Complementing Science: Lessons from the History and Philosophy of Thermal Physics

Abstract In this paper, I argue that the history and philosophy of thermodynamics of computation in 1998–2024 has contributed to gain scientific knowledge about Maxwell’s demon and Landauer’s principle, thus successfully implementing Hasok Chang’s model of integrated HPS as complementary science. Here, I distinguish three senses in which this HPS field has contributed epistemically to thermal physics. Firstly, by reopening a closed research programme based on Smoluchowski’s 1912 fluctuation-based exorcism of Maxwell’s demon Norton (2013d) expanded and diversified the knowledge gained by the orthodox informational programme by developing novel arguments, hypotheses, and evidence. Secondly, several studies in this field such as Earman and Norton (1999) and Hemmo and Shenker (2012) have generated critical awareness among special scientists by challenging the soundness and universal validity of some ideas, thus increasing knowledge quality by refuting unjustified beliefs, sophisticating the proofs for and against Landauer’s principle, and therefore improving their objections. Finally, I argue that conceptual engineering proposals in this field have contributed to improve how special scientists understand problematic and poorly-defined notions such as ‘thermodynamic irreversibility’ or ‘implementing logic operations’.