|
|
Post by Anglesqueville on Aug 22, 2023 21:53:36 GMT
1. Presentation In the context of the transition from Anthro to GenArch, I have felt compelled for days to open a thread around the proto-Germanic question. On the other hand, I don't have the courage (nor the time) to transport here all the work done over there over many months, and moreover, it would be necessary to purify this work of logical dross, inaccuracies, and other adventurous shortcuts, all by-products of polemics. I will therefore limit myself to the essentials, taking as a "leverage point" the summary made by Valter Lang of the work carried out in his 2015 text: "Formation of Proto-Finnic – an archaeological scenario from the Bronze Age / Early Iron Age". The word "archaeological" must immediately attract your attention. Lang is not a linguist, although he takes the work of the Estonian and Finnish schools of linguistics very seriously. I will therefore have, and this is essentially what the first part of my job will consist of, to develop and source the linguistic aspects of his thesis. Because the logic of my remarks here will in fact be the opposite of Lang's: first linguistic, it will allude to archaeological and genetic dimensions only for the sake of convergence. The subject of the present thread makes its first appearance in this text in 5. It will be necessary to bring some nuances and additions to it, in particular by examining what happened "on the other side", that is with pre-proto-Saami (what happened and when it happened). Let us not forget that Lang's concern here is Estonia, not Finland, nor proto-Germanic. What emerges here is this, which Lang seems to take for granted: when speakers of pre-proto-Finnic began to settle in the regions that correspond to present-day Estonia, southwestern Finland, and the Mälaren region in Sweden, they met there, among others, a population which spoke, Lang tells us, proto-Germanic. It is particularly assured for him that the people who in these regions buried their most precious representatives in monumental stone-cist graves and stone ships, were speakers of proto-Germanic. In another text, Lang writes: On the other hand,later, there is every reason to identify the builders of the Tarand Graves with the Proto-Finnic immigrants of the southwestern route.The intersection of the distributions of these modes of burial, to which we could add the distribution of the Akozino-Mälar axes, would give us an idea precise enough of the area where pre-/proto-Germanic and proto-Finnic cultures met, and perhaps hybridized.  That said, what are these "historical linguistic comparisons" on which Lang relies to identify those autochthonous people ("autochthonous" when proto-Finnic arrive in the zone) with Germanic-speaking people? 2. Interrelatedness Proto-Germanic - proto-Finnic (1) The signs that point to those "intensive contacts" are firstly (but perhaps not only) lexical. Indeed, and here is the crucial point of the whole problem, the Balto-Finnic languages and Finnish itself before any other, borrowed from Proto-Germanic an enormous mass of words. (I say "proto-Germanic" for short. The oldest ones are indeed difficult to date with certainty and could have been borrowed from a late pre-proto-Germanic. In any case, these are indeed words prior to the proto-Scandinavian). This lexical mass is currently estimable at more than 500 terms, but it is constantly growing. It was estimated that it represents for the Finnish language a lexicon comparable to the Franco-Norman superstrate of the English language. The reference, indispensable although already quite old, is the huge three-volume dictionary that specialists and amateurs affectionately call "The" Läglös: "Lexikon der älteren germanischen Lehnwörter in den ostseefinnischen Sprachen" (Kylstra et al. 1991). For those who don't have the courage to delve into it, Wikipedia offers 581 pages (but some are devoted to Proto-Scandinavian borrowings): en.wiktionary.org/wiki/Category:Finnish_terms_derived_from_Proto-GermanicIt is important, before continuing a little on this theme, to insist on this: there is nothing speculative here. The proto-Germanic lexicon of Finnish is an element of an absolutely factual nature. It goes without saying that such a massive loan could not have resulted from occasional contacts, such as those resulting from simple commercial relations. But, beyond the simple quantitative observation, the semantic ubiquity of this proto-Germanic lexicon makes such a hypothesis untenable. It is particularly symptomatic that it contains most of the vocabulary relating to advanced techniques of agriculture (unlike primitive techniques where proto-Baltic borrowings dominate) and animal husbandry. To take just a few examples: multa:soil, aidun:pasture, field niittää:to cut hay lammas:sheep aura:plow lauma:herd humala:hop nauta:bovine mallas:malt kana:chicken ruis:rye juusto:cheese kaura:oats rasva:grease laiho:growing grain akanat:chaff kello:(cow) bell turve:peat pelto:field vainio:largish field lanta:manure tunkio:dunghill It is also remarkable that the word "äiti" (mother) is a Proto-Germanic borrowing (while "tyttö" = daughter and "sissar" = sister are probably Proto-Baltic). Remarkable as well that the word "häpy" (shame, and by derivation human female genitalia) is one too. I could go on like this for pages, which I might do later because walking around the Läglös has become one of my favourite hobbies. We cannot recoil from the only plausible explanation: the existence of communities where bilingualism was the rule. Why at a certain time did proto-Finnic take precedence by assimilating a whole borrowed lexicon? This question, certainly fascinating, is beyond the scope of this work. 3. Saami The Saami languages have their share of proto-Germanic lexical borrowings. Even if this share is considerably less important than that of Proto-Finnic, it is nonetheless rich in lessons (and questions). In his 2006 text ("On Germanic-Saami contacts and Saami prehistory") Ante Aikio notes: In detail the study of the problem of Germanic loanwords in Saami is difficult, and I'll refer to the texts of Aikio. It is, however, safe to say that the pre-proto-Saami speakers, who had travelled by the North-Western route, first settled throughout southern Finland where they encountered the native Germanic speakers. The time of this meeting cannot be very different from that when the common "German-Finnic" history began. But they may not coincide. One can in particular think that the first Germano-Saami meetings occurred before the pre-proto-Finnic speakers established, coming from Estonia, their first Finnish colonies. This is what Aikio suggests:  I strongly advise studying the founding texts of Aikio, namely "An essay on Saami ethnolinguistic prehistory" and the one already quoted, which contains many fascinating etymologies. But it's time now for bringing an answer to the question implicitly contained in the title of this thread. 4. Proto-Germanic I would like, at this point, to quote once again Saarikivi and Holopainen, in the abstract of a conference held in 2017 on the occasion of the 50th Annual Meeting of the Societas Linguistica Europaea (and unfortunately not yet published): In fact, it emerges from all the elements briefly mentioned here that the initial area (the "cradle", to use a usual metaphor) of the proto-Germanic language was necessarily in the immediate vicinity of that of proto-Finnic ( and proto-Saami). By far the most economical option is in Sweden, the region of Mälaren. remark: In the same text Saarikivi and Holopainen note: "Quite problematic" is a polite understatement. The "Jastorf" thesis is in fact not based on anything. Schwantes was an archaeologist, knew nothing of the East-Baltic side of the question (and probably didn't give a damn about linguistics, a common failing in his corporation). For us today, if there is one question that arises, it is how such a meaningless thesis could have become a tradition. But this question is objectively of little interest. |
|
|
|
Post by Anglesqueville on Aug 23, 2023 15:41:05 GMT
5. Interrelatedness Proto-Germanic - proto-Finnic (2)
The Consonantal Gradation ( in its two faces, namely Rythmic Gradation,RhG, and Syllabic Gradation) is a remarkable feature of Finno-Saamic languages and, at the other extremity of the Uralic expansion area, Nganasan. RhG in Finnish has been largely obliterated by the analogy, but its traces are detectable in the partitive case of substantives. Examples:
proto-Finnic *puu (wood) has partitive *puu-ta (as in modern Finnish)
proto-Finnic *kala (fish) has partitive *kala-da (> kala-a in modern Finnish)
In Saami and moreover, in Nganasan (where it's productive) the CG leads to much more complex mechanisms. The study of these mechanisms is far beyond the scope of my post, and I'll refer to Viktor Helimski's grounding text: "Proto-Uralic Gradation. Continuation and traces", 1995. In this famous work, Helimsky proves that CG traces back to proto-Uralic. In particular, he had to challenge the theories of some of his predecessors, who saw in the CG and in particular in the RhG the product of the influence of the first Germanic consonant shift, Verner's law.
