According to a news story, the history of the Basque language has been pushed back centuries:
Archaeologists have unearthed inscriptions in the Basque language that could date from as early as the third century, a find Basque linguists hailed as extraordinarily important…
Until now, a text written by a monk in both Castillian Spanish and Basque had been the oldest written example of the language, dating from the year 1040.
The new inscriptions were found at a Roman site near the Basque town of Vitoria in northern Spain, and included the names of colours, verbs and references to God, Christianity and the Holy Family etched into bricks, bones and pieces of glass.
The head of the excavation, Eliseo Gil, said the pieces would not be dated exactly until October or November, but members of the Academy of the Basque Language, Euskaltzaindia, said the find was extraordinary
Among the words inscribed were the colours “urdin” (blue), “zuri” (white) and “gorri” (red), verbs “edan” (drink) “ian” (eat) and “lo” (sleep), the excavation team said.
Another piece read “Iesus, Ioshse ata ta Miriam ama” (Jesus, the father Joseph and the mother Mary) while another had the greeting “Geure ata zutan” (May the Father be with you).
I’m afraid my first response is skepticism—Larry Trask instilled in me a deep suspicion of any and all claims pertaining to Basque—but if it’s true, it’s exciting. Anybody know more about it?
(Thanks for the link, Edward!)
Update. The texts are fake.
In case anyone is interested, here is a link to what seems to be the original article in El Pais.
And another one, from the Basque edition. My knowledge of Basque is limited, so all I understand is that the good people of El Pais did the right thing and asked some experts. Anybody got any more?
I believe there is a world of difference between an archeological discovery like this and the usual wild claims Larry Trask refers to. In other words, this looks like the real thing. The only aspect of this discovery I am worried about is dating. First estimates tend to be a little off.
Yeah, I didn’t mean to compare it to the lunatic stuff; all I meant to say was that Trask had sensitized me to the prevalence of misinformation and nonsense where Basque is involved. As you say, it’s the dating that’s problematic here, but even if it’s only a couple of centuries earlier than the 11th century, it’s still great.
I assume that this is the same find: http://adrianmurdoch.typepad.com/bread_and_circuses/2006/06/irunaveleia.html
If so, it does seem as if the dating is reasonably reliable.
Wikipedia article for the site, with various links.
Doesn’t the third century seem early to find Christian inscriptions in Basque? I’d always assumed it took a while for Christianity to spread into the Western part of the Empire.
Ah, if only Larry could have seen this. It doesn’t seem too implausible.
I am curious (with eyebrows) about this though:
The Egypt expert of the University of Barcelona Montserrat Rius has explained that some Latin inscriptions refer to the ancient Egyptian history and its divinities, and has noted there are also hieroglyphic inscriptions “with a perfect layout” that make experts think they were taught to children.
nw: This lengthier original(?) version, has (with adjustments for the spam filter):
Y qué decir del exótico origen egipcio del preceptor que impartía allí sus amplios conocimientos clásicos, añadiendo t@mbién temas específicos sobre su propia historia, escritura, cultura y creencias. Así nos encontramos con la presencia de escritura jeroglífica clásica, atestiguada por vez primera en un @mbiente tan norteño en el occidente europeo.
So I get a picture more of “show-and-tell” than hieroglyphic class.
Vanya, no, archaeological museums throughout Spain have plenty of Christian artifacts from the third century.
Wonderful! I hope they publish the inscriptions soon.
If you’re looking for photos, this article has pulled them together in one place, including (I think) hieroglyphic signs, Calvary, and two shards with Basque.
I would like to believe that this is an authentic discovery but I suspect it is a forgery much like the alleged Phoenecian inscriptions found in Brazil that were supposedly left by the Carthaginian explorer, Hanno on his way back to Carthage from South Africa.
For one thing. The Basque words mentioned in the article look to much like modern Basque words. However the Basque language has certainly changed over the past 2,000 years just as Irish Gaelic, Welsh and Armorican Celtic (Breton) all have (They don’t resemble the Celtic of Gallic inscriptions very much).
“Ioshe” for Jesus? The ‘sh’ sound was unknown to the Ancient Greeks and Romans and most likely would not have been transliterated that way by anyone usuing the Roman alphabet even if they had an ‘sh’ sound in their native language. This is pretty much a later English spelling of the sound.
