Seeing language from the DNA

[Anglesqueville]

Completely anecdotal of course.

[Æsir]

Æsir:

Btw, has anyone looked more closely the DNA from families with surnames like Lappalainen?

Yes. There are at least two N-subhaplogroups within the last 2000 years, which have both surname Lappalainen and Saami results in FTDNA Discover.

Also Birkarlar families?

Birkarlar and Sámi – inter-cultural contacts beyond
state control: reconsidering the standing of
external tradesmen (birkarlar) in medieval Sámi
societies

https://www.tandfonline.com/doi/pdf/10.1...16.1154676

[Jaska]

Also Birkarlar families?

Birkarlar and Sámi – inter-cultural contacts beyond
state control: reconsidering the standing of
external tradesmen (birkarlar) in medieval Sámi
societies

https://www.tandfonline.com/doi/pdf/10.1...16.1154676

It is possible.

[Jaska]

From a blog comment:
"The line of reasoning here goes in the opposite direction... We know what the genetics of people speaking Slavic looked like in the years 1000-1200 AD. If we look for the same or very similar genetically people from around 1300 BC, i.e. the time of the first Slavic innovations, we will find Proto-Slavics."

So you claim. But Slavs around 1000 CE have several genetic roots. Why do you think that you can just follow the majority autosomal ancestry and claim that language followed it? Language does not always follow majority autosomal ancestry, as you can easily see: take any old and widespread language family, and in different regions in the speaker populations of different branches there are different majority ancestries. Within Uralic: Estonians vs. Nganasans; within Turkic: Turkish vs. Yakuts; within Indo-European: Norwegian vs. Sardinian vs. Iranian, etc.

The problem in your method is that you make a presupposition which is not valid in this reality ("language always follows the majority root"). Only if it was valid, your retrospective method could provide reliable results. Now it cannot.

The only scientific way to find out, when and where distant Pre-Proto-Slavic was spoken, is to accept the linguistic results and then try to find a genetic (and/or archaeological) match for them, agreeing upon time, place, and direction of expansion. There is no law requiring that Pre-Proto-Slavs over two full millennia earlier must have been genetically similar to Slavs in 1000 CE. Language shift and expansion are socially conditioned processes, and they cannot be reliably read from the DNA or the material culture (see the examples above).

Moreover, seeing migrations reliably from the material culture has also proven to be impossible. Here are a couple of examples:

1. Funnel Beaker Culture
”The arrival of farming in northern Europe around 4000 B.C. changed substantially the life of prehistoric communities in the Baltic Sea basin and southern Scandinavia. Archaeologically, this event was marked by the development of the Funnel Beaker (also known as Trichterbecher, or TRB) cultural tradition—indeed, in the view of most scholars, Funnel Beaker culture arose as a result of the adoption of new farming practices and cultural traditions (such as new ways of making pottery, production of long-bladed flint and of polished stone tools, and new burial practices and house construction) by the local hunter-gatherer communities after a prolonged period of contact with the first farmers to the south, in central Europe.”
(Ancient Europe 8000 BC–1000 AD, vol. 1 [2004]: page 431)

Ancient DNA has shown that only ca. 40 % of the ancestry of the Funnel Beaker population was inherited from the earlier hunter-gatherers, and 60 % was inherited from the farmers (Coutinho et al. 2020: The Neolithic Pitted Ware culture foragers were culturally but not genetically influenced by the Battle Axe culture herders).

2. Corded Ware Culture

”Many different views have been voiced concerning the genesis of the Corded Ware culture. There is a division between archaeologists who allow for participation in this process by pastoral societies of the steppes near the north shore of the Black Sea and those who think that Corded Ware is a core central European phenomenon. In both camps, there are many differing views. Among the advocates of a steppe origin, the differences center on the degree that the “steppe factor” played in the genesis of Corded Ware culture, while those who favor central European roots are divided as to where specifically in central Europe the genesis of Corded Ware took place.”
(Ancient Europe 8000 BC–1000 AD, vol. 1 [2004]: page 468)

Ancient DNA studies have shown that ca. 70 % of the ancestry of the Corded Ware population was inherited from the Yamnaya population, and only ca. 30 % from the population of Globular Amphora Culture (Saag et al. 2021: Genetic ancestry changes in Stone to Bronze Age transition in the East European plain).

