Talk:Insular Celtic languages

Sprachbund vs. genetic unit (cleanup request)

 * the section "Language contact vs. common origin (cleanup request)" below supersedes the comments i've made here earlier --Green ink 11:22, 1 August 2006 (UTC)

I don't see that this article addresses the idea of a "Sprachbund" at all; nor have I ever heard of such a suggestion before. As I read it, the article is entirely about Insular Celtic as a genetic unit. User:Angr 16:25, 31 July 2006 (UTC)


 * copied from User talk:Angr:

on the article's talk page, you say that the article wasn't about the sprachbund situation of that same set of languages. i disagree.
 * 1) you're right, the article primarily is about the genetic unit, not about the sprachbund.
 * 2) but yes, the sprachbund needs to be mentioned. after all, what proponents of the i.c. hypothesis say is that of the innovations shared between goidelic and brythonic languages, not all can be attributed to the sprachbund situation.
 * 3) for practical purposes, such as organizing the celtic studies department of a university, both the sprachbund and the genetic unit contribute to making it useful to treat these languages together as a topic of study.

i don't think the sprachbund issue warrants a page of its own. i do think it should be separated out from discussion in the present state of the article; but after that, it should be merged into the insular celtic languages article. put more simply, the article needs to be clarified.

all i've been planning to do was express my (i think: relevant) confusion, and not to do anything else about it after that. i don't do flame wars, and with very few exceptions, i only read wikipedia. no hard feelings. --Green ink 17:36, 31 July 2006 (UTC)


 * I'm sorry if I gave the impression I was being defensive or offended by your comment! If there's anything verifiable by means of reliable sources on Insular Celtic as a Sprachbund, by all means put it in! All I'm saying is that when I read the article in its current state, it seems to me to be only about IC as a genetic unit. I don't see anything in the current text that suggests otherwise. But the current article certainly needs sources for the information about the genetic unit IC too! User:Angr 17:48, 31 July 2006 (UTC)


 * partial and abridged copy from User talk:Green ink:

User:Angr [has been] right, "sprachbund" was off the mark. [...] at the same time, i would maintain most of what i've said, after, of course, substituting my erroneous uses of the term "sprachbund" simply by "mutual language contact". [...]  --Green ink 10:47, 1 August 2006 (UTC)

Language contact vs. common origin (cleanup request)
The article appears to confound two issues: Insular Celtic language contact and an Insular Celtic common origin. For that reason, I've requested a cleanup. However, it does make sense to have them together in the same article, only they should be explained separately before discussing how these issues come together. I'd suggest having a separate section for each in the introduction of this article. In detail:
 * Insular Celtic language contact :Due to language contact, these languages have shared innovations after they were or became separate languages. This could be an outright sprachbund (emergence of traits originating from neither parent language).  More plausibly, this could be heavy adstratal influence of Brythonic on Goidelic.  Example: absence of a third (=neuter) gender, which Old Irish still had but modern Irish, like the Brythonic languages, doesn't.
 * Insular Celtic genetic unit :These languages share a common ancestral language (=I.C. proto-language) that is not an ancestor of any other known language. Therefore, they may also share innovations because those occurred before they split up into separate languages.

Importantly, these two issues are not mutually exclusive. The relevance of language contact has never been disputed, as its effects have continued into historic times, see above. The Insular Celtic Hypothesis, however, turns on evidence that some shared innovations cannot as plausibly be explained as the result of language contact and therefore must have occurred before the languages even separated. Demonstrating this would presumably take too much space for an encyclopedic article. Nevertheless, the general line of argument should be summarized and suitable references cited (also) for further reading. --green ink (t) 08:32, 1 August 2006 (UTC)


 * My answers are mostly at User talk:Green ink. User:Angr 11:26, 1 August 2006 (UTC)

Substrate?
There are features of the Insular Celtic languages that do not correspond to IE and have been suggested as coming from a pre-Celtic substrate. Should this problem be addressed here? If not, where? Adresia (talk) 11:01, 25 January 2008 (UTC)


