ABSTRACT
Modern technology has made it uncertain as to when exactly death occurs, and this has put us in a quandary over when we can initiate behaviour traditionally deemed apt if and only if a patient is dead. In the light of this, there is general agreement that death should be redefined, but wide disagreement remains about how. I argue, against this, that it is a mistake to redefine death in any way: (1) redefining death will not help to settle the question of when traditional death-behaviour becomes appropriate, (2) any attempt to redefine death is attended with significant disutilities and no compensating utilities, and (3) the practical problems generated by the indeterminacy can be better handled in other ways.
