1. Introduction
It seems obvious that we (the authors) and you (the reader) are animals, members of the species, Homo sapiens. But in the contemporary philosophical debate on personal identity, the claim ‘we are animals’ must be distinguished from a more controversial thesis: Animalism. The claim ‘we are animals’ is a statement about our current biological nature. Animalism, by contrast, is a robust thesis about our persistence over time: that our identity is fundamentally that of a biological organism and we persist just in case that organism persists.
Why does this distinction matter? The stakes concern what we are, fundamentally, and what our survival consists in. If Animalism is true, you are the unthinking foetus you developed from, and you would remain the same individual even in a persistent vegetative state, while the biological life of that organism continues. Your psychology is not essential to your survival. If Animalism is false, your persistence might instead be tied to psychological continuity (as neo-Lockeans hold), or, as we hold, a more complex hybrid of biological and psychological continuity. This debate gets to the heart of what it means to be a person.
The most influential argument for Animalism is the ‘thinking animal argument’. Its basic form is as follows: (1) There is an animal sitting where you are. (2) Every animal sitting where you are is thinking. (3) You are the only thinking being where you are. So, (4) You are an animal.
Many philosophers have taken this not only to establish that we are animals but also as a sound argument for Animalism.
This paper’s central claim is that this is mistaken. We do not dispute the thinking animal argument itself; in fact, we believe it is a sound argument, given the assumptions stated below. The significance of our paper lies in showing why its conclusion does not entail Animalism.
We argue that one can accept the thinking animal argument without commitment to the far stronger claims about the necessary conditions of persistence that Animalism requires. By showing that the argument is sound, given the assumptions made, but that Animalism is false, we detach the theory from its primary pillar of support. This move defangs the argument’s power, opening the door to alternative models of personal persistence, such as the ‘hybrid view’ we defend, which can fully accept the biological fact that we are animals without accepting that our persistence is merely biological.
However, we do not think that the thinking animal argument is good for nothing. Its conclusion, together with the empirical facts about the origins of every human animal that has so far existed (including you), is inconsistent with the neo-Lockean psychological continuity account of personal identity, or, better put, personal persistence.
So, in clarification, we argue that the thinking animal argument fails as an argument for Animalism for two related reasons:
(i) Animalism, in the philosophically interesting sense at issue in the contemporary debate, is not entailed by, and does not entail, the proposition that we and you are animals.
(ii) One can accept the thinking animal argument without being an Animalist by accepting a plausible position (which, following Olson, we call ‘the hybrid view’) now under active discussion in the literature.
Regarding reason (i), the philosophically interesting sense of animalism, we think, is the following:
(AN) Necessarily, any person that ever materially coincides with an animal (e.g., of the species Homo sapiens) is (identical with) a biological animal.
(AN) entails that persons coincident with animals have the persistence conditions of animals. We explain and defend this formulation below. But note that (AN) is clearly not entailed by the thesis that we and you, in particular, are animals (since it is the necessitation of a universal generalization), and also does not entail that we and you are animals (since it is compatible with our non-existence).
Below we conclude not only that (AN) is logically independent of the thesis that we and you are animals, but also that its philosophical interest is entirely due to its being entailed by a more general thesis discussable without mention of us, or persons in general, at all – the thesis we call Animalism Generalized:
(ANG) Necessarily, any sentient being that materially coincides with an animal is (identical with) a biological animal.
Regarding reason (ii), the hybrid view is adopted by those who reject both Animalism and the neo-Lockean psychological continuity account of personal persistence. It is defended, in various versions, by several people. We only sketch it here. But we say why we think it is preferable both to Animalism, as we understand it, and to the psychological continuity view as an account of personal (or sentient) persistence.
After elaborating on reasons (i) and (ii) below, we conclude by explaining why we do not think the rejection of Animalism absurd, though it requires accepting that, in certain circumstances, there are many coincident thinkers.
In summary, then, we argue that the thinking animal argument establishes that we and you are animals, but that Animalism, understood as saying something interesting, is false.
2. The Thinking Animal Argument and its core assumptions
Animalism is generally formulated as the thesis that we are animals. So, we begin there, though, as noted, we think more needs to be said before this thesis becomes a philosophically interesting thesis (Animalism deserving of a capital ‘A’). The most-discussed argument for it, repeated here, is the ‘thinking animal argument’ (Olson Reference Olson, Martin and Barresi2003: 325).
