The Argument from Species Overlap
by Jesse Ehnert
This thesis is submitted to the faculty of the
Virginia Polytechnic Institute and State University
in partial fulfillment of the requirements for the degree of
Master of Arts
in
Philosophy
Harlan B. Miller, Chair
William FitzPatrick
James C. Klagge
July 15, 2002
Blacksburg, Virginia
Keywords: ethics, animals, nonhuman animals
Copyright 2002, Jesse EhnertThe Argument from Species Overlap
by Jesse Ehnert
ABSTRACT
The ‘argument from species overlap’ (abbreviated ASO) claims that some human
and nonhuman animals possess similar sets of morally relevant characteristics, and are
therefore similarly morally significant. The argument stands as a general challenge to
moral theories, because many theories hold that all humans possess greater moral
significance than all nonhuman animals. In this thesis I discuss responses to the ASO,
primarily those of Peter Carruthers, Tom Regan, Evelyn Pluhar, and Peter Singer.
Carruthers denies the conclusion of the ASO, while the other three do not. I argue that
the ASO is a sound argument, and that Carruthers’s attempts to counter it via his
contractualist theory are unsuccessful. I next discuss the rights-based theories of Regan
and Pluhar, which agree with the conclusion of the ASO but which, I believe, encounter
significant theoretical difficulties. Finally, I address the ASO from a utilitarian
perspective, first from Singer’s utilitarian formulation and then from a ‘welfareutilitarian’ formulation. I answer a number of critical objections to welfare utilitarianism,
and argue that the theory is most successful in facing the challenge of the ASO.iii
To my family.iv
ACKNOWLEDGEMENTS
I am indebted to many people for helping make this work possible. I thank my
thesis committee members for their support, advice, and patience. My adviser, Dr.
Harlan B. Miller, deserves substantial credit for leading me to this project and for
providing invaluable assistance as I made my first steps into the subject matter. Dr.
William FitzPatrick and Dr. James C. Klagge each provided important insights into my
work, revealing to me some of the more serious challenges to my arguments.
Many thanks go to the entire Virginia Tech philosophy department, the students,
faculty, and administration. It has been an extraordinary pleasure to work with such a
thoughtful, dedicated group of individuals.
I must also express my sincere appreciation to many of my friends and family
members who were surprisingly willing to listen to and respond to my seemingly endless
philosophical ramblings. Among these are my former classmates Bryan Baltzly, Seth
Fairbanks, Zane Rogers, and Jason Rosencrantz, my brothers Brian and Terence, my
mother Sherry and stepfather Al, and friends Jesse Fuchs, Elizabeth Owens, and Jeff
Stern.
Finally, I would like to acknowledge Dr. Patrick Croskery, formerly of Virginia
Tech, in whose class I discovered my love of moral philosophy.v
TABLE OF CONTENTS
ACKNOWLEDGEMENTS .............................................................................................iv
CHAPTER 1: INTRODUCTION.....................................................................................1
PRELIMINARIES .................................................................................................................4
CHAPTER 2: THE CONTRACTUALIST RESPONSE ...............................................7
VARIOUS OBJECTIONS TO THE ASO..................................................................................7
CARRUTHERS ..................................................................................................................11
CONTRACTUALISM AND ANIMALS........................................................................................................ 12
CONTRACTUALISM AND HUMANS......................................................................................................... 17
SUMMARY.................................................................................................................................................. 23
CHAPTER 3: THE RIGHTS RESPONSE ...................................................................24
REGAN ............................................................................................................................24
REGAN’S THEORETICAL METHOD ......................................................................................................... 27
REGAN AND THE ASO............................................................................................................................... 33
PLUHAR ..........................................................................................................................35
PLUHAR’S THEORETICAL METHOD....................................................................................................... 35
PLUHAR AND THE ASO............................................................................................................................. 37
CHAPTER 4: THE UTILITARIAN RESPONSE........................................................44
UTILITARIANISM’S THEORETICAL FOUNDATION.............................................................44
UTILITARIANISM AND THE ASO......................................................................................49
VARIETIES OF UTILITARIANISM ......................................................................................50
UTILITY....................................................................................................................................................... 50
MAXIMIZATION......................................................................................................................................... 55
CRITICISMS OF UTILITARIANISM.....................................................................................56
STRICT DECISION MODEL........................................................................................................................ 57
FAILURE TO PROVIDE ADEQUATE PROTECTION, PART ONE............................................................. 59
FAILURE TO PROVIDE ADEQUATE PROTECTION, PART TWO............................................................ 63
REPLACEABILITY...................................................................................................................................... 69
WELFARE UTILITARIANISM AND A FINAL RESPONSE TO THE ASO .................................77
CONCLUSION ..................................................................................................................79
BIBLIOGRAPHY............................................................................................................81
VITA .................................................................................................................................821
“I thought, ‘Oh my God, it's like eating my niece.’”
-- Cameron Diaz on why she stopped snacking on bacon after she was told that pigs have
the same mental capacity as a 3-year-old, in Esquire.
1
CHAPTER 1: INTRODUCTION
A typical method for criticizing any moral theory is to test its adequacy in
handling difficult, unusual, or extreme cases. After all, moral theories tend to agree about
the basics, what might be termed ‘commonsense morality’. Typical elements of
commonsense morality include presumptions against lying, harming, and killing, as well
as presumptions in favor of their opposites: honesty, benevolence, and the prevention of
death. A theory that contradicts too many of our common beliefs is a highly questionable
theory, just as any scientific theory would be likewise suspect if it failed to account for
what we saw in the world around us. Despite the fact that different theories of morality
agree on many issues, they can and often do differ greatly about what they consider the
ultimate grounding of moral value. These differences produce widely varying results
when the theories are applied to cases for which they were not designed, cases that are
often difficult, unusual, or extreme. If one theory more adequately handles those special
cases, while simultaneously explaining commonsense morality, then we have reason to
prefer that theory to the others: the others have been shown to be superficially successful,
but not ultimately successful in all cases.
In the ongoing debate regarding the moral status of nonhuman animals (hereafter
usually shortened to ‘animals’), philosophers have found themselves in exactly the sort of
situation I described, one where differences in moral theories produce radically different
positions on the issue. The rival theories all provide some justification for the (very high)
moral status of most or all human beings, but arrive at a variety of conclusions when
applied to the subject of animals. Proponents on one side of the debate attempt to prove
that at least some animals deserve consideration equal to that all human beings. Others
argue just the opposite, that animals have no moral worth whatsoever. Of course, there
are plenty of positions that propose some middle ground, holding that animals have some
moral value, but not to the extent that humans beings do. As long as each theory is
1
Bill Zehme, “Cameron Diaz Loves You.” Esquire April 2002: 78. Quoted from Salon.com on the World
Wide Web (http://www.salon.com/people/col/reit/2002/03/12/nptues).2
generally acceptable, the debate ends here, at a stalemate. What is needed to move
forward is a focusing issue for the debate, a narrower point within the debate that forces
those on each side to scrutinize their own theories. Enter the ‘argument from species
overlap’.
The argument from species overlap—or ASO, as I will abbreviate it—has played
a major role in the debate because it provides a specific challenge for all moral theories.
2
As I will show, the ASO is a very effective tool for evaluating those generally acceptable
rival theories, and revealing some to be, in fact, unacceptable. The following is the ASO
in a very general form. This rendering of the ASO is based on the form used by Evelyn
Pluhar in her book Beyond Prejudice.
3
1. Individuals who possess similar sets of morally relevant characteristics are
similarly inherently morally valuable.
2. Some nonhuman animals possess sets of morally relevant characteristics
similar to those of some humans.
3. Therefore, some nonhuman animals and some humans are similarly inherently
morally valuable.
The first premise of the ASO is nearly tautological: if two individuals share all
characteristics upon which their moral value is founded, then they will have equivalent
moral value. The contrapositive is even more obvious: if individuals do not share
equivalent moral value, then the individuals cannot be identical in terms of morally
relevant properties.
The second premise is more controversial. The claim is that whatever those
properties are that determine the moral worth of an individual, there are some humans
who possess those properties only to the extent that some nonhumans do. For example,
many theories hold that the possession of a sufficiently rich mental life is the determining
property. If so, then consider the fact that there are anencephalic humans born with
almost no brain, but only a brain stem.
4
This human is utterly without consciousness, and
therefore has a far simpler mental life than a great number of animals. There are many
2
Although the argument is typically labeled the ‘argument from marginal cases’, I prefer this alternate
wording, originally proposed by Harlan B. Miller (“A Terminological Proposal” in SSEA Newsletter 30
(March 2002)), as it is both more descriptive and less likely to be misinterpreted and cause offense.
3
Evelyn Pluhar, Beyond Prejudice: The Moral Significance of Human and Nonhuman Animals (Durham
and London: Duke University Press, 1995), 65. In my rendering of the argument I have chosen to omit
Pluhar’s expression “maximally morally significant,” in order to reach a more general conclusion. The
argument loses none of its force in the process.
4
Ibid., 8.3
other humans who are far more fortunate, but who are nonetheless mentally undeveloped
(such as a newborn infant) or defective. We can imagine a continuum on which human
beings exist, ranging from those with no mental life whatsoever, on up to the epitome of a
full human mental life. At many points along that continuum, we find not only humans,
but nonhumans as well. Therefore, there are some humans who, in terms of possession of
a mental life, are similar to some nonhumans.
5
If possession of a mental life constitutes a
complete set of morally relevant characteristics, then the second premise is true. If
indeed the second premise is true, and given the relatively unproblematic first premise,
then the conclusion follows. Those humans and nonhumans who possess sets of morally
relevant characteristics are similarly morally valuable.
If the ASO is a sound argument, then the proposition that all humans possess
greater moral value than all animals is false. If the proposition is false, then any theory
that produces this proposition is flawed. All other things being equal, we ought to reject
those flawed theories and favor theories that successfully explain the conclusion of the
ASO. On the other hand, if the ASO is shown unsound, then we should prefer those
theories that explain the flaw in the ASO. It is in this manner that the argument serves
the purpose I spelled out in my opening paragraph.
I will discuss four principal moral theories that have been applied to the ASO in
this manner. They are: Peter Carruthers’s contract theory, the rights-based theories of
Tom Regan and Evelyn Pluhar, and the utilitarian theory of Peter Singer. I critique each
candidate theory in turn. What I hope to show is that, while Singer’s brand of
utilitarianism runs into significant challenges, a certain sort of utilitarian theory, ‘welfare
utilitarianism’, is ultimately the most successful at handling the demands of the ASO.
In this paper I will specifically focus on how each theory handles the morality of
the death and killing of animals and humans. I choose this particular focus for a number
of reasons. First, applying a theory to the subject of death is a clear and easy way of
illustrating the relative moral worth of different individuals. Second, the morality of
killing provides some of the most difficult cases for a utilitarian to handle. As I will be
5
This notion of a continuum is an oversimplification, and does not capture the complex dimensions of
mental abilities, but I think it nonetheless indicates something true about relative capacities between
individuals.4
defending a utilitarian theory, it serves my purpose to address the more difficult
challenges to my own theory.
Preliminaries
There are a number of facts that play a role in each of the four positions discussed
in this paper. These facts are important because they underlie the discussion, but they are
not at the focus of the debate. For that reason, I will enumerate them here and henceforth
consider them presupposed in the debate.
First, there is the fact that many animals are sentient. It is a commonsense notion
that ‘higher’ animals such as dogs and cats are conscious and can experience pleasure and
pain. It is also likely that many birds, fish, reptiles, and amphibians also possess
consciousness, though animals with simpler nervous systems may not be conscious to the
same extent.
Peter Carruthers proposes an argument against animal consciousness in his The
Animals Issue.
6
Conscious mental experiences, according to this argument, are those that
can be the object of second-order beliefs. Animals that do have thoughts—mental
experiences, desires, and beliefs—but do not possess second-order beliefs about those
thoughts, are not conscious, and therefore their pleasure and suffering cannot matter
morally. Carruthers argues that no animals, not even (nonhuman) primates, possess
second-order beliefs. Therefore, the pleasure and suffering of any and all animals cannot
matter morally.
Enough has been said in response to Carruthers’s position and similar arguments
that I have nothing to add to that debate. One example is David DeGrazia’s Taking
Animals Seriously, which explores animal minds in depth, and presents strong evidence
in favor of animal consciousness not only in primates but most or all vertebrates.
DeGrazia also responds directly to Carruthers’s argument.
7
Pluhar has a detailed
objection to Carruthers as well.
8
Carruthers is aware that his argument is controversial,
and is himself willing to put the debate aside and assume animals are conscious until
6
Peter Carruthers, The Animals Issue (Cambridge: Cambridge University Press, 1992), chapter 8.
7
David DeGrazia, Taking Animals Seriously (Cambridge, UK: Cambridge University Press, 1996), 112-
115.
8
Pluhar, 37-46.5
proven otherwise: “Until something like a consensus emerges, amongst philosophers and
psychologists concerning the nature of consciousness, and amongst ethologists over the
cognitive power of animals, it may be wiser to continue to respond to animals as if their
mental states were conscious ones.”
9
A second presupposition is the fact that some humans are not ‘full persons’,
rational agents possessing a rich mental life. Some humans are not agents at all: they
have no preferences or goals whatsoever. Fetuses, at least up to some point of
development, fall into this group, as do victims of comas or severe brain damage. The
ASO, of course, depends upon the existence of these atypical humans to draw its
conclusion about animals who similarly fail to qualify as full persons. The existence of
atypical humans is not in itself a moral claim, but moral theories often propose the moral
relevance of some set of mental properties. The fact that not all humans possess these
properties is what gives the ASO its moral force.
A third presupposition is the fact that the set of moral agents and the set of moral
patients are not necessarily identical. I am following Harlan B. Miller in my use of the
terms ‘moral agent’ and ‘moral patient’: “To be a moral agent is to be an entity capable
of actions that may appropriately be evaluated as right or wrong. To be a moral patient is
to be an entity of such a sort that what is done to that entity by a moral agent is per se,
subject to moral evaluation.”
10
It may be the case that all moral agents are moral patients,
and vice-versa, but this case must be argued for; it is not the case by definition. (Miller,
incidentally, holds that the two sets differ significantly: on his view, the set of moral
patients includes human infants and children, who are not moral agents, while the set of
moral agents includes abstract entities such as corporations, which are not moral
patients.)
11
This distinction is important, because my use of ‘moral value’, ‘morally
significant’, ‘the moral community’ and similar expressions regarding individuals
throughout this essay indicate membership in the set of moral patients, and not
necessarily the set of moral agents. My point here is not to depart from the sense of these
9
Carruthers, 192-193.
10
Harlan B. Miller, “Introduction: ‘Platonists’ and ‘Aristotelians’,” in Ethics and Animals, ed. Harlan B.
Miller and William H. Williams (Clifton, NJ: Humana Press, 1983), 12-13.
11
Ibid., 13.6
expressions as they appear in Pluhar’s discussion of the ASO; rather, I am clarifying the
sense she must intend.
A fourth presupposition is that the commonsense idea that at least some humans
are more morally valuable than some animals reflects something true about morality.
Each of the four philosophers’ theories arrives at this claim by one means or another.
The real disagreement between them concerns which humans are more valuable than
which animals, and to what extent. Their disagreement on these points can be understood
in terms of their answer to the ASO, and is the essential issue of this thesis.
Finally, a brief note about how moral significance and consideration is
understood. There are two categories I will discuss, absolutism and gradualism. An
absolutist morality is one that divides the world into two groups, those individuals that
matter morally, and those that do not. To matter morally is to matter maximally; there is
no hierarchy of moral value, and no individual matters more than any other. A gradualist
morality is one that begins with the absolutist division, but further divides the group of
individuals that matter so that some individuals are ‘higher up’ the moral ladder than
others. There are numerous varieties of gradualism, as there are many ways one can spell
out the hierarchical scheme. For instance, one might consider the relative values of
individuals’ lives, insofar as their deaths are morally important, to lie along a gradualist
spectrum, but view pleasure and suffering as counting equally across all individuals.
Regan, for one, proposes just this view. Sophisticated theories can and often do contain
both absolutist and gradualist elements. For instance, a gradualist theory can propose a
‘top’ of the hierarchy, or an imaginary line above which individuals possess maximal
moral value, providing them with a special form of consideration not granted to
individuals ‘below’ the line. As we shall find, a moral theory’s absolutist or gradualist
elements play an important role in the theory’s handling of the ASO.7
CHAPTER 2: THE CONTRACTUALIST RESPONSE
Various Objections to the ASO
Of the major positions on the ASO that I am considering, only the contractualist
view argued by Carruthers flatly rejects its conclusion. However, there are many other
philosophers who argue against the ASO. The reason I present Carruthers as the lone
opponent of the ASO is primarily because the arguments of other opponents have already
received a thorough response by Evelyn Pluhar,
12
and there is little that I can add. In
addition, while Pluhar does defend the ASO against Carruthers’s objections, I believe that
his position remains important to consider. As part of my criticism of Carruthers’s case, I
will be making additional arguments that Pluhar didn’t provide. Also, Carruthers makes
interesting points regarding both Regan and Singer, several of which I incorporate into
my discussion of their theories.
As a prelude to a discussion of Carruthers’s position, it will be helpful to review
the methods philosophers have used to oppose the ASO. Pluhar provides a
comprehensive list of these objections in her book Beyond Prejudice. They can be
classified into two categories. Objections in the first category attempt to reject the
conclusion of the ASO—that some animals and some humans are similarly morally
valuable—without clearly and directly opposing either premise. Typically, these
objections can be dismissed by revealing an underlying misunderstanding of the ASO on
the part of the objector.
For example, one objection to the ASO is that the conclusion could be seen as a
justification for lowering the value we place on humans who are not full persons. Pluhar
responds that, of course, that is one possible outcome. It is exactly that possibility that
serves as a challenge to those who would deny consideration to animals.
13
Others misunderstand the second premise of the argument, and claim that it does
not accurately compare the abilities of atypical humans to the abilities of animals.
Depending on which philosopher is making the objection, the comparison is found to be
12
Pluhar, 67-107, 140-178.
13
Ibid., 72.8
insulting either to atypical humans or to animals.
14
Still others misunderstand what is
meant by the ‘moral value’ of an individual, and mistake it to mean moral agency.
15
Pluhar exposes each misunderstanding in turn, and in doing so does away with each of
these objections.
Objections in the second category deny the second premise of the ASO: that some
nonhuman animals possess sets of morally relevant characteristics similar to those of
some humans. You will recall that these morally relevant characteristics are typically
agreed to be one or more mental features, such as consciousness, a rich mental life, or
perhaps rational agency. Now, for every such mental feature, there is some animal who
possesses it to at least as great an extent as does some human. Therefore, in order to deny
premise 2, the objector must add another item to the list of morally relevant features.
This item ends up being species itself.
Defenders of the ASO often label their opponents ‘speciesists’, a term meant to
carry the same kind of moral disapproval as ‘racists’ or ‘sexists’. Racists and sexists
discriminate against certain individuals based on characteristics that are not morally
relevant. Similarly, a speciesist is supposedly discriminating against individuals based on
another morally irrelevant feature, one’s species. However, if species can be shown to be
morally relevant, then discrimination based on species would be justified, and the
speciesist label would lose its bite.
Most people would agree that, species aside, if an individual was capable of
rational thought or had some sort of mental life at the level of normal humans, then that
individual would be as morally valuable as humans. For this reason, a commonly
accepted formulation of the speciesist position is that, to receive moral consideration
equal to a normal human, one must either a) possess a certain set of mental properties, or
b) be a member of a species characterized by those mental properties. This formulation
clearly refutes the second premise of the ASO by drawing the conclusion that all humans
possess a morally relevant characteristic not shared by animals, that characteristic being
either of a) or b), above.
14
Ibid., 77-85.