Furthermore, historians tell us that throughout Europe, the Near East and North Africa, Christianity started out in the cities first. The countryside was more consevative and remained pagan for longer periods of time (The same pattern occurred during the Reformation 1500 years later between Protestantism and Catholocism). The Basques have certainly always lived in the more rural parts of Spain and Southern France.
Finally, other sources I’ve read claim that Spain (Hispania) was one of the most Romanized provinces in the Empire outside of Italy and very pagan until almost the middle of the 4th century A.D. Conversion there to the new religion was very slow because Christianity was associated by most of its citizens with treason against the state just as it was in Rome. The Christianization of Spain actually happened very rapidly during the middle and latter parts of the 4th century.
Nevertheless, we’ll wait and see what the experts have to say about it in the months ahead.
Catholocism > should be Catholicism
Above posting was by me.
The Spanish Wikipedia article mentioned above has a different version of the inscription: the much more reasonable iesus, iose ata ta mirian ama (and with a no nasty final [m]). The other words quoted are ones that are remarkable for being identical or almost with modern Basque, implying there are others in the inscriptions that are different (and therefore less interesting to the casual reader, argh!).
Language Hat : « Larry Trask instilled in me a deep suspicion of any and all claims pertaining to Basque »
Stanford professor Merritt Ruhlen asserted that Basque, together with Caucasian languages, Ienissean, Sino-Tibetan and Na-dené, belonged to a same family dubbed “Dené-Caucasian”. Do you know if nowadays this theory is still mentioned and how far it has been proven or refuted?
Ruhlen is completely untrustworthy on such matters.
A comment from Hartza:
I wish it were true… I don’t mean it is not: to date I have no more information than you (well, maybe actually yes, but not very much more) but, in any case, these kind of forgeries do not stand the analysis more than six months AND the archeologist team leading the study is one of the most relevant ones in Spain, so…
… wait and see.
Anyway, I would like to underline a couple of things:
First, here we are speaking about TWO different set of items: one related to some christian inscriptions which also included some hieroglyphs. It doesn’t seem to be hieroglyphic writing at all, but just a sort of ‘show and tell’, as someone has hinted before.
In any case, finding some IIIrd century christian inscriptions doesn’t mean at all that all the region, or even the north of Hispania were christian at that time…
Second, the inscriptions in Basque belong to a second and different finding in the same city (Iruña Veleia), which also seems to be quite older than the previous one, maybe dating even from the Vth century.
I am not in a position to defend or attack its authenticity, but please remember that the Basque-Aquitane inscriptions studied by Gorrochategui are dated around 1st century aC and include words (like CISON -> GISON, ‘man’; UMME -> ume, ‘child’; ANDEREA -> andere, ‘lady’) which continue being almost completely identical in present Basque, 2000 years after… As they belong to the so-called ‘nuclear vocabulary’.
True? I don’t know, but it could be.
Finally: Up to know, the first written Basque sentence (in fact, just 2 sentences, and quite difficult to interpretate) appears in a 10th century document (the ‘Glosas Emilianenses’) which happens to be written not in early Castilian (Spanish) but in another romance, now disappeared: navarro-aragonese.
I promise to keep you all informed on these thingies. Keep in touch.
Hartza, proud to be Basque.
Signatus, keep in mind Larry Trask’s statement about Ruhlen:
“Ruhlen is not recognized by anybody in linguistics as a member of the profession. Every single linguist who is acquainted with his work regards him as a crackpot and a charlatan.”
Basque is clearly related to the Dravidian languages, and (now that Pictish is extinct)is the westermost relic of the Dravidian diaspora. (Earlier theories that the Algonquian languages were brought to American by Pictish refugees have not found empirical support.)
Re: “”Ruhlen is not recognized by anybody in linguistics as a member of the profession. Every single linguist who is acquainted with his work regards him as a crackpot and a charlatan.” — Christopher Culver
Christopher,
Remember that English scientists were saying the same thing about Charles Darwin in the 1860’s (except for Thomas Huxley rather wisely) but now we know that they are all wrong. I’m not saying that history will record Ruhlen being as great as Darwin, but I’ve read Merritt Ruhlen and most of his statements and theories are plausible. At most, they might need a little brushing up. For example, he tends to think that all American Indian languages descend from a single Proto-Amerindian language. Yet experts on American Indian anthropology and languages claim that they arrived in two migrations, one 12,000 years ago and one 8,000 years ago. The two groups were not genetically or linguistically related to each other.