Linguistics, genetics, and archaeology are independent disciplines, which can only study their own objects – nothing else. it is not possible to ignore the results of any of these disciplines and make claims of its study object based only on the results of other disciplines. Language cannot be seen from the DNA or from the material culture.

[alanarchae]

From a blog comment:
"The line of reasoning here goes in the opposite direction... We know what the genetics of people speaking Slavic looked like in the years 1000-1200 AD. If we look for the same or very similar genetically people from around 1300 BC, i.e. the time of the first Slavic innovations, we will find Proto-Slavics."

So you claim. But Slavs around 1000 CE have several genetic roots. Why do you think that you can just follow the majority autosomal ancestry and claim that language followed it? Language does not always follow majority autosomal ancestry, as you can easily see: take any old and widespread language family, and in different regions in the speaker populations of different branches there are different majority ancestries. Within Uralic: Estonians vs. Nganasans; within Turkic: Turkish vs. Yakuts; within Indo-European: Norwegian vs. Sardinian vs. Iranian, etc.

The problem in your method is that you make a presupposition which is not valid in this reality ("language always follows the majority root"). Only if it was valid, your retrospective method could provide reliable results. Now it cannot.

The only scientific way to find out, when and where distant Pre-Proto-Slavic was spoken, is to accept the linguistic results and then try to find a genetic (and/or archaeological) match for them, agreeing upon time, place, and direction of expansion. There is no law requiring that Pre-Proto-Slavs over two full millennia earlier must have been genetically similar to Slavs in 1000 CE. Language shift and expansion are socially conditioned processes, and they cannot be reliably read from the DNA or the material culture (see the examples above).

Moreover, seeing migrations reliably from the material culture has also proven to be impossible. Here are a couple of examples:

1. Funnel Beaker Culture
”The arrival of farming in northern Europe around 4000 B.C. changed substantially the life of prehistoric communities in the Baltic Sea basin and southern Scandinavia. Archaeologically, this event was marked by the development of the Funnel Beaker (also known as Trichterbecher, or TRB) cultural tradition—indeed, in the view of most scholars, Funnel Beaker culture arose as a result of the adoption of new farming practices and cultural traditions (such as new ways of making pottery, production of long-bladed flint and of polished stone tools, and new burial practices and house construction) by the local hunter-gatherer communities after a prolonged period of contact with the first farmers to the south, in central Europe.”
(Ancient Europe 8000 BC–1000 AD, vol. 1 [2004]: page 431)

Ancient DNA has shown that only ca. 40 % of the ancestry of the Funnel Beaker population was inherited from the earlier hunter-gatherers, and 60 % was inherited from the farmers (Coutinho et al. 2020: The Neolithic Pitted Ware culture foragers were culturally but not genetically influenced by the Battle Axe culture herders).

2. Corded Ware Culture

”Many different views have been voiced concerning the genesis of the Corded Ware culture. There is a division between archaeologists who allow for participation in this process by pastoral societies of the steppes near the north shore of the Black Sea and those who think that Corded Ware is a core central European phenomenon. In both camps, there are many differing views. Among the advocates of a steppe origin, the differences center on the degree that the “steppe factor” played in the genesis of Corded Ware culture, while those who favor central European roots are divided as to where specifically in central Europe the genesis of Corded Ware took place.”
(Ancient Europe 8000 BC–1000 AD, vol. 1 [2004]: page 468)

Ancient DNA studies have shown that ca. 70 % of the ancestry of the Corded Ware population was inherited from the Yamnaya population, and only ca. 30 % from the population of Globular Amphora Culture (Saag et al. 2021: Genetic ancestry changes in Stone to Bronze Age transition in the East European plain).