15
Ibid., 74-77.9
The burden of proof in this matter is on the opponent of the ASO; it must be
shown that species membership is indeed morally relevant. However, as Pluhar reveals,
no one has yet produced a satisfactory case. There have been extremely unsatisfactory
attempts which I will not bother to go into here.
16
Of the better arguments, the common
tactic is to show some important link between the atypical human and the mental
properties that are characteristic of the human species.
One such attempt is to refer to the individual’s potential.
17
Unlike nonhuman
animals, humans who are not full persons at least had the potential to possess a rich
mental life. According to this line of thought, because animals never even had the
potential for such a mental life, they are not candidates for consideration the same way
atypical humans are. Pluhar indicates the main problems with this move. First, having
the potential for possessing something is logically distinct from actually possessing it. A
potential employee receives no paychecks; a potential threat is not necessarily a threat. It
is not enough to point to potential; what needs to be shown is that there is something
morally relevant about the possession of the human’s potential full personhood.
Moreover, many humans have lost this potential. Therefore, it must also be shown that
past possession of potential is morally relevant as well. Finally, some humans cannot
even be said to have had potential. In extreme cases such as that of an anencephalic
human, the individual at no point in time was a potential full person, and therefore gains
no benefit from arguments from potential.
Arguments from potential are problematic, and in some cases utterly useless,
when used to defend the moral significance of atypical humans. Another tactic, one that
avoids mention of potential, instead refers strictly to a relationship between the individual
and the normal functioning of members of the individual’s species. I will refer to this as
the telos argument. Pluhar finds philosophers with vastly different views on the moral
status of animals—these include Joel Feinberg, Bernard Rollin, Alan Holland, and A. I.
Melden—who make some version of telos argument.
18
One argument by Rollin, for
16
Ibid., 140-146, 162-171. Arguments found in the second group of pages cited (appeals to kinship and
opportunities for interaction) are actually considered the most plausible by Pluhar, but I disagree. To say
the least, they are utterly unsatisfactory because they fail to confer a higher value on all humans than on
any nonhumans, which is the goal of the speciesist.
17
Ibid., 146-150.
18
Ibid., 150-162.10
example, says that we ought to respect the telos of an individual’s species regardless of
what species it is.
19
Rollin does not make a distinction between humans and other
animals, but others do make the distinction, based on the fact that the human species,
unlike the other species we know of, has full personhood as part of its telos. Because
they view full personhood as the initial source of moral value, and because the human
species uniquely (as far as we know) has full personhood as part of its telos, all humans
deserve consideration above all animals.
There are as many different varieties of the telos argument as there are
philosophers who utilize it, and Pluhar addresses each one on its own terms. While some
of the minor objections she raises against specific arguments do not seem entirely
justified,
20
her arguments successfully reveal the major flaws of all such attempts to
distinguish humans from other animals. First, the telos argument claims that there is
some sort of morally relevant relationship between individuals and the proper functioning
of its species. But, as Rollins pointed out, this relationship, if it exists, exists both for
humans and for nonhumans. When opponents of the ASO argue that the relationship is
different for humans on account of their species, they are begging the question. They
have inserted their conclusion—that species-membership is morally relevant—as a
premise.
21
Additionally, there is the question of what constitutes the ‘norm’ of a species.
22
If
one determines the telos of a normal member of a species by taking a statistical sample of
the members of a species, and finding the average condition, one sees some very bizarre
results. For instance, Pluhar suggests a possible future world where few humans are full
persons.
23
If the species were changed in this way, then non-full persons born at that
point in time would not benefit from the telos argument. But this means that two
individuals who are identical in all ways—even species membership—possess very
different moral value, and merely because of the time at which they were born.
19
Ibid., 154. The argument is summarized by Pluhar.
20
See for example ibid., 153: Pluhar argues that an individual with no moral value cannot suffer nonmoral
evil. It is not entirely clear what she means by ‘nonmoral evil’, other than ‘harm’. If this is what she
means, I do not see why other individuals, such as plants or manmade machines, could not be the objects of
nonmoral evil.
21
Ibid.,, 154-159.
22
Ibid., 159-161.
23
Ibid., 160.11
Alternatively, one could determine the telos of a species to be represented by the most
successful conditions (in terms of full personhood) found within the species, so that if at
least one member of a species achieves full personhood, then full personhood is the norm.
But this method runs into a similar problem: as soon as one member of a species is found
to be a full person, all members of that species gain moral significance. If we try to avoid
these problems by rejecting the notion that the norm of a species is based on empirical,
statistical facts, then it is unclear how one would go about determining what the norm is.
I see no recourse aside from some kind of metaphysical proof, and no such proof has, to
my knowledge, been made.
We now turn to Carruthers who launches an impressive attack on the ASO, but
whose arguments, like the others Pluhar addresses, ultimately fail.
Carruthers
Few philosophers have opposed the ASO as directly and thoroughly as Peter
Carruthers. In his book, The Animals Issue, he argues that indeed we ought to make a
moral distinction between all humans and all animals, and that this distinction is nonarbitrary. However, unlike the attempts I have described above, Carruthers does not
exactly argue that species membership is a morally relevant feature. He agrees that the
second premise of the ASO is essentially true: both animals and some humans lack
rational agency, which Carruthers argues is the fundamental source of moral value. But
his moral principles are constructed in such a way that moral value can appear even
where rational agency does not.
Empirical facts—facts about psychology in particular—play an important role in
Carruthers’s construction of principles regarding our treatment of non-full persons.
While these facts lead him to grant a certain kind of moral value to animals, they lead
him to find a much higher degree of moral value in atypical humans. In this way,
Carruthers denies the conclusion of the ASO.
Carruthers begins by comparing his Rawlsian contractualist theory to Peter
Singer’s utilitarianism and Tom Regan’s rights view. Arguing that the utilitarian and
rights theories are too problematic to be acceptable, he goes on to show why
contractualism does not provide any source of rights for animals. Animals, in his view, 12
possess no real moral value, but only extrinsic value. We may seem to have duties
toward animals, but these duties are derived either from respect for pet owners and
animal lovers, or from a duty to our own virtuous characters. On the other hand, all
humans have equal and maximal inherent moral value. Carruthers argues that this moral
distinction between species, while not a fundamental principle of contractualism, does
indeed follow from the theory, and is non-arbitrary. He concludes his book with a social
criticism regarding the contemporary animal rights movement. He says that the recent
increase of concern for animals, both inside and outside of philosophy, is based on a
“thoroughly misguided” morality.
24
Not only is the movement misguided, it is morally
reprehensible.
25
As I will argue, Carruthers does not successfully refute the ASO. His arguments
granting even extrinsic value to animals fall short, and his attempts to bring all humans
up to equal, maximal moral value are even more problematic.
CONTRACTUALISM AND ANIMALS
Carruthers proposes a contractualist theory much like that of John Rawls, but
extends the application of the theory beyond Rawls’s primary goal of determining
societal structures and institutions.
26
According to Carruthers, the theory can be applied
to the whole of morality. Carruthers’s contractualism is based on the notion of an
imaginary contract between all rational agents.
27
The contract is hypothetical, a
construction founded on an imaginary ideal: the theory generates a set of principles that
we would agree to if we were perfectly rational. These moral principles are chosen in
light of ‘broadly self-interested desires’, by which Carruthers means those desires that we
would have regardless of the particular natures of our individual lives. Because these
desires would primarily regard freedom, power and the like, a fundamental moral
principle is respect for autonomy.
28
After all, given that the imaginary contractors do not
know the particularities of their lives while agreeing to the moral principles, they would
value their rational agency above all else. This fundamental principle would handle
24
Carruthers, 196.
25
Ibid., 168-169.
26
Ibid., 37.
27
Ibid., 35-38.
28
Ibid., 40.13
much if not all of our negative duties to one another. Many other principles might be
agreed upon as well: for example, the perfectly rational agents would agree to principles
of beneficence, in order that they could rely on aid when in need.
The contractualist excludes animals from consideration. The reason is clear, says
Carruthers: animals are not rational agents.
29
Since only rational agents choose the moral
principles, and since they do so out of broad self-interest, only rational agents are granted
moral consideration. And since human beings are the only rational agents in the world so
far as we know, only human beings receive moral consideration. An additional claim is
that consideration is given to all human beings; this claim I will address later.
At this point, I will explain what kind of duties to animals come out of the
contractualist theory. Since there can be no direct duties to animals—for this would
entail that they deserve consideration—they can only be the object of indirect duties.
Carruthers claims that there are three possible indirect duties. The first relates to property
rights.
30
Since some animals have human (rational agent) owners, others would have the
indirect duty not to damage the animal, just as they have a duty not to damage a person’s
car, or pet rock. This duty is quite weak, of course. This provides no protection to
‘unowned’ animals, and does not even protect the owned animals from their owner
(After all, I have the right to slash my own tires. Slashing my cat is no different insofar
as this duty pertains.) Carruthers calls the second indirect duty a duty of ‘legitimate
public concern’.
31
Since there are rational agents in the world who sympathize with
animals, there should be general rules protecting their (that is, the rational agents’)
interests. These rules would not be terribly strong; they would only be as strong as rules
protecting similar human interests, such as those concerning public decency or the
preservation of historical buildings. And, like those other rules, they would have little
control over what is done in private, to one’s own possessions. I can do what I like in the
privacy of my own home, decent or not; the ‘public concern’ duty only demands that I
not negatively affect those people who are more sensitive. So, in terms of the animals,
they are protected in public, but not in private, by both the first and second types of
indirect duties.
29
Ibid., 98-99.
30
Ibid., 105-106.
31
Ibid., 106-107.14
These two duties fail to do credit to our everyday moral sensibilities, a fact
Carruthers admits.
32
After all, despite whatever pain we might feel for the pet owner
when the pet is injured, it is not that the wrongness of injuring animals is based on
whether or not the animals are owned. The difference between our typical moral
evaluation of the kicker of a domesticated dog and that of the kicker of a wild dog is
slight, if a difference is there at all. Moreover, there is little if any moral difference
between a public display of such violence and a similar private indulgence. Therefore,
Carruthers must rely on his third indirect duty, the duty ‘to develop and maintain a
virtuous character,’ to do the bulk of the work.
33
In addition to the principles so far
mentioned, principles directed at a virtuous character can be derived from contractualism.
While the imaginary contractors are perfectly rational, human beings are not. Therefore,
while a perfectly rational agent could accurately apply moral rules to every particular
situation that she found herself in, it is a practical necessity for imperfectly rational
agents that they cultivate virtuous dispositions. With the aid of a well-developed,
virtuous character, an agent’s actions will likely approximate the principles of the
perfectly rational agent. Because right actions will be more likely to occur, the imaginary
contractors would agree to a duty to one’s character. All well and good; so, how does
this apply to animals? Carruthers argues that to intentionally injure an animal, regardless
of whether the animal is domesticated, and regardless of whether the act is done publicly
or privately, is a demonstration of one’s cruelty and indifference. Such vices violate this
third type of duty, and so are against the contractualist account of morality.
This is a very strong argument for the contractualist to use, because it not only
escapes the problem of being too narrow, evident in the first two duty-types, but it also
introduces some notion of giving animals direct consideration. It would seem that to do
so would violate contractualism’s first principle about who counts. But it doesn’t.
Strictly speaking, direct consideration is extended only to rational agents. However, the
nature of virtue is such that the exercise of it brings about a sort of secondary direct
consideration, which is extended both to rational agents and to others.
34
The virtue of
beneficence, for example, is the disposition to feel sympathy for someone’s suffering and
32
Ibid., 108-110.
33
Ibid., 146-169.
34
Ibid., 154.15
to act to relieve that suffering. Since an animal is capable of suffering, even if not
unjustly since that would require primary considerability, the virtue of beneficence would
have us give direct consideration to the animal. It is not moral consideration in the
primary sense, but it is what I will call virtuous, or secondary, consideration. Because
contractualism demands virtuous characters for no reason other than the well-being of the
rational agents, virtuous consideration is merely a secondary effect of real moral action.
Despite this argument’s appeal, I do not think Carruthers achieves a significant
amount of success from this third attempt. My first objection involves the source of these
virtues. Again, the virtues related to duties of beneficence, non-maleficence, honesty,
etc. ought to be cultivated in rational agents in order to approximate the requirements of
duty when precise moral calculations are too difficult. Recall that the contractualist
requirements of duty regard the direct consideration of rational agents and no others. My
objection is this: could the contractualist not limit the scope of the virtuous dispositions to
those who truly matter? It seems like a sufficiently beneficent character need only
sympathize with one who is suffering not from pain, but from an obstacle to autonomy.
The pain felt by a rational agent would be, of course, one example of such an obstacle,
but pain qua pain is not the right kind of object of sympathy, given the enormous amount
of things that feel pain and yet do not matter morally (under contractualism).
Consider this virtue, then: there should be, by the lights of contractualist theory, a
virtue that approximates the principle of non-interference with the autonomy of others.
Persons should be in the habit of promoting and facilitating the freedom of others. This
disposition falls under the general heading of virtues of non-maleficence. Persons in the
habit of controlling other persons—physically restraining them and so on—possess a vice
contrary to the demands of contractualist morality. With that in mind, consider the scope
of this virtue. Consider the variety of things in the world that rational agents physically
control and restrain: the vehicles they operate, the tools they use, and, of course, the
animals they keep. Insofar as these things have the capacity for motion (and independent
motion in some cases), they are similar to the rational agents themselves. It does not
follow, of course, that the presence of this shared property requires us to treat those
things as we do rational agents for the sake of our virtuous character. If indeed we did
have to exercise virtues beyond the scope of consideration, and virtues of non-16
maleficence overflowed into our treatment of animals, one would think that the
contractualist would forbid the very sorts of things that Carruthers considers permissible,
such as the caging of animals. So, to return to the point about the virtue of beneficence:
if we ought to be sympathetic to the pain of others, it ought to be enough that we are
sympathetic to those others who actually matter.
There are two options for the contractualist. The first option, following the above
reasoning, is to narrow the scope of virtuous activity to best approximate the real objects
of duty. If the contractualist chooses this option, then the account of virtues would
provide no consideration for animals. Animals would benefit only from the duties related
to property rights and ‘public concern’, and as we saw, there is not much benefit there.
The other option is to claim that such narrowing is impossible, that humans are
psychologically unable to reliably distinguish between those who matter, and those who
don’t. But this option is not plausible: certainly we can at least distinguish between
humans and animals. On that point, Carruthers and I agree: in an argument regarding a
separate issue—his defense of the inclusion of all humans into the moral community—he
relies on the fact that it is easy for humans to psychologically separate their treatment of
humans from their treatment of animals.
35
For example, the ability to make the
distinction is what makes employment in factory farms and animal laboratories possible.
In Carruthers’s words,
That someone can become desensitised to the suffering of
an animal need not in any way mean that they have become
similarly desensitised to the sufferings of human beings—
the two things are, surely, psychologically separable.
36
This being the case, Carruthers must admit that our sympathy for animals, if it is to be
understood as a virtue, is a flawed virtue.
It should be clear by now that contractualism cannot even provide the secondary
sort of direct consideration for animals that Carruthers wanted. The moral intuition
regarding the wrongness of harming or killing an animal cannot be explained by appeal to
the indirect duties of respect for property, ‘legitimate public concerns’, or even by appeal
35
Ibid., 115.
36
Ibid., 160.17
to duties of character. If one wishes to subscribe to Carruthers’s contractualism, one
must bite the bullet and give up this intuition.
CONTRACTUALISM AND HUMANS
Carruthers’s theory fails to grant any significant moral consideration of animals,
despite his best efforts. However, he believes that all humans possess equal, maximal
moral value. If he is correct, then the conclusion of the ASO—that some nonhuman
animals and some humans are similarly morally valuable—is false. However, he
recognizes that bringing all humans into the moral community is a serious challenge.
37
As I have mentioned before, the first premise of the ASO is unproblematic, and the
second premise, that some humans are in all morally relevant ways similar to animals, is
not easy to deny. It is certainly not easy for Carruthers to deny, since his central morally
relevant characteristic is rational agency. Infants and young children are not rational
agents, and other humans have lost the capacity for rational agency altogether. Because
contractualism requires moral agency for every member of the moral community, it
appears that these groups of humans must be excluded, just as animals are.
Carruthers believes he can handle the cases of the very young more easily than
other cases of non-full persons. Since the injuring of a child can harm the future rational
agent, it appears that contractualism may provide some protection to them. We can
assume that the imaginary contractors would agree to a principle of allowing the young to
develop in healthy ways, as it serves the self-interest of the rational agent that child
becomes. However, contractualism has a problem explaining what is wrong with injuring
a child if the injury results in death. If the child dies, there is no future rational agent |
Likewise, if the injury results in the child’s inability to become rational, then no rational
agent was harmed. It would seem that injuring a child is wrong if and only if there will
exist a future rational agent who is thereby harmed. This result is, of course,
unsatisfactory for Carruthers. Even if it made sense, somehow, to provide protection not
only to actual but also to possible future agents—thereby explaining the immorality of
killing children, etc.—this protection would likely be too drastic. Abortions and even
contraception would be impermissible. Moreover, it seems that regular attempts to
37
Ibid., 110.18
produce offspring would become mandatory | Failures to make such an attempt would
often prevent a future rational agent from ever existing. Clearly, the demands of this kind
of moral principle are unacceptable by any reasonable standard. Therefore, Carruthers
must look elsewhere to account for the moral value of non-full persons.
He relies on two arguments: a slippery slope argument and an argument from
social stability. I will comment on each in turn. First, his slippery slope argument runs
as follows. There is no magic line between rational agent and non-rational agent. While
our society does in practice establish age-based restrictions, such as a twenty-one year
drinking age, no one believes that a human becomes rational at any specific age. There
is, as we all know, a large gray area. Strictly speaking, a contractualist would agree that
the moral standing of a human improves between birth (where there is no rational
agency) and full adulthood (assuming the adult has full rational agency). Despite the
differences in the real moral worth of humans, based on their rational agency, there are
practical issues that must be accounted for. Most people, he thinks, should not be trusted
to handle a moral principle that distinguishes between rational and non-rational humans.
They would tend to misuse the principle, and deny consideration not only to non-rational
humans but also to rational humans who are considered ‘deviant’ for some (morally
irrelevant) reason. The slippery slope lies between actual non-rational humans and
humans who are fully rational but appear otherwise.
This argument is not without its problems. First, it is not clear how potential for
misuse invalidates a moral principle. One might think that we have a moral duty to drive
our cars below the speed limit, to protect our lives and the lives of others. However,
there is frequent misuse of this principle. Ought we for that reason conclude that the
principle is immoral? Of course not. But Carruthers would have a reply to the speed
limit analogy. It is not that a principle treating different humans differently would be
knowingly disobeyed, as in the case of the speed limit. Instead, people would
unintentionally misuse it, due to a lack of intellectual ability. Carruthers believes that
“most people are not very deeply theoretical,” and would not be able handle a principle
that requires them to recognize rationality in others.
38
Since the rational contractors
38
Ibid., 116.19
would be aware of the lack of theoretical ability in these people, the revised principle of
universal human considerability is necessary.
His reply might reasonably justify extending protection to non-rational people
who are almost indistinguishable from rational agents. It would be disastrous to allow
the killing of a senile or similarly incapacitated adult human, for instance, if indeed most
people would be unable to determine whether such people were not rational. It is
unclear, however, how this applies to the treatment of infants. Certainly no one has
difficulty telling an infant from an adult. While there is certainly a gray area between
infancy and adulthood in regard to the development of rational agency, no gray area
exists between, say, a newborn and a six-month-old. Neither is a rational agent in the
slightest. The slope does get slippery sometime after that, so, to avoid becoming
slippery, we could adopt a principle protecting all humans except those below the age of
six months. This principle should be perfectly acceptable to Carruthers.
