THE STRONGEST LINGUISTIC ARGUMENT FOR ELIWLAD SON OF MADOG = MATOC AILITHIR

Eliwlat in Bodleian Library MS Jesus College 20

Eliwlad in Cardiff MS. 2.83

I've been promising this piece for some time, and felt compelled to finally "get it out there."

The linguistic support from top Celtic linguists for Eliwlad as some form of a word denoting 'pilgrim', from elements meaning 'other land', has been quite favorable.  See, for example, the relevant portion of https://mistshadows.blogspot.com/2019/12/a-fairly-radical-revision-of-my-earlier.html.

However, one major stumbling block keeps cropping up - and this is one I actually thought of myself.  It bothered other scholars as well and was, perhaps, best stated by Professor Barry Lewis of Dublin:

"The name Eliwlad, used for a son of Uther's son Madog, has never been explained, so far as I know. An adaptation of Irish ailithir is possible, but one issue that gives me pause is that, if the borrower understood Irish tír as `land', why did he not use the identical Welsh word tir rather than gwlad? I am not sure how to get round this problem.

It is an interesting coincidence, though, that there is a Matóc [i.e. Madog] Ailithir attested in Ireland."

I do not wish to downplay this "problem."  It is significant.  And it is so despite the fact that Eli- itself has been shown to be a perfectly normal Welsh rendering of an Irish aile by none other than Professor Peter Schrijver and colleagues.

So is there any good way around the problem?

Well, we could opt for proposing a purely theoretical Irish word (not now extant, but possible, according to Professor Jurgen Uhlich) *aile(f)laith or, with regular unstressed processing, *ailelaid.  But this is a very weak case for Eliwlad.

Although we have a fairly late W. allwlad, 'other land', which according to Dr. Simon Rodway of The University of Wales could be from an earlier, unrecorded compound, we cannot derive Eliwlad from this word.

And this brings us back to what Dr. Richard Coates at the University of the West of England at Bristol shared with me:

“It looks perfectly possible to me that Eliwlad represents British *Aljowlatos 'other land'.  Eliwlad/t is a plausible rendering of Eilwlad. One certainly finds occasional <e> for <ei> in MW, and metathesis is always possible. If it’s not from *aljo-, I have no idea.”

A correction from Dr. Rodway: "<e> for /ei/ is quite common in Old Welsh, but not in Middle Welsh.  Where it does occur in MW it is probably either due to scribal error or evidence for an OW exemplar which has not been correctly modernized." [1]

When I asked Dr. Rodway whether such a metathesis could have occurred, he responded:

"Well, i and l are reasonably similar, especially if a hair-stroke dot on an eye is not read by a copyist.  Off the top of my head, I can’t think of an example, but it might have happened."

This idea from Coates is really rather brilliant.  We begin with a name that would have been spelled Eilwlad quite early on in Welsh.  Ail (or eil) originally shared a meaning identical with Welsh all and Early Irish aile:

other, second *aljo-, SEMANTIC CLASS: measure, Celtiberian ailam ‘f (As)’, Gaulish alios ‘other’, Early Irish aile ‘other’, Scottish Gaelic eile ‘other, another’, Welsh ail ‘second; like, similar, comparable with; son, grandson, heir; race’, Cornish eyl ‘one of two’, Breton eil, il (Old Breton), eyl (Middle Breton), eil ‘other’

other *allo-, SEMANTIC CLASS: measure, Gaulish alla; allos; Allo- ‘other; second’, Early Irish all- ‘other’, Welsh all- ‘other’, Breton all ‘other, next (time)’

We need only assume one simple error occurred in the transmission of the hypothetical name Eilwlad.  It became Eliwlad through metathesis.  

Dr. Rodway is willing to accept this only if Eliwlad was a 'one off' for Eilwlad.  Yet there are many 'one off' Celtic names that cannot be properly etymologized. And the poem is admittedly late and contains mistakes.  To quote from Nerys Ann Jones' ARTHUR IN EARLY WELSH POETRY on "The Dialogue of Arthur and the Eagle", p. 168:

"Its first editor, Ifor Williams, has shown that although its exemplar was no older than the thirteenth century, some of its linguistic forms... suggest a manuscript tradition going back at least another hundred years.  The poem was written using medieval Welsh spelling but some errors indicate that the scribe was struggling with an orthography and also possibly a script that was alien to him."

Given that this is so, I asked Dr. Rodway if this might make an Eliwlad from Eilwald more attractive.  He replied "Perhaps, if you can show that all extant references stem from this one exemplar."  Well, we can't do that with any certainty.  However, if the error occurred early enough and in just the once source, the incorrect spelling Eliwlad for Eilwlad could easily have become the standardized form.  The authorities agree that Eliwlad/Eliwlat is the early spelling.  We also find Liwlad, Liwlod, but they are later.

I would submit, therefore, the Eliwlad was originally Eilwlad and that this was a Welsh name or word equivalent to Irish ailithir, and that the son of Uther Pendragon is the son of Sawyl Benisel.

The degree to which we must allow a name alternation here is extremely minor - especially compared to the major and often unjustifiable emendations offered by top Welsh scholars for many words found in early literary sources, e.g. the poems attributed to Taliesin.  

If we can allow this one metathesis, and then factor in the nice contextual fit as I've described in detail in previous research, I think we cannot ignore the real possibility - or plausibility? - that the Terrible Chief-dragon, father of Arthur, is Sawyl Benisel of the Sarmatian Roman fort at Ribchester.

