There are many questions concerning the nature of personal identity, but perhaps the most discussed is the problem of persistence. Each of us was once a young child, and many of us will survive into old age. The common sense view is that, despite many superficial changes, that young child is the same person as you; and that you will become that older person. This raises the philosophical question: exactly what is necessary for a past or future being to be you? Or to put it another way: what type of changes can a person undergo without ceasing to exist?
Any answer to this question will have far-reaching implications. Would we be the same person if we were in a vegetative state? What if someone undergoes a profound alteration such as amnesia, religious conversion or sex reassignment surgery? Can we be held responsible for the things we did decades ago? The problem of what constitutes survival is also crucial for those who believe in the possibility of life after death.
In each case, the type of identity under discussion is numerical identity: that is, whether your earlier self and your later self are one thing rather than two. The issue is not one of qualitative identity – two things being exactly similar – because of changes that occur to people over the course of time. We can assume that you and your childhood self are qualitatively different (in terms of size, shape, strength and so on) but are numerically identical (you are the same person). Without this assumption, the notion of personal identity persisting over time would be incoherent.
The so-called ‘simple view’ of personal identity is a consequence of substance dualism. Philosophers like Plato and Descartes identified a person as a non-reducible, non-physical element (a ‘soul’). The persistence of the soul, according to such philosophers, is what guarantees identity over time. While the theory neatly avoids issues of physical change, its basis in the problematic and unpopular theory of dualism renders it difficult to defend.
When we ordinarily speak of survival or persistence over time, we tend to refer to physical things. If someone survives an accident, or if a missing person returns home, it is the survival or persistence of their body that matters. This seems to be true even in extreme cases of amnesia or profound changes in personality: the identity of the person is determined physically. In fact, it is hardly possible to think of a single real-world case where the criterion has failed. This also accords with another intuition. We all recognise that we were once a foetus; similarly, we accept that in the future we could be in a persistent vegetative state. In both cases there is no shared mental activity, and therefore the only criterion to explain identity with our current self seems to be physical continuity.
Despite the strength of these intuitions, it is difficult to explain exactly what such a physiological criterion might be. As John Locke noted, we speak of the same oak tree from the sapling to the full-grown tree, but there seems little in common between the two. The ancient puzzle known as the Ship of Theseus draws attention to these difficulties. If Theseus sets sail in his ship, and then over time replaces each part of his ship with a new one, is it the same ship when he returns? Furthermore, what if someone were to gather all the old parts and reassemble them? Would this second ship be (numerically) identical to the original? The two later ships cannot both be identical to the original ship as they are themselves different (identity is a ‘transitive’ relation, so if A = B and B = C then A = C). If we reflect on the changes that our bodies go through over time, including natural cell regeneration, organ transplants, and so on, the relevance of the analogy with the ship becomes obvious. It is not easy to identify what (if any) physical elements must remain the same for two things to be considered identical.
Some ingenious thought experiments by Derek Parfit explore this problem further. Imagine a teleporter station on Earth, in which your molecules are disassembled and then sent to Mars, where they are perfectly reassembled. It seems that the person on Mars, and the person who left Earth, are identical as they are physically the same (although not exactly continuous). Parfit then goes on to ask, what if only 99% of the body was sent from the first station, and the shortfall was made up by a store of molecules on Mars? Or 50%? Or 5%? Is it possible to draw a line and say that a particular proportion of our bodies must remain the same to ensure identity? This seems unlikely, and raises awkward questions about amputation and other serious physical injury.
Now imagine a different type of teleporter station in which your molecules are scanned but not sent: blueprints of your physical organisation are transmitted instead, and your body is duplicated on Mars – perfectly but from different molecules. By any physical criterion, the duplicate would not be you since all physical traces have been replaced (like Theseus’ ship). But our intuition tells us that the person stepping out of the station on Mars would feel like the same person, as if they had merely stepped off a bus. In this scenario, it also seems relevant to ask what happens to the original body. Traditionally it is destroyed, but Parfit wonders what would happen if that destruction were delayed for a few minutes: during that time, the Earth-you could communicate with the duplicate on Mars, making the notion of identity very difficult to capture. For those few minutes at least, the Earth-you seems to be the ‘real’ you because of an uninterrupted physical continuity.
It is worth noting the relevance of these scenarios to the conception of physical resurrection after death. If such a process exists, it could not plausibly follow the pattern of the first station: a body sheds too many particles over time, and many of them will have become a part of other people. But if resurrection involves the creation of a replica, as in the second teleporter example, then there appears to be no bodily continuity at all. Moreover, in such cases there is also a gap – perhaps a long one – between death and resurrection, making any connection between the original and the resurrected person even harder to understand.
Another thought experiment seems to lead our intuitions away from a criterion of bodily continuity. Bernard Williams has asked what would happen if two people ‘exchanged bodies’: that is, if the entire mental contents of A could be transferred to B’s body, and vice versa. Imagine you are A before the operation. The surgeon announces that after the operation, one of you will receive a reward, and the other will be tortured. What outcome would you wish for? If you are A, you will probably reflect that ‘you’ will be in B’s body after the exchange, and therefore will choose the A body to be tortured. After the operation, you will be watching from the B body, glad that you made the right choice. In this case, you appear to be wherever your mind is and can leave your body behind without losing your identity.
