We discuss some features of the decoherent histories approach. We consider
four assumptions, the first three being in our opinion necessary for a soun
d interpretation of the theory, while the fourth one is accepted by the sup
porters of the DH approach, and we prove that they lead to a logical contra
diction. We discuss the consequences of relaxing any one of them. (C) 1999
Published by Elsevier Science B.V. All rights reserved.