He does not accept it, of course. He does not believe that people can handle the
distinction between such obviously different people. The strange fact is that he does
believe that people can make a distinction between humans and animals, and avoid a
slippery slope between our treatment of the two groups. He writes, “someone who argues
that since animals do not have rights, therefore babies have no rights, therefore there can
be no moral objection to the extermination of Jews, Gypsies, gays, and other so-called
‘deviants’, is unlikely to be taken very seriously.”
39
His claim is that no slippery slope
exists between animals and human infants. But remove the ‘animals’ bit from his
sentence, and we still have a sentence that is unlikely to be taken seriously. Someone
who argues that since babies (unlike all other humans) have no rights, therefore there can
be no moral objection to the extermination of (adult) Jews, etc., is likewise making an
obviously lousy argument. There appears to be nothing blocking my proposed rerevision of Carruthers’s principle, though it places infants in the same morally precarious
situation as animals.
Finally, Carruthers has made a fundamental error in his argument that renders this
entire line of thought unacceptable. According to contractualism, the only beings who
matter morally are rational moral agents. It is simply contrary to the theory to grant
39
Ibid., 115.20
primary, direct consideration to any non-rational beings. I use ‘primary, direct
consideration’ in the same manner as it was used in the earlier discussion of virtuous
behavior; I contrast it with secondary, direct consideration, such as that granted to
animals for the sake of virtue. Such consideration is required only for the sake of those
who deserve primary consideration. Despite Carruthers’s attempt to present our reason
for protecting atypical humans as different from our reason for protecting animals, the
reasons are the same. Our treatment of any non-full person matters only insofar as it
affects the treatment of rational agents. This is the moral fact of the matter, regardless of
anything contingent facts may do to alter the final form of the moral principles. Do we
really believe that humans who are not full persons deserve protection not for their own
sake, but only because injury to rational agents might result? No, this does not even
approach our moral intuitions. We protect the infant for the sake of the infant | To drive
the point home, consider what Carruthers would have to say about a world where most
people are not so terribly lacking in theoretical ability. If proper contractualist moral
education could be provided, it would no longer be immoral to kill a child or conduct
painful medical experiments on the mentally impaired. Perhaps Carruthers would bite
the bullet here, and agree that such actions would in fact be permissible, given the
different conditions, but it is a tough bullet to bite.
His second attempt to grant moral standing to all humans is, unfortunately, worse
than the first. His ‘argument from social stability’ relies on the premise that many people
would find it psychologically unbearable to live in a world where non-rational humans
are not morally considerable.
40
Because people would be unable to accept a principle
that denies full consideration to certain humans that they care about, such as their
children, and because their inability to accept the principle would result in social
instability, it is necessary to replace the principle with something more acceptable. A
principle that grants full consideration to all humans is the best substitute; therefore, it is
the moral principle to which the imaginary contractors would agree.
Carruthers entertains the objection that there are in fact socially stable
communities where differential treatment of humans occurs, and that therefore his claim
about human psychology is false. He responds by pointing out the differences between
40
Ibid., 117.21
those other cultures and our own. First, in some cultures, religion and tradition play a
much stronger societal role than they do in ours, and allow discrimination between
different humans while maintaining social stability. Without the powerful religious
tradition, stability would be impossible. Second, some cultures are “teetering on the edge
of survival,” and so allow practices such as infanticide only because it is necessary for
survival.
41
To sum up Carruthers’s response: Our culture, unlike some other cultures, is
such that the majority of us cannot psychologically handle any principle that denies
consideration to all human beings. Therefore, we must grant consideration to all human
beings.
One way to counter Carruthers’s response is to deny the empirical claims he
makes about other cultures. Pluhar, for one, has done so.
42
I choose to take another
route, and examine how the claim, if true, should be understood. There are two possible
ways to interpret the claim he is making about our culture, and I will address each one.
On the one hand, he might be claiming that our culture produces people who,
unlike the people of other cultures, are psychologically unable to deny moral
consideration to any humans. This is not a moral claim. However, a moral claim seems
to follow from it: a contractualist should prefer a culture where we can make all morally
relevant distinctions, and, so, not prefer ours. Carruthers, if he were making this claim,
would have to admit that we really ought to deny non-rational humans primary, direct
consideration, and that we are currently in an unfortunate situation where true morality is
impossible. His imaginary contractors ought to construct principles that work toward
ending this situation.
Carruthers attempts to correct this. He argues that we cannot modify our
psychological states, and so the imaginary contractors would not demand that we do so.
Instead, moral principles would conform to our psychology. This, he says, is what the
argument from social stability is really saying. But consider the following implication of
his argument. People in a strongly religious or traditional culture, who are
psychologically able to kill or injure some kinds of non-rational humans, and who can
make the proper distinctions between individuals so that no rational humans are harmed,
41
Ibid., 119.
42
Pluhar, 94-95.22
are morally permitted to harm non-rational humans. The culture need not be, as
Carruthers puts it, “teetering on the edge of survival.” So long as social stability is
maintained, the contractualist does not judge harming non-rational humans to be morally
wrong whatsoever. On this view, the real moral significance of non-rational humans
simply varies from culture to culture. This fact stands in stark contrast to the
contractualist view that all rational humans are necessarily maximally morally valuable,
regardless of facts about their culture.
Carruthers should want to object with this line of thought. One would think that
fundamental moral principles should not be based on contingent particularities within a
certain culture. This smacks too much of cultural relativism. Of course, Carruthers
cannot make this objection, because his moral principle protecting non-rational humans is
undeniably based on cultural norms. Moreover, Carruthers does not seem entirely willing
to admit that non-rational humans deserve no consideration except in cultures where
rational humans happen to have a certain favorable disposition towards them. I expect
most people would be similarly unwilling to make that concession. For this reason, it
would be sensible to reject this interpretation of his argument and examine the other.
The other interpretation is that Carruthers believes that our cultural attitudes
toward children and other non-rational agents are in fact morally correct, despite the
conclusions of contractualist theory. If this is what he means, we are led to a bizarre
result: contractualist morality, based on the imaginary agreement between perfectly
rational contractors, would include principles that are somehow morally correct prior to
the agreement | There would have to be some non-contractualist moral standard for this
interpretation to be the case, and it is clear that Carruthers does not admit such a thing.
This interpretation fails more quickly than the first.
In conclusion, the argument from social stability is a mess. At best, it implies that
moral principles ought to be compromised for the sake of contingent cultural biases. Yet
it makes more sense to think that someone is psychologically unable to fully meet the
demands of morality than to think that principles of morality can so easily turn on
cultural norms. Moreover, in the case of our treatment of non-rational humans, it seems
clear that our commitment to their well-being (when we have it) stems not from some
nonmoral inclination, but because we know that they do matter morally. The killing of a 23
small child is morally repugnant not because of its effect on social stability, but because
the child really matters. Like Carruthers’s slippery slope argument, his argument from
social stability does not adequately explain the moral considerability of all humans.
SUMMARY
Carruthers has attempted a refutation to the ASO. He agrees that the second
premise is essentially true, because animals and some humans lack rational agency, the
fundamental source of moral value. To rescue the non-rational humans, he locates
secondary sources of value. Empirical facts—facts about psychology in particular—play
an important role in the construction of principles regarding our treatment of non-full
persons. While these facts lead the contractualist to grant some reasons to treat animals
well, they support even more protection for humans. In this way, Carruthers denies the
conclusion of the ASO.
A strength of the contractualist defense is that it aims to produce real moral value
in all humans. Carruthers is not arguing for the mere practical consequence that we treat
non-rational humans as if they had maximal moral value; that is tantamount to admitting
that they in fact do not. Instead, he brings the empirical, psychological facts to bear on
the imaginary set of rational bargainers at the foundation of contractualism. In this way,
the empirical facts affect the moral principles, and not just the practical consequences.
As I have argued, however, it nonetheless does not seem possible for the
contractualist to produce principles adequately protecting either animals or non-rational
humans. Even where contractualism does grant a modicum of moral considerability to
non-full persons, it does so for the wrong reasons. Compounding that problem, any
protection that is given is in the form of principles precariously resting on contingent,
empirical propositions that may one day not hold true. When such time comes, the
contractualist would be led to deny moral consideration to whole multitudes of creatures
(humans and otherwise) who were previously morally considerable.
In the end, the problem contractualism has in answering the ASO is its principal
requirement of rational agency for membership in the moral community. As we shall see,
other moral theories do not share this obstacle, and produce more acceptable responses to
the ASO.24
CHAPTER 3: THE RIGHTS RESPONSE
Contractualism is just one ‘rationality-based’ theory, taking rational agency as the
basis for inherent moral value. Kant’s moral theory famously does this as well. There is
at least one reason why rational agency is so important for moral theorists: note that,
regardless of the theory, rational agency is a necessary condition for morality to function
at all. In order for us to act morally, to follow any moral principle, we must be capable
of conceptualizing principles and acting deliberately. That is, all moral agents are
rational agents. But the set of moral agents need not be identical to the set of moral
patients. As we found with contractualism, making the two sets identical denies many
humans and animals full moral standing, or any moral standing at all.
In this chapter, I examine the responses to the ASO developed by Tom Regan and
Evelyn Pluhar. Regan and Pluhar hold similar, rights-based moral theories. As I will
show, Regan proposes a plausible direction for an answer to the ASO, but runs into
several problems. Pluhar attempts to recover Regan’s theory, and produces a more
acceptable answer to the ASO. She is somewhat successful. I will argue, however, that
the shortcomings of both Regan’s and Pluhar’s theories warrant a second look at the
theory considered by both to be the ‘runner-up’ theory: utilitarianism.
Regan
Tom Regan has played a large role in popularizing the challenge the ASO
presents. In his book, The Case for Animal Rights, he applies the ASO to Kantian
morality, and shows that Kant’s withholding of moral value to all but rational agents
excludes not only animals, but also many humans from the moral community.
43
Nonrational humans may receive indirect consideration, but as we’ve seen already with
Carruthers’s contractualism, indirect consideration is insufficient. He rejects Kant’s
theory, and seeks out one that does not suffer from the same problems.
The theory he seeks will be something other than rationality-based. Reviewing
other possible morally relevant characteristics, he considers and rejects the alternative of
a being’s simply being alive. While the equal inclusion of all living things in the moral
43
Tom Regan, The Case For Animal Rights (Berkeley and Los Angeles: University of California Press,
1983), 174-185.25
community would avoid contradiction with the ASO, there would be many other results
that are counterintuitive to say the least.
44
So Regan turns to his notion of a ‘subject-of-alife’ as the source of inherent value.
45
Not as narrow as the rational agency requirement
of Kant and Carruthers, Regan intends the term to capture most humans and other
mammals aged one year or more, and probably others.
46
A sufficient condition for being
a subject-of-a-life under Regan’s definition is possession of the following features:
1. Beliefs and desires;
2. Perception, memory, and a sense of the future, including one’s own future;
3. Emotions, and the ability to feel pleasure and pain;
4. Preference and welfare interests;
5. The ability to initiate action in pursuit of one’s goals;
6. Psychophysical identity over time; and
7. A welfare of one’s own.
47
To summarize the criteria, one might say that a subject-of-a-life has an experiential
welfare. Regan seems to require possession of all of the above features, though he is not
entirely strict. For instance, he intends to count non-rational humans among his subjectsof-a-life, despite the fact that some cannot initiate actions. In any event, Regan holds that
all subjects-of-a-life have equal inherent value. He does not make an argument for this
claim, aside from pointing out that a theory based on his claim fares better than
rationality-based theories when it comes to handling issues surrounding the ASO.
48
He
also does not give an answer to whether or not individuals possessing some but not all of
the required features have any moral value, or how that would be determined.
49
These
omissions have negative implications for the theory, as I will show later.
The notion that being a subject-of-a-life is a morally relevant characteristic is not
limited to any one specific theory. Many moral theories use something similar to
Regan’s list of features as the basis for inherent value. I will refer to any such theory as a
‘sentience-based’ moral theory (bearing in mind that some theories in this category
demand something more than bare sentience for moral significance) to be contrasted to
‘rationality-based’ theories such as those of Kant and Carruthers. Proponents of any
44
Ibid., 241-243.
45
Ibid., 243-248.
46
Ibid., 73-81.
47
Ibid., 243.
48
Ibid., 247.
49
Ibid., 264.26
sentience-based theory are going to be in a better position to handle the ASO than are
those who argue for a rationality-based theory, because they are not forced to contrive
devices to handle non-rational beings.
However, different sentience-based theories handle the conclusion of the ASO in
different ways. The conclusion, recall, only says that some animals and some humans are
similarly inherently morally valuable. It does not specify to what extent any individual is
morally valuable; that issue is left undetermined by the argument.
50
A full response to
the ASO is not tantamount to mere agreement with its conclusion; it must also make the
further determination of extent. For this reason, Regan examines various sentience-based
theories, including several forms of utilitarianism.
51
In the end, for reasons I will discuss
in more detail in the next chapter, Regan rejects all forms of utilitarianism. As an
alternative, he proposes a theory that is Kantian in appearance, taking on the language of
rights and with attention paid to respect for autonomous action and treating others as
ends-in-themselves. However, it is a sentience-based theory, not a rationality-based
theory.
Regan’s defense of his theory is largely of a negative sort: he develops his
principles out of a rejection of the most plausible available theoretical alternatives, most
notably utilitarianism. He believes that the best utilitarian theory will ultimately fail to
respect the inherent value of any individuals. For this reason, he postulates a ‘respect
principle’: treat those individuals who have inherent value in ways that respect their
inherent value. Spelled out, this principle consists of two main ingredients: a Kantian
notion of an end-i n-itself, and, following from that, a rejection of utilitarianism.
52
Specifically, Regan claims that we violate the respect principle when we make utilitarian
calculations about morally significant beings.
53
So, just as his focus on the subject-of-alife is the upshot of rejecting rationality-based theories, so his moral principles are built
(in part) upon the rejection of utilitarianism. As we shall shortly see, this method of
theory construction leaves something to be desired.
50
Incidentally, Pluhar utilizes a second formulation of the ASO to conclude that non-rational humans and
animals have inherent value equal to rational agents. In this paper, I do not focus on this formulation as the
ASO, though it appears in my discussion of Pluhar’s response to the ASO.
51
Regan, 140-143, 200-235, 250-258.
52
Ibid., 248-250.
53
Ibid., 248-249.27
REGAN’S THEORETICAL METHOD
Having established the underlying principle of respect, Regan follows the
following process. First, he derives secondary principles of action from the respect
principle. Then, he examines the implications of those principles. From there he
develops additional principles, also based on the respect principle, to handle any
unsatisfactory implications.
The first principle he derives from the respect principle is a principle against
harming, what he terms the “harm principle.”
54
Out of respect for others, we must regard
their welfare as important, because having an experiential welfare is what makes
subjects-of-a-life morally valuable in the first place. Therefore, we have a prima facie
duty not to harm others, as this would detract from their welfare. It is at this point that
Regan first introduces the notion of rights into his theory; he defines rights as valid
claims with corresponding duties.
55
The duty against harming, then, corresponds to a
right not to be harmed.
From the harm principle Regan adds two corollaries. First there is his so-called
“miniride principle,” which states that, given a situation where we must either harm one
group of beings or similarly harm another group, we ought to choose to harm the smaller
group. The second principle is called the “worse-off principle,” and states that, given a
situation where we must either harm one group of beings or harm to a greater degree a
smaller group, we ought to harm the larger group. This latter principle follows from the
rejection of utilitarian calculations found in the respect principle. I will have more to say
on this later on.
Next, he spells out some of the implications of these secondary principles. The
principles are not absolute, he argues, because they may come into conflict with other
principles, which themselves can similarly be derived from the respect principle.
56
Situations that exemplify such a conflict include situations such as acting in self-defense
against an aggressor, punishing guilty moral agents, and allowing for non-obligatory,
supererogatory actions. In these cases, argues Regan, we must balance the harm principle
against other principles that come into play in each specific case.
54
Ibid., 262-263.
55
Ibid., 271-273.
56
Ibid., 286-287.28
If his rights theory succeeds where other theories, notably utilitarianism, do not,
then all that is left to do is discover what answer his theory gives to the ASO. However, I
do not believe it has succeeded. Primarily, it fails in regards to the theoretical method I
have just described. It also has some bizarre implications regarding the ASO, which I
will also discuss.
Problem #1: Weak foundation.
Regan claims that his proposed fundamental principle, the respect principle,
“illuminates and unifies” many of our well thought out moral beliefs.
57
It is important to
note here that the beliefs he mentions in association with this claim are actually nothing
more than a rejection of a utilitarian morality. Are we to accept Regan’s theory on the
basis that we reject someone else’s? It would appear so. It is as if the failure to make a
positive case for utilitarianism has resulted in the establishment of the alternate theory.
Aside from being the wrong way to go about constructing one’s moral theory, it
underestimates the theoretical possibilities within moral philosophy. Surely more than
one option is available to the opponent of utilitarianism.
Naturally, Carruthers—whose contractualist theory we examined in the previous
chapter—is in just the sort of position that would lead him to attack Regan on this front.
And Carruthers is very instructive on this point; he indicates a major flaw in Regan’s
theory. The flaw is its lack of what Carruthers refers to as a ‘governing conception’. A
governing conception is the explanation a moral theory provides as to what, indeed,
morality is. It explains the following things: 1) the nature of morality, what moral ideas
are about; 2) how we come to have moral knowledge; and 3) the basis of moral
motivation.
58
After all, to develop a moral theory, one must presuppose that morality is a
valid subject of rational inquiry, and something about which we can have knowledge
(points 1 and 2). And it is similarly clear that moral principles, when we come to know
them, have some sort of motivating force compelling us (with varying success) to act
morally (point 3). If a theory cannot make sense of these most basic moral data, then the
theory is groundless.
57
Ibid., 259.
58
Carruthers, 23.29
Pluhar has a related argument against Regan, in her Beyond Prejudice.
59
Regan’s
theory stems from the respect principle, which is itself generated by his notion that all
beings with an experiential welfare, all subjects-of-a-life, possess inherent value. This
latter notion is controversial for two reasons. First, we need not agree with Regan’s
judgment that experiential welfare is a morally relevant characteristic. While we may
very well run into difficulty with the ASO if we commit to a rationality-based theory, that
does not automatically entail that we should move to a sentience-based one. That move
is not by default; rather, it must be argued for. Second, we need not agree with Regan’s
specific notion of how we are to treat inherent value merely because it rejects
utilitarianism. As I stated before, we can reject Regan’s notion of inherent value even if
we reject utilitarianism. Pluhar, who does indeed reject utilitarianism, follows her
criticism of Regan’s theory with an attempt to bolster the rights theory with additional
arguments.
Problem #2: Vague decision model
A second weakness of Regan’s theory is that it lacks a clear method for handling
practical conflicts that result from the application of the respect principle and its
derivatives. As I mentioned earlier, the duties Regan derives from the respect principle,
e.g., the harm principle, are not absolute. There are situations when it is permissible to
harm a moral patient, as long as we find that other principles, similarly derived from the
respect principle, act more strongly upon us. Unfortunately, we are never told how to
determine which principles act more strongly, and when. Instead, it appears that Regan
falls back on moral intuitions as the basis for resolving conflicts between principles.
While these intuitions play some role in the development of a theory, they cannot act as
our sole mechanism for moral decision-making, lest we end up with a theory that mirrors
whatever we happen to believe about individual moral cases.
For example, notice Regan’s handling of cases of self-defense.
60
According to the
harm principle, an innocent victim of an aggressor would not be permitted to harm the
aggressor. He calls this notion, that it would be immoral to harm someone even in self-
59
Pluhar, 231-240.
60
Regan, 287-290.30
defense, the ‘pacifist principle’, and argues that it runs sharply against our moral senses.