1. There is an animal sitting where you are
2. Every animal sitting where you are is thinking
3. You are the only thinking being where you are
4. You are an animal
Another argument, with which we shall compare this, is the following (Trixie is a dog, an actual, currently alive dog, which matters, as we will see later):
1. There is an animal sitting where Trixie is
2. Every animal sitting where Trixie is sitting is thinking
3. Trixie is the only thinking being where Trixie is
4. Trixie is an animal
Since Trixie is not a person, this is not an argument for Animalism. But it relates to (ANG) as the thinking animal argument relates to (AN). As indicated, we think that it is (ANG) and not (AN) we should be discussing when discussing Animalism.
We now get back to the original thinking animal argument. It is obviously valid. It is also, we think, sound, given two assumptions. Assumption 1: Endurantism is true. As Olson notes (Olson Reference Olson1997: 5), for perdurantists, or four-dimensionalists, the thinking animal argument is not worth discussion. This is because perdurantists maintain that as well as persons there are person-stages, shorter-lived person-like things coinciding with persons that can do everything that persons do, or, at least everything that persons can do over a short time. So, perdurantists think there is an abundance of thinkers wherever there is a thinking animal, and so will reject premise 3. Assumption 2: You are a typical human person with no more than one hundred years left to live. Why this matters will become clear later. But, given these assumptions, we think that Olson’s defence of the premises of the thinking animal argument (in, e.g., his (2003)), is perfectly coherent.
These comments carry over, mutatis mutandis, to the ‘Trixie argument’. It too is sound (given the assumptions listed, amended where necessary). But it does not establish (ANG), since its conclusion is just a claim about Trixie. And (ANG), we think, is false.
We now explain.
3. Defining Animalism and the problem of coincidence
First, we must get clearer what animals are. We accept much of what Olson says. We accept that the persistence of animals requires satisfaction of no psychological criteria whatsoever, and that animals persist if they satisfy purely biological persistence conditions.
This is a significant choice point. Other opponents of Olson’s animalist position do not accept these things. Shoemaker (Reference Shoemaker and Gallagher2011: 317; Reference Shoemaker, Blatti and Snowdon2016: 128) does not. Shoemaker argues that human persons are animals in the familiar sense, which must be a respectable one, in which dogs are. He also argues that dogs, like persons, go where their cerebra go in the (science-fictional) cerebrum transplants discussed in the literature that preserve full psychological continuity (but, of course, would not go where their cerebra went if the cerebrum transplant did not preserve psychological continuity). So, he maintains, the persistence conditions of both persons and dogs are partly psychological. What is left behind (if life support is provided) in the transplant case is something, he allows, which may be called an animal in a ‘technical’ sense (2016: 129); he calls it a (human or canine) biological animal. What Shoemaker calls the ‘technical’ sense is the sense in which Olson uses the term ‘animal’. We, following Olson, always use the term in this way in this paper (so we used it in this way in formulating the thinking animal argument above).Footnote 1 We sometimes speak of ‘biological animals’ as a reminder of this.
We also accept that, as a matter of fact, at present, all animals begin their lives as foetuses devoid of any psychology, and that some end them as unthinking comatose things with no more psychological traits than a tree.
Next, we note that having accepted all that, we must also accept that the conclusion of the thinking animal argument is inconsistent with the psychological continuity account of personal persistence (i.e., that psychological continuity and/or connectedness are both necessary and sufficient for continued existence). Given the facts of animal conception which have obtained throughout human history so far, you are now biologically continuous with an unthinking early-stage foetus. Assuming that you are a biological animal it follows that you were that unthinking foetus. But according to the psychological continuity account of personal persistence, you were not. So, the conclusion of the thinking animal argument conflicts with the psychological continuity account of personal persistence.
However, rejecting the psychological continuity account based on what it says about you does not commit us to (AN).
