Ethics that favour humans have been under attack for the last few decades. The
critics accuse their opponents of being prejudiced, and brand favouritism toward
humans as ‘human chauvinism’, ‘anthropocentrism’, or ‘speciesism’, viewing it
as on a par with racism and sexism. The most renowned exponent of this view
is Peter Singer. In this essay I shall examine his own carefully considered ethics,
and try to show that there is a lack of coherence in Singer’s views. In spite of an
avowed and widely announced aspiration to equality across sentient species,
Singer’s ethics eventually favours humans. No impartial and plausible justification
is provided for this, thus his favouring exemplifies speciesism.
It is convenient to start with a clarification of the notion of speciesism.
‘Speciesism’, Singer wrote in Animal Liberation, is an ‘attitude of bias toward
the interests of members of one’s own species and against those of members of
other species’.1 This limitation to attitudes and interests is unnecessary, and in
Practical Ethics he applies ‘speciesism’ to views or opinions. He does not
identify speciesism with the view that the border of ethical concern coincides
with the border of our species. Speciesists, he says, ‘give greater weight to the
interests of members of their own species when there is a clash between their
interests and the interests of those of other species’.2 The view that the interests
of one’s species are more important, though, would be speciesism only if held
‘on the basis of species itself’.3 The most plausible and interesting interpretation
of this is, as I have argued elsewhere, that speciesism has to do with a bias in the
meta-ethical reasons that are put forth.4
Singer devotes pages to the refutation of different meta-ethical reasons for
favouring humans, dismissing the refuted writers as speciesists. Yet, he advocates
the view that an acceptable reason can be given, namely that we are persons
in the sense of biographical beings.5 Evidently he does not believe himself to be
speciesist, but why not? As I want to show, Singer does use human-favouring
criteria in his crucial ethical principles and arguments. If I am right in this, one
could say that Singer is at least an indirect speciesist, in the sense of the
expression that Raymond G. Frey has captured:
Am I not, it might be suggested, an indirect speciesist, in that, in order to
determine the quality and value of a life, I use human-centred criteria as if they
were appropriate for assessing the quality and value of all life?6
Validational criteria, as I take it, are articulated in meta-ethics. That criteria
are human-centred should mean that their construction is such as to give
precedence for at least normal humans against the members of all other species.
But even if Singer uses human-centred criteria, I think it is possibile for his
position to avoid indirect speciesism, in the case where such use is justified by
non-partial and otherwise plausible reasons at a higher level of meta-ethics. At
that level the conditions on the validational tools would be fixed; ultimately we
would be dealing with the very choice of evaluative point of view. Singer,
however, does not present a non-partial and plausible defence of his humancentred
meta-ethical criteria, as I intend to show.
To allow Singer the best possible case, I shall operate with a weak and rather
uncontroversial interpretation of ‘plausible reasons’ as reasons that would not be
implausible to contemporary philosophical common sense. As such, this condition
would be fulfilled if the majority of philosophers (regardless of philosophical
affiliation) could agree that the reasons provided were not illogical, for
instance, nor impossible to prove by normal means, nor strongly counterintuitive.7
But Singer’s high level meta-ethical reasons fail to achieve even this, I shall
maintain, and his ethics therefore can be said to exemplify (indirect) speciesism.
THE MAIN STEPS OF SINGER’S ARGUMENT
I shall focus on what I understand to be Singer’s central argument in regard to
the different ethical status of humans and non-humans. It has three main steps,
where [1] is supported by [2], and [2] is supported by [3], although it is possible
to view [3] as providing support for [1] independently of [2]:
[1] It is directly wrong to take a person’s life, but not directly wrong to take a nonperson’s
life8
[2] A person’s life is valuable in a way that a non-person’s life is not9
[3] A person has important interests (preferences) that a non-person does not
have10
When in the course of [1] Singer uses the expression ‘directly wrong’ it is not
clear what he, as a utilitarian, means by this. One interpretation is that it just
means ‘wrong’. However, to thwart a preference for future experiences, for
instance, is not always wrong according to a utilitarian. It depends on the
consequences for all of the beings concerned, compared to the consequences of
the alternative acts. Another interpretation is that ‘directly wrong’ means that we
have a prima facie reason against the act in question. But this interpretation
considerably weakens [1], since any frustration of a preference could be said to
give a prima facie reason against it. Besides, it is also the case that the killing of
a non-person frustrates its preferences for the future, as Singer recently has
conceded (agreeing that it gives rise to a reason not to kill such a being, see note
24). One might counter that more, more intense, or more important preferences
are frustrated when persons are killed than when non-persons are killed. Yes, but
then the difference in [1] seems to change from a difference in kind, between
directly wrong and not directly wrong acts, to a difference in degree of direct
wrongness. This would change the picture and transfer the interest to [2] and [3],
where grounds for a radical difference would be given.
