English only · Odia translation in progress

Chapter 1: The Other Country

On August 19, 2013, in a clearing in the Niyamgiri hills of Rayagada district, a group of Dongria Kondh men and women sat under a temporary shamiana and voted. There were no electronic voting machines. No party symbols. No candidates. The gram sabha — the general assembly of every adult in the village — had been convened under a Supreme Court order that directed the Dongria Kondh to have a “decisive say” in whether Vedanta Resources, a London-listed mining conglomerate, could blast the top off their sacred mountain to extract 70 million tonnes of bauxite. This was the twelfth and final gram sabha. All eleven before it had voted the same way: unanimously against mining. The twelfth was no different. Not a single hand was raised in favour. Vedanta had spent a decade and hundreds of crores preparing for this mine. The Dongria Kondh — approximately 9,659 people with a literacy rate of 6.35% — said no, and that was that.

The Niyamgiri vote is celebrated, rightly, as a landmark in indigenous rights. But there is a detail that gets lost in the celebration: the Dongria Kondh have no paramount chief. No king. No single leader who speaks for the community. They have clans with their own leaders, villages with their own beju and bejuni (male and female priests), and a decentralised structure where nobody holds centralised political power. The unanimous “no” across twelve separate gram sabhas was not orchestrated by a tribal authority figure giving orders. It emerged from a governance architecture that predates the Indian Constitution by centuries — a system where decisions are made through consensus within small, autonomous units that share a common spiritual relationship with the land. The Indian state, which had designed the gram sabha as a tool of democratic decentralisation, accidentally gave the Dongria Kondh a mechanism that mapped perfectly onto governance principles they had been practising since before the East India Company existed.

This chapter is a map of a country within a country. Not a metaphorical country — a real one, with its own languages, its own governance systems, its own land relations, its own spiritual geography, and its own relationship with power. Odisha’s 62 tribal communities, totalling 95.9 lakh people (22.85% of the state’s population as of Census 2011), inhabit nearly 44.71% of the state’s land area. They are not a minority clinging to the margins. They are a civilisation — or rather, dozens of civilisations — occupying almost half the territory. The Indian state treats them as one category: Scheduled Tribe. One checkbox on a form, one column in a budget, one line in a manifesto. In software terms, the state built a monolithic API and tried to force 62 independent microservices to conform to it. This chapter is about what those microservices actually look like, why the monolith keeps failing, and what we miss when we refuse to see the architecture for what it is.

Sixty-Two Is Not One

Here is a number that should stop you: 62.

Odisha recognises 62 tribal communities, including 13 Particularly Vulnerable Tribal Groups (PVTGs) — the highest number of both tribal communities and PVTGs of any state in India. And yet, in the public imagination, in policy documents, in political speech, and in the vast apparatus of the Indian state, these 62 communities are routinely collapsed into a single noun: “tribal.” One word. One demographic category. One set of policies. One budget line.

This is not a minor semantic complaint. It is an architectural error of the most fundamental kind.

Consider what is actually being grouped together. The Gond of western Odisha speak Gondi, a Dravidian language. The Santhal of Mayurbhanj speak Santali, a Munda language from the Austroasiatic family — as linguistically distant from Gondi as English is from Mandarin. The Saora of Gajapati speak Sora, another Austroasiatic language but from a completely different branch than Santali. The Bonda of Malkangiri speak Remo, yet another Munda language, and genetic studies link them to the first wave of human migration out of Africa approximately 60,000 years ago. These are not dialects of a common tongue. These are separate languages from separate language families, carrying separate cosmologies, separate oral traditions, and separate ways of understanding the relationship between human beings and the world.

In software engineering, there is a well-known architectural pattern called microservices. Instead of building one massive monolithic application that handles everything — authentication, payment processing, inventory management, user notifications — you build small, independent services, each responsible for one thing, each with its own database, its own logic, its own API. The services communicate with each other through well-defined interfaces, but each is autonomous. If one service fails, the others keep running. If one needs to scale, it scales independently. The alternative — the monolith — puts everything into one codebase, one deployment, one failure domain. Change one thing, and you risk breaking everything.

