indi-
vidual as conscious wouldn’t prevent us from fulfilling our obliga-
tions to any being that is definitely or probably conscious, we ought
to treat it as definitely conscious.
This principle avoids the problem that sunk that Middling Precaution-
ary Principle. By rejecting fine-grained probabilities of consciousness, it
prevents refinements of the .49 versus .01 counterexample. And by add-
ing the clause about needing to first fulfill our obligations to beings that
are more likely to be conscious, it blocks any implication that babies and
bees are on a moral par.
The “Rough Estimates” Precautionary Principle seems fairly plausi-
ble. Unfortunately, it’s of no use to the defender of the precautionary
argument, since it implies that our obligations to definitely and probably
conscious beings trump our obligations to perhaps conscious beings. The
animals harmed by plant agriculture are in one of the former two catego-
ries; insects are, at best, in the latter. Again, it looks like the “plants and
bugs” diet is the morally preferable one.
The upshot is this. If we harm animals in plant production, then it
becomes much harder to run precautionary arguments against eating
insects. We aren’t in the position of comparing harmful and harm-free
diets, but of comparing diets that harm different beings, some of which are
clearly sentient, but others about which there are varying levels of uncer-
tainty. It seems to me that this fact tells in favor of eating insects. Does it
also tell in favor of eating other invertebrates, such as shrimp? Yes, at least
if we accept the “Rough Estimates” Precautionary Principle, and if we
put shrimp in the perhaps sentient category. After all, given the “Rough
Estimates” Precautionary Principle, our obligations to probably sentient
beings—such as rabbits and field mice—trump our obligations to perhaps
sentient beings. But if we place rabbits and shrimp in the same category,
then the conclusion won’t follow. So everything will turn on those assess-
ments. However, it seems eminently likely that bivalves will end up in that
lower group, and so in addition to eating insects, even proponents of the
rights view will have good reasons to eat oysters, mussels, and scallops.
7.5 Hunting
The argument for the permissibility of eating insects and bivalves depends
on there being harms to animals in plant agriculture. As it turns out, the
rights view requires some hunting for the same reason, as Dan Demetriou
and Fischer (2018) have argued—though I’ll present the argument quite
differently here.
Let’s begin with the basic structure of the argument. Any version of
the rights view is going to have to take a stand on what you ought to
do when the claims of rights holders conflict. One of the principles that
Regan offers is the “worse-off principle”:
[W]hen we must decide to override the rights of the many or the

rights of the few who are innocent, and when the harm faced by the
few would make them worse-off than any of the many would be if
any other option were chosen, then we ought to override the rights
of the many.
(2004, 308)
Now, suppose that we’re going to be complicit in some rights violations:

either the violation of the rights of hunted animals or the violation of the
rights of the animals killed in plant agriculture. If it turns out that the
animals killed in plant agriculture are made worse off than are hunted
animals (a claim for which I’ve not yet argued, but will shortly), then
even if the “harm footprint” of hunting is larger than the harm footprint
of plant agriculture—that is, even if we are complicit in a greater number
of deaths by hunting than we would be by eating plants—we should vio-
late the rights of hunted animals. In other words, we should hunt.
To be clear, if this argument works, it’s because the proponent of the
rights view has a solution to the sorts of problems that we explored in
the previous chapter. If you aren’t responsible for the various intended
and unintended effects of animal agriculture, then surely you aren’t
responsible for the various intended and unintended effects of plant
agriculture. But if you are responsible for the intended and unintended
effects in the former case, then you are responsible for them in the lat-
ter. Moreover, note that it makes perfect sense for a proponent of the
rights view to consider whether she ought to grow her own food on the
basis of worries about conventional plant agriculture. That is, it makes
perfect sense for her to consider whether she ought to try to reduce her
harm footprint by, say, having a veganic vegetable garden, versus pur-
chasing vegetables at the grocery store. But given that she’s a proponent
of the rights view, she should recognize that her harm footprint needs to
be assessed not simply in terms of impact on welfare, but also in terms
of rights violations. If it turns out that hunting actually involves fewer
rights violations, then that’s a mark in favor of hunting. Of course,
the proponent of the rights view will strongly doubt that hunting will
involve fewer rights violations. That’s fair enough; so far, I haven’t said
anything to dissuade her of that view. All I’m after at this stage is the
in-principle observation: given the supposition that animals killed in
plant agriculture are made worse off than are hunted animals, limited
hunting is permissible.
Why might anyone think that animals in plant agriculture are made
worse off? There are two reasons. The first and more mundane one is
related to the differences in the way that the relevant animals die. As
far as we can tell, most of the animals killed in plant agriculture die as
a result of being exposed to predators when crops are harvested (and
so their shelter is eliminated) and through trapping and poisoning (Fis-
cher and Lamey, 2018). (Farmers intend all these deaths: exposing crop-
consuming animals to predators is a reliable way of reducing crop loss,
and farmers welcome predators for that reason.) As Brian Tomasik (2015)
argues, deaths due to predation are likely to be quite awful, and there is
good reason to think that death via low poison dosages is prolonged
and painful. By contrast, hunters can kill their quarry relatively quickly,
and arguably in less painful ways. Obviously, not all hunters. Some take
shots they shouldn’t; others fail to track and finish off injured animals.
So I’m not defending all hunting. I’m only defending those who walk the
line—skillfully—between (1) minimizing suffering and (2) giving animals
a genuine opportunity to evade being killed.
The latter point brings us to the second reason why we might think
that the animals in plant agriculture are made worse off, namely, unlike
hunted animals, those animals are killed as pests. They are not treated
as beings worthy of any respect, but rather as problems that need to be
managed. This is an affront to their dignity. By contrast, some hunting
does treat animals as worthy of respect. Consider, for instance, the per-
spective of Jim Posewitz, a hunter and prominent advocate for the “fair
chase” ethic:
Fundamental to ethical hunting is the idea of fair chase. This con-

cept addresses the balance between the hunter and the hunted. It is
a balance that allows hunters to occasionally succeed while animals
generally avoid being taken.
(1995, 57)
This perspective has been adopted by the most prominent US hunting

organization, the Boone and Crockett Club, and it’s taught as central to
the practice:
[S]hooting a wild turkey perched on a limb of a tree is not illegal, but

to those who consider the art of decoying and calling a wary bird
into range to be turkey hunting, taking such a shot would be neither
fair chase nor turkey hunting. There are no laws against shooting a
game animal that has become partially domesticated or habituated to
humans. Fair chase, however, would not take advantage of animals
whose natural instincts have been compromised in this way. . . . A
sporting approach recognizes the advantage of human capabilities,
including technologies, and a desire to constrain ourselves. More
often than not, this means prey will avoid the hunter. Knowing what
improper advantage means comes from experience, but if there is
any doubt, the advantage should go to the animal. That is fair chase.
(2016, 4)
So it’s built into this idea that hunting should be costly for hunters; it
should be rare that a hunt is successful. Moreover, animals should be
able to take advantage of their species-specific excellences to escape.
Dignity-respecting hunting involves recognizing animals as beings who
deserve the chance to evade harm and so rejecting the hunter’s view of
herself as at liberty to treat the relevant animals as mere resources, avail-
able on demand and at her whim. Granted, this remains an instance of
respect in imperfect circumstances: no one is claiming that the situation
is anything other than nonideal. But if animals are going to be harmed
either way, and the conditions specified in the worse-off principle obtain,
then we ought to take into account animals’ natures, and our own self-
understanding, when deciding how best to relate to them. All the relevant
animals are, of course, predator-evading creatures. Moreover, it seems
valuable to have hunters view animals as beings for whom it’s worth tak-
ing on significant costs.
Of course, the fair chase ethic is not equivalent to respecting animal
dignity. Crucially, you are only engaging in rights-respecting hunting if
it’s motivated in the right way. Respecting animal dignity isn’t simply
about giving animals a sporting chance, but also about trying to respect
animals as beings with rights, killed directly and intentionally only
because of the problems in the food system. This kind of hunting would
be readily abandoned if indoor, harm-free agriculture were suddenly to
take off, or if there were good reasons to think that farmers had come
to see field animals as having inherent value and were doing their best
to minimize injury to them.
Tovar Cerulli, a vegetarian-turned-hunter, comes close to the kind of
motivation that I have in mind:
In hunting, I was aiming to be mindful about the outer conse-

quences of my diet. I was aiming, someday, to confront one of those
consequences—the death of animals—with my eyes open. Killing an
animal myself—mindfully, swiftly—seemed to be the most conscien-
tious path, perhaps even the path of least harm. And I was aiming to
better understand the land and my non-human neighbors. I was on
the same quest that led me to vegetarianism, the quest for a respect-
ful, holistic way of eating and living in relationship with the world,
for right dietary citizenship.
My inner aim remained the same, too: integrity. Having concluded
that I needed some animal protein in my diet and that some harm
to animals was inevitable in even the gentlest forms of local agricul-
ture, moral wholeness and alignment could only come from taking
responsibility for at least a portion of the killing.
Earlier, I suggested that this dual attention—to both inner and
outer consequences—is common among vegetarians. I won’t make
a parallel claim on behalf of most hunters. I don’t know how many
hunters share these sensibilities. Some clearly do not. When I over-
hear a hunter say that he goes to the woods to shoot rabbits for the
sheer challenge of it, with no intention of eating them, or when I’m
told about a moose shot in deer season and left to rot just because
some guy wanted to kill something, I experience the same old revul-
sion. I still wonder what kind of sickness possesses such people.
(2010, 52)
This passage doesn’t provide a perfect model of the dignity-respecting

hunter. For instance, there isn’t anything here about fair chase (though
it’s clear from what Cerulli says elsewhere that he regards fair chase as
important), and his hunting is motivated partially by nutritional con-
cerns. But it certainly contains the thought that animals deserve respect,
and the thought that the legitimacy of harm stems from its unavoidabil-
ity, rather than from some claim on the bodies of animals. Finally, it’s
notable that Cerulli has a conception of integrity that involves taking
responsibility for harm. I read this as a commitment to animal dignity:
animals are the kinds of beings such that, if we’re going to harm them, we
should look at them when we do it. They deserve our attention. We can
aspire not to harm them in the first place, of course, but it isn’t clear how
buying vegetables at the grocery store alone amounts to such an aspira-
tion. If veganism is coupled with a certain kind of agricultural activism,
then so be it. But now we aren’t simply talking about an obligation to be
vegan, but (yet again) an obligation to do much more.
In any case, perhaps the best objection to this view is that even if farm-
ers are intentionally killing field animals in plant agriculture (whether
directly or indirectly), it remains the case that you, as the consumer, don’t
intend those deaths. By contrast, you do have to intend the death of
the animal you hunt. And that difference in intention makes a moral
difference.
Note, however, that if you buy plant-based products at the grocery
store, you do intend to support an industry that you know to cause vari-
ous harms to animals and that has no plans to stop causing those harms
for the foreseeable future. Moreover, it isn’t obvious that purchasing
plant-based products at the grocery store results in a lower harm foot-
print on a per-calorie or per-gram-of-protein basis. Let’s suppose that
you eat a vegan diet with the exception of the animal products that you
secure by hunting large animals, such as deer or feral hogs. Mature males
of both species can weigh 300 pounds, about 40% of which is meat. If
you use this meat to get the USDA’s recommended 5.5 ounces of “protein
foods,” then you can meet nearly a year of your dietary requirements
from a single animal. Even if we adopt a fairly modest estimate of the
number of animals killed in plant agriculture, we would get an annual
harm footprint that’s at least that high.9
Someone might dispute the claim that animals suffer less when hunted
than they do when killed by predators, poisoned by farmers, or mauled
by machinery. And then they might say that the kind of dignity on which
the argument trades isn’t worth wanting. The value of dignity is deter-
mined by its contribution to welfare; it isn’t an independent good. And if
all that’s right, then the argument does indeed fall apart.
However, the move wouldn’t be without cost, as it would create an
uncomfortable dilemma for the proponent of the rights view. If she staves
off the pro-hunting argument by insisting that animals don’t have an
interest in dignity, then she can say that they don’t have a right to be
respected in that regard. But that will be hard to motivate without going
in for an interest-based account of rights generally, and taking on that
commitment will leave her with the task of fending off Cuneo’s challenge.
Recall: that argument depends on the interest-based approach to rights,
as you need it to explain why it’s permissible to shorten the lives of ani-
mals. However, if the proponent of animal rights gives up the interest-
based approach, then she has a ready objection to Cuneo on small-scale,
family-run animal agriculture. She can simply maintain that animals have
a right to continued existence even if they lack the psychological capaci-
ties that partially ground the human interest in continued existence. But
then she’s going to be stuck with the pro-hunting argument, according
to which animals have dignity rights that are better respected by hunting
them than by killing them as pests.
A different sort of objection is that I’m downplaying the costs of hunt-
ing. For instance, the offspring of hunted animals may be more vulnerable
to predators or starvation after their parents are killed, and having fire-
arms in human hands may result in a greater number of human deaths,
as accidents (and non-accidents) predictably occur. I think there are two
things to say in about this. The first is that harm footprint considera-
tions aren’t actually decisive here. Suppose it turns out that, if we hunt,
a greater number of beings have their right to life violated. According
to the worse-off principle, that’s an acceptable tradeoff as long as the
alternative involves violating that right and some other for a different set
of animals: in this case, their right to life and their right to be treated as
having dignity. And as long as being treated as a pest isn’t being treated
with dignity, then that’s the situation in which we find ourselves.
Second, suppose that we haven’t yet assessed the full costs of hunting,
either in terms of body count or rights violations. (With respect to the
latter, for instance, we might wonder whether the offspring of hunted
animals have a right not to have their parents killed.) Still, it doesn’t
follow that hunting is impermissible and veganism is required. Instead,
what follows is that it’s an open question how best to minimize your
harm (or rights violation) footprint. Maybe the argument for hunting
will only work for solitary animals, ones who don’t live in the kinds of
family groups that plausibly ground certain noninterference rights. Or
maybe it only works for much larger animals, so that white-tailed deer
shouldn’t be hunted, but moose may be (which can weigh four or five
times as much, as so will replace far more plant-based food). Or maybe
the argument for hunting works only because when you add guns to a
community, they save more human lives than they take. I’m not defend-
ing any of these claims, nor do I find them all plausible. The point is just
that once we delve into the complex task of sorting out these footprint
questions, there’s no particular reason to think that the calculations will
favor veganism. And given that, it isn’t clear that veganism is obligatory
in the meantime.
7.6 The Doctrine of Double Effect
All these cases—backyard chickens, insects, hunting—involve either sup-
porting or taking an active role in intentional harm to animals. (This
might not be obvious in the case of backyard chickens, but we shouldn’t
simply focus on the chickens you raise; additionally, there are the male
chicks that are killed because no one wants roosters, as well as the lefto-
ver birds who aren’t sold as layers, and so are sold to be killed for food.)
And we might think that the rights view is part of a tradition that has a
ready reply to these arguments: its proponents can appeal to the doctrine
of double effect (DDE). Very roughly, the classic version of the DDE
says that, in pursuit of some sufficiently weighty good, you may permit a
foreseen but unintended bad outcome to occur, as long as there’s no other
way of securing the relevant good. The most famous problem for this ver-
sion is that it seems to make it too easy to avoid responsibility for awful
outcomes. Consider a standard case in the trolley problem literature:
Large Man: You’re standing on a bridge over a train track. A large

man is sitting on the edge of the bridge, enjoying the view. He
doesn’t notice that five people are tied to the track, and that a runa-
way train will kill them if it isn’t stopped. He’s big enough that if
you were to push him off the bridge, his body would stop the train
and the five would be saved. But the large man would certainly die.
Many people tend to think that it would be wrong to push the large man
to save the five. On the face of it, the DDE explains why: the death of the
large man isn’t merely foreseen, but intended. You’re using him to save
the five, and that’s wrong. However, you could easily say something like
this: “Look, I don’t intend to kill the large man; I only intend to stop the
train. If, by some miracle, he were to survive being pushed in front of
the train, then I’d be thrilled. So I merely foresee his death, and the DDE
excuses me.”
Cases like this inspire revised versions of the DDE. Here, for instance,
is Warren Quinn’s (1989) revised DDE (DDE-R):
DDE-R: In cases in which harm must come to some in order to achieve

a good (and is the least costly of possible harms necessary), the
agent foresees the harm, and all other things are equal, a stronger
case is needed to justify harmful direct agency than to justify equally
harmful indirect agency.
According to Quinn, this principle
distinguishes between agency in which harm comes to some vic-

tims, at least in part, from the agent’s deliberately involving them in
something in order to further his purpose precisely by way of their
being so involved (agency in which they figure as intentional objects),
and harmful agency in which either nothing is in that way intended for
the victims or what is so intended does not contribute to their harm.
(1989, 343)
In the case involving the large man, there’s no way to redescribe your
intentions so that they don’t involve using the large man; he is clearly
required for your plan, even if his death isn’t. And the idea is that the
justification for such use needs to be stronger than the justification for
a plan that doesn’t involve him, even if (for some complicated reason) it
were to bring him harm.
How much stronger, exactly? Quinn doesn’t say. However, we might
think that it’s a virtue of the account that it’s possible to justify direct
harmful agency. Suppose that a thousand people were tied to the track,
and only pushing the large man to his death could save them. Then, per-
haps direct harmful agency would be justified. But not for five.
As Andy Lamey (2019) point outs, though, the DDE-R needs to be
supplemented before it might apply to the kinds of cases I’ve discussed.
This is because it’s motivated by broadly Kantian considerations that
seems to exclude animals. Here’s Quinn (1989, 350) again:
The doctrine reflects a Kantian ideal of human community and inter-

action. Each person is to be treated, so far as possible, as existing
only for purposes he can share. This ideal is given one natural expres-
sion in the language of rights. People have a strong prima facie right
not to be sacrificed in strategic roles over which they have no say.
This amounts to a tacit restriction on the application of the principle:

you’re only protected by it if you’re an agent, since only agents are beings
to whom we owe it to act in ways that they can recognize as justified.
Since most animals aren’t agents in this sense, they aren’t covered.
Lamey grants that this is a fine motivation for the DDE-R, but argues
that we can still introduce a supplementary principle, which we might
call “DDE-R-Animal.” This principle is worded in exactly the same way,
but there’s a different rationale for adopting it. Instead of being moti-
vated by Kantian considerations, it’s motivated by the Capacity Principle:
The Capacity Principle: The presumption against harmful direct agency

derives from the fact that such agency can infringe an independent
right not to be made to serve others’ purposes at the expense of the
capacity to form purposes of one’s own.
The idea is straightforward. Suppose, with Lamey, that animals have a

right not to be made to serve others’ purposes at the expense of their
lives—which, of course, they need to have the capacity to form purposes
of their own. If so, then you need a stronger argument for lethal “harm-
ful direct agency” toward animals—intentionally killing them to achieve
your ends, however good those ends may be—than you do for acting in
ways that unintentionally cause their deaths. The upshot is this. If you’re
thinking about directly killing some animals (insects, deer) to save some
others (field animals), it isn’t enough to say that the numbers break in
favor of direct killing. After all, it’s direct killing, and that requires a
stronger defense than, “It will cause a slightly lower total body count.”
And with DDE-R-Animal in hand, Lamey can make short work of some
standard anti-vegan arguments.
Consider, for instance, Davis’s (2003) argument, which we discussed
earlier. Davis compares the harm footprints of (1) beef from pasture-
raised cattle and (2) the plants required for a vegan diet. Based on his
(problematic) estimates of the harms associated growing plants (see
Lamey and Fischer, 2018), he concludes that beef from pasture-raised
cattle has a lower harm footprint. But with DDE-R-Animal available,
Lamey can claim that it isn’t enough to have the numbers come out just
in favor of beef. Raising cattle for food involves harmful direct agency,
whereas growing plants doesn’t.10 The justification for harmful direct
agency needs to be stronger than the one that Davis provides, and if we
use the large man case as a guide, the numbers would have to favor beef
by more than a factor of five.11
To be clear, Lamey never claims that DDE-R-Animal undermines the
kinds of arguments that I’ve made in this chapter. However, if double
effect reasoning does undermine them, it’s probably going to be based on
something like DDE-R-Animal. So, we should consider: does that princi-
ple create problems for the arguments I’ve made?
No. First, it makes no trouble for my argument about raising insects.
Even if we grant the Capacity Principle, we should note that it doesn’t
straightforwardly apply to insects, since we don’t know that they have
morally relevant purposes of their own. In other words, while there may
be an extra burden to justify harmful direct agency, that burden is miti-
gated by the evidence against insects being conscious.
Second, DDE-R-Animal makes no trouble for my defense of hunt-
ing. The crucial part of that argument is that hunted animals have fewer
rights violated than do animals killed as pests in plant agriculture. In par-
ticular, I’m claiming that the dignity rights of hunted animals are being
respected in a way that isn’t happening for animals killed as pests. Pre-
venting those additional rights violations provides the stronger case that
DDE-R-Animal requires.
Finally, and most importantly, DDE-R-Animal doesn’t imply anything
at all about what consumers should do. Suppose I grant DDE-R-Animal
and I agree that farmers are acting wrongly when they raise cattle on
pasture.12 It doesn’t follow that I’m acting wrongly when I buy meat
from cattle raised on pasture, at least if I’m motivated to do so because
I justifiably and correctly believe that this is the way to minimize my
harm footprint. Compare: I’m a military officer who knows that there
are two pilots I could send out on a bombing mission. One of them is
well-intentioned and highly competent, whereas the other has bad inten-
tions and is less competent. The well-intentioned and highly competent
pilot would never dream of intentionally bombing civilians, though as a
result of dropping his bombs so as to maximize the destruction of tar-
gets, it’s in fact the case that a greater number of civilians will die. The
less competent pilot with bad intentions is happy to bomb civilians, but
because his aim is poorer, fewer civilians die. At the very least, it seems
permissible for the officer to send out the less competent pilot with bad
intentions. I’m actually inclined to say that he ought to do it. But even
if that isn’t true, and it’s merely permissible, it would at least be the case
that it’s permissible to act, as a consumer, in a way that minimizes your
harm footprint, even if that involves relying on the harmful direct agency
of others. And given that the aim here is to show that there isn’t a gen-
eral obligation to be vegan, permission is enough. I conclude, then, that
appeals to double effect don’t save such a general obligation.
7.7 Conclusion
So it looks like the rights view doesn’t oblige much of anyone to be vegan,
based on the difficulty of forging a link between the ethics of production
and the ethics of consumption. And if I’m wrong about that, it still looks
like the rights view sanctions some small-scale, family-run animal agricul-
ture, enjoying eggs from backyard chickens, eating insects and bivalves,
and very limited hunting. This certainly isn’t the abstinence-only policy
that we would have expected. We might react by wondering, yet again,
whether we’ve gotten off on the wrong foot. To silence that thought, we’ll
take one more look around for arguments against eating animals.
Notes
1. See 351–353 for the application to lifeboat cases specifically.
2. The sense of “viability” matters here. If it’s economic viability, then some
small-scale operations might make the grade, though it’s very difficult to pull
off without welfare compromises. For details, see Norwood and Lusk 2011
and McWilliams 2015.
3. If this gap can be filled, then Cuneo might be wise to join forces with George.
If we aren’t violating animals’ rights, then we can combine George’s observa-
tions about, say, the nutritional needs of infants and children with the con-
text that Cuneo imagines. So, it might be permissible for infants and children
to consume nourishing and tasty animal products, at least if they’re sourced
in ways that sustain valued social practices, provide a viable alternative to
factory farming, and give animals excellent lives. And crucially, it would
be much harder to argue that either they or their parents are vulnerable to
the charge of selfishness, eliminating an important challenge to rights-based
defenses of eating animals.
4. It’s difficult to find empirical evidence that speaks to how common these
problems are. But speaking from my experience, I’ve never seen anyone take
a lame bird to a vet. Instead, I’ve only seen people (1) allow their chickens to
hobble around or (2) decide that they should kill them for their own sake.
5. Not incidentally, this is essentially the way to reply to a different objection,
which is that purchasing chickens treats them as property.
6. Gary Francione, for instance, is committed to giving insects the benefit of the
doubt. See, for instance, what he says here: www.abolitionistapproach.com/
sentience/.
7. Again, we’re assuming that insects are being given recycled feed. If that isn’t
true, then the harms of insect farming are dramatically greater, and plant-
based diets are preferable. But as I’ve said, many insect farmers do use recy-
cled feed (and clearly advertise this fact, so consumers can know).
8. Sebo uses different numbers, but this formulation makes the point clearer.
9. I’m making certain assumptions about how we ought to generate our esti-
mate of the relevant harm footprint. For all the choice points, see Fischer
and Lamey 2018. One of the crucial ones is that we bear some responsibility
for the field animals who are exposed to predators due to human activities.
This is, in part, because it’s a foreseen consequence of unnecessary activi-
ties (because we could produce/secure food in other ways), but also because
farmers want field animals to be killed by predators to reduce crop losses.
The predators are, in a sense, instruments in farmers’ plans. The other crucial
assumption is this: while we can’t assume uniformity across types of agricul-
ture, we’ve got enough evidence of harm to justify making some generaliza-
tions. Given these assumptions, it would be surprising, in my view, if it were
to turn out that plant agriculture has, in general, a lower harm footprint than
dignitarian hunting.
10. This isn’t quite right, as farmers kill lots of animals intentionally. But I’ll
ignore this complication here.
11. I don’t buy the move from “you shouldn’t directly kill one human to save
five” to “you shouldn’t directly kill one nonhuman animal to save five,” but
I have a hard time disentangling my skepticism about it from my general
doubts about the DDE. So, I’ll set this issue aside.
12. I don’t think we need DDE-R-Animal to respond to Davis: he’s wrong about
the numbers, and that’s enough. My position here implies, though, that if
Davis weren’t wrong about the numbers, it would be permissible to raise and
consume cattle on pasture.
8 Beyond Utilitarianism and the
Rights View
In this chapter, we consider a few non-utilitarian, non-rights-based ways

of defending veganism. Will other theoretical frameworks do better? Or
might we be best off dispensing with theoretical frameworks altogether,
making arguments by analogy instead?
I can’t survey every non-utilitarian and non-rights-based defense of
veganism here. I can, however, discuss some of the prominent options.
My goal, then, is not to be utterly exhaustive, but rather to illustrate
the difficulty of closing the gap between the ethics of production and
the ethics of consumption regardless of where we start, pointing out
some approach-specific problems as they’re relevant. Then, I’ll consider
an argument that isn’t already in the literature, but that I once thought
promising. Finally, I’ll take stock of the discussion so far.
8.1 Virtue Ethics

Let’s begin with virtue ethics, according to which we ought to act as
the virtuous person does, and where the virtuous person is the one who
manifests the appropriate virtue or virtues in the situations she faces.
Rosalind Hursthouse (2006) offers what’s probably the best-known
virtue-based argument for veganism, which begins by calling the reader’s
attention to the suffering involved in intensive animal agriculture.1 She
then writes this:
Can I, in all honesty, deny the ongoing existence of this suffering?

