They cannot see that the Irish are Irish whether or no they are Celtic, whether or no there ever were any Celts. And what misleads them once more is the _size_ of the theory; the sense that the fancy is bigger than the fact. A great scattered Celtic race is supposed to contain the Irish, so of course the Irish must depend for their very existence upon it. The same confusion, of course, has eliminated the English and the Germans by swamping them in the Teutonic race; and some tried to prove from the races being at one that the nations could not be at war. But I only give these vulgar and hackneyed examples in passing, as more familiar examples of the fallacy; the matter at issue here is not its application to these modern things but rather to the most ancient things. But the more remote and unrecorded was the racial problem, the more fixed was this curious inverted certainty in the Victorian man of science. To this day it gives a man of those scientific traditions the same sort of shock to question these things, which were only the last inferences when he turned them into first principles. He is still more certain that he is an Aryan even than that he is an Anglo-Saxon, just as he is more certain that he is an Anglo-Saxon than that he is an Englishman. He has never really discovered that he is a European. But he has never doubted that he is an Indo-European. These Victorian theories have shifted a great deal in their shape and scope; but this habit of a rapid hardening of a hypothesis into a theory, and of a theory into an assumption, has hardly yet gone out of fashion. People cannot easily get rid of the mental confusion of feeling that the foundations of history must surely be secure; that the first steps must be safe; that the biggest generalisation must be obvious. But though the contradiction may seem to them a paradox, this is the very contrary of the truth. It is the large thing that is secret and invisible; it is the small thing that is evident and enormous.