The Indian state’s approach to tribal communities is the monolith. One category. One legal framework (the Fifth Schedule). One extension of democratic governance (PESA). One forest rights law (FRA). One budget mechanism (Tribal Sub-Plan). These are not bad policies in isolation. But they are monolithic policies applied to a microservices reality. The Gond garh system, the Kondh mutha system, the Santhal Manjhi-Pargana system, the Juang pirh system, the Bhuiyan pirha system, the Bonda Naik system — each is its own service, with its own internal logic, its own data structures, its own governance API. Treating them as one system is not just politically convenient. It is architecturally catastrophic.

The Map They Do Not Show You

Pull up a map of Odisha. The images that dominate the public imagination are coastal: Puri’s Jagannath temple, Bhubaneswar’s Lingaraj, Konark’s sun wheel, Chilika’s lagoon. This is the Odisha that tourists visit, that gets discussed in national media, that appears in the popular consciousness. It is also, roughly, the eastern third of the state — the coastal plain where the major rivers deposit their silt, where wet rice cultivation supports dense populations, where cities grew, where British administration took root earliest.

Now look at the rest. Behind the coast, to the west and south, the land rises. The Eastern Ghats — not the dramatic peaks of the Himalayas, but a crumpled landscape of forested hills, deep valleys, and laterite plateaus — run through the interior like a spine. This is where the tribal populations are concentrated. Not uniformly, but with a pattern so stark that it divides the state into two separate demographic realities.

Seven districts are fully designated as Scheduled Areas under the Fifth Schedule: Mayurbhanj, Sundargarh, Koraput, Malkangiri, Nabarangpur, Rayagada, and one more from the reorganised undivided Koraput territory. Six more are partially designated: Sambalpur (Kuchinda tahasil), Keonjhar (Keonjhar, Telkoi, Champua, Barbil tahasils), Kandhamal (Kandhamal, Balliguda, and G. Udayagiri tahasils), Ganjam (R. Udaygiri tahasil and parts of Parlakhemundi and Suruda), Kalahandi (Thuamul Rampur and Lanjigarh blocks), and Balasore (Nilagiri block). Together, these Scheduled Areas cover 44.71% of Odisha’s total land.

The geographic pattern tells a political story. The coastal districts — Puri, Cuttack, Jagatsinghpur, Kendrapara, Jajpur, Bhadrak — have minimal tribal populations. These are the districts that have dominated Odisha’s politics, economy, and cultural production since the province’s formation in 1936. They are where the state capital sits, where the major roads lead, where the universities and courts operate, where the media is headquartered. They are, in the language of The Long Arc, the product of a different institutional inheritance: 130 years of British administration that, for all its exploitation, at least produced courts, revenue records, and a thin layer of bureaucratic infrastructure.

The interior and western districts — the Scheduled Areas — are the product of a different inheritance entirely. These territories were, until 1948-49, princely states. Kalahandi, Keonjhar, Mayurbhanj, Gangpur (now Sundargarh), Koraput — they entered the Indian democratic state with land, forests, minerals, and people, but no institutions. The rajas had provided whatever they chose to provide, which in most cases was very little. When the princely states merged with Odisha, they entered as what a software team would call “undocumented legacy systems” — running, but with no one who understood their internal logic, no tests, and no documentation.

The tribal-interior overlap is not coincidental. The princely states survived as princely states precisely because they were remote, forested, and difficult to administer directly. The British left them alone because the cost of direct administration exceeded the revenue they could extract. And the communities that lived in these territories — the Gond, the Kondh, the Santhal, the Juang, the Saora, the Bonda — survived as distinct cultures precisely because the terrain protected them from the homogenising pressures that flattened cultural distinctiveness in the plains.

Geography is not destiny, but it is infrastructure. The hills and forests that allowed tribal communities to maintain their governance systems for centuries also ensured that when the Indian state arrived with its development programmes, its block development officers, and its five-year plans, it arrived into terrain that resisted the state’s own operating assumptions: that land should have clear individual title, that governance should flow through elected representatives, that forests are a “resource” to be “managed,” and that modernity means integration into the dominant system. Every one of these assumptions, when applied to the tribal interior, collided with a reality that operated on different principles.

The Gond: A Federal Architecture

Start in western Odisha — Kalahandi, Nuapada, Balangir, Sambalpur, Sundargarh. This is Gond country.

