Experimental evidence, the heuristics of indistinguishability, and its logical inconsistency with quantum formalism all argue against the existence of a quantum mixture uncorrelated with the exterior, that is, argue for the postulate "The state of a system uncorrelated with its exterior is pure." This is shown to be equivalent with "The state of a system describable in terms of indistinguishable pure states is pure," and with "The state of the universe is pure"; further, it yields a quantitative expression of the traditional relation of welcher Weg information to partial coherence. It is concluded that all mixtures are "improper," the trace-reduction of a composite system's pure state.
Experimental evidence, the heuristics of indistinguishability, and its logical inconsistency with quantum formalism all argue against the existence of a quantum mixture uncorrelated with the exterior, that is, argue for the postulate "The state of a system uncorrelated with its exterior is pure." This is shown to be equivalent with "The state of a system describable in terms of indistinguishable pure states is pure," and with "The state of the universe is pure"; further, it yields a quantitative expression of the traditional relation of welcher Weg information to partial coherence. It is concluded that all mixtures are "improper," the trace-reduction of a composite system's pure state.
All quantum mixtures are what d'Espagnat has termed "improper." His "proper" mixture cannot be created -- if welcher weg, or distinguishing, information exists, an improper mixture results, while in the absence of such information, the resulting "mixture" is a pure state. D'Espagnat has claimed that an interpretation of the improper mixture in terms of subensembles leads to logical inconsistency; this claim is shown to be incorrect, as d'Espagnat's argument fails to account for the indistinguishability of the pure-state subensembles.
This paper argues that the ignorance interpretation of mixtures is physically unrealistic. The ignorance interpretation is the orthodox interpretation for mixtures, and should not be confused with the ignorance interpretation for superpositions, which has been largely abandoned. Mixtures, unlike superpositions, do not interfere. They are represented by mixed (or non-idempotent, i.e. W2 ≠ W) operators; superpositions, by pure (or idempotent) operators or by vectors. In the minimal interpretation both pure and mixed operators may be taken to describe collections. Any pure state, ψ, may be expressed as a sum of other pure states, ϕ 1, ϕ2, …, ϕn. Yet we cannot postulate that the members of the collection described by ψ are each in one of the pure states ϕ1, ϕ2, …, ϕn. This is because of the interference between the pure ϕ1, …, ϕn. On the other hand, if we have a mixture of ϕ1, …, ϕn, we can consistently postulate that the members of the collections are each in one of the pure states ϕ1, …, ϕn.
Quantum information is nonlocal in the sense that local measurements on a composite quantum system, prepared in one of many mutually orthogonal states, may not reveal in which state the system was prepared. It is shown that in the many copy limit this kind of nonlocality is fundamentally different for pure and mixed quantum states. In particular, orthogonal mixed states may not be distinguishable by local operations and classical communication, no matter how many copies are supplied, whereas any set of N orthogonal pure states can be perfectly discriminated with m copies, where m<N. Thus mixed quantum states can exhibit a new kind of nonlocality absent in pure states. We also argue that a set of orthogonal quantum states may be said to be maximally indistinguishable iff the set is not conclusively locally distinguishable with multiple copies.
We show that several classes of mixed quantum states in finite-dimensional Hilbert spaces which can be characterized as being, in some respect, ``most classical'' can be described and analyzed in a unified way. Among the states we consider are separable states of distinguishable particles, uncorrelated states of indistinguishable fermions and bosons, as well as mixed spin states decomposable into probabilistic mixtures of pure coherent states. The latter was the subject of the recent paper by Giraud et al. [Phys. Rev. A 78, 042112 (2008)], who showed that in the lowest-dimensional nontrivial case of spin 1, each such state can be decomposed into a mixture of eight pure states. Using our method, we prove that in fact four pure states always suffice.
This paper gives a thorough critique of the foundations of quantum physics in its mainstream interpretation (i.e., treating pure states as primitives, without reference to hidden variables, and without modifications of the quantum laws). This is achieved by cleanly separating a concise version of the (universally accepted) formal core of quantum physics from the (controversial) interpretation issues. The latter are primarily related to measurement, but also to questions of existence and of the meaning of basic concepts like 'state' and 'particle'. The requirements for good foundations of quantum physics are discussed. Main results: * Born's rule cannot be valid universally, and must be considered as a scientific law with a restricted domain of validity. * If the state of every composite quantum system contains all information that can be known about this system, it cannot be a pure state in general.
We study improper mixtures from a quantum logical and geometrical point of view. Taking into account the fact that improper mixtures do not admit an ignorance interpretation and must be considered as states in their own right, we do not follow the standard approach which considers improper mixtures as measures over the algebra of projections. Instead of it, we use the convex set of states in order to construct a new lattice whose atoms are all physical states: pure states and improper mixtures. This is done in order to overcome one of the problems which appear in the standard quantum logical formalism, namely, that for a subsystem of a larger system in an entangled state, the conjunction of all actual properties of the subsystem does not yield its actual state. In fact, its state is an improper mixture and cannot be represented in the von Neumann lattice as a minimal property which determines all other properties as is the case for pure states or classical systems. The new lattice also contains all propositions of the von Neumann lattice. We argue that this extension expresses in an algebraic form the fact that—alike the classical case—quantum interactions produce nontrivial correlations between the systems. Finally, we study the maps which can be defined between the extended lattice of a compound system and the lattices of its subsystems.