The similitude of the Uralic RhG and the Germanic Verner's Law is indeed impressive:
As Helimski's text doesn't aim on Germanic but on Uralic, he doesn't go farther from this point and goes on to argument his own thesis, namely that CG traces back to PU. On this very point starts, many years after, the work of the Dutch germanist Peter Schrijver, which consists essentially at inverting the terms of the bolded sentence: P.Schrijver makes an attempt to go further than just comparing Verner's Law to gradation, and to attribute its origin in Germanic to the proto-Finnic influence.
Schrijver writes:
Of course "unreasonable" is not a definitive argument to dismiss the possibility of an independent development in pre-Germanic of an Uralic-like feature. But as in the same time Germanic and proto-Finnic phonologies seem affected by strikingly converging evolutions ( without forgetting, in Germanic, likely shortly after Verner's law the loss of the IE mobile accent and its fixation on the initial syllable, as in all Uralic languages), the probability of such a ressemblance by chance seems really very low.
I prefer to continue here to consider this hypothesis, namely that the evolution leading from pre-Germanic to proto-Germanic took place under the influence of pre-proto-Finnic (or/and pre-proto-Saami?) phonology, as speculative, whereas I think I have insisted enough on the factual character of the Proto-Germanic lexicon in Proto-Finnic and Proto-Saami. Nevertheless, however speculative it may be, I believe I can assign it a high probability coefficient. However, whether we accept it or reject it, this will not affect the thesis stated above on the location of the Proto-Germanic area.
However the acceptance of this hypothesis could perhaps help to solve a slightly annoying mystery, that of the asymmetry of lexical borrowings. I would like to insist on the fact that whereas the preceding developments owe nothing to my own reflection, for I only echoed the work of specialists otherwise competent than myself, all that I write now was born in my poor amateur brain. Let's sum up the question:
1) There is a large lexical flow in the Germanic > Finno-Saamic direction.
2) Conversely, the lexical borrowings to Proto-Germanic from Finno-Saamic are only a handful at best (see for this Hyllestedt's thesis, and the communication by Saarikivi and Holopainen if it is published one day).
3) We observe (still under the hypothesis stated above) a phonological influence in the Finno-Saamic > Germanic direction.
4) Conversely, if there was a phonological influence (or rather phonotactic in this case, as Kallio suggests) it is minor and caused precisely by the integration of Germanic words.
This definitely leads me to speculate on not one linguistic interference but two. If one accepts Thomason's distinction between interference by imperfect learning (which often affects phonology but is accompanied by little or no lexical transfer) and interference by borrowing (which often has little effect on phonology ), we see that observations 1) and 4) seem to converge towards an interference by borrowing, whereas points 2) and 3) would converge towards an interference by imperfect learning. I had formerly on AG advanced the idea that these two interferences were separated by a geographical distance. I wonder now if they could not be separated by a temporal distance and involve different actors. Precisely interference by imperfect learning (which Thomason explains takes place over a short time, often no more than a few generations) may have involved the earliest groups of pre-proto-Saami speakers, whereas interference by borrowing would have taken place over an incommensurably longer time, and would have been the hallmark of common Germanic-Finnic history.
|
|
|
|
Post by JonikW on Aug 23, 2023 16:16:25 GMT
Thanks for posting this in such detail Angles. Your earlier posts on Anthrogenica actually succeeded in persuading me that PGmc arose in a much more northerly region than the Jastorf cultural area where I'd originally assumed it probably had its home.
So I'm glad to see the arguments distilled into one place here, and I found your list of loan words, which I read this morning, particularly interesting. If I'd visited Finland and noticed that turve = peat I'm pretty sure I would have wrongly assumed it was a modern borrowing from English turf.
Altogether it looks like case closed as far as I'm concerned, and for me as an interested but sometimes sceptical layman the information you've provided is a surprisingly powerful illustration of the ability of linguistics to illuminate the distant past.
|
|
|
|
Post by Anglesqueville on Aug 23, 2023 17:01:44 GMT
Thank you, friend. I don't remember if I've ever told this on AG, but when, so many years ago now, I realized the decisive importance of the Finno-Germanic lexical argument, it was for me a tremor. Before that, I stuck to the old Germano-Celtic stories. When, much later, I read Schrijver's book, I first brutally refused his thesis about the "origins". In fact, it took me several more years to convince myself that this thesis definitely had a good chance of containing at least a significant part of the truth. What actually happened is probably much more complex than our models. But at the present time, the Helimski-Schrijver model seems to me much more solid than what I was saying, cautiously, in my big post. I will re-post later the excellent qpAdm Scandinavian_IA + Volga-Oka-IA models I got for Finns and Saamis. This will put some genetic spice in the sauce.
|
|
|
|
Post by folcwalding on Aug 23, 2023 19:13:22 GMT
Nice read Angles!
Nevertheless I stay very very sceptic about your that proto-Germanic originates necessary close to the Saami.
It's totally taken out of context-it's only based on linguistics- and leaves many questions unanswered...
|
|
|
|
Post by Anglesqueville on Aug 24, 2023 7:20:23 GMT
A few days ago I emailed a more detailed version of what I said here to three well-known Finnish linguists (but whose names I will withhold at their request), asking them to respond. Last night I received two replies, which were very friendly and quite detailed. They are both in full agreement with the first part of the thesis that I defend here (points 1. to 4.), even going so far as to affirm that there would be full consensus on their subject (at least among the Finnish specialists).
It is not the same for point 5. (say the Helimski-Schrijver thesis), about which one of them admits to being very skeptical, and which the other rejects categorically for reasons of chronology. More precisely for him, Verner's law would have started to affect the pre-Germanic language (or languages?) spoken in the regions concerned long before the first speakers of Balto-Finnic (pre-proto-Saami included) appeared there. Verner would therefore have nothing to do with this story, one way or the other, and all the linguists concerned, from Posti to Schrijver, would have been fooled by a simple fortuitous resemblance. I do not feel entitled to add a comment to this opinion, which I deliver to you for the sake of honesty.
In his long email (whose seriousness and attention surprised and honoured me), he advises me to devote a post to Mikko Heikkilä's book "Bidrag till Fennoskandiens språkliga förhistoria i tid och rum", which I had briefly mentioned formerly on AG. I will do it one of these days, but that will require me to read it again thoroughly. He also told me that it would be necessary all the same to devote a little time to dismantling the thesis-Jastorf which continues to pollute the discussions. The problem here is that it is difficult to dismantle a thesis which is never seriously argued anywhere.