A third possible interpretation of an act being ‘directly wrong’ is that it
produces negative intrinsic value. Then [1] would imply that the death of a
person has a negative intrinsic value that the death of a non-person lacks, which
is one way to understand what [2] is saying. Again, the core of the issue is found
to be in [2] and [3]. Since I think that the last two interpretations are the most
appropriate, I shall focus on [2] and [3].
In classical utilitarianism all voluntary acts, killing included, are judged only
on the basis of the amount of pleasure and pain they (tend to) bring into the world.
If all sentient beings were to be counted, and counted equally, this utilitarianism
would have rather dramatic effects. Singer claims that the mental capacities of
humans sometimes make them suffer more than non-humans, but the contrary
is also true according to him.11 Still, humans would have to accept relatively
modest places in an endless queue for resources of pleasure and avoidance of
pain, in competition with pigs, elks, cats, mice, etc. Classical utilitarianism even
gives us reason to ask whether the human race ought to be radically diminished
or exterminated.
Perhaps conclusions like these made Singer turn to preference utilitarianism.
He hints that the greater weight that preferentialism gives to the distinction
‘between killing a person and killing some other type of being’ speaks for this
rather than classical utilitarianism. When Roger Crisp forwards this interpretation
of his change of view, Singer has no objection12 – even though this could be
seen to give his move a speciesist appearance.
In the first edition of Practical Ethics Singer used both a ‘total’ and a ‘prior
existence’ version of preferentialism. The first, which focuses on the total
balance of present and future satisfactions, is to be applied to beings that are not
self-conscious. These beings could be killed and eaten if replaced by other
individuals having at least the same amount of satisfactions who would not
otherwise have existed. The second, which focuses on the satisfactions of
already existing creatures, should apply to self-conscious beings, and
preferentialism in this version would make them irreplaceable and render it
unacceptable to kill and eat them. (In some passages Singer even seems to
consider applying preferentialism to persons and hedonistic utilitarianism to
non-persons.)
This appears to be an ad hoc measure to assure a preconceived moral
precedence for persons. Note, Singer’s move was made in spite of the fact that
the prior existence version, by referring to individuals localised in space and
time, violates a requirement of semantic universality on ethical principles that is
both well entrenched in the tradition and something that Singer otherwise
stresses.13 Admittedly, non-universal properties can often be reconstructed into
universal ones, but it is hard to see how this would be done with the property of
existing-now. In the preface to the second edition of Practical Ethics Singer
conceded that using two versions was unfounded and that he presently believed
one version of preference utilitarianism permits us to draw a sufficiently sharp
distinction between the two categories of beings.14 But Singer is unclear as to the
content of such a unified preferentialism; he just expresses a hope of it. That there
should be an ethical distinction between the two kinds of beings is still his
presupposition.
The favouritism toward persons soon turns out to be favouritism toward
human persons. Supporting [2], Singer suggests that there is an ‘impartial
standpoint’ for evaluating lives. He imagines himself to be successively transformed
into individuals of different species, to come out of it with exact
memories of the different lives, and to be obliged to choose between optimal
versions of the lives of different species. After the transformation he would then
decide ‘between the value of the life of a horse (to the horse) and the value of the
life of a human (to the human)’.15 A principle for the ranking of lives would even
be found:
In general it does seem that the more highly developed the conscious life of the
being, the greater the degree of self-awareness and rationality and the broader the
range of possible experiences, the more one would prefer that kind of life, if one
were choosing between it and a being at a lower level of awareness.16
Singer does not let us hesitate as to the outcome of this principle: a human life,
not merely a person’s life, ranks highest. He began with a question that spoke of
humans and not persons: ‘Is it speciesist to judge that the life of a normal adult
member of our species is more valuable than the life of a normal adult mouse?’
