Talk:Celtic languages

From Wikipedia, the free encyclopedia
Jump to: navigation, search
          This article is of interest to the following WikiProjects:
WikiProject Languages (Rated C-class, High-importance)
WikiProject icon This article is within the scope of WikiProject Languages, a collaborative effort to improve the coverage of standardized, informative and easy-to-use resources about languages on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.
WikiProject Celts (Rated B-class, Top-importance)
WikiProject icon Celtic languages is within the scope of WikiProject Celts, a collaborative effort to improve Wikipedia's coverage of Celts. If you would like to participate, you can edit this article or you can visit the project page, where you can join the project and see a list of open tasks or take part in the discussion. Please Join, Create, and Assess.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
WikiProject Europe (Rated B-class)
WikiProject icon This article is within the scope of WikiProject Europe, an effort to build a comprehensive and detailed guide to European topics of a cross-border nature on Wikipedia.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.
WikiProject Cornwall (Rated B-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Cornwall, an attempt to improve and expand Wikipedia coverage of Cornwall and all things Cornish. If you would like to participate, you can edit the article attached to this page, or visit the project member page, where you can join the project and/or contribute to the discussion.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.

P/Q Celtic vs. Insular/Continental Celtic[edit]

"Since the realization that Celtiberian was Q-Celtic in the 1970s, the division into Insular and Continental Celtic is the more widespread opinion (Cowgill 1975; McCone 1991, 1992; Schrijver 1995)."

This sentence is a bit confusing, and could do with some explanation. Is the reasoning behind Cowgill, McCone and Schrijver's opinion that Goidelic and Brythonic are so similar (much more similar than Goidelic is to Celtiberian and Brythonic is to Gaulish) meaning that they cannot have evolved seperately from the time of Celtiberian and Gaulish splitting, and therefore must form one "insular" branch? If so, perhaps the quote could be changed to "Since the realisation that Celtiberian had Q-Celtic characteristics in the 1970's..." (Dragonhelmuk (talk) 18:10, 15 June 2008 (UTC))