|
|
|
|
Post by folcwalding on Aug 24, 2023 12:55:46 GMT
A few days ago I emailed a more detailed version of what I said here to three well-known Finnish linguists (but whose names I will withhold at their request), asking them to respond. Last night I received two replies, which were very friendly and quite detailed. They are both in full agreement with the first part of the thesis that I defend here (points 1. to 4.), even going so far as to affirm that there would be full consensus on their subject (at least among the Finnish specialists). It is not the same for point 5. (say the Helimski-Schrijver thesis), about which one of them admits to being very skeptical, and which the other rejects categorically for reasons of chronology. More precisely for him, Verner's law would have started to affect the pre-Germanic language (or languages?) spoken in the regions concerned long before the first speakers of Balto-Finnic (pre-proto-Saami included) appeared there. Verner would therefore have nothing to do with this story, one way or the other, and all the linguists concerned, from Posti to Schrijver, would have been fooled by a simple fortuitous resemblance. I do not feel entitled to add a comment to this opinion, which I deliver to you for the sake of honesty. In his long email (whose seriousness and attention surprised and honoured me), he advises me to devote a post to Mikko Heikkilä's book "Bidrag till Fennoskandiens språkliga förhistoria i tid och rum", which I had briefly mentioned formerly on AG. I will do it one of these days, but that will require me to read it again thoroughly. He also told me that it would be necessary all the same to devote a little time to dismantling the thesis-Jastorf which continues to pollute the discussions. The problem here is that it is difficult to dismantle a thesis which is never seriously argued anywhere. I guess it's very hard to dismantle or to proof that Jastorf in fact the Elbe region (with some extension towards the Weser) was the area were the historical sound change in the Proto-Germanic language for the first time has taken place. Same is the case for the Malären. Simple: no source, no evidence. So no possibility to pinpoint it in time and place. So we can make high-flown speculations about it, but at the end of the day the sun goes down to rise again tomorrow, no more and no less  So you have to make the story plausible (second best option without sources, evidence). And then you just have a problem with the Malären, because that is not central in the supposed "Germanic world". So you will have to make it plausible that what started in the periphery, so say on the interface with the Saami, has caused a furore up to, say, the Weser (or Rhine)? And that also includes making it plausible that the Suebi - along the Elbe - initial spoke a non-proto-Germanic language. Schrijver and a bunch of Finnic linguist has only looked purely linguistically. Linguistically based on reconstructed proto languages. So it's full of assumptions. And it is completely detached from social reality. But that is of course an idea fix. Language is a social thing and a social product. To declare something an "Urheimat" you need the context in addition to the linguistic possibility. Especially if, as said, you have no evidence, no source. I cannot emphasize that enough. You can ignore that, but then it becomes either an implausible narrative or a story of esoterica, hocus pocus, faith....guys with long beards etc. "This hypothesis continues to meet sympathetic response [Korhonen 1981: 142; Lehtinen 1994: 70–72]. But how can Nganasan facts be combined with this hypothesis? Perhaps only by a daring statement that certain yet unknown Germanic tribe migrated as far east as to the Yenisei river with the generous aim of telling the rules of consonant gradation to the forefathers of the Nganasans — and, on accomplishing its mission, got frozen to death…" yeah right  |
|
|
|
Post by Orentil on Aug 24, 2023 14:29:57 GMT
Hi Anglesqueville, there is one thing that I'm curious about. Gothic is said to be the most archaic Germanic language and I read on Wikipedia: Whereas the North Germanic and West Germanic languages clearly show the effects of Verner's law, those patterns seldom appear in Gothic, the representative of East Germanic. This is usually thought to be because Gothic eliminated most Verner's law variants through analogy with the unaffected consonants.(R.D. Fulk 2018)
On the other side we know that Goths have a certain genetic Uralic shift in the PCA, if I remember the placement of the Weklice samples right. It was again confirmed by Stolarek that the Goths came from the North, someone claimed that the samples in her study (Maslomecz) are most similar to Oland (this of course could be a sample bias).
How can this be explained, shouldn't Gothic be more affected from Uralic than West Germanic? Do I overlook or misunderstand something?
|
|
|
|
Post by folcwalding on Aug 24, 2023 14:35:47 GMT
The frozen Germanics?
Iain mc Donald's survey:
"R-U106>Z381>L48>Z9>Z2>Z7>Z8>Z338
Likely MRCA data range: 1700-800 BC
Likely origin: continental North Sea coast, probably between the Rhine Delta and Jutland
Culture: Norwestblock, e.g. Elp, if early; possible northern Tumulus or Urnfield if later. Nordic Bronze Age if later.
Narrative evidence: R-Z338 shows more enhancement in the Netherlands, Denmark, Finland and Latvia (R-FGC1954) compared to R-Z1. Like R-Z1, it is much more frequent than normal in England. Enhancement also exists in Norway and Sweden. The basal clades are strongest in the Netherlands and Denmark, with a reasonable presence in Germany and France, but not in England. The major sub-clade, R-Z341, is the haplogroup showing the English prominence (by a factor of 1.76 compared to R-Z381 as a whole). The notable anomaly in R-Z338 is the high frequency in Finland. This is difficult to reconcile with the western European frequency, and lack of presence in countries further east. It is also present in the the "nephew" group R-Z346, but not R-Z346's brother group, R-Z344, indicating that the split point between the two dictates when the relevant migration occurred. Hence, a solution to the Finnish problem is a migration of a largely homogeneous population sometime shortly before the R-Z344/R-Z346 split (i.e. before 1500-500 BC)."
|
|
|
|
Post by Anglesqueville on Aug 24, 2023 14:56:47 GMT
What "yeah right"? That quotation only expressed the fact that for Helimski the CG cannot be a copy of Verners'law (Posti's thesis). About Verner's law according to Heikkilä in his "Bidrag ..." ... two words. Heikkilä contests that Verner took place before Grimm. Precisely he writes:
(I kept the following sentence, with which I completely agree, just for fun). Note: I put in red a sentence that I still have to work on, but which intrigues me a lot.
For the rest (namely the "history of Fennoscandia's settlement), he writes:
I'm not comfortable with the words "senast från och med bronsåldern", that I translate as you see. I would be tempted to write rather: "but not before the end of the Bronze Age". If some Swedish friends read this they will feel free to correct and possibly to give another translation. For the rest the sentence I bolded is nothing but the thesis I expressed in my main posts.
|
|
|
|
Post by Pylsteen on Aug 24, 2023 15:13:59 GMT
A detail in the discussion, although important IMO - let's see if I remember well, because I am busy with all kinds of labour these days except linguistics unfortunately...
Verner's law may have occurred before Grimm's instead of after, if we take into account Kluge's Law as well (revived by Kroonen, but somehow it keeps getting unfair criticism),
IMO, the following order may be logical as well:
I. Proto-Indo-European situation:
"fortis" *t, "lenis" *dh, "inbetween" *d (according to Kortlandt (pre-)glottalized, but that's another story)
II. Verner's Law: "fortis" stops become "lenis" after an unaccented syllable
(*t > *dh)
III. Kluge's Law: clusters of stop+n before an accented vowel become geminates:
*-dn-' and *dhn-' > *-dd-
IV Grimm's Law: chain shift of stops:
*dh > *d, *d > *t, *-dd- (from Kluge) > *-tt-, *t > *th,
|
|
|
|
Post by folcwalding on Aug 24, 2023 15:41:53 GMT
@angles, ok thanks clarified by the explanation.
"but that there must have been Germanic colonies settled in present-day Finland, whose population over time was assimilated into the majority population."
Could well be the case. See also the R1bU106 link.
And I can and will not rule out an impact on proto-Germanic from the Germanic-Finnic interface language. Imo it's no CG or FG both can "be true". It's not mutual exclusive imo.
Nevertheless I don't think that it has resulted in a convincing narrative that the proto-Germanic origin can be totally narrowed down to that. As said for exact pinpointing you need original sources. None of that of course, so stays speculative.
