STEPHEN COLEMAN
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY (Accepted in revised form 28 May 1998)
Thought experiments are profitably compared to compasses. A compass is a simple but useful device for determining direction. Nevertheless, it systematically errs in the presence of magnets . . . it becomes unreliable near the North Pole, in mine shafts, when vibrated, in the presence of metal . . . experts will wish to use the compass as one element in a wider portfolio of navigational techniques. Analogously, thought experiments are simple but useful devices for determining the status of propositions. Sadly, they systematically err under certain conditions and so are best used with sensitivity to their foibles and limited scope (Sorensen, 1992, pp. 288–289).
Thought experiments have been used by philosophers and scientists for centuries, yet it has really only been in recent times that their use has been questioned. Sorensen points out that they are best used ‘with sensitivity to their foibles and limited scope’ (1992, p. 289) yet they appear instead to have been used extensively and indiscriminately. The area where this seems most apparent is in the study of personal identity. Thought experiments have been used so extensively in this area, and the thought experiments themselves have been so unusual, and even bizarre, that Kathleen Wilkes felt constrained to apologise for not using them in her book Real People (1988, p. 1). Despite their prevalence, the use of thought experiments in this area of philosophy has been criticised in recent times. Bernard Williams criticises the conclusions that are drawn from some experiments, and retells one of these experiments from a different perspective, a retelling which leads to a seemingly opposing result (1973, pp. 46–63). Wilkes criticises the method of thought experimentation itself, suggesting that the results drawn from the experiments are tainted by a faulty method (1988, pp. 1–48). Another influential critic of thought experiments in personal identity is Mark Johnston (1987), who included criticisms of both these types. I Philosophical Studies 98: 53–69, 2000. © 2000 Kluwer Academic Publishers. Printed in the Netherlands.
54
STEPHEN COLEMAN
would like in this paper to examine these two types of objections, and see if there is any basis to these claims. I will however, only be dealing with the specific objections raised by Williams and Wilkes, as I feel that Johnston’s position is only an extension of the views held by the other two writers, and any criticisms made against them will apply equally to Johnston. In order to achieve my aim, it will be necessary to enter into some detailed examination of thought experiments, and see why Wilkes finds their use problematic. Thus it is with this question that I will begin this paper, before moving on to Williams’ more particular criticisms. In her book (1988), Kathleen Wilkes makes an examination of thought experiments, and divides them up into several types. A similar study is undertaken by Roy Sorensen in his book (1992). In beginning my examination of thought experiments and their role and interpretation in personal identity, I would like to combine these two discussions, and divide thought experiments into various different types. This is important in understanding Wilkes’ objections, as it is only one particular type of thought experiment that she finds problematic. The first major division of thought experiments is into scientific and philosophical thought experiments. Wilkes suggests (1988, pp. 1–6) that scientific thought experiments usually involve themselves with theories of the way that things actually are, and are often devised to point out flaws and inconsistencies in existing theories, while philosophical thought experiments on the other hand, tend to be involved with questions of what humans actually think and believe about certain situations. I believe that trying to differentiate between thought experiments in this way is unhelpful, and tends to mislead. As Sorensen points out (1992, p. 11) there are a great variety of scientific thought experiments, used for many different purposes, and the same applies to philosophical thought experiments. Some scientific thought experiments are used in a way more typical of philosophical experiments, and some philosophical experiments are used in a way more typical of the scientific experiment. Thus it does not really seem to be useful to divide thought experiments in this way, as any division will be somewhat arbitrary, ignoring the aim of the experiments, and concentrating
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
55
instead on the classifier’s inevitably biased view of what is and is not scientific. All thought experiments can more usefully be divided into those experiments that are physically possible, and those that are logically possible but not physically possible (sometimes due to the limits of current technology). Thought experiments that are physically possible are quite common in both science and philosophy. These types of thought experiments usually illustrate problem cases, or hypothetical situations, and generally pose few problems in method, though they can certainly cause great disagreement over the results that they may suggest. An example of this type of thought experiment, in a philosophical form, would be Thomas Reid’s example of the brave officer1 (1975, pp. 114–115). Reid was concerned about Locke’s memory principle, and came up with a thought experiment to illustrate its problems. 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.
