… we could not find out what happens even if we could actually perform this [experiment]. Suppose, for example, that I do survive as one of the resulting people. I would believe that I have survived. But I would know that the other resulting person falsely believes that he is me, and that he survived. Since I would know this, I could not trust my own belief. I might be the resulting person with the false belief. And, since we would both claim to be me, other people would have no reason to believe one of us rather than the other. Even if we performed this [experiment], we would therefore learn nothing.In this case, however, I suspect that Parfit is at best half right. Suppose that you do find yourself waking up on Mars. Parfit says that you would be unable to tell whether you were a freshly-minted person or the same person as the one who entered the booth. This claim still seems fair enough to me. But it also strikes me that you would be able to find out something else. You would be able to find out in which of the two bodies you have woken up. This would be true even if you could not answer the previous question. And what you discover here would not be negligible. E.g., if exactly one of the two bodies was shortly going to be in pain, then it would matter to you in which body you had woken up. Remarkably, Parfit says nothing about this. If he thinks that you could not find out even in which body you have awakened, as his remarks occasionally suggest, then I think that he is wrong. In this case, if you do wake up on Mars, then there is something that you could find out, even if Parfit is right that there is something else that you couldn’t.
Whatever happened to me, we could not discover what happened … (Reasons and Persons, p. 258)