DINAS EMRYS AS CAER DATHAL: A LATE FOLKLORE RENAMING OF A FAMOUS HILLFORT IN NORTHWESTERN WALES?

Gelert's Grave at Beddgelert, Gwynedd, Wales

In the past, I have written a great deal about both Dinas Emrys and Caer Dathal, two famous Arfon forts in early Welsh tradition.  The location of the first is well-known, while that of the second is not.  In this article I would like to suggest that both places are, in fact, the same.

As always, names are important.  Dathal is now thought to be an Irish name.  If we have it in its more or less proper form and it is not for another Irish name (Tuathal), then I have this on its meaning from Dr. Jürgen Uhlich, Department of Irish and Celtic languages, Trinity College. Note that there were two similar names, and it is the second one we are interested in:

"Daith + gal and Daith + -(w)al, ‘having quick valour’ (possessive = Bahuvrīhi compound with final noun) and ‘quickly ruling’ (verbal rection/governing compound with verbal root at the end)."

Other Irish onomasticists agree that Dathal is to be derived from daith.  See, for example, Donnchadh o Corrain of University College Cork in his IRISH NAMES has

"DATHAL (do-hal) m. Perhaps from daith, 'swift, nimble'.  An early name which occurs amongst the people of Fermoy."

On Gelert of Beddgelert (for all spellings of that place-name, enter it into the head-name search blank at http://www.e-gymraeg.co.uk/enwaulleoedd/amr/cronfa_en.aspx), I have proposed that it comes from a Latin celer derivative.  Although the story of the hound is merely a folktale, the name itself may be quite significant.  Prof. dr. P.C.H. Schrijver, Departementshoofd Talen, Literatuur & Communicatie, Keltische Talen en Cultuur, Departement Talen, Literatuur en Communicatie, Universiteit Utrecht, responded to my idea thusly:

"I never thought about this before your message, but Celert has a cluster -rt that cannot have arisen in any native British word (old *rt should have become rth, and old *rd should have become rdd). The only reasonable way in which this could have come about is on the basis of late Latin syncope in a Latin word with antepenultimate stress, which then entered Welsh after native *rt had become *rth (so that it could no longer take part in that development; Jackson dates *rt > *rth to the mid to late 6th century). So not celerá:tus but celéritas ‘speed’ or celériter ‘speedily’."

Richard Coates, Professor Emeritus of Onomastics, Bristol Centre for Linguistics, University of the West of England, agrees with Schrijver:

"He is right, especially about the requirement of antepenultimate stress in the Latin source word. Celeritas (and not the oblique form celeritatem, which if an early borrowing would give something like modern *cylerdawd > *cylerdod - compare awdurdod from autoritatem) is formally possible."

Dr. Simon Rodway, Department of Welsh and Celtic Studies, Aberystwyth University, said merely 

"I have nothing to add to what Peter [Schrijver] says.  As far as I can see, he is right."

celeritās ātis, f

celer, swiftness, quickness, speed, celerity

http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0060%3Aentry%3Dceleritas

Cf. Irish daithe, 'swiftness' (eDIL)

If this is correct, then we could argue that Latin Celeritas is merely an effort to translate the Irish personal name Dathal.  And that would mean that the original name of Dinas Emrys next to Beddgelert was Caer Dathal.  Emrys or Ambrosius became connected to the site in another folktale.  Although, Ambrosius is said in Galfridian tradition to be the brother of Arthur's father, Uther.  Dinas Emrys became confused with Amesbury and its nearby Stonehenge in the work of Geoffrey of Monmouth. 

Uther himself was related to the Sons of Iaen ('ice-sheet') at Caer Dathal.  Iaen as a descriptor points to a fort that was prone to be ice-covered in the winter and that only fits places like Dinas Emrys, high up in Snowdonia.  It cannot have been used to describe any of the lowland forts in Arfon, even though some rivers in the valleys below Snowdonia could freeze over (information courtesy NRW fisheries officer Walter Hanks, https://naturalresources.wales/?lang=en).

I have elsewhere shown that Caer Dathal may have been relocated by Geoffrey of Monmouth to Tintagel in Cornwall. 

The Celeritas-Dathal substitution need not seem an outlandish notion.  We have a great deal of evidence that the Welsh royal genealogists sought to cover up Irish roots in several kingdoms.  The Dessi pedigree for Dyfed has ancestors with Irish names being replaced by good old Roman names.  The same is true of Cunedda.  The Irish are present at Dinllaen and the Lleyn Peninsula (Laigin), and the Ui Liathain were in southern Wales.  Brycheiniog was also an Irish-founded kingdom.  

At least one example in stone displays an attempt to render at least part of a king's name in Latin.  By coincidence (?), I am referring to the stone found in Dyfed honoring Votepor.  There "PROTICTORIS and *votep give the same semantic meaning of `shelter, refuge'." See the following website for details:

https://www.ucl.ac.uk/archaeology/cisp/database/. 

I have many times before pointed out the fact that all Arthurs subsequent to the Arthur of Nennius and the Annals belong to Irish-founded dynasties in Britain.  In the past, the only reason I could come up with to account for this was that the original Arthur had himself been at least part Irish.  If Dathal is an Irish name and Uther was related to the men of that fort, then there may well have been Irish in Arthur's ancestry.