In order to avoid the principle, he resorts to a consequentialist consideration: he claims
that when defending against an aggressor and causing harm results in the prevention of a
greater amount of future harms, the harm we cause is justified. He adds that, aside from
the harm principle and the consequentialist consideration, there is also a ‘proportionality
principle’ at work: we should inflict harm on the aggressor that is proportional to the
harm being prevented.
Consider also how Regan allows for supererogation.
61
He presents an example
where a racecar driver is seriously injured. Without medical assistance, he will die.
There are a number of medical personnel in the area, but they all have patients who will
suffer some serious harm (paralysis, or a lost limb) if not immediately treated. Regan’s
theory—and specifically the worse-off principle—produces an obligation for the patients
to forego treatment in order to save the driver. But that sort of choice is one that most of
us would consider to be above and beyond the call of duty. Regan agrees, and shows that
the situation actually involves a conflict of principles. The worse-off principle is the first
one. The second principle involves situations such as that of the racecar driver, who
voluntarily choose to enter into a high-risk activity. In these cases, Regan argues, the
person knowingly waived the full protection of principles like the worse-off principle,
and so we ought to treat the person accordingly.
62
As a third and final example, consider how Regan explains the rightness of
punishing the guilty.
63
It is a widely held notion, of course, that a person guilty of a
crime ought to be punished. However, the punishment—whether a fine, imprisonment,
death, or otherwise—counts as a harm, and so the harm principle tells us that punishment
is wrong. Regan notes this, but offers no real answer. He mentions that we ought to
respect the rights of the criminal, and that the proportionality principle would hold here
as much as in the case of self-defense. But this does not tell us why we should punish in
the first place. He finally offers the suggestion that his rights theory is “sympathetic to”
the notion of punishment, but he fails to explain why.
61
Ibid., 320-322.
62
Ibid., 322.
63
Ibid., 290-291.31
In each of these examples, Regan offers a supplementary principle to account for
cases where the harm principle fails to provide an acceptable answer. And he tries, with
varying success, to show how those supplementary principles are derived from the
respect principle. Unfortunately, he fails to adequately explain how these principles
interact. For instance, the consequentialist answer to the self-defense example shows
that, at times, the consequences of one’s actions take precedence over the strict duty not
to harm. But when do consequences take precedence? In that one case, at least. But not
always. At one point Regan declares that “side effects” of an action are irrelevant to the
action’s rightness or wrongness.
64
It is unclear which consequences are understood as
side effects, but in any event, some consequences are trumped, morally, by the harm
principle. But nowhere are we told how and when one principle overrides another.
A similar point can be made in regards to the supererogation principle. When
someone enters into a high-risk activity, how are we to determine the extent to which we
must continue to obey the harm principle? For instance, what would happen if we
construed the racecar example so that, rather than a racecar driver, the near-death victim
was a pit crew member working out on the racetrack? Would this still qualify as a highrisk activity? It is unclear how we are to weigh the risk accepted by the victim against
the worse-off principle. Additionally, I should point out the strangeness of the example:
the idea is that the other injury victims, not the medical personnel, are making the
supererogatory choice to forego treatment in order to save the driver. But this setup is a
little hard to believe; one would think the medical staff would normally be the ones
making treatment decisions. But if it were up to the medical staff, an important factor in
the supererogatory nature of the choice, i.e., sacrificing one’s own good for that of
another, is removed from the dilemma. What would Regan say about the medical staff’s
moral obligation? Would it be similarly supererogatory for them to assist the racecar
driver, even though they are not making a sacrifice in doing so? Regan never provides us
with the decision mechanism to address this.
The problem I am describing, the lack of any clear decision model in Regan’s
rights theory, is closely related to the initial problem I mentioned, the theory’s weak
foundation. In both cases, the cause of the problem is Regan’s heavy reliance on moral
64
Ibid., 312-31532
intuitions. Certainly our intuitions matter—indeed, my own paper evaluates how well the
various theories can handle commonsense morality—but a theory must rely on more than
intuitions in its explanations, lest it become a mere enumeration of prevailing attitudes.
The very basis of Regan’s theory, the respect principle, was created out of only two
things: 1) a rejection of utilitarianism, largely on intuitive grounds, and 2) the assertion
that all subjects-of-a-life have inherent value, a moral intuition. With nothing else to
direct the theory, Regan simply throws out various moral problems and invents new
principles as needed to account for that particular problem. And while some principles
do seem to be derived from the respect principle, others (e.g., the proportionality
principle) seem made out of whole cloth. In fact, most of the secondary principles
contain elements that are introduced without explanation; the miniride and worse-off
principles, for example, refer to harming innocents, specifically. Regan never explains
where the notion of innocents comes from, or how it was derived from the respect
principle.
Even worse, the vagueness of Regan’s intuitive foundation occasionally leads him
to remain absolutely silent on an issue. Take for instance his answer to the morality of
abortion.
65
When Regan speaks of subjects-of-a-life, he is careful to note that he is
referring to mammals aged one year or more. Therefore, fetuses as well as the newly
born are not assumed to be subjects-of-a-life (though they may be). But Regan makes it
clear that being a subject-of-a-life is a sufficient, but not necessary, condition of
possessing inherent value. Therefore, the inherent value of fetuses and the newly born is
left unresolved. While that fact alone is not particularly problematic, its implications are.
It turns out that anything that is not a subject-of-a-life is, possibly, inherently valuable.
The lack of a necessary condition for inherent value makes for a mysterious moral theory.
If, indeed, some non-subjects-of-a-life, whether newborn humans, fish, trees, lampshades,
or anything else, turn out to be inherently valuable, how would we discover that fact?
Would a new moral intuition become suddenly available to us? Regan gives us no clue.
65
Ibid., 319-320.33
REGAN AND THE ASO
At best, Regan’s theory is a partial theory. At worst, he has only systematized a
list of moral intuitions. Nonetheless, we can apply it to the ASO and see what sort of an
answer it provides. Immediately, it is clear that Regan is not going to have the difficulty
Carruthers has providing at least an initially acceptable answer regarding the inherent
value of both non-rational humans and animals. Of course, not all humans and animals
are covered by Regan’s theory (non-subjects-of-a-life are, again, an open question), but I
will not linger on that point. What is more interesting is how Regan explains the
differential treatment that is due to different individuals.
Like Carruthers’s contractualism, Regan’s rights theory is absolutist, rather than
gradualist. You either have inherent value or you don’t; no one has more inherent value
than another. Therefore, for Regan, animals (at least, mammals of one year or more)
deserve consideration equal to a normal adult human. In this respect, the theory is a clear
departure from contractualism. Many find Regan’s conclusion too radical, as most
people believe that a human being has greater inherent value than some, if not all, other
mammals. If it turns out that non-mammalian animals also qualify as subjects-of-a-life,
as could very well be the case with some birds and reptiles, then his conclusion would
appear all the more extreme. While this point alone is certainly not enough to lead one to
reject the theory, adding this point to the underlying theoretical problems surely could be.
Interestingly, Regan does temper his theory’s principles in a way that at first
glance seems to approach a more commonsense moral perspective. While he maintains
that all subjects-of-a-life are equally inherently valuable, he does not hold that all have an
equal right to life. Recall that all have an equal right not to be harmed—due to the harm
principle. And the worse-off principle requires that, given a choice where some harm
must be inflicted on one group of beings or another, we should choose the lesser harm,
regardless of the number of beings in each group. What Regan suggests is this: a death
constitutes a loss, or harm, that varies in magnitude depending on which individual dies.
Regarding a lifeboat thought-experiment where we must sacrifice one of four normal
adult humans and a dog, he writes, “no reasonable person would deny that the death of
any of the four humans would be a greater prima facie loss, and thus a greater prima facie 34
harm, than would be true in the case of the dog.”
66
To obey the worse-off principle, then,
means that some lives are indeed worth more than others.
So, while the rights theory does offer a great deal of protection to the lives of both
humans and animals, to the extent that in some respects all moral patients are considered
equal, it also differentiates between different sorts of lives in a way that makes it
acceptable to most people. However, there are glaring problems with this answer to the
ASO that must be considered.
First, there is an arbitrariness about the harm principle and the manner in which it
is derived from the respect principle. Regan argues that in order to show respect to a
morally valuable individual, one acknowledges a prima facie duty not to harm the
individual. Yet, it also follows that one would have a similar prima facie duty not to kill
the individual. Regan gives us no reason why we ought to handle the two duties
differently. He describes killing as a category of harming, but there is no reason to do so
except that he derived the harm principle first. If he had begun by deriving from the
respect principle the principle against killing (or, a principle allowing others to live), he
could have then derived the harm principle (a principle allowing others to live well) from
that. This sort of arbitrariness is the result of the theory’s vague decision model, as
described earlier.
Second, Regan’s notion that the harm of death varies among individuals, and his
subsequent conclusion via the worse-off principle about how to handle lifeboat dilemmas,
is less acceptable than it appears at first glance. Concerning the choice between a normal
dog and a normal human, Regan himself admits that his theory requires us to sacrifice
any number of dogs if it would save a single normal human. After all, the worse-off
principle tells us that the numbers do not matter. Because the death of each dog is less of
a harm than the death of the normal human, the worse-off principle tells us to save the
human, and thereby choose the lesser harm. Perhaps many people would consider this a
reasonable conclusion. However, consider the fact that Regan is no speciesist: he accepts
the conclusion of the ASO, and is therefore willing to swap out the normal dogs for any
other similar beings, including human non-full persons. The startling claim that the lives
of a million non-rational humans should be allowed to die in order to save a single
66
Ibid., 324.35
normal human is a far cry from commonsense morality, yet it clearly follows from
Regan’s arguments.
All told, the primary difficulty with Regan’s answer to the ASO is the lack of
theoretical grounding and the consequent failure to provide a clear decision model. Even
if he had managed to avoid his highly questionable answer to the lifeboat dilemma, he
would still lack the tools necessary to make a persuasive case. Evelyn Pluhar, another
defender of a rights-based moral theory, argues similarly against Regan, but attempts to
present a similar theory while avoiding Regan’s pitfalls. We turn to her theory now.
Pluhar
The most serious flaw with Regan’s case, according to Pluhar, is that it relies too
heavily on moral intuitions.
67
Recall that it was this flaw that Peter Carruthers also
exploited in his attack on Regan. Pluhar also finds fault with Regan’s applications of his
theory, specifically with how he applies his worse-off principle to the question of killing
humans and animals. She presents an argument on that subject similar to mine, though it
is not identical. I shall say more about that later, when I discuss the application of her
theory to the ASO. First, let us consider her attempt to derive a more successful
foundation for a rights-based theory.
PLUHAR’S THEORETICAL METHOD
Pluhar avoids the pitfalls of Regan’s theory by appealing to our rationality rather
than our intuitions. She borrows an argument from Alan Gewirth, from his books Reason
and Morality and Human Rights. The argument, which I will call the ‘argument from
purposive agency’, proceeds from the point of view of a rational, purposive moral agent,
and shows that consistency demands that the agent must acknowledge the rights of
others.
68
1. I am a rational purposive agent with goals that matter to me, that I hold as (nonmorally) good, and that I want to achieve.
2. In order to accomplish any of my goals, necessarily I must have freedom and
well-being, which I therefore must hold as necessary (but again, nonmoral) goods.
67
Pluhar, 236-240.
68
Ibid., 241-4.36
3. All other agents ought not to remove or interfere with my freedom and wellbeing; that is, I have rights to freedom and well-being.
4. Grounding my rights claim is the fact that I am a purposive agent, that I have
goals that matter to me and that I want to achieve.
5. If being a purposive agent is a sufficient ground for having these rights, then all
purposive agents have these rights.
6. Therefore, all purposive agents have rights to freedom and well-being.
Beginning with the reflective agent’s conception of himself as an agent, the argument
takes purely prudential premises and produces a moral conclusion. This move can
initially appear illicit, and has been criticized as such.
69
However, I believe the argument
is quite persuasive if read correctly.
What Pluhar and Gewirth are not saying is that rational purposive agents must
attribute rights to themselves, and therefore all purposive agents have rights. This thirdperson way of spelling it out fails to capture the nature of the argument, because the
argument is necessarily from the point of view of a rational, purposive agent. The first
two premises are my conception of myself as a purposive agent, with certain necessary
goods that I require. I therefore insist upon and approve of the non-interference, on the
part of others, with my possession of these goods. My insistence and approval is
equivalent to my claim that I have rights (premise 3). At this stage in the argument, I am
not making a moral claim; that is, I do not need to be able to justify my rights-claim to
others. All that matters is that I accept it. In premise 4 I recognize that my acceptance of
my rights-claim stems from the basic fact of my being a purposive agent. If I were not a
purposive agent, and so did not have goals and necessary goods, then I would not have a
reason to attribute rights to myself. Therefore, (via premise 5, a formal truth,) and still
within the first-person perspective, I must claim rights for all purposive agents. I accept
the conclusion, a moral claim about all purposive agents, because of my necessary rightsclaim for myself.
If the argument from purposive agency works, it is easy to see why the conclusion
is moral, despite the nonmoral nature of the premises. Since all rational, purposive
agents must accept the conclusion from their own point of view, it is imperative upon
each of them to recognize the rights of each of the other purposive agents.
69
See for example R.M Hare’s criticism in his “Do Agents Have to be Moralists?” in Gewirth’s Ethical
Rationalism, ed. Edward Regis, Jr. (Chicago and London: University of Chicago Press, 1984), chapter 4.37
I find the overall structure of the argument highly persuasive, and have only one
major concern with the content, which regards the nature of rights. As Pluhar describes
the argument, my claiming rights to freedom and well-being (in premise 3) seems to be
the logical correlative to the expression, “all other agents ought not to remove or interfere
with my freedom and well-being.” It is likely that Pluhar wants the notion of rights to
contain more substance than this in order to locate rights (rather than correlative duties) at
the center of her theory. It also bears mentioning that Gewirth has argued specifically for
the primacy of rights before duties.
70
Unfortunately, Pluhar never explains exactly how
we are to understand this introduction of rights. This problem, I believe, is central to the
difficulties Pluhar ultimately faces regarding the ASO.
Nonetheless, Pluhar’s use of the argument from purposive agency introduces a
critical element that is not present in Regan’s case: a theoretical foundation. As we saw,
the lack of a sufficient foundation and decision model was the primary flaw in Regan’s
answer to the ASO. In order to assess Pluhar’s solution, recall Carruthers’s ‘governing
conception’ requirement for an acceptable moral theory. A governing conception
explains 1) the nature of morality, what moral ideas are about; 2) how we come to have
moral knowledge; and 3) the basis of moral motivation.
71
I believe Pluhar’s theory
answers those requirements in the following manner:
1) Morality is a set of principles respecting each individual’s rights to freedom
and well-being.
2) We come to know moral truths by recognizing our concept of ourselves as
purposive agents, and rationally constructing principles consistent with that
concept.
3) There is a basic human need to justify our actions; justification requires
consistency, and consistency demands moral behavior.
Theoretical foundation firmly in place, we can now assess the way Pluhar handles the
ASO.
PLUHAR AND THE ASO
The argument from purposive agency demonstrates that purposive agency itself is
the primary morally relevant property. Purposive agency, the possession of goals and the
70
Alan Gewirth, Human Rights (Chicago and London: The University of Chicago Press, 1982), 14-15.
71
Carruthers, 23.38
ability to act, is found not only in full persons, but also in many non-rational humans and
animals. Accordingly, Pluhar accepts the second premise of the ASO, that some animals
possess sets of morally relevant characteristics similar to those of some humans. She
does not rule out the possibility of circumstances that require us to favor one individual
over another, but, circumstances aside, it is the fact that the individual has goals that
matter to it that generates moral value.
Pluhar refers to two versions of the ASO: a ‘categorical’ version and a
‘hypothetical’ one.
72
The latter version is very similar to the one I have employed in this
paper. The categorical version contains an additional premise (3), and a stronger
conclusion:
1. Individuals who possess similar sets of morally relevant characteristics are
similarly inherently morally valuable.
2. Some nonhuman animals possess sets of morally relevant characteristics
similar to those of some humans.
3. Non-rational humans are as inherently morally valuable as normal adult
humans.
4. Therefore, nonhuman animals possessing sets of morally relevant characteristic
similar to some humans are as inherently morally valuable as normal adult
humans.
73
One would think that Pluhar introduces this version of the ASO in order to prove its
soundness, yet it is unclear that she does so. The additional premise (3) is ambiguous. Is
it the claim that all non-rational humans are as inherently valuable as normal adults, or
merely that some are? Pluhar does not tell us. If the claim is the former, then her theory
does not cohere with this argument: some humans—fetuses in an early stage of
development, the irrevocably comatose, and other extreme cases—are not purposive
agents. Pluhar admits that these humans are not granted moral status by her theory.
74
However, if the claim (and the wording in the conclusion) regards only some non-rational
humans, then her theory does support the argument.
Like Regan’s rights theory, Pluhar’s explains the wrongness of cruelty to animals
and non-rational humans more successfully than Carruthers’s contractualism. Carruthers
must rely on arguments showing that the cruelty is wrong without it being wrong to the
72
Pluhar, 63-66.
73
Ibid., 64.
74
Ibid., 259.39
animal or non-rational human, because he holds rational agency as the morally relevant
characteristic. Regan’s and Pluhar’s theories do not face this challenge. And Pluhar
improves upon Regan by including a moral foundation. It is now time to examine how
her theory leads to specific applications. As we shall see, she differs with Regan on
certain specific issues. While I agree with some of her criticisms of Regan, I believe that
significant difficulties also arise under her theory.
Regan’s worse-off principle, remember, states that given a situation where we
must either harm one group of beings or harm to a greater degree a smaller group, we
ought to harm the larger group. He holds the additional belief that animals are not as
harmed by death as humans are, so that, given a lifeboat case where either an animal or
human must be sacrificed, he believes we should prefer the death of the animal. I noted
earlier that Regan’s position also requires that we sacrifice a million dogs, or a million
non-rational humans, to save a single full person. I also pointed out that Regan’s position
is contingent upon his view of the right not to be killed as derived from the right not to be
harmed, which seems arbitrary and just as easily could have been the reverse.
Pluhar does not lodge these criticisms against Regan, but she, too, disagrees with
him on this point. The death of a dog and the death of a full person, she argues, are
equivalent harms.
75
Supporting this contention is the argument from purposive agency,
which locates moral value in the individual’s possession of goals. Both individuals, in
death, lose the opportunity to pursue goals at all. Therefore, their losses are equal,
regardless of the nature of their particular goals. Pluhar notes that Regan should have
arrived at this conclusion as well, given his claim that all subjects-of-a-life are equally
inherently valuable, despite the richness of their experiences.
76
I am not sure her
arguments on this point are altogether successful—more on that shortly.
The worse-off principle served two purposes for Regan. First, it generated an
anti-utilitarian decision model that better protects individuals’ rights. In combination
with the claim that animals are harmed less by death than humans, his principle also
introduced a significant element of gradualism into an otherwise absolutist morality.
Though Pluhar most likely would stand by the worse-off principle for the first reason
75
Ibid., 289-292.
76
Ibid., 289.40
(though she never explicitly says this
77
), she does not believe it serves the second,
gradualist purpose. She does arrive at some positions that have a gradualist tone, but via
other routes that maintain the absolutism at her theory’s core. Focusing on dilemmas
where one or another individual must die, Pluhar locates three factors that can make the
death of one of the individuals morally preferable. These three factors are: distress, the
relative complexity of simpler beings, and nearness to natural death.
First, it is possible that one individual would be more harmed by death than
another because one would experience a greater level of distress in the process.
78
Since
Pluhar holds that death itself harms each individual equally, any prior distress for one of
them would tilt the scales. The death of the individual who would suffer less in the
process is morally preferable.
Second, the death of an extremely simple being may be morally preferable to that
of a more complex one. Faced with the need to kill an animal (e.g., for food, to survive
when edible plants are unavailable), Pluhar believes it is morally preferable to kill a clam
than a fish, and killing either of them is preferable to killing a chicken.