 * The fact that some of the more exotic features of Insular Celtic have been attributed to a possible substrate effect could be mentioned, if it's well sourced, but I wouldn't want to go into detail about the speculations involving those possible substrate effects, because most of the work that's been done on it has ventured very close to the pseudoscientific and would probably fall afoul of WP:FRINGE. —Angr If you've written a quality article... 15:32, 25 January 2008 (UTC)


 * That and unsophisticated, using simple models, or inventing them. But most importantly there is no support from population genetics. Attempts have been made to explain the features in the context of extreme (P)PIE archaisms, but it would be a fairly monstrous hassle for me to assemble the material so I won't be the one here. Nora lives (talk) 07:51, 28 June 2011 (UTC)


 * Well, I'd hardly call that "most importantly", since there is no correlation between the genetic relationship of people and the evolution of languages. The absence of a DNA link between speakers of Celtic languages and speakers of the languages asserted to have had a substrate effect on Insular Celtic (e.g. Afro-Asiatic languages) doesn't actually provide evidence against a possible connection. And if DNA evidence were present, it would not provide support for any alleged substrate effect. The two issues are simply completely unrelated to each other. —Angr (talk) 13:33, 28 June 2011 (UTC)


 * That is a minority view, and one several years old. In Europe and elsewhere we do find haplogroups and haplotypes associated with cultures and language families. The situation is complex. I see you don't follow it, hence "simply", so I don't know what to say except ask you to respond to what else I said. Nora lives (talk) 17:07, 28 June 2011 (UTC)


 * The idea that haplogroups and haplotypes in people's DNA can be associated with the languages they speak is pure pseudoscience, or is there some genetic affiliation between English-speaking Native Americans, African Americans, Australian Aboriginals and white people from England that is distinct from some other genetic affiliation between Spanish-speaking Native Americans, Afro-Peruvians, Filipinos and white people from Spain? Geneticists have shown that the English-speaking population of England is biologically very closely related to the Celtic-speaking population of neighboring countries, but that doesn't make English a Celtic language or Welsh and Gaelic Germanic languages. People's language can change over time so that people in one time period are speaking a language that is genetically completely unrelated to the language of their great-grandparents, despite the fact that they've inherited all their great-grandparents' DNA. I know it's currently very fashionable to look to genetics to contribute to questions of historical linguistics, but all it can do is show that a known linguistic relationship (say between Dutch and German) is paralleled by a genetic relationship (say between Dutch people and German people). The genetic evidence can't be used to influence the linguistic hypotheses. So if genetics were to show that Dutch people and German people are not closely related, that wouldn't mean the languages are suddenly less closely related than we thought; and if genetics shows that German people and Polish people are closely related, that doesn't mean the German and Polish languages are suddenly more closely related than we thought. Of course it's gratifying when a proposed linguistic grouping (especially a geographically distant one like Dené-Yeniseian) is paralleled by a genetic affiliation (I don't know whether that's the case with Dené-Yeniseian, but it would be gratifying if it were), but genetic evidence is not linguistic evidence and it cannot honestly be used either in support of or as counterevidence against linguistic hypotheses. The absence of any genetic affiliation between speakers of Insular Celtic languages and speakers of Afro-Asiatic languages simply does not contribute to the question of whether there is an Afro-Asiatic substrate in Insular Celtic (and not just because it's an argument from silence anyway - even if there were a genetic affiliation, it would still have no bearing on the question). —Angr (talk) 06:25, 29 June 2011 (UTC)


 * Quite the treatise. Write the paper. I don't know who you think you are trying to lecture me here. When I entered above I was agreeing with you, but maybe you didn't get that, and now you're calling me fashionable. That's telling me I don't know what I'm talking about (fighting words), when you can't even respond to what I know because you don't even have an amateur background. We're finished. Nora lives (talk) 14:24, 29 June 2011 (UTC)


 * I know we agree with each other; I thought you were the one who didn't realize that. And you were the one who started with the fighting words "minority view", "several years old", and "I see you don't follow it", and have followed it up with more fighting words "amateur background" (I have a Ph.D. and a habilitation in linguistics, though that doesn't make my Wikipedia contributions more valuable than those of others who don't). —Angr (talk) 14:39, 29 June 2011 (UTC)