And after stating his principle Singer continues with a famous quote from Mill,
beginning with the following statement: ‘Few humans would consent to be
changed into any of the lower animals, for a promise of the fullest allowance of
a beast’s pleasures; no intelligent human being would consent to be a fool, no
instructed person would be an ignoramus…’ Singer summarises the lesson from
Mill thus:it is not easy to embrace the preference for the life of a human over that of a nonhuman,
without at the same time endorsing a preference for the life of a normal
human being over that of another human at a similar intellectual level to that of
the non-human in the first comparison.
Singer has a problem with this in that it is difficult to reconcile with classical
utilitarianism, but he suggests that preference utilitarianism has a better prospect
of ‘defending the judgments Mill makes’.17
There are several oddities in the method that Singer proposes.18 It is so
uncertain that it leaves too much room for guessing. For instance, the judge
would have to live through not only a single life of each species but all varieties
of life in order to find the optimal versions. More important, however, is that in
taking the judge to be a human person (‘Imagine that I have the peculiar property
of being able to turn myself into an animal…’) Singer introduces a bias. If a
human person were transformed in this way, s/he would probably prefer an
optimal human life. The result, then, is loaded in the setting of the judgement.
With a judge stripped of initial species characteristics it is not at all certain that
a human life would be the highest ranked. But what would a being stripped of
species characteristics be like? On what basis would it react to different lives?
Singer does not bother to examine the possibility.
One can also ask why the method focuses on the optimal versions of the lives
of different species – was not the issue to evaluate the life of a ‘normal adult
member’ of various species? If the judge would choose a human life on the
premise that its optimal version is superior to all other species’ optimal lives, her/
his life might still turn out worse than many normal lives of many other species,
and the judge could feel that s/he had been fooled by Singer’s method.
Let us suppose that it is the normal lives of the different categories that
Singer’s method applies to, and let us further suppose that we adopt his method
in an evaluation of human lives also (as Singer’s lesson from Mill invites us to
do). This would probably result in a classing of the lives of the unfortunate,
unclever, sick and poor as less valuable than the lives of the fortunate, clever,
healthy and rich. Paralleling the case of persons and non-persons, this conclusion
would have an impact on the direct wrongness of the killing of people belonging
to the different categories. If such a position is repugnant when humans are
considered, as most of us would say, probably also Singer (except for a few
extreme abnormalities), why should it not be equally so when persons and nonpersons
are considered? This question shows, I think, not only how problematic
Singer’s method of valuing lives is, but also that his application of it presupposes
what it was intended to prove.
THE IMPORTANCE OF BIOGRAPHICAL LIFE
According to Singer the difference in value of a person’s life and a non-person’s
life, and the difference in direct wrongness of killing persons and non-persons,
can be supported in another way. In a complex passage, which unfortunately is
poorly developed, he writes:Rational, self-conscious beings are individuals, leading lives of their own and
cannot in any sense be regarded as receptacles for containing a certain quantity
of happiness. They have, in the words of the American philosopher James
Rachels, a life that is biographical, not merely biological. In contrast, beings who
are conscious, but not self-conscious, more nearly approximate the picture of
receptacles for experiences of pleasure and pain, because their preferences will
be of a more immediate sort.19
Several questions are fused here. The first question is whether or not beings that
are not self-conscious are individuals. Well, they are, every being is unique and
is perceived and met as such by others, for instance, their offspring. Across the
species we are all original copies. Still, we all have a unique past, present and
future due to our particular equipment and to the series of circumstances under
which we live. This means that we all have a unique story, a biography one might
say, which could be told.
The second question is whether or not beings that are not self-conscious have
biographical lives. In one sense they have, as we saw above. Singer could be
taken to use ‘biographical life’ in a sense that would be so weak as to apply to
all mammals, and maybe birds also. Rachels is possibly on that line, characterising
biographical life as ‘the sum of all we hold dear: our projects, our activities,
our loves and friendships, and all the rest’.20 These things might well be
described in such a way that both mammals and birds can be said to have
biographical lives. Still, we would hesitate to call them all persons in any
qualified sense. It also seems feasible to apply ‘biographical life’ in such a weak
sense to, for instance, fish – beings which Singer considers to be clearly not selfconscious.