Linguistics, genetics, and archaeology are independent disciplines, which can only study their own objects – nothing else. it is not possible to ignore the results of any of these disciplines and make claims of its study object based only on the results of other disciplines. Language cannot be seen from the DNA or from the material culture.

imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA. It’s a beer bad approach to do the opposite i.e. using a theory that a certain culture represents proto languages then trying to shoehorn  that theoretical origin model on and wondering why it doesn’t fit. So when it comes to languages, look first at the period and place when wf to know where a language was spoken, look at their DNA in that area and that time (usually the iron age or even as late as early AD) then work backwards from ‘what we know to what we don’t know’. Just blindly deciding a culture in an area must represent some first origin of a language/people must originate there and then is stupid when their is no attestatim of the language during the existence of the culture/not until many many cevturies later. You have to work back from a point of time where have evidence of language, culture and genes. 

And of course genes and language spread will be more likely to correlate in simple societies and much less so in more complex civic sorts of societies

[Jaska]

[One more example of a voluntarily blindness. Concerning the Yakutia_LNBA ancestry and the Uralic language, a commentator wrote:]
“Your view is ignoring when the DNA and the language overlay perfectly.”

You got that one wrong, too: there is no perfect overlay. There is only one Uralic-speaking population – the Nganasans – which consists mostly of the Yakutia ancestry, and it is more similar to non-Uralic populations (Dolgans and Tundra Yukaghirs) than to the other Uralic or even to the other Samoyedic populations.

Moreover, the linguistic results show that it is impossible that Proto-Uralic was ever spoken in Yakutia. And as you should know, you cannot see the language from the DNA, because language is not inherited genetically. So, make your math: clearly the Nganasans are the aberrant population here, one of the farthest ones from the Proto-Uralic population.

[Perfect overlay/lap? When it is actually just the opposite.]

[Jaska]

Alanarchae:

imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA.

Yes, if you want to trace back the roots of culture and population. But if you want to trace back the root of language, you cannot do that. There is no way how you could see the language from the material culture or from the DNA.

Alanarchae:

Just blindly deciding a culture in an area must represent some first origin of a language/people must originate there and then is stupid when their is no attestatim of the language during the existence of the culture/not until many many cevturies later.

Nobody is just blindly deciding a culture in some area representing certain (proto-)language community. Locating proto-languages is done by certain methodology, and only when the location has been found out, it is possible to find matches in the archaeological or genetic data. 

Alanarchae:

You have to work back from a point of time where have evidence of language, culture and genes.

You have to work back with the linguistic evidence, when you try to trace language. You cannot trace language from cultural or genetic evidence – that is impossible.

Alanarchae:

And of course genes and language spread will be more likely to correlate in simple societies and much less so in more complex civic sorts of societies

Language probably has spread together with certain genetic trait. The problem here is that you cannot just decide which genetic trait it was. Please tell me, by which method you could reliably identify what was the correct genetic trait associated with the spread of certain language? Explain your method to me step by step.

[Rodoorn]

Alanarchae:

imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA.

Yes, if you want to trace back the roots of culture and population. But if you want to trace back the root of language, you cannot do that. There is no way how you could see the language from the material culture or from the DNA.

@Jaska neither you can do with performative assumptions.

Alanarchae:

Just blindly deciding a culture in an area must represent some first origin of a language/people must originate there and then is stupid when their is no attestatim of the language during the existence of the culture/not until many many cevturies later.

Nobody is just blindly deciding a culture in some area representing certain (proto-)language community. Locating proto-languages is done by certain methodology, and only when the location has been found out, it is possible to find matches in the archaeological or genetic data. 

@Jaska
"Done by certain methodology" and that methodology doesn't need validation? Without it it is like cutting with a knife in fog....


