This paper appeared in: Philosophy, vol 69 no 268, 1994

Human Beings and Giant Squids

David Cockburn

1. A television nature programme a year or two ago contained a striking sequence

in which a giant squid was under threat from some other creature (no doubt a
human'being with a video camera). The squid responded in a way which struck
me immediately and powerfully as one of fear. Part of what was striking in this
sequence was the way in which it was possible to see in the behaviour of a creature
physically so very different from human beings an emotion which was so
unambiguously and specifically one of fear.
I would guess, but have not been able to confirm, that most who saw this film
would see in the giant squid's response something similar to what I saw. I cannot,
however, offer a description of the squid's behaviour which might help to
convince someone who has not seen the film and who is sceptical: a description
such that they are likely to agree that if it really did behave like that then it was
certainly correct to ascribe fear to it. That I cannot do so is not, I think, merely a
reflection of my very limited descriptive powers. It is a reflection of the fact that
there is no more fundamental description of the squid's response than 'fleeing in
fear': no other form of description which underlies, and so might provide support
for, this one. My aim in this paper is to establish this point and to bring out its
significance for two very different ways in which philosophers have spoken of
our ascriptions of sensations and emotions to non-human creatures.
In the first chapter of Animal Liberation Peter Singer offers what he takes to be
an essential preliminary to the argument of the rest of the book. If we are to show
that we have a moral obligation not to cause unnecessary suffering to animals we
must, according to Singer, first establish that animals are capable of suffering. He
attempts to do this in the following way:
'Do animals other than human beings feel pain? How do we know? Well,
how do we know if anyone, human or non-human, feels pain? We know that
we ourselves can feel pain. We know this from the direct experience of pain
that we have when, for instance, somebody presses a lighted cigarette against
the back of our hand. But how do we know that anyone else feels pain? We
 cannot directly experience anyone else's pain, whether that "anyone" is our
best friend or a stray dog. Pain is a state of consciousness, a "mental event", and
as such it can never be observed. Behaviour like writhing, screaming, or drawing
one's hand away from the lighted cigarette is not pain itself; nor are the recordings
a neurologist might make of activity within the brain observations of pain itself.
Pain is something that we feel, and we can only infer that others are feeling it
from various external indications. . . . This is an inference, but a perfectly
reasonable one, based on observations of their behaviour in situations in which
we would feel pain, and on the fact that we have every reason to assume that
our friends are beings like us, with nervous systems like ours that can be
assumed to function as ours do, and to produce similar feelings in similar
circumstances.
'If it is justifiable to assume that other human beings feel pain as we do, is there
any reason why a similar inference should be unjustifiable in the case of other
animals?
'Nearly all the external signs which lead us to infer pain in other humans can be
seen in other species, especially the species most closely related to us—other
species of mammals, and birds. Behavioural signs—writhing, facial contortions,
moaning, yelping or other forms of calling, attempts to avoid the source of pain,
appearance of fear at the prospect of its repetition, and so on, are present. In
addition, we know that these animals have nervous systems very like ours, which
respond physiologically as ours do when the animal is in circumstances in which
we would feel pain "
The traditional attempt to justify my conviction that other human beings have a
mental life by appeal to my experience of my own sensations, emotions and thoughts—a
proposed justification most clearly articulated by Mill 2—has been widely criticized; the
criticisms focusing largely on the way in which the justification offered gives my
experience of myself the fundamental position in my understanding of what it is for a
being to be susceptible to sensations, emotions and so on. By contrast, Singer's aim is to
give grounds for our conviction that members of other species feel pain by appeal to
their bodily and behavioural similarities to human beings. That human beings feel pain
is, for the purposes of this argument, to be taken as an indisputable given. Perhaps, then,
this use of the argument can escape many of the familiar objections.
Singer accepts that most of his readers will not need this argument; for most of his
readers will already accept what the argument aims to prove: namely, that animals
do suffer. Similarly, Mill presumably recognized that most of his readers did not need
the argument from analogy in order to convince them that other human beings have a
mental life like their own. It is felt, however, that it is important that we should be able
