Seeing language from the DNA

[Kale]

Not to get off topic, but ChatGPT pontificates very well, however that only covers the fact that it is utterly useless for obscure/specialized knowledge.
Some humorous results from questions I've asked it before (paraphrasing)...
Q: 'How does ZlatyKun relate to Basal Eurasians'
A: ZlatyKun does not relate to Basal Eurasians, ZlatyKun is a type of horse.

Q: 'How tall is the tallest active sumo wrestler?'
A: Ichinojo is the tallest active sumo wrestler, standing 6 feet 12 inches tall.
...Ichinojo is A) not the tallest, B) not as tall as specified, C) who the hell says 6 feet 12 inches!?

[jdean]

: )))))

[jdean]

Not to get off topic, but ChatGPT pontificates very well, however that only covers the fact that it is utterly useless for obscure/specialized knowledge.
Some humorous results from questions I've asked it before (paraphrasing)...
Q: 'How does ZlatyKun relate to Basal Eurasians'
A: ZlatyKun does not relate to Basal Eurasians, ZlatyKun is a type of horse.

Q: 'How tall is the tallest active sumo wrestler?'
A: Ichinojo is the tallest active sumo wrestler, standing 6 feet 12 inches tall.
...Ichinojo is A) not the tallest, B) not as tall as specified, C) who the hell says 6 feet 12 inches!?

It's also pretty crap at helping you out with your Ebay order : )

[JonikW]

Not to get off topic, but ChatGPT pontificates very well, however that only covers the fact that it is utterly useless for obscure/specialized knowledge.
Some humorous results from questions I've asked it before (paraphrasing)...
Q: 'How does ZlatyKun relate to Basal Eurasians'
A: ZlatyKun does not relate to Basal Eurasians, ZlatyKun is a type of horse.

Q: 'How tall is the tallest active sumo wrestler?'
A: Ichinojo is the tallest active sumo wrestler, standing 6 feet 12 inches tall.
...Ichinojo is A) not the tallest, B) not as tall as specified, C) who the hell says 6 feet 12 inches!?

That's hilarious. On the first one, to be fair it's just translating it as a golden horse.* I wonder what might happen if you're more specific. Angles got some impressive responses there. Looks like AI really might match all the hype one day.

*ADD: sorry, for clarity I mean zlatý kůň = "golden horse" I only knew because of the similarity to Russian...

[Anglesqueville]

I specified that I had tried ChatGPT on these subjects "out of curiosity". Perhaps I should have added that I expected to roll on the floor laughing, which didn't happen. On other subjects, I too, like everyone else, got such absurd answers that I can't imagine where the algorithm went to look for them (I remember a hilarious etymology of the Norman toponym "Fécamp" as "battle of the fjord"). In this case, I find the response "clean", as we say in Normandy to describe something which is neither scandalous nor exciting.

[JonikW]

Not to get off topic, but ChatGPT pontificates very well, however that only covers the fact that it is utterly useless for obscure/specialized knowledge.
Some humorous results from questions I've asked it before (paraphrasing)...
Q: 'How does ZlatyKun relate to Basal Eurasians'
A: ZlatyKun does not relate to Basal Eurasians, ZlatyKun is a type of horse.

Q: 'How tall is the tallest active sumo wrestler?'
A: Ichinojo is the tallest active sumo wrestler, standing 6 feet 12 inches tall.
...Ichinojo is A) not the tallest, B) not as tall as specified, C) who the hell says 6 feet 12 inches!?

That's hilarious. On the first one, to be fair it's just translating it as a golden horse.* I wonder what might happen if you're more specific. Angles got some impressive responses there. Looks like AI really might match all the hype one day.

*ADD: sorry, for clarity I mean zlatý kůň = "golden horse" I only knew because of the similarity to Russian...

I'm not bigging up the current technology but I added the words "the" and "skeleton" to the question and got a more interesting response. How does the ZlatyKun skeleton relate to Basal Eurasians?:

"The ZlatyKun skeleton is a 7,500-year-old individual found in the Czech Republic. It is one of the few well-preserved ancient DNA samples from the Mesolithic period in Europe. Genetic analysis of the ZlatyKun skeleton suggests that it belonged to a population with genetic affinity to both Western hunter-gatherers and Eastern hunter-gatherers.

"Basal Eurasians, on the other hand, are a hypothetical population that represents an early split from other Eurasian populations. They are thought to have diverged from other Eurasian populations around 36,000 years ago and contributed to the genetic makeup of modern Europeans and Asians.