So, why formulate Animalism as (AN)? As said, animalism is often initially expressed in the words ‘we are animals’. Olson sometimes does this, suggesting that it is a thesis about us, rather than about (singular) you (see, e.g. Olson Reference Olson, Martin and Barresi2003). But expressed this way the thesis is not at all clear. What is the scope of the plural indexical ‘we’? In each context of utterance, the reference of ‘we’ refers to a class of individuals (including the speaker) saliently related to the speaker. Unless this class is known it is not known what is being said. So, when animalism is expressed by saying ‘we are animals’, explication is necessary (as noted by Snowdon Reference Snowdon2014: 26). Olson, recognising this, says what he means by ‘we’ is ‘human persons’. But what does ‘human persons’ mean? Not ‘persons who are animals of the species Homo sapiens’, because then animalism would be the trivial thesis that persons who are animals of the species Homo sapiens are animals.Footnote 2 In his (2003) Olson considers this point and says:
Is animalism the mere tautology that all animal people are animals? No. I say that you and I and the other people who walk the earth are animals. If you like, all human people are animals, where a human person is roughly someone who relates to a human animal in the way that you and I do, whatever way that is. (Olson Reference Olson, Martin and Barresi2003: 320)
But later Olson notes that there is no one relation you stand in to a human animal. He writes:
When we see you, we see a human animal; when you move a human animal moves; you perceive the world via a human animal’s sense organs, and so on. This makes you human…. A human person is a person who relates to a human animal in this way. It is, we might say, someone with a human body. (Olson Reference Olson2007: 10)
Perhaps, then, ‘human person’ can mean ‘person whose body is the body of an animal of the species Homo sapiens’ or, assuming rejection of dualism, ‘person who materially coincides with an animal of the species Homo sapiens’. If we interpret the explication in the first way, to avoid rejection of dualism, we need to explain what it is for a body to be the body of a person. This is difficult (for an attempted answer see Shoemaker, Reference Shoemaker1976). Assuming rejection of dualism we can perhaps just read ‘materially coincides with’ as ‘shares its matter with’.Footnote 3 This is not problematic for our purposes. Animalism can thus be expressed non-indexically. Here we follow Snowdon (Reference Snowdon2014: 26), who offers what he calls a ‘rough’ formulation:
(APA) Necessarily, if we have a person in the same place as an animal then that person is the animal.
Tidying up, we get to the formulation (AN) given in Section 1 above:
(AN) Necessarily, any person that materially coincides with an animal (e.g., of the species Homo sapiens) is (identical with) a biological animal.
(AN) is trivially true if it is impossible for a person to materially coincide with an animal. Dualists and materialists will disagree over this. But the interesting current debate about animalism is between philosophers (like Shoemaker and Olson) who accept the possibility of this coincidence.
The formulation of Animalism as (AN) is consistent with the modesty of Animalists. They do not deny that our universe may be teeming with Cartesian egos, disembodied souls, inorganic robotic persons, and so on.
So, we take (AN) to be what Animalists defend, or at least are committed to, even if they prefer the indexical formulation.
In Section 4 we explain why we think both (AN) and (ANG) are false. As a reminder, we will do so by explaining their inconsistency with our preferred view of personal (and sentient) persistence, the hybrid view which we sketch in Section 5. As a further reminder, despite this, we think the thinking animal argument is sound, given the assumptions made (and hence our main point, that the thinking animal argument is not a sound argument for Animalism). First, though, we explain what is at issue in the traditional debate about personal identity. We do this because it is crucial to recognise that this problem concerns only de dicto necessities. It does not concern de re modality at all.Footnote 4 We explain this point in section 4, before formulating and defending the hybrid view in de dicto terms in Section 5.
4. Personal identity and de dicto modality
De dicto modal statements are statements concerning propositional necessities and possibilities and have the modal operator governing the entire proposition (e.g., necessarily, for any x, if is a bachelor, then x is unmarried). De re modal statements attribute, e.g., necessities directly to an object, and so have a modal operator inside the scope of the proposition (e.g., ‘Socrates is necessarily human’ or ‘any man is necessarily mortal’). To see that the traditional problem of personal identity concerns only de dicto statements, consider the following formulations.