I tried to clarify the sentence. The Insular Celtic hypothesis is indeed that the Goidelic and Brythonic branches evolved from a common ancestor that the Continental languages did not evolve from. The fact that Celtiberian is "Q Celtic" is not really necessary for the Insular Celtic hypothesis, nor is it actually particularly damaging to the Gallo-Brythonic hypothesis, though, so the sentence is something of a non sequitur. —Angr 20:19, 15 June 2008 (UTC)
The edits you've made make the text a little simpler, but I still feel the Celtiberian part could use some clarification. If Celtiberian having Q-Celtic characteristics is really not relevant here, then editing it out would probably make the text clearer (i.e. implying that the change in opinion has not been due to research on Celtiberian, but just a general shifting of thought.) The passage currently reads:
"There are legitimate scholarly arguments in favour of both the Insular Celtic hypothesis and the P-Celtic/Q-Celtic hypothesis. Proponents of each schema dispute the accuracy and usefulness of the other's categories. Since the discovery in the 1970s that Celtiberian was Q-Celtic, the division into Insular and Continental Celtic has become the more widely held view (Cowgill 1975; McCone 1991, 1992; Schrijver 1995)."
Would the citations support a simpler statement, e.g.: "There are legitimate scholarly arguments in favour of both the Insular Celtic hypothesis and the P-Celtic/Q-Celtic hypothesis. Proponents of each schema dispute the accuracy and usefulness of the other's categories. However, over the last thirty years the division into Insular and Continental Celtic has become the more widely held view (Cowgill 1975; McCone 1991, 1992; Schrijver 1995)." (Dragonhelmuk (talk) 22:47, 15 June 2008 (UTC))
Yeah, except I'd say "since the 1970s" rather than "over the last thirty years", because we expect Wikipedia to be around a long time, and we don't want to have to remember to come back to the article in ten years and change it to "over the last forty years" and then ten years later and change it to "over the last fifty years" and so on! —Angr 07:31, 16 June 2008 (UTC)
Agreed and changed. (Dragonhelmuk (talk) 16:59, 16 June 2008 (UTC))
I do not agree with you : the choice of the sources is partial. Other good specialists of Celtic (P-Y Lambert, Fleuriot, Schmidt for instance) think exactly the contrary and show that Brythonic and Gaulish are much more closer, than we thought before (without the fact that they are both P-languages). Such conclusions are not dead theories, but based on the discovery of new Gaulish inscriptions. They call it Gallo-Brythonic. Before stating something, it would be good to explain why such specialists (Schrijvers, Cowgill, etc.) make a difference between insular and continental Celtic (phonetics, word stock, etc..). Such differences as the mutations are not very convincing arguments, because they could have appeared in the very late Gaulish language, when it was not recorded anymore and on the contrary, it might have not existed in antique Brythonic. Nortmannus (talk) 10:56, 4 April 2010 (UTC)
It is very simple-minded and old fashioned to reduce the common points and the differences between on one hand Gaulish / Brythonic and on the other hand to Goidelic / Celtiberian to this single fact P-Celts and K-Celts. They are two or three other common "innovations" between Gaulish and Brythonic, so it is not accidental. for exemple -nm- > -nu- Gaulish anuana Old Welsh enuein same evolution of *anmana. They have been isolated particularly in the "Plomb du Larzac" (discovered 1983) and new words were found very closed to the Brythonic one. Nortmannus (talk) 11:28, 4 April 2010 (UTC)
Schrijver has pointed out that because of Middle Welsh mynwent "cemetary" (a borrowing from Vulgar Latin *monimenta, instead of Classical Latin monumenta), the sound change *nm > nw must postdate the syncope in Brythonic (which must be relatively late), and therefore cannot be older than about 500 AD, while the Plomb du Larzac is certainly considerably older than that and the Gaulish change was by all appearances already complete in 200 AD. Perhaps the sound change was borrowed from Late Gaulish into Brythonic, but because of the temporal difference, it cannot be a common innovation and therefore has to be abandoned as argument in favour of Gallo-Brittonic.
Another sound change which has been proposed as dividing Celtic along the same lines is the development of the PIE syllabic nasals: PIE * > Gaulish and Brythonic an, but supposedly Celtiberian (as well some Gaulish dialects) and Irish en, as Irish cét : Latin centum "hundred" (with original *) and Irish cét : Latin cantum "sung" (with original *an) develop identically (even giving the exact same result), because Vulgar Latin *ancora was borrowed into Old Irish as ingor and because PIE *-eh2-m > *-ām (suffix for accusative singular in the eh2-stems) exerts the same palatalising effect on the preceding consonant in Old Irish as does PIE *-m̥ (suffix for accusative singular in the other consonant stems), implying a development PIE *-ām > *-am > Primitive Irish *-en. Therefore, it is now commonly assumed that the PIE syllabic nasals gave *am and *an already in Proto-Celtic.
A final sound change which has been proposed as Gallo-Brittonic is the assimilation of Proto-Celtic *o to a before a resonant, as in *torano- "thunder", which yielded Gaulish or Gallo-Latin Taranus and Taranis etc. (attested already in the 2nd and 1st centuries BC) as well as Middle Welsh taran, but Old Irish torann. However, the Christian loanword Middle Welsh manach from Latin monachus – which for historical reasons cannot have been borrowed before the 4th or possibly the end of the 3rd century – display the same sound change which was already complete in Gaulish in 100 BC! So again, at most it is a sound change that spread from the central language Gaulish to the more peripheral Brythonic.
This means that the only common Gallo-Brittonic innovation left is Proto-Celtic * > p, and as Schumacher points out in his Die keltischen Primärverben (2004), that innovation is typologically quite trivial and is also well-known from Greek and Osco-Umbrian, as well as Romanian, and Latin * even yielded *b in Greek, Osco-Umbrian, Romanian and Sardinian (just as it did in Proto-Celtic). The sound change is likely an areal phenomenon in Celtic.
On the other hand, there is at least one and very characteristic morphosyntactic Insular Celtic innovation, namely the development of absolute vs. conjunct endings in the verb, which presupposes an entire complex of innovations, in part phonological, in part morphological and in part syntactical (such as the rise of the Insular Celtic verbal complex and the development of the verb-initial constituent order). There is also evidence for an independent, exclusively Insular Celtic sound law missing from Gaulish, namely Proto-Celtic */s/ [z] > [ð] before voiced stops.
In sum, the case for Insular Celtic could be clearer (but an Insular Celtic protolanguage can be quite neatly reconstructed from the Old Irish and Brythonic evidence and is very helpful for purposes of IE and Continental Celtic studies), while the Gallo-Brittonic hypothesis is, with Schumacher, outdated. --Florian Blaschke (talk) 22:05, 5 August 2011 (UTC)
When did these special insular Celtic innovations take place ? After Gaulish died out ? Where are the antique insular Celtic inscriptions ? As far as history is concerned, I don't see any logic in this statement. Gaulish spoken in the northwestern part of [what is today] France and brittonic spoken in [what is today] England and Corwall should be very close to each other, if you view things from a different angle. First, from a geographical point of view : the English channel is not wide : the distance between Calais in France and Dover is only 30 km and all along the coast maximum 150 km. Sailing across the English channel : around 12 hours (to compare riding to Paris from the Norman coast : 3 days). You suppose probably, that the English channel was a sort of Maginot line and the Celts were extremely bad sailors and they never crossed the channel. On the contrary, they could run to Noricum (1500 km away) and Galatia (3000 km) to avoid any phonetical change in the "continental" Celtic. Second, Archelogy demonstrated the extension of very specific Celtic artefacts in the northwestern part of Europe on both sides of the English channel : the Fécamp rampart, typical of Northwestern France and Southeastern England ; the graves with 2 wheeled carriages of Yorkshire, the Seine valley and Picardy, Western Germany. Similar tribes : Parisii in Yorkshire and in Paris region, where these tombs with carriages are located, Atrebates around Arras and in Hampshire, Surrey, Sussex, etc. So, same origin, same culture, same religion and they would have spoken a different language ? On the contrary, where are the common artefacts between the Celtiberians and the northwestern Gaulish tribes ? These two cultures have significant differences, and what about Noricum or Galatia, except what the Romans or the Greeks told about them ? I haven't the slightest doubt about the phonetical phenomenons that you describe, but, what does it mean really, concerning a "Brittonic language", that would have been different of a "continental Celtic language" in olden days ? Nortmannus (talk) 08:21, 11 August 2011 (UTC)
When these Insular Celtic innovations took place is not known with precision, but Schrijver has pointed out that besides the P/Q difference, no phonological difference between British Celtic (according to the absolute chronology in Language and History in Early Britain) and Goidelic seems to be older than the 1st century AD, hence, these innovations would seem to date to the last centuries BC. That is, Proto-Insular-Celtic was contemporary with Gaulish (and very similar to it, especially in the nominal domain; but then, all the ancient Celtic languages were still quite similar to each other), and place names from Britain recorded in the 1st centuries BC and AD were essentially Proto-Insular-Celtic. (Note that the presence of Celtic in Ireland need not be older than the 1st century BC; the tribal names and toponyms reported from ancient Ireland aren't really as old as Pytheas's voyage.) The Insular Celtic innovations are not attested in any inscriptions, of course, but only reconstructed; don't be silly. I'm not sure what the problem you have with the Insular Celtic scenario is. Surely there was immigration from Gaul to Britain, but we don't know all the dialects spoken in ancient Britain; there may well have been dialects closer to Gaulish in Southeastern Britain, imported through later immigration, but which have not survived Romanisation (or Gauls who later migrated to Britain could also have taken over Insular Celtic dialects). Only a single Celtic dialect of Britain, Proto-British, has survived, and it was most likely not spoken in Southeastern Britain; Proto-British was most likely surrounded by more and less closely related varieties, and only part of a large dialect continuum, just like any proto-language.
Continental Celtic, of course, was very varied at the time of Proto-Insular-Celtic and is not a true genetic node, anyway, so your wording misrepresents the idea. In the 5th century BC, you may have had a Pre-Proto-Insular-Celtic, (Pre-)Proto-Gaulish (perhaps), Lepontic and early Celtiberian dialect, and other lost dialects, but not a Proto-Continental-Celtic dialect, as there are no innovations shared by all Continental Celtic varieties. --Florian Blaschke (talk) 21:03, 27 November 2012 (UTC)

