I haven't read the papers in detail, but can someone explain how genetics can be used to trace spread of languages? For context, you don't need population movements for a language to spread (it is similar to religion). See this article for a logical explanation: https://medium.com/incerto/a-few-things-we-dont-quite-get-ab...
You can’t. But if population A and population B share a ancestor X years ago, and they also speak languages that appear to have drifted apart by X many years, the inference that their ancestor spoke a common proto-language is the simplest explanation.
You are correct that the spread of genetics and the spread of language do not have to coincide. However, in this case, it seems that they do.
If you study the genomes of the populations of Europe as well as parts of Central and South Asia, you can reconstruct a very broad family tree rooted in a shared genetic ancestry from in a population who lived somewhere in Eurasia at a certain point in time. If you also study the languages of those same populations, you can independently reconstruct a family tree of languages that culminates in the reconstructed Proto-Indo-European language that would have existed at the same point in time. The simplest explanation for this is the spread of Indo-European-speaking populations, and not merely the language itself, from a single ancestral population.
Well, you can't. In this case I believe they're already pretty confident about who the PIE speaking people are (the "Yanmaya") and this study is about tracking down where they originally lived. And they have shown that they mostly replaced the previous European population rather than transferring the languages to them.
David Reich is aggressive about these genetics results though. IIRC I read a NYT story once where he came in and claimed to have upended all of Polynesian history based on the genetics of a few historical skulls they found, but it didn't seem like strong enough evidence to me.
I can't find it now, but I've seen at least one claim that the ancestral-human-DNA world inside biology is fairly dominated by a clique, and if you're not seen as fully on the team you can't expect to be funded, published and so on. Which isn't to say that any specific claim is wrong, of course, and on the whole it seems very unlikely that they're far wrong on the bare facts, as opposed to more speculative interpretations.
I wonder how this lower Volga group interacted with the earliest known civilization (5000-3000 BCE), in modern day Ukraine, the Cucuteni–Trypillia [1].They had cities with between 20-40k people that overlapped with the Yamnaya people’s discussed in the article (3500-2500 BCE).
They had agriculture as well as wheels for transportation and pottery. All predating middle eastern civilizations.
They also burned down their own cities every 50-100 years.
This culture was in constant threat from the nomads of the steppe and they learned to live in large groups as protection. This hypothesis is discussed at the end of a recent publication [2: p219-220].
how do archaeologists know they burned down their own cities? One would expect the more neutral statement 'Their cities were burned every 50-100 years".
I recently came across this presentation of Kristian Kristiansen, University of Gothenburg: "Towards a New European prehistory: genes, archaeology and language" (2023): https://www.youtube.com/watch?v=bxTVSwt-jsU [video], which I enjoyed very much. Prof. Kristiansen is a leading researcher in this area.
David Reich, one of the principal authors of the study in question, wrote a book a few years back titled "Who We Are and How We Got Here", which I quite liked (https://www.goodreads.com/review/show/2605841954). It predates some of this research, obviously, but it does have a chapter on the Indo-European origin question, along with chapters on a lot of other interesting paleo-DNA research.
He was on the Dwarkesh podcast last August to provide some lay person friendly synopsis and updates to “Who We Are.” Worth listening to even if you have read the book (in my mind at least).
Warning, link has an auto play when I opened it (but don’t let that minor obnoxiousness dissuade you from listening).
Can someone smarter than me explain how it's even possible to use DNA to identify the origin of a language, given that e.g. if this were tried with a language like German (or maybe any Western European language) the puzzle would look very confusing and is not DNA based.
The story with the Indo-Europeans is basically as follows:
1. By intersecting ancient word sets of ancient Indo-European languages using comparative phonetics we can try and reconstruct the words of the proto-IE language, both their approximate sounds and approximate meanings. This gives us some information about the society. E.g., the PIE language very likely had a word for wheel, which puts the common PIE community in the period after the wheel was invented. Other words can help us guess what landscape the PIE people lived in, and it has been generally assumed for almost a century now that it strongly resembles Southeastern Europe, essentially the Ukrainian steppe. Two alternative hypotheses (modern-day Turkey and the area to the north, in modern-day Poland/Ukraine) had different drawbacks. We can also look at the locations of the earliest historically attested IE groups (Europe, Middle East, Punjab, Anatolia) and try and guess where they all may have had come from, given the time frame.