Alanarchae:

You have to work back from a point of time where have evidence of language, culture and genes.

You have to work back with the linguistic evidence, when you try to trace language. You cannot trace language from cultural or genetic evidence – that is impossible.

@Jaska
Evidence is evidence. Linguistic evidence is some form of recording.

Alanarchae:

And of course genes and language spread will be more likely to correlate in simple societies and much less so in more complex civic sorts of societies

Language probably has spread together with certain genetic trait. The problem here is that you cannot just decide which genetic trait it was. Please tell me, by which method you could reliably identify what was the correct genetic trait associated with the spread of certain language? Explain your method to me step by step.

@Jaska already the difference between the spread of Single Grave Culture and Germanic spread (see mc Coll) is differentiated, in the last case much more different Y-DNA was involved in stead of  the first case.

In sum, I guess the first quote of alanarchae "imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA" is imho a reasonable way. 

[alanarchae]

Alanarchae:

imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA.

Yes, if you want to trace back the roots of culture and population. But if you want to trace back the root of language, you cannot do that. There is no way how you could see the language from the material culture or from the DNA.

Alanarchae:

Just blindly deciding a culture in an area must represent some first origin of a language/people must originate there and then is stupid when their is no attestatim of the language during the existence of the culture/not until many many cevturies later.

Nobody is just blindly deciding a culture in some area representing certain (proto-)language community. Locating proto-languages is done by certain methodology, and only when the location has been found out, it is possible to find matches in the archaeological or genetic data. 

Alanarchae:

You have to work back from a point of time where have evidence of language, culture and genes.

You have to work back with the linguistic evidence, when you try to trace language. You cannot trace language from cultural or genetic evidence – that is impossible.

Alanarchae:

And of course genes and language spread will be more likely to correlate in simple societies and much less so in more complex civic sorts of societies

Language probably has spread together with certain genetic trait. The problem here is that you cannot just decide which genetic trait it was. Please tell me, by which method you could reliably identify what was the correct genetic trait associated with the spread of certain language? Explain your method to me step by step.

In simple societies there is no mechanism or institutions where an entire language can spread without significant human movement. 

The only possible other way I can think of a common dialect being shared widely for 1000s of years without migration in simpler societies is if  there is practically unbroken intense interaction between groups who  all shared an earlier prior mother language so they all basically evolved in the same direction to share v similar descendant dialects 
I think both migration and dialect convergence might be at play with the Celts. 

But I don’t agree that you can’t infer anything about language from DNA, isotopes and  archaeology. These all can demonstrate  human permanent or temporary movements and contacts and a dialect cannot spread or converge without a human component in simple tribal societies.

[Jaska]

Rodoorn:

In sum, I guess the first quote of alanarchae "imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA" is imho a reasonable way.

Rodoorn, you still know nothing about the methods of historical linguistics. As you are not interested in adding your knowledge on that topic, perhaps you should not comment on that?

Please, tell me your method step by step: how can you see the language from the DNA or from the material culture?

[Jaska]

Alanarchae:

In simple societies there is no mechanism or institutions where an entire language can spread without significant human movement.

Nobody claims that language spreads without human movement. But we know that language spread does not require overpowering majority of immigrants. Therefore, you cannot just decide that majority ancestry is the one connected to the language. You will only get contradicting results by that method: try to follow the language from the DNA in cases like Estonians vs. Nganasan, Turkish vs. Yakuts, Norwegians vs. Sardinians vs. Iranians, etc. Please try – only then you will understand the problem.

Alanarchae:

The only possible other way I can think of a common dialect being shared widely for 1000s of years without migration in simpler societies is if  there is practically unbroken intense interaction between groups who  all shared an earlier prior mother language so they all basically evolved in the same direction to share v similar descendant dialects

What are you trying to say with this?