to provide reasons for what we believe on instinct.
Unfortunately, as so often in philosophy, when one sees reasons being offered for
what one believed without reason one's natural conviction begins to look rather shakier
than it did before 3. Mill suggests that I should conclude from the fact that these bodies
which I observe are like mine, and behave much as mine does, that these are beings
which feel pain, anger, fear and so on much as I do. Now, of course, no other body is
exactly like mine. Even within the realm of human bodies many are different from
mine in ways, such as those which distinguish different races, which some are inclined
to see as being of enormous significance. When we are concerned with comparisons
between human and non-human creatures there is, on the face of it, an unambiguous
sense in which the differences become more and more marked. Few have any hesitation
in attributing pain to dogs. Yet when one stops to reflect on the similarities between
canine and human bodies and behaviour it is not difficult to be more struck by the
differences than by the similarities. Dogs have four legs and a tail, are covered in hair,
have ears and noses up to six inches long, and a 'voice' which, to say the least, would
sound decidedly odd in a human being. When we move to giant squids or flies matters
do, of course, get very much worse.
My conviction that this is a being that is capable of suffering is to be grounded in
the similarity between its body and behaviour and my body and behaviour; or
between its body and behaviour and the bodies and behaviour of human beings in
general. To the extent that this is so we have, surely, no choice but to concede that the
weaker the similarities the weaker the conviction that is in place. While this has
obvious implications for my thought about members of other races my immediate
concern is with its implications for our thought about non-human species. If my
conviction that other species feel pain or fear is to be grounded in their simi larities to
human beings then we must, it seems, concede that my conviction that dogs feel pain
stands on much shakier ground than does my conviction that other human beings do;
and my conviction that that giant squid was afraid may begin to look decidedly fragile.
There is, of course, plenty of room here for debate about the extent of the bodily
and behavioural similarities and differences between species. But in the absence of some
reasoned account of which of these similarities and differences are relevant to our ana -
logical reasoning we will, it seems, have to concede that the man who is very hesitant
about ascribing pain to dogs, not to mention giant squids, has a rather more respectable
view than we might previously have thought. It may, at any rate, be very difficult to
avoid the conclusion that our attempts to relieve the suffering of others should be
concentrated on those about whom we can be most certain that they do suffer: those
who look and behave most like us4. In appealing to the argument from analogy it seems
that Singer may be giving a fundamental, justifying position to the notion of a human
being in a way which is deeply at odds with the vision of our responsibilities towards
animals which he aims to defend.
It is no use at this point insisting that it is just obvious that certain similarities are
relevant and others are not: that, for example, skin colour, number of legs, or the
particular arrangement of eyes, nose and mouth which is characteristic of human beings,
have no bearing on the force of the analogical reasoning; and that only somebody with
inherently damaged sensibilities or blinded by prejudice could fail to see this. For,
while all of that may be true, the whole point of the argument from analogy was to
move us beyond such instinctive judgements to something that might ground them.
The appeal to the idea that it is obvious that the dog's having four legs is not a
relevant difference between it and us is simply an appeal to our conviction that dogs
are, for example, capable of suffering. It will cut no ice with the very people who are
supposed to need the argument from analogy here; for if somebody seriously suggested
that having four legs might well be a relevant difference what could we do but point to,
for example, a dog in pain and ask: 'Does the fact that he has four legs really make our
ascription of pain more dubious?'
The converse of this point applies to the appeal to the great similarities between what
goes on in the nervous systems of a human being in pain and of a dog that has been hit
by a car. Certainly this appeal might move somebody who had doubt about pain in
dogs. But he might equally conclude from the similarity, and 'the obvi ous fact that
dogs don't feel anything', that what happens in the nervous system is a far less
decisive mark of pain than he had previously supposed. No doubt he would have to be
crazy or corrupt to draw this conclusion. But then he would have had to be crazy or
corrupt in the first place to doubt that dogs feel pain. The appeal to this 'decisive
similarity' does not provide us with an 'argument' where before we had 'only common
sense'.

