Page:Grimm's Household Tales, vol.1.djvu/48

 Indo-European märchen, there remained the process of slow filtration and transmission on one hand, and the similarity of the workings of the human mind (especially in its earlier stages) on the other hand. But Mr. Max Müller had already discredited the hypothesis that märchen "might have been invented more than once" (Chips, ii. 233). "It has been said," writes Mr. Müller, "that there is something so natural in most of the tales, that they might well have been invented more than once. This is a sneaking argument, but has nevertheless a certain weight. It does not apply, however, to our fairy tales. They surely cannot be called 'natural.' They are full of the most unnatural conceptions. . ." Among these unnatural conceptions, Mr. Müller noted the instance of a frog wooing a maiden; and he went on, as we have already seen, to explain such ideas on the hypothesis that they resulted from "a disease of language," from forgetfulness of the meaning of words. Now some little anthropological study had shown us that the ideas (so frequent in Household Tales), which Mr. Müller calls unnatural, were exactly the ideas most natural to savages. So common and so natural is the idea of animal kinship and matrimonial alliance with animals to the savage mind, that stories turning on these data are, of all stories, the most likely to have been invented in several places. We do not say that they were thus separately invented, but only that the belief on which they turn is, of all beliefs, the most widely diffused. Having once attained this point, we soon discovered that other essential incidents in märchen, incidents which seem unnatural to civilised men, are common and accredited parts of the savage conception of the world he lives in. When this was once