Early Hinduism – the epic stratification

This essay is highly speculative in nature and I have many doubts about many of the things stated below, but I have tried to coherently bring together distinct threads of early Indian history into an explanation for the great stratification of Jati-Varna

Ancient history is in general a tricky subject to delve into, but when it comes to ancient Indian history, the tricky becomes almost entirely speculative. The entire narrative is based on a series of texts, from the Vedic canon to Pali texts – none of them are dated precisely in absolute terms. The paucity of inscriptions from ancient India makes dating much more difficult as oral texts are much harder to accurately date.

Ms Sarah Welch, CC0, via Wikimedia Commons
Ashoka’s major rock edict – the earliest inscription from Ancient India.

One of the early inscriptions from Ancient India, Ashoka’s 13th major rock edict from Kandahar reads

Except among the Greeks, there is no land where the religious orders of Brahmanas and Sramanas are not to be found, and there is no land anywhere where men do not support one sect or another.

Here Brahmanas are mentioned but not as a Varna per se but as analogous to various Sramanas (priests / philosophers). Some academics have come to regard the Sramana traditions as somewhat antagonistic to the orthodox Brahmanical traditions. However, the earliest written reference to these traditions, Ashoka’s Rock Edicts mention them always together. Patanjali and others too mostly mention them together and never as quite as antagonistic as later Sramana canon or modern scholarship would have us believe.

However, it is undeniable that one cannot be understood without the challenges presented by the other. Ahimsa and Vegetarianism are generally acknowledged (and contested) to be Sramana influences on Classical Hinduism. A lot of digital and literal ink has been spilled to answer the question of how these two currents have interacted and shaped each other – mostly through the lens of Ahimsa, Dha(r/m)ma, Moksha and only rarely Karma.

Vedic deities Indra and Surya at the entrance of Buddhist cave in Bhaja (200 BCE – 100 CE)

Johannes Bronkhorst’s Greater Magadha thesis offers a tenuous but interesting take on these interactions. The basic premise of the thesis is that the region of Greater Magadha was home to the Proto-Sramana traditions while the Kuru-Panchala region to the Vedic Brahmanas and that many ideas central to classical Hinduism like Karma, Rebirth, and Ayurveda came into it from the Sramana traditions of the Greater Magadha via the esoteric Upanishads (especially the ones which were composed in the horizon of Greater Magadha). The whole thesis rests on the revised chronology which only makes sense if the thesis is true – so I doubt the book is going to convince anyone. But it has catalyzed a rudimentary and dormant theory that came to my mind years ago while reading Ambedkar’s writings.

The composition of the Manava Dharma Shastra (100 BCE to 200 AD) is generally considered to be an indication (or instrument) of Varna ossification. The Varna system in some form ought to have existed (especially in the Gangetic heartland) since the late Vedic period (Purusha Sukta), yet both textual and genetic evidence points to this period as being one of great mixing. Hence it is fair to assume that whatever rudimentary Varna system existed, it was not very rigidly followed in these times. Also its important to note that traditional Varna system may have never been a reality south of the Vindhyas.

It is difficult to pin the Varna ossification to any particular political period. The only pan India ancient empire – the Mauryas are unlikely to have imposed any Varna hierarchy on their subjects as the pedagogic Ashoka doesn’t once mention Varna in his Rock Edicts. The Shungas are seen as the Brahmanical pushback against excesses of the Mauryan state but their power was both too limited in time and too restricted in region to have made any major impact. The same is true for most other political powers in the country for the next 500 years.

Brahmins had begun moving out of the Gangetic heartland as early as the late Vedic period itself. So why did the Varna system, suddenly begin to ossify  centuries later? Surely some metaphysical, philosophical, and/or political explanation is required to make sense of this phenomenon. Also, Jati endogamy which is the true hallmark of the Indian Caste system cannot be explained by the Brahmanical Varna system – even the rigid one prescribed in the Manusmriti. The answers may lie in a core philosophy of the Indic faith systems.

