Is the usual notion of time evolution adequate for quantum-mechanical systems? I

Abstract

Circumstances are described wherein no state at a given time (nor any definite evolution from one time to another) can be ascribed to a given physical system, but wherein the system can nonetheless be associated with definite dispersion-free values of a new sort of observable, which we call a "multiple-time" observable. The description of physical systems in terms of these new observables is discussed. It emerges as a by-product of our work that no experiment whatever (albeit that its result is certain) can be carried out on a system without disturbing the values of other measurable quantities.