Why Indian forms dominated Chinese forms in mainland Southeast Asia


On Twitter Peter Turchin had a question in response to me tweeting a new preprint on bioRxiv:


This was my impression too until a few years ago, but the genetic evidence does point to gene-flow. Here are two recent posts from me, Likely Male-Mediated Indianization In Southeast Asia and Indic Civilization Came To Southeast Asia Because Indian People Came To Southeast Asia. Lots Of Them.

Looking at the data it is clear in genome-wide analyses, and uniparental analyses, particularly Y chromosomes, that there was gene-flow from South Asia. Model-based clustering with Structure/Admixture has long yielded the result that Khmer people have some fraction, on the order of ~10%, Indian ancestry. Initially, the assumption from many researchers that I talked to was that this affinity was due to ancient South Eurasian peoples who were dominant from the Indus to Pacific ocean before the arrival of Neolithic era agriculturalists from the north and west. That is, this was common shared ancestry from the Pleistocene.

But several years ago it came to my attention that a substantial minority, on the order of ~5%, of Khmer carried West Eurasian Y chromosomes, in particular, R1a1a. In all probability, this is a lineage which is intrusive to South Asia after 4,000 years ago, and almost certainly not anciently present in Southeast Asia. Additionally, unlike Myanmar Cambodia is not adjacent to South Asia, and unlike Malaysia, Cambodia does not have colonial-era experience with Indian settlement. In the plot above look at the frequencies for R and H in particular. These are almost always going to be South Asian in a Southeast Asian context (the R is R1a1a usually).

Of course, Cambodia does have an “Indianized” past. From Wikipedia:

Funan (Chinese: 扶南; pinyin: Fúnán), (Khmer: ហ្វូណន – Fonon), (Vietnamese: Phù Nam) or Nokor Phnom (Khmer: នគរភ្នំ) was the name given by Chinese cartographers, geographers and writers to an ancient Indianised state—or, rather a loose network of states (Mandala)…—located in mainland Southeast Asia centered on the Mekong Delta that existed from the first to sixth century CE….

It is also possible that Funan was a multicultural society, including various ethnic and linguistic groups. In the late 4th and 5th centuries, Indianization advanced more rapidly, in part through renewed impulses from the south Indian Pallava dynasty and the north Indian Gupta Empire...The only extant local writings from the period of Funan are paleographic Pallava Grantha inscriptions in Sanskrit of the Pallava dynasty, a scholarly language used by learned and ruling elites throughout South and Southeast Asia. These inscriptions give no information about the ethnicity or vernacular tongue of the Funanese.

Funan may have been the Suvarnabhumi referred to in ancient Indian texts….

The Hindu nature of the classical Khmer Empire is well known to most people. After all, Angor Wat was dedicated to Vishnu by Suryavarman II. The question, as highlighted by Peter Turchin above, is the balance between cultural and demographic diffusion in the propagation of Indic culture in Southeast Asia.

In Victor Lieberman’s excellent Strange Parallels: Southeast Asia in Global Context, c.800-1830 the author observes correctly that “Indic” models of social and political organization have been much more successful in their spread to mainland Southeast Asia than “Sinic” models. Of the major polities of mainland Southeast Asia, only Vietnam is primarily Sinic in orientation, adopting Chinese Confucian bureaucratic norms, forms, and functions over the past 1,000 years. In contrast, Cambodia, Thailand and Myanmar, a more Indian set of systems predominated. What Lieberman terms “solar polities,” where the rulers took upon a sacral role and political cohesion is driven by religious enthusiasm and identity, were predominant to the west of Vietnam.

Liberman assumes that the cause of the success of the Indian model was due to its lower demands on restructuring indigenous forms of self-governance. That is, Indianized solar polities were loose structures which organized themselves atop native chiefdoms. Bureaucratic Chinese polities required a greater mobilization of the populace and the education of a broad swath of the gentry class which would replace local quasi-feudal arrangments.

The genetic data above point to a different possibility: a large number of people of Indian origin migrated to mainland Southeast Asia over 1,000 years ago, and transplanted Indian socio-cultural forms in totality. In contrast, Han Chinese migration to Southeast Asia in large numbers in the pre-modern period was limited to port cities, and, what became Vietnam (which was part of the Chinese polity before 1000 AD).