Alanarchae:

But I don’t agree that you can’t infer anything about language from DNA, isotopes and  archaeology. These all can demonstrate  human permanent or temporary movements and contacts and a dialect cannot spread or converge without a human component in simple tribal societies.

Of course we can see movements of people and cultural influences from the genetic and archaeological results. The problem here is that there are always many of them, yet still only one extant language lineage. You cannot just throw a dice or arbitrarily decide which migration or influence is connected to certain language. There is no way how you can see which trait was connected to the spread of the extant language lineage to this region, instead of some other ancient language lineage.

The only scientific method is that you take the linguistic results and see if any of those genetic or cultural traits is a match for them concerning time, place, and the direction of expansion. Can you understand this?

[alanarchae]

Alanarchae:

imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA.

Yes, if you want to trace back the roots of culture and population. But if you want to trace back the root of language, you cannot do that. There is no way how you could see the language from the material culture or from the DNA.

@Jaska neither you can do with performative assumptions.

Alanarchae:

Just blindly deciding a culture in an area must represent some first origin of a language/people must originate there and then is stupid when their is no attestatim of the language during the existence of the culture/not until many many cevturies later.

Nobody is just blindly deciding a culture in some area representing certain (proto-)language community. Locating proto-languages is done by certain methodology, and only when the location has been found out, it is possible to find matches in the archaeological or genetic data. 

@Jaska
"Done by certain methodology" and that methodology doesn't need validation? Without it it is like cutting with a knife in fog....


Alanarchae:

You have to work back from a point of time where have evidence of language, culture and genes.

You have to work back with the linguistic evidence, when you try to trace language. You cannot trace language from cultural or genetic evidence – that is impossible.

@Jaska
Evidence is evidence. Linguistic evidence is some form of recording.

Alanarchae:

And of course genes and language spread will be more likely to correlate in simple societies and much less so in more complex civic sorts of societies

Language probably has spread together with certain genetic trait. The problem here is that you cannot just decide which genetic trait it was. Please tell me, by which method you could reliably identify what was the correct genetic trait associated with the spread of certain language? Explain your method to me step by step.

@Jaska already the difference between the spread of Single Grave Culture and Germanic spread (see mc Coll) is differentiated, in the last case much more different Y-DNA was involved in stead of  the first case.

In sum, I guess the first quote of alanarchae "imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA" is imho a reasonable way. 

It’s the one that involves the least assumptions. Obviously it can’t be done if there is no record of the language that at least roughly coincides in time with available DNA. But the bottom line is Celts, Germans, Italic, whatever are defined by what they spoke and we can with luck recover ancient DNA that shows the genetics of the speakers around the time when we first know for sure what they spoke. Then of course you can look at DNA and isotopes is steps prior to historical/linguistic records to understand how long there has been population cintinuity/discontinuity and how much people have been moving about, This is valid because the idea a new language can appear or spread in a simple society without human movement (permanent or just temporary lifetime movement) is nonsense. Why would we not use methods like DNA and isotopes to establish permanent r much  temporary lifetime movement of peoples to help understand how languages spread or changed. 

Archaeology other than using it for DNA and isotopes is a bit more hazy in terms of interpretation of whether it is movement or just influence but it still indicates human agency and tells you the direction of that contact. I think throwing away everything but linguistics itself is nihilistic

[Rodoorn]

Rodoorn:

In sum, I guess the first quote of alanarchae "imo you need to work back from what you know from historical data when the language of a people is first recorded and then look at the archaeology and ancient DNA" is imho a reasonable way.

Rodoorn, you still know nothing about the methods of historical linguistics. As you are not interested in adding your knowledge on that topic, perhaps you should not comment on that?

Please, tell me your method step by step: how can you see the language from the DNA or from the material culture?

That posting of you remembers me when I was a 'wise-crack' at high school, in a debate I said: 'did you ever read a newspaper' to my opponent. Way back then I was 15/16. Hopefully I'm now a bit wiser

[alanarchae]

Alanarchae:

In simple societies there is no mechanism or institutions where an entire language can spread without significant human movement.

