I hear you, and I think that your comparison here actually proves my point in a way. That is, according to the general theory of relativity, gravity can be approached as a force emanating from a massive body, or as an acceleration caused by a disturbance of spacetime. (Or, as my professor used to illustrate, for all we know we could be at the bottom of a giant bucket being swung in circles by the Great Earth Goddess!) Regardless of which framework we consider, the parachutist will still fall to the earth, but if we absolutely and totally believe in the Great Goddess's bucket, then the conventional view of gravity need not apply to us; our observations of phenomena are completely explained by an alternative theory.
Similarly, if we choose to take operant theory out of the picture of parenting, the child will still engage in the same behaviors upon which the operant definitions had been placed. But if we are not focusing on *behaviors*, we cannot apply the definitions about which behaviors are increasing or decreasing in frequency, so in that sense we *aren't* punishing or reinforcing no matter what a Skinnerian behaviorist would observe in the same situation.
But it kind of goes into "reality is changed by intention" (which might be true) and "reality is changed by the beliefs of the observer" (hmmm which might be true, too!)
Too heavy before coffee.