From Swinburne:
What are the logically necessary and sufficient conditions for a person P2 at a time t2 being the same person as a person P1 at an earlier time t1, or loosely, what does it mean to say that P2 is the same person as P1? (Swinburne 1984: 3)
From Parfit:
What is it that makes a person at two different times one and the same person? What is necessarily involved in the continued existence of each person over time? The answer to [this question] can take this form: ‘X today is the same person as y at some past time if and only if …’. This answer states the necessary and sufficient conditions for personal identity over time. (Parfit 1984: 202)
More recently, Shoemaker explains the more general problem of the identity (or persistence conditions) of things over time as follows:
When a thing persists over an interval of time there is a series of … property instances that constitutes the career of the thing over that interval … What a career really is a series of are not individual property instances but sets of simultaneous property instances. These sets will consist of instances belonging to a single thing – instances standing in a synchronic unity relation. These can be called ‘thing stages’. If the career is a career of a tree … each tree-stage will be a state of affairs consisting of a single tree’s having a multiplicity of properties at a time. The question of whether a series of tree-stages is the career of a tree is the question whether it is one and the same tree that is involved in these different states of affairs. (Shoemaker Reference Shoemaker, Gasser and Stefan2012: 124)
And in his (Reference Shoemaker2004), comparing questions of diachronic identity and questions of synchronic identity (which are obviously de dicto), Shoemaker writes:
[F]or each kind of persisting subject, it should be possible to give an account of what it is for different states, whether occurring at different times or at the same time, to have the same subject of that sort…. There are clearly informative things to be said about what makes it true that the building having marble columns at its entrance at one location is the same as the building having a rusty fire escape attached to it at another location – equivalently that the instantiation at the one place of the first architectural feature has the same subject as the instantiation at the other place of the second architectural feature. And there are clearly informative things to be said, involving spatiotemporal continuity and relations of causal and counterfactual dependence of later states on earlier one, of what makes it true that a building having certain features at one time is the same as one having certain features at another time. Specifying these synchronic and diachronic unity relations … tells us something about what it is to be a building. The psychological continuity view attempts to do the same thing for persons. (Shoemaker Reference Shoemaker2004: 610)
And whilst Olson’s (Reference Olson1997) concern is not to give an account of personal persistence, he attempts to clarify what such an account, if offered, should provide:
[T]he Psychological Approach is a claim about what it takes for a person to exist at two different times: a criterion of personal identity. (Olson Reference Olson1997: 22)
After some discussion he concludes that part of specifying this will be specifying that psychological continuity is necessary for a person to persist, saying (note the scope of the necessity operator):
[T]he clearest and least misleading way of saying [this is] “Necessarily, for any x that is a person at t, and any y that exists at another time t*, x=y only if x is at t psychologically continuous with y as she is at t*.” (Olson Reference Olson1997: 41)
We submit that these formulations provide evidence that these writers conceive problems of identity over time, including the problem of personal identity, as the problem of specifying, via de dicto modal truths, constraints on the histories of things of the relevant sort. Specification will include single, wide scope, occurrences of necessity operators. That the typical method of argument used in discussions attempts to elicit verdicts on hypothetical cases, like the brain or cerebrum transplant case, which need not be understood as strictly counterfactual, is also evidence for this, since this method has no bearing on de re modal questions.
Of course, some discussants in the debate about persistence conditions hold essentialist views, e.g., that animals are essentially animals (i.e., have the de re property of being necessarily an animal). However, as Olson emphasises in his (Reference Olson2015), such essentialist views are irrelevant to the debate about the persistence of persons:
It obscures the central point of disagreement between animalists and their opponents — whether we are animals — by tying it to the unrelated matter of whether animals are animals essentially. It gives the false impression that animalists and their opponents agree that we are animals and disagree only about the modal status of this claim. Whether we are animals has nothing to do with essentialism. (Olson Reference Olson2015: 97)
For further clarification, Mackie’s discussion (Reference Mackie2006: Ch. 8) is illuminating.Footnote 5
Mackie contests the sortal essentialism of Wiggins (Reference Wiggins2001), according to which, e.g., if person is an ultimate substance sortal, then if Aristotle is a person he is essentially (for Mackie this is equivalent to ‘necessarily’) so. She considers an argument for this via the principle that an individual’s principle of individuation is essential to it. She calls this (EPI) and argues that it is unfounded. She concludes that, for all the extant arguments show, an individual with one principle of individuation (e.g., that of a person), might have had another (e.g., that of a centipede). She explains that she means by a principle of individuation:
[W]hat has traditionally been called a “criterion of identity”, a principle that determines answers to questions about identity at a time and over time…. The idea is that if the concept frog supplies [such a] principle for frogs, then this principle will determine … whether a given portion of matter constitutes, at a given time, one frog or two; whether spatio-temporal continuity is a requirement on the continued existence of a frog, … whether the relation between a live and kicking frog and that of its corpse is identity; and in general what sort of alterations are compatible with the survival of an individual that is a frog. (Mackie Reference Mackie2006: 134)
(EPI) then states that if an individual has such a principle of individuation it could not have existed without it.