Applying Locke’s principle to the case, it can be seen that the officer is the same person as the boy, and the general the same person as the officer, but the general is not the same person as the boy. This experiment thus showed a failure of transitivity of identity, and prompted a revision of the memory principle. Clearly, such an example is physically possible, but equally clearly, it is an experiment that is unlikely to be actually performed. Nonetheless, it is a valid experiment; a physically possible philosophical thought experiment. While the conclusions of such experiments may cause heated argument, the actual method (generally) does not. A scientific equivalent to such an experiment would be something similar to Leibniz’s thoughts about Descartes’ laws of collision.2 Leibniz imagined collisions between balls that differed microscopically in size, and reasoned that the effects of such collisions would be very similar. Descartes’ theories however, suggested radically different results depending on whether the larger ball was moving or stationary before the collision. Thus Leibniz concluded that there
56
STEPHEN COLEMAN
must be some problem with Descartes laws of collision. Such an experiment is clearly physically possible, but Leibniz deemed it unnecessary to actually perform the experiment, since the results appeared to be obvious. The second type of thought experiment is more problematic. These are thought experiments that are logically possible, but physically impossible. To illustrate the potential problems of thought experiments of this type, let us examine two scientific thought experiments, one successful, the other unsuccessful. In the first experiment, Einstein, when contemplating Maxwell’s theory of electrodynamics, wondered what would be observed if one were able to travel in front of a beam of light. Einstein realised that this would result in observations of an impossibility, a stationary oscillating field. He therefore concluded that there must be some problem with Maxwell’s theory. This experiment could clearly not be performed, as an observer could not travel at the speed of light. Yet it is a logical possibility for an observer to travel at this speed, since there is no logical problem in suggesting that one beam of light observes another. The only change from the real world in the experiment is the assumption that an observer can travel at the speed of light. In 1930, Einstein imagined another thought experiment, but this one failed. Einstein imagined an experimental setup that involved a clock that was inside a box, suspended on a balance. The box had a small shuttered hole on one side, the shutter being controlled by the clock inside the box. Einstein imagined that if the shutter could be opened for a small enough time, then one could release a single photon of light from inside the box, and by means of the balance, establish the energy of that photon by using its weight and the famous formula E = MC2 . This would thus allow one to know both the energy of the photon and the time of its escape from the box, in contravention of the uncertainty principle, which states the time and energy are reciprocally indeterminate. However, this thought experiment failed to take into account the theory of general relativity, which unfortunately invalidates the entire experiment. When the photon escapes, it imparts an unknown momentum to the box. This would mean that the clock in the box is now moving in a gravitational field, so the general theory of relativity comes
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
57
into action, affecting the rate of the clock through the time dilation effect.3 These two experiments both appear to make small steps from the real world to a theoretical one, but in one of these thought experiments the step was valid, while in the other it was not. This clearly illustrates a possible problem for thought experimentation; that the step to the real world from the theoretical one may be illicit. But why was the step possible in one of these thought experiments, and not in the other? Wilkes suggests that the answer is in the way that the background conditions are laid out for each experiment. Whenever we are dealing with physically impossible thought experiments, there will be some part of the experiment that will be impossible. Wilkes suggests that the usefulness of the experiment depends on the centrality, or relevance, of the impossibility (1988, p. 9). In the thought experiment about Maxwell’s theory of electrodynamics, the impossibility of a human traveling at the speed of light is of no relevance to the theory that we are examining. In the clock in the box, relativistic effects are central to the experiment, and cannot be ignored. Wilkes’ objection to philosophical thought experiments then, is that in many experiments it is assumed that the step to the real world from a theoretical one is a small one, when this may not actually be the case. If the impossibility in the experiment is of no relevance, then the experiment can give us useful results. If the impossibility is central to the case, then any results we get will be tainted. Let us now focus our attention on the problem at hand, thought experimentation in the area of personal identity. Of all the vast number of thought experiments in the literature, there are some that are undoubtedly central to the entire debate. Virtually no book on personal identity would be complete without some discussion of two famous cases, those of ‘mind swapping’ and ‘tele-transportation’. Wilkes also makes specific criticisms of the method of thought experimentation used in these cases, and Williams’ particular objections are raised against the ‘mind-swap’ so it is upon these two cases that I will focus my discussion for the rest of this paper. In the ‘mind swapping’ thought experiment,4 it is suggested that two people, A and B are about to undergo an unusual experience. The thoughts, character traits and memories of person A will be