Characteristics of Celtic Languages[edit]

This literal translation is incorrect:

(Irish) Ná bac le mac an bhacaigh is ní bhacfaidh mac an bhacaigh leat. (Literal translation) Don't bother with son the beggar's and not will-bother son the beggar's with-you.

an bhacaigh is in the genitive case, therefore its literal translation is "of the beggar". So, it should be:

(Literal translation) Don't bother with son of the beggar and not will-bother son of the beggar with-you. (talk) 20:50, 3 July 2008 (UTC)

There's no difference between "the beggar's" and "of the beggar" (except in word order, which isn't the point here). —Angr 21:03, 3 July 2008 (UTC)

A literal translation (look it up on(in?) the wikipedia!) is a word for word translation. "Verbum pro verbo," in Latin. Same word order. So yes, it's the point, and "Don't bother with son of the beggar and not will-bother son of the beggar with-you." sounds much better. (talk) 05:46, 16 November 2008 (UTC)

If it *is* the point, "the beggar's" is the precise literal translation of "an bhacaigh" anyway, so what's your point? (talk) 22:12, 24 March 2009 (UTC)

Since the article is generally inflected for case, gender and number, and you can tell from context that this occurrence of "an" is genetive, it should be be translated as "of the". MichealT (talk) 22:32, 28 March 2010 (UTC)

Also, modern Insular Celtic and old (mainly Continental) Celtic features are all jumbled up. /Keinstein (talk) 02:01, 30 May 2009 (UTC)

"of the beggar" is grammatically the exact same this as "the beggar's", they're both genitive markers in English. And, yes, that's "genitive", not "genetive."Baininscneach (talk) 17:00, 9 April 2011 (UTC)

P-Celtic/Q-Celtic hypothesis[edit]

I have split out Q-Celtic hypothesis/ P-Celtic hypothesis into brief articles that redirect back here. I have used only information from this page. Lucian Sunday (talk) 15:52, 9 August 2008 (UTC)


It could be good to tell how many words are common between all Celtic languages. Perhaps 10 thousands between Breton, Cornish and Welsh. But for the Celtic languages in general, I would say about 3000. (talk) 06:14, 9 December 2008 (UTC)

Please feel free to add this to the article if you can reference a reliable source. Thanks --Yumegusa (talk) 23:36, 10 December 2008 (UTC)

Orkney and Shetland[edit]

Why are the Orkney and Shetland islands marked with the same colour as Scotland suggesting they are (parto of) Celtic nations? Undoubtedly Celtic was spoken there prior to the vikings making it extinct, but the same (with anglo-saxons instead of vikings) could be said about England. // JiPe ( (talk) 19:44, 29 March 2009 (UTC))

Presumably simply because they're part of the country of Scotland. —Angr 20:02, 29 March 2009 (UTC)
And Brittany is part of the country of France. Yet only Brittany is marked as a Celtic-speaking nation, not the whole of France. A brief look at the wikipedia articles on the Orkney and Shetland islands shows a distinct culture there, whereas the image implies they are wholly Scottish.-- (talk) 09:24, 2 August 2009 (UTC)
Because Brittany is the titular nation described as Celtic. Orkney and Shetland are part of Scotland and so are included as Scotland is the titular Celtic nation. The rest of France is not included because to do so would be to suggest that the rest of France is part of Brittany, as it is Brittany that is the titular nation, not France. The same can be said for Cornwall/England. England is not part of Cornwall - i.e. the titular Celtic nation here is Cornwall, not England. Mac Tíre Cowag 08:58, 2 July 2011 (UTC)

A table with sound changes from Proto-Celtic to each descendant...[edit]

... would be handy. Can't find such a thing anywhere, and it should be something rather basic. The Proto-Celtic article only shows the development to Welsh and Cornish, for some reason.-- (talk) 23:16, 3 July 2009 (UTC)

Passive or Intransitive, Infinitive[edit]

The bullet item "an impersonal or autonomous verb form serving as a passive or intransitive" is a bit off (it's a very common but very erroneous statement about modern Celtic languages) - in modern Goidelic (in all three languages) the form in question is a finite impersonal form, and not a passive (since it exists for intransitive verbs such as "be" and "go" and since it has no subject - having a subject would of course be difficult for an impersonal form - either of which on its own is sufficient to disqualify it from being considered a passive) and is not in general intransitive (since this form of a transitive verb is just as transitive as any other finite form of that verb - maybe the confusion arose through someone mistaking the direct object for a subject?). There is no analytic passive of Goidelic verbs, but several compound passive forms, some using rach as an auxiliary (eg theid mo bhualadh - I will be hit or I am habitually hit, chaidh mo bhualadh - I was hit, rachadh mo bhualadh - I would be hit or I used to be hit) others using a reflexive formation (eg tha mi air mo bhualadh - I have been hit, tha mi 'gam bhualadh - I am being hit).