No, I can’t. I know perfectly well that although there have been some
improvements in the regulation of factory farming, what is going on
is still terrible. Can I think it is anything but callous to shrug this off
and say it doesn’t matter? No, I can’t. Can I deny that the practices
are cruel? No, I can’t. Then what am I doing being party to them? It
won’t do for me to say that I am not actually engaging in the cruelty
myself. There is a large gap between not being cruel and being truly
compassionate, and the virtue of compassion is what I am supposed
to be acquiring and exercising. I can no more think of myself as
Beyond Utilitarianism and the Rights View 129
compassionate while I am party to such cruelty than I could think of
myself as just if, scrupulously avoiding owning slaves, I still enjoyed
the fruits of slave labor. . . . The practices that bring cheap meat to
our tables are cruel, so we shouldn’t be party to them.
(2006, 141–143)
One worry, of course, is that the notion of “being party” to objectionable

practices is hard to spell out. Should it be given a causal interpretation?
Then either most consumers aren’t party to factory farming, or they are
party to too much, as complicity abounds. And if it isn’t given a causal
interpretation, then we face the same sorts of problems that made trouble
for the rights view: we might end up with a principle that successfully
links production and consumption, such as the benefiting account, but
only because it overgeneralizes; or, we might end up with some princi-
ple that involves appealing to intentions that may not be attributable to
many consumers, or what have you.2
However, I don’t think that this is the central problem. Rather, the
central problem is that it’s a mistake to ignore the larger context in which
people eat animals. As you become further and further removed from the
harm—not just in the sense that there are various intermediaries between
you and the harm, but in the sense that you have less and less control
over its occurrence—your reason for not contributing to that harm grows
weaker, which means that other reasons become relatively stronger. Con-
sider the following case:
(Mostly) Vegan Mom: Miranda is a vegan. Her 11-year-old son, who

she has raised vegan, goes to a public school that offers lunch at no
cost to the students. Her son’s friends convinced him to try cheese
pizza, which he very much enjoys. He asks for it all the time at
home, but Miranda generally doesn’t give in. However, after spe-
cial events—for instance, when his soccer team wins a game—she
orders cheese pizza for him, and because they like to eat together,
she has a slice.
Is Miranda failing to be virtuous? Well, it’s true that she isn’t doing eve-
rything she can to minimize her contribution to animal agriculture. How-
ever, she’s doing quite a lot (she’s vegan the rest of the time), and the
exception that she makes isn’t for some entirely trivial reason. She is eat-
ing an animal product as a gift to her kid. Admittedly, she isn’t modeling
the purest devotion to the “Don’t be a party to cruel practices” norm,
and in that sense, she isn’t optimizing for being the most virtuous con-
sumer. But the virtue theorist never thought that should be her aim: she
should be trying to become a virtuous person, which involves balancing
all the virtues in all the situations in which she finds herself. Unsurpris-
ingly, the virtues don’t always point in the same direction, and part of
130 Beyond Utilitarianism and the Rights View
being virtuous is knowing how to prioritize what matters in each context.
It would be surprising if “Don’t be party to the harms associated with
one particular industry” always won that contest.
But even if we concede that Miranda is virtuous, might we still accuse
her of failing to be compassionate? I’m not even convinced of that, as her
non-vegan behavior is itself motivated by compassion. Granted, it’s plau-
sible that being compassionate involves prioritizing: if the alternative is
not doing anything at all, then the compassionate thing might be to bring
some soup to a friend with the flu; however, if the alternative is attend-
ing to someone who just collapsed after complaining of chest pains, then
bringing soup to your friend isn’t compassionate at all. So, we might
think that Miranda ought to prioritize the plight of animals over connect-
ing with her child. However, the degree to which that’s plausible depends
on various factors, including whether her behavior makes a difference for
animals; what she does the rest of the time; whether she takes the time to
talk with her son about animals, so that he can appreciate the relational
significance of what she’s doing; and so on. The upshot: we shouldn’t be
myopic in our assessment of actions from a virtue ethical perspective.
And when we take a wider view of the reasons for which people act, and
the contexts in which those actions occur, there are bound to be cases
where eating animals is compatible with virtue.
Of course, there are plenty of consumers who simply aren’t troubled
by the plight of animals, and therefore eat animal products without a
second thought. Does Hursthouse’s argument condemn them? A great
deal turns on how we identify the virtues. One option is to say, with
virtue consequentialists, that the relevant character traits count as virtues
because of the aggregate effects that people would have if they were to
manifest them. So if it’s the case that, collectively, people could have a
significant positive impact if they were to be more careful consumers,
then assiduousness with respect to food choices might be a virtue. How-
ever, as discussed earlier with respect to rule utilitarianism, that sort of
approach isn’t going to make a virtue of veganism. Instead, it makes a
virtue of something like “humane eating,” where you should only eat
animals from high welfare farms.
Of course, I’m assuming that the virtue consequentialist is really a vir-
tue utilitarian. Instead, someone could be a virtue consequentialist where
the goal is to maximize the satisfaction of interests, and then offers an
objective list of interests that includes, among other things, continued
existence for the relevant nonhuman animals. But even on this sort of
approach, it doesn’t follow that most people should be vegan, given all
the other non-vegan options discussed so far: roadkill, mussels, lefto-
vers, insects, eggs from backyard chickens, some hunted animals, and so
forth—all of which fit with virtue utilitarianism too.
But suppose that we aren’t virtue consequentialists, and we tie our
account of the virtues to something like human flourishing. Then, I think
it will be even harder to make the case for veganism. After all, such
accounts should grant that there’s no essential connection between those
character traits that contribute to individual flourishing and those that
address collective action problems, or maximize utility, or otherwise
achieve some macro-level good. It will be a bit of good luck if individ-
ual flourishing is aligned with that macro-level good. And in nonideal
circumstances—which, of course, are our circumstances—it seems quite
plausible that the traits that promote individual flourishing will lead to
aggregate harms. In a broken world, there’s little reason to think that
the virtues will jointly favor strict veganism, as there will be so many
circumstances in which virtues will pull in the opposite direction. Most
obviously, perhaps, there will be all the times that being gracious and
accommodating to others might suggest being flexible with respect to
diet. Those cases alone spell trouble for an obligation to be a strict vegan.
Not only, then, will it be possible to defend unusual eating in this frame-
work, but lots of ordinary eating too.3
Moreover, when we think about the virtue attributions in nonideal
contexts, we immediately face a host of questions that further compli-
cate a defense of a general obligation. Is virtue attribution sensitive to
what it’s reasonable for people to believe given the epistemic environ-
ment in which they find themselves? Is it sensitive to the differences in
the difficulty of achieving certain virtues, based on the degree to which
the social, political, and economic institutions support that particular
character trait? Is it sensitive to the degree to which people aren’t fully
integrated selves? Is it sensitive to the various moral blind spots for which
they aren’t culpable due to details of their personal histories? If we give
an affirmative answer in each case—as I think we should—it will become
much harder to deny that these people are compassionate simply on the
basis of their consumption choices. Presumably, no one will be fully vir-
tuous (or compassionate) regardless of what they eat, and this means that
some lower bar for virtue-ascription will be appropriate.4
All that said, perhaps what’s most interesting about the way that
Hursthouse runs the argument is that it involves self-assessment rather
than the assessment of others. And when we go back to our (mostly)
vegan mom, Miranda, it’s easy to imagine her feeling conflicted about
her choice, and wondering how compassionate she can really be if she’s
willing to compromise in this way. But I submit that this kind of self-
evaluation only bolsters the case for Miranda’s being compassionate, as
she’s plainly sensitive to the kinds of considerations to which a virtu-
ous person would be sensitive. The virtuous person doesn’t always act
with a clear conscience, confident that she’s identified the one true path
through the moral complexities of life. Instead, the virtuous person often
acts with a palpable sense of the trade-offs involved in being virtuous
when the world is inhospitable to virtue. Miranda passes this test. Like-
wise, when we consider more ordinary cases, it’s probably important
that people feel some twinge of conflict about eating animals. If they feel
nothing—if it is, for them, a wholly uncomplicated activity—then that
does count against their virtuousness. But suppose that they recognize
the reasons to worry about intensive animal agriculture, seeing animal
product consumption as needing some justification, and suppose further
that they’re aware of the tradeoffs involved in all acts of eating: balanc-
ing environmental, nutritional, and worker- and animal-related issues.
Finally, suppose that they’re sensitive to futility concerns, recognizing the
value of symbolic action, but also that symbols only count for so much.
In that case, it’s plausible that their dietary choices don’t undermine their
claim to be virtuous in one respect or another. There are virtuous people
in line at the (non-vegan) burger stand.
8.2 Ecofeminism
Ecofeminists—who see animals as being oppressed by the same patriar-
chal structures that oppress women—offer a different sort of argument
for veganism. The idea is that we should understand veganism as resist-
ance. Carol Adams sums up the view this way:
Veganism is always a question of now. Knowing what I know, now

what will I do? It comes with an insistence! “Pay attention!” Pay
attention, now. The process of objectification/fragmentation/con-
sumption can be interrupted by the process of attention/newness/
compassion. The enlightened humanist subject, the sexual-politics-
of-meat subject insists on history. The carnophallogocentric subject
is made by history (“this is how I like it” or “change is hard”), is
shaped, more than he or she can admit, by the lies of the parents
(“meat is good for you”).
A refusal, a break, a fissure with that dominant history is
needed. . . . We aren’t just working toward a new text; we are work-
ing for a new world. Join [me] on the other side of that oppressive
history. Now.
(2010b, 315)
There’s a lot going on in this passage, not all of which we need to explore
here. The basic points, however, are these. First, there’s reference to the
linked oppression thesis. Very roughly, this thesis says that there’s an
important connection between oppressions, regardless of who’s being
oppressed. So, the oppression of women is somehow tied to the oppres-
sion of racial others, and both are somehow tied to the oppression of
animals. That’s the significance of the “carnophallogocentric subject,”
where there’s a history of the consumption of animals being linked to
male power, and Adams is critical of those who don’t resist the mold
into which that history places them. Second, there’s the thought that
veganism is a crucial part of resisting these linked oppressions. As she
puts it elsewhere:
Eating animals acts as mirror and representation of patriarchal val-

ues. Meat eating is the re-inscription of male power at every meal.
The patriarchal gaze sees not the fragmented flesh of dead animals
but appetizing food. If our appetites re-inscribe patriarchy, our
actions regarding eating animals will either reify or challenge this
received culture. If meat is a symbol of male dominance then the
presence of meat proclaims the dis-empowering of women.
(2010, 241)
By taking animal products off the table, we reject patriarchy—and rac-

ism, and colonialism, and so on.
Some questions. First, why think that the symbolism of animal prod-
ucts (or anything else) is fixed? There seems to be a lot of variation in
how people understand it across cultures and communities. Second, why
think that we’re all obliged to resist patriarchy in this particular way,
regardless of the details of our particular situations? Even if everyone
ought to take some steps to resist patriarchy, it isn’t obvious why this
particular step is so important. Third, why think that veganism will
always be a way to resist patriarchy, even if the linked oppression thesis
is true? It seems plausible that denying women access to animal products
will sometimes retrench patriarchy. Finally, on the assumption that activ-
ism is about changing some objectionable aspect of the world, doesn’t the
causal inefficacy problem give us good reason to worry about veganism
as a strategy?
However, let’s set aside these kinds of concerns and focus on the main
issue. If the linked oppression thesis is true, then that seems like a good
reason to be concerned about animal agriculture, and any good reason
to be concerned about animal agriculture is at least a reason to be con-
cerned about the consumption of animal products. And if we get that far,
then perhaps we can buttress the argument with various other premises
about the moral importance of political action, the scope of the harm
that can be traced back to the various linked “isms,” and so on.
But it isn’t at all clear that the linked oppression thesis is true. In say-
ing this, it’s important to acknowledge that there’s disagreement about
what, exactly, the connection is. Some contend that a common cause
explains each system of oppression, which means that you won’t be
able to address any particular form of oppression without attacking the
underlying issue. Others don’t see a common cause, but argue that the
mechanisms of oppression are still mutually reinforcing, so that male
domination of women (for example) enables human domination of ani-
mals, and vice versa. There are other proposals still. (For a sense of the
range, see Warren, 1990; Gruen, 1993; Kheel, 2004; MacKinnon, 2004;
Adams, 2010a.) But whatever the details, the central idea is this: if the
linked oppression thesis is true, we’d expect that progress on racial issues
will lead to progress on women’s issues, that progress on women’s issues
will lead to progress on animal issues, and so on.
There’s evidence that some of these links obtain. However, I won’t review
all that evidence here, as what matters is evidence for a specific link—
namely, between progress on any sort of human oppression and progress
on animal oppression. I know of only two attempts to examine this issue
directly, and both offer conclusions that are sympathetic to the ecofemi-
nist position. Rothgerber (2013) found that men and women used differ-
ent strategies to defend their meat consumption—for instance, men were
more likely to appeal to hierarchical justifications, to the effect that we’re
at the top of the food chain, whereas women were more likely to employ
dissociative and avoidant meat-eating “justifications,” such as saying that
they prefer not to think about what happens in slaughterhouses. Moreover,
he found a positive correlation between the “male” strategies and stereo-
typically masculine traits, as well as that employing “male” strategies was
associated with greater meat consumption. Relatedly, Allcorn and Ogle-
tree (2019) found that employing “male” strategies is positively correlated
with scores on a benevolent sexism scale, which looks for paternalistic
attitudes toward women and strict differentiation in terms of social roles.
However, we shouldn’t make too much of this research. It would be
difficult to find Americans who behave as though there are normal, adult
humans of any sex or gender who lack the most basic rights—the right
to life and the right to bodily integrity—which impose corresponding
duties on others.5 By and large, we don’t kill or maim one another, nor
do we tolerate having others do so on our behalf. At best, though, vegans
and vegetarians make up five percent of the US population, and the vast
majority of available animal products—around 99%—are derived from
factory farms.6 These points strongly suggest that most Americans have
internalized norms about humans that they haven’t internalized about
animals. More accurately, it suggests that we’ve made massive progress
on human rights while making virtually no progress on animal rights.
That isn’t what we’d expect if the linked oppression thesis were true.
Additionally, if the linked oppression thesis were true, we shouldn’t
expect to find so many cases where there are tensions between the inter-
ests of women (and racialized communities, and members of the LGBT
community, and so on) and the interests of animals. After all, women
benefit, and animals don’t, when we use them and their byproducts for
countless purposes: food; medical research; and to make a remarkable
number of consumer goods, from clothing to condoms to crayons. Addi-
tionally, women benefit whenever we ignore animal interests by con-
structing new buildings, making new roads, and creating new parks, all
of which involve altering local environments in ways that make them less
conducive to animal flourishing.
The same is true when it comes to the global poor, the vast majority of
whom are nonwhite. Surely we ought to improve human access to food,
water, shelter, and medical care. Insofar as we’re successful in this, we’re
likely to increase the human population. And the more of us there are,
the more food we’ll need, the more land we’ll need for our purposes, and
the more resources we’ll extract from the environment. It’s unlikely that
any of these things will be good for animals. Promoting food security in
the developing world, for example, often involves promoting animal hus-
bandry, since in that context animals are, essentially, ways of recovering
and storing energy that would otherwise be lost. Pigs can convert food
scraps, plants that are inedible for humans, insects, and much else besides
into nutrient-rich flesh, for which they can be slaughtered whenever the
need arises. That’s certainly good for people, but less so for pigs.
In sum, it seems unlikely that we can make an ecofeminist argument
for a general obligation to be vegan that’s based on the linked oppression
thesis, as it’s unlikely that the linked oppression thesis is true.
8.3 The Argument by Analogy

Maybe we would do better not to employ a general framework, focusing
on a simpler strategy instead. This has been the thought behind a number
of “commonsense” arguments for veganism—for instance, those by Nor-
cross (2004), Engel (2016), and Huemer (2019). Here, for instance, is the
version that Engel offers. After defining a “HASK practice” as “any prac-
tice that involves intentionally Harming, Abusing, inflicting Suffering on,
or Killing sentient animals for no good reason” (2016, 5, emphasis in
original), he runs this argument:
1. All forms of meat-producing animal agriculture are HASK practices.

2. It is wrong to engage in, or pay others to engage in, HASK practices.
3. When one purchases and consumes meat . . . one is paying others to
engage in HASK practices on one’s behalf.
Therefore,
4. It is wrong to purchase and consume meat.

(2016, 8)
Let’s focus on part of the second premise. Why should we believe that it’s
wrong to pay others to engage in HASK practices? Engel defends this by
providing a case where someone raises a dog (“Mocha”) for food, treating it
in much the way that pigs are treated in factory farms. Then, he writes this:
Suppose that I didn’t abuse and kill Mocha for myself. I did it for
“Carni,” a meat eater who is too squeamish to raise and kill her
own food. Now is my behavior justified? Absolutely not. Abusing
an animal for no good reason is never justified, and all decent people
recognize this fact. Since both Carni and I can easily meet all of our
nutritional requirements with a plant-based diet, my mutilating, con-
fining, and killing Mocha as described above is morally abominable;
the only difference now is that, since Carni has paid me to do it, she
too is morally culpable for that innocent animal’s suffering.
(2016, 4)
Perhaps we can’t specify the general moral principle that explains both
why Carni acts wrongly and why it’s wrong to buy meat from factory
farms. Still, we can see that Carni acts wrongly, and that it isn’t relevantly
different from what consumers do. Hence, we can see that the latter is
wrong.
Plainly, though, there is a difference between Carni’s behavior and the
behavior of ordinary consumers: if Engel didn’t abuse and kill Mocha,
then Mocha wouldn’t be harmed. But no individual has that kind of con-
trol over what happens on factory farms.
Moreover, once we recognize this, we can see a deep problem in all
arguments of this kind, which are quite popular among defenders of
veganism (see, e.g., Singer, 2009; Rachels, 2004). The issue concerns
what counts as a good reason. When Engel explains why you have no
good reason to eat animal products, he does it by arguing that a vegan
diet is good for your health, easier on your wallet, and no more time
consuming to prepare. So, he claims, the only remaining reason is taste,
which he regards as a trivial consideration. And when we compare your
gustatory pleasure to the experience of being a pig in intensive confine-
ment, it doesn’t look like the former justifies the latter. However, given
my earlier arguments for causal inefficacy, you don’t need a particularly
strong reason to justify purchasing and consuming the relevant products.
Granted, you do need a very strong reason to justify torturing a dog or
a pig. But if you’re not doing that, and can’t stop anyone from doing
it, then you don’t need a very strong reason to purchase and consume
a product that’s good for you (in moderation), cheap, easy for you to
prepare, tastes great, and so on. Those reasons, even if weak in the grand
scheme, seem perfectly adequate.
Any commonsense argument is going to have to have some premise
that forges the link between the ethics of production and the ethics of
consumption. And when that premise is defended via an analogy, we also
have to ask about what happens when we play out the scenario that the
author employs. Consider the way that Huemer makes the case for the
crucial premise:
You have a friend named “Killian,” who happens to be a murderer.

One day, you offer Killian $20,000 to get you a new car. Killian could
carry out this task in a perfectly moral manner. But you know that
the way he will in fact do it is by murdering some innocent person
and stealing their car. You know this because Killian has performed
tasks like this for you in the past, he always murders people along
the way, and you always pay him for it afterward. You don’t specifi-
cally tell him to murder anyone; you just know that that’s the way he
does things. So you tell Killian to get a car, he goes off, kills someone,
steals their car, gives it to you, and you pay him $20,000. . . . [That’s]
like buying factory farmed meat. You didn’t tell them to commit acts
of extreme cruelty, but you know that that is how they do things, and
you keep paying them for the product.
It certainly seems wrong to tell Killian to get a car. But now imagine that
I’m a business owner, and I find out that some of my employees use their
wages to hire Killian. I would think that I’d have an obligation not to pay
those employees. I shouldn’t be giving people money to go out and hire
those who will commit murder! (Someone might doubt this. But consider
the same choice in a more immediate context. You hire a plumber to fix
a leaky faucet. As he’s working, he starts talking with you about an old
vendetta, and it becomes clear to you that this isn’t idle chit-chat: he
means to kill someone. He also reveals that he’s been short on cash for
a while, and this job is the one that’s going to finance the gun purchase
for the next “job.” It seems to me that if there’s any way you could avoid
paying him, you should.) However, it doesn’t seem at all plausible that
I shouldn’t pay my employees because they buy animal products with
their wages.
Huemer might reply that there is a crucial difference. A car from Kil-
lian is relatively expensive, but most of a person’s paycheck doesn’t go
toward animal products. So, although we do have a moral reason not to
pay employees because they buy animal products, the reason just isn’t
very strong. That seems plausible enough, but the analogy doesn’t sup-
port it. If I knew that my employees were paying Killian to do his dirty
work, I would feel that I shouldn’t employ them at all. The strength of the
(felt) reason wouldn’t diminish because I was an additional step removed
from the activity, or because it was a relatively small portion of the pay-
check that I was giving them. But I feel no similar impulse when it comes
to those who consume animal products.
Of course, none of this shows that these analogical arguments are use-
less. The above does show, however, that such arguments only get you a
fairly weak principle, for instance, that if it’s wrong to produce x, then
you have a moral reason not to pay money for x. But to determine the
strength of the reason, we would have to appeal to other kinds of con-
siderations, such as causal impact. In the my-employees-pay-Killian case,
the strength of the reason might be fairly strong; in the case of purchas-
ing animal products, the strength of the reason might be fairly weak—or
even nonexistent—given the causal inefficacy problem. And again, this
reason also counts against supporting plant agriculture when harm-free
alternatives are available, and perhaps even when the alternatives involve
harms to beings that are less obviously sentient than the animals harmed
by plant agriculture.
8.4 The Holocaust Argument

Some people think that we can solve this “strength of the reason” prob-
lem by highlighting the scope of the harm to nonhuman animals. Stuart
Rachels (2011, 897–898), for instance, writes this:
Let’s assume, very conservatively, that during the last twenty

years, around five billion animals per year have suffered in Ameri-
can factory farms, which amounts to 100 billion suffering animals.
And let’s assume that the Holocaust caused suffering to 20 million
human beings. This means that, for every single human being who
suffered in the Holocaust, five thousand animals have suffered in
American factory farms during the last twenty years. And really,
this calculation greatly underestimates the ratio. It ignores all the
intensively farmed fish; it ignores all the animals that suffered in
factory farms but died before slaughter; it ignores all the farm ani-
mals that suffered more than twenty years ago; and it ignores all
the human victims of industrial farming. Pain calculations are hard
to make, but a five-thousand-to-one (or much greater) ratio makes
this judgment easy: industrial farming has caused more suffer-
ing than the Holocaust. . . . [Nevertheless, let’s suppose] that, for
whatever reason, human pain is ten times worse than animal pain.
On that assumption, factory farming over the last twenty years has
still caused pain morally equivalent to five hundred Holocausts.
Or suppose there’s only a 10% chance that the [pro-vegan] argu-
ments in this paper are correct. On that assumption, factory farm-
ing, again, has had the expected utility of five hundred Holocausts.
And if there’s only a 10% chance that animal pain is 10% as bad
as human pain, then factory farming has had the expected utility
of fifty Holocausts (or really more, since I’m ignoring a lot of the
suffering caused by industrial farming).
We might summarize this argument as follows, filling in a suppressed

premise:
1. Factory farming has caused far more pain than the Holocaust.
2. You shouldn’t consume products from an industry that’s caused far
more pain than the Holocaust.
3. So, you shouldn’t consume products from factory farms.
The second premise is quite compelling. The felt pressure to disassociate
is powerful. Nevertheless, I still think we have reason enough to deny
that the argument is sound.
Before saying why, exactly, we should note that there are familiar
worries about this kind of argument. While the Holocaust was certainly
bad because of the pain it involved, it wasn’t a moral atrocity for that
reason alone, and it’s worth considering whether our pro-disassociation
intuition is, first and foremost, a response to factors other than the sheer
the quantity of pain. For instance, the Holocaust was a moral atrocity
because it was genocide, because the Nazis were wantonly cruel, because
the Nazis actively tried to humiliate their victims, which involves recog-
nizing and working to subvert their dignity. I agree that animals are often
wronged in intensive systems, but producers aren’t genocidal. I agree
that intensive systems are cruel, but it’s tricky to attribute that cruelty to
particular individuals in the industry. And I agree that intensive systems
violate animal dignity, but not because producers are actively trying to
subvert it. (Indeed, they often think of themselves as providing the best
lives they can for the animals they raise.) So we need to be careful about
making simplistic inferences from our judgments about the Holocaust to
animal farming.
Second, and as discussed at the end of Chapter 6, our duties to disas-
sociate may be tied to the symbolic significance that victims (and those
related to them) assign to our actions. If the victims of the Holocaust (or
their descendants) would have wanted us to abstain from buying some
product, that would be an excellent reason not to buy it. If, for instance,
certain products were made with concentration camp labor, and modern
day Jews want those objects to be in museums rather than sold as col-
lectibles, then it’s plausible that those objects should be placed in muse-
ums. But since animals don’t care about the symbolic significance of our
actions, and only about the effects that those actions have on their well-
being, consequentialist reasoning about disassociation is appropriate.
Third, let’s consider a parallel argument:
1. Colonialism caused far more pain than the Holocaust.