"While the ZlatyKun skeleton does not directly represent Basal Eurasians, its genetic makeup indicates a mixture of genetic components that likely include contributions from both Western and Eastern hunter-gatherers. This suggests that the ZlatyKun population may have had genetic connections to the ancestral populations that eventually gave rise to Basal Eurasians. However, further research is needed to fully understand the genetic relationships between the ZlatyKun skeleton and Basal Eurasians."

I'll have to ask what language it spoke next. :-)

[Jaska]

A response by Minerva Piha, Mikko K. Heikkilä, and Jaakko Häkkinen to the article Archaeology, Language, and the Question of Sámi Ethnogenesis by Asgeir Svestad and Bjørnar Olsen (2023). We correct some major misunderstandings made by Svestad and Olsen concerning the methodology of historical linguistics and its relation to archaeology.
https://www.academia.edu/114870366/Comme...hnogenesis

Same errors as those made when interpreting archaeological results are often made when interpreting genetic results, so this fresh new response fits in the theme of this thread.

[Jaska]

Results concerning distant relatedness or ancient contacts are contradicting: PrePIE could have been spoken in the northeast, close to Pre-Proto-Uralic; or it could have been spoken in the south, close to the Caucasian language families.

I'm not versed in linguistics at all, so I defer to you to assess my understanding here.
Would a perfectly reconstructed proto-language really be accurate in such a way that if someone with complete knowledge of it, plopped back into the right village at the right time, would speak 100% identically to the people in that place? Or was there never a people that spoke that language 100% exactly, but people over a broader region and time would speak dialects that would be largely understandable by it?
If the latter, are the PPU and Caucasian contacts really contradicting? Could not broadly PPIE dialects have encompassed the whole region between them, and those on the frontiers have filtered the influences throughout the region?

This is a good question. The more there are independent branches which have developed to different  directions, the more certain the reconstructions become. Of course there will always be minor details to which will never get absolute certainty, like what was the actual phonetic realization of the Indo-European laryngeals or was the Proto-Uralic non-low vowel in later syllables *i or *ǝ, but there are probably greater instance-related variation in the pronunciation of any single living individual speaking any language of the world (therefore the dialect question weighs less here). 

Just imagine yourself speaking your native language while yawning, while heavily drunk or sedated, between your closed teeth while lifting a heavy weight etc. Still the others can understand you. So, the phonological reconstruction must be close enough so that it does not cause any real problems for intelligibility.

More difficult is the vocabulary, because languages lose words all the time. After those 4000-6000 years since the proto-languages, huge number of words has been lost - or they have been preserved only in so few languages that we cannot recognize their descending from the proto-language. The younger the loanword layer, the more it usually contains words. There are thousands of Swedish loanwords in Finnish, but only few hundred words inherited from Proto-Uralic. 

The secondary extent of some stage of Proto-Indo-European could have been so wide, but the primary narrow homeland is still an inescapable fact, because there can be no uniform language without the narrow primary homeland. In that case it would be interesting to compare, do either the Indo-Caucasian or the Indo-Uralic allegedly shared ancient features indicate greater time depth than the other. For this I cannot say anything at the moment. Perhaps some nostraticists etc. have compared these language families, but unfortunately in that framework the criteria are too loose, so that the data must contain a lot of false positive "cognates."

[Kale]

Another question pertaining to understanding the dynamics of proto-languages...
Could a sequence of events happen as follows?
1) Speakers of a language expand from a small territory to a medium territory and form a range of dialects.
2) Some event happens that increases contact/interaction between the dialect communities (increase in mobility, trade, etc.)
3) Some degree of linguistic convergence occurs (but not total).
4) The 'culture' expands on multiple frontiers, eventually developing dialects (or further developing them) in the new territories.
Would the dialects that developed during stage 1 appear to have developed during stage 4, as a result of the convergence during stage 3?

[Jaska]

Another question pertaining to understanding the dynamics of proto-languages...
Could a sequence of events happen as follows?
1) Speakers of a language expand from a small territory to a medium territory and form a range of dialects.
2) Some event happens that increases contact/interaction between the dialect communities (increase in mobility, trade, etc.)
3) Some degree of linguistic convergence occurs (but not total).
4) The 'culture' expands on multiple frontiers, eventually developing dialects (or further developing them) in the new territories.
Would the dialects that developed during stage 1 appear to have developed during stage 4, as a result of the convergence during stage 3?

If they were true dialects so that there were also separating sound changes, these never level back. It is impossible that all the same sound changes would have later reversed exactly the same way. So the sound changes would still reveal the earlier dialect boundaries, and the newer dialectal differences were layered upon them.