In the weak sense of this expression, therefore, an argument from
biographical life would hardly support [2] and [3].
Singer, however, uses ‘biographical life’ in a much stronger sense, connecting
it to the idea of a person being on ‘a life’s journey’. Such persons are rational,
self-conscious and autonomous subjects of their life. They have, as Singer
writes, ‘desires that project their images of their own existence into the future’,
and their conscious states are ‘internally linked over time’.21 They have embarked
on their life journey, investing hope and desire, time and effort, in order
to reach particular goals or destinations.22 But observe, to see oneself as going
through and planning events linked to each other as in a written or told biography
is to be able (in principle) to tell one’s story, and this would involve such high
cognitive and linguistic capacities that probably both beings that are not selfconscious
and, for instance, the great apes are excluded. In the strong sense of
‘biographical being’, therefore, an argument from biographical life would
support [2] and [3] only if ‘person’ in these propositions is understood in a
human-favouring sense. This is a plausible understanding of Singer, but what
would remain for him to provide is a non-partial and not implausible demonstration
that biographical life/personhood in a human-favouring sense is superior to
other kinds of life. (Rachels, for his part, only argues that biographical life is the
more valuable ‘to us’.23 )
The third question is to consider whether or not beings that are not selfconscious
have nothing but immediate preferences. They do not, if ‘preferences’
is taken in a sense that does not exclude non-humans. They too exhibit behaviour
aiming at future states, both in the short and in the long run, as when they collect
food for the winter, build warm hides or protected nests. All animals can also be
said normally to have an interest in well-being and avoidance of suffering in the
future – they have future-oriented interests. In a recent work Singer actually
agrees to this.24
The fourth question raises the issue of whether or not beings that are not selfconscious
are replaceable by new members of the same species. If ‘replaceable’
means that the loss in a killing can be compensated by the creation of positive
value for the individual and her/his surrounding, then beings that are not selfconscious
are not replaceable, just as persons are not. For instance, a mama
mouse caught in a trap cannot as far as she and her kids are concerned be replaced
by a new mama mouse in the world. But if it means that the loss in a killing can
be compensated by the creation of positive value for the world as a whole, then
humans and non-humans are equally replaceable, on the same utilitarian principle.
This latter point was also made by H.L.A. Hart, who argued that, preferentially
speaking, self-conscious beings are also replaceable.25 Singer’s answer is
ineffective. He says that a package deal that involves creating and then satisfying
a preference does not need to be considered equal to the good of a satisfied
existing preference. Although hesitant, Singer thinks that his view is supported
by universalisability. One would then imagine oneself in various situations and
the creation and satisfaction of a preference would only be desired depending on
its content, which is the reason why in itself it is neither good nor bad.26
It is unclear what universalisability has to do with this point, especially since
it seems to presuppose a surviving non-universal prior existence preferentialism,
concerned with the intra-personal comparability of preferences of existing
persons. Singer denies that creating a preference in a person and then satisfying
it must be thought of as equivalent to the good of the satisfaction of an actual
preference of the person. But Hart’s argument deals with the inter-personal
comparability of satisfaction of preferences regardless of the actuality of the
bearers. Also, all Hart needs is the possibility that satisfaction of non-actual
preferences outweighs satisfaction of actual ones, which could even be due to a
greater number of created satisfactions.27 Singer does not manage to save
persons from the replacability that his preferentialism allots to all others.
The fifth question is whether or not beings that are not self-conscious have
an interest in continued life. Singer says that death, to a being that is not selfconscious,
is just a cancelling of experiences, and therefore ‘death cannot be
contrary to an interest in continued life, any more than birth could be in
accordance with an interest in commencing life’.28 To persons, on the other hand,
death is contrary to their interest, since they are highly future-oriented in their
preferences. Killing a person will very often ‘make nonsense of everything that
the victim has been trying to do in the past days, months, and even years’, writes
Singer.29
Death, however, is a cancelling of experiences irrespective of whether the
beings in question are self-conscious or not, and it puts an end to the futureoriented
interests in both cases. Also, in the case of a being that is not selfconscious
its interest in continued life can hardly be likened to the ‘interest’ of
a non-being in being born. If one supposes the difference to hold on the ground
that persons normally have more and/or in some sense more highly futureoriented
interests than those of beings that are not self-conscious, Singer’s point
would only give a difference in degree between the interests of a person and the
interests of a non-self-conscious being. In [1] and [2], however, there is question
of a difference in kind.
