Physics has a long history of employing demons as thought experiments in order to probe the limits of human knowledge. Maxwell’s demon is such a thought experiment: a molecule-sorting demon invented in 1878 by James Clerk Maxwell in order to con/test the universal validity of the second law of thermodynamics. The second law of thermodynamics not only postulates a universal increase of entropy and an “arrow of time” but also has been based on human ignorance, the assumption that there is an objective world independent of us that we can know only incompletely. Maxwell’s demon figures the scientist’s ability to acquire knowledge about the microscopic molecular world but ostensibly without the expenditure of work. This essay offers a rereading of Leo Szilard’s influential account of Maxwell’s demon (1929), which explores the influence of human intelligence on scientific measurements in relation to an increase of entropy and the irreversibility of measurements. Opening up ambiguities in Szilard’s account of measurement and memory and comparing it to quantum mechanical experiments (the quantum erasure effect), the author reads both through Karen Barad’s onto-epistemological framework of “agential realism” and Jacques Derrida’s ghostly logic of time, proposing that all measurements are haunted and that scientific activities, reconfigured as intra-active work, constitute time, rather than accumulating knowledge in time. At stake in such a deconstruction of the scientific “arrow of time” from within physics is the possibility of making a difference that matters in science or, in other words, accounting for transformation and rearticulation work as physical work.

The text of this article is only available as a PDF.
You do not currently have access to this content.