|
|
|
|
Post by Anglesqueville on Aug 24, 2023 15:56:19 GMT
A detail in the discussion, although important IMO - let's see if I remember well, because I am busy with all kinds of labour these days except linguistics unfortunately... Verner's law may have occurred before Grimm's instead of after, if we take into account Kluge's Law as well (revived by Kroonen, but somehow it keeps getting unfair criticism), IMO, the following order may be logical as well: I. Proto-Indo-European situation: "fortis" *t, "lenis" *d h, "inbetween" *d (according to Kortlandt (pre-)glottalized, but that's another story) II. Verner's Law: "fortis" stops become "lenis" after an unaccented syllable (*t > *d h) III. Kluge's Law: clusters of stop+n before an accented vowel become geminates: *-dn-' and *d hn-' > *-dd- IV Grimm's Law: chain shift of stops: *d h > *d, *d > *t, *-dd- (from Kluge) > *-tt-, *t > *th, I would willingly agree with that but Heikkilä's methodology relies also on Balto-Finnic sound changes. All this is so difficult. When you have time you should read Heikkilä's dissertation. Google translate doesn't make a totally satisfying job on Swedish and I have to thank my memories of Norwegian Bokmål, but you could try. |
|
|
|
Post by Anglesqueville on Aug 24, 2023 16:07:48 GMT
|
|
|
|
Post by Anglesqueville on Aug 24, 2023 16:38:26 GMT
Hi Anglesqueville, there is one thing that I'm curious about. Gothic is said to be the most archaic Germanic language and I read on Wikipedia: Whereas the North Germanic and West Germanic languages clearly show the effects of Verner's law, those patterns seldom appear in Gothic, the representative of East Germanic. This is usually thought to be because Gothic eliminated most Verner's law variants through analogy with the unaffected consonants.(R.D. Fulk 2018)On the other side we know that Goths have a certain genetic Uralic shift in the PCA, if I remember the placement of the Weklice samples right. It was again confirmed by Stolarek that the Goths came from the North, someone claimed that the samples in her study (Maslomecz) are most similar to Oland (this of course could be a sample bias). How can this be explained, shouldn't Gothic be more affected from Uralic than West Germanic? Do I overlook or misunderstand something? As you have seen, the question of the affectation of Proto-Germanic (and a fortiori of its first descendants) by pre-proto-Finnic (or -Saami) is a question... perhaps open, in any case very delicate. Could there be traces of an affectation of Gothic by... a Balto-Finnic language (which would it be? Proto-Finnic? A late pre-proto-Finnic?)? Unless these traces are only retentions of an earlier affectation of Proto-Germanic itself? I have to think, but I'm very reserved. In any case, I see no reason to imagine that, if such a tiny phonological affectation exists, it should be reflected in autosomal genetics. And by the way, I don't remember seeing any "Uralic shift" ( that's to say towards the Volga_Oka_IA cluster?) on the Weklice cluster. To be honest, the only genetic (autosomal) argument (but perhaps the word "argument" is excessive) that I have found in support of all this is in the genetics of modern Finns, which can very easily be summed up as a mix of Iron Age Scandinavians and people from the Volga_Oka group from the same period. I would like to get Finns from the early Middle Ages, but ... perhaps one day? |
|
|
|
Post by Pylsteen on Aug 24, 2023 18:24:34 GMT
I would willingly agree with that but Heikkilä's methodology relies also on Balto-Finnic sound changes. All this is so difficult. When you have time you should read Heikkilä's dissertation. Google translate doesn't make a totally satisfying job on Swedish and I have to thank my memories of Norwegian Bokmål, but you could try. I will take a look at it; Norwegian is not a similar to Dutch as German, but at least on paper quite recognizable. Luckily it isn't in Finnish. In general, if I have time, I might consider to write down what I think about the development of the proto-Germanic language from the Corded Ware beginnings until the Roman age, and yes Finn that includes a context based on both linguistic clues, archaeology and genetics. |
|
|
|
Post by Strider99 on Aug 24, 2023 18:26:05 GMT
I can only interpret "senast från och med bronsåldern" as meaning "since at least the Bronze Age" or "since no later than the Bronze Age" in this case. It is a somewhat strange way to express that sentence. If Heikkilä is a Swedish-speaking Finn, it could explain why he chose to express himself like that. The Swedish dialects in Finland have a lot of charming archaisms that can seem strange, like my personal favorite "hoppeligen" instead of the usual "förhoppningsvis". Personally, I would have chosen to write something like "åtminstone sedan bronsålderns begynnelse", "åtminstone sedan början av bronsåldern" or "åtminstone sedan bronsåldern" if I were Heikkilä, in order to make it more digestible. But in any case, I remain fairly confident in my interpretation of that sentence. If Hygelac reads this thread perhaps he can provide a second opinion about "senast från och med bronsåldern". |
|
|
|
Post by Anglesqueville on Aug 24, 2023 21:15:37 GMT
I can only interpret "senast från och med bronsåldern" as meaning "since at least the Bronze Age" or "since no later than the Bronze Age" in this case. It is a somewhat strange way to express that sentence. If Heikkilä is a Swedish-speaking Finn, it could explain why he chose to express himself like that. The Swedish dialects in Finland have a lot of charming archaisms that can seem strange, like my personal favorite "hoppeligen" instead of the usual "förhoppningsvis". Personally, I would have chosen to write something like "åtminstone sedan bronsålderns begynnelse", "åtminstone sedan början av bronsåldern" or "åtminstone sedan bronsåldern" if I were Heikkilä, in order to make it more digestible. But in any case, I remain fairly confident in my interpretation of that sentence. If Hygelac reads this thread perhaps he can provide a second opinion about "senast från och med bronsåldern". That's of course how I understood it, but I couldn't accept it (and suspected a translation problem on my part). I could not accept it because for me, as I imagine almost everyone, the "förfäder till finnarna och samerna" only began to settle in Fennoscandia at the earliest at the beginning of the Iron Age. Well, that turns out not to be Heikkilä's opinion: "... I still believe that the introduction of the West Uralic languages in eastern Fennoscandia took place in connection with the spread of the textile ceramic culture and the so-called Sejma-Turbino transcultural Bronze Age phenomenon ( ca. 1900–1500 BC. ) from the Upper Volga to the northeastern Baltic Sea region." This is Kallio's thesis. And if I defended it for a while, I don't defend it anymore, for a lot of reasons. Honestly, I would be happy if Kallio and Heikkilä were right, but I believe they are wrong. |
|
|
|
Post by folcwalding on Aug 25, 2023 6:43:36 GMT
I would willingly agree with that but Heikkilä's methodology relies also on Balto-Finnic sound changes. All this is so difficult. When you have time you should read Heikkilä's dissertation. Google translate doesn't make a totally satisfying job on Swedish and I have to thank my memories of Norwegian Bokmål, but you could try. I will take a look at it; Norwegian is not a similar to Dutch as German, but at least on paper quite recognizable. Luckily it isn't in Finnish. In general, if I have time, I might consider to write down what I think about the development of the proto-Germanic language from the Corded Ware beginnings until the Roman age, and yes Finn that includes a context based on both linguistic clues, archaeology and genetics. I'm honestly looking forward to it! Basically I think that we have to keep in mind that proto-Germanic is a reconstruction, it was not a living language. Nor was it likely that it was a "one size fits all language". From the Aller to Uppsala the chance is small that they spoke exactly the same language, could we perhaps state that it was most probably a loose fit language with many dialects? What Angles states about the Finnic-Germanic interaction is really interesting kind of frontier situation in the North Eastern c.q. Scandic part of the supposed Germanic world. Just like there was most probably also a kind of interface situation in the Southwest c.q. Celtic-Germanic frontier at the upper parts of the Elbe, Weser and (may be) Rhine (Meuse) area. I guess there is no reason to be blind towards one side, I guess we must take both linguistic frontiers in account. I guess it's difficult- or better may be impossible- to grasp exactly what happened, how this this effected the "proto-Germanic dialects" in between those frontiers. We only know that during migration time there was a major push from the North-East to the Southwest, unto the border with the Walloon, Welsh and Vlach/Wallachian (etymological probably the same roots)! |
|
|
|
Post by folcwalding on Aug 25, 2023 12:40:10 GMT
About the non Germanic "walkhiskaz". I guess we can recognize- in the proto Germanic period, besides a frontier/interface with the Finns/ Saami also one with the Celto or Celto-Romanic frontier/interface. Koch (2020)  Resulting in de early middle ages/ migration period into these kind of isglosses, on the border with Celto-Romanic: 
And on the other side:  So: "Before that, I stuck to the old Germano-Celtic stories". No need for a stuck. As a matter of fact both interfaces/ frontiers played a part in the (proto) Germanic world. |
|
|
|
Post by Dewsloth on Aug 25, 2023 13:54:54 GMT
