INACCURATE EXPERIMENTS AND ENVIRONMENT-INDUCED SUPERSELECTION RULES

It is argued that approximate superselection rules induced by the environment cannot account for the emergence of definite measurement results in single experiments. The reason for this is that the inaccuracy necessary for an experiment failing to distinguish between exact and approximate mixtures requires the pointer observable to be strictly classical. This is shown in the case that the observables of physical systems generate W*-algebras.