2. By looking at the descriptions of the earliest IE societies (first of all the society of Rig-Veda), we can try and guess what way of life these people had. We can then look at all the archaeological cultures in the roughly appropriate area from the roughly appropriate time frame and see which of those have features of interest (in the IE case, warrior-like culture with social stratification, etc.).
3. We know that IE migrated a lot and provided a lot of genetic material to modern populations in Europe and some other regions. Since quite recently, by looking at palaeo-DNA data from the remains of the people who belonged to these cultures, we can try and check who of them made the biggest contribution to contemporary populations.
All these sources of data are rather imprecise, but if you combine them all together and see a clear pattern, this looks rather convincing.
I fail to understand how the Rigvedic society can be connected to this DNA research. Rigveda never mentions anything beyond the Punjab/Swat/Haryana region in any of the hymns. The flora and fauna mentioned in it is also exclusive to this region. Lastly there is no mention of an ancient homeland both in Rigveda and Avesta.
PIE reconstructions are very interesting peaces of linguistic, but they seems often mistaken. One great analogy, I first saw presented in some Linguisticae[1] video I think, is "what if we had no direct trace of Latin and we were looking to recreate proto-Romance roots." Of course Latin itself refers to very wide set of linguistic practices, with all the diversity we can imagine through time, space, individuals and even for a given individual there are difference as they age and depending of context they will use different sociolects and language register, plus of course not everyone is mono-linguistic.
It's about the origin of a population whose widely dispersed descendants often speak a language whose primary features descend from the language spoken by the original population (albeit changed via thousands of years of drift and borrowing from other languages).
That doesn't mean that a) all features of the descendant language come from the origin language or b) all speakers of the descendant language have ancestry from the original population.
This study is about prehistoric Steppe peoples, there are no Indo-European inscriptions from this time period nor would there be any until several millennia after this time.
My dad has literally just published a book (in Russian) with about 850 words with near identical sound and meanings in Russian and other Slavonic languages. :)
It was interesting hearing him talk about how you can see pieces of the original proto language preserved in the different languages. E.g. Russian has 6 cases, Sanskrit has some of these but also others and the original language had something like 12 (I don’t have any particular knowledge on the subject so might be misremembering).
For me it was interesting that the original language seemed to be more complex than the modern descendants, like there is a general trend towards simplification with time. In my mind then there is the question as to where the original complex language came from and why would a culture that we would consider more primitive that ours would need and come up with one.
The complexity of natural human languages comes in different forms, but as a general rule, whenever you see something that's built into another language and "missing" from your own, you can express it by using more words. For example, PIE had a lot of noun cases that aren't in English, but you don't need the instrumental case to precisely express its purpose. You can say something like "by means of a forklift."
Some studies actually suggest that literacy systematically pressures languages to use longer, more complex sentences, thus disincentivizing complex inflection rules.
Since you seem to be quoting the Sanskrit words in their root forms, (to which the case-lacking English and German equivalents most closely correspond) your spellings are incorrect. The correct forms are:
pitr
mātr
bhrātr
duhitr
No thematic 'a' on the end.
You might be confusing it with the nominative plural case forms:
Similarities like these, especially with Latin in the mix, were the clue that originally put early linguists on the scent of the IE language family several centuries ago. Since then, extensive research has been done into how exactly these languages developed from their common ancestors. Some modern dictionaries, like Wiktionary, contain entire family trees comparing the divergent development of these cognates and many, many others.
Although it is about numerals, there are words in a few languages, on the right side.
And Sanskrit is the ancestor of many Indian language, such as the regional languages of most of the northern (e.g. Punjabi, Haryanvi, Himachali, Hindi and its dialects), central (e.g. Hindi), eastern (e.g. Bengali, Odiya) and western (e.g. Gujarati, Marwadi) Indian states. To a rough approximation, only the languages of the 4 (now 5, with Telangana added) southern states, and of the 6 / 7 north-eastern states (Assam, Manipur, Mizoram, Meghalaya, etc.) and maybe a few aboriginals' / forest tribals' languages, like Bhil, Gond, etc., don't descend from Sanskrit.
