Quantum theory is plenty strange, but one of the strangest discoveries is the realization that there’s a limit to how much you can measure at any one time. This was famously described by Werner Heisenberg, with his uncertainty principle: how you can never know both the position and motion of a particle at the same time.
We record Astronomy Cast live every Monday at 12 pm PST / 3 pm EST / 2000 GMT. If you want to join in our recording, just make sure you’ve got Fraser circled on Google+, then the show will show up in your stream. You can also watch us live at Cosmoquest.
The most mysterious aspect of quantum mechanics is we are not sure what we mean by reality. Quantum mechanics is about as much a physical logic as it is a theory. We have curious problems of what we mean by reality, for there exist in QM complementary sets of observables, say A and B, where physical systems can be described by A or B as an exclusive or, and where the two are incommensurate or contradictory. A classic case is observables corresponding to position and momentum. The complementary observables obey a noncommutative algebra
[x, p] = i?,
where ? is the fundamental unit of action. A measurement may give information about a system as A or B, but not both, and so prior to such measurements we have a hard time saying there is anything ontological about the quantum system. Prior to a measurement a quantum system can only be said to be epistemic, but a measurement is what gives ontology. Piles of papers have been written on quantum interpretations, where any interpretation is just a formal equivalence between sets of rules for operating on experimental data. This makes quantum interpretations ineffective, or not something which can be determined by empirical testing. I think that quantum gravity makes the situation far worse, for on top of there being a complementarity of observables mathematically represented by noncommutative operators, there is a duality between causal descriptions (S-matrices) separated by event horizons (quantum horizons) where this extends into nonassociativity.
LC