Nobody claims that language spreads without human movement. But we know that language spread does not require overpowering majority of immigrants. Therefore, you cannot just decide that majority ancestry is the one connected to the language. You will only get contradicting results by that method: try to follow the language from the DNA in cases like Estonians vs. Nganasan, Turkish vs. Yakuts, Norwegians vs. Sardinians vs. Iranians, etc. Please try – only then you will understand the problem.

Alanarchae:

The only possible other way I can think of a common dialect being shared widely for 1000s of years without migration in simpler societies is if  there is practically unbroken intense interaction between groups who  all shared an earlier prior mother language so they all basically evolved in the same direction to share v similar descendant dialects

What are you trying to say with this?

Alanarchae:

But I don’t agree that you can’t infer anything about language from DNA, isotopes and  archaeology. These all can demonstrate  human permanent or temporary movements and contacts and a dialect cannot spread or converge without a human component in simple tribal societies.

Of course we can see movements of people and cultural influences from the genetic and archaeological results. The problem here is that there are always many of them, yet still only one extant language lineage. You cannot just throw a dice or arbitrarily decide which migration or influence is connected to certain language. There is no way how you can see which trait was connected to the spread of the extant language lineage to this region, instead of some other ancient language lineage.

The only scientific method is that you take the linguistic results and see if any of those genetic or cultural traits is a match for them concerning time, place, and the direction of expansion. Can you understand this?

It depends what geographical area you are talking about. It would indeed be difficult in a busy highway kind of area of historical movements but there are other areas where DNA only shows one or two significant post-Neolithic population changes prior to the first linguistic/historical attestation of language . So the choice of when’s the later attested historical language arrived or evolved is very small. 

By the way I suspect isotopes may prove as important as DNA because there is clear evidence from the bronze age into historic times of the practice of the elites fostering each others children over long distances. Basically swapping their kids in infancy and then returning to their home tribe as young adults. That is a great way to ensure dialects among the elites stay converged/divergence is seriously slowed artificially. I actually think this is likely a key factor in at least some of the distributing of Celtic as a dialect spoken by elites. ie lots of temporary migration among the elites of bronze  age Europe where future kings and chiefs etc were fully immersed since infancy in the language, culture and beliefs of other tribes before returning home in adulthood more like a member of his foster tribe than his biological one. That imo, other than permanent migration, is likely the most important mechanism in language in later prehistory

[Jaska]

Alanarchae:

It depends what geographical area you are talking about. It would indeed be difficult in a busy highway kind of area of historical movements but there are other areas where DNA only shows one or two significant post-Neolithic population changes prior to the first linguistic/historical attestation of language . So the choice of when’s the later attested historical language arrived or evolved is very small.

Define “significant population change”? And why do you think that language shift requires such a significant population change? In any case, it only works to one direction: when there is a significant population change, the probability of a new language is high; but even if there is no significant population change, there is a possibility for a new language.

Also, can you give a concrete example of your method for tracing language through the DNA or material culture?

Alanarchae:

By the way I suspect isotopes may prove as important as DNA because there is clear evidence from the bronze age into historic times of the practice of the elites fostering each others children over long distances. Basically swapping their kids in infancy and then returning to their home tribe as young adults. That is a great way to ensure dialects among the elites stay converged/divergence is seriously slowed artificially. I actually think this is likely a key factor in at least some of the distributing of Celtic as a dialect spoken by elites. ie lots of temporary migration among the elites of bronze  age Europe where future kings and chiefs etc were fully immersed since infancy in the language, culture and beliefs of other tribes before returning home in adulthood more like a member of his foster tribe than his biological one. That imo, other than permanent migration, is likely the most important mechanism in language in later prehistory

Even in such cases we cannot see from the DNA or the cultural remains which of these languages truly prevailed: the assumed elite language or the assumed commoner language.