58
STEPHEN COLEMAN
extracted from his brain, and then his brain will be wiped clean, so to speak, so that all the information that was previously stored there will be erased. The same thing is simultaneously happening to person B. Then the memories, character traits and thoughts of person A will be transmitted into the brain of the body that was formerly B’s, and the memories, character traits and thoughts of person B will be transmitted into the brain of the body that was formerly A’s. Thus we now have an A body-person with the character and memories of B, and a B body-person with the character and memories of A. Before this swapping takes place, both A and B are told that one of them will be given a large reward (say a million dollars5) after the experiment, and the other will be tortured. Each is asked which body they would like rewarded, and which body they would like tortured. After the experiment, and the reward and torture, each is asked if he received his wish. The usual intuition of such an experiment, is that person A, before the experiment, would ask for the body A to be tortured after the experiment, and for body B to be rewarded. If this was to take place, then when questioned after the experiment, body-person B would say that he had got what he asked for. In other words, the common intuition is to say that after the experiment, person A exists in body-person-B, and person B exists in body-person-A. In the ‘tele-transportation’ experiment,6 you are asked to imagine yourself entering into a booth and pressing a large green button. The machine then will destroy your body and brain, while recording the exact states of all the cells in your body. This information is then transmitted to Mars, where a new body is created by another machine, from the information transmitted from earth. This entire situation is seen by the people who use it as a wonderfully efficient means of transportation. One day, I go to the tele-transporter, and press the button, but nothing appears to happen. I ask the attending staff about the problem, and they inform me that the new scanner records my blueprint without destroying my brain and body, but unfortunately, it has damaged my heart, and I will die in a few days. But my new body on Mars will be able to carry on with my life as if nothing had happened. This experiment tends to stretch intuitions in several different ways. In the case where the original body is destroyed, most people
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
59
are at least willing to entertain the possibility that the duplicate on Mars might be the same person. However, when the original is not destroyed, inevitably people’s intuitions suggest that the person on earth is the same person as before, but that the duplicate on Mars is a new person. Kathleen Wilkes suggests that both of the above thought experiments can be completely disregarded, as they contravene the rules of thought experiments that she has established, by including impossibilities that are central to the matter being discussed. For example, ‘mind swaps’ such as that suggested in the thought experiment above, are not a valid form of thought experiment for Wilkes, since she considers the mind to be too complex to be swapped around in this way. As she points out, the average brain contains something of the order of 1010 cells, each containing up to 2000 synaptic links, in a pattern that is unique to each individual person. Since an identical memory trace is recorded differently on two different brains, the difficulty of exchanging memories between two brains can clearly be understood. Also, since every memory trace affects every future memory trace, it would be necessary to download the entire contents of the brain in order to record a single memory. Wilkes thus concludes that it is not even theoretically possible to take a mature human brain, and map its contents onto another system (1988, p. 42). She further suggests that it is no more possible to copy a brain cell for cell and thus copy the brain as well as the contents as would be the case in the ‘teletransportation’ experiment, since she feels that the structures could not be copied at the sub-atomic level, which may well be necessary to capture all the information stored by the brain. Wilkes therefore concludes that both of these thought experiments are completely invalid, as the impossibilities involved are central to the experiment, and thus nothing at all can be gained from the study of these experiments. I would disagree with Wilkes conclusions on this point, mainly because I do not think that she has adequately proven that the experiments are physically impossible. Let us firstly examine the ‘mind swap’. Certainly the human brain is a vastly complex system, but if computer technology continues to develop at the same rate that is has over the last thirty years, then it certainly appears possible that
60
STEPHEN COLEMAN