The bullet item "no infinitives, replaced by a quasi-nominal verb form called the verbal noun or verbnoun" is wrong too. There's nothing quasi-nominal about the verbal noun, it's a noun just like any other noun and is not in any sense a verb. It's used to construct various phrases which in other languages are called gerundives, present participles, infinitives, and in many other ways. If you say that modern Goidelic has no infinitives you have to say the same about English because the English infinitive is a phrase just as the Gaelic one is: "Thainig mi 'Ghlaschu a dh'iarraidh leanan" contains a very clear infinitive of intent. And of course most people will say that "there are sights to be seen" contains a passive infinitive, and I can't see how that can be true if "tha seallaidhean ri'm faicinn" doesn't contain the same passive infinitive.MichealT (talk) 22:32, 28 March 2010 (UTC)

Old Celtic ?[edit]

I just did a definition for a given name, and it mentioned that it was composed of several "Old Celtic elements". Where should Old Celtic be redirected too?--Brianann MacAmhlaidh (talk) 08:06, 7 April 2010 (UTC)

Probably Proto-Celtic, although given names are usually identifiably either Goidelic or Brythonic and should be labeled as such. +Angr 08:51, 7 April 2010 (UTC)
or more likely, your definition should not use the term "Old Celtic". --dab (𒁳) 10:51, 29 January 2011 (UTC)

Celtic Language ancestry and descent from ancient Greek[edit]

Its worth noting in here that the parent of Celtic is almost certainly a type of Dorian/Anatolian Greek spoken in the North Eastern Mediterranean around the time of the Trojan war. All evidence points to a vast exodus from Greece/Anatolia after the destruction of the Trojan Super Power of the time. The "Sea peoples" as they were known were free to maraud and colonise whole swathes of the Mediteranean all the way around the Atlantic Sea Board to the West Coasts of Great Britain and France. Of course previous Indo-European colonisers had been there before but the Post Trojan war exodus was large enough to spawn a type of creolic ancient Greek which became Celtic. See books by David Rohl, Steven Oppenheimer, Louis Waddell, etc etc for more evidence. Rohl and others argue for an 800BC Trojan war rather than the tired 1200BC date given by traditionalists. The 800BC date again matches to the approximate date that many believe Celtic took shape.-- (talk) 09:24, 25 April 2010 (UTC)

Just to clarify: there is not the slightest bit of evidence that the Proto-Celtic language is descended from any stage of Greek. It's linguistically utterly impossible. Proto-Celtic and Greek descend from a common ancestor, yes, but they are "sisters", not "mother and daughter". +Angr 05:45, 26 April 2010 (UTC)

Welsh Native Speakers[edit]

"62% speak it daily 88% speak it fluently" - Poorly written, unsourced, and the numbers given in the next column suggest otherwise. Having lived in Wales for the last 7 years, I would be shocked if 88% of Welsh natives actually speak Welsh fluently. Sordyne (talk) 15:28, 15 November 2010 (UTC)

I also thought these figures were a little suspicious. Maybe they have travelled here from the year 1950 or so. --dab (𒁳) 10:46, 29 January 2011 (UTC)
The Census use the measure(Proficiency-in-Welsh in Census 2011 from ONS) "can speak Welsh" as a self-certified characteristic, rewriting that as "fluent in Welsh" is unsupported. It would be quite a surprise if all those from 3yo upwards who described themselves as able to speak Welsh were in fact completely fluent. I've not been able to find a source for the level of fluency at all. I've altered the table to better reflect "can speak Welsh" rather than claiming an unsupported level of fluency. For comparison 0.27% don't speak English at all in Wales; Polish is the likely 2nd main language. Pbhj (talk) 23:08, 12 January 2015 (UTC)

Were British really Celtic ?[edit]

More and more studies show difficulties to characterize as Celtic the British Isles. And why should the Celtic languages not be a modern feature (in Middle-age I mean) in these areas, and not since Antiquity like a lot of current studies say ?--Ghosthammer (talk) 22:38, 28 January 2011 (UTC)

Mnya... Historically, the first clearly identifiable linguistic layer in all of the British Isles is Celtic. I hope you are not referring to "The Origins of the British" (or so) by Mr Oppenheimer, because linguistically, there is not much to be said in his defence... Trigaranus (talk) 09:33, 29 January 2011 (UTC)
oh for crying out loud, Ghosthammer, we have an extensive article about the topic right here. Is it really so difficult to sit down and read the article instead of clicking the "Discussion" tab directly to leave a comment to make sure everybody knows you did not read it.
A "lot of studies" say that the Celtic langauges came to Britain in the Middle Ages? Pray give us a list of such studies. Also, you may want to discuss this at Insular Celts, which is, as the title suggests, our article about Insular Celts. --dab (𒁳) 10:45, 29 January 2011 (UTC)

The order is verb subject object (VSO) in the second half - compare this to English or French (and possibly Continental Celtic) which are normally Subject Verb Object in word order.[edit]

Isn't this irrelevant for imperative constructions? —Preceding unsigned comment added by (talk) 08:03, 30 March 2011 (UTC)

== VSO only accounts for Insular Celtic, Continental Celtic had a word-order of SOV; it was only as a result of Wackernagel's Law, (whereby clitics were always placed in 2nd position, subject being in null position, and such is the fact that the verb was moved from the end of the clause to second position when a suffixed or infixed object pronoun was required), that the verb was transported in Insular Celtic. There is also a case of this in one Transalpine Gaulish inscription, namely siox=ti, whereby the verb has been moved to the front of the clause as a result of the clitic being attached.== Baininscneach (talk) 16:52, 9 April 2011 (UTC)

Use of Flags[edit]