 * I recognize your competence in linguistics. Personally I have some, but I'm better known for being a drunk and a drug addict. What I do have is a competent amateur knowledge of population genetics, at least what you might call "suitable for informed contributions", and you don't seem to be read in it at all. Plus "linguistics" can mean anything. You might be a specialist in Proto-Baltic or Basque with fifty published papers and not know a thing about anything else. Nora lives (talk) 14:56, 29 June 2011 (UTC)


 * If you have a competent amateur knowledge of population genetics, then you should know that the fact that two groups of people (ethnicities or whatever) are genetically related to each other doesn't actually tell you anything at all about the languages they speak, because languages are not passed down from gametes to zygotes. If linguists find that language A and language B are related, and geneticists find that ethnicity A and ethnicity B are also related, that's a very gratifying conclusion, but if geneticists find that ethnicities A and B are not related, that in no way weakens the linguistic hypothesis that the languages are related. And if linguists are trying to decide whether or not languages A and B are related (or were once in contact), they can use only linguistic evidence: genetic evidence that ethnicities A and B are or are not related doesn't enter into it. We agree there's no evidence for an Afro-Asiatic (or Basque or Etruscan or whatever) substrate in the Insular Celtic languages, but we disagree on what would qualify as evidence. You said, "most importantly there is no support from population genetics", but the lack of support from population genetics is not, in fact, important. For one thing, it's an argument from silence, and those are never very strong. Furthermore, even if there were Afro-Asiatic (Basque, Etruscan, ...) genes found in the British Isles, that would not strengthen the hypothesis of an Afro-Asiatic (Basque, Etruscan, ...) substrate in the Celtic languages at all. Of course it's interesting when linguistic relationships and genetic relationships match up, but linguistic hypotheses can only be tested against linguistic evidence, just as genetic hypotheses can only be tested against genetic evidence. —Angr (talk) 15:30, 29 June 2011 (UTC)


 * You're not teaching me anything, and all I see are straw men, put up by a guy who doesn't like his language dirty with too much culture or women and babies. For some people it's a religion.
 * I've pretty much lost hope you have anything to say about the Insular Celtic languages. If you've seen it, do you happen to recall the title and author of a paper explaining where the VSO came from, from within Celtic? I can't remember. At least one local dialect of Italian is compared because it's creeping into VSO or might be. Then there is the argument that Proto-Balto-Slavic or Proto-Slavic was/were. I got very sick from alcohol poisoning over the course of '07-'08 and lost a lot. I know I'll never get back in, but I was moving away from pure linguistics anyway. Most of it is conceptually unsophisticated, including anything generative of course, and poorly informed by other disciplines. And slow. Lehmann got it better and I enjoyed him. His article should mention he came to believe the generative approach was vacuous and said so in his Theoretical Bases of Indo-European Linguistics. That I remember.
 * You do know that Bernard Comrie is interested in working with population geneticists? I win. Nora lives (talk) 17:50, 29 June 2011 (UTC)


 * There are lots of papers speculating where VSO word order might have come from; I don't know which one you mean. As for Comrie, I'm glad if he wants to work together with population geneticists. Maybe he'll be able to keep some of their more outrageous speculations in check. —Angr (talk) 05:13, 30 June 2011 (UTC)


 * There is without any doubt a class system, with the comparative linguists at the top, the archaeologists in the middle, and the population geneticists at the bottom. Twenty years ago Mallory went on and on about the power of the Indo-Europeanists and their arcane knowledge. Yes, they are a powerful cult, the one that inspired the Nazis by accident, which as you know almost destroyed them. The Jews still hate them, which is unfortunate. Historical linguists and their close kin are intimidating to many. I almost became one and remember the awe. You get to play with the names of gods like toys and forget what it sounds like to the uninitiated. But what am I talking about? (I have some interesting followers.)
 * I own a copy of Comrie's Language Universals and Linguistic Typology and he is essentially a comparativist. His contribution to Ramat & Ramat introduced me to Jerzy Kuryłowicz (brilliant but the slowest read ever). However his influence with the pretentious dickheads in population genetics I can't tell because I've been out since he started to get in. Have a look at what he's been up to. That paper I haven't seen yet.
 * I'll get back to you about the other. And there is another I want even more, where an Italian linguist reclassifies Goidelic as something closer to Italic than the official Celtic. Seen that one? Four or five years old. I want that one badly. Nora lives (talk) 06:52, 30 June 2011 (UTC)


