ordeals, rivalling in severity those of the American Indians, and not improbably genuine though exaggerated reminiscences of actual tests of endurance and agility. Once admitted he had to observe certain geasa or "tabus," e.g. not to choose his wife for her dowry like other Celts, but solely for her good manners, not to offer violence to a woman, not to flee when attacked before less than nine warriors, and the like.
All this may represent some genuine tradition with respect to a warrior band, with many exaggerations in details and numbers. Some of its outstanding heroes may have had names derived from or corresponding to those of the heroes of an existing saga. But as time went on they became as unhistorical as their ideal prototypes; round their names crystallised floating myths and tales; things which had been told of the saga heroes were told of them; their names were given to the personages of existing folk-tales. This might explain the great divergence between the "historical" and the romantic aspects of the saga as it now exists. Yet we cannot fail to see that what is claimed as historical is full of exaggeration, and, in spite of the pleading of Dr. Hyde and other patriots, little historic fact can be found in it. Even if this exists, it is the least important part of the saga. What is important is that part—nine-tenths of the whole—which "is not true because it cannot be true." It belongs to the region of the supernatural and the unreal. But personages, nine-tenths of whose actions belong to this region, must bear the same character themselves, and for that reason are all the more interesting, especially when we remember that the Celts firmly believed in them and in their exploits. A Fionn myth arose as all myths do, increasing as time went on, and the historical nucleus, if it ever existed, was swamped and lost. Throughout the saga the Fians are more than mere mortals, even in those very parts which are claimed as historical. They are giants; their story "bristles with the supernatural"; they are the ideal figures of Celtic legend throwing their gigantic shadows upon the dim and misty background of the past. We must therefore be content to assume that whether personages called Fionn, Oisin, Diarmaid, or Conan, ever existed, what we know of them now is purely mythical.
Bearing in mind that they are the cherished heroes of popular fancy in Ireland and the Scottish Highlands, we have now to inquire whether they were Celtic in origin. We have seen that the Celts were a conquering people in Ireland, bringing with them their own religion and mythology, their own sagas and tales reflected now in the mythological and Cúchulainn cycles, which found a local habitation in Ireland. Cúchulainn was the hero of a saga which flourished more among the aristocratic and lettered classes than among the folk, and there are few popular tales about him. But it is among the folk that the Fionn saga has always been popular, and for every peasant who could tell a story of Cúchulainn a thousand could tell one of Fionn. Conquerors often adopt beliefs, traditions, and customs of the aboriginal folk, after hostilities have ceased, and if the pre-Celtic people had a popular hero and a saga concerning him, it is possible that in time it was accepted by the Celts or by the lower classes among them. But in the process it must have been completely Celticised, like the aborigines themselves; to its heroes were given Celtic names, or they may have been associated with existing Celtic personages like Cumal, and the whole saga was in time adapted to the conceptions and legendary history of the Celts. Thus we might account for the fact that it has so largely remained without admixture with the mythological and Cúchulainn cycles, though its heroes are brought into relation with the older gods. Thus also we might account for its popularity as compared with the Cúchulainn saga among the peasantry in whose veins must flow so much of the aboriginal blood both in Ireland and the Highlands. In other words, it was the saga of a non-Celtic people occupying both Ireland and Scotland. If Celts from Western Europe occupied the west of Scotland at an early date, they may have been so few in number that their own saga or sagas died out. Or if the Celtic occupation of the West Highlands originated first from Ireland, the Irish may have been unable to impose their Cúchulainn saga there, or if they themselves had already adopted the Fionn saga and found it again in the Highlands, they would but be the more attached to what was already localised there. This would cut the ground from the theory that the Fionn saga was brought to Scotland from Ireland, and it would account for its popularity in the Highlands, as well as for the fact that many Fionn stories are attached to Highland as well as to Irish localities, while many place-names in both countries have a Fian origin. Finally, the theory would explain the existence of so many Märchen about Fionn and his men, so few about Cúchulainn.
