CAER DATHAL AND THE IRISH PROBLEM

Llyn Nantlle Uchaf

When we look at the 'vythneint elei' of the PA GUR poem and settle on 'raptors (?) of Ely [River in Glamorgan]' as the proper translation, the identification of Uther as servant of Mabon - one of those raptors - becomes rather easy.

BUT (and there is always a "but", it seems in Arthurian research!)... such an identification of Uther (with St. Illtud) is rendered profoundly more difficult when we take two factors into consideration.

A.

1) The "Stanzas of the Graves" places the god Mabon's grave in Nantlle, the location of Caer Dathal ( = Caer Engan).

2) From Note 288, Page 96 of Rachel Bromwich and Daniel Simon Evan's CULHWCH AND OLWEN: AN EDITION AND STUDY OF THE OLDEST ARTHURIAN TALE, The University of Wales, 1992:

"Gwynn Gotyuron: Gwin Godybrion is found in Pa Gur.  In a corrupt form he appears as Gwyn Goluthon among the sons of Iaen [Hanesyn Hen tract]."

The Sons of Iaen, of course, resided at Caer Dathal.  

3) Uther was said to have relatives at Caer Dathal in Culhwch and Olwen.

4) Arthur took a wife (Eleirch) from Caer Dathal, according to the Hanesyn Hen tract.

5) Kysceint son of Banon (corrupted to Iscawyn son of Panon) is properly Cysteint, 'Constantius', a name associated with Segontium/Caernarvon, a dozen kilometers north of Nantlle.  Another son of Iaen at Caer Dathel - Siawn - is in the Stanzas of the Graves buried at Hirael hard by Bangor, just a little farther north from Caernarvon.  

6) Eliwlad the eagle son of Madog son of Uther, is said to be in the wooded glen of Cornwall (Cernyw).  This has been identified as the Bodmin valley.  However, the death-eagle in an oak is a motif borrowed directly from the god Lleu as death-eagle in an oak at Nantlle.  And Cutmadoc in Bodmin is the exact Cornish equivalent of Coed Madog in Nantlle.  My feeling is that Eliwlad's placement in Cornwall may be due to the CANU HELEDD's Eagle of Eli, as Eli is a place-name found in the heart of the old Cornovii kingdom (see https://mistshadows.blogspot.com/2022/08/the-location-of-eagles-eli-another.html). Cunorix/Cynric son of Maquicoline/Ceawlin/Cunedda was buried at Wroxeter in Cornovii, the later Powys.  Or Eliwlad was just transferred to Cornwall because Arthurian associations there became common during the development of Arthurian legend.  

7) I've written before, rather extensively, on the Galfridian Tintagel being a relocation for Caer Dathal.  The linguistic argument, based on early forms and possible development of the Tintagel place-name, is quite reasonable (see https://mistshadows.blogspot.com/2022/06/notes-on-cornish-place-name-tintagel.html).

Conclusion:

Other than the PA GUR reference to Elei, no other source places the vyth[n]eint or Uther (or Arthur, for that matter) in that location.  

B.

We must be able to account for why the subsequent Arthurs in Dyfed and Dalriada belonged to Irish-descended dynasties in Britain.  We cannot do this if we opt for St. Illtud as Arthur's father.  On the other hand, if Caer Dathal were inhabited by the Irish or by Hiberno-British, we can account for this.  I have elsewhere suggested this fort may have belonged to Cunedda.  

DISCUSSION

To make vythneint elei work for us as a corruption of gwyth nant lleu, "the battle-furious of Nantlle", there are some conditions that must apply.

First, the /e/ of elei must be accounted for.  This is not so difficult, as in one of the two MSS. containing CULHWCH AC OLWEN the scribe has rendered Nant Lleu as 'Nant y Llew' ("Valley of the Lion"). 

From Dr. Simon Rodway:

"The '6' character in medieval Welsh MSS is sometimes transcribed as w and sometimes as v. Different scribes used it in different ways to represent the sounds we now write as w and u in Welsh."

https://storymaps.arcgis.com/stories/ca7687a56bb343ab9716af16db3f80dc:

Also, the PA GUR line in question has to rhyme in -i.  We can see the rhyme scheme for the relevant lines of the poem:

Ym ty ny doi.

Onysguaredi.

Mi ae guardi.

Athi ae gueli.

Vythneint elei.

Assivyon ell tri.

Of course, this question is moot if we simply propose that elei was incorporated into the poem as an error.  But we also possess (admittedly late spellings) of Nantlle which show -y and -i endings (these may be found in the Melville Richards Archive), e.g. Nantlley, Nantlli.  As these are post-medieval, they aren't of much value.  I only mention them because it shows a tendency in the later period, at least, to alter the ending of the place-name in such a way as to accord with the PA GUR's end-rhyme requirement for elei. 

We need a five-syllable line to match the other lines of the poem.  Again, once the elei form was adopted by the author, the five-syllable requirement was met.

Perhaps the biggest barrier of all is the problem of lenition.  In other words, can we justify the mutation of gwyth to wyth? 

Welsh scholars seem to have no problem with this possibility.  The following two notes are from Sir Ifor Williams' Taliesin poems and Nerys Ann Jones' ARTHUR IN EARLY WELSH POETRY:

Prof. Peter Schrijver explained this problem to me thusly:

"In Old Welsh, unlenited gw- (which in MW is spelled gw, gu, gv) was initially spelled as u-, sometimes uu-. It is conceivable that if an OWelsh exemplar of the poem existed and the MWelsh scribe copied from that, he may have failed to modernize the spelling in this particular instance, spelling v- rather than gv- (gu-); perhaps because he no longer understood what the word/line meant."

My next blog post will focus on Caer Dathal, Cunedda, and why it is, exactly, that the move away from Illtud on the Ely should be preferred.