[1]

If /ei/ and become /e/, we must consider one other possibility for Eliwlad.  Dr. Simon Rodway actually prefers the later spelling Eliwlod or Eliwlad.  This would necessitate a -llawd terminal.  Lladd (strike, slay) won't work because it ends in /dd/.

llawd, from Proto-Celtic *lāto-, Early Irish láth ‘heat, rut’, Welsh llawd ‘heat (of sow), sow’s desire for boar, sow’s mating season’,

llawd [in the GPC]

[Gwydd. C. láth ‘awydd rhywiol anifail’: < Clt. *lāto-, Islandeg Diw. lóða, cf. Wcraineg lit; ceir yr un elf. yn aelawd, llodig, trallod (tra-llawd); daw’r geiriau aeled, anllad, lled4, llid o bosibl o’r un gwr.]

eg. a hefyd fel a.

Awydd hwch am faedd, y tymor pan fo hwch yn ei gwres:

heat (of sow), sow’s desire for boar, sow’s mating season. 

And from Matasovic's ETYMOLOGICAL DICTIONARY OF PROTO-CELTIC:

*lato- 'furor, ardor [Noun]
GOlD: OIr.lath [o m]
W: W llawd [m] 'heat (of sow), sow's desire for boar'
SEE: *layko- 'warrior'
ETYM: The first element in the Gaulish Ethnonym Lato-bici might be
derivable from this Celtic etymon (Delamarre 198, Meid 2005: 54). W llawd
is attested very late (l7th century) and shows somewhat unusual semantic
development.
REF: GPC II: 2106, Delamarre 198, Meid 2005: 54, Zimmer 2000: 288.

Lawdd, according to Rodway, "obviously could denote passions other than lust, as can be seen by the semantic range of compounds like trallod, etc." But in opting for -lawdd in Eliwlod, we are restricted to two possible first elements:

eilyw1, eiliw2,

eg. a hefyd fel bf. 3 un. pres. myn.

Tristwch, galar, gofid, poen; pair ofid, poena:

sadness, grief, pain; he (she, it) grieves, pains. 

eiliw1 [found spelled

[< Brth. *ad-līu̯o-; fel y ceir rhif ac eirif, felly lliw ac eiliw]

eg. ll. eiliwiau, eiliwoedd, a hefyd fel a.

Golwg, gwedd, ymddangosiad, cyflwr, ffurf, llun, delw; lliw, arlliw, paent; llewyrch, ôl, argoel, rhithyn; tebyg, hafal, cyffelyb, cyfliw (â), unlliw (â):

appearance, aspect, guise, state, form, shape, figure, image; colour, hue, paint; glimmer, trace, vestige, sign; like, similar, of the same colour or hue (as). 

I cannot make anything intelligible in terms of a personal name out of the various combinations of these elements.  Unless we go with something like '[He of] lustful aspect.' Dr. Rodway thinks this eiliw "is reasonably close, but not a match.  "

It is just the semantics.  To my mind ‘passion pertaining to sadness/grief, i.e. grieving/sad passion’ seems more likely than ‘passion pertaining to form’ or some such.  In close compounds, it is the first element which modifies the second, not the other way round. "

The same is true for my last idea for Eliwlod, which connects this personage with Lleu in his oak tree.  In the englyns of MATH SON OF MATHONWY, we are told of the 'ulodeu lleu' of 'flowers of Lleu.'  This have been thought to be figurative of feathers, although they have also been related to Lleu's wife Blodeuedd, 'flowers'/Blodeuwedd, 'flower aspect.'  I proposed to Dr. Rodway eiliw, 'form, appearance, aspect', etc., plus blawd, 'flower'.  His opinion?

"I think formally this is possible (lenited b disappearing in a consonant cluster, cf. lledrith < lledfrith).  However the semantics are challenging.  It couldn’t mean ‘flower aspect’, because in compounds, as I’ve already said, it is the first element that qualifies the second, thus ‘aspect flower’."

For this reason he instead prefers the other eiliw, writing to me the following:

"This is my current thinking on Eliwlod/lad.

Perhaps a compound of eiliw ‘sadness, grief’ + llawd ‘passion, lust etc.’ which has undergone some analogical remodelling at various stages. (1) eil- > el-, perhaps due to failure to correctly modernize an Old Welsh form with e for /ei/ and under the influence of the many names in El- (Elidir, Elfed, Elgan etc.). (2) –wlod reinterpreted as (g)wlad ‘country etc.; prince’ in the poem about Arthur and Eliwlod (but not necessarily in the original as the name does not appear to rhyme), and thence to Bleddyn Fardd (where it does rhyme).  The original form of the second element (with llawd regularly reduced to –lod in a polysyllabic name) survived to surface in the late poetry.

This is rather speculative, but seems possible."

I would point out that the Irish cognate of Welsh llawd, lath, had the meaning of warrior (see the eDIL listing for the relevant word).  If the Welsh llawd ever conveyed such a meaning, we could propose something like Eiliw-llawd, 'with a warrior's aspect', for Eliwlod.  However, there is no evidence whatsoever that this meaning for lath was anything other than a peculiarly Irish development.  Dr. Rodway agrees with me on this point.

And, once again, I believe all of this to be a moot point, as I take Eliwlad to be the earlier spelling of the name.

Eilwlad as a dim folk memory or reflection of Ailithir is still the most promising etymology, even if the required metathesis is deemed philologically improbable.

I also don't think we can ignore the fact that Eliwlad is a spirit in the form of an eagle in an oak tree. Whether the motif was copied from that of Lleu in the Mabinogion is uncertain, but the strong possibility is there. In other words, the apparition belonged to the Otherworld, and as Professor Stefan Zimmer has remarked to me, use of a name which like Ailithir meant 'other land' may well have suggested to the composer of the tale that Eliwlad belonged to or came from such a spiritual place.

ADDENDUM:

See the new post on this subject at https://mistshadows.blogspot.com/2020/01/precedence-is-all-eil-found-spelled-eli.html.