2. I have focused so far on differences between the bodies of human beings and
those of other species. It may be said that what is relevant to the analogical reasoning
is not so much the bodily form of the other creature as its behaviour, and that here the
similarities between human beings and members of other species are inescapable.
In assessing this suggestion we can usefully begin with a case within the human
realm: a case where this move might seem to have unambiguous force. Consider the
white slave owner who insists that his black slaves do not suffer as 'we' do when one
of their children dies. Can we not point out to him that, while there are bodily
differences between his slaves and those people—members of his own race—who, he
agrees, feel deep grief on the death of a child, the grieving black mother behaves just as
the grieving white mother does? And will that not make it very difficult for him to hold
on to his claim that what the two women feel is very differ ent? Well, perhaps he can be
made to concede that the two women behave in the same way in the sense that both are
off their food, walk more slowly than usual, utter similar words, and so on. But he
may add that if you really want to know what they are feeling you need to focus on
more fine grained features than these. You must look into their faces. And when he
looks into the black mother's face he sees nothing which he is inclined to describe as
'deep grief. Any more, perhaps, than we can see 'deep grief in the face of, or hear
'deep grief in the bleating of, the sheep whose lamb has just died5.
Our racist can, perhaps, be compelled to concede 6 that the movements of the black
face and the white face described in geometrical terms are very similar: 'The pupil
contracted by % mm, the left hand corner of the mouth rose by %mm . . .'. Suppose
then that we kept a record of these movements over a period of a month after the death
of the child. Would our racist not have to grant that these unambiguous behavioural
similarities provide an overwhelming case for the claim that the two women feel
much the same?
It is not at all clear that he would; for it is quite unclear why he needs to accept that
it is similarities at this level of description which are relevant to a well grounded
ascription of mental states to others. At any rate, in practice we tend to attach much
greater significance to the similarity between these faces:

than to that between these:

 The insistence that our racist is wrong not to be impressed by the similarity in the
facial contortions described in geometrical terms does, surely, stand in some need of
defence; and, as the above drawings should make clear, it is also likely to have some
startling implications for our normal ascriptions of emotion on the basis of what we see
in a person's face. But what concerns me more here is this. If it is held that it is
similarities at this level which are crucial to the analogical reasoning then the fragility
of the move from human beings to other species will become uncomfortably clear.
Singer speaks of 'pain behaviour' which is common to human beings and many other
species: 'writhing, facial contortions, moaning, yelping or other forms of calling,
attempts to avoid the source of pain, appearance of fear at the prospect of its repetition,
and so on' (op. cit., 11). The last two similarities listed here are unlikely to do much
work in our argument with the sceptic: one who needs convincing that dogs are capable
of feeling pain will no doubt also need convincing that they 'attempt' to do certain
things, or that they show signs of fear. Thus, if he could see fear in the dog's behaviour
he is unlikely to need an argument to convince him that there is also pain to be seen
there7. The same point may well apply to our description of the dog as 'writhing',
'moaning' and 'calling'. We need to move to a level of description which is closer to the
'geometrical' if we are to have a serious chance of securing our sceptic's agreement that
the dog, or the squid, does behave in the ways described. However, the closer that we
get to-this level the less clear it will become that these creatures do 'behave in ways
like us'; rather, the more clear it will become that they do not.
Certainly we can hear pain in the dog's yelp. But described in 'geometrical' terms—in
terms of pitch, tone and so on—the sound that the dog produces is quite different from
that produced by a human being in pain. I am not sure how we would react if a
human being made that noise, but I very much doubt that we would take it to be
straightforwardly clear that this was a sign of pain. The case of the dog's 'facial
contortions' raises a more fundamental difficulty. At the level of a geometrical
description the dog's face is radically different from a human face. With many breeds
of dog the differences start at the most basic structural level: in place of the relatively
flat surface in which eyes, nose and mouth are embedded we have a protruding cone.
This difference has immediate implications for the possible placing and structure of
the mouth: described in geometrical terms a spaniel's mouth is, and inevitably so,
radically different from a human mouth 8. And that difference inevitably makes it
obscure what it could mean to say that 'the dog's mouth moved just as that human
being's mouth moved'. Similar points have equally clear application at the level of
total bodily form. Given the radical difference in structure, in geometrical
terms, between the body of a human being and that of a giant squid it is
quite obscure what it could mean to say that, in these terms, 'the squid's
body moved very much as does the body of a human being who is afraid'.
We may find these radical differences surprising when they are brought to
our attention, because what we see on a dog is so unambiguously a face and
what we see in the giant squid is so unambiguously fear. The lesson,
however, is that radically different geometrical arrangements of matter may
all be faces, and radically different movements of matter may all be instances
of a creature fleeing fearfully. My point, then, has not been that dogs do not
sometimes behave as human beings do when they are in pain, that squids do
not sometimes behave as human beings do when they are afraid, and so on.
On the contrary, what struck me about the giant squid might be expressed in
the words: 'It behaved so like a frightened human being'. No doubt the truth
that most of us will see in those words contributes to the illusion that we
might use this similarity, as Singer hopes to, to provide an argument for our
conviction that giant squids suffer fear. But this is an illusion; for the
similarity appealed to will only be seen by one who does not need that
argument.