Flags are used in the table of languages to supplement the description of the areas in which the languages are spoken. To me the flag represents the origin of the language in broad terms, rather than where it is spoken, and is a useful visual addition. Even if the language is used elsewhere, as in the case of Irish, I suggest we still use the flag of Ireland to denote the origins of the language. WizOfOz (talk) 16:39, 8 May 2011 (UTC)

The Irish language originated in Ireland over a thousand years ago. It didn't originate in the 26-county state (Republic of Ireland), which was only created in 1921.
The flags in the table ar merely decoration; ther is no need for any of them.
We need to note exactly what that colum represents (the one named "area"). I suggest it be used to list the areas wher the highest numbers of speakers ar found. For Welsh it would be "Wales, England, Argentina (Chubut Province)" for example. ~Asarlaí 17:10, 8 May 2011 (UTC)

Celtic in Patagonia[edit]

Is this info accurate? It´s been stated in the right box, under geographic distribution. — Preceding unsigned comment added by (talk) 20:18, 19 July 2011 (UTC)

Well it's actually modern Welsh in Patagonia, but Welsh is a Celtic language, so it's reasonable to claim that a Celtic language is currently spoken there, though the number of speakers is relatively few. The speakers are descendants of the Welsh immigrants since 1865. Details can be found in the article Y Wladfa. Dbfirs 22:23, 1 August 2011 (UTC)

Clading chart doesn't match article's prose[edit]

The article states "Rather, in the Insular/Continental schema, Celtiberian is usually considered to be the first branch to split from Proto-Celtic, and the remaining group would later have split into Gaulish and Insular Celtic." Yet this is not reflected in the clading diagram (immediately following "How the family tree of the Celtic languages is ordered depends on which hypothesis is used").

Also, it is stated that, in the P/Q model, Gaelic branched off first. I thought it was considered that Celtiberian branched off first also in this model. Jamesdowallen (talk) 12:21, 4 November 2011 (UTC)

It's easier to understand if only the uncontroversial units are used. Disregarding Pictish, which isn't known well enough to classify it with certainty (and thus fit it into the tree with authority) but may have formed a group together with British (Welsh-Cornish-Breton), the five consensus units of Celtic are Celtiberian, Lepontic, Gaulish, British and Goidelic. The two models group them like this:
Insular Celtic model
  • Celtic
    • Celtiberian
    • Lepontic
    • Gaulish
    • Insular Celtic
      • British
      • Goidelic
P/Q-Celtic model
  • Celtic
    • Q-Celtic
      • Celtiberian
      • Goidelic
    • P-Celtic
      • Gaulish (with Lepontic usually lumped into it)
      • British
That's it, in the most basic terms. It's not even known with certainty whether Lepontic reflects Proto-Celtic * as p or kept it because the known inscriptions exhibit no clear examples. (Also, some seemingly Gaulish names such as Sequani or Quariates look Q-Celtic.) The idea that either Celtiberian or Lepontic may have split off first (which doesn't necessarily imply that all the other units have undergone common innovations which could justify the postulation of a separate branch for the rest of Celtic), or that Gaulish and Insular Celtic may have formed a "North Celtic" branch, which is hard to justify with linguistic evidence and rather intuited on historical, archaeological and geographical grounds, are just a refinement of the Insular Celtic model. For the P/Q-Celtic model, they are irrelevant, anyway. --Florian Blaschke (talk) 21:48, 27 November 2012 (UTC)

Updade[edit] — Preceding unsigned comment added by (talk) 22:34, 24 April 2013 (UTC)

Article needs to be bought.

Still here are some parts of the text:

Here is part of the text:

From around 2800 BC, the LNE Bell Beaker culture emerged from the Iberian Peninsula to form one of the first pan-European archaeological complexes. This cultural phenomenon is recognised by a distinctive package of rich grave goods including the eponymous bell-shaped ceramic beakers. The genetic affinities between Central Europe’s Bell Beakers and present-day Iberian populations (Fig. 2) is striking and throws fresh light on long-disputed archaeological models3. We suggest these data indicate a considerable genetic influx from the West during the LNE. These far-Western genetic affinities of Mittelelbe-Saale’s Bell Beaker folk may also have intriguing linguistic implications, as the archaeologically-identified eastward movement of the Bell Beaker culture has recently been linked to the initial spread of the Celtic language family across Western Europe39. This hypothesis suggests that early members of the Celtic language family (for example, Tartessian)40 initially developed from Indo-European precursors in Iberia and subsequently spread throughout the Atlantic Zone; before a period of rapid mobility, reflected by the Beaker phenomenon, carried Celtic languages across much of Western Europe. This idea not only challenges traditional views of a linguistic spread of Celtic westwards from Central Europe during the Iron Age, but also implies that Indo-European languages arrived in Western Europe substantially earlier, presumably with the arrival of farming from the Near East41.

It seems that genetic evidence supporting the Iberian hypothesis, paired with archaelogy, is ever-growing. A lot has been already published concerning the Iberian-Basque-British Isles connection. Now this seems to continue in other European areas like Germnay.

