The Southern Journal of Philosophy

Volume 52, Spindel Supplement

2014

ANIMALISM AND THE LIVES OF HUMAN ANIMALS

Paul Snowdon

ABSTRACT: It is suggested that the best way to interpret animalism is as an identity thesis
saying that each of us is identical to an animal. Since there are disagreements about
the nature of animal persistence, this means that animalism itself not does not explic-
itly propose criteria of identity for persons. It implies the negative claim that features
that have nothing to do with animal persistence have nothing to do with our persis-
tence. Thinking of it as an identity thesis also makes sense of the nature of the
arguments surrounding the thesis. Central to such arguments are claims about the
persistence of animals and persons in certain imagined scenarios. To adjudicate such
arguments, we need a secure grip on some claims about animal persistence. Often
these are generated by a theory of animal persistence. In the second part of the paper,
it is argued that the attempt to build such theories on the assumption that life is
essential for animal existence is implausible. In the way we speak, we seem not to
recognise death as the ceasing to exist of an animal. No better way to think of animals
is proposed in this paper.

I have adopted the title of the Spindel Conference itself as the title of my own
paper, since that title matches rather well my own themes. I want to present
some thoughts about animalism, about human life and existence, and about
the connections between them. These reflections are, sadly, rather less con-
clusive that I would have wished.1

Paul Snowdon has been the Grote Professor of Mind and Logic at University College London
since 2001. Before that he was, for 30 years, a Fellow and Lecturer in Philosophy at Exeter
College, Oxford. He has written about perception and the philosophy of mind more generally,
personal identity, and the history of twentieth century philosophy. He is shortly bringing out a
book on animalism with OUP, entitled Persons, Animals, Ourselves. He is also bringing out a
collection of his essays, again with OUP, entitled Essays on Perceptual Experiences. He and Stephan
Blatti are also editing a collection entitled Essays on Animalism to be published in 2014 by OUP.

1
The background to this paper is a developing sense that the assessment of various
arguments brought against animalism turn on judgments about the persistence of individual
animals in imagined scenarios. How are we to know what the truth is here? What better, then,
than a conference that brings together people who think about animalism and people who think

The Southern Journal of Philosophy, Volume 52, Spindel Supplement (2014), 171–184.
ISSN 0038-4283, online ISSN 2041-6962. DOI: 10.1111/sjp.12074

171
172 PAUL SNOWDON

1. ANIMALISM AND SOME ARGUMENTS

We have, many of us at least, a fascination with knowing what kind of things

we are, that is what we fundamentally are. Humans have revealed a tendency
to mark themselves out as a basic kind of thing in nature, and we are also
prone to adopt theories of our nature which strongly emphasize the differ-
ences between us and other things in the world. However, according to the
conception of the world that we currently hold, one truth is that where I am,
there is also an animal. That animal, we think, can be placed in an evolving
chain of increasingly complex organisms, going right back to the start of
animal life. This same general point, of course, also applies to you. Once we
think about the question—what are we?—in the knowledge that where each
of us is there is an animal, an object of that sort, there has to be a strong
temptation to identify ourselves with that animal. One thesis, then, that the
label ‘animalism’ stands for is that each of us is identical with the animal that
is where each of us is, and I want in this paper to work with that elucidation
of it.
Now, this makes animalism into an identity thesis, and I want to suggest
that to think of animalism as such an identity thesis illuminates the character
of the debate about it in a number of ways.
One point is this, animalism is not, per se, a theory about the persistence
conditions of creatures of our sort, something that is usually called a theory of
personal identity. Whatever the rights and wrongs of the name “theory of
personal identity,” it is surely obvious what kind of theoretical enterprise l
have in mind here. All animalism tells us is that our persistence conditions are
the same as those of the animal we are, since those things are the self-same
thing. We get from this claim a theory of our persistence conditions only if we
can articulate a theory of animal persistence and apply it to ourselves (or a
theory for animals of the kind we are). But clearly there are disagreements
about animal persistence, just as there about personal persistence. If this is
correct, we can also see that it is a mistake in debates about animalism to
assume that the views on animal persistence that are often accepted by
animalists are themselves definitive of animalism. If animalism is the identity
claim, it is consistent with any theory of animal persistence. One thing this
means is that the basic character of animalism is different from the way
of thinking that the term ‘Lockeanism’, ancient or modern, picks out. The
latter purports to specify conditions of persistence that apply to us, whereas