3. Wittgenstein writes:

'(O)nly of a living human being and what resembles (behaves

like) a living human being can one say: it has sensations; it sees; is
blind; hears; is deaf; is conscious or unconscious."

This remark has been taken to be part of an attempt to fill the gap left by the
collapse of the argument from analogy. On this reading, Wittgenstein is, as
Singer is, suggesting that we can give someone who doubts that dogs are
creatures of a kind capable of feeling pain, or giant squids of a kind capable
of feeling fear, reason for thinking that they are by pointing to the
similarities between the behaviour of these creatures and the behaviour of
human beings. He differs from Mill and Singer, however, in holding that the
connection between pain or fear and certain patterns of behaviour is not
simply an empirical correlation. Writhing, crying out, facial contortion and
so on are 'criteria' for pain. That is to say, part of grasping the meaning of
talk of 'pain' is grasping that if a creature behaves in these ways then, other
things being equal, pain or fear is to be ascribed to it. Thus, someone who does not
agree that dogs feel pain, or giant squids feel fear, is guilty of a failure to grasp the
meanings of the words 'pain' or 'fear'.
A view of this kind would face the same objection as I have levelled against the
argument from analogy. It is now generally agreed, however, that Wittgenstein's aim
is quite different. He is not attempting to justify my willingness to ascribe sensations
and emotions to other human beings, or our willingness to ascribe these to non-
human creatures. These features of our lives neither stand in need of, nor can be given,
a justification. The philosopher's task is simply to describe our practice, and, in
particular, to resist intellectualist distortions of it. My thinking of other human beings in
these terms, and our thinking of non-human creatures in these terms, is grounded, not
in reason, but in primitive responses of, for example, pity towards the other.
This, however, leaves a puzzle for our interpretation of the above passage.
Wittgenstein insists that the philosopher must describe carefully how we do speak; not
legislate how we ought to speak. He warns us against saying how things 'must' or
'cannot' be10. What, then, are we to make of the claim that only of what resembles living
human beings can one say these things? The problem is rendered particularly acute by
the fact that some apparently do ascribe sensations to things, such as trees and flowers,
which one might have thought manifestly do not resemble living human beings".
We might try to make sense of Wittgenstein's remark by connecting it with another:
'My attitude towards him is an attitude towards a soul. I am not of the opinion
that he has a soul.' 12 In so far as someone's words 'It is in pain' are not a mere
mouthing they will be connected in some way with her attitude towards that
of which she is speaking. The connection with, for example, pity or, less
attractively, sadistic pleasure is at least part of what enables us to say that she
is using the word 'pain' as we do; and so is part of what enables us to say
that she is having the thought 'That animal is in pain'. Of course, we need to
be careful about exactly how we state that connection for there is such a
thing as recognizing that another is in pain but simply not caring.
Nevertheless, if a person speaks, for example, of trees as being in 'pain' yet
never responds to them with the pity, horror and so on which are
characteristic of our responses to others in pain there will, to put it mildly, be
room for doubt about how seriously we should take her use of this word here.
Now, to the extent that something lacks the human form it will not be
clear what it would be to respond to it in the ways in which we
characteristically respond to a human being in pain. We turn away in horror
from the sight of someone's contorted face, close our ears to his cries of
pain, watch his face carefully for a sign of an easing of the pain, look into his
face, into his eyes, in sympathy, put an arm around his shoulder, and so on.
If that of which we are speaking, for example a tree, does not have a face,
does not have eyes, does not cry out, does not have a shoulder, then
whatever I do I will not be reacting to it as I react to another human being in
pain. The less like a living human being another being is, the less place there
will be for responses to it of the kind which are characteristically linked with
the thought that another is in pain; and so the more tenuous will be the claim
that I mean the words 'It is in terrible pain' in just the sense in which I mean
them when I use them of a human being. It is in this sense that we can only
say of what resembles a human being that it is in pain: if someone uses those
words of something that does not resemble a human being, little in her
behaviour will even look like the pity or horror which are part of the place
which these words have in our life.
While there is, I think, something to that, it may not take us as far as we
might hope. One point at which we might feel uncomfortable relates to the
apparent implication that the sense of our words 'It is in pain' necessarily
gets thinner as we move from human beings, to apes, to dogs, to giant
squids, and so on. Where the argument from analogy implies that we ought
to feel less confident that something is in pain the less closely it resembles a human
being, this line of thought implies that our ascription is inevitably less rich as the
resemblance weakens.
I am not certain that, if we understand this point correctly, we ought to feel
uncomfortable about it. It will, for example, carry no implications about the strength of
our obligations to relieve animal suffering. Be that as it may, the point I want to stress is
slightly different. It is that nothing in the argument can show us which features of our
familiar responses to the pains of others are central to the sense of our words 'He is in
pain', and which might be written off as more or less 'inessential accompaniments'.
Thus, if we place at the centre of the picture our tendency to watch, and look into, the
other's face we will feel that the ascription of pain to the giant squid or to the wriggling
fly must be distinctly thin; but if it is said that it is the attempt to relieve the other's
pain that is most central to the sense of the words 'He is in pain' there will be much
greater room for the idea that we may react to such creatures much as we do to the
writhing human being.
Now nothing, I think, in our normal use of the word 'pain' entails that we ought to
give the more central place to one rather than another of these responses. If our aim is
simply to describe our normal practice as accurately as possible we will, I suppose,
have to say that there is some variation in the importance which people attach to
particular responses. Some, such as those who in practice feel no difficulty about the
ascription of pain to trees", clearly give less weight than do others to the range of
responses which involve quite fine details of the human form. While I do not rule out
the possibility of showing that they are wrong (or right) to do so, we will, surely, have
moved beyond any 'purely descriptive' philosophy in doing this. The claim that one can
only ascribe pain or fear to a human being or to what resembles a human being— that
these states 'cannot' be ascribed to, for example, trees—will have a clear ethical
dimension.