2. You shouldn’t consume products from a practice that caused far
more pain than the Holocaust.
3. So, you shouldn’t consume products derived from colonialism.
The first premise is probably true, but the second is implausible. Why?
Two answers present themselves: first, the ubiquity of products derived
from colonialism and, second, the fact that your consumption choices
make no difference to whether various European countries systematically
extracted resources and people from the rest of the world. However, if
these are good answers, then we should note that they tell against the
first argument too. Like the products derived from colonialist practices,
products derived from animals are everywhere, both in food and non-
food items. And if the earlier arguments about causal inefficacy are
sound, then your consumption choices make no difference to whether
animals are raised and slaughtered for food.
We’re sensitive to these points in historical cases, but we tend to ignore
them when the wrong is ongoing. I suspect that this is because we hope—
reasonably or unreasonably—that we can be part of the solution to a
very serious problem. Granted, intensive animal agriculture is a very seri-
ous problem. It’s morally admirable to be willing to sacrifice in hopes of
chipping away at its stranglehold on the food system. But people aren’t
obligated to act in accord with even reasonable hopes. That’s veganism
as a kind of political act, not as a moral duty.
Plainly, it might be a very good thing to take political action. Nothing
I say here is meant to discourage that choice, and indeed, I’d like to see
more people get on board. But in general, people aren’t obligated to join
every political movement that it would be good to join.
8.5 An Epistemic Argument?

Let’s consider one last pro-vegan argument that seems, at least at first
blush, to avoid causal inefficacy worries, as well as any worries about the
production/consumption gap. It isn’t an argument that’s already in the
literature; I’ve only heard versions of it in conversation. I used to think
that it would carry the day, closing the gap between the ethics of produc-
tion and the ethics of consumption. This epistemic argument for vegan-
ism is based on the ethics of belief. In short, meat-eating promotes false
beliefs about animals, and all else being equal, it’s wrong to act in ways
that promote such false beliefs. Thus, you shouldn’t eat meat. However,
I’m no longer so optimistic.
8.5.1 A Stacked Deck

To see why I was so optimistic, though, we need to consider how our
psychologies are stacked against animals.
Psychologists seem to agree that affect plays a significant role in moral
judgment (though the details are controversial). However, our affective
responses aren’t triggered in perfectly consistent ways. This is a special
problem when it comes to animals, as Hal Herzog (2010) points out,
because entire species can be marginalized by our affective tendencies.
For instance, we’re strongly disposed to have positive sentiments toward
anything that looks like a baby, which means that we’re much more will-
ing to act on behalf of cute animals (i.e., those that have expressive faces
or baby-like features, such as cats and dogs) than those that aren’t so
adorable (e.g., snakes and spiders).7 This turns out to be one of the many
factors that conspire against our being sensitive to the interests of fish.
Fish have rigid faces that don’t express emotion in ways that are easy
for humans to discern, they’re slimy, and they live in an environment
that’s largely foreign to us—one that, historically, has provoked fear. As
a result, very few people would describe fish as “cute,” which means
that they tend not to be on our moral radar. Even animal activists tend
not to focus on fish: Compassion in World Farming USA, a major activ-
ist group, is one of many animal advocacy organizations that prioritizes
laying hens above all else, even though fish are killed at exponentially
high rates, and over the course of their longer lives, farmed fish probably
suffer just as much as chickens do.8
Additionally, we tend not to assign moral significance based on a care-
ful assessment of a being’s intrinsic properties. T. J. Kasperbauer (2016)
makes this point nicely by summarizing research that supports two
claims: first, our judgments of moral significance track our judgments
about phenomenal states, and, second, we tend to attribute phenomenal
states to animals based on physical and behavioral similarity to human
beings. In other words, we tend to think that animals matter insofar as
they have experiences like pleasure and pain, and we’re inclined to attrib-
ute the capacity to have such experiences insofar as we judge them to be
similar to us. Unfortunately for many animals, however, we don’t seem
to think we’re that similar.
In defense of the first claim, consider Knobe and Prinz (2008), in which
the authors described a scenario for participants about a researcher who
was studying fish. They told one group that the researcher was doing a
memory study; they told the other group that the researcher was assess-
ing whether fish had feelings. They then asked, “Why do you think he
[the researcher] might want to know this? Why might the question be
important to him?” Everyone in the feelings group answered in terms
of the moral significance of the research, while only ten percent in the
memory group made any such connection. Similarly, Jack and Robbins
(2012) found that people are less concerned about lobsters when lob-
sters are described as intelligent and having good memories, and more
concerned about them when they’re described as possessing the ability
to be depressed or feel anxious. Sytsma and Machery (2012) found that
people’s views about the permissibility of an experiment on primates
were affected by descriptions that included the experiential states of those
primates, but not by descriptions that emphasized agency and cognitive
capacities.
In defense of the second claim—that we tend to attribute phenomenal
states to animals based on physical and behavioral similarity to human
beings—there are studies like those done by Phillips and McCulloch
(2005), who asked people to compare various animals to normal adult
humans in terms of the capacity to feel pain, happiness, fear, and boredom.
The sample was diverse and international. They aggregated the scores to
create a sentience index, which was consistent across countries: monkey,
dog, newborn baby (human), fox, pig, chicken, rat, fish. People thought
that monkeys, dogs, and newborn babies were roughly 80% akin to nor-
mal adult human beings, whereas fish were about 47%. Plainly, animals
on the lower end of the index look and behave less like normal adult
humans than those at the upper end. Likewise, Phillips et al. (2011) asked
people to compare animals to one another as opposed to normal adult
humans and got a very similar ranking. Neither study establishes causa-
tion, but these results are indeed what we’d expect if our attributions of
phenomenal states are based on physical and behavioral similarity.
The indirect evidence just surveyed is bolstered by research indicat-
ing that the presence and absence of physical parallels between humans
and animals predict whether and to what degree the members of a spe-
cies will be assigned moral importance. Westbury and Neumann (2008)
argue that empathic emotional responses to abused animals increased
according to phylogenetic similarity, Plous (1993) found the same result,
and Allen et al. (2002) showed that phylogenetic similarity predicts how
severely people want to punish others for animal abuse.
There are, of course, perfectly reasonable explanations for these ten-
dencies in humans. We can’t get inside the minds of animals, so we have
to attribute mentality to them via some method or other. As it happens,
our method appears to be analogical reasoning, the results of which feed
into a framework for assigning moral importance. But imagine that we
discovered analogous mechanisms in humans for assessing the moral
importance of racial others. We would probably regard this as evidence
of serious bias against those others—bias that it would be difficult to cor-
rect without addressing the underlying cognitive mechanisms. And that,
it seems, is our situation vis-à-vis animals.
Our affective tendencies can marginalize animal interests, and that’s
compounded by the way we assign moral significance. Things are made
yet worse for animals due to an underlying feature of our cognitive
architecture. Warrington and Shallice (1984) were the first to describe
unusual deficits in four patients who were recovering from encephalitis:
they’d all become much poorer at naming and recognizing the names
of living things, even though they did fine with non-living things. J. B.
R., for example, could only name or accurately describe 2 of 48 ani-
mals and plants, while he was able to name or accurately describe 45
of 48 inanimate objects. A number of studies have found similar deficits
in other patients—see the references in Caramazza and Shelton (1998,
2006)—and there are cases where people struggle specifically with iden-
tifying animals. E. W., for instance, could only name 12 of 22 animals,
while she could name 18 of 22 plants and 100% of non-living items.
She also struggled to distinguish real animals from fictional ones (e.g.,
a horse-bear chimera), though she could distinguish normal household
objects from strange hybrids (e.g., a water pitcher with an ax handle),
and she was equally bad at judging whether statements about animals
were true versus statements about non-animals (e.g., “Does a whale
have eight legs?”). Perhaps most strikingly, this deficit held across sen-
sory modalities: she was no better at identifying animals by their char-
acteristic sounds—though again, she did fine with non-animals. These
unusual deficits constitute part of Caramazza and Shelton’s evidence
for the hypothesis that the brain has innate, domain-specific knowledge
systems. This hypothesis receives additional support from the fact that
the relevant domains would have been fitness-enhancing for our ances-
tors (e.g., animal, plant, conspecific); it also fits nicely with fMRI studies
indicating that different parts of the brain are activated by human faces,
animals, and inanimate objects, as well as that the same brain regions
are activated when sighted people view animals and when congenitally
blind people hear the names of animals. It may well be the case that
we have certain cognitive modules that are specifically equipped to deal
with animals, predisposing us to carve up the world in certain ways. Of
course, it’s possible to see beings as different and yet morally equal, but
the deeper the difference is perceived to be, the harder it is to preserve
judgments of moral equality.
We can see the outcome of these factors (among others) in some adap-
tations of the trolley problem. Petrinovich, O’Neill, and Jorgensen (1993)
and O’Neill and Petrinovich (1998) did both US-based and cross-cultural
research on how students respond to cases like these:
An out-of-control trolley is headed toward a group of the world’s last

five remaining mountain gorillas. You can throw a switch and send it
toward a twenty-five-year-old man. Should you?
The trolley is speeding toward a man whom you do not know. But
you can throw a switch and send it hurtling toward your pet dog?
Should you?
It turns out that people almost always choose to save the human in cases
like these, and that of all the rules that help us predict people’s responses
to trolley scenarios, “Save people over animals” is the most reliable one.9
Thus far, the upshot is that our sympathies don’t seem to track morally
relevant properties and our ways of divvying up the natural world look
to be stacked against animals. What’s more, these tendencies seem to help
explain a fairly deep pro-human bias in our moral judgments.
Additionally, there seem to be reasons to be concerned about eating
animals specifically. Consider some studies that appear to show that eat-
ing animals is associated with forming less accurate beliefs about them.
For instance, Bilewicz, Imhoff, and Drogosz (2011) did three studies
that brought out differences in the ways that vegetarians and omnivores
attribute mental traits to animals. First, they found that vegetarians were
more likely to attribute secondary emotions to animals (e.g., love and
depression) versus primary emotions (e.g., happiness and pain). Second,
vegetarians were much less likely to regard human beings as having
unique character traits, such as being assertive, anxious, and generous,
that distinguish humans from animals. Third, the researchers asked par-
ticipants to assess how likely it is that either dogs or pigs experience
a dozen primary or secondary emotions. Vegetarians were more likely
to say that both animals experience all those emotions, but the striking
finding was this: omnivores were about as likely to think that dogs expe-
rience all the primary and secondary emotions as vegetarians were, but
they were much less likely to think that pigs experience those emotions.
We might wonder whether these effects are explained by different
knowledge bases in the two populations: for instance, people are gener-
ally more familiar with dogs than they are with pigs, and it seems reason-
able to assume that many vegetarians would have learned more about
the capacities of pigs than many omnivores.10 However, the literature
suggests otherwise. Bratanova et al. (2011) told participants about a
marsupial native to Papa New Guinea, suggesting to some that the mar-
supial is a food animal that humans hunt, while not giving such hints to
the control group. Suggesting that the marsupial is a food animal was
enough to lead participants to judge that the animal is less likely to have
the capacity to suffer and less deserving of moral concern. It looks like
simply categorizing animals as food has an impact on our judgments
about them. Bastian et al. (2012) found something similar. They did a
number of studies showing that people think you shouldn’t eat animals
that have minds, that people are prepared to deny that the animals they
eat have minds, and that the denial of mental states to animals increases
when people think that they are going to eat some meat rather than a
piece of fruit. Finally, Loughnan, Haslam, and Bastian (2010) did a study
in which omnivorous students were randomly assigned to consume either
beef jerky or cashews before assessing both the mental capacities and
moral importance of a cow. The students who ate beef jerky were less
likely to attribute mental states to the cow, and less likely to say that it
would be bad to harm the cow. (For other results in the same vein, see
Ruby and Heine, 2012; Rothgerber, 2014a, 2014b; Kunst and Hohle,
2016.) These all seem like ways that omnivores disengage from the ani-
mals they eat: they deny them mental states, and they deny them moral
standing.
When we put these points together, it seems that we have good reason
to worry about the reliability of our judgments about the mental states
and moral importance of animals. Might that potential unreliability be
morally significant?
8.5.2 The Badness of Eating Animal Products

All else being equal, it seems morally bad to form false beliefs about
animals, and so morally bad to act in ways that increase the likelihood
of forming false beliefs about animals. Here’s an argument that takes us
partway toward that conclusion:
1. It’s epistemically bad to act in a way that makes you more likely to
form inaccurate beliefs.
2. Eating meat makes you more likely to form inaccurate beliefs.
3. So, it’s epistemically bad to eat meat.
4. If an action is epistemically bad, and if the affected beliefs (or the
beliefs that would be affected, were you to form them) concern a
matter of moral importance, the action is morally bad in that respect.
5. The affected beliefs (or the beliefs that would be affected, were you
to form them) concern a matter of moral importance.
6. So, there is one respect in which it’s morally bad to eat meat.
The first premise—that it’s epistemically bad to act in a way that makes you
more likely to form inaccurate beliefs—is the least controversial. If anything
is of epistemic value, it’s truth, and so whatever makes you more likely to
believe falsely has epistemic disvalue. Imagine, for example, that you had an
opportunity to take a pill that would temporarily strengthen your inclina-
tion to rationalize your actions rather than scrutinize them more objectively.
Presumably, there’s an epistemic consideration against taking the pill: taking
it will make it harder for you to form accurate beliefs about whether your
behavior is justified, and that’s (epistemically) bad. Of course, the experi-
ence might be relaxing or entertaining, so it may be the case that, all things
considered, the pill is worth taking. (I take it that this is roughly the way to
defend having a few drinks.) At this juncture, my point is just that there is an
epistemic cost, whatever other benefits there may be.11
Much of the evidence for the second premise has already been stated.
However, we need to say a bit more, because although the studies described
earlier make a compelling case that meat consumption affects how we
think about animals, they don’t show that eating meat increases the odds
that our beliefs about animals are inaccurate.12 To make that conclusion
plausible, you need three further claims: first, that animals have the men-
tal capacities that omnivores are less likely to attribute to them; second,
that animals are of greater moral importance than most omnivores tend
to think; third, that if the former claims are true, we have a moral reason
not to eat meat (though not necessarily a decisive moral reason not to
eat it). These claims aren’t particularly difficult to defend. There is ample
work on the cognitive abilities of standard food animals—chickens, pigs,
cattle, and various species of fish—that reveals their varied emotions,
complex social lives, problem-solving skills, impressive memories, and
so on. On fish, chicken, and pigs, see Brown (2015), Marino (2017), and
Mendl, Held, and Byrne (2010), respectively. No similar review exists
for bovines, though there’s plenty of evidence that they are surprisingly
sophisticated creatures—see, for instance, Hagen and Broom (2004).
Moreover, there are familiar arguments for the view that animals
deserve more moral consideration than ordinary people tend to give
them, though I won’t survey them here. Instead, I’ll just mention that
even defenders of meat consumption almost never defend intensive ani-
mal agriculture, which supplies the vast majority of the animal products
available in the US.13 So, while there certainly isn’t consensus among phi-
losophers about exactly how important animal interests are relative to
human interests, the literature suggests that there is a near consensus that
the dominant view of animal importance in the wider culture is mistaken.
And on the ethics of eating meat specifically, we should note that Schwit-
zgebel and Rust (2016) did a survey of philosophers on this issue (among
others), finding that 60% of ethicists, and 45% of philosophers gener-
ally, assessed meat-eating as morally bad—which is actually a stronger
claim than the one I need, which is only that there is a moral reason not
to eat meat, even if it’s morally good to do so all things considered.14
That’s hardly consensus, but it’s still a very significant difference between
the views of philosophers and the views of the general public. And, of
course, the inclination to say that meat-eating isn’t morally bad may well
be explained by not having reflected seriously on the arguments for the
view that it is. All told, that seems like enough evidence to make the
second premise plausible: eating meat increases the odds that you’ll form
inaccurate beliefs.15
Given the first two premises, we get the first conclusion: it’s epistemi-
cally bad to eat meat. We now enter the second stage of the argument,
where the third premise forges a link between epistemic badness and
moral badness:
If an action is epistemically bad, and if the affected beliefs (or the

beliefs that would be affected, were you to form them) concern
a matter of moral importance, the action is morally bad in that
respect.16
I suspect that a principle along these lines is behind certain feminist cri-
tiques of pornography—for instance, Hill (1987) and Langton (2009).
The more familiar charge against pornography is that it increases the
risk of sexual violence against women, but that certainly isn’t the only
objection that’s been leveled. Additionally, many feminists have thought
that viewing certain kinds of pornography makes it more likely that
you’ll form inaccurate beliefs about women. For instance, perhaps some
violent pornography increases the odds that male viewers judge women
to be mindless, masochistic nymphomaniacs. That seems morally bad,
and it seems morally bad independently of whether it has any negative
consequences for women. Among other things, it’s disrespectful to dis-
pose yourself to believe false things about people, especially where those
beliefs, if formed, would concern morally significant dimensions of their
identities. So if it’s indeed the case that watching violent pornography has
this consequence, then watching it is morally bad in at least one respect.
Likewise, I think that a principle like the one above is behind certain
arguments about the ethics of humor. Consider jokes about race, gender,
and sexual identity. You might worry that telling these jokes can make
you less sensitive to the plight of those the joke concerns, that telling
them can make you more likely to believe that the challenges these peo-
ple face are less significant than they really are. If telling these jokes does
have this effect, then it seems morally bad to tell them.
Of course, it’s an empirical question whether violent pornography dis-
poses viewers to judge women to be mindless, masochistic nymphomani-
acs, just as it’s an empirical question whether telling jokes disposes us to
have false beliefs about the people those jokes concern. I take no stand
on either issue here. The point isn’t that the feminist critiques succeed,
or that it’s morally problematic to tell jokes about members of particular
racial groups. Rather, the point is that there seems to be something to
those arguments on the assumption that the empirical claims are true. In
the present context, what matters is whether viewing violent pornogra-
phy would be morally bad if it were to have that effect, and likewise for
telling such jokes. It seems to me that they would be. That is, even if the
criticisms of certain forms of pornography and humor are unsuccessful,
I’m inclined to explain their appeal by saying that each combines a plau-
sible moral principle with a plausible, but highly controversial, empirical
hypothesis. Given as much, I’m happy to rely on the principle here.
This brings us to the fourth premise:
The affected beliefs (or the beliefs that would be affected, were you to
form them) concern a matter of moral importance.
Here’s something that’s of obvious moral importance: the question of

how important animal interests are. After all, if animals aren’t terribly
morally important, then there may not be anything wrong with animal
agriculture. But if philosophers like Singer and Regan are correct, then
animals are wildly more important than most people take them to be,
and an entire industry is guilty of very serious moral wrongdoing. More-
over, given that people assess moral importance partially based on cogni-
tive and emotional capacities, beliefs about those capacities are of moral
importance too. Given all the above, we get the second conclusion: there
is one respect in which it’s morally bad to eat meat.
For all I’ve said so far, however, there is nothing morally bad about
eating any other animal products. Can we go further?
I think so. Admittedly, all the studies I cited focus on meat, not eggs
or dairy, and there is no research of which I’m aware that focuses on
these other foods. Therefore, we lack direct empirical evidence that eat-
ing those products has the same cognitive effects. What’s more, we kill
animals to get their flesh, whereas we don’t kill a hen when we take its
egg. Obviously, there is plenty of suffering and death associated with
egg production, but because the connections aren’t as obvious, it may be
more difficult to avoid associating meat with suffering and death. This
increases the incentive to rationalize meat consumption by denying that
animals have morally relevant cognitive capacities. Given all this, the
cognitive consequences of egg and dairy consumption may not parallel
those of meat consumption.
However, I think the latter point helps us see what might be wrong
with the studies that psychologists could pursue concerning these foods.
If we rely on the background knowledge that people have about ani-
mal agriculture—which is obviously poor—we may not find that egg and
dairy consumption are associated with the same cognitive consequences.
However, it seems very plausible that such consumption would be associ-
ated with the same cognitive consequences were participants to be edu-
cated about the harms involved in those parts of animal agriculture with
which they are less familiar. After all, if the effects are explained by an
inclination to rationalize behavior—to say whatever it takes to legitimize
the harm in which meat-eaters are complicit—then we should expect
similar effects when the pressure to rationalize is just as strong. And, pre-
sumably, it would be just as strong if participants were better informed.
Given this point, even if it isn’t epistemically bad to eat eggs or dairy
now, it’s easy to make it epistemically bad by providing people with a
bit more information about contemporary animal agriculture. And since
animal activists are making great efforts to do just that, the argument will
work against an increasingly large group of individuals.
That said, we don’t need to wait for the activists to do their work: for
most consumers, it probably is epistemically bad to eat eggs and dairy
now. This is because it’s epistemically bad to be epistemically negligent,
which may explain why most ordinary consumers don’t know what hap-
pens in egg and dairy production, and so why they are more likely to
believe, falsely, that there are few reasons to object to the associated
products. However, that negligence is surely motivated, at least in part,
by egg and dairy consumption itself. Consumers don’t learn what they
don’t want to know—and they don’t want to know about egg and dairy
production because of the way they benefit from it. So, the act of eating
itself is still epistemically bad, and the argument goes through.
8.5.3 The Wrongness of Eating Animal Products?

It’s tempting to add another premise at this juncture:
If an action is morally bad in one respect, then that action is morally

wrong unless the badness is outweighed by some good(s) (includ-
ing, but not limited to, moral goods).
Unfortunately, it’s difficult to infer that it’s morally wrong to consume
animal products on this basis, since the argument above doesn’t tell us
how morally bad it is to eat animal products. And again, if we rely on
considerations about the impact that any individual has based on her
potentially false beliefs, it’s likely that it isn’t very bad at all, as the indi-
vidual’s impact is minimal. Worse, it would be a mistake to attribute
whatever impact she makes entirely to her false—or potentially false—
beliefs. Instead, we should consider the difference that those beliefs make
to her action, which may be relatively minor (perhaps the odds were
already high that she would have purchased and consumed those animal
products). So, the impact is going to be even smaller still.
This problem might be patchable. We could think about the badness
as stemming from a failure to meet a general responsibility to act in ways
that promote true beliefs rather than false ones. Notice that this way of
framing things doesn’t rely on any assumptions about the interests of
animals. After all, we might think that human beings have an interest
in having others believe truly about them because we care about our
reputations, and the beliefs others have about us are obviously relevant
to whether our reputations are good. Presumably, the animals we tend to
eat have no such interests, as they aren’t able to form the relevant beliefs
about the beliefs of others. However, if we frame things in terms of our
general epistemic-cum-moral obligations, we evade this issue entirely.
Nevertheless, we should recall that the degree of badness is deter-
mined by the likelihood of forming false beliefs in virtue of what you eat.
Those already committed to anti-speciesism (for instance) are probably
much less vulnerable to these kinds of effects, and it will be relatively
easy to have the badness outweighed by other sorts of considerations.
These “other sorts of considerations” could, of course, include things
like health, taste, politeness, tradition, and the like. But they could also
be tied to the notion that we have reason to eat unusually. After all,
someone convinced by the standard arguments for veganism may still
have good reason to eat, say, a dented can of beef soup that was recov-
ered from a grocery store dumpster. By consuming the soup you can take
one small step toward reducing support for the harms to animals that are
involved in plant agriculture, reducing food waste, and avoiding spend-
ing funds that can be directed toward morally valuable ends.
At the same time, while committed anti-speciesists may be less vulnera-
ble to the epistemic consequences of animal product consumption, there’s
no reason to think that they’re immune. After all, it isn’t easy to tell
how our thinking is being subtly affected by our behavior, so we should
be wary of claims to the effect that we can maintain our moral outlook
however we act. Moreover, vegans may also see themselves as having
more to lose, morally speaking, than omnivores, since their “vegan wit-
ness” is valuable: even if they don’t regard it as a “slip up” to eat those
animal products, vegans risk being perceived as hypocritical, and so as
unworthy of emulation. Insofar as that’s relevant to whether you have
standing to make the case for veganism, and insofar as it’s important to
convince others that they ought to be vegans, there is risk involved. So if
you have reason to believe that eating animal products will increase the
odds of your having inaccurate beliefs about animals, and you have the
long game in mind, it may not be worth it to take any epistemic chances
that are associated with consuming animal products.
Is this sort of observation enough to show that there is a duty to be
vegan? Not obviously. First, the reasoning in the last paragraph doesn’t
rely very heavily on the epistemic-cum-moral argument that I’ve been
developing, but rather on strategic considerations for those concerned to
convert others to veganism. So it’s only if we ought to be animal advocates
that we ought to be this sensitive to the way that we’re perceived by others.
But again, we are considering whether you have a duty to be vegan, not
whether you have a duty to be an animal advocate. Second, even if you do
have good reasons to be an animal advocate, without some assessment of
the strength of those reasons, we won’t yet know whether they outweigh
the other kinds of goods that may be at stake. Moreover, we have yet to see
an argument for the conclusion that the only way to be an animal advocate
is by being a vegan animal advocate. So, though the epistemic argument for
veganism is an intriguing one, I don’t think that it succeeds.
8.6 Eating Unusually?

