If you remember one distinction from this whole site, make it this one. Varna and jati are not the same thing, and almost every muddled claim about caste — in either direction — comes from treating them as one.
Two different words for two different things
- Varna is the four-fold theory: Brahmin, Kshatriya, Vaishya, Shudra (with the untouchables outside it). It is pan-Indian, scriptural, and tidy. There are four of them.
- Jati is the lived reality: the thousands of endogamous birth-groups people are actually born into — Reddy, Kamma, Mala, Madiga, Kummari, and so on. Jati is the unit that governs whom you marry, what work is “yours,” whom you eat with. There are thousands of them, and they are intensely local.
Varna is the map. Jati is the ground. They line up only loosely.
How the old code tried to fold jati into varna
The Dharmaśāstra was aware of the gap and papered over it with a myth of descent: the jatis, it claimed, arose from mixed varna unions. Manu spends a whole chapter deriving despised groups from cross-varna marriage — the theory of varṇa-saṃkara:
10.36–40. From a Nishāda springs… a Karāvara, who works in leather; and from a Vaidehaka… an Andhra and a Meda, who dwell outside the village.
[Documented] — Bühler’s Manu, verified against the scan. It is a fictional genealogy, but its purpose is real: to rank every actual birth-group by slotting it into the four-varna frame, and to justify endogamy by treating its violation as the source of degradation.
At the bottom, the theory turns into geography. The groups outside varna are ordered to live outside the settlement:
10.51. But the dwellings of Chaṇḍālas and Svapachas shall be outside the village… 10.52. Their dress (shall be) the garments of the dead, (they shall eat) their food from broken dishes… and they must always wander from place to place.
[Documented] Untouchability is a jati-and-outside-varna phenomenon, not a varna one — which is exactly why “Shudra” and “untouchable” must never be merged.
The outside observer saw it too
Al-Biruni, around 1030, described precisely this two-layer structure: the four varnas, then the Antyaja service guilds (“eight classes of them, who freely intermarry with each other”), then the groups “not reckoned amongst any caste.” [Documented] He was watching jati and varna operate as different layers nine centuries ago.
Why the distinction is the key to everything here
Three payoffs, and they are the reason this essay sits under the wizard:
- A “low” varna can be a dominant caste. The Reddys and Kammas are varna-Shudra (fourth) yet, as jatis, they have been landholders and rulers. Varna rank and social power came apart precisely because jati, not varna, is where power actually sits. This is the mismatch the wizard is built to show.
- Reservation runs on jati, not varna. The SC, ST and OBC lists are lists of jatis (and outside-varna groups), not of varnas. That is why a Shudra-varna jati can be either “forward” or “OBC” depending on the specific community — the new code sorts by jati-based category, not by the old four-fold rank.
- Jatis move; varna is the fixed idiom. Sociologists (M. N. Srinivas’s “Sanskritization”) documented jatis raising their status over generations by adopting higher-caste practice — and the colonial census then froze these fluid jatis into hard, listed categories (Bayly, Dirks). The theory stayed rigid; the reality was always shifting underneath it. [Debated] on how much movement there really was.
Keep varna and jati apart and the confusions dissolve: caste-by-birth is jati; the four “colours” are varna; untouchability is outside both; and the reason the wizard shows you two axes — old-code varna and new-code category — is that the old order’s neat theory never did match the ground it ruled.