Christopher Culver > “Ruhlen is not recognized by anybody in linguistics as a member of the profession. Every single linguist who is acquainted with his work regards him as a crackpot and a charlatan.”
Ok. But isn’t it rather surprising then to see that this academic still appears on Standford’s website (Department of Anthropological Sciences) as a “Lecturer in Anthropological Sciences and Human Biology”? Isn’t this university supposed to have a good reputation?
Siganus, but notice he’s not in a Department of Linguistics.
Ah, he could be seen as a kind of stray cat then. It’s nonetheless rather surprising to note that Stanford University doesn’t bring him back on the “right track”. Or did they? Maybe they can’t…
However, at the Santa Fe Institute, an institution intending to promote “multidisciplinary collaborations”, in a project focusing on the Evolution of Human Languages, he is presented as from Stanford University. And this project apparently endorses the Dene-Caucasian theory: http://ehl.santafe.edu/maps5.htm
Ruhlen sweeps a bunch of languages into the Sino-Tibetan Na-Dene group, including three previously regarded as isolates (Basque, Burushaski, and Yeniseian.) He includes Yukaghir, another isolate, with Finno-Ugric and Gilyak, still another, with Korean and Japanese.
These judgments are all controversial resolutions of difficult, embattled questions. It seems very unlikely that he has done is more than just lumping.
As I understand, LH house policy is unfriendly to all supergroups of this kind, not just Ruhlen’s.
I’m sorry to see the Santa Fe Institude involved. It reminds me of a recent physicist’s attempt to apply statistical analysis to romanized transcriptions of various languages, with the goal of finidng unknown relationships. As I recall, this method missed the very close relationship between Portuguese and Spanish, leading me to suspect that any positive relationships it found would be bogus.
John Emerson: Pictish is Dravidian? What should I read about this, please?
Dearieme — “Dravidian origins” is a standing joke of mine here.
I’ve spent a certain amount of my own time dabbling in far-fetched historical-linguistic theories, but by now I agree with Hat here that almost all of that stuff is bogus. (The Tokharians are a rare exception).
Just a comment about the Breton/Gaulish distinction…most people working with Breton would reject the claims that it’s closely related to Continental Celtic. This theory was popular for a time, but even a cursory glimpse at Welsh and Breton shows how closely related they are (and middle Welsh and middle Breton were even closer). So the fact that Breton doesn’t look like Gaulish is unsurprising, in that there’s a much longer point of separation between them than the Breton/Continental Celtic theory would have allowed…there’s at least as much separation time as with P- and Q- Celtic, and there is obviously tremendous divergence between the two groups.
Re: Breton and Gallic (Gaulish):
I believe in the continuity theory myself that Armorica was feebly Romanized and that Celtic has always been spoken there. In fact, this was the prevailing view in Europe for centuries even among linguists. The British origins theory of Breton dates only from the 19th century from what I’ve read and has never been accepted by most linguists and scholars in France.
One must not forget that the British Celts themselves came from the French and Belgian mainlands before arriving in Britain and spoke nearly the same language as both the Gauls and the Galatians (in Asia Minor). This has been proven partially by the close similarities of some place names and personal names among all three peoples.
Breton is quite clearly Island Celtic.
Wikipedia has a good article on Aquitanian.
I don’t quite understand what people have against supersupersupergroups like Dené-Caucasian. It’s certainly no more surprising than Indopacific! Do people really prefer having no hypothesis over having any hypothesis? (Trask’s allergy is easy to understand, given the fact that he has to deal with real pseudoscience all the time.)
That said, cladistic analyses should be done much, much, much more often in linguistics. When it comes to phylogenetics, the linguists are way behind the biologists.
I realize that some will misunderstand my motive for posting the following, especially if they discover that I was born in Wales.
It is my understanding that although the Vasconic language attested to by this article is sometimes called Basque, it would perhaps be more instructive to call it by another name (my preference is for Aquitani), because around the third century AD, and indeed up to at least 476, this language was sufficiently different from modern Basque to warrant a separate classification. With the postulated but relatively well developed pre-proto-Basque language in existence from the middle of the first millennium BC through to at least 100 AD, it seems inordinately obtuse to think that this branch of an already somewhat isolated language group would go through two more generations of languages in the next 100 to 150 years.
The complete absence of bilingual Basque and Latin inscriptions in the face of some well-attested Aquitani/Latin penmanship suggest that Basque at least post-dated 476 AD, and more likely replaced or perhaps more slowly developed from Aquitani around the seventh or eighth century.
