Wayne+Radinsky+%28new+account%29

2025-02-20 03:41:51

Using DNA to trace the origins of the Indo-European language family, the language family that includes English, Russian, Greek, Bengali, and "dozens more" (Spanish, French, Portuguese, German, Persian, Hindi, Marathi, Punjabi, Gujarati, etc).

Basically the article says, 6,400 years ago, people from south and north of grassy plains north of the Black Sea (a place they call "Caucasus-Lower Volga cline") came together and invented a language together, which didn't have a name but we today call it "Proto-Indo-European". Around 5,300 years ago, some of these people migrated to Yamnaya, which today is part of eastern Ukraine and southern Russia (and not a happy place as there is a war going on there -- a fact that apparently made this research hard to do, according to the article). Yamnaya is the part of this region that is a steppe, which is to say, flat glassland. The people of the Yamnaya began herding cattle and sheep with wheels and wagons and experienced explosive population growth, resulting in migration outward in all directions, bringing their language with them.

The one thing that makes me a bit skeptical of this story is how, in modern times, we see people learn languages from other groups of people not genetically related to them all the time. For example, the population of English speakers is something like 1.5 billion, but the population of the Island we call "England", for which the language is named, is only 61 million. (Technically the island is called "Great Britain" and has multiple countries, none of which are called "England": there's the UK, Scotland, and Wales -- but never mind that, the point is, most people who learned English didn't learn it because they were born into an English speaking group, one that migrated or otherwise.) It seems like the assumption here is that the migrating people took their language with them and didn't learn the languages of the places they migrated to or vice-versa. How reliable is that assumption?

Another interesting bit in the article is, it also says the Hittite language was the first Indo-European language to be written down (on clay and bronze tablets). The Hittites lived in part of what is today Turkey. That was much more recent that the other events described in the article -- 1650 BCE vs 4400 BCE (for the Caucasus-Lower Volga cline people -- 2,750 years earlier) and 3300 BCE (for the Yamnaya people -- 1,650 years earlier).

The long-term trend seems to be languages going extinct and the world consolidating on a small number of "major" languages. Does modern machine translation and the advent of large language models (LLMs) change or accelerate this trend?

Who first spoke Indo-European? DNA points to Eurasian herders 6,400 years ago

#solidstatelife #genetics