Pipon — Preceding unsigned comment added by (talk) 23:04, 24 April 2013 (UTC)

Given that genetic and archaeological data cannot be directly tied to languages, these data cannot support nor refute any particular hypothesis regarding Celtic origins. In fact, if we take the genetic and archaeological data seriously, they would rather militate against an identification of the "Beaker folk" as Indo-European (in short: highly improbable), much less Celtic speakers (implausible to the extreme). The (linguistic) ancestors of the Basque would be a much better fit. Please refer to my comments at Talk:Beaker culture#Latest update.. --Florian Blaschke (talk) 20:55, 2 July 2013 (UTC)


I think the map is currently somewhat misleading. The solid colours in Cornwall and IoM give the impression that these geographical areas are entirely Cornish/Manx speaking, which they're not. Both these revived languages are special cases, and its true that there is no geographical focus within Cornwall like there is with Welsh/Gaelic/Irish - but I think perhaps using the striped colour coding or using a different pattern would be more appropriate for both. Also, I'm not quite sure why the whole of Ireland is striped, but the whole of Wales isn't. Cardiff isn't predominately Welsh speaking, but it's still spoken by 11% of the Cardiff population (2011 census), so having a solid colour in Cornwall and none at all in Cardiff/SE Wales does not reflect the reality. 幽Sweorcan (talk) 09:14, 29 March 2014 (UTC)

I agree. Having the whole of Ireland striped whilst the South east corner of Wales is blank is incorrect. And having Cornwall and the Isle of Man solid colour is also incorrect. I have no idea how to alter such maps, but surely deleting them and leaving nothing there would be preferable to leaving the incorrect, very misleading information on the site?mUnwilling to do that without the thoughts of anyone else on the subject, though. Ceiniog (talk) — Preceding undated comment added 00:34, 15 April 2014 (UTC)
Will wait for a bit to see if anyone has anything to say before deleting the map. Ceiniog (talk) 00:40, 15 April 2014 (UTC)
I agree completely. Articles and their accompanying illustrations/tables/graphs should be informative and fact-based. This image is completely misleading and should be either corrected (likewise, I do not know how to do that) or deleted. Mac Tíre Cowag 11:40, 16 April 2014 (UTC)


A fair amount of the section on classification as it now stands is taken directly and without attribution from A Grammar of Modern Indo-European (ISBN 1461022134v, see here). — Preceding unsigned comment added by Tzetzes (talkcontribs) 23:39, 16 April 2014 (UTC)

map deleted[edit]

As no one objected, I've deleted the map. — Preceding unsigned comment added by Ceiniog (talkcontribs) 23:24, 19 May 2014 (UTC)

Comparison table[edit]

The comparison table in the article could use some revision - way too many loan words in it and not enough directly cognate words for a proper comparison. There should also be a column for Proto-Celtic antecedents. Anyone want to pitch in to create something better? Cagwinn (talk) 05:07, 12 June 2014 (UTC)

Problems under "Subdivisions" in box[edit]

Hi, I don't want to just jump in and edit it without discussion, but there are some problems with the content under "subdivisions" in the box at the top right.

1) It shows a single classification scheme, where (as the good work in the body of the article shows) there are two competing classification schemes. One divides Celtic languages between Continental (Continental Celtic) and Insular, and then divides Insular between Goidelic (Irish/Scots/Max Gaelic) and Brythonic/Brittonic (Welsh/Breton/Cornish/Cumbric). The other classifies between P-Celtic languages (most Continental Celtic plus Welsh/Breton/Cornish/Cumbric) and Q-Celtic (Irish/Scots/Manx Gaelic).

The classification scheme shown under "subdivisions" is a mixture of the two. If it is necessary to go with a single scheme, it would be preferable to use the Continental/Insular one, as this is the more widely accepted (and is reflected in the phylogenies linked from the page). However, it would be better to show both schemes as alternatives to each other, as the matter is not settled. Either way, it would be good to show accurate correspondences between the two schemes.

2) The set of languages within Insular Celtic is not the same as the set within Q-Celtic. The Brythonic languages are insular but P-Celtic in form. (A commonly used example is that the Brythonic Old Welsh for son is Map, where the Goidelic Old Irish for son is Maq or Mac).

3) P-Celtic languages are not extinct. Welsh and Breton are still alive, and there are serious efforts to revive Cornish.

I shouldn't need to provide references for these points as I think they are supported by a careful reading of the main article, but if there are any details in the above that people think are not supported by the existing text I will provide suitable references.

Edit: See the Brittonic Languages page for clarity about Brittonic/Brythonic languages being P-Celtic in form and Welsh, Breton and Cornish being among them. Greggjc (talk) 11:49, 21 February 2015 (UTC)

"Old European" and ?[edit]

This article refers to a language called "Old European". What language is that? It looks very much like original research or, given that it comes from a source, as a bad fringe theory that is WP:UNDUE. Same thing with the hypothesis that Q-Celtic originated in Iberia. There is currently not one linguistic source in the article supporting that, yet it is reported as a possibility.Jeppiz (talk) 18:11, 18 April 2015 (UTC)

"Old European" is what some linguists in the 20th century dubbed the pre-Germanic/Celtic/Illyrian/Italic/etc. language(s) of Europe, which has allegedly only left scant traces in place/river names. Cagwinn (talk) 17:39, 19 April 2015 (UTC)

Unclear map[edit]

The map in the infobox is unclear to say the least. What does dark green and bright green represent? And regardless of what they do represent, where are the sources behind these representations?Jeppiz (talk) 21:09, 24 July 2015 (UTC)

