to grammar, which bubble up continually in pragmatist writings. "For illustration take the former fact that the earth is flat," says one, quite innocently; and another observes that "two centuries later, nominalism was evidently true, because it alone would legitimise the local independence of cities." Lest we should suppose that the historical sequence of these "truths" or illusions is, at least, fixed and irreversible, we are soon informed that the past is always changing, too; that is (if I may rationalise this mystical dictum), that history is always being rewritten, and that the growing present adds new relations to the past, which lead us to conceive or to describe it in some new fashion. Even if the ultimate inference is not drawn, and we are not told that this changing idea of the past is the only past that exists--the real past being unattainable and therefore, for personal idealism, non-existent--it is abundantly clear that the effort to distinguish fact from theory cannot be successful, so long as the psychological way of thinking prevails; for a theory, psychologically considered, is a bare fact in the experience of the theorist, and the other facts of his experience are so many other momentary views, so many scant theories, to be immediately superseded by other "truths in the plural." Sensations and ideas are really distinguishable only by reference to what is assumed to lie without; of which external reality experience is always an effect (and in that capacity is called sensation) and often at the same time an apprehension (and in that capacity is called idea).