79
This is not
based on the comparable intelligence of each animal, but rather on each individual’s
ability to care about its goals, since purposive agency is the morally relevant property.
Now, there are two ways one can read Pluhar here, and I will comment on each reading.
On the first reading, Pluhar is claiming that a fish may care about its goals, but
that, if it does, it cares less than the chicken cares. This reading produces a situation
where some individuals are more morally valuable than others, based on their level of
concern with themselves. Pluhar seems agreeable to this possibility: “[W]e should not
kill without good reason. How good the reason must be depends, of course, on how
morally significant the ‘killee’ is.”
80
However, it seems to me that this line of thinking
runs afoul when we consider choices between killing two similar individuals, say, two
normal adult humans. Should we try to ascertain which human cares more about her
77
Her wording of the worse-off principle concerns only choices involving the harm of one individual or
another; she does not specifically address situations where different numbers of individuals are involved. It
nonetheless seems reasonable for her to agree with Regan on this matter.
78
Pluhar, 292-293.
79
Ibid., 259.
80
Ibid., 258-259.41
well-being? I imagine Pluhar would strongly oppose this result, especially because of her
firm conviction that all purposive agents have an equal right to life.
On the second reading, she attaches moral value to the probability that an
individual is a purposive agent. This reading is, I think, the right one. It is clearly
supported by Pluhar’s claim that “beings more likely to be consciously purposive than
others, even if they are not clear cases, should…be spared if we have the option”
(emphasis hers).
81
However, her notion of “how morally significant the ‘killee’ is” does
not seem appropriate on this reading; perhaps she means to say, “how likely to be morally
significant.” But how are we to handle the relative probabilities? Consider the choice
between killing a clam or killing a fish. The fish is far more likely to be a purposive
agent, but one clam might not provide a full meal. If the choice is between one fish and
several clams, is it still preferable to kill the clams? Pluhar does not tell us. It seems
wrong to kill too many possible moral patients in order to spare a single, more likely
moral patient: the harm done, if indeed the simpler individuals turn out to be moral
patients, is tremendous.
The third factor able to tip the scales in favor of one life over another is the age,
or nearness to death, of the individuals: “An octogenarian who has had a lifetime to
formulate and fulfill goals is harmed less by death than a teenager, even if the teen has
developed far fewer interests at that point.”
82
The same goes for the terminally ill, and
others who are close to death. The harm is unavoidable for those individuals, and this
fact mitigates the wrongness of killing.
Pluhar does not fully explain this point, and leaves some worries unanswered.
She is granting additional moral value to the projected length of one’s life, a move
seemingly in conflict with her earlier declaration that all lives are of equal worth,
regardless of their content. Does this mean that we should attach moral value only to
quantity of experiences, and not quality? One would think she would attach moral value
to neither quantity nor quality of experiences; after all, she said that death deprives every
individual equally, that each individual loses everything. It does make intuitive sense
that an individual who will die tomorrow is not suffering a much greater harm by dying
81
Ibid., 259.
82
Ibid., 292.42
today, but Pluhar does not tell us what grounds this judgement. If the ground is indeed
the remaining quantity of experience, then we can apply this line of thought to many
more lifeboat dilemmas than just those containing octogenarians and the very near-death.
Between an average teenager and an average thirty-year-old, the former’s life is more
valuable. Between an average thirty-year-old and any age dog (as dogs do not normally
live more than twenty years), the thirty-year-old’s life is more valuable. And a young
Galapagos tortoise (assuming it counts as a purposive agent, a possibility Pluhar leaves
open), who may live two hundred years, has a more valuable life than any human | I
doubt Pluhar would accept any of these judgements of relative moral value, but she does
not give us a reason not to arrive at them.
Of the three factors mitigating the harm of death, two are at least questionable.
Moreover, Pluhar’s acceptance of Regan’s worse-off principle implies that we should
apply Regan’s worse-off principle to cases where two groups of individuals are involved.
Recall Regan’s argument that the numbers do not matter; it was this aspect of Regan’s
principle that led to his conclusion that a million non-full persons ought to be sacrificed
to spare a single full person. Therefore, we ought to be able to apply the million-to-one
result to Pluhar’s three factors. A million painless deaths are preferable to a single
painful death; a million simpler individuals should be sacrificed instead of a more
complex individual; a million octogenarians or a million dogs should be sacrificed rather
than a young, healthy human. These results are not so counterintuitive as to be fatal to
Pluhar’s theory, but it should be noted that she has not ultimately overcome the sort of
problems we encountered examining Regan’s theory.
Pluhar’s moral foundation, on the other hand, is her significant contribution to the
rights theorist’s answer to the ASO. The argument from purposive agency provides the
right kind of grounding for an acceptable theory, and effectively counters the most
serious criticisms lodged against Regan. There remain some lingering doubts about how
Pluhar arrives at some of her secondary principles,
83
but we can nonetheless credit her
83
Pluhar never explains how we might arrive at the worse-off principle via the conclusion of the argument
from purposive agency. She also proposes the existence of acquired rights and duties, without any specific
argument in their favor (Pluhar, 283 and elsewhere). Other principles, such as her ‘liberty principle’
(which I do not address in this paper) similarly lack arguments in their favor (ibid., 298-300). Moreover,
some applications of her theory, such as her line on abortion (ibid., 253) are in conflict with these same 43
with a theoretically compelling answer to the ASO. And compared to the answer based
on Carruthers’s rationality-based contractualist theory, Pluhar’s answer is undeniably
more acceptable.
principles. Most of these principles and applications are developed strictly out of intuitive notions about
moral rights, and are therefore vulnerable to the same criticisms as those made against Regan’s theory.44
CHAPTER 4: THE UTILITARIAN RESPONSE
In the preceding chapters I introduced three non-utilitarian theories and evaluated
their respective answers to the ASO. Contractualism denies the conclusion of the ASO,
but fails to account for a wide range of moral beliefs regarding non-full persons. The
contractualist must conclude that all non-full persons, whether animal or human, are
hopelessly excluded from real membership in the moral community, as none of the
attempts to grant them secondary consideration are ultimately successful. Tom Regan’s
rights theory proves more successful in handling non-full persons, but lacks a theoretical
foundation. The arbitrary nature of Regan’s moral principles leads to his failure to
answer the ASO adequately. Finally, Evelyn Pluhar’s rights theory combines the best
aspects of the first two theories: a solid theoretical foundation like that of Peter
Carruthers’s contractualism, and the persuasive theoretical application that Regan had
sought.
It is remarkable that each of these philosophers, before defending their own
respective theories, spends considerably more effort arguing against utilitarianism than
any other rival theory. In the writings of all three, utilitarianism appears to play the role
of ‘runner-up’ theory. In this chapter, I describe the theory of utilitarianism in its various
formulations, and produce an initial response to the ASO. In doing so, I discuss the
theory’s initial appeal, the source of utilitarianism’s ‘runner-up’ status. Next, I note the
major criticisms against the theory, and answer each charge in turn. Much of my
discussion refers to Peter Singer’s utilitarian formulation, which is largely successful
against its opponents, but I also attempt to revise the theory in order to handle the most
difficult challenges.
Utilitarianism’s Theoretical Foundation
Utilitarianism is a moral theory that has as its core principle, ‘maximize utility’,
the term ‘utility’ understood to mean nonmoral goods, such as pleasure, happiness,
desire-satisfaction, or well-being. Underlying a moral agent’s acceptance of the core
principle is the agent’s recognition of the fact that her own interests are no more
important, from an objective perspective, than the interests of anyone else. The agent
who does not recognize this fact is not thinking ethically. Peter Singer argues along this 45
line. He claims that, to think ethically, we must “go beyond the ‘I’ and ‘you’ to the
universal law, the universalisable judgme nt, the standpoint of the impartial spectator or
ideal observer, or whatever we choose to call it.”
84
Intuitively, Singer’s claim makes sense. But as we have seen, appeals to
intuitions are often not persuasive. I believe a more solid case can be made for a
utilitarian starting point, if we reexamine Pluhar’s argument from purposive agency.
While it may seem ironic if not implausible to use her theory’s foundation in order to
ground utilitarianism, I believe it is the correct line to take. Let us begin by presenting
her argument, again:
1. I am a rational purposive agent with goals that matter to me, that I hold as (nonmorally) good, and that I want to achieve.
2. In order to accomplish any of my goals, necessarily I must have freedom and
well-being, which I therefore must hold as necessary (but again, nonmoral) goods.
3. All other agents ought not to remove or interfere with my freedom and wellbeing; that is, I have rights to freedom and well-being.
4. Grounding my rights claim is the fact that I am a purposive agent, that I have
goals that matter to me and that I want to achieve.
5. If being a purposive agent is a sufficient ground for having these rights, then all
purposive agents have these rights.
6. Therefore, all purposive agents have rights to freedom and well-being.
85
I believe the following revision to the argument is both more persuasive, and can point
the way toward a utilitarian morality:
1. I am a rational purposive being with goals that matter to me, which are
therefore (nonmorally) good.
2. All agents ought not to interfere with the achieving of my goals; rather, they
ought to promote them.
3. Grounding my ought-claim is the fact that I am a purposive being, that I have
goals that matter to me and that I want to achieve.
4. If being a purposive being is a sufficient ground for these ought-claims, then
those ought-claims apply to all purposive beings.
5. Therefore, as a rational being I ought not to interfere with the goals of all
purposive beings, but rather promote them.
The most significant item that I have added to the argument is the notion of the
promotion of goals in addition to non-interference with them. I do not believe this
addition runs counter to Pluhar’s version of the argument, but builds upon it in an
84
Peter Singer, Practical Ethics, 2
nd
ed. (Cambridge, UK: Cambridge University Press, 1993), 12.
85
Pluhar, 241-244.46
uncontroversial way.
86
I have also changed ‘agent’ to ‘being’, except where the notion of
agency was critical (I cannot make an ought-claim to non-agents). This is a reasonable
modification. Pluhar means nothing more by ‘agent’ than a being with goals it wants to
fulfill.
87
But if being a purposive agent merely entails the possession of goals, then the
expression ‘purposive agent’ is redundant. More important is what I have removed: the
notion of rights.
88
Although it might be argued that the introduction of rights in Pluhar’s
third premise necessarily follows from preceding premises, Pluhar offers no such
argument. She refers to rights as the “logical correlative” of ought-claims,
89
but clearly
rights are meant to add something more than the ought claim; if not, my version of the
argument, which includes ought-claims, subtracts nothing from her version. Note that
Pluhar uses her version of the argument to generate rights-claims that are intended to be
more demanding than the ought-claims available to the utilitarian. But, again, it is
unclear where the strong notion of rights has its source in her argument.
Perhaps my version loses the notion of rights by arriving at a conclusion
regarding the treatment of the goals of individuals, rather than the treatment of
individuals themselves. Purposive beings, not their goals, are the proper objects of
rights-claims. However, I believe this counts as an advantage of my version of the
argument. Both versions have as their premise the claim that my goals are what I hold as
good; Pluhar’s addition of freedom and well-being as necessary goods seems
unnecessary. If my desire for freedom and well-being arises out of my desire to achieve
my goals, why should the former desire become the focus of the argument? This is left
unclear. But since both our versions introduce an initial ought-claim that regards not
myself, but either my goals or the combination of my freedom and well-being, it seems
that my version is more straightforward by maintaining the focus on those goals, rather
86
As Pluhar specifically spells out the argument, agents “ought at least to refrain from” interfering with the
freedom and well-being of purposive agents (emphasis mine). This suggests that agents perhaps should
promote these things as well.
87
Pluhar, 248-249.
88
I have also removed a phrase from Premise 1. It is enough that the goals matter to me. The fact that I
want to achieve them is incidental, and may not even be the case, as I draw out in my formulation of
utilitarianism. The fact that some things matter despite my not wanting them seems to be neglected by
Pluhar’s theory.
89
Pluhar, 242.47
than shift the focus to myself. For this reason, I do not see any obvious objection to my
formulation.
Indeed, I believe my revision of the argument from purposive agency can serve as
a persuasive but generic ethical foundation. It can serve as the basis not only for
utilitarianism, but also for many other (though not all)
90
ethical theories. All I have
meant to accomplish in my revision is a more generalized conclusion, one that underpins
ethical reasoning without arriving at any particular moral theory. In doing so, I arrive at
Singer’s claim about the universal nature of ethics—going beyond the ‘I’ and ‘you’—
without reliance on intuitions.
If we accept Singer’s claim, the next step is to determine a specific moral
principle or set of moral principles. An initially appealing principle turns out to be, in
fact, a utilitarian principle. The conclusion of my version of the argument from
purposive agency does produce such a principle, if we equate the non-interference with,
and promotion of, the goals of purposive beings to the maximization of utility (nonmoral
goods). This equivalence is perfectly reasonable given the first premise of the argument,
that goals are what count as nonmoral goods. Depending on how we spell out the goals
of purposive beings, we can arrive at a variety of utilitarian principles. Later, I will
present and discuss these varieties.
Singer arrives at the same principle, though he talks of ‘interests’ instead of
goals.
91
The difference at this stage is unimportant; later on I will make distinctions in
order to refine the theory. What is important is that, at this point, utilitarianism is
revealed to be what Singer calls a ‘minimal’ theory: when we reason from the pre-ethical
to the ethical, “we very swiftly arrive at an initially utilitarian position.”
92
That we do so
is largely agreed upon. Carruthers, Regan, and Pluhar all recognize the initial plausibility
of utilitarianism; it is a primary reason why utilitarianism remains their runner-up theory.
“Utilitarianism has the reputation of being a theory with considerable appeal,”
writes Singer. He continues,
Some are attracted to it merely by its simplicity, but there is more to its
appeal than simplicity, as is shown by the fact that those who defend
90
For instance, the premise that purposive agency grounds ought-claims would not be acceptable to a
contractualist such as Carruthers, who grounds ought-claims in rational agency.
91
Singer, 13-14.
92
Ibid., 14.48
pluralistic ethical theories almost always include some kind of utilitarian
principle among the things they value or regard as duties. While it is
common for writers in ethics to deny that utilitarian considerations are the
only valid moral considerations, it is quite rare for them to deny utilitarian
considerations any place at all in their moral systems.
93
Here we find another factor behind utilitarianism’s runner-up status. Regan, for instance,
admits that consequences do matter morally, though his theory is on the whole nonconsequentialist.
94
Pluhar also allows consequences to play a role, as evidenced in my
discussion of her theory in the previous chapter; the notion that being close to death
reduces an individual’s claim against being killed, for instance, introduces a
consequentialist attitude. And none of Regan, Pluhar, or Carruthers would deny that
utilitarian considerations are sometimes appropriate: causing only a few individuals to
suffer a specific harm is undeniably preferable to causing many to suffer the same harm,
just as causing a single individual to suffer a lesser harm is preferable to inflicting a
greater harm on (either the same or another) individual.
Carruthers praises utilitarianism for its ‘governing conception’. In the previous
chapter I explained Carruthers’s use of that term, and proposed a governing conception
for Pluhar’s rights theory. The following is utilitarianism’s governing conception,
according to Carruthers:
95
1) The nature of morality: Morality is a set of principles aimed at maximizing
utility.
2) Moral epistemology: We come to know moral truths by empirically
determining how best to maximize utility.
3) The basis of moral motivation: There is a basic human feeling of sympathy for
others with whom we have contact. The faculty of reason motivates us to broaden
our sympathy into consideration beyond our immediate surroundings, to universal
consideration.
Carruthers thus determines utilitarianism to have a theoretical foundation equally
persuasive as his own contractualism or—given my analysis in the foregoing chapter—
Pluhar’s rights theory.
93
Peter Singer, “A Utilitarian Population Principle,” in Ethics and Population, ed. Michael D. Bayles
(Cambridge, Massachusetts: Schenkman Publishing Company, 1976), 85.
94
Regan, 310.
95
Carruthers, 26-27.49
Utilitarianism’s theoretical foundation is compelling. The reason behind its
rejection by opponents is therefore not the foundation, but the alleged failure of its
application. We turn now to its application.
Utilitarianism and the ASO
It should be noted that I have not yet addressed a specific utilitarian theory, but
only a generalized version, without a specific definition of utility or an explanation of
how it is to be maximized. Nonetheless, we can use the generalized version to produce a
partial answer to the ASO. For whether utility can be measured in terms of pleasure,
happiness, desire-satisfaction, or well-being, the utilitarian must accept the second
premise of the ASO: Some nonhuman animals possess sets of morally relevant
characteristics similar to those of some humans. Some nonhumans are capable of feeling
pleasure and pain, of having desires, and of having a well-being to the same degree as
some non-rational humans or even full persons. Therefore, the utilitarian must accept the
conclusion of the ASO: some nonhuman animals and some humans are similarly morally
considerable.
Now comes the question: and how morally significant is that? We saw that
contractualism was led to the uncomfortable answer: not at all. The rights theories
delivered more compelling responses, and in absolute terms: all purposive agents (or
subjects-of-a-life) are equally morally considerable. The utilitarian response is typically
gradualist: the moral significance of an individual is contingent upon the extent to which
that individual contributes to overall utility.
96
If there are three individuals, two
possessing five units of utility (regardless of how that might be defined) and the
remaining one possessing ten utility-units, then the moral significance of the third
individual is equivalent to that of the other two.
This is an admittedly simplistic utilitarian response, but it concurs to some extent
with contemporary popular moral convictions about the relative moral status of huma ns
and animals. As long as animals are considered to have a lesser capacity for possessing
nonmoral goods, we can view them as less important, morally, than humans. However,
96
It might be more appropriate to say that moral value is contingent upon the extent to which the individual
can contribute to overall utility. At this point, however, I am merely establishing a simplistic utilitarian
starting point.50
the ASO demands that any humans relevantly similar to those animals are similarly less
valuable. It is on this point that the utilitarian view is sharply opposed to the popular
view. This situation is a natural result of the ASO, which demands similar consideration
of similar beings, despite the contemporary view of a large divide between all humans
and all animals. There are two directions, each unpopular, that the defender of the ASO
can take: either hold that some animals are much more morally valuable than commonly
thought, or hold that some humans are much less morally valuable than commonly
thought. Both paths have been defended; R. G. Frey is famous for taking the latter
route.
97
I, like Singer, take the first route. Of course, both routes may be taken in
combination, and, as we will later see, Singer’s and my views may be perceived to do so.
Now we shall move beyond the simplistic, generalized utilitarian answer to the
ASO, and toward the specific utilitarian formulation I want to propose.
Varieties of Utilitarianism
The utilitarian’s fundamental principle is ‘maximize utility.’ There are therefore
two questions that one must answer in order to arrive at a specific utilitarian theory: 1)
What is ‘utility’? and 2) How does one ‘maximize’ it?
UTILITY
What is utility? Classic utilitarians such as Jeremy Bentham and John Stuart Mill
refer to happiness or pleasure as utility. Mill calls the utilitarian principle the “greatest
happiness principle,” and describes happiness as “pleasure and the absence of pain.”
98
But happiness and pleasure are not merely to be understood as physical pleasure; Mill
stresses the importance of intellectual pleasures as well, noting that they are generally
considered preferable to ephemeral physical pleasures.
99
Nonetheless, the classic
formulation has come to be understood as hedonistic utilitarianism, due to its focus on
pleasure and pain.
97
See his line on vivisection in his Rights, Killing, and Suffering (Oxford: Basil Blackwell, 1983), 115-116.
98
John Stuart Mill, Utilitarianism, Oskar Piest, ed. (New York: Macmillan Publishing Company, 1957),
10.
99
Ibid., 11-15.51
Peter Singer’s formulation is a departure from hedonistic utilitarianism. His
formulation, called preference utilitarianism, refers to interests rather than pleasures:
actions resulting in the satisfaction of interests are right actions, and actions resulting in
frustrated interests are morally wrong.