We think traditional problems of persistence over time are concerned with principles of individuation in exactly the sense Mackie identifies and that one cannot infer from the claim that an individual has such a principle of individuation that it has it in all possibilities, though, as Mackie notes, an individual cannot have different principles of individuation at different times.
It does not follow that the statement of the de dicto necessary truths that specify the persistence conditions of a person has no de re implications. The de dicto constraints are of the form ‘Necessarily, nothing is a person and is not F’. This entails (assuming quantification into modal contexts intelligible) that no actual thing could have been a person without being F. But to specify the persistence conditions for persons one need list only de dicto necessary truths. Of course, given relevant essentialist theses, e.g., that all actual persons are necessarily persons or that all actual animals are necessarily animals, other de re modal claims follow.Footnote 6, Footnote 7
The traditional problem of personal identity, then, concerns how to specify, in de dicto terms, what kinds of changes persons can and cannot survive. Such a specification can be given by identifying which relations R satisfy the following de dicto schemas:
1. Necessarily, for any x, if x is a person, and if x exists at t1 and t2, then R(x,t1,t2).
2. Necessarily, for any x, if x is a person, and if x exists at t1, then if some y exists at t2, and R(x,t1,y,t2), then x=y.Footnote 8
Instances of 1 express ‘passing away’ conditions for persons. They tell us that it is necessarily true that if such-and-such happens to a person, then that person goes out of existence. Instances of 2 express ‘preservation conditions’ for persons.Footnote 9 They tell us that it is necessarily true that if such-and-such happens to a person, then that person survives.
5. The hybrid view and multiple occupancy
With the traditional account of personal identity so understood, we give a sketch, and brief defence, of the hybrid view we favour. We contrast it with Animalism and the psychological continuity account then consider and reply to some objections. Then, in Section 6 we show how it is consistent with the soundness of the thinking animal argument, which reinforces the advertised conclusion that this argument does not entail Animalism.
To see why we favour the hybrid view over Animalism and the psychological continuity account of personal persistence, note that the following questions are the focus of the personal identity debate:
Can persons survive the destruction of their bodies?
Can persons survive an irreversible loss of consciousness?
Animalists answer ‘No’ to Q1 (in the case of those persons that have bodies) and ‘Yes’ to Q2, whilst psychological theorists answer ‘Yes’ and ‘No’ respectively. But the philosophical literature contains excellent reasons for answering both affirmatively.
The main reason in favour of answering ‘Yes’ to Q1 is our reaction to thought experiments involving transference of a person’s psychology into another body, like Shoemaker’s Brown-Brownson case (Reference Shoemaker1963), as cases in which the person goes where their psychology does. This is the transplant intuition.
And the main reason in favour of answering ‘Yes’ to Q2 is that it is plausible that we may end up in an unthinking comatose state in hospital. Call this ‘the coma intuition’. It is persuasively developed by, e.g., Olson (Reference Olson1997) and Snowdon (Reference Snowdon2014). There is also the very powerful ‘foetus intuition’, viz. that it is plausible that we were once unthinking foetuses.
So, because the hybrid view accords with these intuitions (and thus allows us to answer both Q1 and Q2 affirmatively) we favour it.
To elaborate on this in the terms laid down in section 4, consider that those who endorse Animalism will endorse (very roughly) the following view:
• Necessarily, for any x, if x is a person coincident with an animal and if x exists at t1 and t2, then x has a body at t1 and a body at t2 that are linked by chains of spatiotemporal/biological continuity and/or connectedness.
• Necessarily, for any x, if x is a person x is coincident with an animal and x exists at t1, then if some y exists at some t2, and x at t1 has a body that is linked by chains of spatiotemporal/biological continuity and/or connectedness to the body of y at t2, then x=y.Footnote 10
And those who endorse the psychological account (i.e., that persons survive if and only if their psychology continues somewhere) endorse (very roughly) the following view:
• Necessarily, for any x, if x is a person, and if x exists at t1 and t2, then x has a psychology at t1 and a psychology at t2 that are linked by chains of psychological continuity and/or connectedness.