The numbers of one to ten across the three main Indo-European namely Sanskit, English and German just confirmed they are from the same language tree.
The same goes to Malay-Austronesian language family that is spoken in Taiwan, Malay archipelago and further away in Polynesian islands including native people of New Zealand and Hawaii, their numbers of one to ten are very similar accross very wide geographical area confirming they are from the same language tree. Fun facts their most common word is (nyior/nyiur) which further cemented their status as the community with largest number of islands because coconut tree is trademark of their islands environment.
This one is slightly more interesting than a mere cognate as it is believed that the Proto-Indo-European speakers worshipped a sky god with the reconstructed name *Dyḗus ph₂tḗr ("sky-father") which is the ancestor of these (also Tyr and the like on the Germanic side). See:
https://en.wikipedia.org/wiki/*Dy%C4%93us
"*Dyēus is considered by scholars the most securely reconstructed deity of the Indo-European pantheon, as identical formulas referring to him can be found among the subsequent Indo-European languages and myths of the Vedic Indo-Aryans, Latins, Greeks, Phrygians, Messapians, Thracians, Illyrians, Albanians and Hittites."
Could you explain in non-specialist language how similarities between these modern languages now has anything to do with their relationship from some earliest common ancestor? How is that explanation better than convergent evolution or overfitting hallucinations?
When I look at the difference between modern and “old English” they seem to have changed quite a bit [0]. When I read an etymological explanation [1], it sounds like a just so story.
English is a bit special in that it's a relatively modern mix of Old English (aka Anglo-Saxon) and what the invading Normans spoke (a Romance language), plus some more. So when you compare words it's maybe better to look at the origins of the modern English words. "Ignite", for example, is from Latin "Ignitus", via the Normans. It's fine to include English when comparing words from different IE languages, but perhaps not as the only "Western" example. Wikipedia has a much broader list which is more interesting: https://en.wikipedia.org/wiki/Indo-European_vocabulary
But it's not as good as I would wish. English is included as the only modern western European language. No German, no Swedish, no Icelandic, no Dutch etc.
The explanation is better if it allows you to explain a large number of similar words arising from a common source by a systematic process.
If you have to make up a new just-so story for every pair of words, of course you're not gaining much, but if the same story works for many words at the same time, positing a common origin isn't too far-fetched.
Readit News
If you study the genomes of the populations of Europe as well as parts of Central and South Asia, you can reconstruct a very broad family tree rooted in a shared genetic ancestry from in a population who lived somewhere in Eurasia at a certain point in time. If you also study the languages of those same populations, you can independently reconstruct a family tree of languages that culminates in the reconstructed Proto-Indo-European language that would have existed at the same point in time. The simplest explanation for this is the spread of Indo-European-speaking populations, and not merely the language itself, from a single ancestral population.
David Reich is aggressive about these genetics results though. IIRC I read a NYT story once where he came in and claimed to have upended all of Polynesian history based on the genetics of a few historical skulls they found, but it didn't seem like strong enough evidence to me.
Primarily the male population. Genetically much higher proportion of the female population survived.
Of course that’s an exaggeration as well. In much of Southern Europe and other areas the replacement was far from full.
https://en.wikipedia.org/wiki/Kurgan_hypothesis
This is absolutely true.
Dead Comment
They had agriculture as well as wheels for transportation and pottery. All predating middle eastern civilizations.
They also burned down their own cities every 50-100 years.
This culture was in constant threat from the nomads of the steppe and they learned to live in large groups as protection. This hypothesis is discussed at the end of a recent publication [2: p219-220].
[1] https://en.wikipedia.org/wiki/Cucuteni%E2%80%93Trypillia_cul...
[2] https://pure.tudelft.nl/ws/portalfiles/portal/144861667/enig...