100
However, preference utilitarianism is not
drastically dissimilar from classic, hedonistic utilitarianism. After all, our preferences
and our happiness/pleasure tend to go hand-i n-hand. It is unusual to hold a preference for
something the obtaining of which does not bring us pleasure, or at least decrease pain.
Singer is aware of the similarity between the two formulations, and suggests that they
may even be identical.
101
There are a whole host of standard objections to utilitarianism that are applicable
to either utilitarian formulation. One such objection targets the idea that the good is what
we happen to prefer, or what makes us happy. The philosopher Robert Nozick captures
this objection in his discussion of an ‘experience machine’.
102
This imaginary machine
completely removes you from reality, but produces in your mind the experiences of doing
the things you love, and fills you with a sense of extreme happiness and well-being. If
the resulting happiness would outweigh the happiness of living in the real world, one
would think that a utilitarian would recommend that you plug yourself into the machine.
(We are assuming that the necessities of staying alive in the real world—food, shelter,
etc.—are accounted for.) This conclusion, however, tends to fly in the face of what is
understand to be morally good activity.
This objection works best against a truly hedonistic notion of utility, as it
presupposes that what we desire is merely a sense of well-being, and not actual wellbeing. If being happy, being satisfied emotionally, is the true end of our desires, then this
objection carries significant weight. However, preference utilitarianism does not lead us
to this view. What we really prefer (or most of us, anyway) is our actual well-being, and
therefore the experience machine would not satisfy our preferences. Of course, the
machine would seem to satisfy them, but then all we conclude is that plugging into the
machine seems to be morally right.
100
Singer, Practical Ethics, 2
nd
ed., 94.
101
Ibid., 14.
102
Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, Inc., 1974), 42-45.52
A stronger objection, directed at preference utilitarianism, is that we do not
always prefer what is good. I may in fact prefer to use the experience machine: maybe I
am a lazy person who would rather feel good than work to achieve real goals. Worse yet,
a person might prefer to do bad things: someone might enjoy lying, stealing, or killing.
Should these preferences be counted towards overall utility?
One response to this objection is to bite the bullet, and agree that people who
prefer to use the experience machine should do so, all other things being equal. And
people who prefer to harm or kill others ought to do so as well, all things being equal. Of
course, in the latter instance all things will not be equal, since the harm inflicted
(assuming the victim is not a masochist) will make the action morally wrong. But insofar
as any action satisfies a preference, it is a right action.
Another response to the objection is to refer to ‘rational preferences’ rather than
our actual, often irrational preferences. Spelling out the notion of rational preferences is
no easy task, but many utilitarians work in that direction in order to avoid the
counterintuitive consequences of a simpler notion of preferences. Generally speaking, a
rational preference is a preference developed with perfect knowledge about one’s wellbeing and long-term goals. If it turns out that the preference of using the experience
machine, or of harming others, is irrational, then those preferences do not count toward
real utility. We see here a way for the utilitarian to deny that all preferences are good,
regardless.
The view of utility I propose is similar to the rational preferences model, and even
a model based on interests, as long as ‘interests’ are spelled out a certain way. My
utilitarian formulation is ‘welfare utilitarianism’, and equates utility with actual wellbeing, rather than preferences or perceived well-being. I am not the first to put forth this
formulation. Robert E. Goodin is among those who support this view of utilitarianism,
and he notes that the seeds for welfare utilitarianism can be found in Mill.
103
The
welfare-utilitarian considers the moral patient’s well-being, what will benefit the life of
the moral patient, rather than what the individual happens to desire. Of course, the
individual’s preferences may largely coincide with what is truly in the individual’s
103
Robert E. Goodin, “Utility and the Good,” in A Companion to Ethics, ed. Peter Singer (Oxford:
Blackwell Publishers Ltd., 1993), 243.53
interest. We often develop long-term plans under the impression that the success of the
plan will result in the kind of life that is good for us. However, in cases where a
preference is in conflict with what is in our interests, we ought not satisfy that preference.
I have used the term ‘interests’ to mean something different than Singer’s notion
of interests, which appears to be synonymous with preferences. When I mention an
individual’s interests, or something being in an individual’s interest, I am talking about
the individual’s welfare. The same can be said for the goals of an individual. As a
welfare-utilitarian, I will be indicating those goals the attainment of which contributes to
well-being. Recall my version of the argument from purposive agency: the initial claim
is that I have goals that matter to me. Welfare utilitarianism points to a specific
understanding of this claim and the subsequent claims in the argument. For something to
‘matter to me,’ it must promote my well-being. Therefore, what I hold true is that I have
goals that promote my well-being, and that the attainment of those goals is for that reason
good.
There are objections to welfare utilitarianism, and one objection indicates a
disadvantage it has compared to the classic or preference formulation. Whereas Singer
can determine and rank preferences simply by asking the individual possessing them, the
welfare-utilitarian cannot determine well-being by asking. Bear in mind, all of these
formulations encounter difficulty when utility is compared between two different
individuals. Even Pluhar’s and Regan’s rights theories suffer from this problem when
their harm principle is applied. But the welfare-utilitarian’s ability to compare two states
of well-being, even within a single individual, appears more limited than the preferenceutilitarian.
Goodin argues that the welfare-utilitarian is actually in an advantageous position
compared to other utilitarians, especially when comparing utility between two
individuals. He writes,
[T]he problem [of interpersonal utility comparisons] is a problem only for
hedonistic or preference utilitarians. They are the ones asking us to get inside
someone else’s head. Welfare-utilitarians, by abstracting from people’s actual
preferences, definitely are not. We can know what is in people’s interests, in this
most general sense, without knowing what in particular is inside their heads.
Furthermore, at some suitably general level at least, one person’s list of necessary
basic resources reads much like anyone else’s. Whereas preferences, pleasures 54
and pains are highly idiosyncratic, welfare interests are highly standardized. All
that goes a very long way toward helping to solve the problem of making
interpersonal utility comparisons.
104
Something similar can be said for comparing different states of well-being in a single
individual. Subjective harms such as pains can be measured in terms of how they detract
from the individual’s welfare overall, rather than trying to determine just how much
something hurts. This kind of measurement is what one makes when taking a child to the
doctor: the child may experience the pain of a needle, but the child’s welfare is served in
doing so, regardless of the magnitude of pain involved. (Of course, one could invent an
odd case where the pain was so intense that the child suffered long-term physical or
psychological damage as a result. But these consequences can be added to the utility
calculation without concern for exactly how the pain feels.) The welfare-utilitarian
therefore has a real advantage over many other utilitarians, and this is especially true
when we apply it to the ASO, since we are then dealing with individuals who very well
may not be able to communicate to anyone their specific preferences, or pleasures and
pains.
Singer’s preference utilitarianism leads him to consider morally relevant any
beings with preferences. This set of beings is virtually identical to the set of conscious
beings. (It is hard to fathom a conscious individual with no preferences—though one
might exist.) The welfare-utilitarian faces a more difficult challenge marking out the
morally relevant characteristic; it simply will not do to draw the line at individuals with a
welfare. After all, all living things have a welfare. For that matter, so do corporations,
works of art, kitchen appliances, and friendships. All of these things can flourish (or at
least work properly), be damaged, or be destroyed. An additional argument is required to
narrow the focus of morality to a reasonable set of individuals.
Consider the theoretical foundation of utilitarianism: I begin by considering the
value of my own well-being (or happiness, or preferences) and move to a universal
standpoint, from which I conclude that all relevantly similar individuals also deserve
consideration. In the first premise of the argument from purposive agency I necessarily
have a well-being, so that is one necessary characteristic. The fact that my well-being
‘matters to me’ is not a subjective claim under the welfare-utilitarian view, as I have just
104
Ibid., 246.55
noted. But there is another element in that premise as well: the ‘I’ itself. In recognizing
the value of my well-being, I am presupposing my own consciousness. It is my
awareness, combined with my having a well-being, that leads to my insistence that my
well-being be promoted. Without consciousness, I lose any reason to make the oughtclaim. Therefore, consciousness as well as possession of a welfare are required for moral
considerability. The foregoing argument narrows the focus of welfare utilitarianism to
the set of conscious, purposive beings. Incidentally, this set of beings successfully
includes those hypothetical beings who have a welfare but no preferences, beings who are
excluded by preference utilitarianism.
MAXIMIZATION
Given a concept of utility, how is it maximized? First and foremost, to maximize
utility is to generate the largest amount of nonmoral goods among all moral patients.
What counts as nonmoral goods depends on one’s concept of utility; the welfareutilitarian holds that the set of nonmoral goods consist of the fulfilling of welfare-goals.
For example, all other things being equal, a world with no one going hungry is preferable
to a world with hunger.
If we imagine a fixed population of moral patients, there is an obvious general
answer as to how utility is maximized: consider the total welfare of all moral patients,
and determine what change in total welfare would result from each of our choices of
action. The action that brings about the highest total utility is the morally correct action.
However, fixed populations are only commonly found in the short run, and many moral
decisions must consider lives coming into and out of existence. For this reason,
calculating utility becomes more complex, as does calculating how to maximize it.
One formulation of maximizing utility is the ‘total view’. On this view, utility is
totaled regardless of the number of moral patients contributing to that total. The total
view produces some readily acceptable results: for instance, killing a moral patient is bad
to the extent that doing so decreases total utility. The converse result, that increasing
utility by bringing more lives into the world is morally right, is more controversial. Later
on I will address specific criticisms that attack this aspect of the total view.56
I believe the total view correctly describes how utility ought to be maximized, but
utilitarians dissatisfied with the total view’s implications have proposed the ‘average
view’. According to this view, utility is measured as an average among all moral
patients, regardless of the total number of individuals. While bringing one more
individual into the world may increase total utility, the average utility may decrease as a
result of overpopulation or some other factor. In this case, the average view tells us not
to introduce more individuals into the world. This result lends credibility to the average
view.
However, there is another aspect of the average view that renders it unacceptable.
Given a world with average utility U, the average view must hold as morally wrong the
introduction of a new being with utility below U, even if that being has a life worth
living.
105
The rightness of bringing another being into the world therefore depends on the
current well-being of the world’s inhabitants, a counterintuitive situation to say the least.
The fact that the dependency is such that individuals in a world filled with miserable
beings are morally right to add another miserable being, as long as the new being is at
least slightly less miserable, whereas individuals in the best-off world are prohibited from
adding another very well-off being who is slightly less well-off than average, reveals the
absurdity of the average view. We should therefore prefer the total view, which faces its
own criticisms but does not generate undeniably unsatisfactory principles.
Under both the total and average views, the distribution of goods is irrelevant. All
that matters is the aggregate of nonmoral goods, regardless of whether it is totaled or
averaged. Therefore, if we can generate the largest aggregate of nonmoral goods with an
unequal distribution—even if only one individual benefits while all others suffer losses—
then that distribution is morally preferable. This fact about utilitarianism has been the
target of many of the theory’s opponents, as I will discuss in the next section.
Criticisms of Utilitarianism
In this section I review major criticisms against both utilitarianism and the
utilitarian’s answer to the ASO.
105
Derek Parfit makes this point in his Reasons and Persons (Oxford: Clarendon Press, 1984), chapter 18.
We are assuming that the well-being of others does not increase as a result of the introduction of the new
being to the point where they counterbalance the utility of the new being.57
STRICT DECISION MODEL
Carruthers raises a general, theoretical objection to utilitarianism.
106
Unlike his
contractualism, under which moral agents agree upon a set of moral principles,
utilitarianism produces one and only one duty: to maximize utility. Actions are therefore
divided into two categories: obligatory and prohibited. According to Carruthers, there is
no category of permissible actions. Moreover, there is no possibility of supererogatory
action; our duty is to maximize utility, and we cannot rise above the call of that duty.
This simplistic nature of duties fails to give us what we need from a moral theory, and so
utilitarianism must be rejected.
This objection is a reaction to the utilitarian’s focus on states of affairs: from any
given point in time there are a set of possible future states of affairs, and the utilitarian
demands that we aim for the best state of affairs. The contractualist, on the other hand,
holds that “the lives or interests of individuals cannot be interfered with merely to
subserve greater overall utility,” and leaves open “a substantial domain within which they
remain free to get on with their own lives, and to develop their own concerns and
interests, without direction from morality.”
107
Because Carruthers (as well as Regan and
Pluhar) concentrate on negative rights—the right not to be harmed or killed, the right to
be free to pursue one’s goals—utilitarianism appears to face a unique problem here.
One utilitarian response is hinted at (but rejected) by Carruthers: part of
maximizing utility includes allowing individuals to be free to pursue their own interests
and desires.
108
As a welfare-utilitarian, my focus on the well-being of individuals
requires the acknowledgement of the kind of things we are. Without a biological and
psychological understanding of humans and animals, it is impossible to determine what
constitutes their welfare. Psychologically, it is clear that humans who are moral agents
cannot be made to operate as utility-calculators, to “run around doing good all the time,”
(as Carruthers puts it,) without detracting significantly from their welfare.
109
The
106
Carruthers, 32-34.
107
Ibid., 40.
108
Ibid., 33-34.
109
Ibid., 32.58
utilitarian adds this consideration into the moral calculus, and concludes that some
amount of freedom of action is absolutely necessary.
Carruthers finds this answer lacking. He replies that what the utilitarian is doing
is making it obligatory for moral agents to do what they want.
110
This is a strange
criticism. He seems to think that an obligation for him to do as he pleases—without
concerning himself with morality—is a coherent obligation. What the utilitarian
proposes is that some amount of our lives must be devoted to self-interest. After all, I am
often in the best position to know how to improve my own welfare. In addition to, and
often prior to moral considerations, I must look after my own health, shelter, livelihood,
and other necessities of life. The particular manner in which I secure these is very much
up to me; there is no obvious best way for any individual to live his life. That I look after
my own interests never feels like a moral obligation, despite the fact that I recognize my
own well-being as contributing to overall utility. Utilitarianism therefore does not
enforce free time at certain intervals upon moral agents; rather, it ‘obligates’ us to freely
choose whatever path we like to flourish as individuals. Real obligations only arise after
we are living in a manner that allows time for moral thinking.
Moreover, utility-calculations can hardly generate the sort of precision that would
always indicate one particular action as the clear means to maximize utility. Of course
there are clear cases. We certainly know that killing a moral patient (with a life worth
living) detracts from utility, and that saving them from being killed, when no similar
harm would result, prevents a loss of utility. But there are an infinite number of less clear
cases: the career path one takes, contributing to one benevolent charity instead of another,
going to one social gathering instead of another—these are all choices where we have no
way of gauging with precision the relative utility. It may well be that one choice is
preferable to all alternatives, but until we have perfect knowledge, there is substantial
moral freedom.
Something similar can be said about supererogatory action. If we do not know
with certainty the comparative utility resulting from two choices, but one choice detracts
significantly from one’s own welfare, then the less selfish choice can reasonably be
considered supererogatory. In fact, choices seen as supererogatory often have this
110
Ibid., 33.59
feature. If someone sacrifices her own life to save the life of another, the resulting utility
may be no higher than if she allowed the victim to die. But the decision to put someone
else’s welfare before one’s own is both morally good (utility-maximizing) and beyond
the call of duty (because utility-maximizing alternatives are available). Furthermore, the
choice reflects a virtue in the person making the sacrifice—selflessness—a virtue any
utilitarian would seek to promote in moral agents. Here we see how utilitarianism
provides substantial room for freedom of action and supererogation, without turning
either into a strict duty.
On this issue we can even turn the tables on the anti-utilitarians who bring up this
charge. To the contractualist, we might ask why the imaginary bargainers would not
agree to principles aiming at the best possible world. If maximizing utility is nothing less
than bringing about the best possible world (and that’s really the whole point), then why
would we adopt principles that do less? The notion that we ought to respect rights, even
when many are worse off as a result, is a common view but one that I think needs serious
reexamining. Much has already been said in this regard by Singer, who has written a
great deal in criticism of the emphasis on principles respecting rights.
111
FAILURE TO PROVIDE ADEQUATE PROTECTION, PART ONE
The bulk of the objections against utilitarianism indicate applications of the
theory that fail to provide adequate protection for individual beings. Rights theories and
contractualism aim for stronger principles against harming others, regardless of whether
these principles would maximize utility. I have already noted that something seems
amiss in a moral system that aims at something other than the best possible world. But
this is certainly not enough. More must be said in order to address specific objections to
the application of utilitarianism.
One objection is made by Carruthers, and involves the relative moral value of
humans and animals. He introduces a thought-experiment wherein you enter a burning
house, and find one human unconscious on the floor and five dogs locked in cages.
112
You estimate that you only have enough time either to drag the human out, or release the
111
See for instance his Practical Ethics, 2
nd
ed., chapter 8, especially 232-246.
112
Carruthers, 9.60
dogs from their cages and lead them out. He writes, “no one would maintain that you
ought to place the lives of many dogs above the life of a single human,” and claims that
there is a “common-sense belief that human and animal lives cannot be weighed against
one another.”
113
Therefore, any theory that suggests otherwise is highly suspect, and we
have good reason to dismiss it.
It is worth pointing out that Carruthers’s use of this argument to bolster his
contractualist theory, and then to apply that theory in order to persuade us of his beliefs
about the moral value of animals, is an undeniably circular strategy. However, if the
popular moral belief he appeals to is “central to morality,” as he claims it is, then perhaps
he makes a reasonable case.
114
I seriously doubt, however, that he has done so.
First, is it really true that we cannot weigh human and animal lives? When an
animal research lab claims that the suffering and deaths of their animals are justified by
the benefit to human lives, is this not the weighing of lives against one another, even if
the result is in favor of the humans? If only one person, rather than many, would ever be
saved by killing millions of animals in research, it does not seem that their work would
be justified, or at least as justified. Consider Carruthers’s thought-experime nt. What if
there were not five dogs, but five thousand? Or five million? I find it hard to believe that
the commonsense view is that the value of these lives never approaches the value of the
human’s life.
It was on this very point that I found Regan’s worse-off principle suspect. Recall
that his principle instructs us to sacrifice an infinite number of animals in order to save
one (normal) human. Of course, the principle was suspect because Regan also holds the
view that all moral patients are equally morally valuable. Carruthers does not hold the
latter view; rather, he holds that animal life has no inherent value whatsoever, a view that
leads (as we saw) to its own set of problems.
Pluhar, on the other hand, insists that the right action is to save the five dogs
rather than the human. She accepts the worse-off principle, but does not agree with
Regan that animals are harmed less by death. Given that the lives of animals have any
moral value, it appears there are three possible answers as to how to weigh their lives
113
Ibid., 9.
114
Ibid.61
against human lives. We can view their lives to be equal (Pluhar), view their lives as
infinitely less valuable (Regan), or recognize a sliding scale of value, with normal
humans more valuable—but not infinitely more valuable—than animals.
I believe the third route is the proper choice under utilitarianism, though a
utilitarian could conceivably argue for any of the three. As a welfare-utilitarian, my
evaluation of the life of any individual comes out of an analysis of its welfare. Now,
there are two ways one can understand comparative overall welfare of individuals. One
choice is to take the view of Pluhar, who argues that in death, all is lost for the victim,
and so each life is equally valuable.
115
Her assessment is derived from her theoretical
foundation, which bases moral value on the fact that an individual cares about his or her
well-being. The other choice is to view overall welfare in such a way that what is lost in
death is an amount that can be compared across individuals. That is to say, yes, the
victim loses everything in death, as Pluhar puts it. But the amount signified by
‘everything’ depends on the individual: some have more to lose than others. On this
latter view, some lives are indeed more morally significant than others.
The idea that individual lives can be more or less valuable surely agrees with
popular morality when it comes to weighing lives in Carruthers’s thought-experiment, but
we are far less likely to want to compare human lives in this way. Pluhar is firmly
against the notion that lives are more or less valuable, though she did make exceptions for
the very old or near death. Yet, if we can weigh human lives based on the comparative
time one has left to live, why not on the quality of that time as well? I see no reason why
we cannot justifiably make such comparisons.
