892 REPORT 1900. 



The Ulster men liad a tellef, you see, in the return of the heroes of previous 

 generations to be born again ; but we have here also two social systems face to face. 

 Accordino; to the one to which Ciichulainn as a Celt belonged, it was requisite 

 that he should be the father of recognised oiFspring, for it was only in the person 

 of one of them or of their descendants that he was to be expected back. The 

 story reads as if the distinction was exceptional, and as if the prevailing state of 

 things was wives more or less in common, with descent reckoned according to birth 

 alone. Such is my impression of the picture of the society forming the background 

 to the state of things implied by the conversation attributed to the noblemen of 

 Ulster. Here again one experiences difficulties arising from the fact, that the 

 stories have been built up in the form in which we know them by men who 

 worked from the Christian point of view ; and it is only by scrutinising, as it were, 

 the chinks and cracks that you can faintly realise what the original structure 

 was like. 



Among other aids to that end one must reckon the instances of men being 

 designated with the help of the mother's name, not the father's : witness that of 

 the king of Ulster iu Ciichulainn'stime, namely Conchobar mac Nessa, that is to 

 say, Conor, son of a mother named Nessa ; similarly in Wales with Gwydion son 

 of Don. Further we have the help of a considerable number of ancient inscrip- 

 tions, roughly guessed to date from the fifth or the sixth century of our era, and 

 commemorating persons traced back to a family group of the kind, perhaps, 

 which Cffisar mentions in the fourteenth chapter of his fifth book. Within these 

 groups the wives were, according to him, in common (inter se communes). Take 

 for instance an inscription from the barony of Corcaguiny in Kerry, which com- 

 memorates a man described as ' Mac Erce, son of Muco Doviimas,' where Muco 

 JDovvinias means the clan or family group of Dovvims or Dubin (genitive Duibne), 

 the ancestress after whom Corcaguiny is called Corco-Duibne in Medieval Irish. 

 We have the same formula in the rest of Ireland, including Ulster, where as yet very 

 few Ogams have been found at all. It occurs in South Wales and in Devonshire, 

 and also on the Ogam stone found at Silchester in Hampshire. The same kind 

 of family group is evidenced also by .in inscription at St. Ninian's,in Galloway; and, 

 to go further back — perhaps a good deal further back — we come to the bronze 

 discovered not long ago at Colchester, and dating from the time of the Emperor 

 Alexander Severus, who reigned from 222 to 235. This is a votive tablet to a 

 god Mars Medocius, by a Caledonian Pict, who gives his name as Loseio Veda, 

 and describes himself further as Nepos Vepogeni Caledo, He alludes to no father, 

 and Nepos Vepogeni is probably to be rendered Vepogen's sister's son. At any 

 rate, the Irish word corresponding etymologically to the Latin nepos has that 

 sense in Irish ; and, so far as I know, it has never Ijeen found meaning a nephew in 

 the sense of brother's son. That may serve as an instance how the ideas of 

 another race penetrated the fabric of Goidelic society ; for here we must suppose 

 a time to have come when there was no longer any occasion for a word meaning a 

 brother's son, which, of course, there never was in the non-Celtic society which 

 ranked men and women according to their birth alone. 



Now this Caledonian Pict was not exceptional among his kinsmen, for they suc- 

 ceeded in observing a good deal of silence concerning their fathers down, one may 

 say, to the 12th century. It is historical that the king of the northern Picts was 

 not wont to be the son of the previous king. In short, when the Celtic elements 

 there proved strong enough to ensure that the son of a previous king should 

 succeed, a split usually took place, the purer Picts being led by the rule of succession 

 by birth to set up a king of their own. The fact is not so well known that the 

 same succession prevailed also some time or other at Tara in Ireland ; it is proved 

 by a singular piece of indirect evidence, the existence of a tragic story to explain why 

 ' no son should ever take the lordship of Tara after his father, unless some one 

 came between them.' The last clause is due, I should say, to somebody who 

 could not understand such a prohibition based on the ancient rule that a man's 

 heir was his sister's son. This would be, according to Irish legend, in the lifetime 

 of Conor mac Nessa. 



It is curious to notice hew the stories about the Pictish minage seem to have 