Warning, link has an auto play when I opened it (but don’t let that minor obnoxiousness dissuade you from listening).
https://www.dwarkeshpatel.com/p/david-reich
1. By intersecting ancient word sets of ancient Indo-European languages using comparative phonetics we can try and reconstruct the words of the proto-IE language, both their approximate sounds and approximate meanings. This gives us some information about the society. E.g., the PIE language very likely had a word for wheel, which puts the common PIE community in the period after the wheel was invented. Other words can help us guess what landscape the PIE people lived in, and it has been generally assumed for almost a century now that it strongly resembles Southeastern Europe, essentially the Ukrainian steppe. Two alternative hypotheses (modern-day Turkey and the area to the north, in modern-day Poland/Ukraine) had different drawbacks. We can also look at the locations of the earliest historically attested IE groups (Europe, Middle East, Punjab, Anatolia) and try and guess where they all may have had come from, given the time frame.
2. By looking at the descriptions of the earliest IE societies (first of all the society of Rig-Veda), we can try and guess what way of life these people had. We can then look at all the archaeological cultures in the roughly appropriate area from the roughly appropriate time frame and see which of those have features of interest (in the IE case, warrior-like culture with social stratification, etc.).
3. We know that IE migrated a lot and provided a lot of genetic material to modern populations in Europe and some other regions. Since quite recently, by looking at palaeo-DNA data from the remains of the people who belonged to these cultures, we can try and check who of them made the biggest contribution to contemporary populations.
All these sources of data are rather imprecise, but if you combine them all together and see a clear pattern, this looks rather convincing.
I fail to understand how the Rigvedic society can be connected to this DNA research. Rigveda never mentions anything beyond the Punjab/Swat/Haryana region in any of the hymns. The flora and fauna mentioned in it is also exclusive to this region. Lastly there is no mention of an ancient homeland both in Rigveda and Avesta.
[1] https://www.youtube.com/@Linguisticae
It's about the origin of a population whose widely dispersed descendants often speak a language whose primary features descend from the language spoken by the original population (albeit changed via thousands of years of drift and borrowing from other languages).
That doesn't mean that a) all features of the descendant language come from the origin language or b) all speakers of the descendant language have ancestry from the original population.
https://www.biorxiv.org/content/10.1101/2024.04.17.589597v1
https://www.biorxiv.org/content/10.1101/2024.04.17.589600v1
Fun facts, the most common words of Indo-European Family are surprisingly very similar across Sanskrit (S) <--> English (E) <--> German (G) [3].
Pitara (S) <--> Father (E) <--> Vater (G)
Matara (S) <--> Mother (E) <--> Mutter (G)
Bhratara (S) <--> Brother (E) <--> Bruder (G)
Duhitar (S) <--> Daughter (E) <--> Tochter (G)
[1] New insights into the origin of the Indo-European languages (147 comments):
https://news.ycombinator.com/item?id=36930321
[2] Ancient genomes provide final word in Indo-European linguistic origins (16 comments):
https://news.ycombinator.com/item?id=42515584
[3] Turandot and the Deep Indo-European Roots of “Daughter” (15 comments):
https://news.ycombinator.com/item?id=29450507
https://borissoff.wordpress.com/2025/02/06/russian-sanskrit-...
For my part I built the web based editing tool, DB and LaTeX generation system that he used to assemble this massive undertaking over the years. :)
https://borissoff.wordpress.com/2015/10/30/first-public-pres...
It was interesting hearing him talk about how you can see pieces of the original proto language preserved in the different languages. E.g. Russian has 6 cases, Sanskrit has some of these but also others and the original language had something like 12 (I don’t have any particular knowledge on the subject so might be misremembering).
For me it was interesting that the original language seemed to be more complex than the modern descendants, like there is a general trend towards simplification with time. In my mind then there is the question as to where the original complex language came from and why would a culture that we would consider more primitive that ours would need and come up with one.
Some studies actually suggest that literacy systematically pressures languages to use longer, more complex sentences, thus disincentivizing complex inflection rules.
