In his discussion of personal identity, written in the seventeenth century but still, probably, the first piece about the problem that students read, Locke is concerned to draw a distinction between what makes something the same man and what makes something the same person (or self). Locke summarises his theory of persons in the slogan that “… personal identity consists …. In the identity of consciousness”. This is usually taken to mean that according to him a person’s life stretches back only to periods the person can remember, as it is said, “from the inside”. Locke is remarkably cagey about what the identity of a man consists in. This uncertainty derives from Locke’s, commendable for his time, professed ignorance as to whether there are spirits (or souls), and if there are, how such entities are involved in human life. However, Locke is clear that whatever is involved in human identity it is not a matter of consciousness. Locke’s theory then seems committed to what has recently been called the man/person distinction.

According to him, as one might put it, where I, the person, am there is a man (or human), but the person (or self) is not the same thing as the man, since they have different conditions of persistence. A thing’s “condition of persistence” is what is required for that thing to remain in existence. For example, it is a condition for a house to remain in existence that the bricks making it up stay together and are not scattered over the ground. It is generally assumed that people as a sort have their own persistence conditions, and so have animals.

Locke’s defence of the man/person distinction was so successful that philosophers simply took the contrast for granted, losing all interest in the notion of man or animal when discussing their own persistence conditions, and so representing any disagreement, to whatever extent, they had with Locke as disagreements about the concept of personal identity.

In what might be thought of as the recent classic period of discussion of personal identity, culminating in Parfit’s brilliant Reasons and Persons, together with the ensuing debate about it, the notion of an animal (or man) is virtually invisible. Now, this invisibility ended in the 1980s when a number of philosophers were independently struck by how unsatisfactory this neglect is. This insight struck, amongst others, David Wiggins, Michael Ayers, Eric Olson and myself in Britain, and Peter van Inwagen and William Carter in the United States. We responded to this thought in different ways, given the contrasting metaphysical frameworks we worked within, but the idea of us as animals had returned to a debate from which it might be said Locke had removed it.

Having noted how central the idea of man (or the human animal) is to Locke’s discussion, but how invisible it became in subsequent discussions of personal identity, it needs to be asked whether Locke’s conception of the relation between the person and the human animal is correct. The choice is between holding that the human animal (or man) is a distinct thing from the person, and holding that the animal and the person are the same thing. The view that, contrary to the longstanding Lockean framework, they are the same thing, is the view currently called “animalism”. Before presenting some reasons which indicate, to varying degrees of strength, that adopting the distinctness thesis has real drawbacks, I want to clarify animalism and the questions to which it is offered as an answer.

Animalism is an identity thesis; it says that a certain thing is the same as another thing. Now, normally when philosophers discuss the problem of personal identity they are searching for an informative specification of what is involved by way of links across time in the persistence of a person. So a typical (but I do not mean to imply plausible) candidate answer would be: a person P who does X at time t remains in existence so long as there is a person P* who can recall, from the inside, doing X at t. Clearly, the animalist identity thesis does not have that structure. It does not explicitly pick out links over time. However, it does so indirectly, because we have what we might call a shared proto-theory of animal persistence, and animalism implies an answer to the normal problem of personal identity by requiring that persons fall under and conform to that proto-theory.

Of course there are disagreements among animalists as to what the correct theory of animals is. But what seems clear is that even normally minded animals do not cease to exist if their mental capacities are destroyed, so long as they remain physically intact and alive. In such circumstances they remain in existence even though damaged. So the animalist identity thesis implies that any claim that the survival of people requires the presence of mental links over time is incorrect. It also seems clear that no one could think that some processes which purport to preserve mental links over time, without preserving that physically substantial thing which is the animal, could seriously count as preserving the actual animal. The animalist should think, therefore, that simply generating mental links cannot be enough to ground real animal persistence, and hence not enough to preserve the persistence of the person as they think of the person. This explains how a thesis with the structure of animalism can imply substantial things about personal persistence.

A second clarification is needed. So far I have adopted a way of expressing animalism that regards it as saying that the person is the human animal. But if we are prepared to allow there might be entities which merit being described as persons who are not human – say God, or angels, or Martians, or robots, – then animalism should not rule them out. The content of animalism is better formulated as the claim that we, the persons hereabouts discussing these issues, are human animals. That is the nature that we have. Indeed, it is fairly clear, I suggest, that questions about the nature of persons as they interest us really are questions about the nature of ourselves. I have argued that we should understand animalism as a thesis about ourselves, rather than a thesis about persons in general, and also that, despite its form as an identity thesis, it has significant implications about the conditions of what is called personal identity.

