Relatedly in Mencius on the Mind: Experiments in Multiple Definition (1932):
can we maintain two systems of thinking in our minds without reciprocal infection and yet in some way mediate between them? And does not such mediation require yet a third system of thought general enough and comprehensive enough to include them both? And how are we to prevent this third system from being only our own familiar, established tradition of thinking rigged out in some fresh terminology or other disguise ? There is nothing new about the problem, that is obvious. What would run some risk of novelty would be an attempt to discuss it explicitly, to bring it out of the realm of midnight dubiosities and initial misgivings into the field of arguable methodology or technique. The problem seems to grow still more formidable as we realize that it concerns not only incommensurable concepts but also comparisons between concepts and items which may not be concepts at all. (87)