So what is the common characteristic that defines Indian religious thought? The answer is easy – the concept of Karma, Rebirth and Dharma. Even if we reject the thesis of Greater Magadha, we have to accept that the concepts of Rebirth and Karma are explored in far more detail in the Sramana schools – namely Buddhists, Jainas, and Ajivikas. The whole philosophical aim of the Sramana schools is to avoid Bad Karma to primarily get Good Karma and finally Moksha. This is in clear contrast with mainstream Vedic thought. Though the early Upanishads (Chandogya and Brihadaranyaka) touch the Karma doctrine it’s in no way as critically dissected as by the Sramanas. The lengths to which the Jainas and Ajivikas go to avoid all Karma; the detailed linking of the intention of the “Actor” to the Karma done by Buddha illustrate that the Sramanas, in general, were way more focused on Karma than their Brahmana counterparts. More importantly, the concept of Karmic retribution in Rebirth is much more detailed in early Sramana traditions than the Upanishads (Yajnavalkya doesn’t link Karma directly to Rebirth but discusses both separately). So it remains fair to assume that even if doctrines of Rebirth and Karma didn’t come into Classical Hinduism as an import from Sramana traditions, it can surely be thought that the Sramana innovations in the Karma and Rebirth doctrines challenged the more “this-worldly outlook” of the Vedic Brahmanas.

But how does this matter to the Jati Varna matrix? The initial conception of Varna sees it as a natural order of things (not unlike stratification seen in most ancient societies). Moreover, this conception is in no way rejected by the Sramana traditions even Buddhism – thought Buddha did not give Varna the emphasis it received from the Vedic Brahmanas.  Even today caste is practiced in the Jainas. So how did the conception of Karmic retribution affect this system? The answer seems obvious enough. It meant that the position of one in the Varna hierarchy could be justified as the fruit of Karma of previous births and not only as a Natural order. In other words, the ritual status was awarded to certain births for their good Karma and vice-versa. In many ways, Karmic retribution is a fundamental shift from the “this-worldly” ways of the composers of Rigveda.

This change is captured in the Bhagavat Gita, arguably the most important book of the Hindu canon. While there continue to be many interpretations of the doctrine of Karma espoused in the Gita, the one reading tells us to fulfill the Dharma (of your Varna/ Position/ Situation) with the implication that it would result in Good Karma and better Rebirths – the ultimate aim of Moksha notwithstanding. That indeed seems to be one of the simplistic messages of the Gita which would have begun spreading in the society with the final versions of Mahabharata. The prescriptive Manusmriti is one thing, but the bonafide revelation of Gita is another (though it is not my point that Karmic retribution is the core message of Gita but it is hard to argue against it being a vehicle of the spread of these memes). This doesn’t mean that Varna became birth-based at this moment in history – it is fair to assume it always was at least partially birth-based though more flexible. But we can state that at this stage, one’s Birth became Karma-based and Varna also became inextricably linked to Karma. 

This could have resulted in two primary effects:

  • It would mitigate the sense of injustice perceived by sections of the society who had it tough. The injustice of birth was not injustice but the karmic justice of previous births.
  • It associated “ritual Varna hierarchy and division of labor” with moral dimension (Karma of previous birth). Potentially this moral dimension would buttress the existing Varna hierarchy.

It’s easy to imagine how this would in turn result in decreasing porousness between Varnas. Incidentally, this is attested through the first/second-century inscription near Nasik by Brahmana Satavahana Queen Gotami, which praises how her son prevented the mixing of the Varnas. This is one of the most solidly dated references against the mixing of Varnas (as it is an inscription) issued by a political authority (not just religious abstractions).

