Maybe I'm off-base here, but I always thought that, by definition, a scientific theory makes testable predictions.... and if those predictions are shown to be true, that's the scientifically accepted "proof" that the theory is valid.
Or, to put it another way, just because you can't reproduce something in a laboratory under controlled conditions does NOT mean that evidence supporting a theory's validity does not exist. Seems to me that the quoted person is getting hung up on the word "proof".