4. Despite my criticisms I take it that there is something right in the argument which
connects the possibility of responding to another in the ways distinctive of our
relations with other human beings with the bodily form of the other. To the extent that
that is so we need partially to reverse the relation of ideas which we find in Singer. For
Singer, seeing the similarity to the human body and behaviour is prior to the ascription
of fear or pain to a creature, and that in turn is prior to responses of pity, refraining
from injury and so on. On the picture that I have derived from Wittgenstein, the
possibility of responses such as pity is directly dependent on the bodily form of the
other; and the ascription of pain or fear follows from, or goes with, that. But if what I
have argued in earlier sections is correct the reversal which we have to make to Singer's
view is more radical than that.
We can approach this through a consideration of another possible area of difficulty in
Wittgenstein's remark: '(O)nly of a living human being and what resembles (behaves
like) a living human being can one say: it has sensations; it sees; is blind; hears; is deaf;
is conscious or unconscious.' At least on one natural reading, this remark shares
something important with the argument from analogy: it shares the assumption that our
ascriptions of sensations and so on to non-human species are, in one way or another,
dependent on there being similarities between members of those species and human
beings. The argument from analogy suggests that our right to ascribe fear to a giant
squid or pain to a fly depends on such similarities; and Wittgenstein that the possibility
of ascribing these conditions is so dependent.
Mill's use of the argument from analogy gives a fundamental position to my
experience of myself in a way which seriously distorts my relationship to others.
Singer's use of the same argument gives a fundamental, justifying position to the notion
of a human being in a way which is deeply at odds with the vision of our relationship to
non-human species which he wishes to defend. It might equally be argued that
Wittgenstein's remark presupposes that human beings are the paradigm of beings to
which pain, fear and so on can be ascribed; and that that is an assumption which we have
no reason to accept.
It is certainly not my intention to suggest that the central position of the notion of a
human being in Wittgenstein's thinking is wholly confused; on the contrary, I am
convinced that what is of most importance in Wittgenstein's later philosophy is
inseparable from this. Nevertheless, the suggestion that Wittgenstein's thought is human-
centred in a way of which we should be suspicious may contain an element of truth. For
the fact that we, who are wondering whether, say, jellyfish may feel pain, are human
beings does not imply that that question is to be answered by reflecting on the simi larities
and differences between jellyfish and human beings. There is no reason to suppose, as
Wittgenstein sometimes appears to, that the primary application of the notions of 'pain'
and 'fear' are those in connection with human beings: that other applications must spread
out from this centre and are dependent on perceived similarities between other creatures
and human beings. Of course, if someone confidently asserted of slugs and flies that 'It is
in pain', and, with that, apparently pitied them and so on, but generally hesitated when
confronted with human beings doubled up in agony, we might have doubts about the
beliefs which should be ascribed to her. But in the lives of many people animals occupy a
place which is, in certain respects, as central as that occupied by other human beings. In
particular, certain specific animals have a quite fundamental place in the lives of many
young children; and a child's use of the words 'pain', 'fear' and so on may be acquired as
immediately in connection with the pet cat as in connection with human beings.
My central point, however, is this. I suggested that any attempt to use the argument
from analogy to convince someone that a giant squid may feel fear faces a dilemma. If
we are working with descriptions of the squid's behaviour in what I called 'geometrical
terms'—in terms of 'matter in motion'—it seems unlikely that we will find similarities
between the squid's behaviour and ours of a kind which could possibly licence an
inference to the conclusion that it almost certainly, or even quite probably, feels fear;
the radical differences in physical structure, described in 'geometrical' terms, are
sufficient to rule out this possibility. On the other hand, someone who is working with
descriptions of a form such that it can be said that there are important similarities
between the behaviour of the giant squid and that of human beings will be someone
who already accepts the conclusion of which the argument is supposed to convince
him.
A similar dilemma faces Wittgenstein's remark if we understand it as stating a
condition which something must satisfy if we are to be able to ascribe to it fear or pain.
If we are speaking of resemblances at the level of descriptions in geometrical terms then
it is false, or else too open-ended to do any serious work, to say that we can only ascribe
fear or pain to what resembles a living human being. As Wittgenstein himself
remarks, we find a clear foothold for ascriptions of pain to a fly 14. Yet the similarities at
this level are likely to be, to say the least, decidedly tenuous. In these terms, perhaps the
closest likeness we could find in a human being to the fly's behaviour would be
someone lying on their back, with their arms and legs waving about in the air, and
buzzing. And that is hardly what we would regard as the paradigm behaviour of a per-
son in pain. (I do not say that this would be the closest imaginable similarity. As I have
noted before, the radical differences in physical form leave little sense to dogmatic
claims of this kind.) On the other hand, if we are working with descriptions of a kind
such that it can be said that there are important similarities between the behaviour
of the fly, or of the giant squid, and that of human beings then we cannot say that it is
the similarities which enable us to ascribe pain to the fly or fear to the squid. For exam-
ple, it is not the fact that flies writhe or squids flee that makes these ascriptions
possible. That we can see these similarities between the behaviour of flies and
squids and that of human beings is a reflection of, not a condition of, our ability to
ascribe the pain or fear. We might then, with some justice, reverse Wittgenstein's
remark, writing instead: 'Only of what has sensations; sees; is blind; hears; is deaf; is
conscious or unconscious can one say that it is a living human being or resembles
(behaves like) a living human being.'15
The argument from analogy is an attempt to state a condition such that if something
satisfies it we ought to ascribe to it pain, fear and so on. Wittgenstein, on one natural
reading, attempts to state a condition which something must satisfy if we are to be
able to ascribe to it pain, fear and so on. But there are no such conditions. We cannot
place limits in advance on the forms of behaviour in which we might be able to see
pain, grief or fear 16; and that is closely connected with the fact that we cannot place
limits in advance on what might move us in the ways in which we are moved by
another's pain, grief or fear 17 . In particular, 'behaves much as does a human being
who is feeling pain, grief or fear' is not such a condition.

5. My discussion has been fairly abstract: I have been concerned with a general
philosophical model in terms of which we are tempted to understand the relation
between what we can see in an animal's behaviour and the sensations, emotions and so
on that we are able to ascribe to it. This abstractness has led to a crudeness in the
dichotomies with which I have worked. There are, I take it, forms of description which
cannot be neatly placed in either of the two categories which I have distinguished; and
ways of thinking about particular species which embody neither a denial that they feel
anything nor an unambiguous conviction that they experience pain, fear and so on.
Consider, for example, the following description of the response of three young
chimpanzees to the death of their mother:

The next morning dawned cold and grey with rain pouring steadily from a
leaden sky. Passion was dead. She had fallen during the night and hung,
caught by one arm, from a tangle of vines. Her three offspring, who had been
her constant companions during the last weeks of her life, were around her
now. Pom and Prof, for the most part, just sat staring at their mother's
body. But Pax repeatedly approached and tried to suckle from her cold, wet
breasts. Then, becoming increasingly upset, screaming louder and louder, he
began to pull and tug at her dangling hand. So frenzied was he in his distress
that eventually he succeeded in pulling her loose. As Passion sprawled
lifeless on the sodden ground, her three offspring inspected her body many
times. Occasionally they moved off a short distance to feed, listlessly, then
hurried back to their dead mother again. As the day wore on Pax gradually
became calmer and no longer tried to suckle, but he seemed ever more
depressed, crying softly and, occasionally, pulling at Passion's dead hand.
Eventually, just before darkness fell, the three moved off together. 18

Goodall manages to convince us, or, at least, to convince me, that we can, indeed that
we must, ascribe to chimpanzees an emotional life which is very much richer than
perhaps most of us would have suspected. She does not achieve this by pointing to
similarities between chimp and human behaviour which we can be brought to
acknowledge quite independently of our readiness to ascribe a rich emotional life to
them; but nor does she simply throw us in at the deep end with an invitation to describe
the young chimp as emotionally devastated by its mother's death. Goodall eases us
along through an interplay between descriptions of their behaviour which very few
would find controversial, and descriptions which embody richer and richer ascriptions
of emotion. In being eased along in this way we need to place a good deal of trust in
Goodall: not simply in her competence as a 'scientific observer', but in her ability to offer
descriptions of what she has observed which are not coloured by sentimentality. For an
important aspect of the process by which she brings us to accept her descriptions of the
chimps' emotions lies in the fact that she offers descriptions which move us. Our readiness
to ascribe grief or depression to the chimp is a reflection of the fact that we are moved in
a certain way, perhaps by a form of pity, by our observation, or by her descriptions, of
them19. And our acknowledgement of close similarities with the behaviour of human
beings flows from our readiness to employ such a rich language of emotions in our
descriptions of their behaviour.
Peter Singer presents us with a relation of ideas which might be summarized in this
way: the observed similarity between the behaviour of certain creatures and that of
human beings creates the possibility of our ascribing fear, pain and so on to such crea -
tures; and that, in turn, creates the possibility of our responding to them with concern. It
might be closer to the truth to say: our concern for creatures of a certain kind creates
the possibility of our ascribing fear, pain and so on to them; and that, in turn, creates
the possibility of our observing a similarity between their behav iour and that of
human beings. And my astonishment that it should be possible to see fear so
unambiguously in the behaviour of a giant squid—a creature whose bodily form is so
radically different from that of a human being—may, then, have been the reflection of
a philosophical prejudice; or, perhaps, a pair of prejudices. It may have been a reflection
of the prejudice that all similarities are in some way parasitic on similarities at the
geometrical level; and of the prejudice that a fleeing human being provides us with our
paradigm of a creature in a state of fear. 20