However in a pre-modern subcontinent without a strong centralized state, these ideas would have spread very slowly through the network of Brahmins and various (particularly) Vaishnava sects through the vehicle of Gita. The Hindu Golden age of the Imperial Vaishnavite Guptas – who ruled the second-largest and arguably the richest empire of ancient India, in the fourth/fifth century AD nicely correlates with these timelines. Thus we could say that by the time of the Huna invasions of the 5th and 6th century the Varna ossification was prevalent, but even that doesn’t explain the complete story. Still, we have no philosophical or scriptural basis for Jati endogamy.

Irawati Karve – a pioneering Indian Anthropologist / Indologist

Anthropologist Irawati Karve in her book “Hindu society” was one of the earliest to claim that the Jati system was a pre-Aryan reality upon which the abstraction of the Aryan Varna system was imposed. Academically her work has been contested and not accepted in mainstream Indology, but her case is very compelling, given that it is based on her immense fieldwork in “Non-Aryan” tribes who have maintain very strict endogamy. But how does her thesis map onto what we know from genetics? Endogamy in India roughly seems to have ossified between 0 AD and 500 AD but who is to say that less rigid endogamy (not detectable) was not the norm earlier? Is it possible that the self-conception of Jatis is indeed is an ancient Pre-Aryan reality that was less rigid during the Vedic times? Clearly, there are no easy answers as all we can do is speculate and wait for Ancient DNA from India to show if there existed any pre-Aryan structure in the populations of the Indus valley.

Many tribal (hunter-gatherer) societies have endogamy baked into their cultures. But generally, as these tribal societies get integrated into the agricultural societies, this endogamy tends to break down – as evident for recent genetic findings (particularly Europe). But what if the tribal societies which integrated into the emerging Urban civilizations (first the Indus and then the Ganga) , never fully gave up their tribal/clan identities? The hundreds of excavated IVC villages point to sophisticated trade/occupational specialization. If both the sexes work in their ancestral trades per se, it would naturally result in tribal endogamy as it makes occupational sense. But that would not necessarily lead to rigid endogamy to the levels we see in the subcontinent- probably because this doesn’t exist elsewhere in the world. Though the identities of groups by kinship (precursor to Jati) may have existed even before the Varna system began to take form (let alone become rigid).

But why does this Jati endogamy become sharper with the ossification of the Varna system? Some take the explanation as Jatis arising out of the mixing of Varnas seriously, but that thesis (ludicrous imo) can be jettisoned without a second thought as Jatis exists even in those who are outside the Varna hierarchies. A potential answer may again lie in the doctrine of Karmic retribution.

Unlike the original simplistic Varna hierarchy – the concept of Karmic retribution enables hierarchies within hierarchies. Every Jati can be ranked within the Varna hierarchy based on the perceived moral inheritance (Karma) of their profession. Additionally, better births and even salvation are promised to the ones following their Jati-Varna Dharma. Thus Jatis would have both religious as well as occupational/cultural reasons for enforcing stricter endogamy which is far more believable than assuming these norms were somehow imposed across the subcontinent in pre-modern times by machinations of Dvija Varnas.

None of the above points are sufficient but all are necessary to explain the great vivisection of Indian society. Chronologically first the kinship-based (not gotra) groups were integrated into the expanding Aryavarta both culturally and genetically while the late Vedic abstractions of Varna and ritual purity began to take root in the orthodox Vedic traditions. When the rudimentary conceptions (Vedic or non-Vedic) of Karma and Rebirth were taken up by the Sramanas, taking them to a complex, philosophical, and rigorous extreme, they began to affect the Vedic philosophies.

In essence, the religious innovations of Karma, Rebirth and Dharma when coupled with pre-existing concepts of Varna, ritual purity, and tribal occupational endogamy conjure up a perfect storm that continues to flow through the blood of around 1/4th of humanity, in form of thousands of distinct streams. 