What’s the upshot of the last five chapters? The most important points,
I think, are these:
1. It’s difficult to supply a premise such that you can infer the wrongness
of purchasing or consuming all animal products from the wrongness
of producing them. This is largely because of the causal inefficacy
problem: if we accept that individual consumers don’t make a dif-
ference (even if they do collectively), then many arguments against
purchasing and consuming animal products are either weakened or
undermined entirely, even if they aren’t explicitly consequentialist
arguments. But it’s difficult even on non-consequentialist approaches,
partially because bracketing causal considerations often results in
arguments that overgeneralize. It’s often still possible to argue that it
would be good to abstain from animal products, but that’s simply a
different conclusion.
2. Even if we grant that animals have rights, it takes further argument
to show that those rights rule out any particular way of using or
relating to animals.17 Not all use is exploitative. In particular, we
need to grapple with the possibility that you can respect rights while
causing or supporting harm. Agrarians and their ilk may overstate
what follows from this, but the point remains.
3. Animals are harmed in conventional plant agriculture, and this
makes trouble for arguments that otherwise seem to have straightfor-
ward pro-vegan implications.18 This doesn’t support eating grass-fed
beef, pace Steven Davis. Still, if we aim to minimize harm—or avoid
supporting practices that cause extensive and unnecessary harm, or
respect the rights of every experiencing subject-of-a-life—then we
have to factor in the costs of eating plants.
4. Any case for consuming animal products shouldn’t focus exclusively
on traditional animal products—meat, dairy, eggs, and fish—or even
on animal husbandry. In addition, we need to consider “unusual”
animal-based foods: those derived from insects, mollusks, eggs from
backyard chickens, animals that die either accidentally or naturally,
animal products that will be discarded if they aren’t consumed, hunt-
ing, and so on. This means that even if it isn’t possible to defend
many traditional animal products, veganism doesn’t follow either de
facto or as an ideal. Moreover, unusual animal-based foods are unu-
sual because we often don’t eat them, not because they’re particularly
uncommon or difficult to access. So, while it’s certainly significant if
people shouldn’t purchase or consume conventional animal prod-
ucts, we should also recognize that eliminating them is compatible
with a diet that clearly isn’t veganism.
Given these points, I think that there are two things to say. First, given
the causal inefficacy problem, as well as the difficulty of linking the ethics
of production to the ethics of consumption, it’s unlikely that there is a
general obligation to be vegan. Rather than being the moral baseline, as
some have claimed, being vegan is just a good thing to do—among many
good things to do. And second, even if you are skeptical about causal
inefficacy and optimistic about there being some way of bridging the pro-
duction/consumption gap, we still shouldn’t conclude that we ought to be
vegans. Instead, we should think that we ought to eat unusually.
Recall Bruckner’s argument. If there’s an available alternative, we
shouldn’t support practices that cause extensive and unnecessary harm
to animals. By eating a strict vegan diet, we support practices that cause
extensive and unnecessary harm to animals—namely, plant agriculture—
when eating roadkill is an available alternative. So, we shouldn’t be strict
vegans.
Bruckner’s argument generalizes. As far as I can see, the best rea-
son not to eat insects and various bivalves is based on a precautionary
principle: even though the evidence suggests that they aren’t sentient,
they might be; since they might be, and the cost of being wrong would
be significant, we shouldn’t harm them unnecessarily. But we need to
weigh the odds of insect and bivalve sentience against the known costs
to animals involved in plant agriculture. Granted, it would be very bad
if we were wrong about insects and bivalves, and we then began raising
them for food. However, it would be equally bad if we were wrong
about plant sentience, and yet we rightly accept this risk: the evidence
for plant sentience is weak, and the considerations that tell against pos-
iting it are strong. Likewise for the creatures in question. The upshot is
that we should weigh the risk of making a mistake in the line-drawing
problem (i.e., excluding insects and bivalves when they ought to be
included) against the harms to those creatures that are clearly one side
of the divide (e.g., the rabbits and field mice that are harmed in crop
production). The aim isn’t to limit our moral concern, but to balance
moral caution against the moral imperative to respond to plant agricul-
ture’s costs.
The same points apply to animal products that will be discarded if they
aren’t consumed (your roommate’s leftover Kung Pao chicken, which is
low-hanging fruit for a freegan). A reasonable reservation about the lat-
ter concerns your ability (1) to limit yourself to eating such products just
when they really would be thrown away and (2) to influence others to
adopt more animal-friendly diets. These are questions to which there are
no general answers. Some people have the willpower to opt out based on
the provenance of the food; others don’t. Those who do may eat; those
who don’t probably shouldn’t (assuming causal efficacy). Likewise, some
people will be willing and able to explain to those nearby why they’re
consuming animal products, turning the occasion into an opportunity to
advocate for animals. They’re willing to communicate with words what
vegans signal by their abstinence (and surely there are some people who
are more likely to be influenced by those who make thoughtful excep-
tions than by those they perceive as rigidly adhering to a rule). Of course,
others either won’t be willing or won’t be able to have those conversa-
tions, and so should think twice before saving leftovers from the trash
(again, assuming causal efficacy).
Concerns about plant agriculture may also mean that some hunting
is permissible. Plainly, this is much more controversial than the other
exceptions to a general vegan norm. However, if animal dignity matters,
and if animals are going to be killed one way or another, I don’t think
it’s obvious that it’s better to preserve the illusion of nonviolence than
to take full responsibility for your role in animal death—especially if it’s
done in a way that treats animals as beings worthy of respect, rather than
as mere pests.
Quite apart from concerns about plant agriculture, I think we can
defend some animal husbandry. Minimally, this means raising “liber-
ated” animals. I once had a student whose family ran a chicken sanctuary
where they took in birds that Austinites no longer wanted. (Backyard
chicken farming isn’t all it’s cracked up to be.) The family fed and pro-
tected the chickens well, they allowed the birds to live out their natural
lives, and they ate some of the eggs that the hens laid. Crucially, this
family operates outside the meat industry: there are no concerns about
exploiting animals for meat, since they aren’t shortening the chickens’
lives to get access to their bodies. Moreover, there are no concerns about
where the male chicks went, as the family couldn’t have prevented their
deaths. It seems to me that if there are any examples of use without
exploitation, then this is one. And if it is, then it’s permissible. Finally,
I can’t see any reason why they shouldn’t sell those eggs to neighbors—
at least as long they’re able to resist any temptation to make welfare
compromises—since there’s nothing wrong with selling something that
it’s permissible to own (assuming that the selling doesn’t lead to other
harms). Obviously, it’s more controversial to claim that small-scale fam-
ily farming of the sort that Cuneo envisions is permissible. However,
I don’t think it’s easy to show that Cuneo’s view is mistaken without
taking on some controversial assumptions about the cognitive capacities
of animals, at least if that’s the way that we attempt to argue that death
robs them of significant goods.
So even if we set aside causal inefficacy concerns, as well as the diffi-
culty of bridging the production/consumption gap via some deontologi-
cal principle, there is a good case for the permissibility of eating many
animal products, and for the duty to eat some of them, albeit not most
of the ones you’ll find at your local restaurant or grocery store. These
include roadkill, bugs, bivalves, animal products that will be wasted, and
the bodies and byproducts of animals who are either hunted or who live
full, pleasant lives. That is, it seems we have good reason to eat unusu-
ally. What’s more, we can justify all this consumption without ignoring
the welfare and respect-based concerns that have long motivated those
who advocate for animals.
Nevertheless, I don’t expect anyone to accept this conclusion. It’s
implausible on its face. Surely people aren’t obligated to eat oysters, or
hunt, or start munching on crickets. We don’t have decisive moral rea-
sons to start picking up roadkill. Rather than showing that we have these
obligations, these arguments show that we’ve gone wrong somewhere. If
there weren’t independent reasons to worry about the production/con-
sumption gap, a different conclusion might be appropriate. But given
such reasons, the arguments just mentioned serve as further confirmation
that we should reject any simple link between the ethics of production
and the ethics of consumption. If we accept that link, then we have to
say unbelievable things, including that people are failing in their moral
duties if they eat plants instead of mealworms. The upshot is that the
arguments for eating unusually count doubly against an individual duty
to be vegan: first, they count against it directly if we accept that there is
a relatively simple link between the ethics of production and the ethics
of consumption, which can be forged via an expected utility calculation,
or some relatively familiar deontic principle; second, they count against
it indirectly, providing motivation for a modus tollens argument against
there being such a link at all.
To be clear, I don’t find it implausible that it might be morally good
to eat unusually, just as I don’t find it implausible—and indeed believe—
that it’s morally good to be vegan. I just don’t think that either is obliga-
tory. Harm and rights-violation footprint considerations just aren’t that
strong.
Nevertheless, I do think that some people ought to be vegans, albeit
not for the reasons that people usually cite. I explain what I mean in the
next chapter.
Notes
1. For others, see Abbate 2014 and Alvaro 2017.
2. Someone might say that the virtuous person just knows, at least in some
cases, when she would be a party to wrongdoing. If that’s right, then I think
she also just knows, at least in some cases, how important it is to avoid
being party to particular wrongdoing. If she didn’t, then it would be dif-
ficult for her to tell how to prioritize among all the options available to her.
And as I argue below, I doubt that animal-related concerns are always most
important.
3. Someone might say that we should tie our account of the virtues to collective
human flourishing. But now I want to hear much more about why we should
think that our collective human flourishing requires veganism. To the con-
trary, it seems to require meat production without the negative externalities
of intensive animal ag. And if someone says that we should tie our account
of the virtues to the flourishing of all sentient beings, then I have trouble
distinguishing the view from virtue consequentialism. There are, of course,
other forms of virtue ethics still, but whatever their implications for ordinary
consumption practices, I see no reason to think that they wouldn’t sanction
unusual eating.
4. Might someone argue that these animal product consumers aren’t virtuous,
but incontinent? Yes. But whether the virtue theorist can make this move
depends on (1) the strength of the reason that people take themselves to have
to abstain and (2) whether we think of the other reasons at stake as moral,
so that the person can still plausibly be described as aiming for the good. If
people take themselves to have a strong reason (relative to their other rea-
sons in that moment) and they partake, then they’re incontinent. And if their
other reasons are non-moral, then ignoring even a relatively weak moral
reason might make them incontinent. But if you think, as I do, that health
and collective eating and such can give you moral reasons to eat animal
products, then I don’t think the meat-eater is always, or perhaps even often,
incontinent.
5. Plenty think that right can be waived, which is relevant to the (otherwise
very different) justifications that tend to be offered for capital punishment
and physician-assisted suicide. And plenty are prepared to deny less basic
but still important rights, such as the right to define your gender identity. But
for the purpose of comparing people and animals, it’s the most basic rights
that are of interest. Someone might object that some police officers behave
as though people of color lack the most basic rights, and therefore it isn’t so
difficult to find such Americans. But I think that this is the wrong interpreta-
tion of police behavior. These police officers don’t think that persons of color
lack certain basic rights, nor do they behave as though they do. Rather, they
behave as though persons of color pose a much greater risk to their safety
than is supported by the evidence. These officers aren’t to blame for denying
rights, but for acting on stereotypes that have no basis in reality.
6. For this generous estimate of the number of vegetarians and vegans
(as well as evidence that meat consumption is up), see: www.vox.
com/2016/8/18/12248226/eat-less-meat-campaign-fail. This estimate is
based on a survey that asked how people self-identify, and more fine-grained
surveys often show that self-identifying vegetarians and vegans eat small
amounts of the animal products that you’d think they’d avoid. The Humane
Research Council’s 2014 survey, which is sensitive to this concern, estimates
that only two percent of the US population is either vegetarian or vegan (see
https://faunalytics.org/wp-content/uploads/2015/06/Faunalytics_Current-
Former-Vegetarians_Full-Report.pdf). For information about the dominance
of industrial agriculture, see the USDA’s 2012 Census of Agriculture: www.
agcensus.usda.gov/Publications/2012/Full_Report/Volume_1,_Chapter_1_
US/usv1.pdf.
7. However, it doesn’t follow that our actions on behalf of cute animals are
always in their interest. We’ve bred French bulldogs in a way that makes
them cuter, but as a result they suffer from respiratory problems: their tiny
snouts make it hard for them to breathe. What’s more, we may be hardwired
in ways that incline us to fear certain animals that posed threats to us in the
past. It turns out, for example, that young children are better at detecting
snakes than centipedes in pictures (Hoehl et al. 2017). When you combine
that sensitivity with lessons from elders that snakes can be dangerous, we
may be especially likely to develop snake-related phobias. When you com-
bine the absence of cuteness with our being wired to avoid certain threats,
those animals are particularly likely to be victimized.
8. For CIWF’s explanation of its choice to focus on chickens, see www.ani
malcharityevaluators.org/resources/conversations-and-interviews/conver
sation-leah-garces-executive-director-compassion-world-farming-usa/. For
information on the incredible number of fish killed each year—which, it
turns out, is very difficult to estimate—see www.fishcount.org.uk/published/
standard/fishcountfullrptSR.pdf and www.countinganimals.com/the-fish-
we-kill-to-feed-the-fish-we-eat/. For an excellent discussion of fish pain, see
Braithwaite (2010) and the many replies to Key (2016) in Animal Sentience.
Finally, I should note that there are groups, such as Anishinaabe Native
American tribes, that claim to see fish as members of their communities, and
so see themselves as having obligations toward fish. However, it takes signifi-
cant historical, environmental, and cultural forces to create those attitudes.
Plainly, those forces aren’t present, or aren’t strong enough, to create similar
attitudes in most coastal communities, and they appear not to be operative
in most non-coastal communities. So, though we can be sensitive to fish, the
deck remains stacked against them.
9. Not incidentally, this fits with a standard criticism leveled against animal
advocates: namely, “Why aren’t you worried about all the ways that people
are being harmed?”
10. Of course, it might also turn out that having less interest in learning about
the capacities of pigs is itself caused by eating pork. Insofar as any inclination
not to know is an epistemic vice, there may be a problem here even if the
effects could be explained away.
11. There are wrinkles here, but not ones that detract from the main point.
The key thing to observe is that I’m only making a claim about what’s epis-
temically bad in one respect—not overall. To assess whether an action is
epistemically bad overall, we need to know about its other epistemic con-
sequences. After all, you might choose to employ an epistemic strategy that
increases your odds of forming true beliefs at the cost of increasing your odds
of forming false ones, as James thought we should when it comes to matters
of ultimate concern. Someone might argue that even if eating meat increases
the odds of forming inaccurate beliefs, it isn’t epistemically bad overall, since
it makes available other truths that are closed off to those who abstain: for
instance, knowledge of certain flavors, or of the relative goodness of different
recipes, or whatever. However, most philosophers seem to be of the opinion
that we should be more concerned about avoiding false beliefs than about
gaining true ones, so I’m not sure how promising this line is going to be. And
that issue aside, it’s worth noting that even if eating meat is epistemically
good overall, it may still be morally bad overall, which invites us to consider
whether we want to say that securing epistemic goodness is more important
than avoiding moral badness.
12. Granted, the studies aren’t sufficiently fine-grained to establish, for example,
that a committed defender of animal rights would be affected epistemically
by consuming meat scavenged from a dumpster. More on this later.
13. The exceptions: Peter Carruthers (1992), Loren Lomasky (2013), and Hsiao
(2017). It’s worth noting that these philosophers establish their conclusions
based on fairly unpopular moral positions—austere forms of contractualism,
in the former cases, and a version of natural law theory, in the latter.
14. By contrast, only 19% of non-philosophers made the same assessment.

Thanks to Tyler John for bringing this study to my attention.
15. It’s important to note that you don’t need to have false beliefs about both
empirical and moral matters for the argument to go through; it just matters
that meat-eating is associated with a greater likelihood of inaccurate beliefs
about one or the other. Consider the meat-eating animal scientist, who has
perfectly accurate beliefs about the mental capacities of animals. Still, she
may well have grievously mistaken beliefs about the moral importance of
animals, and her meat-eating may well be responsible for that.
16. Someone might claim that not all errors are equally morally problematic,
and some may even be morally good. Suppose, for instance, that animals
aren’t as important as people, but are still much more important than they
are generally thought to be. And suppose that becoming a vegan increases the
odds that you’ll overvalue animals, believing them to be exactly as important
as people (as some philosophers in fact believe—e.g., Mark Bernstein). And
suppose that this leads you to perform all sorts of altruistic acts that you
otherwise wouldn’t, such as running an animal sanctuary. Someone might
think that this is a much more tolerable error than one that leads us to under-
value animals, and perhaps even a morally good one, but the premise I’ve
offered draws no distinction between them. However, I’m inclined to think
that the neutral version of the premise is the right version. After all, given
the assumption with which we began, overvaluing animals is tantamount to
undervaluing humans, and that seems like a fairly significant error. What’s
more, if animals don’t matter as much as humans, then it’s at least worth
considering whether someone who devotes his life to running an animal
sanctuary is misusing his resources, since those could otherwise go to beings
of much greater moral worth.
17. Of course, a right’s implications often seem to be straightforward: the right
to bodily integrity probably implies the right not to have your tail docked or
beak trimmed, and it probably follows that people have an obligation not to
place pigs and chickens in circumstances where tail-docking and debeaking
seem like good ideas. But consider a parallel: the right to bodily integrity
probably implies the right not to be spayed or neutered, and it probably fol-
lows that people have an obligation not to place animals in circumstances
where spaying and neutering seem like good ideas. PETA, for example, advo-
cates for both the right to bodily integrity and the importance of spaying and
neutering, so I presume that its leadership sees no tension there. If not, then
its leadership must concede that we may do some things to an animal that
aren’t directly in its interests. The interesting question concerns the limits of
that permission. For more on this, see Boonin 2003.
18. There are, of course, alternatives to conventional plant agriculture. The
standard one is veganic agriculture, which attempts not to harm any ani-
mals whatever. Unfortunately, it faces significant practical hurdles; see
Cerulli 2010. Moreover, even if these can be cleared, veganic agriculture is
so uncommon that those who want to source their food this way will prob-
ably have to become agriculturists themselves, thus dramatically increasing
the burden of going vegan.
9 Activist Ethics
Despite my doubts that most people are obligated to be vegans, I do think
that some people have that obligation. I think that people ought to be
vegan because they take very weak agent neutral reasons—such as the
outside chance of making a difference, or the thought that animal products
symbolize patriarchy—and strengthen them into obligation-generating
agent relative reasons. They do this in many ways. Among them, by
forming a particular practical identity, or joining a certain movement,
or opting into a certain sort of relationship with animals. My goal in
this chapter is to say something about how this happens, and then to say
something about why it might make moral sense to make the sorts of
commitments that generate obligations.
9.1 Relationships, Identities, and Participation

in Movements
Let’s begin with a detour through sexual ethics. It seems plain that some
relationships involve the presumption that you’ll respect certain norms.
If you’re in a monogamous romantic relationship, you act wrongly if you
have an affair. This isn’t because it’s inherently wrong to have sex with
someone other than the person to whom you are romantically attached,
or just because such behavior will upset (or otherwise harm) your part-
ner. Granted, the consequences for your partner matter. But even if your
infidelity went unnoticed, you would still be acting wrongly: part of what
it is to have a monogamous relationship is to submit yourself to a certain
moral norm. In having an affair, you fail to meet it.
Suppose we grant this. Still, it doesn’t follow that we should commit
ourselves to monogamous romantic relationships. There are two issues
here: the first concerns what’s wrong given a particular relational struc-
ture (e.g., being a partner in a monogamous romantic relationship); the
second concerns whether we have reason to select that relational struc-
ture from the many that are available. And if we decide that we don’t
have particularly good reasons to select it, we aren’t bound by its norms.
Activist Ethics 159
If, for example, we choose to have open relationships, it won’t be wrong
to have sex with people who aren’t our primary partners.
Relationships aren’t the only things that work this way. We find a simi-
lar distinction with respect to the practical identities that we take on. If
I decide to become a teacher, I might acquire a range of obligations that
wouldn’t apply to me otherwise, and that are distinct from any institu-
tional responsibilities I might have. For instance, it might be the case that
there are circumstances in which I really ought to provide feedback on
a draft of a paper, even though my university doesn’t insist that I do it,
and the student has no right to my comments.1 But of course, I have no
duty to become a teacher: the reasons I have for being a teacher, and the
reasons I have once a teacher, are distinct.
I can make the same point about participating in a movement. My wife
is quite politically active; I am not. I do have reasons to be much more
active than I am, but it’s hardly obvious that those reasons generate any
obligations to be much more active than I am. However, in virtue of her
backing a certain campaign—not in a formal role, but simply as a private
citizen who is now known around our city—she feels obligated to show
up at certain events, respond to various criticisms of candidates on social
media, and so on.
Importantly, this last example brings out something else that we should
note. The scope of my wife’s obligations seems to depend on specific
social expectations, which vary for individuals at different levels of politi-
cal activity. She is now expected to be present and vocal in ways that she
wasn’t previously; she isn’t expected to sell all she has and give the pro-
ceeds to the campaign. Moreover, the logic of her obligations isn’t always
straightforward. She certainly thinks that she ought to put up signs and
do some leafleting. She doesn’t think that she has an obligation to put a
bumper sticker on her car. Is there some principle that explains exactly
why she has some duties rather than others? Probably not. Instead, the
story is going to appeal to somewhat contingent traditions. History could
have gone such that people would have had rather different expectations
of campaign participants, but it didn’t. Still, the expectations are no less
real for that.
The same is true when it comes to relationships and practical identi-
ties. Focusing on the former, we can see this when we ask about the
precise location of the line between fidelity and infidelity in the context
of monogamous relationships. Plainly, people don’t simply expect their
partners to avoid having sexual intercourse with others; there are a lot
of lines that have been crossed well before that point. Although there
are reasons that can be given for considering any one of these lines to be
the bright one that divides permissible from impermissible behavior, we
won’t be able to tell the full story without appealing to the expectations
of others. If partners agree that flirting isn’t cheating, then that counts
160 Activist Ethics
against its being cheating. And if they don’t agree, then that counts in
favor of its being impermissible.
9.2 The Duty to Be Vegan