about animals in general, and human beings in particular? This was Stephan Blatti’s excellent
goal for the Spindel Conference. My aim in this paper is to explain this perspective on how the
issues are joined and to advance some claims about how we should think about animals. I am
still in the process of thinking through what emerged in the conference itself.
 ANIMALISM AND THE LIVES OF HUMAN ANIMALS 173
animalism does not. What does, perhaps, follow from the basic identity claim
is that our persistence conditions do not include or rest on elements that
obviously do not apply to animals. It seems to me that it is fairly obvious that
animal persistence has nothing to do with mental relations over time, or with
the retention of mental capacities. Think simply of the kinds of experiments
on animals that can result in an animal losing these capacities without the
animal being destroyed. So the adoption of animalism implies that such
features have nothing to do with our persistence conditions, and, of course,
that implication might be true even if animalism is not. What positively should
be said remains to be determined, even given acceptance of animalism.2
Next, the fundamental commitment of an identity claim that A = B is that
A and B must share all their properties. Hence, the identity can be disproved
if a property P can be found which A possesses but B lacks. The strategy of
anti-animalists is to persuade us that they have found some such property.
Any detected property difference is enough to disprove the identity, but when
comparing ourselves to the cohabiting animals what is striking, surely, is how
similar we are. Thus, the physical properties I think of myself as possessing are
also ones I think of the animal here possessing. We have the same shape,
weight, color, hair length, and so on. I also think of myself as alive, breathing,
digesting food, and suffering ailments, and I think of the animal as the same.
The resemblances, it must be said, are very striking.
But the resemblances seem not to end with the physical and biological
features already mentioned. If asked about the human animal here whether
it (or he) has sensations, perceives, acts, thinks, and communicates by speech,
the obvious answer is that it does. Supposing that is correct, a style of
argument in favor of animalism suggests itself. If we deny that I and the
animal here where I am are identical, we seem committed to the existence of
two things both of which have the same mental properties. It would further
seem that if one is a person in virtue of that mental life then so is the other.
So a denial of animalism seems committed to the existence of two psycho-
logical lives, one of the animal, the other of me (in the case under consider-
ation), and, perhaps, the existence of two persons. Now, neither implication
seems to be acceptable. Can there really be two psychologically endowed
entities here? This is, of course, simply a sketch of the style of argument now

2
At the conference it was suggested by some, including John Dupré, that animalism is
committed to the idea that animals are continuing objects, but that really they should be
thought of as processes. However, if it is correct to regard animals as belonging in the general
category of processes, then the identity claim is not threatened since we can regard ourselves as
processes. The identity thesis leaves even that open. Further, it cannot be said, I believe, that
thinking of animals as processes was established by those who believe it.
174 PAUL SNOWDON

known as the two lives problem. It alleges that here is something close to a
paradox in the denial of animalism.
The current popularity of animalism flows from the realization by various
philosophers more or less simultaneously in the 1980s that there are animals
where each of us is and that denying there is an identity at least seems
paradoxical. Now, opponents of animalism have devoted considerable efforts
to show there is nothing paradoxical here, and that debate remains, as far as
I can see, still alive. However, what I want to stress here is that it is a serious
mistake to regard the so called two-lives argument as the only pro-animalist
argument and to think that rebutting it would leave animalism unsupported.
In fact, the long-standing debates about the psycho-physical identity theory in
the philosophy of mind provide one possible argumentative pattern in favor
of identities that could in principle be employed by animalists. Thus, one can
argue that A = B, given premises of the form: A = the F and B = the F. So, if
one could find such an identity that applies to me and also applies to the
human animal, here one would have another pro-animalist argument.3
Again, identities can be supported in inferences to the best explanation. Thus
we might conclude in a case where there is some uncertainty that a certain
man is identical to the man married to someone else, because it is the best
explanation as to why he has certain information (about the marriage cer-
emony, say). Similarly it is worth considering whether the truth of the identity
claim that is animalism might not provide the best explanation for certain
otherwise unexplained facts about ourselves. Further, there might be plau-
sible general principles that apply to animal evolution that support the con-
clusion that the new capacities that emerge in evolution must belong to
emerging animals.4 This list, I suggest, reveals the extent of other possible
arguments for animalism.
However, what most philosophers seem to be struck by are not the simi-
larities between us and the animals, nor the plausibility of these arguments,
but rather some supposed differences. The major candidates for such differences
concern the contrasting possibilities for survival that we and the animals have.
One supposed sort of difference is that the animal can survive despite a total
loss of mental functioning, whereas we cannot survive that. A potential

3
It should be pointed out that arguments with this pattern can be generated to support
views which are inconsistent with animalism. Thus, at the Spindel Conference, Carl Gillett
proposed an argument along the following lines to show that each of us is identical to his or her
brain. (1) I am the thing doing this thinking here and now. (2) My brain is the thing doing this
thinking here and now. (3) So, I am my brain. This argument plainly has the same structure.
Reflection on this candidate argument, I think, brings out why it is difficult to give persuasive
examples of it.
4
This is, in outline, the thought that Blatti (2012) is attempting to articulate.
 ANIMALISM AND THE LIVES OF HUMAN ANIMALS 175
example would be when an accident leaves a human alive but in a state of
what is called persistent vegetation. Many think this amounts to the survival
of the animal but not one of us (or, as it is said, the person). A contrasting sort
of possible case is one where one of us (the person) can survive due to a
significant part of us being transplanted and continuing to function, but it
being too small a part to count as yielding survival for the animal. Tradition-
ally with this sort of case the focus has been on brain-transplants. Now, it is
true that these two argumentative patterns are not the only ones that anti-
animalists can or do employ, but they do seem to me to be the most exploited
ones (in some form or other) in standard debates.
A crucial question is, of course, whether there are good counterexamples
to animalism amongst these sorts of cases. Fundamentally, what they involve
is the judgment that in certain sorts of supposedly possible cases either the
animal is still there but the person is not, or the person is still there but the
animal is not. Clearly, anyone making such judgments supposes themselves to
be in a position to make judgments about animal persistence and what we
might call person persistence. Now, in the supposed cases where the animal
survives but the person (or the thing of our kind, whatever that is) does not,
there really is no dispute about whether the animal does survive, since
nothing in the cases even seems to remove the animal. It is there, say, asleep
in the hospital bed or on the experimentalists table. In these examples what
is at issue is the other judgment to the effect that the person is not there.
However, by contrast in the cases where there are significant subtractions
from the animal, it is not always obvious what has happened to the animal or
what has happened to the person. The imagined cases are not easy for us to
judge because they are not actual, and it is hard to be sure what to think about
them. An example to illustrate this kind of difficulty would how to trace the
animal if the head of the animal were removed and kept alive in some
artificial structure. Would that head count as a severely truncated animal or
not? It is clear that arguing against animalism on the basis of such an example
relies on being confident that it would not.
One reaction at this point is to search for a theory of animal persistence, or
what might be called animal identity. Such a theory cannot, though, be
regarded as necessary in order for people to propose or to reject this style of
anti-animalist argument. For in order to support the adoption of a general
theory, the theory must rest on accepted judgments about particular cases. It
must in some way generalize from more particular judgments. The presence
of a general theory is, therefore, not necessary to make judgments that we can
rely on for argumentative purposes. But philosophers in this debate have been
prone to suggest general theories of animal persistence. And in the second
part of this paper I want to reflect on the role of the notion of life in such a
176 PAUL SNOWDON

theory. Frequently such theories center themselves on the assumption that

animal persistence is tied to life. They build up their theories from that central
assumption. So an important question is whether that assumption is true. It
needs to be added that even if it is not a prelude to constructing a general
theory, the issue as to what the relation is between animals and life is an
interesting one.

