Page:The Science of Fairy Tales.djvu/266

 traces of sacrifice. The offering of an innocent child is familiar, if not comprehensible, enough to any one who has the most superficial acquaintance with savage rites. We have already seen that an unbaptized child is regarded as a pagan, and is an object of desire on the part of supernatural beings. The same reasons which induce fairies to steal it would probably render it an acceptable offering to a pagan divinity. No words need be wasted on the torture, or the tale of the mice. But the personal violence, if indeed the remnant of a tradition of sacrifice, involves the slaughter of the divinity herself. This might be thought an insuperable objection; but it is not really so. For, however absurd it may seem to us, it is a very widespread custom to sacrifice to a divinity his living representative or incarnation, whether in animal or human form. It is believed in such cases that the victim's spirit, released by sacrifice, forthwith finds a home in another body. The subject is too vast and complex to be discussed here at length; the reader who desires to follow it out can do so in Mr. Frazer's profoundly interesting work on "The Golden Bough." Assuming, however, the custom and belief, as here stated, to be admitted, it will be seen that the underlying thought is precisely that which we want in order to explain this mode of disenchantment. For if, on the one hand, what looks like murder be enjoined in a number of stories for the purpose of disenchanting a bewitched person; and if, on the other hand, the result of solemnly slaughtering a victim be in fact held to be simply the release of the victim's spirit—nay, if it was the prescribed mode of releasing that spirit—to seek a new, sometimes a better, abode in a fresh body, we may surely be satisfied that both these have the same origin. We may then, go further, and see in this unspelling incident, performed, as in the Enchanted Princess stories, in this way, at a haunted spot, frequently on a day of special sanctity, one more proof that