 * So that that will make more sense above, I was getting into population genetics, had made connections, and was beginning a major study of the lineages in Scandinavia. My professor of Germanic philology and mythology, who studied under Werner Winter, was impressed and wanted to set me up with John Lindow of Berkeley once I got to Uppsala University, where I was supposed to attend for a year and then either stay there for graduate school or go to Ireland. Having done a huge amount of indepedendent study (for an undergrad) of all things Indo-European, I was coming in particularly strong. One major issue of the time was the age of the various haplogroups in Europe and this was affecting how even some linguists viewed the spread of the languages. Since then R1b has been dated using a superior method and the majority of western europeans are no longer thought to be male line descendants of Paleolithic non-Indo-Europeans. That fight is over. Nora lives (talk) 03:36, 1 July 2011 (UTC)


 * I still consider myself some breed of Indo-Europeanist or Indo-Germanicist in a way but I hate the Aryan nonsense you still see in different forms on occasion. A lot of people still think that way, probably. For those readers who do not understand, Indo-European, Indo-Germanic, and Aryan were once all synonymous, and the Aryan Race Theory was the accidental child of Indo-European studies and German nationalism. Nora lives (talk) 04:10, 1 July 2011 (UTC)


 * It's not about the body, it's about the spirit. Nora lives (talk) 04:27, 1 July 2011 (UTC)

First, I am not a linguist myself, but I am a writer and a writing consultant. The text beneath the heading "Possible Afro-Asiatic substratum" doesn't say anything important about either the possible substratum or Insular Celtic Languages. For example, who are these "well known linguists" who are adherents to the theory? What did they say? Is it controversial? Is there any evidence? The "further work on the theory by Shisha-Halevy," what was this work? What did it add to the others' hypotheses? Why should we believe it--(i.e. it's not enough for them to be well-known linguists, since the audience of wikipedia are readers who are uninitiated to the community). Lastly, what were the criticisms of the theory by Isaac and McCone? What are Gensler's twenty points and Isaac's responses?

What I'm getting at here is not that this section is somehow improper, but that it says functionally nothing about the subject. If this subject is important, it needs to be drastically expanded. If not, it needs to be deleted or moved to it's own article. Cuthalion1 (talk) 21:24, 18 May 2012 (UTC)


 * In my opinion, the paragraph portrays the issue accurately, even though I agree more details would be desirable. The subject is certainly of some importance in Celtic studies, but still a marginal issue overall. The idea that the typological aberrance of the Insular Celtic languages directly goes back to the influence of a substrate language is or was supported by some notable linguists indeed and seems to retain some popularity outside expert circles, but as far as I understand the issue, the evidence is just insufficient to decide if the idea has any merit. The problem is that while we can, for principal reasons, be certain that non-Indo-European languages (possibly also non-Celtic Indo-European languages) must indeed have spoken in the British Isles before the spread of Celtic gradually drove them to extinction, it seems we cannot reconstruct anything reasonably certain about them. There are a couple of only tantalising hints, but the evidence so far adduced is just mostly far too weak or ambiguous.
 * The problem with the hypothesis that a substratum of a certain type has somehow influenced the development path of the Insular Celtic languages, while it makes sense in principle and is even plausible (analogous developments along these very lines can be demonstrated to have taken place in different regions and periods), as there is so little evidence known that could help further its credibility, in this specific case, it appears to be a frustrating dead-end of historical-linguistic enquiry – so far at least. Should anyone come up with a superior, even compelling explanation for the typological drift that has apparently occurred in the British Isles, I'm sure historical linguists would be delighted. That said, perhaps there is already something amiss at the foundation of the idea in its traditional form: I believe it was Schrijver who once pointed out that hallmark traits such as the Insular Celtic verbal complex (which led to the typical absolute/conjunctive distinction and incorporated object markers, as well as presumably the verb-initial word order) have actually evolved while Celtic was still spoken on the continent, and that the Insular Celtic languages have drifted away from this type since then. --Florian Blaschke (talk) 18:22, 22 June 2013 (UTC)