______________Speech communities:
Stage 1 dialects: AAAABBBBBBBBBBBBBBBBBCCCC
Stage 4 dialects: 111111 222222 3333333 444444

In the present languages we would still be able to see that the earlier dialect boundaries were in different spots (between communities A1 and B1; between communities B4 and C4) than the later dialect boundaries. 

Finnish and Estonian languages are examples of languages having the present language boundary very different from the ancient dialect boundaries. Until recently, South Estonian dialects have been considered part of Estonian language due to their long convergence, but still the ancient sound changes show that South Estonian was the first unit to diverge from Proto-Finnic. North Estonian and Finnish share the same development (*kakti '2'):

__Nominative : genitive : partitive : essive__
1. *kakti : *kakte-n : *kakte-ta : *kakte-na 
2. *kakci : *kakte-n : *kakte-ta : *kakte-na 
3. *kakci : *kahte-n : *kahte-ta : *kahte-na 
4. *kaksi : *kahde-n : *kahte-da : *kahte-na 
5a. kaksi : kahde-n : kaht-a : kahte-na (Finnish)
5b. kaks  :  kahe  :  kaht  :  kahe-na (Estonian)

On the other hand, South Estonian development has been different since the stage 3:

__Nominative : genitive : partitive : essive__
1. *kakti : *kakte-n : *kakte-ta : *kakte-na 
2. *kakci : *kakte-n : *kakte-ta : *kakte-na 
3. *kakci : *katte-n : *katte-ta : *katte-na 
4. *katsi : *katë-n : *kattë-da : *kattë-na 
5. *katś : *katë-n : *kattë-da : *kattë-na 
6. katś  :  katõ  :  kattõ (South Estonian; essive is not productive)

Shared convergence development between North and South Estonian (loss of final vowel; loss of genitive *-n; development of consonant gradation) cannot prevent us from seeing that there also were early sound changes separating South Estonian from North Estonian (*kt > *tt; progressive palatal harmony *a-e > *a-ë).

[Jaska]

Svestad & Olsen: Archaeology, Language, and the Question of Sámi Ethnogenesis
https://www.academia.edu/106210245/Archa...hnogenesis

This is the original article containing several misunderstandings (response to this linked a few messages earlier). Is here anybody who wants to try to defend their view?

[Anglesqueville]

^^

The Siberian genetic component is substantially higher in the individuals from Bolshoy Oleni Ostrov, and Lamnidis et al. (2018, 6, 8) suggest that it was introduced to the area not later than 4000 BP and possibly earlier. If this is right, though caution should be taken, it means that these genetic ancestors of the Sámi have lived in the northeastern Sámi settlement area since at least the beginning of the Early Metal Age.
This would accordingly speak against the claim that the Sámi had no ancestry here before the early common
era arrival of the Proto-Sámi speakers, who from their homeland in southeastern Finland are presumed to
have moved north and replaced the original population (e.g. Heikkilä 2014, 135–136).

Reading this I almost fell off my chair. I agree, we will have to work on this text.

[Parastais]

Buuut, can we be sure they are not (part of) genetic ancestors of Sami?
I mean they are not writing “linguistic ancestors of Sami”..

[Anglesqueville]

Buuut, can we be sure they are not (part of) genetic ancestors of Sami?
I mean they are not writing “linguistic ancestors of Sami”..

Yes, I accept the criticism. My reading was "THE genetic ancestors", and that reading was rushed and abusive. If I replace it with "part of genetic ancestors", then this morning I will simply answer that there is work to be done. The latest published study (for which we are still waiting for the data to be published on edmond.mpg) characterizes BOO as a fairly balanced mix of Siberian genetics and EEHG. If I add the already published samples of BOO to the sources which give good qpAdm models for modern Saamis or those from Levänluhta (Allentoft's Minino_IA, Peltola's Volga_Oka_IA, and a Scandinavian Germanic source, like Allentoft's Falköping_IA), BOO is preserved with a significant coefficient and excellent tail_prob. But should this necessarily be interpreted as a participation of BOO in the genetic constitution of the targets? I will only say this morning that I am not sure. My intuition is that BOO intervenes in these models mainly as a vector of an indigenous EEHG bonus. But this intuition needs to be verified.

[Æsir]

Sami are originally "Bjarmian" long distance traders setting up seasonal camps in suitable locations for metal working to exchange tools and weapons for furs from the native peoples of Fennoscandia.
After working and trading during the winter they would return trough the river systems in the spring with boats loaded with furs and ivory, until some stayed as they married in to the tribes and the dialects separated.

You are welcome.