About the non Germanic "walkhiskaz". I guess we can recognize- in the proto Germanic period, besides a frontier/interface with the Finns/ Saami also one with the Celto or Celto-Romanic frontier/interface. Koch (2020) [snip] So: "Before that, I stuck to the old Germano-Celtic stories". No need for a stuck. As a matter of fact both interfaces/ frontiers played a part in the (proto) Germanic world. 800-500 BCE buried in NW Bohemia (~middle band of the lines), but probably from north of the middle band, perhaps already Baltic coastal cultural contacts diffusing? Distance to: CZE_IA_Hallstatt_low_res:I17607 0.04608915 Netherlands_LBA 0.04753983 England_MBA_lowEEF 0.04874972 Sweden_IA.SG 0.05022306 England_C_EBA 0.05023995 Greenland_LateNorse.SG 0.05097497 Netherlands_MBA 0.05097845 England_EBA_BellBeaker 0.05098995 Denmark_IA.SG Target: CZE_IA_Hallstatt_low_res:I17607 Distance: 4.3180% / 0.04318026 | ADC: 1x RC45.4 Netherlands_LBA 34.6 Sweden_IA.SG 20.0 England_MBA_lowEEF |
|
|
|
Post by folcwalding on Aug 25, 2023 15:48:52 GMT
About the non Germanic "walkhiskaz". I guess we can recognize- in the proto Germanic period, besides a frontier/interface with the Finns/ Saami also one with the Celto or Celto-Romanic frontier/interface. Koch (2020) [snip] So: "Before that, I stuck to the old Germano-Celtic stories". No need for a stuck. As a matter of fact both interfaces/ frontiers played a part in the (proto) Germanic world. 800-500 BCE buried in NW Bohemia (~middle band of the lines), but probably from north of the middle band, perhaps already Baltic coastal cultural contacts diffusing? Distance to: CZE_IA_Hallstatt_low_res:I17607 0.04608915 Netherlands_LBA 0.04753983 England_MBA_lowEEF 0.04874972 Sweden_IA.SG 0.05022306 England_C_EBA 0.05023995 Greenland_LateNorse.SG 0.05097497 Netherlands_MBA 0.05097845 England_EBA_BellBeaker 0.05098995 Denmark_IA.SG Target: CZE_IA_Hallstatt_low_res:I17607 Distance: 4.3180% / 0.04318026 | ADC: 1x RC45.4 Netherlands_LBA 34.6 Sweden_IA.SG 20.0 England_MBA_lowEEF Imo you are mixing up genetics and culture. Netherlands LBA are in cultural sense fare from being what we now define as Germanic ( actual more kind of Celto-Germanic, "twillightzone" . For the rest already Single Grave/ Corded Ware had a departure from NW Bohemia, see the oldest R1b U106 PLN001 (about 2800 BC) at the upstream of the Elbe, pure Steppe like, Bohemia was a bridgehead towards NW Europe with the Elbe river as a main link, also for the genes.... |
|
|
|
Post by JonikW on Aug 25, 2023 21:06:36 GMT
I'd like to briefly ponder here the potential role of the I1 haplogroup in the spread of PGmc southwards from Sweden, if we accept as I now do that the language group expanded from there, as the I-M253 haplogroup itself also seems to have done based on the evidence we have so far.
I'll sum up what we know about the spread of I1 here, and in my view it's suggestive. To start at the potential I1 source region, we had 10 I-M253+ samples from Sweden and Denmark in "Population Genomics of Stone Age Eurasia," Allentoft et al. As many here will remember, the paper identified "a final stage from c. 3,800 BP onwards, where a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears."
And importantly the study concluded on the subject of I1: "Using individuals associated with this cluster (Scandinavia_4000BP_3000BP) as sources in supervised ancestry modelling (see “postBA”, Extended Data Fig. 4), we find that it forms the predominant source for later Iron- and Viking Age Scandinavians, as well as ancient European groups outside Scandinavia who have a documented Scandinavian or Germanic association…"
Please take a moment to digest those last words. Now, the question of when I1 started spreading beyond Scandinavia to a truly significant extent that reflects in some way the later distribution hasn't yet been fully answered by aDNA studies, although there are already some big clues. We've had a lone BA sample from Ukraine in Chyleński et al, (poz643 - Komarov/Trzciniec culture, Beremiany, West-Ukraine); Brunel et al also had a lone sample, this time IA (BES1248 from La Monédière, Bessan, Hérault); and Allentoft et al in their big BA/IA study also had one sample (I16453. 800-1 BCE England_IA, Cornwall, St. Mawes, Tregear Vean, who looks like a Scandinavian tin trader).
Aside from those three, who can all comfortably be linked to known trading activity, and the later, 2nd or 3rd century AD sample NWC010 from Scheib et al's study of Roman Cambridgeshire – a time and place where there had long been documented Germanic movements of auxiliary troops and others – we have to wait for the period covered by the Gretzinger et al Anglo-Saxon paper before we see significant numbers of I1 men outside Scandinavia (55 I1 samples, which is remarkable if we stop and think what that tells us in comparison with the above non-Allentoft studies that feature lone I1 samples.)
Then, covering a later period, we had the Magaryan et al Viking study, which was the first really significant paper for I-M253 studies. That demonstrated for the first time that, in the case of I1 at least, modern distribution really does reflect the past to a significant extent.
By my count there were 92 I1 samples out of the 442 total male and female samples in that study. The authors pointed out that "I1 is the most well represented haplogroup in our ancient dataset."
So to cut to the chase, I suspect I1's spread beyond Sweden will end up reflecting very closely the spread of PGmc after about 500 BC if we ever get sufficient data of whatever kind to tell us either way. A lot more work is needed, and I hope it will be forthcoming, but I think that the basic picture can already clearly be seen.
|
|
|
|
Post by folcwalding on Aug 26, 2023 7:40:44 GMT
I'd like to briefly ponder here the potential role of the I1 haplogroup in the spread of PGmc southwards from Sweden, if we accept as I now do that the language group expanded from there, as the I-M253 haplogroup itself also seems to have done based on the evidence we have so far. I'll sum up what we know about the spread of I1 here, and in my view it's suggestive. To start at the potential I1 source region, we had 10 I-M253+ samples from Sweden and Denmark in "Population Genomics of Stone Age Eurasia," Allentoft et al. As many here will remember, the paper identified "a final stage from c. 3,800 BP onwards, where a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears." And importantly the study concluded on the subject of I1: "Using individuals associated with this cluster (Scandinavia_4000BP_3000BP) as sources in supervised ancestry modelling (see “postBA”, Extended Data Fig. 4), we find that it forms the predominant source for later Iron- and Viking Age Scandinavians, as well as ancient European groups outside Scandinavia who have a documented Scandinavian or Germanic association…" Please take a moment to digest those last words. Now, the question of when I1 started spreading beyond Scandinavia to a truly significant extent that reflects in some way the later distribution hasn't yet been fully answered by aDNA studies, although there are already some big clues. We've had a lone BA sample from Ukraine in Chyleński et al, (poz643 - Komarov/Trzciniec culture, Beremiany, West-Ukraine); Brunel et al also had a lone sample, this time IA (BES1248 from La Monédière, Bessan, Hérault); and Allentoft et al in their big BA/IA study also had one sample (I16453. 800-1 BCE England_IA, Cornwall, St. Mawes, Tregear Vean, who looks like a Scandinavian tin trader). Aside from those three, who can all comfortably be linked to known trading activity, and the later, 2nd or 3rd century AD sample NWC010 from Scheib et al's study of Roman Cambridgeshire – a time and place where there had long been documented Germanic movements of auxiliary troops and others – we have to wait for the period covered by the Gretzinger et al Anglo-Saxon paper before we see significant numbers of I1 men outside Scandinavia (55 I1 samples, which is remarkable if we stop and think what that tells us in comparison with the above non-Allentoft studies that feature lone I1 samples.) Then, covering a later period, we had the Magaryan et al Viking study, which was the first really significant paper for I-M253 studies. That demonstrated for the first time that, in the case of I1 at least, modern distribution really does reflect the past to a significant extent. By my count there were 92 I1 samples out of the 442 total male and female samples in that study. The authors pointed out that "I1 is the most well represented haplogroup in our ancient dataset." So to cut to the chase, I suspect I1's spread beyond Sweden will end up reflecting very closely the spread of PGmc after about 500 BC if we ever get sufficient data of whatever kind to tell us either way. A lot more work is needed, and I hope it will be forthcoming, but I think that the basic picture can already clearly be seen. Of course bearers of I-M253 were a factor in the spread of proto-Germanic. If only I1 plays a part of PGmc and exclusively "southwards from Sweden" con be questioned that's too restricted I guess. Because that would mean that Anglo-Saxons - from the bottleneck- with I1 couldn't spread it for example? And besides that it's reasonable to assume that for example R1b U106 also played a part. And besides the haplotypes I don't think you can state that the spread from PGmc was exclusive from Sweden, what kind of language was spoken in LBA/IA in the Jastorf c.q. Elbe area Jonik? I guess also a kind of PGmc.....So nothing against your description, as such true! But I don't think you can pinpoint PGmc exclusivel to one Y-DNA type-in casu I1- and not exclusively Sweden. Just a few considerations.... |
|
|
|
Post by Strider99 on Aug 26, 2023 9:19:13 GMT
