measurement said to have been adopted by St. Patrick for the monasteries built under his direction at Ferta.

According to Miss Stokes, it is always easy for an experienced eye to distinguish between the ecclesiastical and the military cashels in Ireland.

Whether, in Wales, there were any conversions of stone forts into monastic settlements I do not know. In Cornwall there must have been something of this sort. St. Denys is a church planted in the midst of a dinas. When the church was re-consecrated by one of the mediæval bishops of Exeter, he dedicated it to St. Denys, through misconception of the original name, Landinas. At Hellborough, a stone caer near Camelford, is a chapel to St. Itha, the Bridget of Munster, on a cairn in the midst; and St. Petrock's at Lydford is in the midst also of a strongly fortified cliff castle.

In Wales, the stone hut circles are attributed by tradition to the Irish Gwyddels. May not the stone fortresses there be also due to them? They occur in those parts of North and South Wales that were overrun by the Irish. It is greatly to be desired that Tre'rceiri, Carn Ingli, and Carn Gôch, at least, should be thoroughly explored, to settle the many questions that are asked concerning these castles. But, unfortunately, the digging out of a camp is a peculiarly costly work; and for such undertakings money is not readily forthcoming. Nevertheless, it is to be hoped, that some day pick and shovel will force from them their story.

EARLY SOCIAL LIFE IN WALES.'

BY

DAVID BRYNMÔR JONES, Q.C., M.P.

SUCH exaggerated notions have prevailed as to the antiquity of the Cymric race that it is necessary to make an observation as to the use of the word "early" in the title I have given to this paper. The period in my view covered by that term is the time that elapsed from the first emerging of our race as a separate nation or state after the departure of the Romans, to the Norman conquest or a little later. I say the emerging of our race as a separate nation because the tribes which joined together under Cunedda and his successors to resist the Teutonic invaders had been for many years under the rule of the Roman Empire. Each of them had had of course its own history, though very little is known about that of any of them. So far as I can find out no one of the tribes in the island called themselves Cymry. The word "Cymro" means compatriot, and only came into use after the legions had departed, and the island was left to defend itself as best it could. It looks as if it was employed to designate the Celtic tribes and kindreds who acknowledged Cunedda as their leader after he had conquered North Wales. If this be so, the Cymric kingdom is not very ancient, and it was only in the fifth century that the Cymry began to regard

Read before the Honourable Society of Cymmrodorion, at 20 Hanover Square, on Thursday, the 23rd of February, 1899; Chairman, Sir John T. D. Llewelyn, Bart., M.P.

themselves as one people. I therefore feel justified in applying the word "early" to the life in Cymru from say the fifth to the eleventh century.

Now it so happens that there is ample material of trustworthy character for the construction of a picture of life in Wales during these centuries. The two principal sources of information are the Ancient Laws and Customs of Wales, published under the editorship of Aneurin Owen, under the auspices of the Government, in 1841, and the works of Giraldus Cambrensis.

Of the legal treatises the most helpful are the so-called three codes (supposed to have been in use in Gwynedd Deheubarth and Gwent respectively) and the Latin versions. All these books derive their origin from a tenth century compilation which was known as "hen lyfr y Ty Gwyn," in which the laws were set down in writing at, or as a result of, an assembly convened by Howel Dda, who was king of a large part of Cymru for many years (907-950). From them it is possible to give an outline of the Cymric legal system. As to the works of Giraldus I speak below. It will be observed that both these sources are comparatively late; but in the times with which we are dealing laws, customs, and habits of life changed only very slowly. The general complexion and leading features of Welsh society and character were much the same in the sixth as in the tenth, and in the tenth as in the twelfth century. Progress there was; but not progress at such a rate as to involve any essential or revolutionary change.

I propose first to sketch briefly the political and legal organisation of the Cymry; and next to give some account of their way of life and their national characteristics. I will commence with the ancient divisions of Cymric land.

Cymru was divided into districts called cantrefs and cymwds. The exact significance of the cantref it is very

difficult to determine, for in the laws of Hywel Dda it is the cymwd which is the unit of organisation. In the time of Hywel the boundaries of the cantrefs and cymwds were evidently known and settled for practical purposes. For the purposes of government from day to day the cymwd is the area on which one must fix one's eye. The cantref, as it then existed, was in all probability a district over which a lord (arglwydd), appointed by the king of the country (gwlad) of which it formed part, ruled with a set of officers whose rights and duties corresponded with those of the king. The lord of a cantref or cymwd must not be confounded with another kind of chieftain, the head of a kindred (cenedl), with whom the laws make us acquainted. The lord might, of course, be a penkenedl in reference to his own kindred, but his position as arglwydd was due, as it would seem, to his appointment by the king of, or the royal kindred ruling over, the country in which the cantref or cymwd was situate. Sometimes several cantrefs were combined under one lord, who called himself tywysog (prince) or brenin (king), but in any case, if we may judge from the laws, each cymwd and cantref maintained its separate organisation. The lord delegated to certain officers the discharge of some of his functions. In every cymwd there was a maer (in the Latin text, præpositus) and a canghellor (in the Latin text, cancellarius), discharging prescribed governmental duties, and in each cymwd a court was held by them with the aid of other officers.

As might have been expected, the Codes disclose communities containing different classes of person, or castes. Speaking broadly, braint (status) depended on birth. The primary distinction is between tribesmen and non-tribesmen, between men of Cymric and those of non-Cymric blood. The Cymry themselves were divided into: (1) a royal class consisting of men belonging to families or

kindreds (cenedloedd) of kingly or princely braint (status) who had over divers areas of Cymru special rights; (2) a noble class called in the codes sometimes uchelwyr (literally, "high-men"), sometimes breyr, sometimes gwyrda, and in the Latin versions nobiliores and optimati; and (3) innate tribesmen styled boneddigion (gentlemen).

Below the tribesmen in the scale were unfree persons denominated taeogion or eilltion (in Latin, nativi or villani), corresponding roughly to the villeins of English law. Lowest of all was a class of menial or domestic slaves (caethion).

But quite apart from these-the primary classes contemplated-forming the legal organisation, the laws deal with strangers residing temporarily in or settling within the limits of a Cymric area. Such strangers were called alltudion, and though there was some similarity in the position of the two classes, they must not be confounded with the eilltion.

The degree of the alltud in his own country made no necessary difference to his position in the Cymric system. If a Mercian, whether noble or non-noble, settled in Gwynedd, he was in either case an alltud. For the individual, the line that separated him and the Cymro could not originally be passed.' But there is evidence to show that, in regard to South Wales, the residence in Cymru of an alltud and his descendants continued till the ninth generation conferred Cymric status upon the family; and also that intermarriage with innate Cymruesau generation after generation made the descendants of an alltud innate Cymry in the fourth generation. Late texts give also examples of artificial methods of securing Cymric kinship,

It would seem, however, that if the king conferred office on him, he assumed the braint (status, privilege) attaching to it.