The third clarification is that the animalist identity claim (we are identical to animals) should not be thought of as solely, or mainly, a thesis about what is known as “personal identity over time”. One reason for not saying that it is a theory of personal identity full stop is that, as we have seen, animalists can disagree about the persistence conditions of animals. More importantly, we should think of animalism as a general, so to say, metaphysical characterisation of our nature, with myriad implications.

For example, if we are animals, it is also true to say that we are subjects of experience, and so it implies that there need be no more unity to the mind of a subject than there need be to the mind of an animal. On the face of it, an animal’s mind can be radically dis-unified – imagine the sorts of surgical interventions that can break connections consistent with the animal remaining in existence and having experiences. So animalism also leads to a distinctive way of thinking about subjects and psychological unity.

Now I want to present some reasons that incline some of us to accept animalism. There is something artificial in dividing the reasons into those in favour and then subsequently considering those against. Part of the appeal of animalism, assuming it has an appeal, derives from a sense that the grounds that have convinced philosophers that animalism is hopeless are not as strong as their standard assessment supposes. However, in a preliminary presentation things simply have to be artificial. Now, I am assuming, as I hope any reader will allow me to, that where each reader of this article is, there is also a human animal. Animalism claims that the person reading and the animal (presumably also reading) are one and the same thing. Now, if the man/person distinction is accepted it seems to follow that the same space is occupied simultaneously by two distinct things – the animal and the person.

This is sometimes called the thesis of the possibility of coincidence – the idea that two distinct things can coincide in the space they occupy. It is a thesis that strikes some as absurd. How can there be two things precisely coinciding in a single space (possibly, indeed, throughout their entire histories)? Now, if it is absurd then the conventional man/person distinction would have to be wrong, and there would be considerable pressure in favour of animalism. However, although this is a line of argument that has persuaded some, it rests on the conviction that coincidence is not a possibility, which is a conviction that is not obviously correct. So this reason strikes me as unreliable.

What is more striking, I believe, is that on any normal conception of them, there is a massive degree of similarity between the individual human animal and the individual person who occupy the same place. They seem to have the same spatial extension, the same parts, the same history (starting at the same point, and ending at the same point), the same causal roles – they seem to be the same in more or less every specific way. For example, if the animal is breathing then the person is breathing, and vice versa. Now, the most obvious explanation for this virtually exception-less similarity is that the animal and the person are, in fact, the same thing.

The weakness of this argument, though, is obvious. Opponents of animalism hold that there are differences between the human animal and the person. What these are alleged to be I’ll consider shortly, but they will reject the basic premise of the present argument. There is, according to them, no complete similarity, and so no explanation is needed. This reservation does mean that there is no proof here, but it can be suggested that a weaker conclusion looks plausible. Given the manifest extent of the similarity between person and animal, the animalist claim seems to be the view that counts as the default view. Discussion should start from it, and if we are not to accept it some strong reasons need to be provided. If that is correct the neglect of the view for quite a period must count as extremely regrettable.

A third reason or type of reason in favour of animalism is that there are difficulties in the idea of the man/person distinction. There are different ways to develop this idea, and here I shall sketch two closely related versions of the problem. The problem starts from the question whether the animal has mental states. It is obvious that the person or self does, but does the animal?

It surely seems quite clear that the human animal does have mental states, so initially I want to assume that is the correct answer to our question. Indeed, it is this conviction that fuels the appeal of animalism. What mental states does the animal have? It is hard to deny that the animal itself has the same complex array of mental states that the person has. The human animal can think about itself, can talk, can reason, as well as have experiences and perform actions. If that is so then maintaining the man/person distinction seems to carry the implication that there are two subjects of experience and two mental lives being lived. Now, that is not a verdict that we accept as we start thinking about the nature of persons (or ourselves).