I'd like to briefly ponder here the potential role of the I1 haplogroup in the spread of PGmc southwards from Sweden, if we accept as I now do that the language group expanded from there, as the I-M253 haplogroup itself also seems to have done based on the evidence we have so far. I'll sum up what we know about the spread of I1 here, and in my view it's suggestive. To start at the potential I1 source region, we had 10 I-M253+ samples from Sweden and Denmark in "Population Genomics of Stone Age Eurasia," Allentoft et al. As many here will remember, the paper identified "a final stage from c. 3,800 BP onwards, where a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears." And importantly the study concluded on the subject of I1: "Using individuals associated with this cluster (Scandinavia_4000BP_3000BP) as sources in supervised ancestry modelling (see “postBA”, Extended Data Fig. 4), we find that it forms the predominant source for later Iron- and Viking Age Scandinavians, as well as ancient European groups outside Scandinavia who have a documented Scandinavian or Germanic association…" Please take a moment to digest those last words. Now, the question of when I1 started spreading beyond Scandinavia to a truly significant extent that reflects in some way the later distribution hasn't yet been fully answered by aDNA studies, although there are already some big clues. We've had a lone BA sample from Ukraine in Chyleński et al, (poz643 - Komarov/Trzciniec culture, Beremiany, West-Ukraine); Brunel et al also had a lone sample, this time IA (BES1248 from La Monédière, Bessan, Hérault); and Allentoft et al in their big BA/IA study also had one sample (I16453. 800-1 BCE England_IA, Cornwall, St. Mawes, Tregear Vean, who looks like a Scandinavian tin trader). Aside from those three, who can all comfortably be linked to known trading activity, and the later, 2nd or 3rd century AD sample NWC010 from Scheib et al's study of Roman Cambridgeshire – a time and place where there had long been documented Germanic movements of auxiliary troops and others – we have to wait for the period covered by the Gretzinger et al Anglo-Saxon paper before we see significant numbers of I1 men outside Scandinavia (55 I1 samples, which is remarkable if we stop and think what that tells us in comparison with the above non-Allentoft studies that feature lone I1 samples.) Then, covering a later period, we had the Magaryan et al Viking study, which was the first really significant paper for I-M253 studies. That demonstrated for the first time that, in the case of I1 at least, modern distribution really does reflect the past to a significant extent. By my count there were 92 I1 samples out of the 442 total male and female samples in that study. The authors pointed out that "I1 is the most well represented haplogroup in our ancient dataset." So to cut to the chase, I suspect I1's spread beyond Sweden will end up reflecting very closely the spread of PGmc after about 500 BC if we ever get sufficient data of whatever kind to tell us either way. A lot more work is needed, and I hope it will be forthcoming, but I think that the basic picture can already clearly be seen. It's quite fascinating that this remarkable pattern, as you mention, has persisted even as large studies like Patterson et al. were published. At the risk of sounding like a broken record, I also think it's worth mentioning that out of the roughly 45 male samples from the Mierzanowice, Strzyżów, and Trzciniec cultures, none carried I1. Out of 16 samples from Bronze Age Leubingen in Thuringia, there was no I1. None of the Unetice samples from the multiple other tested sites carried I1, either. We also have samples associated with the Urnfield culture from Lower Saxony, and we only see R1b and I2 among those as well. Additionally, what might have been a Lusatian man from Halberstadt carried R1a. As did the potential Lusatian autosomal outlier at Weltzin in Bronze Age Mecklenburg-Vorpommern. The other 15 samples from Weltzin were all R-P312 and I2. Furthermore, the samples from multiple sites dated to the Bronze Age in the Netherlands were all P312 and U106. Similarly, there is a complete dearth of I1 in the British and Baltic Bronze Age cultures. As you point out, JonikW, this is in stark contrast to the Nordic Bronze Age samples from Denmark and Southern Sweden, where more than 60% of the currently available samples carried I1. We can also see some interesting patterns if we go slightly further back in time, to the Late Neolithic Dagger Period in Southern Scandinavia. If we look at the 8 Late Neolithic samples from Denmark, only one of them carried I1 while most of the others carried R1b. The situation in Late Neolithic Sweden appears to have been almost the complete opposite: of the 10 Late Neolithic samples from Sweden, 7 were I1, while two carried R1b and one R1a. If we look at the Bronze Age samples from Denmark, however, I1 is at a much higher frequency than it was during the Late Neolithic. Then again, we are not dealing with huge sample sizes here. It is also worth noting that Late Neolithic Skåne is, thus far, our earliest evidence for coalescence between I-DF29 and R-U106 (as well as being our oldest case of DF29). When BAM files from Allentoft et al. 2022 become available later this year that could change, however. In any case, the idea that DF29 diversified in the Scania-Sjælland region remains by far the most likely hypothesis in regards to the origin and expansion of I1 based on both ancient and modern distribution and diversity, in my opinion. It's possible that I'm being slightly pedantic when emphasizing "Scania-Sjælland" to such an extent, but this is merely because it is of course pointless for me to speak of Sweden or Denmark during the Bronze Age. More important is this important interaction zone and "base" of expansion. In light of the very early samples from Västergötland in Allentoft et al. 2022 I'm not entirely opposed to an origin further north, but for now at least I think Scania or thereabouts is more parsimonious. And in contrast to the complete absence of I1 among the samples from (what is now) Germany, the Netherlands, and Poland during the Bronze Age, it is ubiquitous among later Germanic-associated samples in those same regions. This fact is made more interesting by the Copenhagen lab's evidence for IBD-sharing between their 4000-3000 YBP cluster and later Germanic groups within and outside of Scandinavia. Among Stolarek's recently published East Germanic samples, it has not only a strikingly high frequency but also high subclade diversity. Among the Anglo-Saxons we see lower subclade diversity (although it's still there) but nonetheless a very high frequency. In short, I very much agree with you that the data currently available to us strongly suggests that there was an initial, large expansion of I1 lineages out of Southern Scandinavia that appears to have started no earlier than around the Late Bronze Age/early IA, although it certainly didn't stop then. I should probably include the mandatory disclaimer that there were multiple patrilineages involved in that expansion, not just I1. I also think that an origin of L22 in East-Central Sweden in the early BA which you suggested multiple times on Anthrogenica is going to turn out to be pretty spot-on. This may be of some relevance to the topic of the long-standing linguistic interactions between Scandinavia and Finland, as the apparent migration of L258 from East-Central Sweden to Åland and Finland during the Iron Age could have some implications in that sense. There will be more Iron Age DNA from those regions, so I suppose we'll know in time. |
|
|
|
Post by Anglesqueville on Aug 26, 2023 10:13:42 GMT
For some reasons I can't quote folc where I want to quote him. The quotation is :" If only I1 plays a part of PGmc and exclusively "southwards from Sweden" con be questioned that's too restricted I guess. Because that would mean that Anglo-Saxons - from the bottleneck- with I1 couldn't spread it for example?" Anglo-Saxons didn't spread PGmc, but West Germanic. For them PGmc was already "old story". Well, in this post I'm going to be even more restrictive. Because this post is a logical ... ... Focus I believe it is necessary to define within the thesis that I published above something like a nucleus, a core (the reactor core  ), expressed with as few words as possible so that in the face of any criticism I can ask: "Okay, but do you dispute the core of my point?". Failing that, we risk reproducing on this new forum what we produced on the old one, namely several hundred pages of a labyrinthine controversy where readers were lost, often torn between boredom and exasperation (as some had told me). I thought a lot about this, weighed my words, and here it is: Simplified premisses: There are several hundred words from the Finnish and Saami lexicons which, when we go back up the chain of phonological mutations specific to the languages concerned, lead to terms that belong to the reconstructed Proto-Germanic lexicon. Archaeological evidence suggests that these words belong to the language spoken by Iron Age Scandinavians in an area containing the Lake Mälar region. (The simplifications relate mainly to the first sentence, which is a scandalously abrupt summary of hundreds of sometimes difficult etymologies, and equally difficult definitions of the phonological mutations mentioned and their chronology. Simplification also relates to the semantic scope of the borrowed vocabulary, which is an essential element. But I said I have to reduce, so I reduce. About the second sentence, there are reasons to suggest that this area also contained Scandinavian colonies permanently settled in the coastal regions of Finland, and obviously nearby Swedish regions, such as Södermanland. In the following lines, I will write "Mälaren" for this geographical zone. It's very likely also that very quickly PGmc spread to the regions in Scandinavia that were in close and regular contact with this region, namely southern Sweden and Norway; You can read for this many texts, including Engedal's thesis). Minimal Thesis (MT)
In the second part of the first millennium B.C. was spoken in Mälaren a language for which our reconstructed proto-Germanic is a good proxy.