I propose that we think about a duty to be vegan in these terms. In my
view, there is no one way to end up with a duty to be vegan. Instead, it
can be grounded in a number of ways, but usually through one of the
mechanisms above. For instance, someone could opt into a relationship
with animals that’s structured by a principle like this:
You ought to treat an animal however you ought to treat a human

being with comparable cognitive capacities unless it would be bad
for that animal to be so treated.
Essentially, this principle says that you ought to treat animals as honor-
ary humans, and it immediately rules out all sorts of things that I’ve
argued are permissible. After all, if someone thinks that it’s wrong to eat
dead human beings, whatever their cognitive capacities, then the above
principle commits them to saying that it’s wrong to eat dead animals. (To
be clear: I’m not suggesting that we should endorse the principle I just
mentioned, and I take no stance on whether it’s wrong to eat all dead
human beings. I’m only pointing out that if someone opts into a relation-
ship with animals based on that principle, and has the relevant moral
belief about cannibalism, then consistency commits them to a certain
view about the consumption of animal bodies.)
Alternately, someone could come to think of herself as “an animal
person,” and come up with various rules for herself on that basis. It isn’t
at all clear what her duties would be just based on that description. But
plenty of people decide that they are “animal people,” and think, par-
tially on the basis of that self-understanding, that they shouldn’t eat veal,
or foie gras, or perhaps even support factory farms.
Finally, someone could decide to become an animal advocate, and then
would be bound by the expectations that are shared in that community.
For instance, questions about when it’s permissible to use products with
trace amounts of animal products in them, or which were produced using
some animal byproduct—such as non-vegan sugar, which is whitened
with bone char—won’t be answerable apart from appeal to the standards
to which animal advocates happen to hold one another, based on a com-
plex set of contingent cultural, historical, and economic factors.
So there’s a sense in which the duty to be vegan is volitional: it comes
from our commitments. But that’s perfectly compatible with the notion
that our commitments are motivated by various non-volitional con-
siderations. For instance, people have good reasons to become animal
advocates: they are responding to the immense suffering that animals
Activist Ethics 161
experience at human hands. My claim is just that the suffering alone
doesn’t ground their duty to be vegan, for reasons that I’ve laid out in
previous chapters. Instead, they decide to relate to animals in a particular
way, or to join the movement, or to rethink their own self-understanding.
When they do, they end up with genuine moral obligations.
To be clear, I’m not saying that you need to have a relationship with
animals in order to have any obligations to them. In a sense, that’s the
broadly relational approach to animal ethics that Cora Diamond (1978)
recommends. Her goal is to criticize the arguments that Peter Singer and
Tom Regan made famous. These arguments begin with the observation
that infants, those with severe cognitive disabilities, and the comatose
deserve moral consideration. Nevertheless, they lack the more sophisti-
cated capabilities that the rest of us enjoy. So, what explains their moral
importance must be some simpler trait, such as being sentient, or being
the subject of a life. The next step is to observe that animals also have
that trait, and so it can only be mere prejudice, that is, speciesism, that
leads us to deny them moral consideration.
Diamond rejects the idea that any non-moral trait can, in and of itself,
explain why an entity matters morally. Instead, she thinks that moral
importance depends on a prior disposition to see an entity as falling into
a category that is already moral. So, for example, to see an entity as
human is to see it as an entity about which it would be bizarre to ask,
“Why are you relieving its pain?” If we see an entity as human, we don’t
see her as a mere entity. Instead, we see her as her, as someone, and as
someone whose suffering provides a pro tanto reason to offer aid (even
if that reason can be overridden).2 If Diamond is right, then a remark-
able conclusion follows: Singer/Regan-style arguments are either self-
undermining or irrelevant. They’re self-undermining if they lead us to
abandon the thick categories that underwrite our moral responses, such
as human and animal and thing. If we make that move, then we’re left
with nothing but the search for consistency, which we can just as well
achieve by denying moral consideration to newborns and the senile.
They’re irrelevant if they work primarily because they involve relaying
the many awful things we do to animals, rather than because of the phil-
osophical framework on which they rely. In that case, the arguments
are implicitly invoking our way of seeing animals—namely, as beings
to which we shouldn’t be cruel—and not because they offer the right
account of when a thing deserves moral consideration.
Diamond’s view is far more radical than mine. On her view, non-moral
traits—like the capacity to feel pain—don’t themselves generate moral
reasons. I’m certainly not saying that. To the contrary, I take for granted
that the mainstream philosophical tradition is correct about there being
certain empirically discoverable capacities that ground moral status. And
so we have all sorts of obligations to animals quite independently of
what we think about them, or how we want to relate to them, or which
162 Activist Ethics
movements we join. You shouldn’t beat your dog because it hurts your
dog, not because you have a relationship with your dog that grants him
moral importance, or that strengthens weak agent neutral reasons.
However, the ethics of eating is not like the ethics of beating your
dog. You don’t have control over how animals are treated in intensive
animal agriculture, whereas you do have control over what happens to
your dog. You are deeply embedded in a society that makes you com-
plicit in innumerable, complex ways to the harms that animals suffer
on factory farms. You are not deeply embedded in a society that makes
you complicit in breaking your dog’s ribs. You could actually minimize
your harm footprint by not being a strict vegan (e.g., by eating mussels),
and yet it seems permissible to be a strict vegan. You would not mini-
mize your harm footprint by being cruel to your dog. And so on. In my
view, we need relationships—or decisions to join movements, or practical
identities—to oblige you never to put certain products in your mouth. We
don’t need them to ground all obligations whatever.
Someone might insist that you can be obligated to relate to animals in
the way that generates a duty to be vegan. I don’t know how someone
might argue for this conclusion, but my suspicion is that any argument
will be vulnerable to the kinds of problems that I’ve already discussed.
For instance, it would be surprising if our obligations to relate to ani-
mals weren’t in some way sensitive to our difference-making capacities.
Moreover, even if we ought to relate to animals in a way that’s shaped by
a recognition of their rights (if, in fact, they have rights), veganism won’t
be required: we can still eat (at the very least) eggs from backyard chick-
ens, bivalves, insects, and comparable crustaceans. After all, it would be
quite surprising if it were to turn out that we were obligated to relate to
animals in a way that led us to harm more of them. A position without
implication could rightly be condemned as fetishistic, valuing a certain
kind of personal purity over the good for animals.
In any case, the position that I’m describing makes it very easy to
understand why it’s so common for people to think that most people
have obligations that, in my view, they don’t have. When people argue
for veganism, they are often arguing, without realizing it, that people
should become animal people, that they should join the animal advo-
cacy movement, that they should change the way that they relate to
animals in a fundamental way. And if people do these things, then they
should indeed be vegans. But activist ethics—that is, the obligations
that bind you when, say, you join the animal advocacy movement—are
not binding on everyone. You can recognize that animals are morally
important, and even that they are wrongly killed, without becoming a
strict vegan. We need to separate activist ethics from our general ethic,
the one that applies to most people. The latter is more modest than the
former.
Activist Ethics 163
9.3 Reasons to Opt In
Let’s suppose, then, that our pre-commitment reasons to be vegan aren’t
strong enough to generate an obligation to be vegan. We then have to
ask: why make the kind of commitment that might generate such an
obligation?
Some people won’t feel the weight of this question. They’ll be willing
to opt into a duty-generating relationship with animals immediately upon
recognizing their moral importance, and they’ll be confused about why
someone might resist. That may well be an admirable response. However,
it’s worth appreciating that it isn’t the only reasonable response. Suppose,
for instance, that the Peter Singer of “Famine, Affluence, and Morality”
is mistaken. He thinks that, minimally, we ought to donate to aid the des-
perately poor whenever it wouldn’t require sacrificing anything of moral
importance. To the contrary, suppose that while it would be very good
to do this, we aren’t actually obligated to do it. This, I take it, is how
most people understand our moral situation. And we should note that
there are, of course, some people willing to make commitments to the
desperately poor that generate obligations to them. Again, that may well
be an admirable response. But many of us need some additional reasons
to make this sacrifice. Presumably, this isn’t be because we secretly don’t
believe that the desperately poor have any moral importance. Instead, it’s
because we have our lives configured such that it isn’t a trivial thing to
give away significant amounts of money.
In any case, let’s suppose that we do want a story about why we might
make a certain obligation-generating commitment regarding animals,
one that demands various sacrifices that probably won’t make a differ-
ence.3 To my mind, perhaps the best reason to do this is because it’s
good to signal to others that it’s worth being part of a coalition that can
address aggregate harms—in this case, the plight of animals in the food
system. And in our present cultural context, it’s plausible that dietary
change is one of the more reliable ways to communicate that animals are
worth some sacrifice: people see what you eat, and they’ve got plenty of
opportunities to see what you eat, since you eat regularly. (You could, of
course, advertise that you make regular donations to Mercy For Animals,
but that’s socially unacceptable outside the world of effective altruism.)
Granted, your rationale for your diet can’t be read off your behavior. It
isn’t obvious, simply from the fact that you order the black bean burger
instead of the beef, that you’re motivated by the plight of animals as
opposed to, say, environmental considerations. Note, however, that this
ambiguity isn’t always troubling. If you are avoiding meat for the sake
of animals, but someone thinks that you’re doing it for the environment,
you’re still signaling that you’re willing to make a sacrifice to form a coa-
lition that can address important aggregate harms.
164 Activist Ethics
The ambiguity might be troubling, though, in cases where someone
thinks your diet is simply driven by health considerations. The right
response here, I think, is to provide an independent reason to think that
dietary change is the right kind of costly behavior, and to argue that this
consideration outweighs any problems created by ambiguity in the sig-
nal. And I think we can provide such a reason. In short, moral judgments
often involve assessments of what it’s reasonable to demand of others, of
those sacrifices that are worthwhile, of how various goods should be bal-
anced. By sacrificing in a way that’s obviously tied to the kind of change
you hope to bring about, you are, essentially, responding to some poten-
tial objections to your call for collective action.
After all, as discussed much earlier, moral norms impose costs on those
who adhere to them, and our sense of when those costs are reasonable
is influenced by a host of personal and contextual factors. Instead of
being able to take what you’d like from the grocery store, following the
Don’t steal norm means that you have to live on what you can afford—
which might not be much. If you’re wealthy, then the costs of following
the Don’t steal norm are low; if you’re poor, they’re much higher. So,
by sacrificing in this particular way, you signal that you understand the
relevant cost that you’re asking others to make in endorsing the moral
norm; at the same time, your example provides some evidence that it isn’t
unreasonable to ask others to internalize the relevant cost too. In other
words, internalizing a cost that’s appropriately connected to the change
you hope to bring about signals that it isn’t merely inexperience, or privi-
lege, or some other factor that explains why you regard this cost to be
an acceptable one.4 And given the importance of addressing this kind of
challenge to a potential signal, I think it’s worth accepting that the signal
needn’t communicate in a perfectly reliable way.
We might worry that, over time, the value of our signaling falls off. The
longer you go without eating animal products, the easier it gets. For many
people, they stop being appealing. And even if they remain appealing,
you become much better at finding substitutes to satisfy your cravings. So
although there might be notable costs to transitioning to a vegan diet, there
are fewer and fewer costs as time goes on, and perhaps that undermines the
signal. Just as we might doubt that the wealthy person really signals that
he understands the sacrifice he’s asking others to make when he follows the
Don’t steal norm, we might doubt that vegans can effectively signal that
they understand the sacrifice they’re asking meat-lovers to make.5
However, it’s a contingent fact that meat-lovers don’t seem to interpret
the veganism of vegans as costless. They still say things like, “I just don’t
know how you do it,” and “Don’t you miss [fill in the blank]?” These
sorts of responses suggest that they struggle to see veganism as anything
but costly, and in this context, that’s a mark in favor of veganism.
Julia Nefsky (2018) points out some serious problems for this way of
defending an obligation to signal. First, most of our behavior probably
Activist Ethics 165
doesn’t influence other people: thankfully, no one notices what we do.
Second, even when it does influence other people, it isn’t clear that it
influences them enough to ground an obligation: if your behavior doesn’t
actually make a difference to whether the coalition forms, it’s unlikely
that the expected utility of signaling is positive. Third, although “butter-
fly effects” are possible, they aren’t probable: so, again, it’s unlikely that
the expected utility of signaling is positive.
What’s more, even if it’s valuable to signal in a costly way, that’s a far
cry from saying that the value of signaling trumps all other moral consid-
erations, such as minimizing harm to animals—assuming that consumer
choices make a difference, and you could minimize your harm footprint
by, say, eating some leftover Kung Pao chicken. Moreover, nothing I’ve
said supports the view that signaling can only take one particular form.
So if you eat the leftover Kung Pao chicken, and you explain to your
roommate that you feel uncomfortable about eating meat due to the way
animals are treated, but want to avoid food waste, you’ve signaled. I can
imagine similar kinds of conversations such that signaling-based consid-
erations favor eating mostly vegetarian, with some flexibility built in for
health considerations, financial burdens, or certain goods associated with
significant cultural traditions, family and/or social obligations, and the
like. So, for instance, if you tell your family that you are only eating
turkey on Thanksgiving for their sake, but generally think we shouldn’t
do it, you may be acting in accord with any signaling-based reasons you
have simply in virtue of pursuing that awkward conversation: the costly
signal is just your willingness to make that point to your loved ones.6
However, we aren’t looking for obligation-generating reasons here: we
are looking for reasons that would make it rational—and ideally, mor-
ally good—to make a certain commitment that would itself generate the
obligation. It does seem to be both rational and morally good to make
such a commitment, insofar as the project itself is morally good. Surely it
is: billions of animals suffer each year, and it’s good to try to prevent that.
9.4 Conclusion
The upshot: the good of signaling isn’t decisive in favor of a commitment
that generates an obligation to be vegan, but it can still motivate one.
And, of course, many people do make this sort of commitment, and so
have a duty to be vegan.
Notes
1. As Elizabeth Harman (2016, 369) says: “My claim about this case is that it
could be a true story: that the normative claims that are part of the case could
be true of a case with these non-normative features. I don’t want to claim
about this case or the other cases in this essay that the non-normative features
166 Activist Ethics
of the case that I lay out make it clear that the normative claims hold in the
case. Rather, I just need the weaker claim about this case and my other cases
that the normative claims could be true in some case like this. Thus, my use of
described cases in this essay is a bit different from the common use in which
a philosopher describes a case and then claims that a certain normative claim
must be true in the case as described.”
2. As she puts it: “We cannot point and say, ‘This thing (whatever concepts it
may fall under) is at any rate capable of suffering, so we ought not to make
it suffer’. . . . That ‘this’ is a being which I ought not to make suffer, or whose
suffering I should try to prevent, constitutes a special relationship to it”
(1978, 470).
3. Some people will object to this question because they don’t think that it’s
costly to give up eating animals. It seems obvious to me that they’re wrong:
animal products taste good, they can be good for you, they fit into traditions
that many people value, and eating them simplifies your life socially and prac-
tically. Perhaps these costs are usually outweighed by the benefits of veganism,
so that veganism isn’t costly all things considered. Again, if we’re assessing
things prudentially, it seems obvious to me that this is wrong: I, for one, derive
very little satisfaction from taking the high road. But there’s no hope of set-
tling the issue here.
4. Holly Lawford-Smith (2015) endorses a similar view. As she points out, many
of the most significant problems we face are collective action problems, which
can’t be addressed unless we form the appropriate coalitions. However, if we
want to form those kinds of coalitions, we need to send signals to the effect
that we’re willing to cooperate with others to achieve the relevant good. For
such signals to be believable, though, they need to be somewhat costly. As
Lawford-Smith puts it, a “public signal of conditional willingness [to take on
a non-trivial cost] by [an individual] ensures that she is not a reason for others
to think that cooperation is impossible, and it simultaneously places norma-
tive pressure on others to signal in a similar way, which raises the chance of
a group capable of acting in pursuit of the relevant end being formed” (2015,
325–326).
5. Thanks to Beth Barker for this objection.
6. There may be limits here, as some signaling strategies are self-undermining.
Conscientious omnivorism faces an uphill battle, even if only for contingent
psychological and social reasons: I’ve never met a conscientious omnivore
who will turn down hotdogs at a barbecue, suggesting that it’s quite difficult
for folks to maintain a strict dietary standard in circumstances where there
would be clear signaling value to refraining. You might think that reducetari-
anism is problematic, and perhaps even more so, for the same sort of reasons:
it doesn’t draw any hard lines between kinds of products, but only asks you
to eat fewer of the objectionable ones, so it’s even harder for observers to pick
up what’s going on. Of course, we can imagine Mr. Talker, who’s willing to
have a lot of conversations about the plight of animals while eating them.
But recall the point I made earlier about internalizing relevant costs: at some
point, those conversations will probably seem hypocritical, as people will rea-
sonably wonder whether Mr. Talker’s views about animals are enabled by his
not having to do much of anything for them. (Granted, though, if Mr. Talker
convinces people that that isn’t the case—as, for instance, is arguably true of
a certain animal advocate I know who regularly eats pork with pig farmers,
as he thinks that he won’t be able to get them to improve the living conditions
for hogs unless they see him as an insider—then there is no problem.)
10 Taking Stock
In a sense, this book is one long argument for a conclusion about what
most people may do. Plainly, though, this doesn’t imply that most mem-
bers of the general public are moral exemplars with respect to animals.
Many people think that assigning a function to another sentient being—
“But that’s what we raise them for!”—is all it takes to justify harming that
being. They’re wrong. Many people make far too much of species mem-
bership, which looks like motivated reasoning. Many people wouldn’t
abstain from consuming animal products even if they thought that doing
so would make a difference; they simply don’t care about animals, which
is a moral failure indeed. Nothing I’ve said in this book should be taken
as an apology for the beliefs and attitudes of the members of the general
public. On my view, then, we’ve misdiagnosed their failings. Their eating
isn’t the problem: their beliefs and values are.
In any case, given that I don’t think most people are obligated to be
vegans, you might wonder what I think they are obligated to do. The
short answer: I don’t know, but probably not much specifically with
respect to their food choices.
On the “I don’t know” side, figuring things out would be easier if
we knew whether we have any collective obligations regarding intensive
animal agriculture—and if so, what they are. If we do have some, and
if we know what they are, then the collective obligation(s) could at least
be a guide to our individual obligations, even if it (they) didn’t strictly
determine them. But I’m not sure whether “we” have any collective obli-
gations regarding intensive animal agriculture. While it’s easy to say that
we consumers or animal product purchasers (or whatever) have a certain
collective obligation, we aren’t organized in anything like the way that a
business or government is, which I take to be the normal bearers of such
responsibilities. Moreover, I’m not at all sure what the collective obliga-
tion would be. As I said at the outset, I’m not someone who thinks that
animals don’t matter morally; I don’t insist that the satisfaction of human
interests is dramatically more important than the satisfaction of animal
interests. And I readily acknowledge that billions of animals are indeed
treated badly in intensive agriculture. But while it might be clear that we
168 Taking Stock
collectively ought not to preserve the status quo, that negative statement
doesn’t tell us what our positive duty is. Is utilitarianism true, and is
the Replaceability Argument sound? If so, then we should be massively
improving the welfare of farmed animals, but not abandoning animal
agriculture, as Torbjörn Tannsjö (2017) argues. In that case, we might
need to consider something like Jayson Lusk’s (2011) proposal to create a
market in animal welfare, which would allow people to buy units of farm
animal well-being, thereby providing producers with a financial incentive
to treat animals well. Is some version of the rights view true, and yet it’s
also the case that the animals we eat don’t have a strong interest in con-
tinued existence? In that case, we should be dismantling intensive animal
agriculture entirely, while—if Cuneo is right—leaving small family farms
intact. Or have I made significant errors, and some moral theory is true
that implies that it’s always and everywhere wrong to kill animals for
food? In that case, we should be dismantling animal agriculture in all its
forms—including bivalve and insect production. To my mind, there are
reasonable things to say on behalf of each of these views, and I often feel
like I could be convinced of any of them. The point here, though, is that
this uncertainty makes it hard for me to be at all confident about what
we ought to be working toward, and so hard to be at all confident about
what individuals should be doing—if anything at all.
On the “Not much” side of things, let me sum up. There are prob-
lems with standard attempts to bridge the production/consumption gap.
And if we could bridge it, then we would probably be required to eat
strange things, at least by the standards of many people in the Western
world: bivalves and bugs, roadkill and refuse, perhaps some big game.
If bridging the gap didn’t lead to this conclusion, then it would be more
reasonable to say: we just need to keep working on strategies for deriv-
ing conclusions about the ethics of consumption from the ethics of pro-
duction. But since it does, and it seems implausible that we’re obligated
to eat unusually, we should accept that we probably aren’t obligated to
eat in certain ways simply because foods were produced wrongly. More
bluntly, with less nuance: we aren’t obligated to be vegan.
Truth be told, if there weren’t such a large body of literature defending
the opposite conclusion, this wouldn’t be a very interesting position to
take. After all, it’s compatible with the view that it’s morally subpar to
eat animals, which seems quite plausible given the arguments that I’ve
discussed. And if it’s subpar, then we can say about eating animals what
we say about many other actions: we probably don’t have specific obliga-
tions to act in some particular way, and yet it’s pretty clear that decent
people will often act in that way, or at least make some effort in the rel-
evant direction. I think this is the norm in consumer ethics cases, but also
in plenty of others: giving to various weighty causes, doing community
service, voting in national elections, being an organ donor, and so on. In
these cases, it isn’t morally admirable, and it’s often morally subpar, not
Taking Stock 169
to give or serve or vote or donate. Likewise, it isn’t morally admirable,
and it’s often morally subpar, not to adjust your diet for the sake of
animals—which is perfectly compatible with it being permissible not to
adjust your diet. Granted, some will want more from their morality; they
may want to deny that there can be cases where the subpar is permissible.
That doesn’t fit well with my own moral methodology. As I said at the
outset, I expect a certain degree of consensus across moral frameworks
before I think it’s appropriate to judge that an action is morally wrong.
When we lack that, we should fall back to a weaker position, such as the
one that I’ve developed here.
Haven’t I shown that there is a consensus among moral frameworks—
namely, that veganism isn’t generally required? No. What I’ve shown, if
anything, is that many standard arguments for veganism aren’t successful,
and so it’s plausible that there is no general obligation to be vegan. Surely
there is some version of virtue ethics that implies that we’re required
to be vegans—or some version of ecofeminism, or consequentialism, or
what have you. So disagreement remains. But as I said at the outset, when
most of our moral frameworks point toward a particular duty, that’s a
good reason to think we’ve got the duty in question. When most don’t,
and only a handful do, that strikes me as evidence that we’re in the realm
of how to do better relative to some important values, rather than what
we’re obligated to do.
Some people don’t value consensus among frameworks as much as
I do. If you’re willing to endorse one of the frameworks that implies that
there’s a general duty to be vegan, then you can have the relevant con-
clusion. But I’m not, so I can’t. Whatever we’re doing when we theorize
about our obligations—across our many disagreements—I don’t think
that it can be something so partisan.
Over the last several years, I’ve spent a lot of time trying to figure out
how to get obligations not to purchase or consume animal-based foods
out of the wrongness of producing them. Here’s one good explanation
for the difficulty of the task: most people don’t have such obligations,
and I was led to think they did by the force of an intuition—namely, that
if something is wrong, then you ought to disassociate from it. But in the
broken world we’ve inherited, disassociation is impossible. So, we should
abandon the intuition, taking more modest positions on the ethics of
purchasing and consuming. Thankfully, we can do this without chang-
ing much about how we try to convince others to eat differently. People
still have reasons to be vegan, or something near enough. And it’s good
to share those reasons. But instead of concluding with, “Therefore you
ought,” we say, “Come and join us.”
And now at the end, let’s ignore the question of our duties: let’s think
about eating better, even if it isn’t required. If we decide to take respon-
sibility for the harms of animal agriculture—not because an argument
says we must, but because we’re moved by the plight of animals—where
170 Taking Stock
might we begin? If our goal were minimizing our suffering footprint, we
might start with chicken, given how much broilers suffer relative to the
amount of meat we extract from them; if our goal were minimizing our
death footprint, we might start with fish, given the enormous problem
of bycatch. If, by contrast, we wanted to begin by avoiding products
from the animals who suffer most overall, we’d eliminate eggs: although
there is more suffering per serving of chicken, each layer hen suffers more
in total, and arguably more than all other animals in intensive systems.
Finally, if we wanted to start where animal and environmental concerns
most clearly overlap, we might start with beef, due to its carbon footprint.
But while these footprint considerations are important, they aren’t
decisive. The reason is familiar: your diet probably won’t affect how
things go for animals. However, your diet does affect how omnivores
understand you, as well as whether vegans grant you entry to the animal
movement. Moreover, some dietary changes are easier to make, given the
world in which we find ourselves. So, my vote is to balance footprint-,
signaling-, and implementation-based reasons.
I’d begin, then, by cutting out pork, being careful to replace it with
plant-based sources rather than other animal products. There are num-
ber of reasons for this, a chief one being that sows in gestation crates
suffer so much. But it’s also because pigs are widely known to be particu-
larly smart mammals, which makes them particularly relatable, as well
as because it’s an easy ask for health reasons. (Bacon isn’t winning any
awards from dietitians.) It’s good to start in a spot where the self- and
other-directed reasons line up. If you’re looking for a next step, I’d elimi-
nate chicken. Layer hens probably suffer more than most broilers, as they
live relatively long lives (roughly 18 months) in tight confinement, during
which they face well-known welfare issues (respiratory problems due to
poor air quality, trimmed beaks, prolapse, etc.). However, broiler welfare
is reliably low (largely due to breeding for rapid growth and overcrowd-
ing), it takes many chickens to provide as much meat as you would get
from a single pig or head of cattle, and it’s simply harder to give up eggs,
as they’re in so many products. Behavior change is easier when there are
successes to celebrate along the way, and while minimizing meat con-
sumption may not be a principled moral position, it’s a recognizable pro-
ject that can be achieved. For that very reason, I’d encourage you to give
up beef next, thereby becoming a pescatarian.
If you’re open to doing more, I’d give up fish next. That gets you to
something like vegetarianism—another recognizable identity. And then,
due to the severity of the harms in egg production, set aside eggs, fol-
lowed by dairy. Throughout the process, though, there’s no need to give
up bivalves. Two servings of mussels per week will help cover your B12
and vitamin D needs, as well as provide a good source of protein. This
will spare you from having to supplement, which isn’t terribly expensive,
but I’ve always found to be an inconvenience.
Taking Stock 171
I wouldn’t rush any of these changes. Take some time per kind of ani-
mal product, giving yourself a bit of time to adjust to new flavors, reci-
pes, and approaches to meal planning. Retraining yourself in the kitchen
takes time, and there’s no sense making the process harder than it needs
to be. And on that note, the transition will be much easier if you find
friends and family members with whom to figure things out. Most veg-
etarians and vegans don’t stay vegetarian or vegan largely because of a
lack of social support. So if you can help it, don’t go it alone. In any case,
if you follow a plan like this, you’ll be all-but-vegan within a short while.
At this juncture, you’ll have to assess how things are going. I haven’t
drawn any distinctions between conventional animal products and those
that come from more animal-friendly operations. I haven’t distinguished
between farm-raised and wild caught fish. I haven’t encouraged you to
explore all the unusual ways of eating that I take to be permissible, even
if they aren’t morally required. This is largely because I think it’s worth
trying to find out whether you can manage easily and happily enough
without food from these sources. If you can, you’ll probably find it easier
to spend time with other vegetarians and vegans. Those relationships
matter because they can help you stick to your commitment, but also
because they may, ultimately, help form the kind of coalition that could
actually make a difference. Individual signaling may be good, as I argued
in the last chapter, but collective signaling is far better.
However, if things aren’t going well, don’t throw in the towel. The
choice isn’t between strict veganism and indiscriminate omnivorism:
there are plenty of intermediate options. So grab some cricket powder.
Care for some chickens who would otherwise be neglected. Nab those
leftovers. And if that isn’t enough, then introduce what you need to feel
okay, relying on unusual and high-welfare sources whenever possible. In
a food system like ours, you don’t need to be a martyr. You also don’t
need to abandon any sensitivity to what’s better and worse. No, you
probably aren’t morally obligated to be a strict vegan. But you can do
better.
References
Abbate, Cheryl. 2014. Virtue Ethics and Animals: A Minimally Decent Ethic for
Practical Living in a Non-ideal World. Journal of Agricultural and Environ-
mental Ethics 27: 909–929.
Adamo, Shelley Anne. 2016. “Do insects feel pain? A question at the intersection
of animal behaviour, philosophy and robotics.” Animal Behaviour 118: 75–79.
Adams, Carol J. 2010a. The Sexual Politics of Meat: A Feminist-Vegetarian Criti-
cal Theory, 20th anniversary edition. London and New York: Continuum.
Adams, Carol. 2010b. Why Feminist-Vegan Now? Feminism & Psychology 20:
302–317.
Al Homidan, A. and J. F. Robertson. 2003. Effect of Litter Type and Stocking
Density on Ammonia, Dust Concentrations and Broiler Performance. British
Poultry Science 44: S7–S8.
Allcorn, Ashley and Shirley Matile Ogletree. Forthcoming. Linked Oppression:
Connecting Animal and Gendered Attitudes. Feminism & Psychology.
Allen, Michael W., Mathew Hunstone, Jon Waerstad, Emma Foy, Thea Hobbins,
Britt Wikner, and Joanne Wirrel. 2002. Human-to-Animal Similarity and Par-
ticipant Mood Influence Punishment Recommendations for Animal Abusers.
Society and Animals 10: 267–284.
Alvaro, Carlo. 2017. Ethical Veganism, Virtue, and Greatness of the Soul. Jour-
nal of Agricultural and Environmental Ethics 30 (6): 765–781.
American Dietetic Association. 2009. Position of the American Dietetic Associa-
tion: Vegetarian Diets. Journal of the American Dietetic Association 1266–1282.
American Heart Association. 2016. Vegetarian Diets. www.heart.org/
HEARTORG/HealthyLiving/HealthyEating/Vegetarian-Diets_UCM_306032_
Article.jsp#.WhRUgFWnGHt.
Balderrama, N., H. Díaz, A. Sequeda, J. Núnez, and H. Maldonado. 1987.
Behavioral and Pharmacological Analysis of the Stinging Response in African-
ized and Italian Bees. In Neurobiology and Behavior of Honeybees, edited by
R. Menzel and A. Mercer, 121–128. New York: Springer-Verlag.
Banhazi, T. M., J. Seedorf, M. Laffrique and D. L. Rutley. 2008. Identification of
the Risk Factors for High Airborne Particle Concentrations in Broiler Buildings
Using Statistical Modelling. Biosystems Engineering 101(1): 100–110.
Barnhill, Anne and Tyler Doggett. Forthcoming. Food Ethics II: Consumption
and Obesity. Philosophy Compass.
Barratt, B. I. P., C. M. Ferguson, D. M. Barton, and P. D. Johnstone. 2009. Impact
of Fire on Tussock Grassland Invertebrate Populations. Science for Conserva-
tion 291: 5–33.
References 173
Barron, Andrew and Colin Klein. What Insects Can Tell Us About the Origins of
Consciousness. Proceedings of the National Academy of Science of the United
States of America 113 (2016): 4900–4908.
Bastian, B., S. Loughnan, N. Haslam and H. Radke. 2012. Don’t Mind Meat?
The Denial of Mind to Animals Used for Human Consumption. Personality
and Social Psychology Bulletin 38: 247–256.
Bateson, M., S. Desire, S. E. Gartside, and G. A. Wright. 2011. Agitated Hon-
eybees Exhibit Pessimistic Cognitive Biases. Current Biology 21: 1070–1073.
Belshaw, Christopher. 2015. Meat. In The Moral Complexities of Eating Meat,
edited by Ben Bramble and Bob Fischer, 9–29. New York: Oxford University
Press.
Bessei, W. 2006. Welfare of Broilers: A Review. World’s Poultry Science Journal
62: 455–466.
Bilewicz, M., R. Imhoff and M. Drogosz. 2011. The Humanity of What We Eat:
Conceptions of Human Uniqueness Among Vegetarians and Omnivores. Euro-
pean Journal of Social Psychology 41: 201–209.
Blackwell, T. E. 2004. Production practices and well-being: Swine. In The Well-
Being of Farm Animals: Challenges and Solutions, edited by G. J. Benson and
B. E. Rollin, 262. Ames, Iowa: Blackwell Publishing.
Boersma, S. 2001. Managing Rapid Growth Rate in Broilers. World Poultry 17
(8): 20–21.
Boland, H. T., G. Scaglia, W. S. Swecker and N. C. Burke. 2008. Effects of Alter-
nate Weaning Methods on Behavior, Blood Metabolites, and Performance of
Beef Calves. Professional Animal Scientist 24: 539–551.
Bonilla-Silva, Eduardo. 2015. The Structure of Racism in Color-Blind, ‘Post-
Racial’ America. American Behavioral Scientist 59 (11): 1358–1376.
Boone and Crockett Club. 2016. The Boone and Crockett Club on Fair Chase.
www.boone-crockett.org/pdf/On_Fair_Chase.pdf.
Boonin, David. 2003. Robbing PETA to Spay Paul: Do Animal Rights Include
Reproductive Rights? Between the Species 13: 1–8.
Borderas, T. F., B. Pawluczuk, A. M. De Passillé and J. Rushen. 2004. Claw Hard-
ness of Dairy Cows: Relationship to Water Content and Claw Lesions. Journal
of Dairy Science 87: 2085–2093.
Bouvard, Véronique, Dana Loomis, Kathryn Z. Guyton, Yann Grosse, Fatiha
El Ghissassi, Lamia Benbrahim-Tallaa, Neela Guha, Heidi Mattock and Kurt
Straif. 2015. Carcinogenicity of Consumption of Red and Processed Meat. The
Lancet Oncology 16 (16): 1599–1600.
Bradshaw, R. H. 1998. Consciousness in Non-Human Animals: Adopting the
Precautionary Principle. Journal of Consciousness Studies 5 (1): 108–114.
Bradshaw, R. H., R. D. Kirkden and Broom, D. M. 2002. A Review of the Aetiol-
ogy and Pathology of Leg Weakness in Broilers in Relation to Welfare. Avian
and Poultry Biology Reviews 13 (2): 45–103.
Bradshaw, R. H., R. F. Parrott, J. A. Goode, D. M. Lloyd, R. G. Rodway and
D. M. Broom. 1996. Behavioural and Hormonal Responses of Pigs During
Transport: Effect of Mixing and Duration of Journey. Animal Science 62:
547–554.
Braithwaite, Victoria. 2010. Do Fish Feel Pain? New York: Oxford University
Press.
Bratanova, B., S. Loughnan and B. Bastian. 2011. The Effect of Categorization
as Food on the Perceived Moral Standing of Animals. Appetite 57: 193–196.
174 References
Brown, Culum. 2015. Fish Intelligence, Sentience and Ethics. Animal Cognition
18 (1): 1–17.
Brown-Brandl, T. M., R. A. Eigenberg, G. L. Hahn, J. A. Nienaber, T. L. Mader, D.
E. Spiers and A. M. Parkhurst. 2005. Analyses of Thermoregulatory Responses
of Feeder Cattle Exposed to Simulated Heat Waves. International Journal of
Biometeorology 49: 285–296.
Bruckner, Donald. 2015. Strict Vegetarianism Is Immoral. In The Moral Com-
plexities of Eating Meat, edited by Ben Bramble and Bob Fischer, 30–47. New
York: Oxford University Press.
Budolfson, Mark Bryant. 2015a. Consumer Ethics, Harm Footprints, and the
Empirical Dimensions of Food Choices. In Philosophy Comes to Dinner: Argu-
ments Over the Ethics of Eating, edited by Matthew C. Halteman, Terence
Cuneo and Andrew Chignell, 163–181. New York: Routledge.
———. 2015b. Is It Wrong to Eat Meat from Factory Farms? In The Moral Com-
plexities of Eating Meat, edited by Ben Bramble and Fischer, 80–98. New York:
Oxford University Press.
———. 2019. The Inefficacy Objection to Consequentialism and the Problem
with the Expected Consequences Response. Philosophical Studies 176 (7):
1711–1724.
Buijs, S., L. J. Keeling, C. Vangestel, J. Baert, J. Vangeyte and F. A. M. Tuyttens.
2010. Resting or Hiding? Why Broiler Chickens Stay Near Walls and How
Density Affects This. Applied Animal Behaviour Science 124 (3–4): 97–103.
Buijs, S., E. Van Poucke, S. Van Dongen, L. Lens, J. Baert and F. A. M. Tuyttens.
2012. The Influence of Stocking Density on Broiler Chicken Bone Quality and
Fluctuating Asymmetry. Poultry Science 91: 1759–1767.
Buzby, Jean C., Bruce Axtman, and Jana Mickey. 2009. Supermarket Loss Esti-
mates for Fresh Fruit, Vegetables, Meat, Poultry, and Seafood and Their Use in
the ERS Loss-Adjusted Food Availability Data. Economic Information Bulletin
No. 44. www.ers.usda.gov/publications/pub-details/?pubid=44309.