• Necessarily, for any x, if x is a person, and if x exists at t1, then if some y exists at t2, and x at t1 has a psychology that is linked by chains of psychological continuity and/or connectedness to the psychology of y at t2, then x=y.Footnote 11
The intuitions noted above, the transplant intuition, the coma intuition, and the foetus intuition together strongly suggest that neither of these views is correct, and that instead we ought to adopt the hybrid view. To state this, following Shoemaker (and, as Shoemaker notes, without commitment to perdurantism), we can introduce the notion of a relation which holds between two person-stages iff they are stages related either by psychological continuity or by biological continuity. We can then consider the ancestral of this relation and call this simply ‘the continuity relation’. The hybrid view can now be stated (again, very roughly):
• Necessarily, for any x, if x is a person, and if x exists at t1 and t2, then the stage of x at t1 and the stage of x at t2 are linked by continuity
• Necessarily, for any x, if x is a person, and if x exists at t1, then if some y exists at t2, and the stage of x at t1 is continuous with the stage of y at t2 then x=y.Footnote 12
The first of these three views is an account of the persistence conditions of things of the kind, person sometime coincident with an animal. The Animalist thinks this is a kind of animal, like a sometime wounded animal. According to Animalism there is simply nothing to say about the persistence conditions of persons as such. The second and third views are competing accounts of the persistence conditions of persons. So the Animalist view is not answering the same question as the other two. But its consequences are incompatible with them.
According to the third view, the hybrid view, the following is a possible history for a person. Conception occurs. Sometime after, a living creature comes into being lacking any psychology (e.g., as a foetus). This develops. Psychology appears and becomes more sophisticated. Later, the living thing acquires the capacity for self-conscious thought. Thus it qualifies as a person. Further growth and development take place. Later there is physical decline. The person’s body ceases to be functional. Her cerebrum is transplanted into a new body carrying her psychology with it. She continues to exist in this new body with this cerebrum. She then falls into a comatose state, losing all psychological continuity with herself as she was previously. She continues to exist in this state, until her biological life ceases.
This is a possibility that neither the psychological continuity theorist nor the Animalist allows. It is the possibility that a creature which is at some time psychologically endowed survives through some period in virtue only of psychological continuity and through another only in virtue of biological continuity.
Arguably the hybrid view is not just a philosophical construction but accords with a natural, ordinary way of thinking. Belshaw, writing for a general audience, says:
Why think that there is any such thing as the essential me? Why not think, instead that we will try to make the best of whatever comings and goings we encounter? Let’s say Jim gets hit by a car. When there’s just a mindless body in the hospital bed, making the best of things involves allowing that Jim is still there, in a terrible condition. But when, as well as this body, there’s some fabricated equivalent (maybe with Jim’s transplanted cerebrum) sitting alongside, claiming not implausibly to be Jim and thanking the doctors for their efforts, we can do better. Both views – ‘animal’ and ‘person’ – suffer in the same way, they are too ambitious. (Belshaw Reference Belshaw2015)Footnote 13
The hybrid view is more complicated than Animalism and the psychological account, for sure. But it is perfectly coherent, and consistent with all the intuitions it seems compelling to endorse. So we endorse it.
It is adopted by several philosophers (e.g., Langford, Reference Langford2014, and Madden, Reference Madden2016) and is implicit, we think, in remarks of McDowell and Wiggins.Footnote 14 It can be understood in several different ways.
One way is as a version of Nozick’s closest continuer account of identity (Nozick Reference Nozick1981). According to Nozick, persistence is to be understood in terms of continuities, in the case of people and animals, in particular, psychological and physical continuities. A later thing may be a closer continuer of an earlier thing than another later thing because it is more continuous in some way with it than the other later thing, or it may be a closer continuer because it is continuous in a way that is more important than the way in which the other later thing is continuous with the earlier thing. So psychological continuity may, for example, trump physical continuity. For a later thing to be identical to an earlier thing is for it to be the closest continuer. One way to understand the hybrid view is therefore as a version of the closest continuer view in which the relevant types of continuity, for people and animals, are physical and psychological continuity. This is how Madden understands it, who in fact presents the hybrid view as a version of animalism.Footnote 15 So understood the hybrid view faces the familiar objection from Bernard Williams (Reference Williams1957) to psychological continuity accounts which include a ‘no branching’ or ‘closest continuer’ clause. This is roughly that such accounts entail that the identity of later x and earlier y may depend on something that happens elsewhere and has no causal impact on what happens in the region where x and y are.
