Relational Dynamics with Periodic Clocks

TL;DR

This paper develops a systematic framework for relational dynamics using periodic clocks in classical and quantum theories, revealing invariance conditions, equivalences between different quantum pictures, and limitations of existing probability definitions.

Contribution

It introduces a unified approach to relational dynamics with periodic clocks, establishing equivalences between quantum formalisms and addressing issues in defining probabilities.

Findings

Relational observables are invariant only if the quantity is periodic.

Quantum relational observables can be derived via partial group averaging.

The three quantum pictures are equivalent and imply periodic dynamics.

Abstract

We discuss a systematic way in which a relational dynamics can be established relative to periodic clocks both in the classical and quantum theories, emphasising the parallels between them. We show that: (1) classical and quantum relational observables that encode the value of a quantity relative to a periodic clock are only invariant along the gauge orbits generated by the Hamiltonian constraint if the quantity itself is periodic, and otherwise the observables are only transiently invariant per clock cycle (this implies, in particular, that counting winding numbers does not lead to invariant observables relative to the periodic clock); (2) the quantum relational observables can be obtained from a partial group averaging procedure over a single clock cycle; (3) there is an equivalence ('trinity') between the quantum theories based on the quantum relational observables of the clock-neutral picture of Dirac quantisation, the relational Schr\"odinger picture of the Page-Wootters formalism, and the relational Heisenberg picture that follows from quantum deparametrisation, all three taken relative to periodic clocks (implying that the dynamics in all three is necessarily periodic); (4) in the context of periodic clocks, the original Page-Wootters definition of conditional probabilities fails for systems that have a continuous energy spectrum and, using the equivalence between the Page-Wootters and the clock-neutral, gauge-invariant formalism, must be suitably updated. Finally, we show how a system evolving periodically with respect to a periodic clock can evolve monotonically with respect to an aperiodic clock, without inconsistency. The presentation is illustrated by several examples, and we conclude with a brief comparison to other approaches in the literature that also deal with relational descriptions of periodic clocks.