Abstract
Do physical influences occasionally travel from future to past? How should this be tested experimentally? (We restrict ourselves to time-travel of the second kind.) Partridge-type experiments, based on absorber theory, lack a clear interpretation since the Wheeler-Feynman theory is circular, and even alternative absorber theories involve complex hypotheses. (Mere verbal reasoning about time is paradox-prone since it simultaneously uses the notion of time already in the tense-structure of language.) Hence, the desirable theoretical starting point is a mathematical model obtained by dropping just the one hypothesis of ``causality''. Functional differential equations (FDEs) are a necessary though neglected consequence of classical electrodynamics and relativity (without any additional hypotheses). Dropping the hypothesis of ``causality'', or admitting a ``tilt'', makes these FDEs of mixed-type---this is the required mathematical model. (These equations imply phenomena such as quantum interference regarded as unique to quantum mechanics, though we do not discuss that here.) Retarded FDEs cleanly resolve the classical recurrence and reversibility paradoxes of thermodynamics, and explain (fine-grained) entropy increase. Mixed-type FDEs imply also occasional spontaneous decrease of entropy. It is mathematically impossible to replicate this mechanically---thus avoiding perpetual motion machines. On mundane observation, living organisms do exhibit occasional non-mechanical behaviour. The value of experimentation, itself, rests on this belief. We conclude that a tilt is consistent with empirical observations, and provides a non-mechanistic physics better suited to model life. This conclusion may be further tested by applying this model to biological macromolecules.
C. K. Raju