Buzby, Jean C., Jeanine T. Bentley, Beth Padera, Jennifer Campuzano, and Cara
Ammon. 2016. Updated Supermarket Shrink Estimates for Fresh Foods and
Their Implications for ERS Loss-Adjusted Food Availability Data. EIB-155,
U.S. Department of Agriculture, Economic Research Service.
Callicott, J. Baird. 2015. The Environmental Omnivore’s Dilemma. In The Moral
Complexities of Eating Meat, edited by Ben Bramble and Bob Fischer, 48–64.
New York: Oxford University Press.
Caramazza, Alfonso and Bradford Z. Mahon. 2006. The Organisation of Con-
ceptual Knowledge in the Brain: The Future’s Past and Some Future Directions.
Cognitive Neuropsychology 23 (1): 13–38.
Caramazza, Alfonso and J. R. Shelton. 1998. Domain-Specific Knowledge Sys-
tem in the Brain: The Animate—Inanimate Distinction. Journal of Cognitive
Neuroscience 10: 1–34.
Carruthers, Peter. 1992. The Animals Issue. New York: Cambridge University
Press.
———. 2005. Consciousness: Essays from a Higher-Order Perspective. New
York: Oxford University Press.
———. 2011. Animal Mentality: Its Character, Extent, and Moral Significance.
In The Oxford Handbook of Animal Ethics, edited by Tom Beauchamp and R.
G. Frey, 373–406. New York: Oxford University Press.
References 175
CDC. 2017. Health of Black or African American non-Hispanic Population.
www.cdc.gov/nchs/fastats/black-health.htm.
Cerulli, Tovar. 2012. The Mindful Carnivore: A Vegetarian’s Hunt for Suste-
nance. New York: Pegasus Books.
Chappell, Timothy. 2011. On the Very Idea of Criteria for Personhood. Southern
Journal of Philosophy 49: 1–27.
Chignell, Andrew. 2016. Can We Really Vote with our Forks? Opportunism and
the Threshold Chicken. In Philosophy Comes to Dinner, edited by A. Chignell,
T. Cuneo, and M. Halteman, 182–202. New York: Routledge.
Ciocchetti, Christopher. 2012. Veganism and Living Well. Journal of Agricultural
and Environmental Ethics 25: 405–417.
Cochrane, Alastair. 2012. Animal Rights Without Liberation: Applied Ethics and
Human Obligations. New York: Columbia University Press.
Cordeiro-Rodrigues, Luis. 2017. Animal Abolitionism and ‘Racism without Rac-
ists’. Journal of Agricultural and Environmental Ethics 30: 745–764.
Cox, Christopher. 2010. Consider the Oyster. Slate, April 7. Accessed July 8,
2015. www.slate.com/articles/life/food/2010/04/consider_the_oyster.html.
Crisp, Roger. 1988. Utilitarianism and Vegetarianism. International Journal of
Applied Philosophy 4: 41–49.
Cuneo, Terence. 2015. Conscientious Omnivorism. In Philosophy Comes to Din-
ner: Arguments Over the Ethics of Eating, edited by Matthew C. Halteman,
Terence Cuneo and Andrew Chignell, 21–38. New York: Routledge.
Curnutt, Jordan. 1997. A New Argument for Vegetarianism. Journal of Social
Philosophy 28 (3): 153–172.
Currah, J. M., S. H. Hendrick and J. M. Stookey. 2009. The Behavioral Assess-
ment and Alleviation of Pain Associated with Castration in Beef Calves Treated
with Flunixin Meglumine and Caudal Lidocaine Epidural Anesthesia with Epi-
nephrine. Canadian Veterinary Journal 50: 375–382.
Daily News. March 13. Accessed February 21, 2010. www.nydailynews.com/life
style/2009/03/13/2009-03-13_he_hogties_abuse_on_hbo_documentary_deat.
html.
Damasio, Antonio and Gil B. Carvalho. 2013. “The nature of feelings: evolution-
ary and neurobiological origins.” Nature Reviews Neuroscience 14: 143–152.
Danbury, T. C., C. A. Weeks, J. P. Chambers, A. E. Waterman-Pearson and S. C.
Kestin. 2000. Self- Selection of the Analgesic Drug Carprofen by Lame Broiler
Chickens. The Veterinary Record 146: 307–311.
Danielle, Melissa. 2009. Nutrition Liberation: Plant-Based Diet as a Tool for
Healing, Resistance, and Self-Reliance. In Sistah Vegan, edited by A. Breeze
Harper, 47–52. New York: Lantern Books.
Davis, Steven L. 2003. The Least Harm Principle May Require That Humans
Consume a Diet Containing Large Herbivores, Not a Vegan Diet. Journal of
Agricultural and Environmental Ethics 16: 387–394.
Deep, A., K. Schwean-Lardner, T. G. Crowe, B. I. Fancher and H. L. Classen.
2010. Effect of Light Intensity on Broiler Production, Processing Characteris-
tics, and Welfare. Poultry Science 89 (11): 2326–2333.
DeGrazia, David. 1996. Taking Animals Seriously: Mental Life and Moral Status.
New York: Cambridge University Press.
———. 2009. Moral Vegetarianism from a Very Broad Basis. Journal of Moral
Philosophy 6: 143–165.
176 References
Demetriou, Dan and Bob Fischer. 2018. Dignitarian Hunting: A Rights-Based
Defense. Social Theory and Practice 44 (1): 49–73.
Diamond, Cora. 1978. Eating Meat and Eating People. Philosophy 53: 465–479.
Dimitrijevic, N., S. Dzitoyeva, R. Satta, M. Imbesi, S. Yildiz, and H. Manev. 2005.
Drosophila GABA(B) Receptors are Involved in Behavioral Effects of Gamma-
Hydroxybutyric Acid (GHB). European Journal of Pharmacology 519: 246–252.
Doggett, Tyler. 2018. Moral Vegetarianism. In The Stanford Encyclopedia of Phi-
losophy (Fall 2018 Edition), edited by Edward N. Zalta. https://plato.stanford.
edu/archives/fall2018/entries/vegetarianism/.
Donham, K. J. 2000. The Concentration of Swine Production: Effects on Swine
Health, Productivity, Human Health, and the Environment. The Veterinary
Clinics of North America: Food Animal Practice 16 (3): 559–597.
Dozier, W. A. III, M. P. Lacy and L. R. Vest. 2001. Broiler Production and Man-
agement. Athens: The University of Georgia College of Agricultural and Envi-
ronmental Sciences, Cooperative Extension Service.
Driver, Julia. 2015. Individual Consumption and Moral Complicity. In The
Moral Complexities of Eating Meat, edited by Ben Bramble and Bob Fischer,
67–79. New York: Oxford University Press.
Duncan, I. J. H. 2001. Animal Welfare Issues in the Poultry Industry: Is There a Les-
son to Be Learned? Journal of Applied Animal Welfare Science 4 (3): 207–221.
———. 2004. Welfare Problems of Poultry. In The Well- Being of Farm Animals:
Challenges and Solutions, edited by G. J. Benson and B. E. Rollin. Ames, IA:
Blackwell Publishing.
Duve, H. and A. Thorpe. 1979. “Immunofluorescent localization of insulin-like
material in the median neurosecretory cells of the blowfly, Calliphora vomito-
ria (Diptera).” Cell and tissue research 200: 187–191.
Edelman, Gerald M. 2003. “Naturalizing consciousness: A theoretical frame-
work.” Proceedings of the National Academy of Sciences 100: 5520–5524.
Eisemann, C. H., W. K. Jorgensen, D. J. Merritt, M. J. Rice, W. Cribb, P. D.
Webb, and M. P. Zalucki. 1984. Do Insects Feel Pain? – A Biological View.
Experientia 40: 164–167.
Eisemann, J. H. and R. A. Argenzio. 1999. Effects of Diet and Housing Density
on Growth and Stomach Morphology in Pigs. Journal of Animal Science 77:
2709–2714.
Engel, Mylan. 2016. The Commonsense Case for Ethical Vegetarianism. Between
the Species: A Journal of Ethics 19 (1): 2–31.
Estevez, I. 2007. Density Allowances for Broilers: Where to Set the Limits? Poul-
try Science 86: 1265–1272.
Everett, Jennifer. 2004. Vegetarianism, Predation, and Respect for Nature. In
Food for Thought: The Debate Over Eating Meat, edited by Steve Sapontzis,
302–314. Amherst, MA: Prometheus Books.
Fairlie, Simon. 2011. Meat: A Benign Extravagance. White River Junction, VT:
Chelsea Green Publishing.
Faria, C. 2016. Animal Ethics Goes Wild: The Problem of Wild Animal Suffering
and Intervention in Nature, PhD thesis, Barcelona: Pompeu Fabra University.
Faria, C. and E. Paez. 2015. Animals in Need: The Problem of Wild Animal Suffer-
ing and Intervention in Nature. Relations: Beyond Anthropocentrism 3: 7–13.
Febrer, K., T. A. Jones, C. A. Donnelly and M. S. Dawkins. 2006. Forced to
Crowd or Choosing to Cluster? Spatial Distribution Indicates Social Attraction
in Broiler Chickens. Animal Behaviour 72 (6): 1291–1300.
References 177
Fisher, A. D., M. Stewart, G. A. Verkerk, C. J. Morrow and L. R. Matthews.
2003. The Effects of Surface Type on Lying Behaviour and Stress Responses of
Dairy Cows During Periodic Weather-Induced Removal from Pasture. Applied
Animal Behaviour Science 81: 1–11.
Fischer, Bob and Andy Lamey. 2018. “Field Deaths in Plant Agriculture.” Jour-
nal of Agricultural and Environmental Ethics 31: 409–428.
Fischer, Bob and Josh Milburn. 2019. In Defense of Backyard Chickens. Journal
of Applied Philosophy.
Fitzgerald, Amy J., Linda Kalof and Thomas Dietz. 2009. Slaughterhouses and
Increased Crime Rates an Empirical Analysis of the Spillover from ‘the Jungle’
into the Surrounding Community. Organization & Environment 22: 158–184.
Fitzpatrick, S. 2017. Against Morgan’s Canon. In Routledge Handbook of Phi-
losophy of Animal Minds, edited by Kristin Andrews and Jacob Beck, 437–
447. New York: Routledge.
Fleischman, Diana. 2013. The Ethical Case for Eating Oysters and Mussels. https://
sentientist.org/2013/05/20/the-ethical-case-for-eating-oysters-and-mussels/
Fox, J. T., D. U. Thomson, N. N. Lindberg and K. Barling. 2009. A Comparison
of Two Vaccines to Reduce Liver Abscesses in Natural-Fed Beef Cattle. Bovine
Practitioner 43: 168–174.
Fregonesi, J. A., D. M. Veira, M. A. G. Von Keyserlingk and D. M. Weary. 2007.
Effects of Bedding Quality on Lying Behavior of Dairy Cows. Journal of Dairy
Science 90: 5468–5472.
Frey, R. G. 1983. Rights, Killing, and Suffering: Moral Vegetarianism and Applied
Ethics. Oxford: Blackwell Publishers.
Gamroth, M., T. Downing and A. Peters Ruddell. 2000. Flame-Clipping Udders
on Dairy Cows. Oregon State University Extension Service EM 8755. https://
catalog.extension.oregonstate.edu/sites/catalog/files/project/pdf/em8755.pdf.
Garcia, L. G., K. L. Nicholson, T. W. Hoffman, T. E. Lawrence, D. S. Hale, D. B.
Griffin, J. W. Savell, D. L. Vanoverbeke, J. B. Morgan, K. E. Belk, T. G. Field, J. A.
Scanga, J. D. Tatum and G. C. Smith. 2008. National Beef Quality Audit-2005:
Survey of Targeted Cattle and Carcass Characteristics Related to Quality, Quan-
tity, and Value of Fed Steers and Heifers. Journal of Animal Science 86: 3533–3543.
Garrett, Jeremy R. 2007. Utilitarianism, Vegetarianism, and Human Health:
A Response to the Causal Impotence Objection. Journal of Applied Philoso-
phy 24: 223–237.
Garvey, James. 2011. Climate Change and Causal Inefficacy: Why Go Green When
It Makes No Difference? Royal Institute of Philosophy Supplements 69: 157–174.
Gentle, M. J. and V. L. Tilston. 2000. Nociceptors in the Legs of Poultry: Implica-
tions for Potential Pain in Preslaughter Shackling. Animal Welfare 9: 227–236.
George, Kathryn Paxton. 1994. Should Feminists Be Vegetarians? Signs 19 (2):
405–434.
Gill, Michael B. 2013. On Eating Animals. Social Philosophy and Policy 30:
201–207.
González, L. A., K. S. Schwartzkopf-Genswein, M. Bryan, R. Silasi and F. Brown.
2012. Relationships Between Transport Conditions and Welfare Outcomes
During Commercial Long Haul Transport of Cattle in North America. Journal
of Animal Science 90: 3640–3651.
González, L. A., K. S. Schwartzkopf-Genswein, N. A. Caulkett, E. JanzEn, T.
A. Mcallister, E. FiErheller, A. L. Schaefer, D. B. Haley, J. M. Stookey and S.
Hendrick. 2010. Pain Mitigation After Band Castration of Beef Calves and
178 References
Its Effects on Performance, Behavior, Escherichia Coli, and Salivary Cortisol.
Journal of Animal Science 88: 802–810.
Grandin, T., ed. 1998. Genetics and the Behavior of Domestic Animals, 323. San
Diego, CA: Academic Press.
———. 2001. Livestock Trucking Guide: Livestock Management Practices That
Reduce Injuries to Livestock During Transport. National Institute for Ani-
mal Agriculture. www.animalagriculture.org/Education/Pamphlets/Livestock
Trucking Guide.pdf.
———. 2007. Recommended Animal Handling Guidelines and Audit Guide.
Washington, DC: American Meat Institute Foundation.
———. 2012 Recommended Animal Handling Guidelines & Audit Guide:
A Systematic Approach to Animal Welfare. Washington, DC: Am. Meat Inst.
[Online]. www.animal-handling.org/ht/a/GetDocumentAction/i/80011.
Greaves, K. and S. Tuene. 2001. The Form and Context of Aggressive Behaviour
in Farmed Atlantic Halibut (Hippoglossus hippoglossus L.). Aquaculture 193
(1–2): 139–147.
Gregory, N. G. and L. J. Wilkins. 1992. Skeletal Damage and Bone Defects Dur-
ing Catching and Processing. In Bone Biology and Skeletal Disorders in Poul-
try, edited by C. C. Whitehead. Abingdon: Carfax Publishing.
Griffin, D. 1984. Animal Thinking. Cambridge: Harvard University Press.
Gruen, Lori. 1993. Dismantling Oppression: An Analysis of the Connection
Between Women and Animals. In Ecofeminism: Women, Animals, Nature,
edited by Greta Gaard. Philadelphia: Temple University Press.
Gunnarsson, Logi. 2008. The Great Apes and the Severely Disabled: Moral Status
and Thick Evaluative Concepts. Ethical Theory and Moral Practice 11: 305–326.
Hagen, Kristin and Donald M. Broom. 2004. Emotional Reactions to Learning in
Cattle. Applied Animal Behaviour Science 85 (3): 203–213.
Hare, R. M. 1993. Why I Am Only a Demi-Vegetarian. In Essays on Bioethics.
Harman, Elizabeth. 2016. Morally Permissible Moral Mistakes. Ethics126 (2):
366–393.
Harper, A. B. 2010. Race as a ‘Feeble Matter’ in Veganism: Interrogating White-
ness, Geopolitical Privilege, and Consumption Philosophy of ‘CrueltyFree’
Products. Journal for Critical Animal Studies 8 (3): 5–27.
Harris, John. 2011. Moral Enhancement and Freedom. Bioethics 25 (2): 102–111.
———. 2012. ‘Pass the Cocoamone, Please’: Causal Impotence, Opportunistic
Vegetarianism and Act-Utilitarianism. Ethics, Policy and Environment 15:
368–383.
Herzog, Hal. 2010. Some We Love, Some We Hate, Some We Eat: Why It’s So
Hard to Think Straight About Animals. New York: HarperCollins.
Hettinger, Ned. 2004. Bambi Lovers Versus Tree Huggers. In Food for Thought:
The Debate Over Eating Meat, edited by Steve Sapontzis, 294–301. Amherst,
MA: Prometheus Books.
Hill, Judith M. 1987. Pornography and Degradation. Hypatia 2: 39–54.
Hoehl, S., K. Hellmer, M. Johansson, and G. Gredebäck. 2017. Itsy Bitsy Spi-
der. . . : Infants React with Increased Arousal to Spiders and Snakes. Frontiers
in Psychology 8: 1–8.
Hölldobler, B. and E. Wilson. 1978. The Multiple Recruitment Systems of the
African Weaver Ant Oecophylla Longinoda (Latreille) (Hymenoptera: Formi-
cidae). Behavioral Ecology and Sociobiology 3: 19–60.
References 179
Hooley, Dan and Nathan Nobis. 2015. A Moral Argument for Veganism. In
Philosophy Comes to Dinner, edited by Andrew Chignell, Terence Cuneo and
Matthew Halteman, 92–108. New York: Routledge.
Horta, Oscar. 2010. Debunking the Idyllic View of Natural Processes. Télos 17:
73–88.
Horta, Oscar. 2014. The Scope of the Argument from Species Overlap. Journal of
Applied Philosophy 31 (2): 142–154.
Horta, Oscar. 2017 Animal Suffering in Nature: The Case for Intervention. Envi-
ronmental Ethics 39: 261–279.
Hourdequin, Marion. 2010. Climate, Collective Action, and Individual Ethical
Obligations. Environmental Values 19: 443–464.
Hsiao, Timothy. 2015. In Defense of Eating Meat. Journal of Agricultural and
Environmental Ethics 28: 277–291.
Hsiao, Timothy. 2017. Industrial Farming Is not Cruel to Animals. Journal of
Agricultural and Environmental Ethics 30: 37–54.
Huemer, Michael. 2019. Dialogues on Ethical Vegetarianism. New York: Routledge.
Hursthouse, Rosalind. 2006. Applying Virtue Ethics to Our Treatment of the
Other Animals. In The Practice of Virtue: Classic and Contemporary Readings
in Virtue Ethics, edited by Jennifer Welchman, 136–153. Indianapolis: Hackett
Publishing Company.
Jack, A. I. and P. Robbins. 2012. The Phenomenal Stance Revisited. Review of
Philosophy and Psychology 3: 383–403.
Johnston, J. 2008. The Citizen-Consumer Hybrid: Ideological Tensions and the
Case of Whole Foods. Theory and Society 37: 229–270.
Julian, R. J. 2004. Evaluating the Impact of Metabolic Disorders on the Welfare
of Broilers. In Measuring and Auditing Broiler Welfare, edited by C. Weeks and
A. Butterworth. Wallingford: CAB International.
———. 2005. Production and Growth Related Disorders and Other Metabolic
Diseases of Poultry—A Review. The Veterinary Journal 169: 350–369.
Kagan, Shelly. 2011. Do I Make a Difference? Philosophy & Public Affairs 39:
105–141.
Kasperbauer, T. J. 2016. Mentalizing Animals: Implications for Moral Psychol-
ogy and Animal Ethics. Philosophical Studies.
Kazez, Jean. 2018. The Taste Question in Animal Ethics. The Journal of Applied
Philosophy 35(4): 661–674.
Kemmerer, Lisa. 2014. Eating Earth: Environmental Ethics and Dietary Choice.
Kendall, Gillian. 2011. The Greater Good: Peter Singer on How to Live an Ethi-
cal Life. The Sun Magazine: The Sun Interview, May (425).
Kennard, D. C. 1930. ‘Chicken Vices’ and ‘Tipping the Beaks’. In Poultry, Ohio
Agricultural Experiment Station. Special Circular 28.
Key, Brian. 2016. Why Fish Do Not Feel Pain. Animal Sentience 3.
Key, N. and McBride, W. 2007. The Changing Economics of U.S. Hog Pro-
duction. Economic Research Report Number 52. Economic Research Service.
United States Department of Agriculture.
Kheel, Marti. 2004. Vegetarianism and Ecofeminism: Toppling Patriarchy with a
Fork. In Food for Thought: The Debate Over Eating Meat, edited by Steve F.
Sapontzis. New York: Prometheus.
Knobe, J. and J. Prinz. 2008. Intuitions About Consciousness: Experimental
Studies. Phenomenology and Cognitive Science 7: 67–83.
180 References
Knowles, T. G., S. C. Kestin, S. M. Haslam, et al. 2008. Leg Disorders in
Broiler Chickens: Prevalence, Risk Factors and Prevention. PLoS One 3
(2): e1545. doi:10.1371/journal.pone.0001545. www.plosone.org/article/
info:doi/10.1371/journal.pone.0001545.
Kolliker, Mathias and Michel Vancassel. 2007. Maternal Attendance and the
Maintenance of Family Groups in Common Earwigs (Forficula auricularia):
A Field Experiment. Ecological Entomology 32: 24–27.
Korsgaard, Christine M. 2018. Fellow Creatures: Our Obligations to the Other
Animals. New York: Oxford University Press.
Kristensen, H. H., G. C. Perry, N. B. Prescott, J. Ladewig, A. K. Ersbøll and C.
M. Wathes. 2006. Leg Health and Performance of Broiler Chickens Reared in
Different Light Environments. British Poultry Science 47 (3): 257–263.
Kunst, Jonas R., and Sigrid M. Hohle. 2016. Meat Eaters by Dissociation: How
we Present, Prepare and Talk about Meat Increases Willingness to Eat Meat by
Reducing Empathy and Disgust. Appetite 105: 758–774.
Lacy, M. P. 2002. Broiler Management. In Commercial Chicken Meat and Egg
Production, edited by D. D. Bell and W. D. Weaver Jr, 5th edition, 856. Nor-
well, MA: Kluwer Academic Publishers.
Lacy, M. P. and M. Czarick. 1998. Mechanical Harvesting of Broilers. Poultry
Science 77: 1794–1797.
Lambooij, E. 2007. Transport of Pigs. In Livestock Handling and Transport,
edited by T. Grandin, 3rd edition, 228–244. Cambridge, MA: CABI.
Lamey, Andy. 2019. Duty and the Beast. New York: Cambridge University Press.
Langton, Rae. 2009. Sexual Solipsism: Philosophical Essays on Pornography and
Objectification. New York: Oxford University Press.
Lawford-Smith, Holly. 2015. Unethical Consumption and Obligations to Signal.
Ethics & International Affairs 29 (3): 315–330.
Lay, D. C., T. H. Friend, R. D. Randel, C. L. Bowers, K. K. Grissom and O.
C. Jenkins. 1992. Behavioral and Physiological Effects of Freeze or Hot-Iron
Branding on Crossbred Cattle. Journal of Animal Science 70: 330–336.
Lean, I. J., J. M. Thompson and F. R. Dunshea. 2014. A Meta-Analysis of Zil-
paterol and Ractopamine Effects on Fedlot Performance, Carcass Traits and
Shear Strength of Meat in Cattle. PLoS One 9: e115904.
Lee-Johnson, C.P. and D.A. Carnegie. 2010. “Mobile robot navigation modu-
lated by artificial emotions.” IEEE Transactions on Systems, Man, and Cyber-
netics, Part B (Cybernetics) 40: 469–480.
Lehman, Hugh. On the Moral Acceptability of Killing Animals. Journal of Agri-
cultural and Environmental Ethics 1 (1988): 155–162.
Leonard, Christopher. 2014. The Meat Racket: The Secret Takeover of America's
Food Business. New York: Simon & Schuster.
Leopold, Aldo. A Sand County Almanac: And Sketches Here and There. New
York: Oxford University Press, 1949.
Liao, S. Matthew. 2010. The Basis of Human Moral Status. Journal of Moral
Linzey, Andrew, ed. 2009. The Link Between Animal Abuse and Human Vio-
lence. Eastbourne: Sussex Academic Press.
Lipscomb, Benjamin. 2015. ‘Eat Responsibly’: Agrarianism and Meat. In Philosophy
Comes to Dinner: Arguments Over the Ethics of Eating, edited by Matthew C.
Halteman, Terence Cuneo and Andrew Chignell, 56–72. New York: Routledge.
Lomasky, Loren. 2013. Is It Wrong to Eat Animals? Social Philosophy and Policy
30: 177–200.
References 181
Loneragan, G. H. and M. M. Brashears. 2005. Effects of Using Retention-Pond
Water for Dust Abatement on Performance of Feedlot Steers and Carriage of
Escherichia Coli O157 and Salmonella Spp. Journal of the American Veteri-
nary Medical Association 226: 1378–1383.
Loneragan, G. H., D. U. Thomson and H. M. Scott. 2014. Increased Mortal-
ity in Groups of Cattle Administered the B-Adrenergic Agonists Ractopamine
Hydrochloride and Zilpaterol Hydrochloride. PLoS One 9: e91177.
Loughnan, S., N. Haslam and B. Bastian. 2010. The Role of Meat Consumption in
the Denial of Moral Status and Mind to Meat Animals. Appetite 55: 156–159.
Lusk, Jayson. 2011. The Market for Animal Welfare. Agriculture and Human
Values 28: 561–575.
MacClellan, Joel. 2013. Animal Size, Contributory Causation, and Ethical Veg-
etarianism. Journal of Animal Ethics 3: 57–68.
MacKinnon, Catharine. 2004. Of Mice and Men: A Feminist Fragment on Ani-
mal Rights. In Animal Rights: Current Debates and New Directions, edited by
Martha Nussbaum and Cass Sunstein. Oxford: Oxford University Press.
MacLean, Douglas. 2010. Is ‘Human Being’ a Moral Concept? Philosophy &
Public Policy Quarterly 30: 16–20.
Maniates, Michael F. 2001. Individualization: Plant a Tree, Buy a Bike, Save the
World? Global Environmental Politics 1: 31–52.
Marder, Michael. 2013. Plant-Thinking: A Philosophy of Vegetal Life. New
York: Columbia University Press.
Marino, Lori. 2017. Thinking Chickens: A Review of Cognition, Emotion, and
Behavior in the Domestic Chicken. Animal Cognition. Online First. http://link.
springer.com/article/10.1007/s10071-016-1064-4.
Martin, Michael. 1976. A Critique of Moral Vegetarianism. Reason Papers 3:
13–43.
Mashour, George A. and Michael T. Alkire. 2013. “Evolution of conscious-
ness.” Proceedings of the National Academy of Sciences 110 (Supplement 2):
10357–10364.
Matheny, Gaverick. 2002. Expected Utility, Contributory Causation, and Veg-
etarianism. Journal of Applied Philosophy 19: 293–297.
———. 2006. Utilitarianism and Animals. In In Defense of Animals: The Second
Wave, edited by Peter Singer, 13–25. Malden: Blackwell.
Mayo Clinic Staff. 2017. Vegetarian Diet: How to Get the Best Nutrition.
www.mayoclinic.org/healthy-lifestyle/nutrition-and-healthy-eating/in-depth/
vegetarian-diet/art-20046446.
McCloskey, Deirdre. 1991. If You’re So Smart: The Narrative of Economic
Expertise. Chicago: University of Chicago Press.
McMullen, Steven and Matthew C. Halteman. 2019. Against Inefficacy Objec-
tions: The Real Economic Impact of Individual Consumer Choices on Animal
Agriculture. Food Ethics 2: 93–110.
McPherson, Tristram. 2015. Why I Am a Vegan (and You Should Be Too). In Phi-
losophy Comes to Dinner, edited by A. Chignell, T. Cuneo, and M. Halteman,
73–91. New York: Routledge.
McWilliams, James. 2015. The Modern Savage. New York: Thomas Dunne Books.
Mendl, Michael, Suzanne Held and Richard W. Byrne. 2010. Pig Cognition. Cur-
rent Biology 20 (18): R796–R798.
Menhinick, Edward F. 1962. Comparison of Invertebrate Populations of Soil and
Litter of Mowed Grasslands in Areas Treated and Untreated with Pesticides.
Ecology 43 (3): 556–561.
182 References
Merker, Bjorn. 2005. “The liabilities of mobility: A selection pressure for the
transition to consciousness in animal evolution.” Consciousness and Cognition
14: 89–114.
Meyers, C. D. 2013. Why It Is Morally Good to Eat (Certain Kinds of) Meat:
The Case for Entomophagy. Southwest Philosophy Review 29 (1): 119–126.
Milburn, Josh. Forthcoming. Death-Free Dairy? The Ethics of Clean Milk. Jour-
nal of Agricultural and Environmental Ethics.
Morin, Alain. 2006. “Levels of consciousness and self-awareness: a comparison
and integration of various neurocognitive views.” Consciousness and Cogni-
tion 15: 358–371.
Mullally, Conner and Jayson L. Lusk. Forthcoming. The Impact of Farm Ani-
mal Housing Restrictions on Egg Prices, Consumer Welfare, and Production in
California. American Journal of Agricultural Economics.
Muller, S. Marek. 2017. For the Good of the Species: Gary Francione and the
Omnipresence of Eugenics in Animal Rights Rhetoric. Communication Studies
68 (5): 588–606.
Nääs, I. A., I. C. L. A. Paz, M. S. Baracho, et al. 2009. Impact of Lameness on
Broiler Well-Being. Journal of Applied Poultry Research 18: 432–439.
Nagaraja, T. G. and K. F. Lechtenberg. 2007a. Acidosis in Feedlot Cattle. Veteri-
nary Clinics of North America Food Animal Practice 23: 333–350.
———. 2007b. Liver Abscesses in Feedlot Cattle. Veterinary Clinics of North
America Food Animal Practice 23: 351–369.
National Pork Board and the American Association of Swine Veterinarians.
Undated. On Farm Euthanasia of Swine—Options for the Producer, 186.
Accessed February 21, 2010. www.aasv.org/aasv/euthanasia.pdf.
Nefsky, Julia. 2018. Consumer Choice and Collective Impact. In The Oxford
Handbook of Food Ethics, edited by and Barnhill, Mark Budolfson and Tyler
Doggett. New York: Oxford University Press.
Nielsen, B. L., K. Thodberg, L. Dybkjaer and E. M. Vestergaard. 2006. Feed-
ing Behaviour in Pigs. In Feeding in Domestic Vertebrates: From Structure to
Behaviour, edited by V. Bels, 156–178. Wallingford: CAB International.
Nienaber, J. A. and G. L. Hahn. 2007. Livestock Production System Management
Responses to Thermal Challenges. International Journal of Biometeorology
52: 149–157.
Nijdam, E., P. Arens, E. Lambooij, E. Decuypere and J. A. Stegeman. 2004. Fac-
tors Influencing Bruises and Mortality of Broilers During Catching, Transport,
and Lairage. Poultry Science 83: 1610–1615.
Nijdam, E., A. R. M. Zailan, J. H. H. van Eck, E. Decuypere and J. A. Stegeman.
2006. Pathological Features in Dead on Arrival Broilers with Special Reference
to Heart Disorders. Poultry Science 85: 1303–1308.
Ng, Y.-K. 1995. Towards Welfare Biology: Evolutionary Economics of Animal
Con-Sciousness and Suffering. Biology and Philosophy 10: 255–285.
Norcross, Alastair. 2000. Contractualism and the Ethical Status of Animals.
Southwest Philosophy Review 17: 137–143.
———. 2004. Puppies, Pigs, and People: Eating Meat and Marginal Cases. Philo-
sophical Perspectives 18: 229–245.
Norwood, F. Bailey and Jayson Lusk. 2011. Compassion, by the Pound: The Eco-
nomics of Farm Animal Welfare. New York: Oxford University Press.
O’Neill, Patricia and Lewis Petrinovich. 1998. A Preliminary Cross-Cultural
Study of Moral Intuitions. Evolution and Human Behavior 19 (6): 349–367.
References 183
Ost, David. 1986. The Case Against Animal Rights. The Southern Journal of
Philosophy 24 (3): 365–373.
Patison, K. P., D. L. Swain, G. J. Bishop-Hurley, G. Robins, P. Pattison and D. J.
Reid. 2010. Changes in Temporal and Spatial Associations Between Pairs of
Cattle During the Process of Familiarisation. Applied Animal Behaviour Sci-
ence 128: 10–17.
Pearse, A. S. 1946. Observations on the Microfauna of the Duke Forest. Ecologi-
cal Monographs 16: 127–150.
Peterson, E., D. Strohbehn, G. Ladd and R. Willham. 1989. Effects of Precondi-
tioning on Performance of Beef Calves Before and After Entering the Feedlot.
Journal of Animal Science 67: 1678–1686.
Perremans, S., J. M. Randall, G. Rombouts, E. Decuypere and R. Geers. 2001.
Effect of Whole-Body Vibration in the Vertical Axis on Cortisol and Adreno-
croticotropic Hormone Levels. Journal of Animal Science 79: 975–981.
Petrinovich, Lewis, Patricia O’Neill and Matthew Jorgensen. 1993. An Empirical
Study of Moral Intuitions: Toward an Evolutionary Ethics. Journal of Person-
ality and Social Psychology 64 (3): 467–478.
Phillips, C. J. C. and S. McCulloch. 2005. Student Attitudes on Animal Sentience
and Use of Animals in Society. Journal of Biological Education 40: 17–24.
Phillips, C., S. Izmirli, J. Aldavood, M. Alonso, B. I. Choe, and A. Hanlon. 2011.
An International Comparison of Female and Male Students’ Attitudes to the
Use of Animals. Animals 1: 7–26.
Plous, S. 1993. Psychological Mechanisms in the Human Use of Animals. Journal
of Social Issues 49: 11–52.
Posewitz, Jim. 1995. Beyond Fair Chase: The Ethic and Tradition of Hunting.
Helena, MT: Falcon Press.
Prescott, N. B., H. H. Kristensen and C. M. Wathes. 2004. Light. In Measuring
and Auditing Broiler Welfare, edited by C. A. Weeks and A. Butterworth. Wall-
ingford: CAB International.
Preylo, Brooke Dixon and Hiroko Arikawa. 2008. Comparison of Vegetar-
ians and Non-Vegetarians on Pet Attitude and Empathy. Anthrozoös 21 (4):
387–395.
Quinn, Warren S. 1989. Actions, Intentions, and Consequences: The Doctrine of
Double Effect. Philosophy and Public Affairs 18: 334–351.
Rachels, James. 2004. The Basic Argument for Vegetarianism. In Food for
Thought: The Debate over Eating Meat, edited by Steve F. Sapontzis, 70–80.
Amherst: Prometheus Books.
Rachels, Stuart. Vegetarianism. 2011. In The Oxford Handbook of Animal Eth-
ics, edited by R. G. Frey and Tom L. Beauchamp, 877–905. New York: Oxford
University Press.
Raj, A. B. M. 2004. Stunning and Slaughter of Poultry. In Poultry Meat Process-
ing and Quality, edited by G. C. Mead. Cambridge: Woodhead Publishing Ltd.
Raj, M. 2008. Welfare of Pigs During Stunning and Slaughter. In Welfare of Pigs
from Birth to Slaughter, edited by L. Faucitano and A. L. Schaefer, 225–242.
Wageningen, The Netherlands: Wageningen Academic Publishers.
Rao, M. A., T. G. Knowles, and S. B. Wotton. 2013. The Effect of Pre-Stun
Shocks in Electrical Water- Bath Stunners on Carcase and Meat Quality in
Broilers. Animal Welfare 22(1): 79–84.
Regan, Tom. 2004. The Case for Animal Rights. Berkeley: University of Califor-
nia Press.
184 References
Reinhardt, C. D., C. I. Vahl, B. E. Depenbusch, J. P. Hutcheson and D. U. Thom-
son. 2014. Feeding Zilpaterol Hydrochloride Is Associated with Decreased Dry
Matter Intake Shortly After Initiation of Feeding Dependent on Season and
Previous Intake. Journal of Animal Science 92: 4751–4760.
Rémy, C., J. Girardie, and M. P. Dubois. 1979. “Vertebrate neuropeptide-like
substances in the suboesophageal ganglion of two insects: Locusta migratoria
R. and F.(Orthoptera) and Bombyx mori L.(Lepidoptera) immunocytological
investigation.” General and comparative endocrinology 37: 93–100.
Rice, Marlin E., Matthew E. O'Neal, and Palle Pedersen. 2007. Soybean Aphids
in Iowa—2007. Agriculture and Environment Extension Publications 89.
Ritter, M. J., M. Ellis, R. Bowman, et al. 2008. Effects of Season and Distance
Moved During Loading on Transport Losses of Market-Weight Pigs in Two
Commercially Available Types of Trailer. Journal of Animal Science 86:
317–345.
Roser, Max. 2016. Global Economic Inequality. https://ourworldindata.org/
global-economic-inequality.
Rothgerber, Hank. 2013. Real Men Don’t Eat (vegetable) Quiche: Masculinity
and the Justification of Meat Consumption. Psychology of Men & Masculinity
14 (4): 363–375.
———. 2014a. A Comparison of Attitudes Toward Meat and Animals Among
Strict and Semi-Vegetarians. Appetite 72: 98–105.
———. 2014b. Efforts to Overcome Vegetarian-Induced Dissonance Among
Meat Eaters. Appetite 79: 32–41.
———. 2015. Can You Have Your Meat and Eat It Too? Conscientious Omni-
vores, Vegetarians, and Adherence to Diet. Appetite 84: 196–203.
Rowlands, Mark. 2009. Animal Rights: Moral Theory and Practice. New York:
Palgrave Macmillan.
Rozin, P., M. Markwith and C. Stoess. 1997. Moralization and Becoming a Veg-
etarian: The Transformation of Preferences into Values and the Recruitment of
Disgust. Psychological Science 8: 67–73.
Rubio, Julie Hanlon. 2014. Animals, Evil, and Family Meals. Journal of Moral
Theology 3: 35–53.
Ruby, Matthew B. and Steven J. Heine. 2012. Too Close to Home: Factors Pre-
dicting Meat Avoidance. Appetite 59: 47–52.
Sabrosky, C. W. 1953. How Many Insects are There? Systematic Biology 2:
31–36.
Saja, Krzysztof. 2013. The Moral Footprint of Animal Products. Agriculture and
Human Values 30: 193–202.
Scanes, C. G., G. Brant and M. E. Ensminger. 2004. Poultry Science, 4th edition,
260. Upper Saddle River, NJ: Pearson Prentice Hall.
Schaefer, G. Owen and Julian Savulescu. 2014. The Ethics of Producing In Vitro
Meat. Journal of Applied Philosophy 31: 188–202.
Schipper, C., T. Church and B. Harris. 1989. A Review of the Alberta Certified
Preconditioned Feeder Program (1980–1987). Canadian Veterinary Journal
30: 736–741.
Schwartz, Judith. 2013. Cows Save the Planet: And Other Improbable Ways of
Restoring Soil to Heal the Earth. White River Junction, VT: Chelsea Green
Publishing.
References 185
Schwartzkopf-Genswein, K. S., K. A. Beauchemin, D. J. Gibb, D. H. Crews, D.
D. Hickman, M. Streeter and T. A. Mcallister. 2003. Effect of Bunk Manage-
ment on Feeding Behavior, Ruminal Acidosis and Performance of Feedlot Cat-
tle: A Review. Journal of Animal Science 81: E149–E158.
Schwartzkopf-Genswein, K. S., L. Faucitano, S. Dadgar, P. Shand, L. A. González
and T. G. Crowe. 2012. Road Transport of Cattle, Swine and Poultry in
North America and Its Impact on Animal Welfare, Carcass and Meat Quality:
A Review. Meat Science 92: 227–243.
Schwartzkopf-Genswein, K. S., J. M. Stookey, T. G. Crowe and B. M. Genswein.
1998. Comparison of Image Analysis, Exertion Force, and Behavior Measure-
ments for Use in the Assessment of Beef Cattle Responses to Hot-Iron and
Freeze Branding. Journal of Animal Science 76: 972–979.
Schwartzkopf-Genswein, K. S., J. M. Stookey and R. Welford. 1997. Behavior
of Cattle During Hot-Iron and Freeze Branding and the Effects on Subsequent
Handling Ease. Journal of Animal Science 75: 2064–2072.
Schwean-Lardner, K., B. I. Fancher, S. Gomis, A. Van Kessel, S. Dalal and H. L.
Classen. 2013. Effect of Day Length on Cause of Mortality, Leg Health, and
Ocular Health in Broilers. Poultry Science 92: 1–11.
Schwitzgebel, Eric and Joshua Rust. 2016. The Behavior of Ethicists. In A Com-
panion to Experimental Philosophy, edited by Justin Sytsma and Wesley Buck-
walter, 225–233. Malden: Wiley-Blackwell.
Scruton, Roger. 2004. The Conscientious Carnivore. In Food for Thought: The
Debate Over Eating Meat, edited by Steve Sapontzis, 81–91. Amherst, MA:
Prometheus Books.
———. 2006. Animal Rights and Wrongs. London: Continuum.
Sebo, Jeff. 2018. The Moral Problem of Other Minds. The Harvard Review of
Shim, M. Y., A. B. Karnuah, A. D. Mitchell, N. B. Anthony, G. M. Pesti and S.
E. Aggrey. 2012. The Effects of Growth Rate on Leg Morphology and Tibia
Breaking Strength, Mineral Density, Mineral Content, and Bone Ash in Broil-
ers. Poultry Science 91: 1790–1795.
Singer, Peter. 2009. Animal Liberation: The Definitive Classic of the Animal
Movement. New York: Harper Perennial.
Sinnott-Armstrong, Walter. 2005. It’s Not My Fault: Global Warming and Indi-
vidual Moral Obligations. In Perspectives on Climate Change, edited by Walter
Sinnott-Armstrong and Richard Howarth, 221–253. New York: Elsevier.
Song, Mingyang, Teresa T. Fung, Frank B. Hu, Walter C. Willett, Valter D.
Longo, Andrew T. Chan and Edward L. Giovannucci. 2016. Association of
Animal and Plant Protein Intake with All-Cause and Cause-Specific Mortality.
JAMA Internal Medicine 176 (10): 1453–1463.
Springmann, Marco, H. Charles J. Godfray, Mike Rayner and Peter Scarbor-
ough. 2016. Analysis and Valuation of the Health and Climate Change Coben-
efits of Dietary Change. Proceedings of the National Academy of Sciences 113
(15): 4146–4151.
Stackhouse-Lawson, K. R., C. B. Tucker, M. S. Calvo-Lorenzo and F. M. Mit-
loehner. 2015. Effects of Growth-Promoting Technology on Feedlot Cattle
Behavior in the 21 Days Before Slaughter. Applied Animal Behaviour Science
162: 1–8.
186 References
Stafford, K. J. and D. J. Mellor. 2011. Addressing the Pain Associated with Dis-
budding and Dehorning in Cattle. Applied Animal Behaviour Science 135:
226–231.
Stock, M. L., S. L. BalDridge, D. Griffin and J. F. Coetzee. 2013. Bovine Dehorn-
ing: Assessing Pain and Providing Analgesic Management. Veterinary Clinics
of North America Food Animal Practice 29: 103–133.
Swanson, Jennifer. 2011. Contractualism and the Moral Status of Animals.
Between the Species 14: 1–17.
Sytsma, J. and E. Machery. 2012. The Two Sources of Moral Standing. Review
of Philosophy and Psychology 3: 303–324.
Tännsjö, Torbjörn. 2017. It’s Getting Better All the Time. In Food, Ethics, and
Society: And Introductory Text with Readings, edited by Anne Barnhill, Mark
Budolfson, and Tyler Doggett. New York: Oxford University Press.
Taylor, L. F., C. W. Booker, G. K. Jim and P. T. Guichon. 1997. Epidemiologi-
cal Investigation of the Buller Steer Syndrome (riding behaviour) in a Western
Canadian Feedlot. Australian Veterinary Journal 75: 45–51.
Terrell, S. P., D. U. Thomson, K. StacKhouse-Lawson, M. D. Apley, C. Larson
and C. D. Reinhardt. 2013. Perception of Lameness Management, Education,
and Welfare Implications in the Feedlot from Consulting Nutritionists, Veteri-
narians, and Feedlot Managers. Bovine Practitioner 48: 53–60.
Tomasik, Brian. 2015. The Importance of Wild-Animal Suffering. Relations:
Beyond Anthropocentrism 3: 133–152.
Tucker, C. B., E. M. Mintline, J. Banuelos, K. A. Walker, B. Hoar, Varga D. Drake
and D. M. Weary. 2014a. Effect of a Cooling Gel on Pain Sensitivity and Heal-
ing of Hot-Iron Cattle Brands. Journal of Animal Science 92: 5674–5682.
———. 2014b. Pain Sensitivity and Healing of Hot-Iron Cattle Brands. Journal
of Animal Science 92: 5674–5682.
Tuso, Philip J., et al. 2013. Nutritional Update for Physicians: Plant-Based Diets.
The Permanente Journal 17 (2): 61–66.
UNDP. 2017. About South Sudan. www.ss.undp.org/content/south_sudan/en/
home/countryinfo.html.
[U.S.C.] United States Code. 1958. Humane Methods of Slaughter Act. 7 United
States Code. §§1901 et seq. Accessed February 21, 2010. http://uscode.house.
gov/download/pls/07C48.txt.
[USDA] United States Department of Agriculture. 1999. Cattle and Swine Truck-
ing Guide for Exporters. Agricultural Marketing Services. [Online]. http://purl.
access.gpo.gov/GPO/LPS18102.
———. 2000. Feedlot 1999, Part III: Health Management and Biosecurity in US
Feedlots. Fort Collins, CO: USDA-APHIS National Health Monitoring System.
———. 2005. Census of Aquaculture, Volume 3, Special Studies Part 2, 2002
Census of Agriculture. www.agcensus.usda.gov/Publications/2002/Aquacul
ture/AQUACEN.pdf.
———. 2007a. Beef 2007–08, Part I: Reference of Beef Cow-Calf Management
Practices in the United States. Fort Collins, CO: USDA-APHIS National Health
Monitoring System.
———. 2007b. Beef 2007–08, Part III: Changes in the US Beef Cow-Calf Indus-
try, 1993–2008. Fort Collins, CO: USDA-APHIS National Health Monitoring
System.
References 187
———. 2007c. Swine 2006 Part I: Reference of Swine Health and Management
Practices in the United States, 2006. National Agricultural Statistics Service.
www.aphis.usda.gov/animal_health/nahms/swine/downloads/swine2006/
Swine2006_dr_PartI.pdf.
———. 2011. Feedlot 2011, Part I: Management Practices on US Feedlots with
a Capacity of 1,000 or More Head. Fort Collins, CO: USDA-APHIS National
Health Monitoring System.
———. 2018a. Livestock Slaughter: 2017 Summary. National Agricultural Sta-
tistics Service, April. http://usda.mannlib.cornell.edu/usda/current/LiveSlauSu/
LiveSlauSu-04-18-2018.pdf.
———. 2018b. Poultry—Production and Value: 2017 Summary. National Agri-
cultural Statistics Service, April. http://usda.mannlib.cornell.edu/usda/current/
PoulProdVa/PoulProdVa-04-27-2018.pdf.
———. 2018c. Quarterly Hogs and Pigs. National Agricultural Statistics Ser-
vice, June 28. http://usda.mannlib.cornell.edu/usda/current/HogsPigs/Hogs
Pigs-06-28-2018.pdf.
———. 2018d. Cattle. National Agricultural Statistics Service, July. http://usda.
mannlib.cornell.edu/usda/nass/Catt/2010s/2018/Catt-07-20-2018.pdf.
Van Dyke, Christina. 2015. Manly Meat and Gendered Eating: Correcting Imbal-
ance and Seeking Virtue. In Philosophy Comes to Dinner: Arguments Over the
Ethics of Eating, edited by M. C. Halteman, T. Cuneo, and A. Chignell, 39–55.
New York: Routledge.
Višak, Tatjana. 2013. Killing Happy Animals: Explorations in Utilitarian Ethics.
New York: Palgrave Macmillan.
Von Borell, E., A. Özpinar, K. M. Eslinger, A. L. Schnitz, Y. Zhao and F. M.
Mitloehner. 2007. Acute and Prolonged Effects of Ammonia on Hematological
Variables, Stress Responses, Performance, and Behavior of Nursery Pigs. Jour-
nal of Swine Health and Production 15 (3): 137–145.
Warfield, Ted. 2015. Eating Dead Animals: Meat Eating, Meat Purchasing,
and Proving Too Much. In Philosophy Comes to Dinner, edited by Andrew
Chignell, Terence Cuneo and Matthew Halteman, 151–162. New York:
Routledge.
Warren, Karen J. 1990. The Power and Promise of Ecological Feminism. Envi-
ronmental Ethics 12 (2): 125–143.
Warren, L. A., I. B. Mandell and K. G. Bateman. 2010. Road Transport Condi-
tions of Slaughter Cattle: Effects on the Prevalence of Dark, Firm and Dry Beef.
Canadian Journal of Animal Science 90: 471–482.
Warrington, E. K. and T. Shallice. 1984. Category Specific Semantic Impairments.
Brain, 107: 829–854.
Watt Poultry USA. 2005. Housing Expansion Plans. Watt Poultry USA, June,
pp. 24–28.
Weary, D. M., J. Jasper and M. J. Hötzel. 2008. Understanding Weaning Distress.
Applied Animal Behaviour Science 110: 24–41.
Weeks, C. A. 2007. Poultry Handling and Transport. In Livestock Handling
and Transport, edited by T. Grandin, 3rd edition. Wallingford: CAB Inter-
national.
Welchman, Jennifer. 2003. Xenografting, Species Loyalty, and Human Solidarity.
Journal of Social Philosophy 34: 244–255.
188 References
Westbury, H. R. and D. L. Neumann. 2008. Empathy-Related Responses to
Moving Film Stimuli Depicting Human and Non-Human Animal Targets in
Negative Circumstances. Biological Psychology 78: 66–74.
White, B. J., J. F. Coetzee, D. G. Renter, A. H. Babcock, D. U. Thomson and
D. Andresen. 2008. Evaluation of Two-Dimensional Accelerometers to Moni-
tor Behavior of Beef Calves After Castration. American Journal of Veterinary
Research 69: 1005–1012.
Wileman, B. W., D. U. Thomson, C. D. Reinhardt and D. G. Renter. 2009.
Analysis of Modern Technologies Commonly Used in Beef Cattle Production:
Conventional Beef Production Versus Nonconventional Production Using
Meta-Analysis. Journal of Animal Science 87: 3418–3426.
Williams, Bernard. 2006. The Human Prejudice. In Philosophy as a Humanistic
Discipline, edited by A. W. Moore, 135–152. Princeton, NJ: Princeton Univer-
sity Press.
Yarali, A., T. Niewalda, Y. C. Chen, H. Tanimoto, and S. Duerrnagel. 2008. ‘Pain
relief’ Learning in Fruit Flies. Animal Behavior 76: 1173–1185.
Zabala, N. A. and M. A. Gómez. 1991. Morphine Analgesia, Tolerance and
Addiction in the Cricket Pteronemobius sp. (Orthoptera, Insecta). Pharmacol-
ogy Biochemistry and Behavior 40: 887–891.
Zamir, Tzachi. 2004. Veganism. Journal of Social Philosophy 35 (3): 367–379.
Zeis, John. 2013. A Rawlsian Pro-Life Argument Against Vegetarianism. Interna-
tional Philosophical Quarterly 53: 63–71.
Index
Abbate, Cheryl 110 Budolfson, Mark 50, 58, 70, 74n22,