This suggests that what matters in identity is rather some type of mental continuity, and that the body is incidental to our continued survival. But consideration of the foetus and the patient in the vegetative state suggests otherwise: we experience the world as embodied beings, as human animals, and our physical continuity seems inseparable from who we are. In Williams’ scenario of the exchange of bodies, it seems that the human animal ceases to exist, while the ‘person’ continues in a new body (as a new human animal). But to say that a human being is two things with different criteria of identity is to violate the transitivity of identity, like the two ships of Theseus.
A more common approach to personal identity is to claim a psychological criterion: I am identical to my younger and older selves because of some kind of mental continuity, consisting in the same beliefs, memories, preferences and so on. This is also the approach of philosophers who claim that identity consists in physical continuity of the brain, since it is the brain’s cognitive capacity that matters in this regard. A theory of psychological continuity sits well with our intuitive reaction to brain-transplant or body-swap thought experiments: ‘I’ would be wherever my mental life takes place. As John Locke put it, if a prince were to wake up in the body of a cobbler he would still be the prince – even though ‘he would be the same cobbler to everyone besides himself’ (Essay II.xxvii.15).
However attractive the theory is, however, it is extremely difficult to provide a formal account of any psychological criterion. For example, some philosophers have considered memory a necessary condition of identity: I am identical with my younger self because I remember being that person. This type of direct connection was famously refuted by Thomas Reid:
Suppose a brave officer to have been flogged when a boy at school for robbing an orchard, to have taken a standard from the enemy in his first campaign, and to have been made a general in advanced life. Suppose also, which must be admitted to be possible, that when he took the standard he was conscious of his having been flogged at school, and that when made a general he was conscious of his taking the standard, but had absolutely lost the consciousness of his flogging. Essays on the Intellectual Powers of Man 3.6
Reid draws the inevitable conclusion that if memory is the criterion of identity, then the boy is the same person as the young soldier, and the young soldier is the same as the general – but the general cannot be the same as the boy since the memory of the flogging has been lost. This violates the transitivity of identity and is wholly counterintuitive. Moreover, if memory is a criterion for persistent identity, it is difficult to account for periods of which we have no memory (early infancy, sleep, coma, amnesia) unless we were not ourselves at those times. There is also something question-begging about the memory criterion: first-person memories by definition must belong to one person, so proving the identity of two individuals (say, a child and the adult they become) on the grounds of such memories is a circular argument.
These objections can be escaped by avoiding direct connections like memories, and instead defining the criterion in more general terms, much as we would define the term ‘person’ itself. John Locke’s definition of a person is a good example:
... a thinking intelligent Being, that has reason and reflection, and can consider itself as itself, the same thinking thing in different times and places. Essay II.xxvii.9
However, appealing to such a broad definition presents problems. If such overlapping, indirect psychological connections constitute the existence of a person over time, then we are forced to agree that the foetus that grew into me is not the same person (since it lacks the properties Locke identifies), and if I am later in a vegetative state, that patient will also not be me for the same reason.
Attempts to explain persistent identity through psychological continuity also struggle with the notion of fission. Imagine that your brain is split into separate hemispheres and each hemisphere is placed into a different body. If psychological continuity is the criterion of identity, then (like with Theseus’ ship) we have two different persons both identical with an original. The problem with all fission scenarios is that they seem to violate the requirement of numerical identity. It is logically possible, of course, that the original ‘you’ contained two persons all along, and they simply forked. This would have very serious implications for our notion of personhood. Other philosophers have claimed that neither of the resulting persons is identical to the original, precisely because identity must be numerical. But then imagine that one half of your brain was destroyed rather than transplanted: in this case, there would be no problem establishing identity. It seems odd that identity only occurs on condition that a separate thing is destroyed. Defenders of the psychological criterion have not yet found a comprehensive answer to these fission scenarios, although it is unclear how seriously to take objections that are the stuff of science fiction (such examples permeate all discussions of personal identity).
Some philosophers reject the psychological criterion because it entails that biological continuity is not a necessary condition of identity over time. This position fails to take account of our existence as biological organisms and raises questions of how I, as a person, relate to my body. Bernard Williams suggests the following scenario: if you are told that you will be tortured, but that your personality and memories will be replaced beforehand, should you be afraid? It seems obvious that you should, just as someone might fear a car crash with a similar result (amnesia and subsequent pain). This scenario once again appeals to our intuition about bodily, not psychological, continuity.
We might also ask about radical mental changes: amnesia, religious conversion, brain damage and so forth (see the famous example of Phineas Gage). The striking element in such cases is precisely the discontinuity in psychology, but we would tend to consider that the subject in each case remains the same person. Or perhaps we would not. After all, the principle of rehabilitation in the penal system seems to imply that a person can radically change and therefore be released from imprisonment for a crime that their earlier self committed. Our intuitions on these matters are far from clear, no doubt because we do not yet have a conclusive theory of personhood.
The conflicting intuitions generated by these thought experiments have led some philosophers to reject altogether the existence of a criterion for personal identity through time. But this itself is puzzling. I am the same person as my younger self but I do not share that identity-relation with any other individual, which suggests that there is something particular about the relationship that can be identified. To reject any criterion is itself deeply counter-intuitive.
A more promising approach to these problems has come from Derek Parfit. Like John Locke, he recognises a distinction between persons and their bodies, although they are closely connected. This allows for the teleporter scenario and psychological continuity to guarantee the persistence of a person. But unlike Locke, he draws a distinction between survival and identity. In a fission case, the original person has survived in both resulting persons, even though identity has not been preserved. Survival, says Parfit, is all we need to worry about. Of course, this requires a definition of survival that does not presuppose identity; and any such definition must account for the role of identity in first-person memory. It is not yet clear that this is possible.