A persistent focus on the concept of emergence as a core element of the scientific method allows a clean separation, insofar as this is possible, of the physical and philosophical aspects of the problem of outcomes in quantum mechanics. The philosophical part of the problem is to explain why a closed system has definite experimental outcomes. The physical part is to show mathematically that there exists a limit in which the contradiction between unitary Schroedinger dynamics and a reduction process leading to distinct outcomes becomes negligible according to an explicitly stated criterion, and to make this criterion as objective as possible. The physical problem is solved here by redefining the notion of a quantum state and finding a suitable measure for the change of state upon reduction. The appropriate definition of the quantum state is not as a ray or density operator in Hilbert space, but rather as an equivalence class consisting of all density operators in a given subspace, the members of which all describe the same experimental outcome. For systems containing only subsystems that are integrated with their environments, these equivalence classes can be represented mathematically by projection operators, and the resulting formalism is closely related to that used by von Neumann to study the increase of entropy predicted by the second law of thermodynamics. However, nearly isolated subsystems are reduced only indirectly, as a consequence of their interaction with integrated subsystems. The reduced states of isolated subsystems are the same conditional states used in the definition of quantum discord. The key concepts of decoherence theory can all be adapted to fit this definition of a quantum state, resulting in a unified theory capable of resolving, in principle, all aspects of the quantum measurement problem. The theory thus obtained is weakly objective but not strongly objective.