Hi Edwin,
A lot of physicists have undertaken to explain why action at a distance isn't "spooky." Aharonov's and everyone else's explanation necessarily relies on a probabilistic interpretation of reality. (Richard Gill is also among those who do not think that nonlocality is necessary to quantum mechanics -- that probability alone determines measurement outcomes.)
All of these explanations end up assuming what they intend to prove, for at least one very elementary mathematical reason (a), and one very elementary physical reason(b): a) the measure space is made of a continuous range of variables, while the measurement result is discrete (this is what makes Lamport's discovery profound); b) given that relativity is true, there can be no probabilistic distinction between past and future, which is equivalent to assuming a privileged coordinate frame.
Like Joy Christian, Aharonov et al propose nonlocality as only apparently true: " ... what appears to be nonlocal in space turns out to be perfectly local in spacetime." So far, so good -- except that it is impossible in principle to perform a probability measure on the spacetime continuum. The difference between purportedly entangled discrete particles and demonstrably correlated continuous wave functions is a strict constraint; spacetime can't be discretized without assuming either nonlocality or a privileged frame of reference. If considered to be a complete theory, quantum mechanics *cannot* discard the assumption of nonlocality, or else the theory is incoherent.
Joy does away with the conundrum, by allowing a natural, constructive, and globally continuous topological condition (all proofs of nonlocality are nonconstructive, and all topology is global) to be realized in the local discrete measurement outcomes -- without invoking probability or time, which in turn shows that quantum correlations do not result from particle entanglement, and both nonlocality and entanglement are illusions.
In other words, while Bell's theorem explicitly demonstrates that no classical theory can be derived from the principles of quantum mechanics -- it does not forbid quantum mechanical results from being derived on classical principles. This latter is what Joy has demonstrated, and which satisfies the completeness criterion for a physical theory; i.e., every element of the mathematical theory corresponds to every element of the physical measure.
Tom