Referring to the language of the Aquitani people during Roman occupation (of southern Gaul) as Basque is surely exactly the same (and at exactly the same time) as referring to the “British” or “Celtic” language of the late Roman occupation of britain as Welsh.
Sorry.
Which is to say that it’s quite a reasonable thing to do. We used to call the earliest written form of English “Anglo-Saxon”, as if it were a different language altogether (which from the point of view of mutual intelligibility it surely is), but we now call it “Old English” to emphasize continuity. Of course, there is such a thing as too much continuity: the name “Old Russian” has mostly been replaced by “Old East Slavic” to convey that not only Russian but also Ukrainian and Belarusian are its modern descendants, and perhaps something similar should be done with the name “Old Irish”.
Christopher Benson: No apology required, your point is quite valid. There is a difference between Welsh and Basque however: the ancestor of Welsh, spoken in Roman times, whether we call it British Celtic or Brythonic Celtic (I prefer the latter), spawned languages other than Welsh (Cornish and Breton), so that it would be inaccurate to use “Welsh” as a label to designate the Celtic language of Roman Britain, as inaccurate as calling it “Breton” or “Cornish”. By contrast, “Old Basque” or “Aquitanian” are acceptable synonyms, since however you wish to call it, the language only has one known daughter (Basque).
Furthermore, even before Aquitanian had been shown to be the ancestor of Modern Basque, the term “Proto-Basque” had already become established to refer to (the ancestor of Modern) Basque back in the days of the Roman Empire, when the first Latin loanwords entered the language (and which shared the sound changes which subsequently affected Basque, thereby giving us a picture of what Basque was like two thousand years ago and how it has changed), so that in effect Aquitanian and Proto-Basque are one and the same language. Since the latter term is more transparent, it is to my mind preferable.
John Cowan: A thought. If Scotland ever becomes a separate country (I believe they have a referendum scheduled this year) “Anglo-Saxon” as a linguistic label might come back in fashion, since after all you could argue that this language could just as reasonably be called “Old Scots” as “Old English”.
Even with Scottish political independence, I doubt if there will be an ideological power struggle for possession of the name English. Some of the English have been known to grumble (contrary to their national maxim) that American English is not English at all, but few are so extreme as that: they seem to be mostly content to complain about how it’s ruining “proper English”.
Wouldn’t most linguists protest against a claim that one modern dialect of English is closer to Anglo-Saxon than the other? After all none of them has avoided neither the vikings, the Normans, the great vowel shift nor later changes. Does any dialect speaker understand more of Beowulf than the other or, by the way – than an Icelander or a low German?
Gone are the days, when people (other than possibly Hindu nationalists) thought of Sanskrit as closer to PIE than Latin, Gothic, ancient Greek or Avestan. So argued e.g. Larry Trask, that English is the only language in the IE family that has preserved the ‘w’-sound (a petit-maître may add that it’s also preserved in some rural Swedish dialects). Similarly I suspect that unique archaisms from Anglo-Saxon can be found all over the English speaking world.
Well, Scots (which counts as a dialect of English diachronically though not synchronically) has experienced rather more Vikings, rather fewer Normans, rather more Parisians, and a different GVS that left /o:/ and /u:/ unaffected.
As for /w/, Larry must have forgotten about “Hen Wlad Fy Nhadau” (he was partly of Welsh extraction).
Strange, if Larry Trask should have missed that. But his remark of course concerned /w/ inherited from PIE. And I’m ignorant, whether any Celtic /w/ is so or an innovation (perhaps under the influence of ‘Englisc’).
Well, initial Welsh /w/ is typically, but not invariably, lenited from /gw/, which in turn comes from Proto-Celtic /w/: in this case wlad is lenited from gwlad ‘land, country’ < ProtoC *wlati- ‘sovereignty. A stronger example is gwlanen ‘woolen cloth’ (probably > English flannen > flannel) < gwlan ‘wool’ < ProtoC *wlana < PIE wele- > English wool. This pattern also applies to Roman borrowings like Gwener ‘Friday’ < L. (dies) veneris.
