Conditions for Lorentz-invariant superluminal information transfer without signaling

TL;DR

This paper explores conditions under which superluminal information transfer can occur without violating Lorentz invariance or enabling signaling, challenging traditional notions of realism and causality in quantum mechanics.

Contribution

It demonstrates that superluminal information transfer can be compatible with Lorentz invariance without enabling signaling, by considering contextuality and the impossibility of perfect state reproduction.

Findings

Superluminal transfer does not necessarily imply signaling.

Contextuality of experiments affects event ordering in relativistic settings.

No-cloning and preparation reproducibility prevent superluminal signaling.

Abstract

We understand emergent quantum mechanics in the sense that quantum mechanics describes processes of physical emergence relating an assumed sub-quantum physics to macroscopic boundary conditions. The latter can be shown to entail top-down causation, in addition to usual bottom-up scenarios. With this example it is demonstrated that definitions of "realism" in the literature are simply too restrictive. A prevailing manner to define realism in quantum mechanics is in terms of pre-determination independent of the measurement. With our counter-example, which actually is ubiquitous in emergent, or self-organizing, systems, we argue for realism without pre-determination. We refer to earlier results of our group showing how the guiding equation of the de Broglie-Bohm interpretation can be derived from a theory with classical ingredients only. Essentially, this corresponds to a "quantum mechanics without wave functions" in ordinary 3-space, albeit with nonlocal correlations. This, then, leads to the central question of how to deal with the nonlocality problem in a relativistic setting. We here show that a basic argument discussing the allegedly paradox time ordering of events in EPR-type two-particle experiments falls short of taking into account the contextuality of the experimental setup. Consequently, we then discuss under which circumstances (i.e. physical premises) superluminal information transfer (but not signaling) may be compatible with a Lorentz-invariant theory. Finally, we argue that the impossibility of superluminal signaling - despite the presence of superluminal information transfer - is not the result of some sort of conspiracy (a la "Nature likes to hide"), but the consequence of the impossibility to exactly reproduce in repeated experimental runs a state's preparation, or of the no-cloning theorem, respectively.