activism 6, 17, 52 – 53, 79, 121, 133, 88n11
141, 148, 158 – 165; agricultural bugs see insects
activism 121; animal activism 6, Buzby, Jean 59 – 60
52 – 53, 141, 148; see also advocacy
Adams, Carol 48n12, 49n24, Callicott, J. Baird 28 – 29
132 – 133 Caramazza, Alfonso 143
advocacy: animal advocacy 53, 79, Carruthers, Peter 21 – 24, 47n2
141, 162; see also activism cattle see cows
agriculture: chicken 12 – 13, 15, 56, causal efficacy 81, 86, 152; see also
61, 63 – 69, 108 – 110, 123, 126, causal impotence; causal inefficacy
152 – 153; cow 9 – 11, 14, 16, causal impotence 73n9; see also causal
65 – 66, 76, 125 – 126, 127n12; inefficacy
pig 11 – 12, 65 – 66; veganic 118, causal inefficacy 76, 87n6, 91, 95, 102,
157n18; see also aquaculture; 136, 140, 151, 153; see also causal
bivalves; chickens; cows; pigs impotence; causal inefficacy problem
Allcorn, Ashley 134 causal inefficacy problem 50 – 72,
Allen, Michael W. 142 72n4, 87n8, 91, 95, 102, 133,
aquaculture 16, 138, 141; bivalves 80, 138, 150 – 151; potential solutions
151; see also agriculture; fish 55 – 63; thresholds 64 – 71; see also
causal impotence; causal inefficacy
Balderrama, N. 81 Cerulli, Tovar 120 – 121
Barratt, B. I. P. 88n17 chickens 2, 12 – 13, 44, 51, 54 – 56, 61,
Barron, Andrew 83 – 84 64 – 65, 67 – 68, 71, 84, 108 – 110,
Bastian, B. 144 123, 126, 127n4, 130, 141 – 142,
Bateson, M. 81 145, 151 – 153, 162, 170 – 171;
Belshaw, Christopher 106 backyard chickens 108 – 110,
Berry, Wendell 29 123, 126, 130, 151 – 153, 162;
Bilewicz, M. 143 consumption of 2, 12, 44, 51,
bivalves 6, 80, 86, 88n10, 113, 117, 54 – 56, 170; see also agriculture
126, 130, 151 – 153, 162, 168, 170; Ciocchetti, Christopher 30
consumption of 6, 86, 88n10, 117, Cochrane, Alasdair 47n9, 106 – 107,
126, 130, 151, 153, 162, 168, 170; 110
see also aquaculture colonialism 46, 133, 139
Bonilla-Silva, Eduardo 37, 38, 41 consequentialism 55, 72n2, 97, 139,
bovine agriculture see agriculture 150, 169; rule consequentialism 97;
Bratanova, B. 144 see also utilitarianism
Bruckner, Donald 77 – 79, 87n8, contractualism 1, 21 – 24, 47n1 – 2,
111 – 112, 151 156n13
190 Index
Cordeiro-Rodrigues, Luis 37 – 40, Hettinger, Ned 27
42 – 46 Hooley, Dan 98 – 99
cows 9 – 11, 14, 16, 24, 55, 62, 65 – 66, Horta, Oscar 19n4
76, 110, 125 – 126, 144 – 145, Hsiao, Timothy 26 – 27
127n12, 151, 170; consumption of Huemer, Michael 136 – 137
24, 55, 62, 110, 125 – 126, 127n12, human exceptionalism 24 – 27
151, 170; see also agriculture hunting 27, 76, 117 – 123, 125 – 126,
Cuneo, Terence 105 – 108, 122, 127n9, 151 – 152; respectful hunting
126n3, 153, 168 119 – 122, 127n9
Curnutt, Jordan 78 Hursthouse, Rosalind 128 – 131
Davis, Steven 76 – 77, 125, 127n12, Imhoff, R. 143