But the problem seems worse than this. On normal conceptions of what qualifies someone as a person, it is that they have (or are capable of having) the higher mental functions – such as self-consciousness and reason. These are the mental capacities that Locke cites in his famous and resonant elucidation of what persons are. But it seems that we have to credit these higher capacities to the human animal, in which case accepting the man/person distinction implies that there are two persons – as one might say, the non-animal person to be distinguished from the animal, and the personal animal. It would seem highly paradoxical to hold that there are two persons. We seem to have derived two paradoxical consequences from the man/person distinction. Clearly the opponents of animalism need to do some serious explaining at this point.

Here is a final line of thought. I have suggested that the question about the relation between the person and the animal can be viewed as a question about ourselves. How am I related to this animal? That is in effect what I want to know. We can get some purchase on the question if we can determine what “I” refers to, as used in that question. Does it refer to the animal or not? But it seems to help with that question to ask how it comes about that there are “I” users in the first place. As we might say, the “I” in the crucial question refers to that thing, whatever it is, that “I” talk exists in order to refer to. But it would seem to be a plausible principle that such cognitive/linguistic devices have emerged in the course of evolution because they enable advanced animals to self-refer. We naturally think of such cognitive devices as akin to, say, other devices that natural selection has generated and preserved; they are preserved because they yield functions for animals that animals benefit from. This means that they are devices for self-reference by the animals that are lucky enough to possess them. In this way, reflections on the emergence of self-reference indicates that the subject of self-reference is the animal. Which is to say that it indicates that the animalist identity should be accepted.

This amounts, then, to a sketch of an argument from general biology to metaphysics. This may seem a surprising route to a metaphysical claim, but philosophers often appeal to considerations about language to support substantive (non-linguistic) conclusions, and in this case we are relying on empirical considerations about the origins of language, rather than brute semantic intuitions. Plainly, questions can be put to this line of thought, but it is not, I suggest, absurd.

There are then strong reasons in favour of animalism. However, I think that it is true to say that most philosophers do not accept it. Why? One way to think of the ground of opposition is to notice that animalism is an identity thesis. The person is the animal (where the person is). Now, an identity cannot be accepted if there is a (detectable) property difference between the two items. The ground for opposition is the general conviction that there are property differences between the person (or self) and the animal.

As we have seen there is a massive overlap or congruence of properties if we survey what we might think of as the ordinary features that our lives exhibit. But what philosophers have tended to think is that there are detectable differences in relation to what is possible for the person and for the animal. These differences come out when considering what might be called possible, even if not actual, dissociation cases. There are fundamentally two kinds of such cases. One sort of possible case is where we start with the person and the animal occupying the same space at the same time, but as the case develops the person is counted as ceasing to exist whereas the animal remains. One candidate for such a case would be where someone suffers a traumatic injury which makes them what we sometimes call a “human vegetable”. There is life (perhaps artificially sustained), and so surely an animal, but no mental capacity. About such cases many judge that the person no longer exists.

Another type of example that has been suggested fitting this pattern is what used to be called cases of multiple personality disorder. As popularly conceived in such occurrences there is associated with a certain human animal a definite personality, linked to a battery of memories and values, and a name to which they answer. After some time this psychological syndrome vanishes and is replaced in the same animal by contrasting personality, set of memories and values, and self professed new name. There is one animal but it is proposed that one person (or self) has been replaced by another. Cases of this sort can be called [A and not P] cases – indicating that they are supposed to be examples where there is the same animal but not the same person.

But the sorts of cases that have had most influence on philosophers are ones which can be called [P and not A] cases. These purport to be cases where we start with a (coincident) person and animal but which develop in such a way that the person remains but the animal is no longer there. The example that has seemed most convincing is that of brain transplants. The standard example was first set out by Sydney Shoemaker. In his example we start with two people, Brown and Robinson. Both have their brains removed (for medical reasons) but, by mistake, Brown’s brain is returned to Robinson’s body, and to avoid extra complications Robinson’s brain perishes. It seems that we now have one person whom we can name Brownson. But is it Brown or is it Robinson?

Shoemaker suggests that the correct judgement is that Brownson is in fact Brown. Assuming that our psychological character is preserved in the brain, we can say that Brownson will remember Brown’s life, have Brown’s character and beliefs, and Brown’s values, and so on, and will be sure that he is Brown. This psychological similarity and continuity between Brown and Brownson, according to Shoemaker and most philosophers, makes it the case that Brownson is Brown. In which case we have the same person, but on the face of it Brown is no longer housed in or attached to the same human animal. So there is the same person but not the same animal. Now, there are, of course, other candidate [P and not A] examples – for example, head transplants – but brain transplants have seemed very powerful anti-animalist cases.