One of the central themes in Strange Parallels is that early polities in mainland Southeast Asia dealt with the shock of the migration of the Tai peoples differently. Modern-day Thailand and Laos were totally culturally swallowed by the Tai, and the political system was reoriented toward the Tai elites (though they adopted many of the religious and cultural norms of the earlier Mon polities). Vietnam rebuffed the Tai, maintaining its territorial integrity, and expanded at the expense of Cambodia. Both Myanmar and Cambodia survived the period of disruption but were strongly impacted. Cambodia lost much of its territory to what became Thailand and Vietnam and arguably survived intact only due to the French intervention. Myanmar had to accommodate large Tai minorities in its upland areas, the “Shan states.”

But one of the facts that Lieberman does not address is that the arrival of the Tai was coincident with a period of Indian influence, and myths of migrations from the west. Before 1000 AD much of Southeast Asian history is lost to us, only perceived through Chinese records of trade missions and delegations. It is not implausible that the stress and chaos that generated the new proto-states in the wake of the Tai shock resulted in the final cultural assimilation of Indian socio-cultural elites within Southeast Asia, who served as vectors for Indic civilization.

6 thoughts on “Why Indian forms dominated Chinese forms in mainland Southeast Asia

  1. Any ideas as to what region these Indians came from. Bangladesh/NE India seems like an obvious choice for Burma. South Indians have migrated to Malaysia recently (post- colonial migration with the Chinese), but Cambodia seems to be ancient. Could it have been South Indian; many Dravidian style architecture, historical ties etc. However, the presence of R1a indicates either a high caste South Indian migration perhaps with the Cholas or a North Indian mugration, also from Bangladesh which has high amounts? Could they be the same cohort that migrated to Burma, but some groups splintered off and migrated further to Cambodia etc.

  2. Fascinating. There was an additional shock to the mainland SEA landscape prior to the Mongol invasion and spread of Tai-speaking chiefdoms, and that was the Nanzhao conquests (from Yunnan) of the 9th century. The Nanzhao elite likely spoke Yi-Burmese and their domination of western and northern mainland SEA may be linked to the emergence of the Burmese language by 1000. Nanzhao’s successor in Yunnan, Dali, was also heavily influenced by Indian culture.

  3. It seems that Vietnamese DNA is similar to that of Tai with Chinese influence and without Indian influence like Thai or Laos

  4. I have to admit, I’ve always been skeptical of a purely cultural transmission with no concurrent or subsequent gene flow from India.

    But I think we certainly need a rich data series to understand how concurrent or subsequent gene flow was.

    Analogy: In Bavaria, we know from last year’s paleogenomics that in late Antiquity that people had less Southern European related ancestry than today, and that there was also a cultural shift towards Christian religion from the south. But did a pulse of migration from Southern Europe concurrently with and essentially to establishing the religion, or did religious conversion happen, and then subsequent, constant, continuous lower level migration dynamics persisting to today increase levels of Southern European ancestry? The usual model is more like the latter, but we will need a rich data series to know.

    Applied to SE Asia, did a pulse of movement establish Indian cultural forms, or were they established without much movement (really small movements that were deeply culturally influential), but their establishment supported commerce and interactions that continuous increased Indian related ancestry through later history?

    Re: Lieberman’s idea that Bureaucratic Chinese polities required a greater mobilization of the populace and the education of a broad swath of the gentry class which would replace local quasi-feudal arrangements hence lower spread, this is the kind of thing which is seems plausible at first glance and I think you can kind of understand it in terms of Chinese interactions with the steppe, where nothing much of their thought about political philosophy or religion at all could transplant outside their context. (Contra Buddhism).

    But when it comes to fairly settled agricultural polities, I do kind of think it’s probably more the case that simple distance from the core of the Chinese and Indian heartlands on trade routes probably matters a lot more. In a context where the Chinese south was far less developed than it became in the 2nd millennium AD.

    Mainly because Japan tried to copy Chinese developments in toto pretty enthusiastically in the mid-late 1st millenium AD, on a base that I’m not sure was a lot more structurally capable of doing so than SE Asian polities of the time. The main outcome was of course that they ultimately failed to do so (unlike Vietnam?), leading them onto a kind of feudal-Confucianised path instead (a “lower development” path that oddly may have allowed them to swerve foibles of Confucian bureaucratic-landlord-gentry governance that may not have supported a rapid modernisation later in history). It’s not the case that the failure to endure of the Chinese gentry-bureaucrat model meant Japan was also not profoundly influenced by Chinese culture, in a way that is much more absent in most of SE Asia.

  5. fake, only Tai yai was attacked by Chinese, Mongol then they moved to Shan and Assam. Tai noi (Thai and laos) had not been attacked by chinese and stayed at present location more than 1000 years ago. Also Zhuang stayed at south china no moving to south.

Comments are closed.