Obviously MT doesn't say anything about the linguistic history that led to this language, nor nothing about the history that starts from this language. These are two different problems (connected but different). MT doesn't say anything neither about languages spoken elsewhere in Northern Europe in the pre-Roman Iron Age (for example Denmark, Northern Germany or Poland). MT doesn't say anything neither about the languages spoken by the "Germanic" tribes in the Migrations Times. These are both different problems (more or less connected but different). Question 1: Does someone contest MT?
Question 2: Is someone able to put an alternative to MT under the form "In the second part of the first millennium B.C. was spoken in ... a language for which our reconstructed proto-Germanic is a good proxy"? (Replace the points with a geographic zone). Please, don't joke by replacing "..." with Tanzania or Iran (I remember reading on Academia a text where on the basis of the vague resemblance of a handful of words the author defended an Iranian origin of proto-Germanic). In order to write you have to argue. |
|
|
|
Post by JonikW on Aug 26, 2023 10:22:28 GMT
I'd like to briefly ponder here the potential role of the I1 haplogroup in the spread of PGmc southwards from Sweden, if we accept as I now do that the language group expanded from there, as the I-M253 haplogroup itself also seems to have done based on the evidence we have so far. I'll sum up what we know about the spread of I1 here, and in my view it's suggestive. To start at the potential I1 source region, we had 10 I-M253+ samples from Sweden and Denmark in "Population Genomics of Stone Age Eurasia," Allentoft et al. As many here will remember, the paper identified "a final stage from c. 3,800 BP onwards, where a distinct cluster of Scandinavian individuals dominated by males with I1 Y-haplogroups appears." And importantly the study concluded on the subject of I1: "Using individuals associated with this cluster (Scandinavia_4000BP_3000BP) as sources in supervised ancestry modelling (see “postBA”, Extended Data Fig. 4), we find that it forms the predominant source for later Iron- and Viking Age Scandinavians, as well as ancient European groups outside Scandinavia who have a documented Scandinavian or Germanic association…" Please take a moment to digest those last words. Now, the question of when I1 started spreading beyond Scandinavia to a truly significant extent that reflects in some way the later distribution hasn't yet been fully answered by aDNA studies, although there are already some big clues. We've had a lone BA sample from Ukraine in Chyleński et al, (poz643 - Komarov/Trzciniec culture, Beremiany, West-Ukraine); Brunel et al also had a lone sample, this time IA (BES1248 from La Monédière, Bessan, Hérault); and Allentoft et al in their big BA/IA study also had one sample (I16453. 800-1 BCE England_IA, Cornwall, St. Mawes, Tregear Vean, who looks like a Scandinavian tin trader). Aside from those three, who can all comfortably be linked to known trading activity, and the later, 2nd or 3rd century AD sample NWC010 from Scheib et al's study of Roman Cambridgeshire – a time and place where there had long been documented Germanic movements of auxiliary troops and others – we have to wait for the period covered by the Gretzinger et al Anglo-Saxon paper before we see significant numbers of I1 men outside Scandinavia (55 I1 samples, which is remarkable if we stop and think what that tells us in comparison with the above non-Allentoft studies that feature lone I1 samples.) Then, covering a later period, we had the Magaryan et al Viking study, which was the first really significant paper for I-M253 studies. That demonstrated for the first time that, in the case of I1 at least, modern distribution really does reflect the past to a significant extent. By my count there were 92 I1 samples out of the 442 total male and female samples in that study. The authors pointed out that "I1 is the most well represented haplogroup in our ancient dataset." So to cut to the chase, I suspect I1's spread beyond Sweden will end up reflecting very closely the spread of PGmc after about 500 BC if we ever get sufficient data of whatever kind to tell us either way. A lot more work is needed, and I hope it will be forthcoming, but I think that the basic picture can already clearly be seen. It's quite fascinating that this remarkable pattern, as you mention, has persisted even as large studies like Patterson et al. were published. At the risk of sounding like a broken record, I also think it's worth mentioning that out of the roughly 45 male samples from the Mierzanowice, Strzyżów, and Trzciniec cultures, none carried I1. Out of 16 samples from Bronze Age Leubingen in Thuringia, there was no I1. None of the Unetice samples from the multiple other tested sites carried I1, either. We also have samples associated with the Urnfield culture from Lower Saxony, and we only see R1b and I2 among those as well. Additionally, what might have been a Lusatian man from Halberstadt carried R1a. As did the potential Lusatian autosomal outlier at Weltzin in Bronze Age Mecklenburg-Vorpommern. The other 15 samples from Weltzin were all R-P312 and I2. Furthermore, the samples from multiple sites dated to the Bronze Age in the Netherlands were all P312 and U106. Similarly, there is a complete dearth of I1 in the British and Baltic Bronze Age cultures. As you point out, JonikW, this is in stark contrast to the Nordic Bronze Age samples from Denmark and Southern Sweden, where more than 60% of the currently available samples carried I1. We can also see some interesting patterns if we go slightly further back in time, to the Late Neolithic Dagger Period in Southern Scandinavia. If we look at the 8 Late Neolithic samples from Denmark, only one of them carried I1 while most of the others carried R1b. The situation in Late Neolithic Sweden appears to have been almost the complete opposite: of the 10 Late Neolithic samples from Sweden, 7 were I1, while two carried R1b and one R1a. If we look at the Bronze Age samples from Denmark, however, I1 is at a much higher frequency than it was during the Late Neolithic. Then again, we are not dealing with huge sample sizes here. It is also worth noting that Late Neolithic Skåne is, thus far, our earliest evidence for coalescence between I-DF29 and R-U106 (as well as being our oldest case of DF29). When BAM files from Allentoft et al. 2022 become available later this year that could change, however. In any case, the idea that DF29 diversified in the Scania-Sjælland region remains by far the most likely hypothesis in regards to the origin and expansion of I1 based on both ancient and modern distribution and diversity, in my opinion. It's possible that I'm being slightly pedantic when emphasizing "Scania-Sjælland" to such an extent, but this is merely because it is of course pointless for me to speak of Sweden or Denmark during the Bronze Age. More important is this important interaction zone and "base" of expansion. In light of the very early samples from Västergötland in Allentoft et al. 2022 I'm not entirely opposed to an origin further north, but for now at least I think Scania or thereabouts is more parsimonious. And in contrast to the complete absence of I1 among the samples from (what is now) Germany, the Netherlands, and Poland during the Bronze Age, it is ubiquitous among later Germanic-associated samples in those same regions. This fact is made more interesting by the Copenhagen lab's evidence for IBD-sharing between their 4000-3000 YBP cluster and later Germanic groups within and outside of Scandinavia. Among Stolarek's recently published East Germanic samples, it has not only a strikingly high frequency but also high subclade diversity. Among the Anglo-Saxons we see lower subclade diversity (although it's still there) but nonetheless a very high frequency. In short, I very much agree with you that the data currently available to us strongly suggests that there was an initial, large expansion of I1 lineages out of Southern Scandinavia that appears to have started no earlier than around the Late Bronze Age/early IA, although it certainly didn't stop then. I should probably include the mandatory disclaimer that there were multiple patrilineages involved in that expansion, not just I1. I also think that an origin of L22 in East-Central Sweden in the early BA which you suggested multiple times on Anthrogenica is going to turn out to be pretty spot-on. This may be of some relevance to the topic of the long-standing linguistic interactions between Scandinavia and Finland, as the apparent migration of L258 from East-Central Sweden to Åland and Finland during the Iron Age could have some implications in that sense. There will be more Iron Age DNA from those regions, so I suppose we'll know in time. Thanks for adding all that detail Strider99. That's a great summary that I'll save for future reference. I agree about the role of Scania and its immediate surrounds including Zealand in the initial spread of modern I1, and think it bodes well that we came to that conclusion independently of each other and both mentioned it on Anthrogenica a few times. I also agree of course (but shouldn't have neglected to mention it yesterday) that multiple haplogroups will have been involved in all these movements, including those that spread PGmc. |