2. LIVES

Before engaging with the issue, I want to distinguish between two employ-
ments of the term ‘life’. The first use, when we are talking about animals, and,
if this is different, ourselves, is that “life” is understood as the period during
which the animal (or we) is (or are) alive. The life of the animal is what
happens to that animal when alive, and in this use the life of the animal is
finished, is over, once the animal dies. Then the animal no longer has a life.
This we might call “life as living.” But there is a second use of ‘life’. On this
use we can talk about the life of a planet or of a building or a natural process
like a hurricane. We then mean, roughly, those things that happen to it while
it exists. The life of a planet ceases when that planet ceases to exist. This use
we might label “life as existing.”5
Now, with that distinction before us, we can note a potential ambiguity in
the term ‘lives of human animals’. Does it mean their lives while alive, or does
it mean their lives while existing? According to one conception of human and
animal existence these things can exist only if alive. So although there is a
difference between the two notions of lives, there is no possibility of them
coming apart with animals. This is, however, I suggest, not something that is
completely obvious and certainly does not go without saying. What I want to
do is to explore what the relation actually is between an animal being alive
and its existing. Given animalism, what this relation is for animals will also be
its relation for us.

3. THE LIVES OF HUMAN ANIMALS

Until recently it was assumed by most philosophers that life is a necessary

condition for animal existence. Such a view seems to have been endorsed by
both Aristotle and Locke, and recent prominent supporters of it are
Dummett, Wiggins, and Olson.6 A number of other philosophers have,

5
When people sometimes advise another to “get a life,” they are using ‘life’ in a third way.
6
The necessity of life thesis was certainly the view that was widely assumed in Oxford in the
1970s and 80s. Probably Michael Ayers’ was the only dissident voice there during that period.
 ANIMALISM AND THE LIVES OF HUMAN ANIMALS 177
though, argued for the opposite claim.7 I am going to defend two propositions
about animals which place me in agreement with this second group of
theorists.
(A) Life processes are not necessary for animal persistence. The reason for
saying this is that it seems we can put an animal into a state of what is called
frozen animation. In this state, life processes do not occur. The animal is there
but frozen solid. Further, it may be possible to unfreeze the animal so that its
life processes resume. The animal surely remains there if that is a possibility.
Although freezing the animal may in fact be the only way to preserve it
without life processes occurring in it so life can resume, it is not unthinkable
that physical processes other than lowering temperature might have achieved
the same result. There seems, therefore, to be nothing problematic in the idea
of an animal existing without life processes occurring in it.8 One comment
that might be made about this example is that if in fact there is no viable
resuscitation process, then freezing the animal does in fact kill it. That may be
a reasonable comment, but it does not touch, rather, it concedes, the point
that an animal can remain in existence without life processes occurring in it.
(B) An animal does not cease to exist at death. This proposition is more
controversial than (A). I propose therefore to put forward first some argu-
ments in favor of thesis (B) and then to consider some arguments against it.
There is an established nomenclature in relation to this debate. The claim
that death is the end of existence for an animal is known as the Termination
Thesis. Employing that name, it can be said that (B) is the denial of the
Termination Thesis. Obviously, if (B) is correct, an alternative positive
account of animal persistence is needed.
What can be said in favor of denying the Termination Thesis? (i) Let us
consider first what our ways of talking and thinking about dead things is.
What sorts of things do we say? We would say, in a certain situation, “This
animal is alive, but that animal is dead.” This seems to mean we regard some
entity as both an animal (“that animal”) and as dead, just as we regard
another entity as an animal which is alive. We would say, “I have a dead
butterfly in a box.” This seems to mean that we think something is both dead
and a butterfly. If it is a butterfly, though, it is surely an animal. We would say
of a particular horse, “Red Rum died last night and is now available to be
seen at the stable.” This also seems to imply that there is an animal called Red
Rum which is still there to be seen but also dead. We can say, “Red Rum is
being cremated today,” where we treat an object as being both Red Rum and
7
Amongst such philosophers are Michael Ayers, David Mackie, and Fred Feldman.
8
It strikes me that the cryogenic project seems odd to people because the technology of
resuscitation is hardly one about which we are confident. It is not immediately felt that freezing
is necessarily equivalent to ceasing to exist.
178 PAUL SNOWDON