The simplest way to set up this objection is in terms of symmetrical fission, i.e., cases in which a person P1 develops in such a way that (i) there are two later persons P2 and P3 who are both equally close continuers of P1, and (ii) if P2 (or P3) alone had existed (because, e.g., only one brain hemisphere had been transplanted and the other destroyed) then P2 (or P3) would have been identical with P1. In such cases closest continuer views entail that P1 is not identical with P2 or P3, that P1 goes out of existence and that P2 and P3 are new people who come into existence. But, the objection goes, this entails that a spatio-temporal region which is intrinsically the same in two possible situations may contain, in one situation (where both hemispheres are transplanted so that there is a spatiotemporal region containing first P1 and then P2), two different people, one ceasing to exist and another coming into existence, and in the other situation (where only the left hemisphere is transplanted and the right destroyed), a single person throughout. This is possible even though in the two possible situations what happens elsewhere has no causal impact whatsoever on what happens in that spatiotemporal region.
Madden does not consider such cases, but they are as forceful against Madden’s closest continuer version of the hybrid view as against the original version of the closest continuer view or ‘no branching’ versions of the psychological view (see Skrzypek & Mangino, Reference Skrzypek and Mangino2021, for an explicit deployment of this argument against Madden). We regard them as decisive against any closest continuer version of the hybrid view, whether or not put forward as a version of animalism.
To exclude the Williams-inspired fission objection the hybrid view can be understood instead along the lines of the ‘multiple occupancy’ view of personal persistence developed by Lewis (Reference Lewis and Rorty1976) and Robinson (Reference Robinson1985), according to which, in the case of fission with two equally close continuers, no one dies, because they were there all along. That is how we understand it.Footnote 16 The multiple occupancy view goes naturally with four-dimensionalism, but can be developed independently, as Robinson explains, and the strongest motivation for it is not anything Lewis argues but avoidance of Williams’ objection.
There are, of course, other objections to the hybrid view. For example, one might object that there are possible transplant cases in which both these things occur: (i) The consciousness of a person P, initially coincident with an animal A1, is transferred into some other animal A2 (this is precisely how transplant cases are typically set up), (ii) person P’s animal A1 then lacks any psychology. And in this case, it might be said, P cannot survive as something materially coincident with A1 and as something materially coincident with A2, for then P would be two things and not one. So, the objection goes, the transplant intuition and the coma intuition cannot both be respected, and the hybrid view cannot be true. We reply by noting simply that the defender of the hybrid view can maintain that the coma intuition does not apply in transplant cases such as this. This intuition is to be considered from the first-person perspective (i.e., from P’s perspective), but so considered the intuition is that P will survive as something materially coincident with A2 and not as something material coincident with A1. The immediately following history of the object of first-person reference is to be identified by tracing the unique (if any) path of psychological continuity from the tokened first-person thought. This, the defender of the hybrid view can say, just is the transplant intuition. So, we say that in cases like this P does survive as something materially coincident with A2 and does not survive as something materially coincident with A1. This is perfectly consistent with the coma intuition, which tells us only that P may have ended up in an unthinking comatose state (i.e., were it not for the fact that P’s psychology was transferred into A2).
One might further object that, according to this view, A1 is a biological animal that exists before the transplant as a thinking thing and afterwards as an unthinking comatose thing. Its existence cannot be denied. So, the objection goes, we must admit that in this case before the transplant there were two thinking things present: (i) the animal A1, and (ii) the person P. We do admit this but note that in doing so we admit to nothing more than those who endorse a multiple occupancy response to the familiar problem of branching. In other words, our view here is no more controversial than a standard responseFootnote 17 to a problem that everybody faces, no matter which theory of personal persistence they adopt. (Animalists, who think that human persons survive if and only if their biological animal survives, face it too, if it is in principle possible for animal bodies to split symmetrically such that two biological continuers result (Perry Reference Perry1976).) We maintain that transplant cases like the one described above have a similar structure to the familiar fission case, except for the complete symmetry stipulated in the latter which makes singular first-person thought impossible (Lewis Reference Lewis and Rorty1976).