Singer tries a ‘test of universalisability’ to support the view that beings which
are not self-conscious have no interest in continued life:
If I imagine myself in turn as a self-conscious being and a conscious but not selfconscious
being, it is only in the former case that I could have forward-looking
desires that extend beyond periods of sleep or temporary unconsciousness.30
But this test works with the idea of a human person being the proper judge, which
builds bias into it. Also, Singer here relies on a view that he later rejected, viz.
that beings that are not self-conscious have only immediate preferences. If both
kinds of beings have future-oriented preferences, do they not also share an
interest in continued life?
A sixth question is whether or not beings that are not self-conscious are only
receptacles for pleasure and pain, contrary to what would be the case with selfconscious
beings. But should they not, according to Singer, be receptacles for
preference and satisfaction of preference instead? If they are, why are persons not
such receptacles? In the passage where Singer concedes that beings that are not
self-conscious do have future-oriented interests, he explains the difference
between killing persons and killing beings that are not self-conscious simply by
saying that with persons ‘there is a personal loss that is not balanced by the
creation of another being’.31 Singer still seems to use a prior existence version
of preferentialism for persons and a total version for non-persons, even though
this is not defended.
INTERESTS AND PERSONS
The notion of interests, or preferences, is central in preferentialism. As Singer
understands these terms they involve use of human-centred criteria. He adopts
the standard preferentialist view, according to which an interest, or preference,
is a desire for something advanced after a rational consideration of the alternatives
open in the situation. This is what has been called rational preference. In
the opening chapter of Practical Ethics Singer makes the following definition:
‘we count anything people desire as in their interests (unless it is incompatible
with another desire or desires)’.32 The character of the incompatibility is clarified
later on, when he writes: ‘we make the plausible move of taking a person’s
interests to be what, on balance and after reflection on all the relevant facts, a
person prefers’.33
Singer’s apparent ambition is to say that non-humans and non-persons also
have interests, yet the words ‘person’ and ‘people’ occur in both definitions. His
apparent ambition is also to say that many non-humans are persons. But ‘person’
is interpreted with reference to John Locke as ‘a rational and self-conscious
being’, which means that it can ‘consider itself as itself, the same thinking thing,
in different times and places’.34 This is such a strong concept that probably only
humans would be persons (it is made even stronger later on in the book, where
‘person’ comes to be identified with ‘biographical being’). The word ‘people’ is
suitable, since the characteristics of interest, or preference, are such that only
humans (if any) can have them: all the relevant facts would have to be reflected
upon, positive and negative aspects weighed, and a balance of these made as a
ground for the preference.
Limiting interest to rational preference in this way sets up human-centred
criteria, and in Singer’s work the propriety of these limits remains unargued. If
the possibility for non-humans and non-persons to have interests is to be kept
open, then another, wider concept is needed. This could be achieved by
relegating the rationality required to an observer; the observer would judge
something to be in a being’s interest if the latter could be assumed to prefer it,
were the being able to make a reasoned choice. Another solution would be to
construe the concept along the lines suggested by Peter Carruthers, viz. that
‘rational desire’ be interpreted in terms of the modes of desire-formation that are
normal for the cognition of the creature involved.35 This would allow us to say
that animals and non-rational agents in general do have rational desires. The
proposed concepts, however, could be said to be weakly human-centred, since
they refer to reasoned choice and cognition respectively. The latter is avoided if
cognition is understood so as to include any means of orientation that a being can
have in a given surrounding.
Yet another concept of preference that is not human-centred is conceivable,
to be characterised through the orientation of movement in an organism that
manifests attraction and repulsion. This construal would permit us to attribute
preferences also to the so-called ‘lower’ animals, like birds, fish, reptiles and
insects. Even very simple organisms such as bacteria would have preferences,
and largely the same as humans.36 Such a behaviourist concept of preference
could be said to have the drawback that in the case of beings that are not selfconscious
the interest in continued life cannot be based on observations of
behaviour, whereas the interest in copulation can, and therefore the former
would not count. It is reasonable, however, to infer such an interest on the basis
of the whole pattern of observable behaviour.