computers will eventually be developed that mimic the human brain in complexity.7 If this was achieved, then it certainly seems to be at least logically possible that the information stored in the brain could be downloaded onto computer. From there it would seem to not be an enormous step for the information to be reloaded into another brain. Wilkes herself admits the possibility of a physical brain swap (1988, p. 37), though she points out that it would need to be a brain and spinal cord swap, since the spinal cord is centrally implicated in many psychological functions. In fact the possibility of this sort of brain swap cannot really be denied, since scientists in the USA have already performed the unpleasant experiment of transplanting monkey heads from one body to another (Wilkes, 1988, p. 37). As for the ‘tele-transportation’ experiment, it is pure guesswork on Wilkes’ part that suggests the human brain would have to be copied to the sub-atomic level or beyond. If it were only necessary to copy the brain to the atomic level, then this certainly seems to be physically possible, though admittedly well beyond our current technology. In The Physics of Star Trek, physicist Lawrence Krauss details exactly what technology we would require to build a Star Trek style transporter (1996, pp. 65–83). While this includes computers that are 1000 billion times more powerful than the ones we have now, and telescopes the size of the earth, the very fact that he is able to state the minimum technology requirements sufficient to build such a device seems to suggest that the device’s impossibility is merely a function of our current technology, rather than any actual physical impossibility.8 If Wilkes cannot prove that these types of thought experiments are even physically impossible, then it is obviously impossible for her to suggest that the impossibilities involved are relevant to the experiment. Ignoring brain and spinal cord swaps seems especially problematic, since this looks like a good candidate for physical possibility within just a few years. Thus her vehement criticisms of thought experiments in personal identity seem to be badly misplaced. What about the other problem with thought experiments in personal identity, that redescription of the experiment can lead to an opposing result? This is the subject of Bernard Williams’ paper ‘The Self and the Future’ (1973, pp. 46–63). In this paper, Williams
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
61
provides an account of the ‘mind swap’ experiment as it might be seen from the perspective of one of the participants. This new perspective appears to completely change the results of the thought experiment, so it is worth some concentrated examination. In the original ‘mind swap’ experiment, the participants are apparently willing to participate in the experiment, and apparently are fully informed of the procedure involved. Williams however, recasts the experiment to include an unwilling participant, about to be set upon by a mad scientist, and unaware of what will happen to him until informed by the scientist. In practical terms, what this means is that the subject of the experiment (who in line with Williams we will call person A) is simply told of all the terrible things that will happen to him, and not informed of the possible benefits (in other words, he is not informed that his memories and character will be transplanted into someone else, nor is he told that he may receive a million dollars). In Williams reconfiguration of the ‘mind swap’ experiment (somewhat paraphrased), person A is told that he will be tortured tomorrow. This prospect obviously fills him with dread. Then he is told that he has nothing to fear, for when the time comes for him to be tortured, he will not remember being told about it beforehand, nor indeed will he be able to remember anything that he knows now, he will have total amnesia. Williams states, correctly I believe, that person A would not find this statement reassuring. Then person A is told that when the time comes for him to be tortured, not only will he have total amnesia, but certain changes will be made to his character, and have new ‘memories’ as well, that correspond to those of another fictitious person. Williams suggests that person A would not be reassured by this either. Finally, person A is told that the memories and character traits which will be implanted into his brain will be copied from the mind of another person who is alive today, and that this will be accomplished by means of a machine to which both person A and the ‘memory donor’ will be connected. This still will give person A no reason to not fear the torture, Williams states. In order to clarify further discussion, Williams lists six possible things that person A could be told, and then examines these pieces of information given to person A to see if there is some significant step
62
STEPHEN COLEMAN
between two of them that would justify person A no longer fearing the torture. Thus he sets out these six possibilities as follows (1973, pp. 55–56): (i) A is subjected to an operation which produces total amnesia; (ii) amnesia is produced in A, and other interference leads to certain changes in his character; (iii) Changes in his character are produced, and at the same time certain illusory ‘memory’ beliefs are induced in him: these are of a quite fictitious kind and do not fit the life of any actual person; (iv) the same as (iii), except that both the character traits and the ‘memory’ impressions are designed to be appropriate to another actual person, B; (v) the same as (iv), except that the result is produced by putting the information into A from the brain of B, by a method which leaves B the same as he was before; (vi) the same happens to A as in (v), but B is not left the same, since a similar operation is conducted in the reverse direction. Williams concludes that A will