 * Also, if the subject is not important, moving it to its own article makes no sense, quite the opposite: Then it should stay here, the way it is now. I believe this state reflects it importance rather accurately: not altogether irrelevant for those interested in Celtic languages but still not terribly important. --Florian Blaschke (talk) 18:26, 22 June 2013 (UTC)

Wagner...?
Did Wagner the chess player also support the substratum theory? Just curious, as he is not introduced in any way as a linguist. Please check that link. Thx. 66.30.4.97 (talk) 04:33, 27 January 2010 (UTC)


 * A different Heinrich Wagner (1923–88), for whom no Wikipedia article yet exists, was meant. I changed the link to the red link Heinrich Wagner (linguist) so an article can be written at that name whenever someone gets around to it. Googling Heinrich Wagner Celtic gets some info about him. Thanks for pointing out the error! +Angr 07:12, 27 January 2010 (UTC)

Level of Detail on Grammar
At http://en.wikipedia.org/wiki/Anglo-Saxon_invasion_of_Britain#Linguistic_evidence, we find: Celtic is claimed as a cause in the change from a synthetic language [in Anglo-Saxon] to an analytic language [in Middle English] (expressing meaning through word endings to expressing meaning through word order), and the use of the progressive (e.g. I am reading). Celtic influence is also used to explain puzzling elements in English, for instance, dependence on the semantically neutral DO ('I don't know' rather than 'I know not') and the lack of an external possessor in English, a construct which is present in all major European languages except Celtic. (English uses 'then someone gouged out his eyes' but never 'then someone gouged him the eyes out' which is the Germanic structure.) The above quote implies that scholars have inferred more details about the grammar of at least one insular Celtic language than is detailed here. It would be interesting to be able to come here (or to the article on the British language) and see something about the grammatical features that may have influenced the evolution of English.

Jack Waugh (talk) 02:40, 2 September 2011 (UTC)


 * Uhm, no. The proposed (Insular) Celtic substratum in Middle (and Early Modern) English is tied to British Celtic specifically, especially the long extinct Cumbrian, but also to now-extinct dialects of Cornish and Welsh in southwestern England, not completely different Insular Celtic languages. If at all, Goidelic and Pictish have influenced Middle (and Early Modern) English only marginally. Check Brittonicisms in English. --Florian Blaschke (talk) 18:31, 29 September 2012 (UTC)

Content fork
Per WP:Content fork, we should not be duplicating our classifications across articles. That inevitably causes problems with maintenance and will end up with us contradicting ourselves. The Brittonic tree is at Brittonic, and the Goidelic tree is at Goidelic; they shouldn't be here as well unless they're relevant to this article, which they're not. (Primitive Irish being the ancestor of Old Irish is off-topic.) Cf. our thousand other language-family articles. — kwami (talk) 19:23, 12 May 2014 (UTC)


 * That isn't a content fork in the usually accepted sense. A family tree of what is comprised in insular celtic seems totally suitable for an article on insular celtic. Most articles on a language family has something like that. Why is this different? DeCausa (talk) 19:49, 12 May 2014 (UTC)


 * Sure they do. But they tend to stop where the child articles split off.  If we didn't have the articles Goidelic and Brittonic, then of course we'd list all the details here.  But because we do have those articles, it's best to leave the details there, with just a basic summary here so readers can follow this article.  Otherwise we end up with duplicate trees, and when editors modify one but not the other, we end up contradicting ourselves.  That is a content fork, and to be avoided.  — kwami (talk) 22:41, 12 May 2014 (UTC)