The result of adopting preferentialism with a concept of (rational) preference
in any of these construals would be that the importance of typically human
preferences decreases drastically. If all that counts is the strength and number of
affected preferences (as Singer often appears to assume), then we cannot be
confident that the preferences of a human person outweigh the preferences of a
non-person. And if they did, one human’s preferences could easily be outweighed
by the preferences of a group of non-persons. A preferentialism of this
kind, however, would generally be considered quite unattractive regardless of
these implications, because preferences that are not reflected do not seem to have
ethical weight. Singer, for one, does not embrace it.
A way out would be to make an axiological shift and to regard satisfaction
of preference (in a wide sense) as a secondary value compared to meaning in life,
as Roger Crisp interprets Singer to maintain.37 Since Singer thinks that meaning
in life requires a concern with long-term goals that involve the interests of (all)
others – that is, requires an ethical point of view – it is something that only
humans could have. But then again human-centred criteria are involved. This
solution would also overdo the job, since it puts people with no meaning in their
life, in this sense, on the level of frogs or dogs. Singer, in a reply, wisely denies
the adequacy of Crisp’s interpretation of him.38
Another solution, patterned on Mill’s way of avoiding a coarse and simplistic
hedonism, would be to value preferences differently, to distinguish between
lower and higher preferences. In this way non-humans should not have to be
swept out from the moral realm, at the same time humans would not lose much
of their original status – provided that typically human preferences would turn
out to be of higher value than non-human preferences. This seems to be his
assumption, as when he describes biographical persons’ highly future-oriented
preferences as ‘a wide range of the most central and significant preferences a
being can have’ (see note 29). It is fairly clear that most people but no non-human
persons, and no non-persons, have the highly future-oriented preferences of
biographical beings – only we can tell our story and fancy or plan its continuation.
39 Singer’s solution would guarantee humans an overall ethical precedence,
but again, his move involves human-centred criteria.
A HIGH LEVEL META-ETHICAL JUSTIFICATION?
Singer’s ethics, then, would be speciesist, unless his use of human-centred
validational tools is non-partially and not implausibly justified on a high metaethical
level that sets the conditions on the validational tools. Singer recognises
the need, as indicated by the mentioned unsuccessful attempts, to find an
impartial standpoint and to universalise. A path that he hints at several times is
one laid out by Richard Hare, distinguishing between an ‘intuitive’ level and a
‘critical’ level of ethics, connected with different validational, or meta-ethical,
approaches. Ethical validity would be established only in critical meta-ethics,
through the application of the universal point of view. 40
Hare’s method has two different interpretations. On one interpretation,
explicitly invoked by Singer, the judge is an ideal being, an ‘archangel’ with
superhuman powers of thought, superhuman knowledge and no human weaknesses,
lacking all partiality to self, friends and relations.41 There are several fatal
problems with this view; and I have developed this theme in some detail
elsewhere.42 The main drawback is that this makes theology of ethics. Even if the
archangel were to embrace every being, we cannot take for granted that s/he
would consider human interests to be the most central and significant interests
a being can have, or that human life has a value superior to that of, say, a butterfly.
Such views would presuppose that archangels favour humans. But assuming this
is a matter of religious belief. We could neither grasp nor have any wellcorroborated
hypothesis about the precepts of godlike creatures. Also, the very
archangel-construct is inconsistent with the meta-ethical setting. If Hare’s
archangel were to lack partiality to self, then her/his ‘I’ would lack the
prescriptivity that Hare considers essential in his method.43 An archangel
lacking self-interests would not have any preferences in any situation of any
being! If, on the other hand, s/he were to have self-interests, what we would get
is not an impartial ethics but an ethics of the party of archangels. Why should we
humans try to live up to that?
On the second interpretation the critical level ethical judge is a somewhat
idealised human being with natural self-interests, but also willing to act on a
principle of equality. It is this version that Hare and Singer employ in practice,
when doing critical-level reasoning.44 Since the judge is to start from what s/he
wants for her/himself, and the procedure builds on her/his ability to understand
and sympathise with others, it makes for an ethical method restricted by human
psychology, even if the latter is idealised.45 This version of Hare’s method, then,
actually adopts a human point of view, inescapably favouring humans. It is
therefore not surprising to find that Singer considers the greater degree of selfawareness
and rationality and the broader range of possible experiences to be the
most preferable for his would-be impartial judge (see note 16).
