It is not the case that Phillips is either asserting the surprising claim that generative theorizing has never been led astray by demonstrably unreliable data, or he is changing the subject.
?Set your confidence on the premises below to see your aggregate.