THE SCOPE OF THERMODYNAMICS

This thesis investigates the application of the laws of thermodynamics to various sub-systems of the universe. I begin by distinguishing between three different second laws: orthodox, statistical and probabilistic. After suggesting that entropy is best understood in means-relative terms, I then show that this interpretation does not imply an epistemic understanding of thermodynamics, pace Jaynes. I conclude the discussion of phenomenological thermodynamics by arguing that thermodynamics can properly be applied to systems containing only a single particle, given the right circumstances. Next I discuss thermodynamics in the context of black holes, and examine the question of whether black hole entropy is genuine thermodynamic entropy, as opposed merely to information-theoretic entropy. I examine the original arguments by Bekenstein and Hawking, and conclude that these are unsatisfactory, but I go on to demonstrate that black holes ought to be considered to be genuine thermodynamic objects by constructing a black hole Carnot cycle. The third chapter investigates thermodynamics in the quantum realm and begins by discussing a recent argument by Hemmo and Shenker against the identification of the von Neumann entropy with the thermodynamic entropy. Their argument is shown to be flawed as it a) allows for a violation of the second law and b) is based on an incorrect calculation of the entropy. I continue by providing a derivation of the laws of thermodynamics from quantum mechanics and a few phenomenological assumptions. This approach is then compared to so-called resource theories of thermodynamics and to single-shot thermodynamics. I end my discussion of quantum thermodynamics with the analysis of an argument made by Cabello et al., who claim that thermodynamics allows for the derivation of an empirical difference between two important classes of quantum interpretation. I provide a counterexample to this claim and show that the alleged heat cost is fully accounted for in the external agent.