Coming to terms with decoherence

In response to Charles Day’s item on the solution to Hund’s paradox (Physics Today, September 2009, page 16 ), Robert Harris and Leo Stodolsky write that their work ¹ from the early 1980s could not yet refer to decoherence because the term was created more than five years later (Physics Today, February 2010, page 10 ). However, Day’s formulation that their results—which we used and cited in our 1985 paper ² when the term decoherence still did not exist—were not yet couched in “the then-nascent decoherence theory” referred not to the name but to the concept. Harris and Stodolsky’s insistence that their terms “quantum damping” and “tunneling friction” are just as appropriate indicates that they adhere to a widespread misunderstanding of the concept of decoherence.

Decoherence and dissipation are described by different terms in the master equation, with decoherence usually acting on a far shorter time scale than dissipation. Although the explicit expression obtained by Harris and Stodolsky does contain decoherence terms, we do not remember that they ever mentioned the importance of those terms for the quantum-to-classical transition, superselection rules, or the measurement problem, connections one of us (Zeh) first discussed conceptually during the 1970s. ³ The term “decoherence” does not describe any fundamentally new physics; in a way the idea is already present in Nevill Mott’s 1929 analysis of alpha-particle tracks. ⁴ The point is that this consequence of the entanglement that unavoidably arises between all systems as they interact had been overlooked since the discovery of quantum mechanics, particularly when physicists tried unsuccessfully to recover classical mechanics for isolated systems in the limit in which Planck’s constant h vanishes. Hund’s paradox was not concerned with the lifetimes of certain chiral states but with the absence of their superpositions. It may appear from the master equation that superpositions are destroyed by decoherence, but that is not so; they are only irreversibly “dislocalized.”