BRITTANY AND «ARMES PRYDEIN VAWR »
PAR
David N. DUMVILLE
Since 1916 all study of the mediaeval Welsh prophetic poem Armes Prydein Vawr has been dominated by the views of Ifor Williams. In a celebrated assault on the published ideas of J. Gwenogvryn Evans he argued that the date of composition should be sought ca 900, with an expressed preference for the late ninth century.1 In the forty years which culminated in the publication of his edition in 19552 Williams refined his arguments on the dating and interpretation of the text. His final published view was that the poem must be dated after the mid-ninth century because of its reference to the vikings of Dublin: on this count he could perhaps have added that the years 902x917 were an unlikely period in which to expect activity by that group, since that was the time of their enforced exile from their city.3 A terminus ante quern was provided by the continuing use of the name Glywysing, a political unit believed not to have been so called after the end of the tenth century.4 More precise dating could be suggested by
1. Y Beirniad 6 (1916) 207-12 ; cf. his book, The Beginnings of Welsh Poetry (Cardiff 1972), p. 128, reprinting a paper of 1932.
2. Armes Prydein o Lyfr Taliesin, ed. Ifor Williams (Cardiff 1955). An English version of this book, lightly revised and with an English translation of the poem, was published as Armes Prydein, The Prophecy of Britain, from the Book of Taliesin, edd. Ifor Williams & Rachel Bromwich (Dublin 1972).
3. The dates are provided by the Annals of Ulster, 902.2 and 917.4. One must not, however, overlook the possibility that the Dublin vikings could be so known even during their years of absence from that place.
4. The Morgan ab Owain from whom the successor-state Morgannwg is named ruled 950-74 ; the territorial term presumably dates from that quarter-century or soon afterwards. If Williams’s argument is correct that for the poet Dyfed and Glywysing constituted the whole of south Wales — -if, that is, the poet was not metrically constrained or did not use the figure of pars pro toto — then the dating can be refined by reference to the political disappearance of Ystrad Tywi and Gwent. On this, see Wendy Davies, An Early Welsh Microcosm. Studies in the Llandaff Charters (London 1978), pp. 90-2, 95-8, 100-1, 157. Brycheiniog, the other independent political unit of the period, is lost to view as a kingdom after 934, in which year King Tewdwr attested a charter of Æthelstan of which an original survives : P. H. Sawyer, Anglo-Saxon Charters. An Annotated List and Bibliography (London 1968), no. 425 ; D. P. Kirby, 'Hywel Dda : anglophil ?’, Welsh History Beview 8 (1976/7) 1-13 (at p. 7), thinks that Hywel seized Brycheiniog. Further¬ more, the reference to Alclud (line 151) suggests a date before the kingdom of Strathclyde lost its independence. And gynhon in line 131 ( ?here 'gentiles’, 'pagans’, rather than ‘tribes’), if not traditional usage, implies a reference to the Dublin vikings not later than the early eleventh century when their dynasty became Christian.