Both greens indicate areas where Modern Celtic languages are spoken (at least traditionally); the dark green indicates regions with highest density of Modern Celtic speakers. Cagwinn (talk) 04:25, 25 July 2015 (UTC)
If it would do that then all would be good (given a source was provided) but the problem is that it doesn't.
  • For Ireland, Wales, Cornwall and Man, the map shows the entire country in light greed; for Brittany and Scotland it doesn't.
  • Just what is meant by the light green area in Scotland is beyond me, I'm afraid. It appears to be the Gàidhealtachd with some random light green patches in other places.
  • For Brittany there's even more of mystery. Some areas that have never been Breton speaking are coloured in light green. If someone has a reliable source that Nantes or its eastern hinterland has ever spoken Breton, it would be nice to see.
  • Still in Brittany, the dark green is as mystifying as the light green. At what period in time was Brest more Breton-speaking than the hinterland of Vannes? No area in Brittany is mainly Breton speaking but some areas that were Breton until very recently and where there are still lots of native Breton speakers are in light green, some areas that have not been Breton speaking for a very long time are in dark green.
  • For Ireland, the map gives the official Gaeltachtaí. They have nothing to do with reality, of course. Irish speakers can be found outside these areas as well. As for the dark green, several of these areas aren't Irish speaking at all. Hearing Irish in eastern Uíbh Ráthach, not to mention the areas east of Galways, is but a long lost memory.
So once again, the map is probably well intended but it is a prime example of WP:OR that should be removed from all articles.Jeppiz (talk) 09:14, 25 July 2015 (UTC)
I agree with the concerns about the map. Mabuska (talk) 11:10, 25 July 2015 (UTC)
Agree too. We do already have maps for Gaelic and Welsh sourced to the 2011 British and Irish censuses [1], [2] and [3]. It shouldn't be too difficult to choose one or more percentages (50%/70%?) and present a composite map. Not so sure about Breton - there's this but I'm unsure of the source. DeCausa (talk) 12:00, 25 July 2015 (UTC)

As per the consensus here, I've removed the map. The only dissenting user immediately reverted the removal, without even providing a reason. Further WP:OWN-violations of that kind will be reported. Not agreeing with a consensus is not a reason to ignore it. Jeppiz (talk) 17:46, 9 September 2015 (UTC)

After Cagwinn's edit war[edit]

Having started the discussion in July and achieved consensus that the map is erroneous, I still waited 1½ month befodre removing it to give the sole dissenting view (Cagwinn) a chance to make an argument. Cagwinn never bothered to comment again, but instead launched a full-out edit war against the consensus, and is currently blocked. Cagwinn's violations of WP:3RR and WP:OWN are problematic, as so is his refusal to WP:HEAR. His solution was to include labels, but labels aren't the problem. The map is utterly wrong, as anyone with even a basic knowledge about Celtic languages sees immediately. Here are the same errors I listed in July and which remain, none of which Cagwinn even bothered to address.

  • For Ireland, Wales, Cornwall and Man, the map shows the entire country in light greed; for Brittany and Scotland it doesn't.
  • Just what is meant by the light green area in Scotland is beyond me, I'm afraid. It appears to be the Gàidhealtachd with some random light green patches in other places. It does not correspond to any given moment in time.
  • For Brittany there's even more of mystery. Some areas that have never been Breton speaking are coloured in light green. If someone has a reliable source that Nantes or its eastern hinterland has ever spoken Breton, it would be nice to see. Some areas that were Breton speaking long ago are in grey.
  • Still in Brittany, the dark green is as mystifying as the light green. At what period in time was Brest more Breton-speaking than the hinterland of Vannes? No area in Brittany is mainly Breton speaking but some areas that were Breton until very recently and where there are still lots of native Breton speakers are in light green, some areas that have not been Breton speaking for a very long time are in dark green.
  • For Ireland, the map gives the official Gaeltachtaí. They have nothing to do with reality, of course. Irish speakers can be found outside these areas as well. As for the dark green, several of these areas aren't Irish speaking at all. Hearing Irish in eastern Uíbh Ráthach, not to mention the areas east of Galways, is but a long lost memory.

So once again, the map is probably well intended but it is a prime example of WP:OR that should be removed from all articles. Jeppiz (talk) 08:33, 10 September 2015 (UTC)