> Matara (S) <--> Mother (E) <--> Mutter (G)
> Bhratara (S) <--> Brother (E) <--> Bruder (G)
> Duhitar (S) <--> Daughter (E) <--> Tochter (G
Since you seem to be quoting the Sanskrit words in their root forms, (to which the case-lacking English and German equivalents most closely correspond) your spellings are incorrect. The correct forms are:
pitr
mātr
bhrātr
duhitr
No thematic 'a' on the end.
You might be confusing it with the nominative plural case forms:
pitarah
mātarah
bhrātarah
duhitarah
>Matara (S) <--> Mother (E) <--> Mutter (G)
Also some roots of the smaller natural numbers, like (E): one, two, three, four, five, six, seven, eight, nine, ten, etc.
(G) eins, zwei, drei, ...
(S) eka, dvi, tri, ...
See the "Table" here:
https://en.m.wikipedia.org/wiki/Devanagari_numerals
Although it is about numerals, there are words in a few languages, on the right side.
And Sanskrit is the ancestor of many Indian language, such as the regional languages of most of the northern (e.g. Punjabi, Haryanvi, Himachali, Hindi and its dialects), central (e.g. Hindi), eastern (e.g. Bengali, Odiya) and western (e.g. Gujarati, Marwadi) Indian states. To a rough approximation, only the languages of the 4 (now 5, with Telangana added) southern states, and of the 6 / 7 north-eastern states (Assam, Manipur, Mizoram, Meghalaya, etc.) and maybe a few aboriginals' / forest tribals' languages, like Bhil, Gond, etc., don't descend from Sanskrit.
The same goes to Malay-Austronesian language family that is spoken in Taiwan, Malay archipelago and further away in Polynesian islands including native people of New Zealand and Hawaii, their numbers of one to ten are very similar accross very wide geographical area confirming they are from the same language tree. Fun facts their most common word is (nyior/nyiur) which further cemented their status as the community with largest number of islands because coconut tree is trademark of their islands environment.
[1] Austronesian peoples:
https://en.wikipedia.org/wiki/Austronesian_peoples
For example, 'to be' - French 'etre' (circumflex over the e indicates old 's' after the e), Marathi 'asane' (pronounced esnay)
'to go', German gehen, Marathi jana (when conjugated the j becomes hard)
'to give', french 'donner', Hindi 'danaa' (pronounced similarly)
'to mix', french 'melanger', Hind 'melaanaa'
Other non-obvious ones:
Vedas and Wisdom / Wit. Alternatively, Latin video (to see)
Dyaus-pitar and Jupiter, Zeus-pater
'that' in English is 'que' (that/what) in french and 'kya' (for what) or 'ki' (for that) in Hindi (pronounced similarly to French 'que').
English burden or 'to bear' and Hindi bhar (burden)
English 'ignite', Latin 'ignis' and Indic 'agni' (fire)
'Raja' and 'regal' or 'royal'
'Dental' and Hindi 'dant' (tooth)
Greek 'polis' and Indic 'pore' / 'pur' / 'puram' (the 'r' is pronounced like a soft l)
This one is slightly more interesting than a mere cognate as it is believed that the Proto-Indo-European speakers worshipped a sky god with the reconstructed name *Dyḗus ph₂tḗr ("sky-father") which is the ancestor of these (also Tyr and the like on the Germanic side). See:
https://en.wikipedia.org/wiki/*Dy%C4%93us "*Dyēus is considered by scholars the most securely reconstructed deity of the Indo-European pantheon, as identical formulas referring to him can be found among the subsequent Indo-European languages and myths of the Vedic Indo-Aryans, Latins, Greeks, Phrygians, Messapians, Thracians, Illyrians, Albanians and Hittites."
Not all similarities between mondern languages are inherited, coincidences do happen.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-Eur...
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-Eur...
Pedar, Madar, Baradar, Dokhtar
When I look at the difference between modern and “old English” they seem to have changed quite a bit [0]. When I read an etymological explanation [1], it sounds like a just so story.
0. https://www.reddit.com/r/etymology/comments/9ouweu/how_engli...
1. https://www.pimsleur.com/blog/words-for-father-around-the-wo...
If you have to make up a new just-so story for every pair of words, of course you're not gaining much, but if the same story works for many words at the same time, positing a common origin isn't too far-fetched.
Dead Comment