Contribution
The paper sets two arguments about the same case against each other and shows that each carries conviction. The first is the familiar third-person case for the psychological view, made sharper by tying it to what A and B would choose and how they would react. The second is new: a first-person series that reaches the same outcome by small steps, at none of which the subject’s fear could rationally stop. It then argues that the standard escapes, “borderline case” and “decide by convention”, cannot be lived with from the first person.
Key insight
From the inside, a prediction of my future pain survives any prediction about what will be in my mind beforehand. Amnesia, a new character, someone else’s memories: none stops it being me who is hurt. The body-swap story looks like “A moves to the other body” only because the experimenter chose to stop at the neatest point of a series, and the step that makes it neat is something that happens to the other man.
Assumptions
- Memory is causal. The memories must be suitably caused by the past experiences; the imagined process copies information out of one brain and into the other (“information-parking”, p. 162).
- Imaginability. A and B are taken to be alike enough for the exchange to be imaginable “in the most restricted sense” (p. 161).
- Selfish choice. A and B choose “on selfish grounds” (p. 163).
- Pain as test. Physical pain is “absolutely minimally dependent on character or belief”, so predicted psychological change cannot remove the reason to fear it (p. 170).
- Fear needs involvement. Some emotions, fear among them, are felt only if I will be “involved” in the expected situation (p. 175).
Key results
- The first presentation (pp. 163–167).
- With the choice of money or torture for the A-body-person and the B-body-person, every pattern of choices yields reactions afterwards that fit “changing bodies”: the B-body-person complains that he chose otherwise, “since he has A’s memories” (p. 164).
- “This seems to show that to care about what happens to me in the future is not necessarily to care about what happens to this body” (p. 164), and that in some sense “I and my body are ‘really distinct’”, though nothing supports existing without a body (pp. 164–165).
- The wooden leg, the anxiety and the distressing memories confirm it (pp. 166–167). Arguments that bodily continuity is necessary “would seem to be just mistaken” (p. 167).
- The second presentation (pp. 167–171).
- Torture announced; then forgetting the announcement; then total amnesia; then a different set of impressions of my past; then impressions matching another living person, copied from his brain. “Fear, surely, would still be the proper reaction” (p. 168).
- The principle behind it: “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). Contrast acrophobia, where predicted change does remove the reason (p. 169).
- The presentation differs from the first in using “you” and in leaving out the other man. Neither seems a fault: “My selfish concern is to be told what is going to happen to me” (p. 171).
- The series (pp. 172–174). (i) amnesia; (ii) plus character change;
(iii) plus fictitious “memories”; (iv) character and “memories”
appropriate to a real person B; (v) produced by copying from B’s brain,
B left unchanged; (vi) the same, with the reverse operation done on B.
- No step from (i) to (v) gives A a reason to stop fearing.
- “From the point of view of A’s expectations and fears, there is less difference between (vi) and (v) than there is between (v) and (iv)” (p. 173).
- To say A reappears as the B-body-person in (vi) implies that A “just does not exist” in (v), since the A-body-person there is “in character, history, everything, exactly the same” (p. 174). Where then to draw the line, between (iii) and (iv)?
- Against indeterminacy (pp. 174–178).
- Ordinary indeterminate expectations (one of five will be hurt; a nameless horror) keep fear in place, because “one thing that is fixed amid the indeterminacy is the belief that it is to me” that it will happen (p. 176).
- Conceptual undecidability whether I am involved leaves fear “neither appropriate, nor inappropriate, nor appropriately equivocal” (pp. 176–177). Projective imagining “implicitly answers the necessarily unanswerable question” (p. 177).
- The analogy of ambivalent feeling toward a transformed sentimental object fails, because it keeps “him and me clearly distinct” (p. 177). “There seems to be an obstinate bafflement to mirroring in my expectations a situation in which it is conceptually undecidable whether I occur” (p. 178).
- Against conventionalism (p. 178). Calling the A-body-person “A” in (v) but not in (vi) is “the sort of thing indeed appropriate to lawyers”, but “as a line to deal with a person’s fears or expectations about his own future, it seems to have no sense at all”.
- The reversal (p. 179). The “mentalistic” conclusion came from an argument “entirely conducted in third-personal terms”; the bodily one from “the first-personal issue of what A could expect”.
- The resolution (pp. 179–180). The first presentation’s neatness “is basically artificial”; the experimenter “has rather produced the one situation out of a range of equally possible situations which we should be most disposed to call a change of bodies”. The principle that fears reach through psychological change “seems positively straightforward … but we need to be shown what is wrong with it”. So pass the pain to the B-body-person: “It would be risky: that there is room for the notion of a risk here is itself a major feature of the problem.”
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | The third-person telling supports describing the outcome as an exchange of bodies | moderate, as an account of what we would say | the three choice patterns and post-experiment reactions (pp. 163–167) |
| C2 | Predicted psychological change does not remove the rational ground for fearing predicted pain | moderate | the torture series and the stated principle (pp. 167–170); contrasted with acrophobia |
| C3 | The step that completes the exchange makes no difference to what A should expect | strong given C2 | (v)–(vi) differ only in what happens to B (pp. 173–174) |
| C4 | If A is the B-body-person in (vi), A does not exist in (v) | strong, as a consequence of the intrinsic sameness of the two A-body-persons | p. 174 |
| C5 | A future situation in which it is conceptually undecidable whether I am involved cannot be represented in my expectations and emotions | moderate | analysis of indeterminate expectation (pp. 175–178) |
| C6 | Conventional decision cannot guide a person’s fears about his own future | moderate | the lawyers contrast (p. 178) |
| C7 | The first-person view favours bodily continuity and the third-person view psychological continuity | illustrated | p. 179; “I will not discuss it further here” |
| C8 | A, choosing selfishly, should pass the pain to the B-body-person | tentative, by the author’s own word (“rather shaky”, “risky”) | pp. 179–180 |
Concepts
- A-body-person / B-body-person — whoever one is confronted with, after the experiment, when confronted with the body that was A’s / B’s (p. 163).
- information-parking — extracting information from a brain into a device and replacing it; not “learning again” (p. 162).
- involvement — figuring in a future situation as doer or sufferer, or being affected by its consequences (p. 175).
- projective imagination — thinking of a future situation “as it will be for me” (p. 176).
Connections
- Parfit’s “Personal Identity” (filed beside this reading) cites it in footnote 11 as showing how natural the beliefs are that the question must have an answer, that the answer is all-or-nothing, and that there is a “risk” of getting it wrong. Parfit proposes to undermine them through their causes. Williams’s C5 is the claim that this undermining cannot be lived from the first person.
- Lewis’s “Survival and Identity” (filed beside it) does not mention this paper. Its tensed identity gives no help with C3, since the case has no branching: there is one A-body-person in each of (v) and (vi).
- Birch (LIT-111, NOTE-183). The no-scan teletransporter is case (iv). See the LIT entry.
- Dennett (LIT-405, NOTE-338). Dennett’s narrator follows his point of view; Williams’s subject has his point of view rewritten and is still the one in pain. Neither cites the other.
- Locke (filed beside it). The prince and cobbler is the first presentation’s case. Locke’s verdict is the first presentation’s.
Bearing on the record
No THEORY document in the record cites it. It is the main source the record now holds for the view that what a subject should expect follows the body, and so bears on any future THEORY that treats copying or re-instantiating an agent as its persistence. It carries no instruction for machine-learning practice.
Note of 2026-10-16: THEORY-531 now cites LIT-462 as a source.
Limitations
What is shown and what is asserted:
- Shown, given C2: C3 and C4. The (v)–(vi) argument is the paper’s strongest step.
- Argued from cases, contestable: C2. The principle is defended by pain’s independence from character and belief, not by an account of what makes a future pain mine. Parfit’s later reply targets exactly this.
- Analysed, not proved: C5. It depends on fear requiring a determinate belief that it is to me that something will happen.
- Explicitly tentative: C8. Williams says he is “not in the least clear which option it would be wise to take”, and that he finds that “rather disturbing” (p. 179).
- Not addressed: fission. In the series there is never more than one candidate at a time on A’s side.
Open questions
- What makes a future pain one’s own, if not psychological continuity? The paper argues that psychological change does not stop it being mine, and does not say what does, beyond the body that “has had the leg on it for some time”.
- Would the first-person argument survive a version in which the body too is gradually replaced? The paper keeps the body fixed throughout.
Corrections
None. The brief’s citation is correct in every detail.