Page:Researches respecting the Book of Sindibad and Portuguese Folk-Tales.djvu/156

 2 and minute points of contact that unite the various Oriental texts with each other disappear almost completely. The relationship between the two groups, and even more precisely the actual derivation of the Western from the Eastern, are easily recognizable; but it is impossible, as in the Eastern versions, to trace out and identify any one particular book as the original. There is no Eastern version which differs so much from the others as the whole Western group differs from the Eastern group, whether it be in the form of the fundamental story or in the tales inserted in it, of which scarcely four are common to both groups. It is all very well to attempt to assimilate these Western texts with one or another of the Eastern texts, as Loiseleur and others have done. The profound differences in question will never be explained by the supposition of intermediate links having been lost, so long as the remark is applied to written versions. Only oral tradition transmutes the contents of a popular book in that manner, and it is that certainly which stands between the Eastern and the Western groups. It should be remarked that the Western texts have given rise also to oral traditions, and from these certainly proceeds the tradition which has been found still existing in the mouth of the people in Hungary. However evident be the derivation of the latter from the Western texts, yet the variation is such that if we were still in the Middle Ages, and a monk or a minstrel got hold of it, and, completing it out of his own head, should make a book of examples or tales out of it, this book would differ as much from all the Western texts as they do from the Eastern.

The Western versions are, therefore, secondary offshoots of the ancient Indian book. They belong, in fact, to this family, and have a place in its history; but of it they