A new proof of the impossibility of reconciling realism and locality in quantum mechanics is given. Unlike proofs based on Bell's inequality, the present work makes minimal and transparent use of probability theory and proceeds by demonstrating a Kochen-Specker type of paradox based on the value assignments to the spin components of two spatially separated spin-1 systems in the singlet state of their total spin. An essential part of the argument is to distinguish carefully two commonly confused types of contextuality; (...) we call them ontological and environmental contextuality. These in turn are associated with two quite distinct senses of nonlocality. We indicate the relevance of our treatment to other related discussions in recent literature on the philosophy of quantum mechanics. (shrink)

A combination of methodological considerations and propositions about the causal structure of spacetime provides a reply to Fine's criticisms of the "factorizability requirement" used in several versions of Bell's theorem. His proposal of "action in harmony" is criticized. Experimental tests are proposed for both the "synchronization models" and the "prism models", which Fine has invented as loopholes to Bell's theorem. A theorem of Suppes and Zanotti which purports to show the impossibility of hidden variables is criticized. One of their crucial (...) premisses seems to be justifiable only if one accepts a Parmenidean metaphysics. (shrink)

Two principles of locality used in discussions about quantum mechanics are distinguished. The intuitive no-action-at-a distance requirement is called physical locality. There is also a mathematical requirement of a kind of factorizability which is referred to as "locality". It is argued in this paper that factorizability is not necessary for physical locality. Ways of producing models that are physically local although not factorizable which are concerned with correlations between the behavior of pairs of particles are suggested. These models can account (...) for all the quantum mechanical single and joint probabilities. (shrink)