Brahminhood Is Not Doghood: How a Buddhist Philosopher Dismantled the Metaphysics of Caste
Dharmakirti does not deny that dogs share a universal, "doghood" — he refuses to grant that "Brahminhood" exists, because Brahmins and Shudras differ in no perceptible way in body, intellect or capacity, and caste produces no causal efficacy at all.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
Caste has two layers: four varnas plus countless jatis
Nilanjan Das first unpacks the structure of caste: one layer is the four occupational ranks — Brahmin, Kshatriya, Vaishya, Shudra — corresponding to different occupations; the other is the endogamous social groups embedded within them, the jatis. The former are called varnas, the latter jatis. The key point is that it is simultaneously a hierarchy of social power and political power, with Brahmins and Kshatriyas at the top and the rest below. And whichever rank you are born into is the one you stay in: caste is transmitted through lineage, not through any other factor. Das also concedes that the history of the caste system's own origins is unclear, and that it appeared roughly alongside the Brahmanical forms of the mid-first millennium BCE.
— Nilanjan DasThe opponent's analogy: Brahminhood is like doghood
Dharmakirti's main opponent, Kumarila, holds a kind of universal realism: when we watch two dachshunds running in a field, besides seeing particular features like brown color and wagging tails, we also see the shared property of "doghood." Properties of this kind, if taken to exist independently of individual objects, are what philosophers of the time called universals, and what they themselves called samanyas or jatis. Kumarila's inference is: just as dogs share doghood, Brahmins share "Brahminhood," and — this is the strangest point, Das says — we can actually see a Brahmin's Brahminhood. Das concedes this looks odd at first glance: the distinction between dogs and snakes is natural, but saying all Brahmins share a property that non-Brahmins lack is something else entirely.
— Nilanjan DasSeeing Brahminhood is not the same as recognizing a Brahmin
Here there is a key distinction: non-epistemic seeing versus epistemic seeing. The former is seeing something without seeing it as anything; the latter is seeing it as something. Das gives an example: if I had never seen a dog, I might see a dog and its doghood yet be unable to identify it as a dog — identification requires the background capacity to classify and categorize. Kumarila would say the same of Brahmins: you might see a Brahmin together with his Brahminhood yet be unable to see him as a Brahmin. So even if you put a group of Brahmins and non-Brahmins into the same identity line-up, wearing the same clothes and with the same haircut, Das says Kumarila would probably hold that he could still pick out who is a Brahmin — but there is an important qualification here.
— Nilanjan DasDharmakirti's test: structure, properties, causal power
What Dharmakirti himself says is actually quite basic: we can grasp a natural distinction between objects only under certain conditions — when the objects differ in physical structure, or differ in some important property, or differ in causal power. He cites cows and horses: a cow's shape and structure differ from a horse's, a cow's milk tastes different from a horse's, and a cow can do things a horse cannot, so we can posit a natural distinction between cows and horses. But on Dharmakirti's view, no such distinction exists between Brahmins and other caste ranks such as Shudras: Brahmins are no different from other castes in body, intellect or capacity. So there is no reason to think we can grasp a natural distinction between castes. Das stresses Dharmakirti's general principle: if you want to posit that something exists, it has to "earn its place," by proving that it has causal efficacy and can produce effects in the world.
— Nilanjan DasCaste is like an occupational surname, not like witchcraft
Asked whether Dharmakirti is saying caste is a social construction, Das gives a careful answer. Dharmakirti does hold that many of the things we think and talk about — tables, chairs, properties like brown — are constructed through our conceptual capacities, yet they are still true in another sense, because without granting them, everyday activity could not proceed. Dharmakirti thinks caste terms like Brahmin and Shudra are much like occupational terms: you call someone a Brahmin either because he performs activities such as sacrifice, or because he was born into a family that performs such activities; by Dharmakirti's time, terms like physician and merchant had already become surnames, marking a family's traditional occupation. Das then distinguishes two kinds of "social construction": one is like the category of witch — socially constructed and nonexistent, because it presupposes that people have magical powers; the other is like money — socially constructed yet still objective in some sense, since the distinction between the dollar and the pound does not track any natural distinction between two pieces of paper. Das thinks Dharmakirti's position may be compatible with the latter.