still fear the torture when told of prospect (i), and suggests, reasonably I think, that everyone would agree with this. Williams agrees that situations (ii) and (iii) will give person A some cause for concern in addition to the torture that he faces, but suggests that they do not give him any reason whatsoever to feel reassured about his torture tomorrow. Situations (iv) and (v) are really identical with (iii) as far as person A is concerned, for they only supply extra detail about the source of the character and memory traits that will be implanted into A, so if his fear of the torture can reach through to (iii), then there is absolutely no reason why it should not reach through to (iv) and (v) as well. The logic of this step (from (iii) to (v)) certainly cannot be doubted. It is in situation (vi) that those philosophers who agree with psychological continuity theories really take issue with Williams, for it is at this point that A finds out about a potential positive in all of this, the fact that his memory and character traits will be preserved in B. Williams responds that the only way to really place any weight on the step from (v) to (vi) is to suggest that A will not fear the torture at point (v). This is easily demonstrated. If (vi) involves the re-introduction of A, then this must logically mean that
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
63
in situation (vi) A is now body-person-B. If A is body-person-B, then A is obviously not body-person-A. If A is not body-person-A in situation (vi) then this is obviously also true of (v), (iv) and (iii) as well (since there is no actual change to A in any of these steps, merely a change in where the implanted memories will be coming from). But in (v) there is no other candidate to be A, so it would seem that A does not at this time exist. If this is the case, then when did A cease to exist? In thinking back to the original thought experiment, none of these questions seemed to be important, all of them have been produced by Williams’ change of perspective. Not only are we no longer sure that A and B have ‘swapped minds’, or perhaps ‘swapped bodies’, we now no longer seem to be sure if A continues to exist at all. Or if A does continue to exist, then in which body does he continue to exist. Either body seems to be a possibility at this stage, as does the possibility that A actually ceases to exist at some stage in the process. Incredible as it may seem, all of these new questions were raised simply by a change of perspective, from the third person to the first person. If a change of perspective can produce such widely differing results in this thought experiment, then perhaps the same problem arises for all thought experiments. This would certainly seem to be a problem for thought experimentation. After all, it would seem to be a fundamental problem in the design of an experiment if the result of the experiment was totally dependent on the perspective of the observer. A good scientific experiment is thought to be totally independent of the location of the observer, and also infinitely repeatable. That certainly does not seem to be the case with these thought experiments. In fact, the situation is even worse than it first appeared, as we can get a different result again in the ‘mind swap’ experiment from another change of perspective. If one views the experiment from Williams’ first person perspective, then it would appear that person A has every right to fear the torture which he faces tomorrow, despite all the changes he will undergo before the torture takes place. But if we take the same formulation of the experiment, and look at it from the perspective of person B, then it does appear that we can achieve a different result. Person B is told that he will be receiving a million dollars tomorrow,
64
STEPHEN COLEMAN
and is naturally quite happy about that fact. He is then told that when he receives the million dollars, he will not remember being told that he would receive it, and his happiness would probably not dim at all. Then he is told that he will in fact not remember anything that he remembers now, for he will undergo a procedure which causes complete amnesia. I would suggest that it is at this point that person B would start to become somewhat concerned. Next he is told that in addition to losing his memory of everything he now knows, he will gain memory impressions that have no connection with his current life, and that his character will be altered to match those memories. Lastly he is informed that the new memories and character traits that he will receive will come from another person A, and that this will be accomplished by means of a machine to which both person B and the ‘memory donor’ person A will be connected. After being informed of all of this, I would suggest that person B would probably be rather unhappy about all that he was going to face, and that the prospect of something positive out of all of this, the million dollars that he will receive tomorrow, would seem small compensation. Why is this situation so different from that of person A? Person A’s fear of torture appeared to be able to extend through all the changes that he faced, yet person B’s joy does not seem to have the same resolve. Perhaps the answer to this question is that person A does not really fear the torture tomorrow, but rather fears all of the changes that he will undergo beforehand, changes which perhaps could be characterised as psychological torture. If it was this torture that person A feared, rather than the physical torture that he had been told he would undergo, then this would seem to explain why person B does not look forward to the million dollars that he will receive tomorrow, for he too will be psychologically tortured beforehand. At what point then can we say that person A stops fearing the physical torture, and starts to fear the psychological torture instead? I would suggest that this occurs somewhere between situation (ii) and (iii), and that even in situation (i) person A will probably fear the psychological torture as well as the physical torture. This would relate well to the situation of person B, who would probably start to feel less happy about the money he will receive when informed of situation (i), and by the time of situation (iii) will almost certainly be