University of Wales: Lampeter

1
Peter Singer, Animal Liberation (New York: Avon, 1978), 10—11.
22
See J. S. Mill, An Examination of Sir William Hamilton's Philosophy,
6th ed. (London, 1889), 243^1-.
3
In the hands of philosophers, though hopefully few others, the shaki-ness may
issue in doubts, or denials, that animals feel anything at all. See, for example, Peter
Harrison, 'Do Animals Feel Pain?', Philosophy 66, No. 255 (January 1991), 25-40.
4
This implication is drawn out in Peter Harrison's paper: ibid, 38—9. 138
5 For a highly illuminating discussion which is relevant to this point
see Raimond Gaita, Good and Evil (London: Macmillan, 1991), 158-65.
6
Assuming that it is true. Part of my point in all of this could be put
by saying that most of us probably have very little idea to just what
extent it is true. I should note here that when I speak of a 'description in
geometrical terms' I am concerned with any form of description which it
is plausible to suppose that someone could be compelled to accept no
matter how sceptical he was about feelings in other races or other species.
I make no judgement on the question of whether geometrical description
is really the most plausible candidate for that role.
7
Closely related points may have an application to the following pas
sage: 'An adult male, reclining in the shade after a good meal, reaching
benignly to play with an infant or idly groom an adult female, is clearly in
a good mood. When he sits with bristling hair, glaring at his subordinates
and threatening them, with irritated gestures, if they come too close, he is
clearly feeling cross and grumpy. We make these judgments because the
similarity of so much of a chimpanzee's behaviour to our own permits us
to empathize.' Jane Goodall, Through a Window (London: Penguin,
1990), 14. It is the last sentence which is suggestive of a mistake.
8
Just where would one place a human shaped mouth on a spaniel's
head if instructed to place it at the same point as it occupies on a human
head?
9 Ludwig Wittgenstein, Philosophical Investigations, G. E. M. Anscombe and
R. Rhees (eds), trans. G. E. M. Anscombe (Oxford: Blackwell, 1968), *281.
10
See, for example, Zettel *134: 'Do not say "one cannot", but say
instead: "it doesn't exist in this game".' Ludwig Wittgenstein, Zettel, eds
G. E. M. Anscombe and G. H. von Wright, trans. G. E. M. Anscombe
(Oxford: Blackwell, 1967).
11
In the paragraph following that which we are considering
Wittgenstein notes that people do ascribe pain to things which do not
behave like human beings; his example, which is rather different from
mine, is the ascription of pain to dolls. He comments: 'this use of the
concept of pain is a secondary one' (op. cit. note 9, *282). Oswald
Hanfling has strongly urged me to accept that this brings out that
Wittgenstein does not face the difficulty which I am pressing. But is not
the claim that this use of the concept of pain is 'secondary' simply the
claim that this use is dependent on that in which pain is ascribed to
human beings and what resembles them: that the former use derives its
character from the latter? And that is hardly the claim that pain cannot
be ascribed to things other than human beings and what resembles
them.
12
Philosophical Investigations, op. cit. note 9, p. 178. I presented the following
line of argument in more detail in 'The Mind, the Brain and the Face',
Philosophy, 60, No. 234 (October 1985), 477-93.
13
I am assuming that there are such people, and that their tendency to see sense in such
ascriptions is not simply the result of misleading philosophical pictures—as talk of
'time travel' may be primarily the result of misleading philosophical pictures. But the
latter assumption may raise quite delicate issues.
14
Philosophical Investigations, op. cit. note 9, *284.
15 I am suggesting only that it could be put this way round with as
much justice as it can be expressed in Wittgenstein's own formulation.
Connected with this is the fact that I am far from clear how much any
thing that I have said should be taken as a criticism of Wittgenstein. It
will be clear that my 'criticisms' have drawn heavily on the general spirit
of Wittgenstein's work; and Oswald Hanfling has tried hard to convince
me that there is no need to take the remark which has been central to my
discussion in a way which lays it open to my objections.
16 I do not mean here to rule out the possibility that we might be able to
place limits of some kind on the type of creature to which such states might
be ascribed. For example, it might be argued that 'living and mobile' is
such a limit. But to say that that is a condition which something must satis
fy if it is even to be a candidate for ascriptions of, say, fear is to say nothing
about the patterns of behaviour in such a creature which we could see as
expressive of fear, and on which our ascription might be based.
17 At least, we can do this no more than we can place limits in advance
on where we might be able to see redness, or, rather differently, beauty.
Presumably we cannot rule out in principle the possibility that scientists
should discover some physical feature shared by all patterns of behaviour
in which we are capable of seeing fear or pain. But such a condition, if it
existed, would not be a feature of the sense of our talk.
18
Jane Goodall, op. cit. note 7, 166.

19 In her paper 'Eating Meat and Eating People' Cora Diamond writes:
'We cannot point and say, "This thing (whatever concepts it may fall
under) is at any rate capable of suffering, so we ought not to make it suf
fer.".'. The Realistic Spirit (Cambridge, Massachusetts: MIT Press, ,
1991) 325. I owe a large debt to this marvellous paper.

20 I am very grateful to Oswald Hanfling, Elizabeth Wolgast and

Maureen Meehan for extremely helpful comments on earlier versions of
this paper. I am indebted in a more general way to staff and, especially,
students at University of Wales: Lampeter for discussions which have
helped to stimulate the ideas in this paper.