We must be cautious not to take this idea too far, as there are clearly dangerous
implications. R. G. Frey, as I have mentioned, has accepted the ASO and its demand that
we view animals and relevantly similar humans as having lives of similar moral value.
Frey’s response, as a proponent of animal experimentation, is to accept similar
experimentation on relevantly similar humans.
116
But Frey bases his response on his own
utilitarian formulation, which rests on a sophisticated notion of interests. According to
Frey, animals and non-rational humans lack these interests to such an extent that their
115
Pluhar, 293.
116
Frey, 115-116.62
lives are far more expendable than normal humans. As a welfare-utilitarian, I do not
recognize a wide gap. There is a gradual scale, surely, but the differences between the
welfare of a normal human and the welfare of a ‘higher’ animal or similar human are
quite small.
I do not see an alternative to the gradualist answer without accepting the unlikely
view that the simplest conscious organism has a welfare of equal value as dogs, apes, and
human beings. Many philosophers rely on a method of cushioning gradualism: they
postulate an imaginary line dividing moral patients on the basis of one or another morally
relevant property. Those individuals above the line are considered maximally morally
considerable; those under the line are placed on the gradualist scale. Under a moral
theory centered on rational agency, rationality may provide the basis for that line. Singer,
to some extent, uses the preference to continue one’s life as the line: individuals with no
conscious preference regarding their own lives fall below the line.
117
Philosophers have
also drawn the line such that all rational agents, as well as all humans (including nonrational agents), are above the line; however, such divisions have been found to be
unjustified, as I mentioned in chapter 2. The welfare-utilitarian does not have access to
any such property to serve as a magic line.
However, I believe some sort of line ought to be imagined nonetheless. Because
the welfare of one human being is impossible to compare to most others, it is best to
consider them equal. This applies to many non-rational humans as well. After all, the
richness of one’s experience is merely one component of one’s welfare. The web of
friendships and other relationships is another. Other components include one’s lifestyle,
moral inclinations, access to necessary resources, and physical health. Facts about an
individual’s mental life surely cannot give us a complete understanding of the
individual’s welfare, nor is any final, precise assessment really possible. So, between
two highly complex animals (human or otherwise), there may indeed be a difference in
moral value. But we have limited means to determine it.
So, unlike Frey, I find reason to object to painful experimentation on many sorts
of animals, and on humans. This does not rule out the possibility that some may be
117
Singer, Practical Ethics, 2
nd
ed., 95-99. I will have more to say on Singer’s position in a later section
where I discuss his ‘replacement argument’.63
justified, but I will discuss this further in the next section, when I reply to the charge that
utilitarianism fails to sufficiently protect many basic rights.
To return to Carruthers’s objection, I think it is the dogs who ought to be saved, as
long as we ignore all welfare considerations outside the six individuals in the thoughtexperiment. In the larger picture, we need to also be aware of the welfare of those whose
lives were tied in with the human, though we must do the same for the dogs. I think it is
safe to assume that in most (but certainly not all) cases the death of five dogs is not going
to have the same repercussions on others as the death of a human. Moreover, the normal
human is a moral agent and, if he is not morally corrupt, might be expected to do more
good in his lifetime than the dogs in theirs. Also, like Pluhar, I see the length of one’s
life as relevant. If the human was young, the longer life expectation becomes another
factor in the calculation.
But now, no one is going to expect the person running into the burning house to
work all this out. No utilitarian insists that a utility-calculation be performed before
every action, especially in an emergency. Sometimes it is necessary to rely on past
reflections to produce an intuitive judgement, and this judgement may very well be to
save the human. If the number of dogs were greater, perhaps the person would choose
otherwise. In no circumstance would I find the person blameworthy; morality does not
provide a final answer to every possible dilemma, a fact resulting from the limitations of
human reason. All told, I think the welfare-utilitarian handles Carruthers’s thoughtexperiment more successfully, and more in keeping with some of our commonsense
moral ideas, than Carruthers himself does.
FAILURE TO PROVIDE ADEQUATE PROTECTION, PART TWO
Carruthers, Regan, and Pluhar all criticize utilitarianism for its failure to protect
rights. When the details of these criticisms are spelled out, I think the utilitarian has the
necessary resources to counter their charges.
Regan argues that the utilitarian treats the individual as a mere means to the end
of maximizing utility.
118
In doing so, the individual is not treated as an end-in-itself, the
Kantian notion of what it is to be morally valuable. The individual does not matter for
118
Regan, 311.64
his or her own sake, but for the sake of utility. Regan describes the utilitarian
understanding of individuals as “mere receptacles,” valueless containers of utility.
119
As
valueless containers, their individual welfares are not important as long as the aggregate
utility is maximized. Therefore, the utilitarian would justify shifting all utility into one
individual, if slightly more utility is created in the process. If utility is already
maximized, no increase in utility would even be necessary to permit such a utility shift.
To illustrate his point, Regan imagines a situation where killing one innocent
victim would maximize overall utility.
120
Without the notion of a right to life, the
utilitarian must permit the killing. What is worse, the utilitarian must find the killing
morally obligatory. Similarly, a principle against lying is good only insofar as it
maximizes utility. If by lying I can improve my own welfare without harming anyone to
an equal degree, then lying is a morally right action. So it goes for all principles of rights
theories; as long as utility is maximized, any rights or principles may be ignored. Pluhar
makes a similar point: if two actions will maximize utility, but one action involves killing
an innocent individual and the other does not, either action is morally acceptable.
121
Carruthers makes it plain: “Apparently anything can be done to a person, provided that it
produces more utility (either total or average) than any alternative course of action.”
122
A frequent utilitarian response is to appeal to side effects. There are several ways
of doing this. For instance, one can appeal to the effects the killing of an innocent
individual will have on the individual’s friends and loved ones. Of course, this appeal
will not work if the individual has no friends or loved ones. And this defense will not
work anyway: we can still suppose that utility is maximized despite any additional
suffering. Another possible side effect is the bad character of the killer. The utilitarian
can argue that even if the killing maximizes utility in this one instance, a person capable
of calculated killing reveals a bad character. In the long run, utility will not be
maximized if we permit bad characters to flourish, so we ought not allow the killing. But
again, this response will not succeed: there are a variety of ways to reject the appeal to
bad character. Perhaps the killer is near death or is otherwise removed from the company
119
Ibid., 205-211. This description is not without merit: Regan credits Singer with the ‘receptacle’
terminology.
120
Ibid., 202-204.
121
Pluhar, 182-183.
122
Carruthers, 28.65
of others, and will not have time for his bad character to cause any harm.
123
Or perhaps
the character of the killer is so well developed that he can distinguish between situations
where killing is justified and situations where it is not. This is no extraordinary claim.
Nations routinely send soldiers out to kill the enemy, but still welcome them back into
society without fear that they have become dangers to society. The response to any
appeal to side effects is the same. Side effects can be accounted for, but as long as utility
is maximized, they will not significantly affect the utilitarian’s evaluation of killing,
lying, stealing, or any other actions strongly prohibited by rights theories.
Another response to the rights theorist’s challenge is to propose the adoption of a
rights scheme as the best means of maximizing utility. Pluhar discusses one such attempt
in her criticism of utilitarianism, and calls the strategy “the best case that can be made for
genuine rights within a utilitarian framework.”
124
This defense begins with a point I
made earlier, that moral agents cannot be expected to make utility calculations quickly
and with precision. Add to this premise the claim that some choices must rely on taking
certain risks, and that what is at risk is sometimes extraordinarily bad. For instance,
suppose that a group of terrorists are planning an operation to kill many civilians. You
are an officer in the military and intelligence reveals that there is a likelihood that the
terrorists are holed up in a certain area containing innocent civilians. Suppose further
that you are assured that the terrorists will disappear if they detect any military presence
in the area. The only way to defeat them would be a remote strike on the area using
missiles or bombs, which will kill innocent civilians as well. Now, the simplistic
utilitarian calculation might be to weigh the number of innocent casualties lost in the
missile strike with the number of lives threatened by the terrorists. If fewer lives are
involved in the missile strike, then we should launch the strike. However, this simplistic
calculation omits the very important risks involved. Perhaps the terrorists are not hiding
there after all. Or, even if they are, perhaps their plans can be foiled in another way, with
less loss of life. The addition of these risk considerations generates a strong presumption
against launching the missile strike. After all, the strike is guaranteed to generate a great
amount of disutility; whatever disutility is prevented is merely a matter of probability,
123
This is suggested in Carruthers, 32.
124
Pluhar 219-221. The particular argument she presents is borrowed from L. W. Sumner.66
and the precise level of probability is only guesswork. The presumption against killing,
in this instance and others with relevantly similar choices, can be construed as rights held
by the innocent individuals. This notion of rights, as strong presumptions against certain
harmful actions, matches up with our common use of the term.
There is merit in this argument for deriving rights from utilitarianism. I think it
can be strengthened by additional considerations regarding the welfare of individuals.
First, there is a frequent misconception about utility calculations, which is captured in
Regan’s criticism that individuals are mere ‘utility containers’ mattering only insofar as
the utility they hold. The idea is that units of utility can be moved between individuals,
even if one individual ends up with most or all of them, and that this would be preferred
if utility was maximized in doing so. To an extent this is a correct characterization, in the
sense that total utility is what counts, in the end. However, one tends to ignore the
complexities contained in the notion of utility. Utility cannot be exchanged like
currency: taking something valuable away from one individual and giving it to another
does not necessarily maintain a constant amount of utility. This is plainly true in the case
of wealth in our society, where more equal distributions allowing more people to rise
above the poverty line result in greater utility than more unequal distributions. The rising
out of poverty simply counts for more than, say, the rising from a comfortable income to
an even higher income.
What we have here is a distinction between trivial and necessary nonmoral goods.
A trivial good contributes to welfare, but not to any important extent. Trivial goods are
often expendable, and easily replaced by similar goods. Necessary goods, on the other
hand, are things like food, shelter, and adequate resources for a healthy life. Necessary
goods often have no suitable replacement, and cannot be sacrificed without extreme
consequences for the individual’s welfare. There is a difference between trivial and
necessary goods other than one of quantity; there is a qualitative difference. The use of a
limb, for instance, is simply is not the same kind of good as excessive wealth. No one
would reasonably trade the use of a limb for a great deal of money, unless the person was
so destitute that similarly necessary goods could be acquired thereby. Trivial goods,
then, are worth infinitely less than necessary goods. There is simply no amount of trivial
goods that can warrant the loss of a necessary good. This is an argument against the 67
notion that utility can easily be exchanged between individuals: if the exchange results in
a loss of necessary goods by one individual and no gain in necessary goods for another,
then the exchange results in great disutility.
There is a related point. Let us judge my losing an arm to equal ten units of
disutility. Therefore, my losing an arm and a similar person losing his arm will add up to
twenty units. The idea of ‘utility containers’ might suggest that I could keep the disutility
constant by transferring all twenty units to the other person, costing him his other arm
and leaving mine intact. But this is not right. Being without both arms is substantially
more than twice as bad as living with one, due to the much more extreme handicap it
causes to an individual. The same is true for necessary goods in general: the loss of
multiple necessary goods adds up to more than the sum of those losses taken individually.
Therefore, we ought to favor a wide distribution of harms among many individuals over a
concentration of the same number of harms on just a few.
This principle comes fairly close to the worse-off principle favored by Regan and
Pluhar. In fact, it can be seen as more successful. Recall that the worse-off principle
prevented us from harming any one individual rather than harming many individuals to a
lesser extent. I have just showed how utilitarianism generates a similar principle via two
methods, a distinction of trivial and necessary goods and a claim regarding the additional
disutility of multiple harms. However, the worse-off principle suffers from problems that
mine does not. For instance, the worse-off principle gives us the same result concerning
my lost limb example above, but would further suggest that it is preferable to cause a
million people to lose a limb than to cause one person to lose two. Even worse, it makes
it preferable to cause a million people to lose a limb than to cause one person to lose a
limb and a finger | Or a limb and a dollar (or any other nonmoral good). One expects
there would be some point where it is acceptable to cause a greater harm to prevent many
lesser harms, at least when both harms involve necessary goods. The utilitarian can
accept this possibility, but a rights theorist who accepts the worse-off principle cannot.
The claim by opponents of utilitarianism that the aggregate utility is all that
matters, regardless of which individuals ‘contain’ the utility, misses a general point about
how welfare is measured. An individual’s welfare is bound up in the individual, and can
only be measured with respect to that individual. The welfare-utilitarian does not simply 68
add up all the happiness (or preferences) in the world to generate a measure of utility;
rather, it is essential to take each individual separately, consider what sort of being it is
and what comprises its welfare, and then evaluate the individual’s overall well-being.
Now, after considering the individual, utility ought to be aggregated. But this summingup does not prevent the individual from being the primary locus of moral consideration.
Rather, it reflects the fact that one individual’s welfare is only as important as that of any
other moral patient, and no more.
Despite these arguments, rights theorists are likely to object that rights remain
inadequately protected. Consider my earlier example of the choice to kill innocents in
order to stop terrorists. In the example I noted that the chance that utility would be
maximized was unknown, but that we should not guarantee the loss of life in order to
prevent a merely possible loss of life. But what if we were sure that the terrorists were in
the area, and what if they were guaranteed to kill many more people if we did not launch
a strike? In the unlikely event that these facts are absolutely known to us, then the
utilitarian would launch the strike.
Pluhar seems to find this result unacceptable. She mentions a parallel situation,
where medical experimentation is guaranteed to prevent more harm than it causes. “Even
if we have decided that sentient nonhuman animals have a right to have their interests
respected, a case could be made [under utilitarianism] for painful, fatal experimentation
on them if overall utility would be maximized.” She goes on to say, “[T]he same would
apply to humans.”
125
Her objection is that utilitarianism justifies rights violations too
easily. Under her rights view, medical experimentation, regardless of benefits, is wrong.
Her argument that the utilitarian fails to provide this stronger kind of protection is
correct. But how weak is the utilitarian’s protection? Pluhar asserts that the utilitarian
prohibits experimentation unless “overall utility would be maximized,” but she does not
provide an example of when this would be the case. I argue that the case is rarely, if
ever, actualized. We can never guarantee the results of experimentation—that is why
they are experiments | If, somehow, we knew that more lives would be spared pain and
death, and we knew that the injury and death spared would certainly outweigh the harms
of experimentation, then I think painful, fatal experimentation would be justified. But it
125
Ibid., 222.69
should be emphasized that this is a highly extraordinary, if not utterly implausible
situation. Under more realistic conditions, utilitarianism has no problem extending the
kind of protection the rights theorist wants.
REPLACEABILITY
A final challenge to utilitarianism is the ‘replacement argument’. Pluhar uses the
notion of replaceability in order to reveal a weakness of utilitarianism.
126
The argument
is supposed to show that one individual can justifiably be replaced by another, so long as
utility is maximized in doing so, and can be understood as follows:
1. Relevantly similar moral patients possess lives of similar moral value
(measured in utility).
2. The existence of one such moral patient is therefore equivalent to the existence
of another.
3. The death of an individual is measured as the loss of whatever utility the
individual possesses, as well as whatever harms the death causes in others.
4. Therefore, if an existing moral patient has a life worth living (i.e., counts as
positive utility), and the existence of a future moral patient depends on the
destruction of the existing individual, then we are permitted to kill and replace the
first individual with the second, assuming no extra disutility (aside from the harm
of death to the individual, but including the suffering of the individual before
death) occurs in the process.
The conclusion, if sound, represents a serious challenge to the utilitarian. Since future
lives add to total utility, they must be counted as part of the utilitarian’s moral evaluation
of actions. The argument is typically applied to such things as the raising of animals for
food. Since future animal lives are created by the same industry that kills them for their
meat, the replacement argument points the way toward preferring a meat-eating society.
This result is not terribly objectionable to many people in our society; however, the ASO
reminds us that we must apply similar results to humans as well. And those implications
are unacceptable to most.
What are the implications? First, let us note what is not implied. Our current
animal food industries are not justified by the replaceability argument, on the grounds
that the existing moral patients in factory farms do not have lives worth living. This fact
has been well documented, and there is little need at this point to rehash the description
of the suffering involved in factory farms. Pluhar claims that hunting (and then replacing
126
Ibid., 185-190.70
the hunted animals) is justified by the replacement argument,
127
but this is clearly wrong.
Hunting fails to satisfy the condition that no additional disutility attend the death of the
individual, disutility such as the fear and pain experienced by the hunted animal.
Replacing individuals is also not acceptable when other individuals will suffer as a result.
Human deaths tend to cause additional disutility in those that knew the victim, and the
same is true for some animals.
However, the argument does apply in cases where animals—or humans—are
raised humanely and killed painlessly, and where no others are harmed by their deaths.
According to the argument, a life as a well-treated food animal, or as a subject of painless
medical experiments ending in an early but painless death, is acceptable as long as the
individual is replaced to avoid a loss of utility. The additional utility generated by tasty
meat or increased chances of new medicines make the replacement choice obligatory,
though only marginally (trivial and probable benefits do not count for much), and only if
utility is actually maximized (other courses of action may do better).
Many find it highly counterintuitive to justify the killing of one individual by
bringing a similar individual into existence. But that justification is exactly what seems
to come out of the ‘total view’ of utilitarianism. Rights views, on the other hand, do not
have the same result: presumably, the existing individual has a right to life that
nonexistent beings do not, and so replacement is an unjustified rights violation. Pluhar
holds that the “upsetting” problem of replaceability by itself is enough to drive many
people away from utilitarianism.
128
There is a temptation for the utilitarian to bite the bullet on this issue. If
replaceability is permissible under very stringent conditions, then at a practical level it is
rarely a real worry. Humans are in even less danger than animals of real replacement,
since popular sentiments will not condone any killing of humans, regardless of whether
they are replaced. Secret killings, of course, would bypass that problem. In the end, the
uncertainty that any practical replacement industry will really maximize utility is
probably enough to warrant a utilitarian prohibition on it.
127
Ibid., 185.
128
Ibid., 190.71
Nonetheless, the opponents of utilitarianism stress that the replaceability issue
indicates a theoretical flaw within the utilitarian theory, and specifically the total view.
The theoretical problem is that utilitarianism seems to allow the goodness or rightness of
creating one individual to cancel out the badness or wrongness of killing another
individual. Our common sense, of course, tells us that the wrongness of killing far
outweighs the rightness of creating a replacement individual.
One attempt to escape the replaceability problem is the ‘prior-existence view’.
129
This is an alternative to the total view, and provides moral consideration only to existing
beings, not future individuals. Singer held this view, but has since discarded it.
130
The
prior-existence view successfully explains the notion that bringing an individual into
existence is not tantamount to benefiting that individual, whereas killing an existing
individual counts as a harm. This view, however, is too flawed to accept. Quite simply,
it is not acceptable to exclude future generations from the moral community. An
exclusive focus on present individuals, with no regard for future generations, is a
shortsighted morality. For instance, under the prior-existence view there would be no
reason to protect the environment, so long as the conditions did not greatly deteriorate in
our own lifetimes. And how can it be that future generations matter in the future, but
deserve no consideration now?
A specific thought-experiment has been used by Singer and Pluhar to illustrate the
problem with the prior-existence view.
131,132
Pluhar calls it the ‘wretched child’ scenario,
and it runs as follows. Suppose that a couple is considering having a child, but they are
told by a doctor that there is a very high probability the child would be afflicted with a
birth defect that would render the child’s life miserable and brief; the child would have a
life not worth living. The couple has a very good reason not to conceive a child, and that
reason is the child’s suffering. However, this child does not yet exist. Under the priorexistence view, there would be nothing wrong with conceiving the child up until the point
that it exists. Once the child exists, it would be wrong to let the child live. Therefore, the
prior-existence view permits the conceiving of a miserable life, but obligates us to
129
Ibid., 190-193.