The initial anti-animalist argument derives then from the conviction, encouraged by reflection on such cases, that the person and the animal can “come apart”. If so, they can hardly be the same thing.

If the arguments on both sides have struck you as powerful and plausible then we have to acknowledge that we have the task of evaluating them and also assembling, if we can find it, new evidence to determine what the truth is about our relation to the animals where we are. I’ll now raise some questions about the anti-animalist case that has been sketched. On the whole, in these anti-animalist examples, it is not really in doubt what the correct thing to say about the animal is. We feel confident that we can trace it in the developing story. The issue that is more unclear is what is happening to the person. What is it right to say about the person in such cases?

What of the [A and not P] examples? There is no general demonstration against the possibility of such cases, and so it is a matter of deciding what to say case by case. It helps, it seems to me, to consider such stories by imagining that they involve people who matter to you. So imagine your father suffers the accident. Would you seriously deny that he, your father, was himself still alive and in the hospital bed? When the hospital authorities ask you when you will be visiting your father would you really react by asking who they are talking about, since your father no longer exists? It seems to me you would not.

Next, imagine in the second example that it is your brother. Do you seriously have any inclination to think that your brother’s whereabouts are unknown and that talking to you is a totally new person? Rather, we think of such abnormal psychological developments as illnesses that befall people. Our aim is to cure the people to whom they happen. More needs to be said, but despite their popularity in philosophical folk-lore such examples do not really seem like genuine [A and not P] cases. The [not P] description is implausible.

Such [A and not P] cases seem, on reflection, fairly weak because we have an established way of thinking of them according to which they really concern the same person, even if it is a way we tend to lose sight of when philosophising. With suggested [P and not A] cases the situation is different. We do not have an established way of thinking about them, since they do not occur. The fundamental question is whether opponents of animalism are right to be confident as to what is correct to say about such examples. Once we ask this, possible reasons for being less confident suggest themselves.

First, even if more or less everyone initially regards the person as going with the brain, it has to be acknowledged that the final verdict needs to be made in the light of all the evidence, and so it may be that some initial and quite strong convictions have to discarded. I have not investigated here lines of reply to the pro-animalist arguments – and that is a weakness in the overall strategy of argument here – but they also surely seem strong and convincing. So there is no entitlement to insist that the brain transplant intuition just has to be accepted.

Second, we can ask whether we are entitled to trust what I am calling our brain transplant intuitions. We can distinguish two broad conceptions of our conceptual practice of tracing ourselves over time. On one conception we do so in virtue of having an understanding of what our kind of persistence involves, an understanding that allows us to generate reliable verdicts about all sorts of imaginary cases. The second conception is that we are entities that are able to trace themselves in the world we find ourselves in, but that we must fix what the nature of our persistence is by investigating what kinds of things we in fact are tracing. Then, but only then, can we determine verdicts about such merely imaginable cases. This is a rough contrast, but people who feel confident about what to say in brain transplant cases need to consider whether such confidence is based on a proper understanding of our conceptual practices.

Third, an alternative way of thinking about such cases can be proposed, and it may not be without merit. Organs can be transplanted. What was one animal’s liver can become another animal’s liver. Similarly, the organ of cognition (or mentation) which sustained cognition for one thing might be able to become the organ of cognition for another creature. On this way of thinking it is not that brain transplants transfer people; rather a single organ for doing something goes from one person to another. Robinson may get a new brain, just as he can get a new liver. Philosophers who attach supreme importance to thought and cognition may find this an unnatural way of thinking, but that may not be how it does, and should, strike everyone.

I have been trying to put some question marks against some anti-animalist arguments. But it is obvious that these moves constitute merely the first moves in a complex debate, and in some respects I have not even sketched the first moves of the debate. What, I believe, we need to cultivate in explorations of our own nature and persistence conditions is the ability to resist being swept away from solid and clear ways of thinking into realms of fantasy, where more or less anything goes. But it is too early yet to say whether animalism is the view that avoids fantasy.

Paul Snowdon is Grote Professor of Mind and Logic at University College London. His forthcoming books include a collection of his own papers entitled Essays on Perceptual Experience, and, jointly edited with Stephen Blatti, Essays on Animalism. Both will be published by Oxford University Press in 2012.