BiGyElLoWhAt said:
So I haven't made it very far into that paper yet, but does the "spooky action at a distance" come from the denial of the assertion that nothing 'exists' in the quantum world in advance of measurement? Am I reading this right?
In order for me to say that there IS spooky action at a distance (in the context of EPR), must I first say that nothing is definite in the quantum world before measurement, but, once measured, the object in question is forced into a state; then must I extend this assertion to not only does my measurement force the measured object into a particular state, but it also forces all entangled objects into some (probably, but not necessarily, different) state? But not in the sense that we don't know what state it's in until it's measured, but that it actually doesn't exist in a particular state until measured?
Yes, and no. Here I will take the answer that "spooky action at a distance" has nothing necessarily to do with the idea that we only get definite results when a measurement is made, and that we are agnostic about whether the wave function is real in any sense.
So I will define "spooky action at a distance" to mean that the Bell inequalities are violated. The Bell inequalities do not assume quantum mechanics. The Bell inequalities hold if special relativity holds, if two experiments are far away enough, and if a complete knowledge of variables in the causal past of each experiment is enough to determine the outcome of the joint results of the experiments. We summarize this by saying that the Bell inequalities hold if local variables describe what we observe. Notice that we did not use quantum mechanics at all in saying what the Bell inequalities mean. So we can do an experiment without knowing quantum mechanics, and if the Bell inequalities are observed to be violated, then we know that a theory that successfully describes our observations cannot contain only local variables.
Quantum mechanics is one example of a successful theory that predicts the violation of the Bell inequalities, ie. spooky action at a distance. There are two technical elements in quantum mechanics that allow it to predict the violation of the Bell inequalities (1) non-commuting variables, which basically means that there are no particles that have classical trajectories (2) entanglement, which roughly means that the superposition principle can be applied to the wave function of a system of interacting particles.
The "non-reality" problem is that in quantum mechanics, we think that the measuring apparatus should also be in principle described by quantum mechanics. And so the measuring apparatus should have a wave function. But we can also have a measuring apparatus to measure the measuring apparatus, which should have its wave function. As far as we can tell, quantum mechanics succeeds and we can give every measuring apparatus a wave function. However, we run into serious problems if we try have a wave function of the universe, which has no measuring apparatus to measure it. It is because of this serious problem, that we say that we are agnostic about the reality of the wave function. Maybe it is just a tool to calculate the results of measurements.
There are other theories that also predict the violation of the Bell inequalities, such as Bohmian mechanics. Although Bohnian mechanics has spooky action at a distance, it does not have any variables in it that don't make sense if we try to extend the theory to the whole universe. Bohmian mechanics does not require that there is a measuring apparatus sitting outside the universe. So everything in Bohmian mechanics can be taken as real. Accordingly, Bohmian mechanics is an example of a theory with spooky action at a distance, but with no "non-reality" problem. Bohmian mechanics reproduces quantum mechanics closely enough that some form of Bohmian mechanics might be useful if quantum mechanics is experimentally falsified. But till then, we user quantum mechanics, because it is generally easier to use than Bohmian mechanics.
There are other approaches to solving the "non-reality" problem. These include the Many-Worlds approach and the Consistent Histories approach. However, I think it is fair to say that there is no consensus at the moment as to whether these really work. Bohmian mechanics itself has a limitation, in that it is not known whether it can describe relativistic observations in a way that the underlying theory is also fully relativistic.
