Humans have been migrating all over the place for thousands of years, yet there are still distinct geographic differences in human phenotypes, and now visible with molecular technology, genotypes. This allows us to look back in time to infer large migrations that happened long before the historical periods, like late and post-Roman times where many migrations in and around Europe are documented, if scantily.
 |
| Out of Africa. A process that started about 70,000 years ago. |
India is a particularly interesting subject for this kind of analysis, because it has been a crossroads for tens of thousands of years, first for the migration out of Africa that led to the peopling of Australia, and then other influxes from all directions, most recently from the north, with the Indo-Europeans and then the Arab invasion. Secondly, India has also had an unusual degree of stasis since these migrations, embodied in its caste system, which may have frozen some of these genetic signals in static communities.
A recent paper continues a body of work that looks at these issues, and concludes that there are four different population signatures detectable in the mainland Indian population, and that the caste system has been in genetic terms a relatively recent development- in the last thousand years or so.
They sample hundreds of thousands of variable genetic markers (snps) from 367 people of 18 recognized ethnic groups of India. This data is put through a traditional statistical analysis of related-ness to come up with 69 sub-populations, and four principal components: those differences in the data that most efficiently explain the most differentiation into the least number of large groups. The main graph is below:
The green elipse identifies Indo-European populations, such as Brahmins from Gujarat and West Bengal, Khatri, and Maratha. The red elipse identifies Dravidian-speaking tribal populations. The turquise elipse identifies South-East Asian-influenced populations, such as Korwa, Birhor and Gond which are a Dravidian mixture. Lastly, the blue elipse identifies Tibeto-Burman originating populations such as Jamatia, Tripuri, and Manipuri, living in the northeast.
None of this is very surprising, given the clear ethnic diversity and the local neighborhood of India. More interesting are their reflections of the stability of these groups. There is very low mixture, though the sampling was modest, with an average of 20 people per ethnic group and 5 per inferred sub-population. The hypothesis, drawn from literary and historical sources, is that there was originally substantial mixing between the North and South Indian populations, but that this ceased with the gradual establishment of the caste system. The researchers use haplotypes to track how much mixture there has been, and how long it has been going on, or been in abeyance. (Though there are other views.)
"We estimated that all upper-caste populations, except MPB from Northeast India, started to practice endogamy about 70 generations ago. The length distributions of the AAA blocks and the ASI blocks within any one of these populations (GBR, WBR, IYR) were very similar. The most parsimonious explanation of this is that the practice of gene flow between ancestries in India came to an abrupt end about 1,575 y ago (assuming 22.5 y to a generation). This time estimate belongs to the latter half of the period when the Gupta emperors ruled large tracts of India (Gupta Empire, 319–550 CE)."
Thus the golden age of India, which happened during this time, seems to correspond with a hardening of the social order. What effect the latter had on the former, or the decline of the former, is perhaps food for thought.
- Our institutions are shabby enough already.
- Republicans without a clue.
- Haidt on group affiliation and the greatest generation.
- One of whom was Jimmy Carter.
- A sad day.
- De-industrialization has been devastating.
- If heartland voters thought Trump will serve their interests, they have another think coming. And if they thought Hillary was corrupt ...
- Some problems with Democrats.