Post Script: 

  • What is not discussed above is the impact on the subcontinent of the violent Huna invasions which along with internal strife resulted in the collapse of the Gupta empire. The rapid de-urbanization which is speculated to have occurred in the fifth and sixth centuries would have also played a crucial role in this ossification. The second millennium with the Turkic invasions would have also played some role in the maintenance of this now-steady state.
  • I continue to have a lot of doubts about the above speculations, but when I read books on Indology and Indian prehistory, I find even more tenuous speculations (made by professional academics) than the ones I have proceeded to make in this essay. At least these speculations seem to align with the history alluded by the genetic data of caste (Or I have made them align).
  • I had thought along these lines even before reading about the interactions of Brahmanas and Sramanas but while reading the Greater Magadha thesis and following a YouTube seminar I thought the thrust of my current argument was staring me in the eye. I expected someone to draw the conclusions I had drawn, but was extremely surprised than no one has gone there.
  • The references for this essay are numerous and diverse to be noted here. Anyone interested please reach out to me.


The Sintashta horses!

The origins and spread of domestic horses from the Western Eurasian steppes:

Domestication of horses fundamentally transformed long-range mobility and warfare1. However, modern domesticated breeds do not descend from the earliest domestic horse lineage associated with archaeological evidence of bridling, milking and corralling…at Botai, Central Asia around 3500 BC3. Other longstanding candidate regions for horse domestication, such as Iberia5 and Anatolia6, have also recently been challenged. Thus, the genetic, geographic and temporal origins of modern domestic horses have remained unknown. Here we pinpoint the Western Eurasian steppes, especially the lower Volga-Don region, as the homeland of modern domestic horses. Furthermore, we map the population changes accompanying domestication from 273 ancient horse genomes. This reveals that modern domestic horses ultimately replaced almost all other local populations as they expanded rapidly across Eurasia from about 2000 BC, synchronously with equestrian material culture, including Sintashta spoke-wheeled chariots. We find that equestrianism involved strong selection for critical locomotor and behavioural adaptations at the GSDMC and ZFPM1 genes. Our results reject the commonly held association7 between horseback riding and the massive expansion of Yamnaya steppe pastoralists into Europe around 3000 BC driving the spread of Indo-European languages. This contrasts with the scenario in Asia where Indo-Iranian languages, chariots and horses spread together, following the early second millennium BC Sintashta culture.

The paper is open access. Basically the Sintashta seem to have triggered the equine revolution across Eurasia.


I’m spending a lot of time reading about the Corded Ware for my series on the steppe. The Corded Ware is a culture that appeared that abruptly in Northern Europe between 2900 and 2800 BC, covering a vast territory of Central and Eastern Europe in a century. The name derives from the unique marks left on their pottery.

For decades scholars have argued whether they were an indigenous development out of the farmers who occupied this region for thousands of years, or whether there was a mass migration out of the steppe. More realistically, there was a synthetic position between at some point. Perhaps the farmers were influenced by a few elite bands migrating out of the steppe?

Today due to ancient DNA we know more. The Corded Ware culture in its mature phase is about 70% Yamnaya and 30% farmer. The farmer’s ancestry almost certainly comes exclusively through women. The Y chromosomes of the farmers were G2. There is very little of that within a few generations. It is almost all R1a.

But that leaves us with the question: where did the new pots come from? The answer is straightforward: the men from the steppe took wives from the farmers. They killed their fathers and brothers and took them to their homesteads to bear them children. These women knew how to make pots because they did not come from nomadic backgrounds. They adapted their techniques to making pots that exhibited marks that made them resemble the baskets that their husbands brought in their wagons.

The miracle of the immediate emergence of a new pottery technique is due to the fact that the nomads didn’t learn to work in clay. Their wives already had the skill.

Why is this posted on this weblog? That’s a question that does have an answer…

Koneswaran, Tenavaram Temples and God Upulvan

This posted started off as a reply to Siddarth’s comment and his visit to two temples in Sri Lanka.

