Caste as an Institution — Part 2 – Indian Political Debate

From lock-in to first install: how India built a social operating system

Abstract

In Part 1, I argued that caste is not “eternal culture” so much as an institution that became the cheapest social technology available—reducing transaction costs in marriage, hiring, credit and ritual services—and then locking itself in through increasing returns and switching costs. (Indian Political Debate) This Part 2 extends that argument by addressing the missing piece: how the early advantage formed in the first place. Drawing on a new provocation from Tony Joseph’s Indian History Congress lecture—scheduled as a special lecture on “Homo ‘Opportunisticus’: The Contingent, Contested Evolution of Caste” at the 84th IHC session (28 Dec 2025) —I propose a continuity model: caste begins as a solution to authority and standardisation, scales as a solution to agrarian surplus extraction, and persists as a solution to coordination under scarcity, even when morally indefensible.

1. What Part 1 already established (and why it still holds)

Part 1’s core claim was straightforward: caste survives not because “Indians loved hierarchy,” but because at critical junctures it delivered coordination gains that outweighed the costs of inequality, and those gains then compounded into path dependence. (Indian Political Debate)

Two elements matter for continuity:

Institutions as technologies with increasing returns. Like QWERTY, a slightly early advantage can snowball into near-permanent lock-in when switching costs exist. (Indian Political Debate)
The tipping logic. Once a critical share of households insists on endogamy (and the social rules that guard it), others follow to avoid losing marriage partners, credit networks, and ritual services. I put it bluntly: “once a quarter of households insist… every additional household has an incentive to follow.” (Indian Political Debate)

That argument remains intact. But it leaves a legitimate question: who created the first quarter—and how?

2. The missing piece: increasing returns explain persistence, not ignition

Lock-in stories can become circular if they don’t explain the first push. In Part 1, I described the payoffs and showed how they reinforce each other; I didn’t fully specify the “founding shock” that produced the initial installed base. (Indian Political Debate)

This is where Joseph’s provocation is useful—not because it ends the debate, but because it supplies a plausible institutional trigger: standardisation of authority.

The 84th Indian History Congress programme explicitly lists a special lecture titled “Homo ‘Opportunisticus’: The Contingent, Contested Evolution of Caste,” delivered by Tony Joseph on 28 December 2025. Whatever one thinks of his conclusions, the framing points to the missing mechanism: caste is not born as ideology first; it can begin as infrastructure.

3. The first install: when a polity needed an authorised canon

Every durable hierarchy starts as a solution to a coordination problem.

One plausible early coordination problem is who is authorised to preserve, perform, and certify a shared religious/ritual archive. If a society’s highest-status knowledge is:

oral (hard to audit),
specialised (hard to replicate quickly),
and politically useful (valuable for unity and legitimation),

then the cheapest way to stabilise it is not exams or open licensing. It is closure.

A guild becomes a lineage. A skill becomes a birthright.
That is the first “install”: the moment an occupational role converts into inherited status.

In Part 1 language, this is how k (the coordination bonus) is manufactured, not assumed. Once an authorised network exists, everyone outside it interacts with it under asymmetric terms—and imitation begins.

4. The coalition that makes caste scalable: power + legitimacy

Part 1 argued that caste becomes especially durable when it fuses with agrarian state formation—because the state’s central obsession becomes predictable labour, predictable surplus, predictable order. (Indian Political Debate)

The continuity move is to see priestly authority and ruling authority as complementary assets:

Rulers want legitimacy that reduces enforcement costs.
Priestly elites want patronage and protection that converts symbolic rank into material security.

Once fused, the system can do something more powerful than classify: it can enforce. That enforcement doesn’t need to be constant violence; it can be cheaper tools—sanctions, boycotts, denial of services, social expulsion. In model terms, the coalition raises c (cost of exit) and raises R (the legitimacy multiplier) without necessarily increasing overt coercion.

5. Agrarianisation as the true accelerator: surplus needs immobility

Part 1 located the scaling event in the shift to intensive agriculture: wet-rice systems, iron tools, inheritance rules, and the labour demands of settled cultivation. (Indian Political Debate)

This is the macroeconomic engine:

Pastoral settings can have rank, but mobility is higher and surplus thinner.
Agrarian settings generate surplus—and therefore generate the political economy of extraction.

Caste becomes attractive to states because it lowers the cost of administering society. As I wrote in Part 1, the “rural tax machine found it cheaper to tag peasants by caste than to audit individual incomes.” (Indian Political Debate) That line is not merely descriptive; it is the institutional heart of the story.

6. Contestation doesn’t negate the model; it strengthens it

A common misunderstanding is that if caste was contested, it cannot have been stable. The opposite is often true: contestation is the price of building a stable equilibrium.

Part 1 already treated Buddhism as a major disruption—raising the benefit of exit (B)—yet failing to dismantle the agrarian coordination logic (k) that caste provided. (Indian Political Debate)

That remains the correct continuity frame. Moral revolutions can delegitimise hierarchy; they rarely dissolve the underlying service bundle unless they build a cheaper replacement: insurance, credit, matchmaking, occupational pipelines, and local governance.

So caste adapts, absorbs, rebrands, and survives—precisely like other exploitative equilibria in history.

7. Codification keeps returning: from scripture to census to algorithm

Part 1 tracked how identity hardens under systems that demand legibility—from colonial enumeration to modern bureaucratic categories and then into the digital age, where surnames, networks, and proxies quietly recreate screening and segregation. (Indian Political Debate)

The continuity point is simple: codification is not an aberration. It is what complex societies do when they lack high-trust universal alternatives. Empires codify for governance, markets codify for screening, communities codify for trust.

Unless the alternative is deliberately engineered, caste returns as a compression algorithm: cheap, fast, and morally catastrophic.

8. The Part 2 conclusion: annihilation requires replacement, not only rejection

If caste is an institution with increasing returns, it won’t die from sermons. It dies when its comparative advantage collapses.

Part 1 ended with policy levers that target the parameters of lock-in. (Indian Political Debate) Part 2 restates them as the only credible “uninstall” strategy:

Lower k by universalising the services caste informally supplies (health, insurance, pensions, legal aid, portable welfare).
Slash c by making exit safe: credible protection against boycott and violence, fast justice for atrocity enforcement, penalties for coercive community “courts.”
Reduce R by attacking sanctification channels—ritualised contempt and everyday segregation—rather than only abstract moralising.
Raise B for openness through opportunity that does not require identity brokerage: fair hiring, transparent admissions, mobility backed by the state, not by caste.

That is the continuity from Part 1 to Part 2:
Caste begins as a coordination solution, becomes an extraction machine, stabilises as a self-reproducing equilibrium, and persists because it still delivers convenience under scarcity. (Indian Political Debate)

If we want to end it, we must build a better social operating system—one that outperforms caste on trust, insurance, and opportunity—without inherited contempt.