The limits of deductive reasoning
Posted this a while back too. You see, physics such as Fermat's Principle of Least Time, and Snell's Law are deductive. Theory fits observation, and vice versa. There are no contradictions allowed, it is always either true or false.
Bell's Inequality is statistical. Whether it is true, false, or both, is a matter of probability. Things get fuzzy, and I'm not sure if theory and observation are mutually derivable and reversible. Hence, in the absence of a reversible logic, it is spooky.