Quantum Foundations

There are a number of arguments in the philosophical, physical and cosmological literatures for the thesis that time is not fundamental to the description of nature. According to this view, time should be only an approximate notion which emerges from a more fundamental, timeless description only in certain limiting approximations. My first task is to review these arguments and explain why they fail. I will then examine the opposite view, which is that time and change are fundamental and, indeed, are perhaps the only aspects of reality that are not emergent from a more fundamental, microscopic description. The argument involves several aspects of contemporary physics and cosmology including 1) the problem of the landscape of string theory, 2) cosmological inflation and the problem of initial conditions, 3) the interpretation of the “wavefunction of the universe,” and the problem of what is an observable in classical and quantum general relativity. It also involves issues in the foundations of mathematics and the issue of the proper understanding of the role of mathematics in physics. The view that time is real and not emergent is, I will argue, supported by considerations arising from all these issues It leads finally to a need for a notion of law in cosmology which replaces the freedom to choose initial conditions with a notion of laws evolving in time. The arguments presented here have been developed in collaboration with Roberto Mangabeira Unger .

I discuss how we can give a satisfactory account of theory confirmation for theories with random data, such as Copenhagen quantum theory, despite the lack of a completely satisfactory definition of probabilistic theories of nature. I also explain why neither this nor any other proposed account of scientific confirmation works for many-worlds theories

I will identify six \'problems of time\' that arise in connection with quantum gravity and review the extent to which some of them can be regarded as solved, highlighting the very different aspects that they assume depending on one\'s starting point: Hamiltonian vs. path-integral, discrete vs continuous.

The evidence for the big bang is now overwhelming. However, the basic question of what caused the bang remains open. One possibility is that time somehow \'emerged,\' placing the universe in an inflationary state. Another, perhaps more conservative possibility, is that the big bang was a violent event in a pre-existing universe. I will describe model calculations employing the AdS/CFT correspondence which show how this is possible, and which point to a new explanation for the origin of large scale structure in the universe.

In recent work with Bob Coecke and others, we have developed a categorical axiomatization of quantum mechanics. This analyzes the main structural features of quantum mechanics into simple and general elements, which admit an elegant diagrammatic representation. This enables an illuminating and effective analysis of quantum information protocols and computational structures. One aspect which is brought to light in this analysis is that protocols such as teleportation use entanglement to achieve a logical information flow which has an apparent retro-causal (or even `backward-in-time\') component. However, there is also a physical or operational description of the same systems, based on an abstract structural characterization of quantum measurements, which is entirely causally consistent. The systematic relationship between these two descriptions suggests a novel perspective on the flow of time and information in the quantum realm.

In recent years there has been a growing awareness that studies on quantum foundations have close relationships with other fields such as probability and information theory. In this talk I give another example of how such interdisciplinary work can be fruitful, by applying some of the lessons from quantum mechanics, in particular from Bell\'s theorem, to a debate on the philosophical foundations of decision theory. I argue that the basic assumptions of the popular causal decision theory -- which was developed partly in response to a puzzle proposed by the physicist William Newcomb and published by the philosopher Robert Nozick -- are analogous to the basic assumptions of a local hidden-variables theory in the context of Bell\'s theorem. Both have too strong a prejudice about the causal structure of the world: there are possible games the world can pose such that an agent who operates by those theories is constrained to choose losing strategies no matter what evidence he or she acquires.

I\'ll sketch of a proposal for unifying classical and quantum probability, arguing first for the need to recognize a measure over phase space as a component of classical theories (indeed, of any theory satisfying certain constraints and capable of generating predictions for open systems) and then showing how to use that measure to define objective chances. Time permitting, I\'ll briefly address questions about the nature and interpretation of the measure.

As is well known, time-energy uncertainty generically manifests itself in the short time behavior of a system weakly coupled to a bath, in the energy non-conservation of the interaction term (H_I does not commute with H_0). Similarly, the monotonic evolution of the system density operator to its equilibrium value which is a universal property of quantum dynamical semigroups (Spohn\'s theorem), e.g., systems with Lindbladian evolution, is in general violated at short (non-Markovian) timescales. For example, frequent, brief non-demolition measurements of the energy states of a two level system (TLS) coupled to a bath, disturbs the thermal equilibrium between them, despite leaving the system and bath states separately unperturbed. For sufficiently short intervals between measurements (Zeno regime) the system and bath heat up immediately following the measurement. It is also possible to have net cooling in an intermediate (anti-Zeno-like) regime. The evolution of the system state away from its equilibrium value, not only violates the Markovian-dynamics version of the 2nd law (Spohn\'s theorem), but also Lindblad\'s theorem on which it rests, which is valid for any evolution described by a completely positive map. This does not imply that the evolution is not completely-positive, but rather that it is not a well-defined map at allthe evolution of the state of the system is not determined by this state alone (nor even together with the reduced state of the bath), but rather by the full joint system-bath state (this indeterminacy was shown previously, by Buzek et al., for special cleverly constructed joint states). Ref: N. Erez, G. Gordon, M. Nest & G. Kurizki, Nature 452, 724 (2008)

In reference [1] R. D. Sorkin investigated a formulation of quantum mechanics as a generalized measure theory. Quantum mechanics computes probabilities from the absolute squares of complex amplitudes, and the resulting interference violates the (Kolmogorov) sum rule expressing the additivity of probabilities of mutually exclusive events.However, there is a higher order sum rule that quantum mechanics does obey, involving the probabilities of three mutually exclusive possibilities. We could imagine a yet more general theory by assuming that itviolates the next higher sum rule.An experiment is in progress in our laboratory which sets out to test the validity of this second sum rule by measuring the interference patterns produced by three slits and all the possible combinations of those slits being open or closed. We use either attenuated laser light or a heralded single photon source (using parametric down conversion) combined with single photon counting to confirm the single photon character of the measured light. We will show results that bound the possible violation of the second sum rule and will point out ways toobtain a tighter experimental bound.[1] R. D. Sorkin, Quantum Mechanics as Quantum Measure Theory,Mod. Phys. Lett. A 9, 3119 (1994).

It has sometimes - though usually informally - been suggested that the psychological arrow can be reduced to the thermodynamic arrow through information processing properties of the brain. In this talk we demonstrate that this particular suggestion cannot succeed, as, insofar as information processing (at least in the sense of a classical computer) has an arrow of time, it is not governed by the thermodynamic arrow.

In contrast to Heisenberg\'s position-momentum uncertainty relation, the status of the time-energy uncertainty relation has always remained dubious, For example, it is often said that \'time\' in quantum theory is not an observable and not represented by a self-adjoint operator. I will review the background of the problem and propose a view on the uncertainty relations in which the cases of position-momentum and time-energy can be treated in the same way.