John D.
Norton


home >> research >> Maxwell's Demon...  
Maxwell's demon is a fictitious, miniscule being imagined by Maxwell as able to reverse the second law of thermodynamics by manipulating individual molecules. In a tradition of work initiated by Szilard in the 1920s, it has become standard to predict the failure of the demon on information theoretic grounds through a connection supposed to obtain between information processing and entropy dissipation. In a study with John Earman, we have suggested that this account of the demon's failure is either based on question begging or groundless supposition.  With John Earman, "Exorcist XIV: The Wrath of Maxwell's
Demon." Studies in the History and Philosophy of Modern Physics, Part I
"From Maxwell to Szilard" 29(1998), pp.435471; Part II: "From Szilard
to Landauer and Beyond," 30(1999), pp.140. Download. 

The present orthodoxy holds that Maxwell's demon must fail to reverse
the second law of thermodynamics because of a hidden entropy cost in
the erasure of information. The analysis is based on Landauer's
principle, which asserts that the erasure of n bits of information is
accompanied by the passage of least k ln n of entropy to the
surroundings. I argue that Landauer's principle is based on the
formation of illicit canonical ensembles in statistical physics that
give the illusion of the necessity of this entropy cost. I also urge
that, even if the principle were correct, the literarure seeks to
establish that it must defeat all Maxwell demons by the inadquate means
of merely displaying a few suggestive examples. 
"Eaters of the Lotus: Landauer's Principle and the Return of Maxwell's Demon." 36 (2005), pp. 375411. Download  
Landauer's Principle asserts that there is an unavoidable cost in thermodynamic entropy creation when data is erased. It is usually derived from incorrect assumptions, most notably, that erasure must compress the phase space of a memory device or that thermodynamic entropy arises from the probabilistic uncertainty of random data. Recent work seeks to prove Landauer’s Principle without using these assumptions. I show that the processes assumed in the proof, and in the thermodynamics of computation more generally, can be combined to produce devices that both violate the second law and erase data without entropy cost, indicating an inconsistency in the theoretical system. Worse, the standard repertoire of processes selectively neglects thermal fluctuations. Concrete proposals for how we might measure dissipationlessly and expand single molecule gases reversibly are shown to be fatally disrupted by fluctuations.  "Waiting for Landauer," Studies in History and Philosophy of Modern Physics, 42(2011), pp. 184198. Download. For my reply to Ladyman and Robertson's reply to this paper, see "Author's Reply to 'Landauer Defended'," Studies in History and Philosophy of Modern Physics, 44 (2013), p. 272. Download. See also Goodies pages: When a Good Theory meets a Bad Idealization: The Failure of the Thermodynamics of Computation. No Go Result for the Thermodynamics of Computation For the latest and best developed version of the "no go" result, see "All Shook Up..." below. 

Entropy creation in excess of that tracked by Landauer's principle is needed to overcome fluctuations in molecular scale computation. This paper is a short account of the "no go" result reported in "Waiting for Landauer."  "The End of the Thermodynamics of Computation: A No Go Result."Philosophy of Science. 80, (2013), pp. 11821192. Download. For the latest and best developed version of the "no go" result, see "All Shook Up..." below. 

Brownian computers are supposed to illustrate how logically reversible mathematical operations can be computed by physical processes that are thermodynamically reversible or nearly so. In fact, they are thermodynamically irreversible processes that are the analog of an uncontrolled expansion of a gas into a vacuum.  "Brownian Computation is Thermodynamically Irreversible." Foundations of Physics. 43 (2013), pp 13841410.Download. "On Brownian Computation" International Journal of Modern Physics: Conference Series. 33 (2014), pp. 14603661 to 14603666. download. 

The most successful exorcism of Maxwell’s demon is Smoluchowski’s 1912 observation that thermal fluctuations would likely disrupt the operation of any molecular scale demonic machine. Informationtheoretic exorcisms fail since these same thermal fluctuations invalidate the molecular scale manipulations upon which the thermodynamics of computation is based. A new argument concerning conservation of phase space volume shows that all Maxwell’s demons must fail.  "All Shook Up: Fluctuations, Maxwell's Demon and the Thermodynamics of Computation." Entropy 2013, 15, 44324483. Download. For a short extension of the exorcism to quantum theory, see ""The Simplest Exorcism of Maxwell's Demon: The Quantum Version." Download. 

Efforts to exorcise Maxwell's demon have focused on the information processing a demon supposedly must do. There is a much simpler exorcism. If the demon and thermal system combined are a Hamiltonian system, then the intended operation of the demon must compress the overall phase space, in violation of Liouville's theorem.  The Simplest Exorcism of Maxwell's Demon No Information Needed In Goodies pages. Based on Section 4 of "All Shook Up..." 

The idea of a thermodynamically reversible process is central to thermodynamics. Yet essentially all descriptions of them over nearly two centuries are internally contradictory. They consist of equilibrium states, which are by definition unchanging in time; yet still they still change in time. I review the history and offer a solution. 
"The Impossible Process: Thermodynamic Reversibility" Draft 

Thermodynamically reversible processes cannot be completed in systems at molecular scales. They are fatally disrupted by fluctuations. This paper reviews the general result and computes two cases in detail. 
"Thermodynamically Reversible Processes in Statistical Physics." Draft 

The received view is that a Maxwell's demon must fail to reverse the second law of thermodynamics for reasons to do with information and computation. This received view has failed, I argue, and our continuing preoccupation with it has distracted us from a simpler and more secure exorcism that merely uses the Liouville theorem of statistical physics. I extend this exorcism to the quantum case.  "Maxwell's Demon Does not Compute." Prepared for Michael E. Cuffaro and Samuel C. Fletcher, eds., Physical Perspectives on Computation, Computational Perspectives on Physics. Cambridge University Press." Download draft.  