to a Semitic root which gives a commonplace explanation. One explanation we called a myth, because it is inserted in what would usually be called a myth, and because it takes the form of a story; the other we call a comment, because it stands at the foot of a page under Homer's text and because it involves no picturesque events. But the mental processes are the same: both the Englishman and the Fijian have lost the tradition which is the key to the problem; like any one else they are worried by any fact that stands apart and will not fall into their general scheme of things, and, like everyone else, in default of a natural explanation which gently absorbs the fact, they will push it and squeeze it till it fits into a place that is not meant for it. In both cases we have loss of meaning, resulting in a totally new meaning; where they differ is in their ideas of evidence and in the latitude which they consider they may lawfully allow to their imaginations.
Loss of meaning and consequent misconstruction play a great part, I believe, in the building up of myths; I do not say the sole part; I am not concerned here with the processes that may conceivably have helped to produce myths, but only those which can be definitely proved to have taken part. I have often been asked, "Don't you think myths may have arisen in this way, or in that way?" But I am still waiting for examples, and until examples are forthcoming it is fruitless to discusss these supposed ways in which myths may arise. You can do so much with a "may," that there ceases to be any fun in the game; it is far more amusing to trace what actually did happen. Now loss of meaning is one of the things that has happened, and pretty often too. We can readily see why, for customs are continually decaying, and with each one that passes away the meaning of some old tradition is lost; for a narrative always assumes that certain customs, or beliefs, or events are known to the audience; the speaker cannot stop at every turn to expound them; he takes them for granted;
