martinbn said:
No, I am saying that it would be less meaningless if it didn't make up unobservable things.
Plain old QCD relies upon all sorts of mechanism and interactions of quarks and gluons that aren't directly observable.
Generally speaking, we don't directly observe individual u, d, s, c, or b hadrons or gluons because they are confined within hadrons or are indistinguishable parts of a quark-gluon plasma.
We infer, from a combination of Occam's razor and observations, that there are three color charges (and three anti-color charges) of quarks, and eight possible color charge (including anti-color charge) combinations in gluons. But it is still impossible to identify the color charge of any particular quark or any particular gluon (in stark contrast to the easily observable electromagnetic charges of quarks, charged leptons and on shell W bosons, or the mass-energy charge with respect to the gravitational force of particular particles up to the limit of the Heisenberg uncertainty principle).
The most interesting parts of QM (regardless of interpretation) happen between measurements, and some of the inferred in between observations parts, like virtual particles (which if they were real particles would violate mass-energy conservation, a bedrock assumptions that holds for all observables in physics), are profoundly important to the end results.
Similarly, one of the things that we do to calculate the path integrals of particle propagators in QED, which is considering amplitudes for photon paths in which photons are moving at more or less than the speed of light in a vacuum "c" as part of the total path integral calculation, involve unobservable things that are fundamentally inconsistent with the bedrock principles of special relativity. Classical special relativity (which QED, QCD, and the weak force all follow perfectly at the level of observables) suggests that all photons must always be on paths between points consistent with velocities of exactly c. Yet, we rely on these unobservable things that are contrary to special relativity, and contrary to all experimental observations (statistically significant evidence of Lorentz invariance violations has never been observed), because when we do, we get the right end results of the overall calculation of observable quantities to the greatest precision of any scientific observations ever made by human beings.
We don't directly observe the intermediate imaginary number quantities that go into lots of ordinary classical electromagnetic calculations.
Like those familiar unobservable things, the unobservable parts of Bohemian quantum mechanics are inferred indirectly from observations in ways that, when you are doing it right, should be indistinguishable from, for example, the Copenhagen interpretation, with experiments devised to date that allow us to make these indirect observations.
So, merely predicting unobservable things isn't very troubling if, as in the case of Bohemian quantum mechanics, the unobservable things imply observable things.
The fact that there is more than one possible set of unobservable mechanisms to produce the same results in quantum mechanics that humans can conceive isn't really all that shocking or surprising.
End results that can arise in more than one possible way are well within the realm of human experience.
For example, there are usually many possible different ways that the pieces midway through a chess game could have ended up there and nobody is very troubled by taking on chess problems that don't include the moves that led up to its starting point.