151 insects 6, 80 – 86, 88n10, 88n13,
Degrazia, David 77 – 78, 87 – 88n8, 88n15, 112 – 117, 123, 125 – 126,
88n12 127n7, 130, 151 – 153, 162, 168;
Demetriou, Dan 117 consciousness 81 – 84, 88n13, 113,
deontology 94, 124, 153; Kantianism 115, 125; consumption of 6, 81,
99, 124 84, 88n10, 86, 112, 114, 117, 126,
Diamond, Cora 161 130, 151, 153, 162, 168; farming
doctrine of double effect 123 – 126, 84 – 86, 125, 127n7, 152, 168; see
127n11 – 12 also entomophagy
Doggett, Tyler 92 inverse relationship hypothesis 62
Driver, Julia 95 – 98
Drogosz, M. 143 Jack, A. I. 141
Jorgensen, Matthew 143
ecocentrism 28
ecofeminism 132 – 135, 169 Kagan, Shelly 56
Eisemann, C. H. 82 Kant, Immanuel 21 – 22; see also
Engel, Mylan 135 – 136 deontology
entomophagy 84 – 86, 88n15, 113; see Kasperbauer, T. J. 141
also insects Kazez, Jean 31 – 32
Everett, Jennifer 27 Kemmerer, Lisa 47n4
Klein, Colin 83 – 84
Fairlie, Simon 87n6 Knobe, J. 141
fish 16, 19n4, 53, 138, 140 – 142, 145, Korsgaard, Christine 19n5, 99 – 102
151, 155n8, 170 – 171; consumption
of 151, 170; see also aquaculture Lamey, Andy 57, 124 – 125
Fleischman, Diana 80 Lawford-Smith, Holly 166n4
Foer, Jonathan Safran 30 Lehman, Hugh 104
Francione, Gary 37, 48n15, 127n6 Leonard, Christopher 63
freeganism 51, 77 – 80, 110 – 112, 152 Leopold, Aldo 27, 29
linked oppression thesis 132 – 135
Garrett, Jeremy R. 72n4, 88n8 Lipscomb, Benjamin 29
George, Kathryn Paxton 35 – 38, 46, Lomasky, Loren 8, 17, 31
49n16, 126n3 Loughnan, S. 144
Gill, Michael 31 Lusk, Jayson 54 – 55, 61 – 62, 168
Gomez, M. A. 81
MacClellan, Joel 86 – 87n1
Halteman, Matthew C. 59 – 60 Machery, E. 141
Harman, Elizabeth 165 – 166n1 Marder, Michael 72n1
HASK practice 135 Matheny, Gaverick 77
Haslam, N. 144 McCloskey, Deirdre 45
Herzog, Hal 113, 140 McCulloch, S. 141
Index 191
McMullen, Steven 59 – 60 98 – 99; Julia Driver 95 – 98;
McPherson, Tristram 94 – 95 Nathan Nobis 98 – 99; Tom Regan
Menhinick, Edward F. 88n17 89 – 92; Tristram McPherson
Merker, Björn 83 – 84 94 – 95
Meyers, C. D. 84 roadkill 51, 77 – 80, 87 – 88n8, 99,
Milburn, Josh 49n19, 108, 110 110 – 112, 130, 151, 153, 168
Mill, John Stuart 31 Robbins, P. 141
mussels see bivalves Rothgerber, Hank 52, 134
Rozin, P. 52
Nefsky, Julia 164 Rubio, Julie 47n7
Neumann, D. L. 142 Rust, Joshua 146
Nobis, Nathan 98 – 99 Ryder, Richard 24
Norcross, Alastair 24, 55 – 56, 61, 80
Norwood, Bailey 54 – 55, 61 Sabrosky, C. W. 88n17
Salt, Henry 87n4
Ogletree, Shirley Matile 134 scavenging 77 – 78, 86
O’Neill, Patricia 143 Schwitzgebel, Eric 146
Order Threshold Solution 55 – 61, 64, Sebo, Jeff 116, 127n8
67, 70 – 71 Sethu, Harish 53, 55
oysters see bivalves Shallice, T. 142
Shelton, J. R. 143
Pachirat, Timothy 14 – 15 signaling 79, 163 – 165, 166n6,
Pearse, A. S. 88n17 170 – 171
Petrinovich, Lewis 143 Singer, Peter 22, 24 – 25, 56, 76, 78,
Phillips, C. J. C. 141 – 142 147, 161, 163; animal liberation
pigs 11 – 13, 15, 55, 65 – 66, 84, 22; speciesism 24, 161
135 – 136, 142, 144 – 145, 155n10, speciesism 24 – 25, 79, 87n5, 101,
166n6, 170; consumption of 55, 112, 149, 161
155n10, 166n6, 170; see also Stoess, C. 52
agriculture Swanson, Jennifer 23
Plous, S. 142 Sytsma, J. 141
Posewitz, Jim 119
precautionary argument 88n10, Tännsjö, Torbjörn 19n5, 168
113 – 117 Tomasik, Brian 118
precautionary principle see
precautionary argument utilitarianism 1, 50 – 72, 75 – 86, 89,
Prinz, J. 141 91, 104, 112, 128, 130, 168;
production consumption gap 51, 53, act utilitarianism 1, 50, 52, 62 – 63,
89 – 104, 128, 140, 151, 153, 168 70 – 72, 75 – 76; rule
utilitarianism 75, 130; see also
QUALY 68, 70 consequentialism
Quinn, Warren 123 – 124
virtue ethics see virtue theory
Rachels, Stuart 78, 138 virtue theory 1, 128 – 132, 154n3,
racism 35 – 42, 44, 49n24, 133 169; virtue consequentialism 130,
Regan, Tom 22, 89 – 92, 98, 103n1, 104, 154n3
112, 117, 147, 161; animal liberation
22; rights theory 89 – 92, 117 Warrington, E. K. 142
Replaceability Argument 76, 87n4 – 5, welfare threshold 64 – 70
168 Westbury, H. R. 142
rights theory 1, 89 – 103, 105 – 126,
128 – 129, 168; Christine Zabala, N. A. 81
Korsgaard 99 – 101; Dan Hooley Zeis, John 47n2

(Routledge Research in Applied Ethics 16.) Fischer, Bob - The Ethics of Eating Animals - Usually Bad, Sometimes Wrong, Often Permissible (2020, Routledge)

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

(Routledge Research in Applied Ethics 16.) Fischer, Bob - The Ethics of Eating Animals - Usually Bad, Sometimes Wrong, Often Permissible (2020, Routledge)

Uploaded by

Copyright:

Available Formats

The Ethics of Eating Animals

Intensive animal agriculture wrongs many, many animals. Philosophers

Bob Fischer teaches philosophy at Texas State University. He’s the

The Injustice of Punishment

Friendship, Robots, and Social Media

The Capability Approach in Practice

The Ethics of Counterterrorism

Disability with Dignity

Media Ethics, Free Speech, and the Requirements of Democracy

Ethics and Chronic Illness

The Future of Work, Technology, and Basic Income

The Ethics of Eating Animals

For more information about this series, please visit: www.routledge.com/

3 Bad Arguments for Eating Animals 20

4 Utilitarianism and the Causal Inefficacy Problem 50

5 Causal Inefficacy Aside, Utilitarianism Requires

6 The Rights View and the Production/Consumption Gap 89

7 Eating Animals the Rights Way 104

8 Beyond Utilitarianism and the Rights View 128

9 Activist Ethics 158

10 Taking Stock 167

2.1 Low-Level Stress, Acute Pain

2.1.5 Taking Stock

Is some factory-farming unduly brutal and morally corrupt?

2.2 What Follows?

1. It’s wrong for producers to bring these animals into existence.

In this chapter, I want to do some ground clearing. My goal here is to

we have no immediate duties to animals; our duties towards them

This sort of problem is going to affect any form of contractualism that

3.2 Human Exceptionalism

1. Moral welfare interests trump non-moral welfare interests.

3.3 Environmentalism and Agrarianism

paradigmatically monistic (duties and obligations are generated by

According to Callicott, the communities in which we’re most deeply

According to Berry, we do have obligations to animals—lots of them,

Imagine a sequence of vegan diets that are a person’s only alternative

there’s something a bit dubious about [the person willing to give up

3.6 Ageism, Classism, Sexism, Racism

In arguing for vegetarian or vegan diets on traditional moral grounds,

Obviously enough, George is claiming that it’s permissible for mem-

When nearly everyone on Earth, except the most privileged class of

The upshot is as follows. Typically, people argue for veganism by claim-

3.6.1 Racism Without Racists

racism is above anything, about practices and behaviors that produce

3.6.2 Limited Access

Low-income census tracts20 where a significant number (at least 500

And if we accept this usage, then roughly 54.4 million people (17.7% of

3.6.3 The Neocolonial Economic System

4.1 The Simple Act Utilitarian Argument for Veganism

1. You ought to act so as to maximize expected utility.

Granted, you aren’t obligated to be a strict vegan. However, it’s very

There are several points to note here.

4.1.1 Causal Inefficacy, a First Pass

To determine the number of animals saved by a vegetarian [in the

Total number of animals killed

He then provides some estimates of those numbers—which is surpris-

Similarly, consider the work of two agricultural economists, Bailey Nor-

It is tempting to think that your food choices have no impact on the

If you give up. . . Total production Per capita consumption of

One pound of beef 0.68 lbs 65.20 lbs

difference. In other words, we can’t immediately infer either the actual

4.1.2 The Order Threshold Solution

Suppose that the industry is sensitive to a reduction in demand for

3 Bad Arguments for Eating Animals 20

4 Utilitarianism and the Causal Inefficacy Problem 50

6 The Rights View and the Production/Consumption Gap 89

7 Eating Animals the Rights Way 104

8 Beyond Utilitarianism and the Rights View 128

9 Activist Ethics 158

10 Taking Stock 167

4.1.3 The Main Problem With the Order