|
|
|
Post by Strider99 on Aug 26, 2023 10:53:19 GMT
I agree about the role of Scania and its immediate surrounds including Zealand in the initial spread of modern I1, and think it bodes well that we came to that conclusion independently of each other and both mentioned it on Anthrogenica a few times. Indeed! Cheers to that  |
|
|
|
Post by folcwalding on Aug 26, 2023 11:27:14 GMT
For some reasons I can't quote folc where I want to quote him. The quotation is :" If only I1 plays a part of PGmc and exclusively "southwards from Sweden" con be questioned that's too restricted I guess. Because that would mean that Anglo-Saxons - from the bottleneck- with I1 couldn't spread it for example?" Anglo-Saxons didn't spread PGmc, but West Germanic. For them PGmc was already "old story". Well, in this post I'm going to be even more restrictive. Because this post is a logical ... ... Focus I believe it is necessary to define within the thesis that I published above something like a nucleus, a core (the reactor core ), expressed with as few words as possible so that in the face of any criticism I can ask: "Okay, but do you dispute the core of my point?". Failing that, we risk reproducing on this new forum what we produced on the old one, namely several hundred pages of a labyrinthine controversy where readers were lost, often torn between boredom and exasperation (as some had told me). I thought a lot about this, weighed my words, and here it is: Simplified premisses: There are several hundred words from the Finnish and Saami lexicons which, when we go back up the chain of phonological mutations specific to the languages concerned, lead to terms that belong to the reconstructed Proto-Germanic lexicon. Archaeological evidence suggests that these words belong to the language spoken by Iron Age Scandinavians in an area containing the Lake Mälar region. (The simplifications relate mainly to the first sentence, which is a scandalously abrupt summary of hundreds of sometimes difficult etymologies, and equally difficult definitions of the phonological mutations mentioned and their chronology. Simplification also relates to the semantic scope of the borrowed vocabulary, which is an essential element. But I said I have to reduce, so I reduce. About the second sentence, there are reasons to suggest that this area also contained Scandinavian colonies permanently settled in the coastal regions of Finland, and obviously nearby Swedish regions, such as Södermanland. In the following lines, I will write "Mälaren" for this geographical zone. It's very likely also that very quickly PGmc spread to the regions in Scandinavia that were in close and regular contact with this region, namely southern Sweden and Norway; You can read for this many texts, including Engedal's thesis). Minimal Thesis (MT)
In the second part of the first millennium B.C. was spoken in Mälaren a language for which our reconstructed proto-Germanic is a good proxy.
Obviously MT doesn't say anything about the linguistic history that led to this language, nor nothing about the history that starts from this language. These are two different problems (connected but different). MT doesn't say anything neither about languages spoken elsewhere in Northern Europe in the pre-Roman Iron Age (for example Denmark, Northern Germany or Poland). MT doesn't say anything neither about the languages spoken by the "Germanic" tribes in the Migrations Times. These are both different problems (more or less connected but different). Question 1: Does someone contest MT?
Question 2: Is someone able to put an alternative to MT under the form "In the second part of the first millennium B.C. was spoken in ... a language for which our reconstructed proto-Germanic is a good proxy"? (Replace the points with a geographic zone). Please, don't joke by replacing "..." with Tanzania or Iran (I remember reading on Academia a text where on the basis of the vague resemblance of a handful of words the author defended an Iranian origin of proto-Germanic). In order to write you have to argue. Ok focus. If you want to pinpoint an Urheimat than you have to have evidence, a source. Otherwise it's impossible to pinpoint it to a certain time and place. Second: PGmc is a reconstruction, it didn't actual exist as a kind of living language. So thé proto-Germanic language didn't exist, there were most probably variants, dialects so to say. Ergo: So I don't deny the MT. But I also say it's impossible to state that, because there is no evidence. But I have an alternative MT: In the second part of the first millennium B.C. most probably was spoken in Southern Scandinavia and the North German Plain (especially along the Elbe riverine) a language for which a variant of a reconstructed proto-Germanic was spoken.
The difference is that it's impossible to restrict it to a relative small area like the Malären. Unless you have direct evidence (a source). It can be done of course see the history and spread of Latin from Rome, so the Mälaren as the Germanic Rome? Explain.... By the way if you state that proto-Germanic came from the Malären to Denmark and NE Germany what kind of language did they speak in Denmark and in Northern Germany before the linguistic influx from Sweden? When we agree that proto-Germanic is also derived from some kind of IE and assume that Single Grave people spoke a kind of IE, does this imply that Single Grave is rooted along the Malären and then came to the North Sea area in the range Netherlands-Denmark? According to Egfjord et al. (2021) they came via the Elbe from Central Europe....And if this was only Battle Axe in stead of Single Grave the impact of the language of the Single Grave people (range Netherlands-Denmark) on proto Germanic was nada? So indeed no need to recycle discussions if we are able to incorprorate the previous discussion: so the possibilities but also some impossibilities. Your MT is imo in fact nuts, because it's impossible to proof it in the first place. Unless you say ok it's my pet theory, I like the idea but that's all....then it's ok for me, but if the MT is more pretentious (^^^) |
|
|
|
Post by Anglesqueville on Aug 26, 2023 12:21:12 GMT
We (not "I") precisely have a source, namely the Saami and Finnic loans. These loans are what makes the difference between Mälaren (or "Mälaren" in the sense I indicated) and any other geographical zone in northern Europe.
It's a very serious question, that I would correct : "What kind of language did they speak in Denmark and in Northern Germany before the linguistic influx from Sweden and Norway?". This question is intimately linked to the question of the split of of W-Gmc in N-Gmc and NW-Gmc. My pet theory is that NW-Gmc formed in the late Jastorf under the influence of non-Germanic languages. But I have only speculative arguments in support of it, and anyway it's off-topic on this thread.
|
|
), expressed with as few words as possible so that in the face of any criticism I can ask: "Okay, but do you dispute the core of my point?". Failing that, we risk reproducing on this new forum what we produced on the old one, namely several hundred pages of a labyrinthine controversy where readers were lost, often torn between boredom and exasperation (as some had told me). I thought a lot about this, weighed my words, and here it is: