IDENTITY,RELATION R,AND WHAT MATTERS

This paper offers a challenge to Derek Parfit's thesis that one ought to have no preference between these two otherwise identical situations: 1. I continue to go on living as before, and 2. I do not survive, but am replaced by a duplicate, psychologically continuous to my present self (i.e. an R-related duplicate). I point out that virtually all psychologically normal persons regard some inanimate objects as being ‘irreplaceable’ (such that no copy could adequately substitute). I then propose that in every such case, this judgment is based on the subject's regarding some person as likewise irreplaceable. It would follow that any such subject would have to reject Parfit's thesis.