Returning to the theory of the historic aspect of the Fians, it should be noted that, while, when seen through the eyes of the annalists, the saga belongs to a definite historical period, when viewed by itself it belongs to a mythic age, and though the Fians are regarded as champions of Ireland, their foes are usually of a supernatural kind, and they themselves move in a magic atmosphere. They are also brought into connection with the unhistoric Tuatha Dé Danann; they fight with them or for them; they have amours with or wed their women; and some of the gods even become members of the Fian band. Diarmaid was the darling of the gods Oengus and Manannan, and in his direst straits was assisted by the former. In all this we are in the wonderland of myth, not the terra firma of history. There is a certain resemblance between the Cúchulainn and Fionn sagas, but no more than that which obtains between all sagas everywhere. Both contain similar incidents, but these are the stock episodes of universal saga belief, fitted to the personages of individual sagas. Hence we need not suppose with Professor Windisch that the mythic incidents of the Fionn saga are derived from the Cúchulainn cycle.
The personages against whom Fionn and his men fight show the mythic nature of the saga. As champions of Leinster they fight the men of Ulster and Connaught, but they also war against oversea invaders—the Lochlanners. While Lochlann may mean any land beyond the sea, like the Welsh Llychlyn it probably meant "the fabulous land beneath the lakes or the waves of the sea," or simply the abode of hostile, supernatural beings. Lochlanners would thus be counterparts of the Fomorians, and the conflicts of the Fians with them would reflect old myths. But with the Norse invasions, the Norsemen became the true Lochlanners, against whom Fionn and his men fight as Charlemagne fought Muhammadans—a sheer impossibility. Professor Zimmer, however, supposes that the Fionn saga took shape during the Norse occupation from the ninth century onwards. Fionn is half Norse, half Irish, and equivalent to Caittil Find, who commanded the apostate Irish in the ninth century, while Oisin and Oscar are the Norse Asvin and Asgeirr. But it is difficult to understand why one who was half a Norseman should become the chosen hero of the Celts in the very age in which Norsemen were their bitter enemies, and why Fionn, if of Norse origin, fights against Lochlanners, i.e. Norsemen. It may also be inquired why the borrowing should have affected the saga only, not the myths of the gods. No other Celtic scholar has given the slightest support to this brilliant but audacious theory. On the other hand, if the saga has Norse affinities, and if it is, in origin, pre-Celtic, these may be sought in an earlier connection of Ireland with Scandinavia in the early Bronze Age. Ireland had a flourishing civilisation then, and exported beautiful gold ornaments to Scandinavia, where they are still found in Bronze Age deposits.507 This flourishing civilisation was overwhelmed by the invasion of the Celtic barbarians. But if the Scandinavians borrowed gold and artistic decorations from Ireland, and if the Fionn saga or part of it was already in existence, why should they not have borrowed some of its incidents, or why, on the other hand, should not some episodes have found their way from the north to Ireland? We should also consider, however, that similar incidents may have been evolved in both countries on similar lines and quite independently.
The various contents of the saga can only be alluded to in the briefest manner. Fionn's birth-story belongs to the well-known "Expulsion and Return" formula, applied to so many heroes of saga and folk-tale, but highly elaborated in his case at the hands of the annalists. Thus his father Cumal, uncle of Conn the Hundred Fighter, 122-157 A.D., wished to wed Muirne, daughter of Conn's chief druid, Tadg. Tadg refused, knowing that through this marriage he would lose his ancestral seat. Cumal seized Muirne and married her, and the king, on Tadg's appeal, sent an army against him. Cumal was slain; Muirne fled to his sister, and gave birth to Demni, afterwards known as Fionn. Perhaps in accordance with old matriarchal usage, Fionn's descent through his mother is emphasised, while he is related to the ancient gods, Tadg being son of Nuada. This at once points to the mythical aspect of the saga. Cumal may be identical with the god Camulos. In a short time, Fionn, now a marauder