obviously dead, since it is about to be burnt. In the way we speak, then, we

clearly seem to regard being dead as something a previously alive entity can
be. We might say, “I saw Red Rum but only after he died,” and “I helped to
move Red Rum after he died.” These examples, many more of which could
be given, seem to involve a commitment to the falsity of the Termination
thesis.
(ii) Although the thesis being discussed is a thesis about animals and
death, it is relevant to consider an obvious phenomenon about human
grieving. I have in mind the way humans seem to regard the dead bodies
of those they have loved as being the loved one themselves. Thus, parents
who lose children to violent deaths want desperately to keep the child from
being buried or cremated so that they can still see their child. The emotion here
seems to rest on the conviction that the dead entity is their child whom they
are still seeing and cannot come to terms with no longer seeing.9 Now it
might be said that it begs the question to cite this as evidence about our
attitude toward animals and death. In response, we can say two things.
First, it is very hard to think we can allow that persons, or ourselves, remain
in existence when dead (and not, as it were, existing in another domain),
whereas we do not think the same about animals. I think it is undeniable
that people do have such feelings in this kind of situation. Although strong
emotions can skew our language and our thought, it is not clear that it does
so here. Second, the same attachment to seeing the animal although it is
dead can be observed in people whose pets have died. It is an attitude that
applies to our engagement with animals.
(iii) A difficulty for the Termination Thesis is that it does not seem to fit
our thinking in analogous cases. I have in mind the way we think about
plants. We draw the living/dead distinction with plants. Thus we would
count a particular rose as dead and another one as alive, but we seem to
show no inclination to hold that the rose no longer exists once dead. We
even have, and talk of, pressed roses. We would talk of where the rose in
question had grown, when it had been picked, how it is still pretty, etc.
Now, surely, it is remarkably odd if that is our attitude toward plants but
not toward animals.
(iv) It does not seem that the predicate ‘x is dead’ behaves like a predicate
that implies that the thing to which it applies does not exist. In the first place,
‘x is alive’ and ‘x is dead’ do not seem to be what we call attributive adjectives.