A native word beginning with unlenited /w/ is wyth ‘eight’, which obviously does not have a PIE ancestor in /w/. Looking this up, I find that the Welsh for ‘week’ is wythnos lit. ‘eight nights’, perhaps reflecting the Roman (or both Roman and Celtic?) habit of counting both ends when specifying time intervals; cf. Latin nundinae ‘eight-day week’, and the day before the day before the Kalends (1st) of each month being known as ‘the third day before the Kalends’.
flannel/flanelle
I had always thought that Eng flannel was a borrowing from Fr la flanelle, but the TLFI confirms that the borrowing went the other way. The English word was apparently a borrowing from Welsh gwalen referring to a kind of woollen garment, as did flannel at the time of borrowing. The English word entered the French language with the ‘(light, soft) woollen cloth’ meaning, later taking on the additional meaning ‘any garment made of this cloth’.
JC: [Welsh] wlad is lenited from gwlad ‘land, country’ < ProtoC *wlati- ‘sovereignty’.
Would you know the answers to these questions: Is the wl- the same root in both words? Does it go back to PIE? If it is one root, what seems to be its accepted meaning? Is the name Gwladys (Eng Gladys) derived from gwlad?
Wiktionary gives no PIE etymology of gwlad. Alas, there seems to be no proper Welsh etymological dictionary (there is a Proto-Celtic one, but it’s Brill = $$$$), so I have no idea if there is a connection between *wlati- and *wlana-. But a connection between ‘sovereign’ and ‘wool’ doesn’t look too promising to me.
There are two theories of the name Glwadys: that it is < glwad and means ‘princess’, or that it is a Welsh version of Latin Claudia ‘lame (f.)’. The first looks more likely to me, but certainly many people named Gwladys, Gwladus used Claudia as their English name in the 19th C. The English name Gladys did not become popular until about 1900, but then for whatever reason it became immensely so.
JC, Thank you for your answers, I had never run into the Gladys – Claudia equivalence. About my first questions, I was not asking about a potential connection between *wlati ‘sovereignty’ and <*wlana ‘wool’, which is most unlikely, but between *wlati and gwlad ‘land, etc’ where a semantic link is more plausible. .
Oh, I see. In fact, gwlad and wlad are the same lexeme: the former is the citation form, and both are descended from Proto-Celtic *wlati. In Welsh, when a noun is preceded by an adjective, the initial consonant of the noun (if any) is lenited, a process known as “soft mutation”. Soft mutation originally occurred in contexts where a final vowel in the preceding word was lost, but is now a morphophonological rather than simply phonological alternation.
In particular, though most adjectives follow their noun in Welsh, hen ‘old, ancient’ (ProtoC senos, cognate with Latin senex) is exceptional: it always appears before its noun, which then undergoes soft mutation. The lenited form of Welsh /g/ is zero, so hen + gwlad = hen wlad. The fact that the /g/ of gwlad was never intervocalic here is irrelevant.
Soft mutation also happens to feminine nouns when preceded by y ‘the’ or un ‘one’. These words once had gender markers that are now lost, with the masculine forms ending in a consonant and the feminine forms ending in a vowel, analogous and cognate to Latin unus, una. Likewise, feminine nouns themselves trigger soft mutation in a following adjective, so y + merch + bach lit. ‘the girl little’ is expressed as y ferch fach (where written f is /v/, and is the lenited form of both /b/ and /m/). Most prepositions also induce soft mutation.
Similarly, the word fy ‘my’ (itself permanently lenited from IE 1sg forms in /m-/) once ended in /-n/ and consequently now triggers a different mutation, the nasal mutation, in the following word tad-au ‘father-s’ (a papa-mama word that is now the standard word in Welsh), making it nhadau (Welsh initial /t/, like English initial /t/, is strongly aspirated). The nasal mutation is rare, so in spoken Welsh it can carry the whole load, allowing fy to be omitted altogether (the only other frequent context for it is after yn ‘in’).
As I mentioned before, /w/ > /gw/ is one of the changes separating Proto-Celtic from Welsh. It is mere coincidence that soft mutation “undoes” this sound change.
Thanks JC. I had heard about those mutations but did not know the details.
I’d be surprised if *wlati- weren’t cognate (or related by borrowing) with Slavic vlad- “rule”, Germanic wald- “power”.
Marie-Lucie: when you reacted to the name Geominy: please mail me!!
Trond: I’d be surprised if *wlati- weren’t cognate (or related by borrowing) with Slavic vlad- “rule”, Germanic wald- “power”.
I don’t know wald- but I too was wondering about Slavie vlad- as in Vladimir (I suppose), without knowing the meaning. I have seen the name written Volodimir (perhaps an older or OCS form ?), and the diminutive of Vladimir is Volodia).
