The asymmetric nature of time, the radical difference in nature between past and future, has often been taken to be the core feature distinguishing time from that other manifold of experience and of nature, space. The past is fixed and has determinate reality. The future is a realm to which being can only be attributed, at best, in an ‘indeterminate’ mode of a realm of unactualized potentiality. We have memories and records of the past, but, at best, only inferential knowledge of a different sort of what is to come. Causation proceeds from past to future, what has been and what is determining what will be, but determination never occurs the other way around. Our attitudes of concern and our concepts of agency are likewise time asymmetric.

Since the late middle part of the nineteenth century there have been recurrent claims to the effect that all the intuitively asymmetric features of temporality are ‘reducible to’ or ‘grounded in’ the asymmetry of physical systems in the world that is captured by the Second Law of Thermodynamics. The suggestion is, as far as I know, first made by Mach. It is taken up by Boltzmann and used by him in his final attempt to show the statistical mechanical reduction of thermodynamics devoid of paradox. The claim is treated in detail in the twentieth century by Reichenbach, and accepted as legitimate by a host of other philosophers, mostly philosophers of science.

The physical asymmetry to which the asymmetry of time is to be reduced is that which tells us that the entropy of an isolated system can only increase in the future direction of time and can never spontaneously decrease.