First, three old Sinhala words out of many. Gokkana, DeviNuwara, Upulvan

Gokkana. The old Sinhala name for Trincomalee. Trincomallee is Derived from the recent name, Tri-Kona-Malaya. Malaya (=Malai in Tamil) is an old word used for hill/mountain. eg the Hill country is Malaya Desaya and first ref to that term that I know around 1BC. I suspect that word is from Pali. The common sinhalese word is Kanda for mountain. eg Kanda-Uda-Rata-Nuwara Hill-Top-Country-City. Now in English Kandy and in Sinhala Nuwara. Is Kanda Tamil or really old Sinhala? Who knows,

Dondara: Devapura and Devanagara. In Sinhalese it has been referred to as Devundara and Devinuwara, meaning City of Gods and Devundara is Southern most point in Sri Lanka
Until the late 16th century a historic temple port town complex housed merchants from around Asia. . A multi-religious site, its primary deity was the Buddhist god Upulvan and at its zenith was one of the most celebrated religious sites of the island, containing a thousand statues of the various sects of Hinduism and Buddhism.

Upulvan*  Upul=Dark Blue Van=Vanna (pali)=Varna. Sinhala: උපුල්වන් ‍දෙවියෝ, Pali: Uppalavanna; Sanskrit: Utpalvarna. Now used to mean Vishnu. First mentioned in Mahavamsa as Vijaya (500 BC) being blessed by god Upulvan (see foot note 2) on arrival in Sri Lanka. Upulvan is considered the god whom the Buddha entrusted with the guardianship of Sri Lanka and Buddhism in the country. At the end of the 15th century, god Upulvan was identified with god Vishnu of Hinduism. Thereafter images of Upulvan as Vishnu were set beside the images of Buddha

So now to the Temples Tenavaram in Dondra/Devundara and Koneswaram in Trinco/Gokkana

Koneswaram in Gokkana
Apparently King Mahasen (277-304 AD) destroyed the Hindu Temple.  According to the Mahavamsa to quote
and founded three viharas, destroying temples of the (brahmanical) gods: the Gokanna (vihara), (and another vihara) in Erakavilla, (and a third) in the village of the Brahman Kalanda;13

King Mahasen, initially a Mahayanist and later went back to Theravada. Built 16 reservoirs including Minneriya. Minneriya reservoir is now the site of the largest gathering of Elephants. Mahasen was regarded as a god or deity, and was called Minneri Deviyo (God of Minneriya).
In that same chapter “At the place of the yakkha Kalavela14 he built a thupa”, which reiterates the claim the Yakkas were real people.

The deep ocean where Portuguese are reputed to topple the Temple, there are Buddhist and Hindu ruins
Just a few years later in 1956, the famous Arthur C Clarke uncovered underwater masonry, architectural and idol images of the original temple. In his 1957 book ‘the Reefs of Taprobane’, Clarke identified at least 3 Hindu temples as having been build on or around Swami rock over the millennia. At that time, he said they were probably the most photographed underwater ruins in the world.
Arthur C Clarke description of first sights underwater.
Recent Dive clip


Tenavaram in Dondra/Devundara/DeviNuwara.
DeviNuwara is dotted with many temples, Buddhist and Hindu. Pothgul and Galge are the lesser known but ancient buildings.
Devundara is multi-religious complex, the Buddhist temple and the Upulvan devale (shrine) was started by King Dappula II in the 7th century AD.

*I suspect a lot of merging of pre Buddhist gods with Hindu Gods. Much like the dynamics in the Caribbean African Gods being merged with Catholic Saints.  See Link below for synopsis.

To the victors go the glory!

I have written an introductory post (it’s free), Entering Steppelandia: pop. 7.7 billion, to a series of posts (mostly paid) that I will write about the Eurasian steppe. So I’m thinking and reading a lot about this topic. This is relevant to “Brown Pundits” because we subcontinental people have been stamped by the steppe.