One might press here by asking: ‘Why isn’t the animal in such cases a person if it is a thinking thing?’ In brief, the answer is this: we use ‘person’ in the way that Locke does. It means ‘a thing that can think of itself as itself’, something with the capacity to think of itself in the first-person way.Footnote 18 Persons are thus references of I-thoughts (i.e. thoughts that are thought in the first-person). When P and A1 materially coincide, they think the same thoughts (they are and always have been microphysically identical, composed of numerically the same particles, how could they not?). Some of those thoughts are I-thoughts. These I-thoughts refer to a thing whose immediately following history is to be identified by tracing the subsequent path of psychological continuity therefrom (that is the transplant intuition).Footnote 19 So, when P thinks I-thoughts, P thinks them about P (i.e. reflexively), and that makes P a person. The animal A1, identically composed, must think about exactly what P thinks about. So A1, when it thinks I-thoughts, also thinks them about P (i.e. non-reflexively). And so it is not thereby constituted a person. In brief, ‘I’ is to be grouped with ‘here’ and ‘now’. The reference of these is determined by the speaker’s intentions and may, because of the overlapping of regions of space and regions of time, be indeterminate. In the standard fission case, where the two later people are related in exactly the same way to the one or ones existing earlier, then according to the multiple occupancy view, because of the overlapping people it postulates, this is the case for the reference of ‘I’ uttered earlier. So there is no singular first-person reference earlier but indeterminate plural reference (‘All of us’/‘At least one of us’ – Lewis Reference Lewis and Rorty1976). In the case we are considering, by contrast, there is an asymmetry. So, despite overlap, there is determinate singular reference.
All the above requires further development. We have given a brief outline here of how to respond to some familiar objections only to give a flavour of the hybrid view. But we emphasise that we are not here concerned to give a full defence of it. Instead, the concern is only to show that Animalism does not follow from the soundness of the thinking animal argument. And now we can show this.
6. Why the Thinking Animal Argument does not entail Animalism
Recall that all the constraints on personhood listed above are given by statements of de dicto necessity (containing a single modal operator with maximal scope).
Now your actual history is consistent with the pair of persistence conditions the Animalist must endorse since you were once a foetus and your future does not contain any cerebrum transplants or the like (since, we assume, you will not reach a hundred years and consciousness transplants are not likely to become widely available within your lifetimes). So, your past and future history is consistent both with the persistence conditions Animalists must endorse and with the formulation of Animalism, (AN). This does not, by itself, entail that you are an animal, of course. But this is not the point. The point is that this, so far, is perfectly consistent with the thinking animal argument being sound.
However, if any future persons undergo cerebrum transplants and go where their psychology goes then (AN) is false. And since the descriptions of transplant cases are consistent and (AN) is a modal claim, it follows from this alone that (AN) is false. So, the fact that the thinking animal argument as addressed to you is sound, if it is, does not establish the truth of AN, i.e., Animalism. It is consistent with the hybrid view.
That establishes the advertised conclusion: the soundness of the thinking animal argument does not entail Animalism.Footnote 20 To finish, it may be said: ‘What if the thinking animal argument were addressed to a person that was going to undergo a cerebrum transplant with consequent transfer of consciousness? Would it still not be valid and sound?’ The answer is ‘No’. It would be valid but not sound. Premise 3 (i.e., ‘You are the only thinking being where you are’), addressed to P, would be false. Either P’s consciousness is transferred from an animal A1 into another animal A2, but A1 nonetheless survives. Or, P is simply reduced to the size of his cerebrum, and A1 is rubbed out of existence, before the transplant takes place. Either way, before the transplant, P and A1 are materially coincident but non-identical. So, premise 3 is false, as there are two thinkers present before the transplant, as the hybrid view, understood in conformity with the standard multiple occupancy story about fission, allows.
Now, we think you know that you will not have a cerebrum transplant; the technology will not develop in time for you. But future people may not be entitled to that confidence. They will then not know whether they are temporarily coincident with other thinkers. If the thinking animal argument is addressed to them they should say that they do not know whether the premises are true. And likewise, to return at last to the ‘Trixie’ argument and Animalism Generalized, future pet owners, aware of the rich technology available to their vets, should not say that the premises of ‘Trixie’ type arguments are true either; they should just say that they do not know. They should not endorse ANG; nor therefore, should we.Footnote 21
The best argument for Animalism is the thinking animal argument. It is not good enough. We should not be Animalists.
Open access