9
This attitude came across very strongly in the TV reporting at the time of the extremely
moving burials of the children killed in the school siege at Beslan. It is, also, surely obvious that
many people have the same attitude to ordinary animals. Thus, pet lovers are reluctant to bury
them because they desire to still see them (even though they are dead).
 ANIMALISM AND THE LIVES OF HUMAN ANIMALS 179
Rather, we can scrutinize an object and either affirm that it is alive or that it
is dead (or, of course, neither). Whether it is alive or dead is not relative to a
type. Further, we seem quite prepared to co-ascribe both being dead and lots
of other properties that seem to imply existence. We can say, “That animal is
dead and I shall eat it/throw it in the river/photograph it.” If I am doing
these things to it presumably it exists. We can also say, “Please make sure the
specimen is dead before bringing it to us.” Here, again, we seem to regard an
existent movable object as dead. There is, it appears, no general entailment
from ‘x is dead’ to ‘x does not exist’.
(v) There are logical complexities with the use of ‘dead’. We might say
that Elizabeth I has been dead for 400 years. Now this seems to mean it is
true to say that Elizabeth I is dead (now). We do not, though, want to say
that Elizabeth I exists now, since surely nothing remains of her. So, ‘x is
dead’ does not imply that ‘x does not exist’, but neither does it imply that
‘x does exist’. In this it is like the predicate ‘x is famous’, which present-
tensedly applies to some things which do exist (for example, some famous
sportsmen) and also applies to some things which do not exist (say, some
long-dead politician).
To these arguments some will wish to add the following objection to the
Termination Thesis. If an animal ceases to exist when it dies, then there is
surely something that was there before death which does not cease to exist,
namely the body of the animal. Since it was there before occupying the same
space as the animal, the Termination Thesis’ doctrine is committed to the
possibility of co-incidence. There is, however, no such possibility. That is not
an argument upon which I am relying. One reply, devised by Olson, is that
there is no such entity as the body of the animal which exists both before
death and afterward. So there is no commitment to the possibility of
co-incidence. This response would need overturning if the objection pre-
sented is to stick. However, my general attitude is that it is a mistake to appeal
in arguments about animalism to very general metaphysical claims, such as
the impossibility of co-incidence. The reason is that there is massive contro-
versy about such metaphysical claims, and it cannot be said that the argu-
ments offered in support of them are conclusive. As a general policy it seems
better to me to eschew reliance on such claims.
I have so far tried to make a case for thinking that an entity can exist and
be dead. It seems, also, that the entity which is dead is the thing that was alive.
Further, we also classify it as being a certain sort of entity, say a type of
animal.
Can, though, something be said for the other point of view? There is
certainly a case to be made, and I shall outline it as best I can, but shall also
comment on the strength of the reasons as I outline them.
180 PAUL SNOWDON

(vi) It needs registering immediately that the proposition that an animal

ceases to exist at death seems to many people a fairly obviously true one. They
would describe it as an intuitively plausible one. It seemed to me, for a long
time, to have that status. We should not ignore this fact, but in my own case
this appearance of truth was weakened by noticing (or so I think) that there
are many other things I also find equally obvious and which seem to imply the
opposite view. (The above arguments articulate some of these claims.) For
me, this weakens the idea that the Termination Thesis is intuitively more
attractive than its denial.
(vii) If an animal remains in existence when it dies, then it seems to follow
that death is a rather considerable change in an object. This, however, seems
a comic and hence unacceptable consequence. Consider what our reaction
would be to someone who said, “I am afraid my uncle has rather changed
since you last saw him; he’s dead.” Such a remark is so incongruous as to be
humorous. This remark certainly is incongruous, but it is not clear that that
is because it is contradictory. It seems equally plausible to assign its incon-
gruity to the first part of the remark being in the circumstances a rather
extreme understatement.10 Anyone who based the statement that someone
has changed on the knowledge that he or she was dead could be accused of
failing to say something which would so relevant that it should have been
said.
(viii) A line of argument that has seemed plausible to many rests on two
premises. They are: (1) our concepts of animals, such as a dog or a cat,
represent a subject matter for laws and explanatory generalizations. (2) If
we allow that after death we still have the same animal, then there will be
no (or very few) laws centered on such entities, because the onset of death
fundamentally alters the nature of the entity. In many respects all dead
animals are pretty much alike and not much different from things that have
always been inanimate. The conclusion is that it is in some way built
into the point of such animal concepts that they can only apply to living
entities.
Now, premise (2) seems correct. The question is whether premise (1) is
correct in a strong enough sense to generate the intended conclusion. The
alternative view is that our animal categories do, and are supposed to,
provide subjects for generalizations and laws, along many dimensions, ana-
tomical, developmental, behavioral, physical, etc., but such generalizations
are understood as being restricted to living cases. Is there anything in this
idea of an implicit restriction that is somehow alien to our employment of