First, there were the Indo-Aryans. About 15% of the ancestry of modern South Asians comes from these people (averaged across region and caste). Then there were the Iron Age Iranian pastoralists, Scythians, and assorted other related groups. There is no strong evidence right now of a major genetic impact, but I think the statistical power is not such that I can definitively ignore this possibility.

Finally, there are the Muslims. They had the least impact. But they are most reviled. Why?

They lost.  The lesson is to not lose. The rest is commentary.

The Mughals!

The Brown Pundits Clubhouse channel hosted a discussion on “the Mughals” yesterday that went on for a while. There seem to be two polarized extreme views

1) The Mughals were great Indians! Long live the Mughals.

2) The Mughals were genocidal colonizers and induced inter-generational trauma.

Most people occupy a position in the middle. As for myself, I think it is clear that the Mughals were to some extent an alien and occupying influence because that is how they viewed themselves more, or less. They were Turanian Muslims of Turco-Mongol provenance. No matter how much Rajput or Persian blood they had, their paternal lineage came down from the Turk Timur. The maternal lineage of Babur was Genghiside. If India had been mostly Islamicized this would have changed. But it wasn’t. Despite the deep cultural synthesis between Mughal culture and that of India and their indigenization of the generations, there remained a connection between ashraf Muslims and Persia and Central Asia. They were not equivalent to Muslim Bengali peasants or Ismaili traders in Gujurat.

And yet the flip side of this is that the Mughals, and Muslims as a whole, in particular Turks, drove change within Indian society. To some extent, the native reaction and response in the dialectical synthesis can only be understood in the light of the Islamic shock. More generally, an Islamicate civilization evolved that extended beyond the Mughals and included the Rajputs and Marathas (reciprocally, the Mughals internalized many Rajput values, but this is to be expected due to their long residence in India and intermarriage with Rajputs).

Those Hindus who are traumatized by the impact of Islam are free to feel this way, but I am genuinely curious about an Indian culture stripped away of Islamic influence. What would that look like? Perhaps Odisha and Sri Lanka might come close?

More generally, the excited and emotional response of both Hindus and Muslims and their inability to engage in epoché makes me think that the prospects for deeper analysis are poor. Emotion has reason by the leash.

On the historicity of the Vedas

BPer Mukunda and I were having a discussion on Twitter, which I want to elevate and push to the blog, because it’s somewhat important.

When I was young (20th century) I read stuff about how the Indo-Aryans described the natives of the subcontinent as dark and “snub-nosed.” That their arrival in some ways was a meeting of two different races.

In the 2000’s I read other books and works that suggested that actually, these descriptions were metaphorical. Terms like “dark” in other words reflect an ideological or tribal conflict, with the descriptions pointing to tropes that signal which side is evil and which side is good. This is not a crazy view. The anthropology is clear that a certain level of fictitious dehumanization occurs with inter-group conflict.

So I accepted this view and moved on with my life.

But in the 2010’s things changed. I am now convinced that 3,000-4,000 years ago a people who resembled what we would term “white” expanded within the Indian subcontinent. If modern Armenians are white, then the Indo-Aryans were white. At least initially. In the subcontinent, they met a variety of people. Some of them, such as in Sindh, were of brownish complexion. Others, to the south and east, would have been considerably darker. I also assume that the Vedas were constructed in situ in the Indian subcontinent. That is, they reflect a milieu of people who were encountering the northwest of the subcontinent, and had recently traversed through BMAC (Indra may actually be a BMAC diety).

What’s the upshot here? I know think that the metaphorical view of the physical descriptions should be set next to the literal view. The reality is probably a mix. But the fact is that groups with very different physical appearances did interact in ancient India. The Aryans were almost certainly very light-skinned, with “sharp features”, in comparison to many of the people they encountered. Though one can construct hybrid scenarios, where Indo-Aryan enemies were described in inaccurate ways precisely because those tropes were associated with tribes and peoples the Indo-Aryans had conquered.

Someone who has deep knowledge of the Vedas in Sanskrit and genetics needs to look into this. That’s obviously not me.