 * Calling it a content fork is an exaggeration, but it probably is sufficient for this article to say that Insular Celtic consists of Brittonic and Goidelic, and allow the articles Brittonic languages and Goidelic languages expand on that. Angr (talk) 19:16, 13 May 2014 (UTC)


 * Kwami has been defacing all of the Celtic language articles and pushing his own uninformed POV. Something has to be done about this.Cagwinn (talk) 21:34, 12 May 2014 (UTC)


 * Yes, it's called "discussion". That's what these pages are for.  — kwami (talk) 22:41, 12 May 2014 (UTC)

Propose deletion of section "Possible Afro-Asiatic substratum"
I don't understand what mentioning a discredited theory and then stating that it is discredited is adding to the article. The source says "There must have been among Cornish speakers a tendency to a somewhat blurred sound of certain letters, as though there were an obstruction of some sort in their vocal organs, not altogether unlike that attributed on the stage and in fiction, with some foundation in fact, to the Hebrew race." This is linguistic claptrap even without the racial overtones. We are getting into Fringe theories territory here. Alázhlis 06:27, 10 January 2017 (UTC) — Preceding unsigned comment added by Alázhlis (talk • contribs)


 * I don't think it's a bad thing to document discredited theories. But I agree that the passage about Cornish/Hebrew is claptrap. It is also irrelevant. I have tried to improve this section (without discarding it), but my edits keep getting reverted. Editor is acting as though his wording cannot be improved on. He has even reverted corrections to the footnotes (Gensler's dissertation is unpublished according to the UCal and LOC library catalogs). He has reinserted the claim "The concept of the Insular Celtic languages being descended from Hebrew was part of Medieval superstition", which has no source. (Though it was widely believed that all languages descended from Hebrew.)
 * The strange things is that Cagwinn writes in an edit summary "Not only not true, the theory has been all but discredited in recent years", yet when I edit the text to make that clearer, he reverts it. --Macrakis (talk) 22:22, 29 January 2018 (UTC)


 * I reverted because the rewritten passage (notably claiming - without any evidence - that Insular Celtic has features not found in any other Indo-European language) was no better. Cagwinn (talk) 23:30, 29 January 2018 (UTC)


 * OK, maybe that claim needs to be better documented (I believe it is actually asserted by both the Afro-Asiaticists and the critics of the Afro-Asiaticists).
 * But how about the other material?
 * Source for medieval theory of Hebrew origin?
 * False claim that Gensler was published by UCal Press?
 * Irrelevant, bizarre, and unduly emphasized passage from Jenner? (cf. WP:UNDUE)
 * Why phrase it as "has been criticised" and "alleges" and "he considers the theory to be not just unproven but also wrong" rather than "the claims are untenable" -- which you seem to agree with. WP policy does say that various points of view should be reported, but it doesn't say that we have to be indecisive.
 * Blanket reverts are generally a bad idea -- correct the things that need correction and improve the article, don't just revert to an earlier version.
 * --Macrakis (talk) 00:24, 30 January 2018 (UTC)