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
65
feeling fearful of the psychological torture that he faces. Yet there is one difference for person B, he can at least look forward to some compensation for his torture, as he will receive a million dollars tomorrow. But perhaps he would not see this as compensation, for perhaps person B would not feel that it would be him that was receiving the money. On the other hand, some people may feel that the money is compensation, and feel that despite the psychological torture at least there is some glimmer of hope in the money they will receive tomorrow. Perhaps we can get some idea of whether or not people would regard the million dollars as compensation for person B by examining their attitudes to death. If they are the type of person who regards any kind of life as valuable, and would rather be kept alive by life support machines even if there was no chance of them ever regaining consciousness, then they would probably think that the million dollars would be some compensation for B. If on the other hand, they were the type of person who would rather have the life support machines turned off if there was no chance of them regaining consciousness, then they would probably not regard the million dollars as compensation. These types of thought experiments are designed to draw out one’s intuitions to reach a specific conclusion. In light of the fact that different people appear to have different intuitions about this thought experiment, it would appear that Williams’ conclusion (that our intuitions support a physical continuity theory of personal identity) is wrong. However, the original conclusion of the experiment reached by Locke, Shoemaker and Parfit (the psychological continuity answer) also appears to be wrong. In fact, there seems to be something fundamentally wrong with the thought experiment, since we have no consistent result. How is this possible? One possible reason is that Wilkes is correct, and that we can get no consistent results from the experiment because the situation described in the thought experiment is impossible in a way relevant to the results of the experiment. However, there is another possibility; that another theory might explain the results of the experiment in a consistent way. One possible candidate for this theory would be Robert Nozick’s Closest-Continuer schema (1981, pp. 29–70). This theory suggests
66
STEPHEN COLEMAN
that in all problem cases in personal identity, we assume that identity continues through the closest continuer of the original. Thus in the ‘tele-transporter’ case for example, when the original person on earth is destroyed, then the duplicate on Mars is the closest continuer, and so we would say that the person on Mars continues the existence of the original person on Earth. In the case when the original is not destroyed however, we would say that the duplicate on Mars is simply a duplicate, and the original is still on Earth. How does this theory cope with the ‘mind swap’ cases that we have described? This will obviously depend upon the particular content we give the schema. Nozick gives us the schema, but he specifically states that the content must be supplied by the individual’s particular values. Thus if we think that psychological continuity is the most important, then we will give this feature the strongest weighting in the schema; if we consider physical continuity to be most important, then we will weight this more heavily. In order to bring the schema into line with our intuitions in the Williams cases, we seem to require precedence of psychological continuity, a sort of ‘mind over matter’ situation.9 Thus in the original ‘mind swap’ case, this version of the schema gels with our intuition that person A now exists within body-person B, and person B now exists in body-person A. In Williams’ redescription of the case, when we have no other candidate as continuer, we are happy (or perhaps in this case unhappy) to say that the body-person A is the closest continuer of person A, and thus we fear the torture that this body will be subjected to tomorrow. However, there does seem to be a problem in the second half of the redescribed case, when we examine person B. If we truly follow the closest-continuer theory, then we would be forced to say that person B does get the million dollars, and should be happy about the prospect of receiving the money tomorrow. But this does not really seem to be the case. Even though we have no other candidate as continuer of person B, I think that most people would agree that person B will not look forward to receiving this money with great excitement, but will rather dread the prospect of the sun rising in the morning. If this is really the case, then there seems to be something missing in the Closest-Continuer account, and so we will need to look for another explanation of the results of the experiment.