130
Peter Singer, Animal Liberation, 2
nd
ed. (New York: Avon Books, 1990), 228. Also mentioned in
Pluhar, 190-191.
131
Ibid.
132
Pluhar, 193-195.72
destroy that life once it appears. A moral principle that allows this is absurd, so the priorexistence view must be rejected.
Singer’s answer to the replacement argument is to retain the total view, but under
his preference utilitarianism. Recall that preference utilitarianism measures utility in
terms of satisfied preferences, rather than happiness or pleasure. Singer argues that
complex beings such as humans have a preference for continued existence, and that this
preference is thwarted when the individual is killed.
133
A thwarted preference is not
compensated by the creation of new, initially unsatisfied preferences; an unsatisfied
preference is understood as negative utility, which disappears when the preference is
satisfied. (Whether we should construe a satisfied preference as ‘resetting to zero’ or as
positive utility is a matter for debate. Singer prefers the latter option.)
134
In a
replacement industry, individuals will consistently have preferences thwarted, so utility is
not maximized.
Such is Singer’s argument against replacing humans, and other animals capable of
a preference for continued existence. However, individuals with no preference for
continued existence gain no protection from this argument. Because conscious
preferences are the issue, individuals with no concept of a continued existence cannot
have a preference for continued existence. Therefore, individuals must be self-conscious,
reflectively aware of themselves, in addition to being merely conscious, to have the
preference for continued existence. We believe many animals lack self-consciousness.
These animals, therefore, are replaceable. Singer mentions fish as an example:
We can presume that if fish became unconscious, then before the loss of
consciousness they would have no expectations or desires for anything
that might happen subsequently, and if they regain consciousness, they
have no awareness of having previously existed. Therefore if the fish
were killed while unconscious and replaced by a similar number of other
fish who could be created only because the first group of fish were killed,
there would, from the perspective of fishy awareness, be no difference
between that and the same fish losing and regaining consciousness.
135
133
Singer, Practical Ethics, 2
nd
ed., 126-127.
134
Ibid., 128-129.
135
Ibid., 126.73
Many will find Singer’s distinction between replaceable and non-replaceable individuals
preferable to the application of the prior-existence view, and to that of utilitarian theories
that equate utility to pleasure or happiness.
There are still problems for Singer’s view. First, the ASO demands that humans
receive consideration equal to the consideration granted to relevantly similar animals. So
Singer must admit that since some animals are replaceable, similar humans are
replaceable as well. Some humans are not self-conscious. Newborn infants are certainly
in this class, as are the most extreme cases of mental retardation and senility. Singer
concludes that these humans are just as replaceable as the fish in his example.
136
This
problem is not fatal to Singer’s position, but many people find it highly objectionable.
On the other hand, Pluhar argues that the class of self-conscious individuals may be
larger than Singer realizes, and may include birds and other ‘lower’ animals.
137
If this is
true, fewer humans are replaceable as well.
Second, Pluhar argues that preference utilitarianism still fails to explain why we
cannot replace even a self-conscious individual.
138
If we kill one individual in order to
replace it with another, we can counteract the disutility of the former individual’s
unsatisfied preferences by satisfying the preferences of the new individual. Now, in the
case of self-conscious beings, the replacement individual cannot itself be replaced,
because the preference for continued existence must be satisfied in order to counteract the
previous disutility. Pluhar notes that this whole line of thought is strange, since death
comes to all of us, and therefore the preference for continued existence is ultimately
thwarted anyway.
139
Singer avoids this criticism by adding a level of sophistication to
the idea of a preference for continued existence: he describes self-conscious lives as
“arduous and uncertain journeys, at different stages, in which various amounts of hope
and desire, as well as time and effort have been invested in order to reach particular goals
or destinations.”
140
The preference for continued existence is therefore not as morally
significant at the end of one’s life as it is in earlier stages. “Towards the end of life, when
most things that might have been accomplished have either been done, or are now
136
Peter Singer, “Killing Humans and Killing Animals,” Inquiry 22 (1979): 153.
137
Pluhar, 203-4.
138
Ibid., 208-11.
139
Ibid., 209.
140
Singer, Practical Ethics, 2
nd
ed., 130.74
unlikely to be accomplished, the loss of life may again be less of [a] tragedy than it would
have been at an earlier stage of life.”
141
Pluhar notes that, at most, this strategy only
prevents replacement of younger individuals.
142
Moreover, she argues, the description of
self-conscious lives as uncertain journeys is a false generalization. Plenty of selfconscious beings have no complicated plans or long-term goals.
143
Most importantly, Singer has failed to show why the creation of new individuals
is not morally obligatory under the total view, regardless of how the notion of utility is
construed. It is a mistake to fall back on the prior-existence view, and hold that the
satisfaction of future preferences does not count in the moral calculus. Future
generations, and their preferences, do matter. Likewise, it is morally wrong to bring the
‘wretched child’ into existence. The total view, then, must be accepted. Yet, are we not
therefore obligated to bring a child into existence, under the condition that his or her
preferences are satisfied? Or, to take things to their final conclusion, should we not
create as many happy lives as possible? Singer admits that he has no satisfactory answer
to this question.
144
“If the pleasure a possible child will experience is not a reason for
bringing it into the world, why is the pain a possible child will experience a reason
against bringing it into the world? A convincing explanation of this asymmetry has not,
to my knowledge, been produced.”
145
I believe that the replacement argument can be successfully addressed by the
welfare-utilitarian. First, something should be noted about the moral ‘asymmetry’ Singer
notes in the creation of the two possible children, one happy and one miserable. Singer
presumes that the conditions of the children are symmetrical. The symmetry is actually
an illusion. In the case of the miserable child, nothing can be done (aside from
euthanasia) to relieve the child of misery. The child cannot be made happy, or have
satisfied preferences. The happy child, on the other hand, is not necessarily happy.
Parents generally succeed in raising children well enough that the children have lives
worth living, but surely not always, and the child’s life is never entirely positive. The
miserable child’s life, on the other hand, is always entirely bad. A situation truly
141
Ibid.
142
Pluhar, 210.
143
Ibid., 211.
144
Singer, Practical Ethics, 2
nd
ed., xi.
145
Singer, “Killing Humans and Killing Animals,” 148. His italics.75
symmetrical to that of the miserable child would require a child that cannot be made
miserable. Now, if the couple’s doctor predicted that a certain couple would produce
such a perfectly happy child, then perhaps it is not so radical to suppose that conception
would be obligatory | Singer’s asymmetry, therefore, has been given at least a partial
explanation.
The welfare-utilitarian does not have the same answer to the replacement
argument as the preference-utilitarian. A salient difference is that the welfare-utilitarian
will not place moral weight on the preference for continued existence. Even if an
individual has no concept of continued existence, the individual’s welfare depends upon
continued existence. Welfare is determined solely by investigating what sort of being the
individual is, not merely the individual’s particular preferences. Therefore, no distinction
between self-conscious beings and merely conscious beings can be highly relevant in
regards to replaceability.
To consider an individual’s welfare, it is not enough to determine how well an
individual is doing at the present point in time. One’s welfare is a measure of one’s life
as a whole. Living a full natural life, being free to develop one’s abilities, and flourishing
as the kind of thing one is all comprise an individual’s welfare. For the welfareutilitarian, then, replacement harms each individual’s welfare in a way that is not
counteracted by the addition of a new being. This is especially true when the replacee
will face the same harm as well. This line of thought is similar to Singer’s application of
preference utilitarianism to self-conscious beings, but, as I noted, the welfare-utilitarian
extends this consideration to all conscious beings, regardless of whether they have a
conscious preference for continued existence.
We can follow Singer’s view further, and hold that killing an individual nearer to
death is not as significant a harm. Pluhar, you will recall, also holds this view. In old
age, or in an otherwise near death condition, one’s welfare is already diminished. There
is simply less to lose. So perhaps replacement is permissible for near-death individuals.
This is not a particularly dangerous view to hold. For one thing, recall that the
replacement argument requires that replacement is only justifiable when the replacement
individual’s existence depends on the killing of the first individual. Replacement
scenarios that truly conform to utilitarian demands are hard enough to come by. One 76
would certainly be hard-pressed to imagine a situation where a near-death individual
must be killed, rather than allowed to die a short time later, in order to create a new life.
Finally, something must be said in response to the criticism that the total view
obligates us to increase utility by bringing more beings (with lives worth living) into the
world. This criticism applies to welfare utilitarianism, at least my version, which utilizes
the total view. In the present conditions in many parts of the world, overpopulation has
caused great suffering, and many people are calling for population control for that reason.
The same can be said for many nonhuman species, an example being deer overpopulation
in many parts of our country. We should therefore understand this criticism of
utilitarianism as applying only to a world where overpopulation is not a problem. In such
a world, is reproduction morally obligatory?
The answer is not simple. If we focus solely on the individuals who are produced,
and a life worth living is guaranteed, then the answer is yes. The total view requires us to
admit that much. And the admission is not counterintuitive: it makes perfect sense that a
world with more happy individuals is a better place than one with fewer. If we imagine
that there is another planet in the universe with a population of beings whose lives are
fulfilling, and we ask if this is a better universe because that planet exists in addition to
ours, few would say no.
However, there are other important considerations that affect the ultimate answer
as it pertains to this planet. First, an individual’s well-being is never guaranteed. In the
right environment, we can consider it likely that the life will be one worth living, so the
risk may not be substantial. But we must consider it, regardless.
Second, recall the asymmetry between the creation of a happy child and a
miserable child: the miserable child cannot be made happy, but the happy child must be
cared for in order to be happy. It is not the case that we can simply bring these happy
individuals into existence. When a new being appears, one or more individuals receive
the duty of looking after the individual’s well-being. This duty may be taken on with
pleasure, as it is with many parents, or it may be highly undesirable. When the duty to
care for a new individual is not desired, we must take into account the disutility generated
by the negative affect on the caregivers’ welfares. How much disutility is generated will
vary greatly, based on the impact the acquired duty has on each individual’s well-being. 77
Moreover, if the duty becomes so much of a burden that the caregivers relinquish their
roles, the child’s well-being is endangered as well.
Finally, even if reproduction is morally obligatory, enforcing reproduction does
not appropriately respect the kind of individuals people (and other animals) are. It would
certainly bring a different tone to family existence if the family unit was created out of
obligation. Our welfare, as I have mentioned, includes our free activity to pursue our
goals, flourish as individuals, and create the kind of lives that are right for us. Enforced
reproduction would surely detract from our welfare.
Problems with welfare utilitarianism’s answer to the replacement argument may
still exist. Why is the limited welfare of a near-death individual sufficient to permit
replacement, yet it is impermissible to replace someone with slightly more time left? I do
not know how one is to draw the line, but we should always err on the side of caution.
Perhaps replacement should always be (legally) prohibited, despite the existence of
exceptional cases where it is morally permissible. Such gray areas always exist for
gradualist moral theories, but gray areas are a common enough feature when applying
any theory that this particular instance is not terribly damaging to utilitarianism.
In contrast to classic utilitarianism and Singer’s preference version, welfare
utilitarianism provides an answer to the replacement argument that serves humans and
animals equally, and in a manner that maintains the utilitarian’s total view without
suffering from the typical obstacles of the total view.
Welfare Utilitarianism and a Final Response to the ASO
Near the beginning of this chapter I produced an initial utilitarian response to the
ASO. The conclusion of the ASO was accepted: some humans and some animals are
similarly morally valuable. Exactly how valuable different individuals are was left
undetermined. One could conclude that animals are more valuable than our current
attitudes suggest, or one could conclude that relevantly similar humans are less valuable.
As a welfare-utilitarian, I believe the former is correct.
Humans are incredible animals, capable of mental activities that no other animal
on Earth can perform. Humans create splendid works of art, vast cities, and even make
valiant attempts at philosophical reasoning. These activities certainly contribute to our 78
welfare, in a way that other animals cannot achieve. At the same time, I do not believe
that these high-order mental activities contribute so much to one’s welfare that the
welfares of ‘lower’ beings (non-rational humans included) pale in comparison. Taking
an evolutionary viewpoint, the complexity of human primates is only a few steps away
from our nearest surviving primate neighbors. Why should we think that the latest
evolutionary step has produced an altogether new and unique form of well-being? An
individual’s well-being is too complex, made up of as many diverse elements as it is, to
be compared to the welfare of any closely related individual, even across species borders.
Distinctions are not impossible to make, so long as the individuals are substantially
different and easily distinguished, in ways that are morally relevant. If Singer’s
description of ‘fishy awareness’ is correct, then the welfare of a fish is simple enough to
be looked after by a mind that fails to persist over time. Simpler animals such as fish
may in some sense be said to flourish over time, but not in the sense that a human, or
even a dog, cat, or pig may flourish. Mental richness does count for something, just not
as much as many moral theories propose.
In extreme cases, it is admittedly possible to compare the relative value of two
humans. An obvious case is an anencephalic human, who is born with no brain
whatsoever, only a terminating brain stem. Another case is a victim of an irreversible
coma. These extreme cases clearly lack a welfare equivalent to a normal human. There
is simply less of an individual present in those cases for us to consider, much less than in
many animals. In the case of brain stem babies, there is no consciousness to speak of.
Welfare utilitarianism provides no moral consideration whatsoever for such individuals.
Neither, for that matter, do Regan’s and Pluhar’s rights theories, which similarly draw the
line at consciousness.
Unlike Regan’s answer to the lifeboat scenario, in which he is willing to sacrifice
an infinite number of dogs (or non-rational humans) to save a single normal human, my
view is closer to Pluhar’s. A dog’s life is not so expendable. While Pluhar holds that
each individual always deserves equal consideration, the welfare-utilitarian’s answer is: it
depends on the individuals. Which humans are we talking about? Which dogs? Often,
welfare comparisons will be highly difficult or impossible. Like Pluhar, I suggest that we
give preference for individuals not near death. Other factors, such as the ability to 79
successfully bring the lifeboat to land, considerations about the individuals’ moral
characters (toss the villain overboard | ), also make a difference. Because such
comparisons will be difficult if not impossible to make, a group in such a terrible
situation may very well prefer to draw lots, and cannot be blamed for doing so.
The foregoing points regarded our moral evaluation of individual lives. It is
important to recognize that the gradualist scheme I am proposing does not produce
gradualist answers in every situation. For example, pain is a disutility that can affect two
very different individuals in a very similar manner. A chicken may be sufficiently simple
that its life is less valuable than mine, yet pain for the chicken detracts from its welfare
just as much as a similar pain detracts from mine. The moral value of one’s life does not
act as a ‘multiplier’ such that harming me matters more than harming a chicken to the
same extent. In fact, some harms may matter more when they are inflicted on the
chicken | I can get by, more or less, with a broken leg. A broken leg for a chicken can
drastically reduce its welfare, perhaps fatally. Moreover, a human’s higher awareness of
the source of pain can often make pain more tolerable than it would be for an animal,
who might experience extreme fear out of ignorance. Only when we consider the death
of an individual, or other harm that detracts from most or all aspects of one’s well-being,
do we grant the more complicated individual additional consideration.
Conclusion
Carruthers, Regan, and Pluhar all describe utilitarianism as an attractive theory,
but find certain implications surrounding the ASO unacceptable. In response, each
develops a theory of rights: Carruthers argues for a contractualist position, and Regan and
Pluhar each argue for a more generic rights theory. Contractualism, while based on a
firm theoretical foundation, generates no consideration for non-rational agents, despite
Carruthers’s effort to include all humans. Regan’s theory has initial appeal, but lacks a
solid foundation and, despite his reliance on intuitions, results in highly counterintuitive
applications. Pluhar’s theory appears most successful, having overcome Regan’s
foundational obstacles while serving much better than contractualism for explaining our
attitudes towards animals and non-rational humans.80
Each of these philosophers devotes significant attention to utilitarianism as the
‘runner-up’ theory, and they view Singer’s preference utilitarianism in particular as
among the most successful formulations. By proposing the alternative view of welfare
utilitarianism, I have attempted to overcome the criticisms that lead to the rejection of
utilitarian theories. I have by no means described a complete theory, and much more
needs to be said in regards to the nature of welfare, and how one can make decisions that
affect it. What I hope to have done is indicate the way toward a moral theory that 1)
provides an acceptable answer to the ASO, 2) rests on a solid theoretical foundation, and
3) overcomes the obstacles both of rights-based theories and of other utilitarian
formulations.81
BIBLIOGRAPHY
Carruthers, Peter. The Animals Issue. Cambridge, UK: Cambridge University Press,
1994.
DeGrazia, David. Taking Animals Seriously. Mental Life and Moral Status. Cambridge,
UK: Cambridge University Press, 1996.
Edward Regis, Jr., ed. Gewirth’s Ethical Rationalism. Chicago and London: University
of Chicago Press, 1984.
Frey, R.G. Rights, Killing, and Suffering. Oxford: Basil Blackwell, 1983.
_____, ed. Utility and Rights. Minneapolis: University of Minnesota Press, 1984.
_____ and Christopher W. Morris, eds. Value, Welfare, and Morality. Cambridge:
Cambridge University Press, 1993.
Gewirth, Alan. Human Rights. Chicago and London: University of Chicago Press, 1982.
Goodin, Robert E. “Utility and the Good.” In A Companion to Ethics, edited by Peter
Singer, 241-248. Oxford and Malden, Mass: Blackwell Publishers Ltd., 1993.
Miller, Harlan B. “Introduction: ‘Platonists’ and Aristotelians’.” In Ethics and Animals,
edited by Harlan B. Miller and William H. Williams, 1-14. Clifton, N.J.: Humana
Press, 1983.
_____. “A Terminological Proposal.” SSEA Newsletter 30 (March 2002).
Mill, John Stuart. Utilitarianism. Edited by Oskar Piest. New York: Macmillan
Publishing Company, 1957.
Nozick, Robert. Anarchy, State, and Utopia. New York: Basic Books, Inc., 1974.
Parfit, Derek. Reasons and Persons. Oxford: Clarendon Press, 1984.
Pluhar, Evelyn. Beyond Prejudice. The Moral Significance of Human and Nonhuman
Animals. Durham and London: Duke University Press, 1995.
Regan, Tom. The Case For Animal Rights. Berkeley and Los Angeles: University of
California Press, 1983.
Singer, Peter. Animal Liberation. 2
nd
ed. New York: Avon Books, 1990.
_____. “Killing Humans and Killing Animals.” Inquiry 22 (1979): 145-156.
_____. Practical Ethics. 2
nd
ed. Cambridge, UK: Cambridge University Press, 1993.
_____. “A Utilitarian Population Principle.” In Ethics and Population, edited by
Michael D. Bayles. Cambridge, Mass.: Schenkman Publishing Company, 1976.82
VITA
Born in New Jersey, Jesse Ehnert received his B.A. from Connecticut College in 1995,
with a major in English and a minor in philosophy. After graduation, he worked at a
variety of computer- and website-related jobs in North Carolina and Virginia, pursuing
his philosophical interests in his free time. In 1997, Jesse happened upon a copy of Peter
Singer's Animal Liberation, and made the startling discovery that his prior moral beliefs
about nonhuman animals had been absolutely, undeniably incorrect. This revelation,
combined with a lack of interest in computer- and website-related jobs, led him to the
philosophy program at the Virginia Polytechnic Institute and State University, where he
received his M.A. in philosophy in 2002. Jesse is currently seeking a career where he can
make use of his philosophical education. Until then, he will likely continue to work at a
variety of computer- and website-related jobs.
المراجع
www.abegs.org/sites/Research/DocLib/%D8%A3%D8%AC%D9%86%D8%A8%D9%8A%D8%A9/The%20Argument%20from%20Species%20Overlap.pdfموسوعة الأبحاث العلمية
التصانيف
الأبحاث
login |