It's not a serious proposition to keep that map - it's amateurish nonsense to have argued for keeping it. It's gone and we can move on. The question is to find a properly sourced map that can replace it. DeCausa (talk) 21:32, 10 September 2015 (UTC)
You're right. It's crazy how much one disruptive user can disturb articles, even when there's a complete consensus among all users with any knowledge of the topic. As you say, the question is where to find a good map. Ireland, Wales, Scotland and Man all have censuses giving exact information. I'm not sure about Cornwall, and I know there is no such census in Brittany. To compound matters, most of those censuses only measure language skills. So a fluent native speaker who speaks Welsh daily will fall into the same category as someone who studied Welsh for a semester 20 years ago and haven't spoken it since. Is the second really a Welsh speaker? If we used the same criteria, we could make all of England seem almost completely French speaking by counting anyone with rudimentary school day memories of French, just like the UK censuses do for the Celtic languages. In Ireland we get around this problem as we have good data for daily speakers, and in Cornwall and Man it's safe (and sourceable) to assume no area has even 5% speakers. But that leaves Scotland, Wales and Brittany, all of which definitely have areas with significant proportions of actual, fluent and regular speakers, but none of which have censuses measuring that. Jeppiz (talk) 23:30, 10 September 2015 (UTC)
The "factual" problems you all are arguing about are more methodological problems. The question of what constitutes a speaker of a language is a difficult one that socio-linguists have not even reached consensus on. The shading of Man as such is fine. There is a Manx speaking presence across island with an immersion school in the center (at St. Johns). Students at this school travel from all over the island. Man's secondary schools also have a Manx speaking presence and while there are no current native speakers, there is a sizable number of second language learners meeting for community conversation groups island wide. As for Scotland, there is a sizable presence of Gaelic in Glasgow and Edinburgh (the two green patches in the south). The rest of the light green region represents the disperse distribution of Gaelic speakers. Ireland is fine. Anyone plugged into the Irish language scene there can tell you that. Cornwall is perhaps the only erroneous bit of the map but until someone wants to take the time to do the original research as to the exact distribution of second language speakers (of which there are many), we can only say that it is spoken there and hence a county wide shading is appropriate. 10:09, 8 December 2015 — Preceding unsigned comment added by (talk)
Even if you would be right (and you're not), that would still constitute original research. It is not based on any source, rather on an individual user's (erroneous) conceptions. Jeppiz (talk) 15:32, 8 December 2015 (UTC)
You say that I am not correct. Please cite me your sources and I will provide you with mine. (You will have to give me a bit as it will take me a while to compile the list, maybe a day or two?) I am more than happy to listen to your argument, but please substantiate it. The previous discussion on this talk page do not sufficiently address primary or secondary sources. Also there appears to be a difference here between an informed and uninformed editor. This is a subject which is highly technical yet unfortunately has taken hold of the popular imagination. Expertise in the field of the modern Celtic languages is quite limited and the number of speakers small. Many questions (such as what constitutes a speaker) are still open ended in the scholarly debates. There is no reason why a talk page on Wikipedia composed of apparently non-experts would be equipped to answer these questions definitively when those of us who do this for a living continue to struggle with some of these ideas. Maps are valuable, but can never be exact. The question comes down to whether one can produce a map that is good enough, not if one can produce a map which is "factual". Unfortunately there is so such thing as "fact" in the world of socio-linguistic inquiry, too many things are uncertain. Also census records do not provide reliable nor exact information on minority languages, take for instance the problem outlined by Jeppiz on 23:30, 10 September 2015. For reliable data one must rely on the work of the very few socio-linguists that exist in this field. I will provide what I can with regard to reliable data in my next post. 12:21, 8 December 2015. — Preceding unsigned comment added by (talk)
The removed map is unsourced and is clearly misleading. If you are able to provide a sourced map, then please do so. But there is no consensus to insert the misleading map. DeCausa (talk) 17:53, 8 December 2015 (UTC)
The map is not misleading at all and there is no consensus here among people WHO ACTUALLY KNOW ANYTHING ABOUT MODERN CELTIC LANGUAGES; rather it is only a weak consensus among busybody Wikipedia bureaucrats with no specialist knowledge on Celtic languages whatsoever (as far as I can tell from reviewing your edit histories). Please feel free to provide us with your credentials on Celtic languages that allow you to label this map as "misleading". Cagwinn (talk) 18:20, 8 December 2015 (UTC)
Well, you are clearly a very great expert because you shout in capitals and tell everyone you are a very great expert. And of course that's how Wikipedia works. Only those that say they are an expert are counted in any consensus (and they should be particularly believed if they shout that they are an expert) and because they say they are experts on a subject they are exempt from producing sources and their opinions should be accepted by everyone else without the tedious inconvenience of having to justify them per WP:V. Yes, that's exactly how Wikipedia works and it works very well: comment on the contributor, not on content. DeCausa (talk) 19:36, 8 December 2015 (UTC)
After 30 years of studying both modern and ancient Celtic languages and, for the past 15 years, moderating multiple academic message boards dedicated to Celtic languages (the membership of which includes many of today's top Celticists) and being cited as a reliable source on Celtic linguistics in several academic books and articles, I have no problem declaring that I am more of an expert on the subject than you are. Cagwinn (talk) 00:03, 9 December 2015 (UTC)
What you fails to understand is that nobody cares. Even if you were a great expert (and you're not) it would not matter one bit, as Wikipedia works by reliable sources. Appeals to authority with no sources don't work, appeals to one's self-proclaimed expertise even less. And number of users have pointed out the errors of the map. It does not accurately represent the present areas where Celtic languages are spoken and it does not represent any given point in history either. Jeppiz (talk) 00:08, 9 December 2015 (UTC)
You just don't know what you are talking about - on all points.Cagwinn (talk) 00:15, 9 December 2015 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────The map is most definitely misleading for the reasons Jeppiz gave back in September. It's clearly made up from an amalgamation of information, not all of which is from the same period, or otherwise jives across all these territories. In some areas, the whole country/territory is shaded light green, but not in others. There was no one point in history that Celtic languages were spoken in all these highlighted areas. What we should have, in my opinion, is a map that highlights the entire territory of Ireland, Scotland, Wales, Man, Cornwall, and Brittany as regions where Celtic languages were spoken into the modern era. Within those we could try to identify places where the highest number of Celtic speakers live, perhaps based on Census data as in this map. In the meantime, we could just use a map like this one identifying these six places as the regions where there was a Celtic language spoken into modern times. The map was a nice effort, but it's got too many problems to be useful.--Cúchullain t/c 22:23, 8 December 2015 (UTC)

The original map - - had dark green indicating regions with a majority usage of the native Celtic language and light green indicating minority usage; you can compare it against the following collection of similar maps:
Thank you, Cagwinn. We can see that it's unlikely there will be maps of these six regions that use the borders shown in this map. Moving on, I think we could actually put together a good map based on the maps you identify. Meaning, one map with the information of these five, and also highlighting Cornwall and Mann, and with each of the sources cited.--Cúchullain t/c 02:19, 9 December 2015 (UTC)
Yes, these are relevant maps, and the contradict the removed map on several points, illustrating nicely why it was removed. If a better map, reflecting these maps, was made, it would of contribute to the article Jeppiz (talk) 02:24, 9 December 2015 (UTC)
Agree. See my post of 25 July 2015 at the beginning of the parent thread to this one were I link to 3 census maps plus the Breton map Cagwinn linked to suggesting just that. DeCausa (talk) 07:47, 9 December 2015 (UTC)
I see that now. This discussion became pointlessly heated and unproductive and it's a shame it happened again.
For now, I'm going to add the "Celtic nations" map so at least people can see where the areas with modern Celtic languages are located. Scratch that, for this article, this map of Celts through time will be better. It's got some problems of its own, but it's sourced. Perhaps we can include the Celtic nations map later in the article.--Cúchullain t/c 14:33, 9 December 2015 (UTC)

External links modified[edit]

Hello fellow Wikipedians,

I have just modified 3 external links on Celtic languages. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

Question? Archived sources still need to be checked

Cheers.—InternetArchiveBot (Report bug) 17:58, 11 September 2016 (UTC)