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
67
Fortunately, I think there is an answer to this conundrum, and it lies in the original solution to the experiment. In the first version of the experiment, we are quite happy to say that the minds have swapped bodies, and this is in line with a psychological continuity theory of personal identity. In Williams’ redescription, we seem to get the result that physical continuity is what is important. Yet I think that examination of the situation of person B reveals that psychological continuity is really the right answer after all. If person A does not actually fear the physical torture that awaits him tomorrow, but rather the psychological torture that will occur beforehand, then this would suggest that it really is psychological continuity that is at work here. The case of person B seems to reveal this even more clearly. He doesn’t look forward to receiving a million dollars tomorrow, because he won’t receive a million dollars tomorrow. What he faces is not reward, but rather psychological torture, followed by death. The fear of person A is fear of torture; but not the physical torture, rather the same mental torture that awaits B. So after a long journey, we return to where we started, with a psychological account of personal identity. Williams’ redescription of the case has muddied the water, but the thought experiment in the end seems to reveal the same result as it originally did. If this is really the case, then redescription does not seem to be the major problem for thought experimentation that it originally appeared, but instead seems to actually confirm the method. Despite changing the perspective, we arrive at the same result, which would appear to make criticism of the method of thought experimentation unfounded.
ACKNOWLEDGMENT
I am grateful to all who offered comments and suggestions on this paper, most especially to Richard Hanley and Steve Matthews at Monash University, and Catriona Mackenzie and Caroline West at Macquarie University.
68
STEPHEN COLEMAN
NOTES 1
This example was suggested by one of the journal’s anonymous referees. See Sorensen (1992 p. 10) for a discussion of this thought experiment. 3 For further discussion see Niels Bohr (1959, pp. 199–241). 4 This experiment has been discussed in various forms for centuries. John Locke proposed a form of it (1975, p. 340) in which the soul of a prince and the soul of a cobbler exchange bodies. Sydney Shoemaker takes a more technically sophisticated line in his formulation of the experiment (1963, pp. 22–28). In his version, the brains of two persons are transplanted between bodies. Derek Parfit also discusses the case, and Williams response to it (1984, pp. 204–243) All of these formulations conclude that psychological continuity is the basic criterion for personal identity. 5 Bernard Williams suggested $100,000 in 1973, but given inflation, I would think that perhaps a million dollars would be more appropriate these days. 6 This experiment was suggested by Derek Parfit (1984, pp. 199–200). 7 An indication of the rapid advance in computer technology is given by the fact that I am typing this paper on a computer with more calculating power and complexity than was possessed by the entire computer system that piloted man to the moon twenty five years ago. 8 Krauss does think it would be impossible to build a transporter that would copy to the sub-atomic level, since the Heisenberg uncertainty principle would then become impossible to ignore. This is not a problem at the atomic level however. 9 I owe this suggestion to Richard Hanley. For a discussion of this idea see Hanley (1997, pp. 135–149). 2
REFERENCES Bohr, N. (1959): ‘Discussion with Albert Einstein on Epistemological Problems in Atomic Physics’, in P.A. Schilpp (ed.), Albert Einstein: PhilosopherScientist, New York: Harper Torchbooks, pp. 199–241. Hanley, R. (1997): The Metaphysics of Star Trek, New York: Basic Books. Johnston, M. (1987): ‘Human Beings’, The Journal of Philosophy 84, 59–83. Krauss, L.M. (1996): The Physics of Star Trek, New York, Harper Perennial. Locke, J. (1975): An Essay Concerning Human Understanding, Oxford: Clarendon (First published in 1689). Nozick, R. (1981): Philosophical Explanations, Cambridge, MA: Belknap. Parfit, D. (1984): Reasons and Persons, Oxford: Oxford University. Reid, T. (1975): ‘Of Mr Locke’s Account of Our Personal Identity’, (first published in 1785) reprinted in Perry, J (ed) (1975) Personal Identity, Berkeley, University of California, pp. 113–118. Shoemaker, S. (1963): Self-Knowledge and Self-Identity, Ithaca: Cornell University. Sorensen, R. (1992): Thought Experiments, Oxford: Oxford University.
THOUGHT EXPERIMENTS AND PERSONAL IDENTITY
69
Wilkes, K. (1988): Real People: Personal Identity without Thought Experiments, Oxford, Clarendon. Williams, B. (1970): ‘The Self and the Future’, reprinted in Williams, B (1973) Problems of the Self, Cambridge: Cambridge University, pp. 46–63.
ARC Special Research Centre for Applied Philosophy and Public Ethics Charles Sturt University Australia E-mail:
[email protected] Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