 * Ranko Matasovic, Celticist and Indo-Europeanist and already cited on this page, has a paper, [The Substratum in Insular Celtic, link], that handles this. Various other sources also discuss the Afro-Asiatic substratum theory for Insular Celtic (even broad ones like Empires of the Word)-- overall, it has much more to do with syntax than phonology typically despite the section on this page discussing phonology more. Here's an excerpt from his conclusion.
 * The syntactic parallels between the Insular Celtic and Afro-Asiatic languages are probably not accidental, but they should not be taken to mean that the pre-Celtic substratum of Britain and Ireland belonged to the Afro-Asiatic stock. It is also possible that it was a language, or a group of languages (not necessarily related), that belonged to the same macro-area as the Afro-Asiatic languages of North Africa. The parallels between Insular Celtic, Basque, and the Atlantic languages of the Niger-Congo family, presented in the second part of this paper, are consistent with the hypothesis that there was a large linguistic macro-area, encompassing parts of NW Africa, as well as large parts of Western Europe, before the arrival of the speakers of Indo-European, including Celtic. The historical origin of this macro-area can be seen in the re-population of Western Europe after the last Ice Age from the Western Mediterranean, or in the much later spread of agriculture along the Atlantic coast, which was probably associated with the archaeological culture of megalithic tombs in NW Africa and the western fringes of Europe in the Neolithic and early Copper Age (Sherratt 1994)....}}
 * Thoughts about DUE-ness for incorporating this? I'm not sure how much is DUE on a page about the language itself rather than its substratum, but I think this is much more useful discussion than the stuff about Cornish speakers "slurring like the Hebrew race". Matasovic's theory is not without controversy, but I think it still has some notability. I think maybe the section should be "Language Contact", with a subsection for "Afro-Asiatic theory" Macrakis, since you're probably wondering, although I do have much interest in substrate topics, I did end up here from glancing at your history after that comment of yours. Hope that's not too weird lol --Calthinus (talk) 04:25, 30 January 2018 (UTC)


 * In criticizing Gensler et al, Isaac's most devastating argument is that for the AA substrate theory to be true, the substrate influence would have had to have "jumped over" hundreds of years, unless the Celtic languages were still in contact with the AA substrate in the 10th century AD. (p. 26). This is based on the relative timing of some sound changes and some syntactic changes. --Macrakis (talk) 04:53, 30 January 2018 (UTC)


 * @Macrakis I hear what Isaac and you yourself are saying, but I don’t agree with Isaac’s thing. I would say that "jumping over" is in fact common. Firstly there is the question of when evidence _comes to light_, time of first attestation is not the same thing as time of a language change. This has been argued for English by Hildegard Tristram. Basically the Old English that we see is the posh conservative variety. At the same time there was also a rubbish variety, learned hurriedly and badly by Brittonic speakers who produced a pool of celticised-English speakers and these people could affect posh English much later on. Compare also English and Norman-French in England where English spoken by the lower classes caused the delayed death of French in England centuries later. Secondly language contact with some other speaker pool can happen really late, and say the other speaker pool has ancestors were L2 learners or the pool was affected by a substrate or adstrate. For Old Irish it seems to me that we have the unrepresented population on Irish speakers in Northern Britain (Argyll) who would have been affected by contact with ‘British and who also triggered the production of a large population of Britons who spoke Primitive Irish really badly. Same is true for the Irish invasion further south into Wales. If some of those people went back and forth to Ireland it could have affected _observed_ Irish in Ireland slowly for centuries and attested Irish is much later than the time of the changes whose later results we get to see. Ó Rathaile says that in the case of Argyll important nobles took accompanying groups back to Ulad with them and he claims that Scottish Gaelic influences Ulster Irish. The most extreme case I can think of is that of Lewis Gàidhlig which even uses the same type of phonetic realisation of the nasalisation mutation as Welsh does. (I have developed an explanation for this particular case.) CecilWard (talk) 07:53, 18 February 2019 (UTC)


 * Isaac's rationale should be included too imo. I think a possible rebuttal that may exist is that it is possible to import the causal pattern that has effects centuries later -- but I don't know personally if this applies to the AA-IC case which does seem like quite a separation. A typical example is that importing a heavy and time-influencing stress pattern can lead to diphthongization of stressed vowels and deletion of unstressed ones much further down the road. Sometimes effects are posited millenia after the point of influence -- if I remember correctly Guiter's 1995 paper Sur le substrate gaulois attributes the word boundary weakness (enchainement, elision, liaison) that is still having effects in modern French to Gaulish language contact that must have occurred at least 1400 years ago (almost certainly more) given when we think Gaulish was dead by (Guiter here would be synth tho, of course).--Calthinus (talk) 05:25, 30 January 2018 (UTC)


