LCTRLocal-comparison theory of time reference

Comment papers

The first two papers examine whether a statistical state determines a time measurable by a clock. The third asks what a boundary entanglement calculation establishes about time in de Sitter space.

Target article · C. Rovelli, Foundations of Physics 41, 1475–1490 (2011)

Comment on “Forget time”: completeness of thermal flows and correspondence with physical clocks

Two counterexamples show why specifying a thermal-flow equation from a statistical state does not establish the existence of time measurable by a physical clock.

In the first example, for every nonzero parameter the thermal flow is undefined on a set of initial points of positive probability. Rovelli’s Eq. (18) therefore leaves a correlation undefined even for bounded smooth observables. In the second example the flow is complete, yet no change of clock that preserves the original motions, conserved quantity and ensemble reproduces its thermal correlation.

Related LCTR reading

Target article · C. Rovelli, Physical Review D 87, 084055 (2013)

Comment on “General relativistic statistical mechanics”

Even when the local Gibbs relation holds, the statistical state need not determine the stationary evolution of the ensemble required to infer equilibrium time.

The counterexample has a canonical family with finite partition functions, mean energies and energy variances at every positive inverse temperature. Yet at every nonzero parameter the domain of the thermal flow has probability less than one. Periodic and reflecting boundary rules both preserve the original distribution but predict different correlations: the state and local equations do not select between them.

Related LCTR reading

Target article · K. Fujiki et al., Journal of High Energy Physics 04 (2026) 072

Comment on “Entropic interpretation of Einstein equation in dS/CFT”

Agreement between the boundary first law of pseudo-entropy and the gravity calculation tests a de Sitter spacetime whose time coordinate is already specified; it does not derive that time.

An explicit perturbation violates Eq. (4.12) in Fujiki and colleagues’ comparison with Euclidean AdS with a spherical boundary. Correcting that equation leaves a separate issue: the width of a boundary interval is a timelike coordinate in an auxiliary space, whereas the bulk calculation already uses physical bulk time in specifying the metric and the future boundary.

Related LCTR reading