Source
Williams (1970), “The Self and the Future”, The Philosophical Review 79(2): 161–180, LIT-462, read in full in NOTE-387.
What was actually shown
The third-person telling (pp. 163–167). A and B are told that one person emerging from the process will get money and the other torture, and they choose selfishly. Whatever they choose, the reactions afterwards (complaints, satisfaction, recalled choices) make “a strong case for saying that the B-body-person really is A, and the A-body-person really is B” (p. 164). So “to care about what happens to me in the future is not necessarily to care about what happens to this body” (p. 164).
The first-person telling (pp. 167–171). I am told I will be tortured; then that I will first forget being told; then that I will have total amnesia; then a new character; then “memories” matching another living person, copied from his brain. “Fear, surely, would still be the proper reaction” (p. 168). The principle: “my undergoing physical pain in the future is not excluded by any psychological state I may be in at the time, with the platitudinous exception of … unconsciousness” (p. 169). Fear of heights is the contrast: there, predicted change does remove the reason (p. 169). Pain is chosen because it is “absolutely minimally dependent on character or belief” (p. 170).
The series (pp. 172–174). (i) amnesia; (ii) plus a new character; (iii) plus fictitious “memories”; (iv) character and “memories” fitting a real person B; (v) produced by copying from B’s brain, B left unchanged; (vi) the same, with the reverse done to B. No step from (i) to (v) gives A a reason to stop fearing. Step (vi) “is absolutely no difference at all in what happens to him, the only difference being in what happens to someone else” (p. 174). If A is the B-body-person in (vi), then A “just does not exist” in (v), though the A-body-person there is the same “in character, history, everything” (p. 174).
Against the escapes (pp. 175–178). Ordinary indeterminate expectations keep fear in place because “it is to me” that something will happen (p. 176). Where it is conceptually undecidable whether I am involved, fear “seems neither appropriate, nor inappropriate, nor appropriately equivocal” (pp. 176–177). A conventional decision, “appropriate to lawyers”, “as a line to deal with a person’s fears or expectations about his own future … seems to have no sense at all” (p. 178).
The reversal and the resolution (pp. 179–180). The mentalistic conclusion came from a third-person argument and the bodily one from the first person (p. 179). The neat exchange is “basically artificial”, the one situation of a range “we should be most disposed to call a change of bodies”. So A, choosing selfishly, should pass the pain to the B-body-person: “It would be risky”.
What could have come out otherwise. Some step in the series could have been one at which fear plainly stops being apt. On the paper’s analysis, none is, and the last step concerns only B.
What this does not say
- What does make a future pain mine. The paper argues that psychological change does not stop it being mine. It does not say what does, beyond the body that “has had the leg on it for some time”. It offers no criterion of identity.
- That the body is the person. The first-person argument leans to the body; the conclusion is a choice under risk. Williams is “not in the least clear which option it would be wise to take” (p. 179).
- That the principle about pain is established. It is defended by cases and by pain’s independence from character, not by an account of ownership. Parfit’s reply targets exactly this.
- Anything about fission. There is never more than one candidate at a time on A’s side, so Lewis’s tensed identity has nothing to do here.
- That the series survives a gradually replaced body. The body is held fixed throughout.
- That ordinary people’s verdicts follow the person of the telling. Williams observes that the mentalistic conclusion came from a third-person argument and the bodily one from a first-person argument (p. 179). That is a remark about his own two arguments. Among Arizona undergraduates, Nichols & Bruno (2010, LIT-1269, NOTE-996) found no first- against third-person difference. In a simplified Pain frame, 68% (N = 50) and 72% (N = 54) agreed that the subject would feel the later pain, χ²(1, N = 64) = 0.062. In a brain-transplant case the differences were t(70) = 0.69 and 1.09. These are small samples with no equivalence test.
- That the third-person verdict is only “the will of the experimenter”. In the same survey, memory still mattered in a case where the memories are erased and go to no one. Williams’s explanation concerned a full exchange, which was not tested, so this undercuts it only weakly. None of this bears on the claim itself: the survey did not ask what anyone would fear, and the promote_when already says such reports cannot settle it.
Connections
- Rival. Parfit (THEORY-447) cites the paper in fn. 11 for the beliefs that “shall I survive?” must have an answer, all-or-nothing, with a risk of error, and proposes to undermine them through their causes. The claim here that indeterminacy cannot be lived with from the first person is what that undermining has to overcome.
- Purpose-relative identity. Floridi (THEORY-650) makes sameness of a self relative to a purpose-fixed level of abstraction. The argument against conventional decision here is the objection such a view must answer for a person’s expectations about his own future.
- Re-instantiation. Birch’s teletransporter “with a twist” (LIT-111, NOTE-183) recreates states without scanning, which is case (iv). Birch makes the causal link decide who persists; Williams doubts that adding it in (v) changes what the subject should expect. The comparison is the filing reader’s (LIT-462).
- Dennett’s narrator (LIT-405) follows his point of view and leaves his body. Williams’s subject has his point of view rewritten and is still the one in pain.
- Folk verdicts. Nichols & Bruno (2010, LIT-1269) put simplified versions of both frames to students. The Pain frame’s bodily verdict is the majority in both persons. The authors favour psychological criteria on reflection, but only conditionally, if intuitions inform the theory.
- Locke’s prince and cobbler (THEORY-546) is the first telling’s case, with the first telling’s verdict.