10
There are jokes the other way too. Think of the absurdity of the following dialogue. “We
are burying Uncle Fred tomorrow.” “Oh, how odd, I thought he was dead.”
 ANIMALISM AND THE LIVES OF HUMAN ANIMALS 181
those categories? It can also be added that it is not quite true to say that all
generalizations fail once death intervenes. Thus, the anatomy of dead
animals of a certain kind will resemble that of living cases especially if care
is taken to preserve them. In fact, investigation of such cases is a source of
information about animal kinds, and in the description of such cases there
is no scruple about calling the investigated specimens animals of the rel-
evant kind.
(ix) The rejection of the Termination Thesis faces or raises a problem. If
an animal does not cease to exist at death, then when does it cease to exist?
In the case of many animals, assuming they are there after death, their
history is one of gradual decay and disintegration. At what stage in this
process does the animal cease to exist? This is a significant question, but
there are two points to make in response. The first is that the Termination
Thesis itself faces a similar question, though on a different time scale. When
exactly does the life of an animal end? That notion, of life ceasing, is not
completely precise. It is not therefore as if we have a completely precise
proposal versus another totally imprecise one. Second, if we are thinking in
terms of the existence of objects after the animal has died, even if it is not
counted as the animal, there is a question to be faced and somehow solved
as to what its conditions for existence are. Thus if we accept the Termina-
tion Thesis, then after the animal’s death there will be an entity thought of
as a corpse or a body. Since there is such an entity we can ask even sup-
porters of the Termination Thesis view when it ceases to exist. Presumably,
there has to be some suitable answer. If so, it can be taken over as the
answer that someone denying the Termination Thesis can offer about the
animal itself.
These assessments conclude the initial and major part of my examination
of the Termination Thesis, arguing that on balance the considerations favor
rejecting it. There is, however, a way of keeping the debate going which needs
outlining and assessing. There seems to me to be no easy way to deny the
linguistic data which is cited in arguments (i) to (v). It is hard not to conclude
that it is true to say given the way we use the word ‘animal’, animals can exist
even though they are dead. The comment might be made, though, that the
metaphysical thesis represented by the Termination Thesis is not a straight-
forwardly linguistic claim, suitable for refutation in virtue of claims about the
use of ‘animal’. It is, rather, a claim that there is significant category of entity
to which the Termination Thesis applies. It is likely to be added that the word
‘animal’ has a use in which it stands for that category, but it may also be, and
perhaps this is what the linguistic data reveals, that ‘animal’ can also stand for
a related category of entity to which the Termination Thesis does not apply.
Now, conventional metaphysicians might have assumed that such a complex
182 PAUL SNOWDON

