be mainly historical, with later romantic accretions; others, like myself, have held it to be mainly ancient myth re-crystallised around a third-century name. According to Professor Zimmer, Finn was an early ninth-century half-Viking, half-Irishman, an opponent of the Dublin Danes, by whom he was slain. Let us first see what this theory postulates respecting the Irish records that have come down to us: (a) That the genuine history of the ninth century, in so far as this hypothetical struggle of Finn against the Danes is concerned, was practically left unrecorded in its chronological place save for the chance entry of the defeat and death of Caitill Find; (b) that the main elements of this history were used as the basis of an historical romance the date of which was thrown back five hundred years; (c) that the statements of this romance reacted upon the genuine historical record of the third-tenth centuries. No reason is assigned for this process, and the only explanation of it vouchsafed is that the genuine history of the early ninth, was strikingly like the genuine history of the third century, so that confusion was made possible.
It is difficult to find in English history an analogy to the process postulated. If, however, we can imagine the stories about Alfred's resistance to the Danes being carried back to the fifth century, and thus originating the story of Arthur's resistance to the Saxons, we can form some idea of what Professor Zimmer assumes to have taken place in Ireland. The analogy, of course, halts in this, that the men who sang of Alfred were of other race than those who told of Arthur; but it will serve the purpose.
That the Irish could create pseudo-history on a large scale we know, but we also know why they did it. The annalists of the eighth and succeeding centuries were Christian monks, and they could only conceive mythic tradition as pseudo-history. It was inevitable that they should euhemerise the national mythology. They obeyed exactly the same impulse as Nennius or Saxo Grammaticus. But