— Nilanjan DasCaste-based pride is a cognitive failure
Das says this is the strongest part of Dharmakirti's critique. At the end of the first chapter of the Pramanavarttika, Dharmakirti lists five marks of delusion, one of which is pride based on belief in caste. Dharmakirti's commentators explain: on Dharmakirti's own account of caste, this pride is a cognitive failure, a rational failure — a failure to respond to the available evidence. They point out that caste pride may rest on two grounds: first, that caste tracks some natural distinction, which would explain why Brahmins are superior; second, that the Brahmin body is ritually purer than other bodies. The commentators say that given Dharmakirti's argument that there is no material or intellectual difference between Brahmins and other caste ranks, both grounds fail, so this pride is entirely unreasonable. Das also cautions that the function of caste in South Asian society has kept changing since Dharmakirti's time, but the hierarchical structure and the attitudes accompanying it are the constant aspects.
— Nilanjan DasThe conclusion is not just fluidity but abolition
The host presses: if caste has no universal, should the concept of caste be far more fluid than it is? Das agrees, and gives the example that as far as he knows South Asians are still hesitant about inter-caste marriage, which on Dharmakirti's grounds is irrational. But Das says more important is the contemporary Buddhist movement, especially the movement inspired by B. R. Ambedkar, which holds not merely that caste should be more fluid but that it should have no place in our lives. Das thinks Dharmakirti's critique lays the groundwork for this eliminativism. Here the two sides of the disagreement emerge: on Kumarila's side, caste tracks a natural distinction and can be perceived; on Dharmakirti's side, it has no causal efficacy and does not deserve to be posited as existing.
— Nilanjan DasHe himself ran into the wall of caste
Das answers the question about personal motivation, giving two reasons. First, he was already interested in the problem of universals, which came from his other interests in metaphysics and epistemology. Second, when he studied Sanskrit texts, he repeatedly hit obstacles because of his own caste background: as a student he could not study Mimamsa and Vedanta texts, because such study requires access to the Vedas, and he is not a Brahmin. He says he comes from a caste group that in Dharmakirti's time would have been treated as Shudras. Das says caste was a reality he was forced to contend with in his younger years, and now much less so. This passage brings the metaphysical argument back down to a concrete situation.
— Nilanjan DasIn their own words · checked verbatim
According to Dharmakirti, Brahmins don't differ either in their bodies or in their intellect or in their capacities from other castes. So there is no reason according to Dharmakirti for us to think that we can grasp this natural distinction between the different castes.
Nilanjan Das10:12
For Dharmakirti more generally, if you want to posit something out there in the world that that thing has to earn its keep.
Nilanjan Das10:12
There's a different notion of socially constructed which I'm more interested in, which is that something can be socially constructed and still have a certain kind of objectivity.
Nilanjan Das13:16
he lists five signs of stupidity. And one of them is arrogance based in a belief in caste.
Nilanjan Das15:17
we can see that this kind of arrogance is an epistemic failure. It's a failure of rationality. Like not responding to the evidence that is available to them.
Nilanjan Das15:17
It's not just there should be more fluid, it should have no place in our lives. And I think Dharmakirti's critique lays the foundations for that kind of eliminativism.
Nilanjan Das16:18
I could not, for example, as a student study Mimamsa and Vedanta texts precisely because they involve engaging in the Veda because I'm not a Brahmin.
Nilanjan Das17:18
Figures
| Dharmakirti's dates | 6th to 7th century CE | 2:05 |
| Length of time Dharmakirti was continuously discussed after his death | about five hundred years | 2:05 |
| Approximate emergence of the caste system | mid-first millennium BCE | 1:05 |
| Number of marks of delusion listed by Dharmakirti | five | 15:17 |
Glossary
- varnas
- The four occupational ranks: Brahmin, Kshatriya, Vaishya, Shudra.
- jatis
- Endogamous social groups, the caste units embedded beneath the varnas.
- universals
- Properties shared by multiple individual things, taken to exist independently of the individuals.
- epistemic seeing
- Seeing something as something, which requires the background capacity to classify and categorize.
- eliminativism
- The view that caste should not merely be more fluid but should have no place in our lives.
How to listen
People interested in Indian philosophy and the metaphysical problem of universals; people who want to understand how the caste system can be dismantled by philosophical argument rather than moral condemnation.
The closing section on interactions between the Indian and Anglo-American philosophical traditions is fairly general and can be skipped.