linguistic state of affairs would not exist. They might, however, be prepared to
countenance this as a possibility in order to defend the metaphysical insight
they seem to themselves to have.
In support of this proposal we can give examples where language
behaves in the way proposed. For example, a visitor to Madam Tussaud’s
might well ask where David Beckham is. It is surely not the case that this
use implies or indicates that the wax object there actually is, as we might say,
the real David Beckham. In fact we clearly use language flexibly, and we
rely on the audience to latch on to the reference (or significance) of what we
are trying to talk about by using a term which does not strictly apply, but
which has a basic meaning that enables the audience to link it to the
intended object or feature. In the Beckham example, the audience naturally
regards the word ‘Beckham’ as picking out the model of Beckham. The
idea, then, is that this is what is happening when we talk of a dead animal
as ‘an animal’.
The contrast between the terms we use and the metaphysical categories
we thereby express is, given the possibility of what we might call this sort of
creative use of language, a genuine one. The metaphysical question is not
settled in a simple way by the language we use. However, it seems to me
that the Termination Thesis remains unattractive in two main respects.
First, the problems for it do not rest solely on our tendency to talk of “an
animal” being there post-mortem, but also on other aspects of our talk. For
example, we give evidence, in our predicative practice, of thinking that the
living entity is the same thing as the dead thing. I might say, for example,
“this (dead) butterfly was caught four days ago.” Second, there seems to be
no strong evidence in favor of recognizing a type of entity that ceases to
exist at its death. A third point is that one should postulate such an ambi-
guity (or usage) as is being advocated with the term ‘animal’ only if there is
strong evidence for it, which, as far as I can see, does not exist. Finally,
although metaphysics cannot be read straight off of language, metaphysics
tends to start with language. Thus, people argue for the possibility of coin-
cidence because we say such things as “the statue has been destroyed but
the lump remains.” Reflection on such a remark grounds the postulation of
coincidence. But the type of proposal being made above could be applied
here. It might be suggested that ‘has been destroyed’ has an extended
significance here and so coincidence is not an implication of what is
said. Once the looseness of relation between language and metaphysics
is granted, the task of the defending metaphysical claims becomes
harder. This indicates that we should be rather cautious about invoking
linguistic flexibility. On balance, then, the Termination Thesis seems
incorrect.
 ANIMALISM AND THE LIVES OF HUMAN ANIMALS 183
Where does this conclusion get us? First, although the Thesis seems incor-
rect to me, I do not think it matters a great deal for the assessment of
animalism whether it is or not. If animals can survive after death, then
animalism implies that so can we, but that implication hardly runs counter to
anything we might feel we know about ourselves. I have pursued the issue,
rather, because it strikes me as interesting in itself.
We might say that the arguments presented in this section support the
idea of existence (but not life) after death!11 If true, that cannot, as I have
just suggested, generate a problem for animalism. It does, however, gener-
ate a problem for some well-known attempts to analyze the persistence
conditions of animals in terms of life. For example, Wiggins (2001) takes as
his basic notion in the account of animal identity that of principles of activ-
ity that animals instantiate. He does not make it entirely clear what is
meant by activity, but it is natural to understand it as features that involve
life. If animals can exist without life, it cannot be right to suppose that
principles to do with aspects of life are the sole things needed to elucidate
animal persistence. The same problem arises for Olson’s (2002) account of
animal persistence. Its central conceptual tool is the idea of what is essential
to sustain life. Again, that is on the wrong tracks if there is existence after
death.
Third, although it may be that animals can exist without being alive, that
does not mean in saying what an animal is we should not mention life. On
the contrary, it is obvious that an animal is caused to exist by living things,
and, as we might say, the processes of animal creation have gone wrong if
the product is not alive in its earliest stages. We might even say that
any animal produced by these processes must be alive during its earliest
stages. In saying what animals are, the idea of life has to be introduced,
but that does not commit us to regarding life as necessary for continued
existence.

4. CONCLUSION

Reflections on the significance of animalism lead us to recognize the impor-

tance in adjudicating debates about it of tracing animals in imagined sce-
narios. We do not need a theory of animal persistence to trust at least some
of our judgments of this sort. However, there has been a tendency to propose
theories of animal persistence that assume life is a necessary condition for
animal persistence. I have suggested both that animalists are not committed

11
I would like to think Arthur Schipper for this amusing way of speaking and for other
points about this topic.
184 PAUL SNOWDON

to this, per se, and, moreover, that it is probably a mistake. Sadly I have, here
and now, no better theory (or proto-theory) to put in the place of the ones
I am rejecting.12

REFERENCES

Blatti, S. 2012. A new argument for animalism. Analysis 72: 685–90.

Olson, E. 2002. The human animal. Oxford: Oxford University Press.
Wiggins, D. 2001. Sameness and Substance Renewed. Cambridge: Cambridge University Press.

12
I would like to thank the participants in the Spindel Conference for very helpful com-
ments on my talk there, which I have not been able to properly engage with, and also for the
stimulating general discussions of animals and animalism. In particular, I wish to thank Stephan
Blatti, both for the invitation to attend, for his excellent organization and for the many points
he made, and also John Dupré, Carl Gillett, Steve Luper, and Eric Olson.