 * it does seem to me that Ranko Matasovich has hit the nail on the head - I am about to read his paper, I’ve only read the comments about it here. The macro-area thing is excellent. Take the question of the epidemics of apocope and analyticity in Western Europe. Apocope affects so many languages even across IE sub-families, Germanic, Romance and Celtic all affected. (Preference for periphrastic analytic syntactic structures rather than person-marked synthetic verb-forms, isn’t that a general development? But what is its geographical range?) When a language leaves its original home and moves into a new territory we could I suppose see the loss of sounds that the new population can not cope with. Take loss of laryngeals in PIE. Presumably PIE was born in an area where neighbours could cope with pharyngeal sounds, glottal stops and so-on and where they were felt to be distinctive and phonemic. Neighbours such as Semitic. In Western Europe and in India presumably populations cannot cope with these sounds. In India IE picks up retroflex consonants from a substratum as a consolation prize. Basque I suppose doesn’t have such pharyngeals? CecilWard (talk) 07:53, 18 February 2019 (UTC)

Cause of Split
Any academic theories on the cause of the P Celtic Q Celtic split in Insular Celtic? Assuming they did descend from a Proto-Insular Celtic. I mean Ireland is not particularly isolated from Great Britain, it's merely 19km across the sea at its closest point and we know the inhabitants of the British Isles were highly marine mobile from prehistoric times.

It's surely not coincidence that the areas Rome occupied are the only areas that ended up P Celtic right? I mean the map of Roman occupation effectively matches the map of Insular Celtic language dispersal perfectly (it's even Brythonic between Hadrian and Antonine which Rome briefly occupied, and even when it did not the tribes between Hadrian and Antonine most assuredly would have interacted heavily and been influenced deeply by Romanized Insular Celts living south of Hadrian's). Pictish zones then form a sort of quasi hybrid language somewhere between Q and P Celtic, or perhaps more aptly contained large numbers of both Q and P Celtic speaking peoples depending on isolation/rurality.

Is there literally any evidence the that Q Celtic/P Celtic split occurred before Roman contact?

Did the Romans just view Brythonic speaking peoples as less worthy or more worthy of conquering? Seems bizarre for them to stop at some arbitrary borders which just happen to contain all P Celtic speaking inhabitants. And would this also perhaps explain why Goidelic and ogham is found in the far western extremities of Great Britain (such as Scotland, Wales and Cornwall). It's traditionally assumed to have been the result of Irish invasions or colonizations of western Great Britain but there's no actual archaeological or contemporary historical evidence for this happening.

Generally speaking it seeems to be the general pattern with invasions of the British Isles pushing the dominant cultures of the time to the western extremities, this even has occurred today with the surviving P and Q Celtic languages of the British Isles. So is it possible all the British Isles were Q Insular Celtic speaking before Rome invades and has a drastic effect on the Insular Celts living there, perhaps pushing the pre Roman Q Celtic speakers into small, rural isolated pockets as they try to resist Roman influence. — Preceding unsigned comment added by 92.4.16.225 (talk) 08:29, 22 March 2019 (UTC)


 * I think it's a coincidence, yes. Your theory makes perfect sense from the standpoint of history, but linguistic evidence is easier to explain if the /kʷ/ > /p/ shift happened long before the Roman invasion.  We could get more information by looking at loandwords between the two proto-languages, which I assume has already been done by at least some scholars. — Soap — 13:52, 27 August 2020 (UTC)

This hypothesis is not generally accepted.
The infobox lists this linguistic hypothesis as "generally accepted". There seems to be few items within the body of text to give sustenance to this categorization, and published works by linguists of Celtic do not favour this. A major RS on the subject, "Celtic Culture - A Historical Encyclopedia" by John T Koch (2006) stresses that there is "no general consensus" as to whether "Insular vs Continental" or "P- vs Q-Celtic" should be favored when drawing the lineage of Celtic languages.

I am willing to engage in discussion with other editors as to whether is is necessary to change this. --JoeyofScotia (talk) 20:09, 18 April 2021 (UTC)


 * There is no consensus. The insular Celtic hypothesis is certainly not "generally accepted". This should be changed to disputed or controversial. Tewdar (talk) 18:53, 6 June 2021 (UTC)


 * "But it's in Glottolog so..." Tewdar (talk) 19:00, 6 June 2021 (UTC)