The problem of how to make sense of the non-classical probabilistic behavior of quantum systems was the subject of an unresolved dispute between Einstein and Bohr. The opposition between their opposing philosophical views has provided the tension for an ongoing debate among physicists and philosophers of physics about the conceptual significance of the transition from classical to quantum mechanics, centering on a measurement problem in the foundations of the theory. The 1990s saw the development of a quantum theory of information, prompted by the realization that certain nonclassical features of quantum systems could be exploited to perform information-processing tasks that would be impossible in a Newtonian world. This has led to an explosion of interest in theoretical and practical applications of quantum computation and quantum cryptography, and a reconsideration of the foundational problems underlying the Bohr-Einstein debate. Intellectual Merit: The PI has published a number of recent papers on these issues, in particular a paper (jointly authored with Rob Clifton and Hans Halvorson) showing that one can derive the basic features of a quantum-theoretic description of physical systems from three information-theoretic constraints. The research project involves working out the details of an information-theoretic interpretation of quantum mechanics in the light of the Clifton-Bub-Halvorson result. A core feature is to split the measurement problem into a problem about truth and a problem about probability, and to set aside the truth problem. The PIs solution to the probability problem involves showing (i) that one can derive from quantum mechanics the conditions for the existence of measuring instruments as sources of classical information, and the existence of information-gatherers with the ability to use measuring instruments to apply and test quantum mechanics, and (ii) that the generalized transition probabilities of the theory can be given a subjective Bayesian interpretation without solving the truth problem of measurement. The PI argues that the possible solutions to the truth problem are well understood, and proposes to show that any such solution produces a modification of quantum mechanics that is explanatorily and methodologically inferior to the standard theory, particularly with respect to accounting for the new results in quantum computation and quantum cryptography. Broader Impacts: The PI proposes to write a book developing these ideas accessible to scientists who work on issues in quantum information and computation, to philosophers of science who work on questions of interpretation, and to a wider group of readers interested in developments in the foundations of physics. The PI has been invited to present papers on topics related to the above information-theoretic interpretation of quantum mechanics at various conferences, and to contribute articles to several Handbooks and Encyclopedias. A number of recent papers and PhD dissertations discuss the Clifton-Bub-Halvorson result and its implications, and three of the PIs PhD students will shortly be working on dissertations on various aspects of quantum information. The results of this research will be incorporated in an undergraduate course on philosophy of physics (Phil 354) directed at physics students. The PI will also disseminate information on the project through an annual "new directions in the foundations of physics conference," and through online archives. 1
