A) Unlike Bell's theorem, the GHZ argument is based on unverifiedassumptions concerning the physical reality of a particular state vector andmeasurability of certain Hermitian operators pertaining to a system of threecorrelated spin-(l/2) particles.B) Unlike Bell's theorem, the GHZ formulation is limited to determinis-tic local theories.C) A direct experimental test of the GHZ argument is probably imposs-ible.

But in theclassical domain it is always possible to assign a priori well-defined values to allobservable quantities. This result of Garg and Mermin is disturbing for thecoherence and rationality of the existing quantum theory, which seems toextend its (~magic,) predictions also to the macroscopic domain where classicalphysics had successfully banished all <(magicab) approaches

Other interesting consequences of local realism were found by Garg andMermin[61] who were able to deduce Bell-type inequalities for two spin-jparticles (with arbitrary j). They could show that the singlet state for twoparticles with spin j leads to violations of local realism for arbitrarily largevalues of j right up and above the threshold of the classical world

We disagree with the contention by Mermin [46] that the GHZ formulation(ds an altogether more powerful refutation of the existence of elements ofreality than the one provided by Bell's theorem~. The reasons are, first of allthat one can talk of refutation only after an unambiguous experimental verdict,and moreover