ML: Russion -olo-, -oro- vs common Slavic -la-, -ra- is all over the place. Cf. Novgorod vs Leningrad, molod- vs mlad- (young), dorog- vs drag- (dear), zolot- vs zlot- (gold), golos vs glas- (voice) etc.
Trond: As you probably know the SAOB is the slowest project on this planet. After well over 100 years they have reached the letter ‘T’ (and after ‘Z’ we have our three diacritics ‘Å’, ‘Ä’ and ‘Ö’).
But Elof Hellquist in his Svensk etymologisk ordbok, http://runeberg.org/svetym/1240.html from 1922 comments upon Sw. våld ‘violence’ and the verb vålla modern ‘cause’, earlier ‘rule’. He connects them to Goth. OSax. and OHG waldan and to OE weldan, modern English wield.
As PIE origin he gives *ǔaltō or a dh-present *ualdhō. This in turn he connects to Lat. valere, ‘be strong’ (cf. valid, value). Then he mentions Oir. flaith, ‘domination’ and the Latv. wala, ‘power’ and Lith. valdýti, ‘rule’ but is at least in the Baltic case unsure whether they are Gmc. loanwords or inherited.
An extremely helpful person has sent me Matasović’s Proto-Celtic dictionary, which I quote (references omitted):
*wlati- ‘sovereignty’ [Noun]
GOID: Olr. flaith [i f, later m] ‘sovereignty, ruler’
W: OW gulat [f], MW gwlad ‘country’
BRET: OBret. guletic ‘prince’, MBret. gloedic ‘count’, MoBret. gwlad ‘inheritance’
CO: OCo. gulat gl. patria, Co. gulas, gwlas
SEE: *walo- ‘prince, chief
ETYM: Gaul. PN Ulattius might be related, cf. also the ethnonym Ulatti. Olr. flaithem [n m] ‘ruler’ represents PCelt. *wlati-mon-. PCelt. *wlati presupposes PIE *wlHti- with shortening of the first vowel by Dybo’s law (see *walo- ‘prince, chief).
*walo- ‘prince, chief [Noun]
GOID: Mlr. fal ‘rule’ (DIL fal ‘king’)
W: OW Con-gual [PN]
BRET: OBret. Conuual [PN]
PIE: *welH- ‘rule’ (IEW: 1112)
COGN: Lat. ualeo ‘be strong’, Goth. waldan ‘rule’, Toch. B walo ‘king’
SEE: *wlati- ‘sovereignty’, *wal-na- ‘rule, govern’
ETYM: OIr. Con-all, OW Con-gual, OBret. Conuual [PN] < *kuno-walo-. Cf. the parallel compound PNs Gaul. Katouualos, OW Catgual, Olr. Cathal ‘strong in battle’ < *katu-walo-. PCelt. *walo- can be derived regularly from PIE *wlHo-. Cf. also W gwaladr ‘ruler’ < *walatro- < /i>*wlhzetro-. Mlr. follamnaigid ‘rules, governs’, follamnacht ‘ruling, government’ presuppose a nominal derivative from the same root, PCelt. /i>*wolna-mon- ‘ruler’ < PIE *wolH-nehz-(mon). The Gaul. ethnonym Catu-vellauni is also often related to this set of words.
*wal-na- ‘rule, govern’ [Vb]
GOID: Olr.fallnaithir, -fallnathar, -follnathar
PIE: *welH- ‘be strong, rule’ (IEW: llllf.)
COGN: Lat. ualeo ‘be strong’, Lith. velditi ‘govern, rule’, Goth. waldan ‘rule’, perhaps also Hitt. hullffi
– ‘smash, defeat’
SEE: *walo- ‘prince, chief
ETYM: Celtic preserves the zero-grade of the PIE root (PIE *wl-nehz- > *wal-na- >> PCelt. *wal-na-).
So that pretty much confirms Trond and Stefan. Note: PN = proper name, DIL = Dictionary of the Irish Language, IEW = Pokorny.
I thought it’s way too poorly attested to tell if it’s a direct ancestor or a closely related separate branch?
(And West Flemish, and the Dutch of Suriname though of course that may be substrate influence, and at least one Walser dialect in the Aosta valley.)
More like *wlh2-.
In particular, when you find the latter forms in East Slavic, they’re always loans from OCS, a South Slavic language.
There seem to be OCR errors. In particular, PIE *hz must be *h2, and there wasn’t any f (let alone llff) in Hittite.