Generally, Hare’s and Singer’s critical-level thinking does not upset traditional
and intuitive ethical views too much. Hare takes it to accord ethical validity
to many partial norms of common life.46 At the end of Practical Ethics Singer
withdraws his recommendation that we act only on principles that adopt the point
of view of the universe.47 In How Are We to Live? his ambition is to ‘reinstate
the idea of an ethical life as a realistic and viable alternative’,48 and in Rethinking
Life and Death he confesses that a viable ethic perhaps ‘must allow us to show
a moderate degree of partiality for ourselves, our family and our friends’.49
Recently Singer (writing together with Leslie Cannold and Helga Kuhse) has
declared that Hare’s critical-level impartialism holds, but that it is a muted
impartialism, not requiring us to be impartialist in our everyday life.50
In the passages conceding an ethical role for partiality, Singer never mentions
partiality to humans. But why would this not be as legitimate as partiality to ego,
family and friends,51 if highly future-oriented interests are the most central and
significant preferences a being can have? The problem, not solved by Singer (or
Hare), is how at all an ordinary-life partiality can be validated non-partially and
not implausibly. One might counter by saying that Hare’s methodology enjoys
such a wide acceptance among moral philosophers that this should be enough to
make it not implausible. There is a point to this, but it does not suffice, since what
is embraced of Hare’s methodology is the human interpretation of it, which is
partial to humans. Even if there were a wide acceptance of the theological
interpretation of it, this would not make it philosophically plausible enough, I
think. Singer’s ethics can therefore be said to be (indirectly) speciesist.
Singer is not alone in illustrating favouritism toward humans in spite of an
ideology of the opposite. On the contrary, this may actually be the rule. Two even
more radical philosophers can be mentioned: Arne Naess and Paul Taylor, who
plea for biospherical egalitarianism. Both constantly move away from their
ethics when it comes to conflict-of-interest cases, as William C. French has
convincingly argued.52 That Singer also exhibits favouritism toward humans
brings into question the possibility of an ethics, employed by humans, that does
not favour humans. Singer’s lack of coherence between ethical ideology and
actual ethical theory is serious, but his shortcomings in meta-ethics to defend his
favouritism need not be serious. Something like his position can probably be
defended by non-partial and not implausible high-level meta-ethical reasons.
The point of this essay is not to reject Singer or his ethics, but to show how risky
and perhaps unfruitful it is to attribute ‘speciesism’ to philosophers. I think it is
an important task to create a cooler climate for the discussion of humans, nonhumans,
and ethics. Peter Singer and other critics of humanistic ethics have
definitely shown how narrow the traditional understanding of ethics has been,
but they have also shown the need for a better underpinning of an ethics
favouring humans.
1 Singer 1975, p. 7.
2 Singer 1993, p. 58. (All references are to the 2nd edition).
3 Singer 1993, p. 61.
4 This interpretation of ‘speciesism’ is argued more in detail in my essay ‘Specifying
Speciesism’ (Fjellstrom 2002 a).
5 Singer 1993, p. 126 and p. 95.
6 Frey 1988, p. 199.
7 For a more detailed treatment, see Fjellstrom 2002 (a).
8 This thought is introduced in Practical Ethics, p. 91, and stated on p. 95: ‘preference
utilitarianism does provide a direct reason for not killing a person’.
9 Singer devotes in Practical Ethics a section to the claim that there is a special value in
the life of a person, ending by affirming that claim, as it seems (pp. 89–95). The following
section starts with this summing up: ‘Although preference utilitarianism does provide a
direct reason for not killing a person […]’ (Singer 1993, p. 95).
10 As is evident in Singer 1993, pp. 94f, the special value of the life of a person is bound
to the kind of interests that a person has.
11 Singer 1993, pp. 17f.
12 Singer’s statement is made in Singer 1993, p. 94. Crisp, in his essay ‘Teachers in an Age
of Transition: Peter Singer and J. S. Mill’, writes: ‘Like Mill, Singer retreats from simple
hedonism because of its unacceptable implications, in Singer’s case the implication being
that there is no weighty difference between killing a person and killing a non-person’
No comments:
Post a Comment