- Language and Society among the Insular Celts AD 400-1000 -
and says nothing of the Deissi.^26 The former existence of Irish settlements in western
Britain is apparently common knowledge. The details, however, are far from being
agreed. The period of Irish power on both sides of the Irish Sea is clearly seen as
something in the past by Cormac, as it is by the Welsh sources, but this is not
surprising since the Glossary, in so far as it is the work of Cormac mac Cuilennain
(ob. 908), is a text of the Viking period.27 The whole body of evidence, British and
Irish, gives the impression of an old but muddled tradition, not propaganda newly
devised on behalf of an intrusive dynasty.28
By the tenth century the Irish settlements were receding into a more remote past.
The Harleian Genealogies ignore the anti-Irish role of Cunedda. For them what
matters is land-division, and in particular the way in which, if a son dies before his
father, the grandson can step into his shoes and share the patrimony along with his
uncles. On the other hand, the grandson is to assume the position of the youngest
son, even if he is the son of the eldest son; it is, therefore, his duty to divide the
patrimony and to take the last share.^29 In terms of the succession of one son to an
undivided kingship of Gwynedd -which can co-exist with a territorial division by
which the other brothers receive lordship but not kingship - the grandson of the
26 K. Meyer, 'The expulsion of the Dessi', § I I; Sanas Cormaic: an old Irish glossary, ed. K.
Meyer (Anecdota from Irish Manuscripts, ed. OJ. Bergin et aI., IV: Halle, 1912) (hereafter
Sanas Cormaic (YBL), no. 883; M. Dillon, 'The Irish settlements in Wales', Celtica 12
(1977): I-II; T. OCathasaigh, 'The Deisi and Dyfed', 1-3, 18-28 (on pp. 18-19 the
versions of Rawlinson B 502 and Laud Misc. 612 are both given); on p. 26 he suggests that
the composite material in the Deisi material suggests a further migration to Dyfed from
Leinster.
27 The most likely interpretation is that the shorter version of Cormac, as found in Lebor
Brecc and fragmentarily in Bodleian MS Laud Misc. 610, is at least of Munster origin about
the time of Cormac: d. P. Russell, 'The sounds of a silence: the growth of Cormac's
Glossary', Cambridge Medieval Celtic Studies 15 (Summer 1988): I-30, especially 10-11.
28 Mrs Chadwick, 'Early culture and learning in North Wales', 34-5, maintains that the
Cunedda story is antiquarian speculation based on the genealogies and stimulated by a
new sense of nationalism associated, in Wales, with the reign of Rhodri Mawr. Against this
is the priority in date of attestation of the versions of the story in the Historia Brittonum
as against the version in the Harleian Genealogies: the genealogical version is the later.
Also the versions in the Historia Brittonum both antedate Rhodri Mawr; and, moreover,
the intellectual character of Merfyn's court, in so far as it can be reconstructed, shows no
sign of any rise in nationalism, if one excludes the Cunedda story itself. David Dumville,
'Sub-Roman Britain', 182, sees the point of the story as being (I) an explanation of the
close relationship between Wales and the British kingdoms of what is now southern
Scotland; (2) to provide a parallel to the intrusion of an outside dynasty into the kingdom
of Gwynedd; (3) to explain how the sub-kingdoms of 'Greater Gwynedd' came into
existence. This view rests on the unspoken assumption that all the versions of the story
can be assimilated into a single tradition. My view is that they cannot.
29 The alternative interpretation is that the new king was supposed to assign shares to his
brothers, as he appears to do in the Vita Sancti Gundleii, § I, Vitae Sanctorum Britanniae
et Genealogiae, ed. Wade-Evans, p. 172; this would imply that Meirion succeeded his
father Cunedda, whereas the line of Maelgwn Gwynedd is traced from Einion Yrth,
EWGT,p. II.