The Gonds are the second-largest tribal community in Odisha, a Dravidian-speaking people who migrated through the western corridor via the Kalahandi-Nuapada region. They speak Gondi, though the Gonds of Sundargarh, Sambalpur, and Balangir have been deeply acculturated into Hindu customs and now speak Odia. This alone — the fact that a single tribal community spans a spectrum from Gondi-speaking forest cultivators to Odia-speaking, Hindu-acculturated warriors who held zamindari rights — should suggest how absurd it is to treat “tribal” as a uniform category.

The Gond political system was built around the garh — a territorial domain, literally meaning fort. When the Chauhan dynasty established its rule over Sambalpur, it did not replace the Gond territorial organisation. It superimposed new administrative nodes on top of it. The result was the Attargarh system: eighteen garhs, each controlled by a particular Gond clan, federated under the Chauhan overlord who bore the title “Atharagada-maudaamani” — the great jewel on eighteen garhs.

Each garh was subdivided into units of 84 villages called chourasi, which were further broken down into barhons of 12 villages each. This created a layered federal structure: the clan chief governed the garh, barhon leaders governed clusters of villages, and village headmen (variously called majhi, patel, or mukhiya) managed daily affairs through traditional panchayats.

In software terms, this is a multi-tier architecture with clear separation of concerns. Village-level governance handles local disputes, land allocation, and daily administration. The barhon aggregates twelve villages for coordination. The garh provides clan-level governance over a territorial domain. The Attargarh federation coordinates across all eighteen garhs. Each tier has its own authority, its own leadership, its own scope. The architecture balanced central coordination (the Chauhan overlord could call on Gond military support) with local autonomy (Gond clan chiefs governed their own territories without revenue obligations to the overlord).

The relationship between the Gonds and the Chauhan rulers was symbiotic, not subordinate. During the rule of Jayant Sai in the 1780s, the Gond chief of Sarangarh served as dewan — chief administrator — of the entire Sambalpur estate. One Chauhan ruler granted the zamindari of Kharsal to a Gond soldier named Udam Singh, decorated with the hereditary title of Sardar. Most Gond zamindars of western Odisha held their zamindari on military tenure: they paid no revenue or tribute but were obligated to provide men and money during emergencies. The Gonds were not subjugated people living under a Hindu kingdom. They were political partners who had negotiated a relationship that preserved their autonomy while participating in a larger political structure.

This matters because the colonial and post-colonial narrative treats tribal communities as people who were passively “ruled” by successive powers — Hindu kingdoms, the British, the Indian state. The Gond evidence tells a different story. These were communities with their own political sophistication, their own negotiating leverage, and their own capacity for alliance-building. The Gonds and Binjhals actively participated in the anti-colonial resistance in the Sambalpur region from 1857 to 1864. Their territory was known as “Hirakhand Samrajya” during the Chauhan era — not a backwater, but a domain of significance.

The Gond social organisation adds another layer: the four-phratry (saga) system. All Gond clans trace descent from one of four groups of gods who emerged from a primordial cave after their release by the culture hero Lingal. The phratries are divided into exogamous, totemic clans. Each clan worships a deity called persa pen (“great god”), and the shrine of this deity lies within the ancestral clan land. The most prestigious division — the Raj Gonds, tracing descent from the elder sister in the origin myth — once maintained an elaborate feudal order and held positions as zamindars in western Odisha.

Sacred geography, political territory, and clan identity are fused into a single system. The shrine is on the ancestral land. The land defines the garh. The garh is governed by the clan. The clan is defined by descent from a specific deity. Pull any thread and the entire fabric moves. When the Indian state treats land as a commodity that can be acquired for “public purpose,” it is not just taking physical territory. It is severing the connection between a people, their ancestors, their deity, and their political identity. The monolithic API has no field for this. It has fields for “compensation amount” and “rehabilitation package.” It does not have a field for “relationship between a people and the ground they consider sacred.”

The Kondh: Decentralisation as Strength

Move south and east into Kandhamal, Rayagada, Kalahandi, Koraput. This is Kondh territory.

The Kondhs (also Kandha, Khond) are the largest tribal group in Odisha by population. They speak Kui and Kuvi, Dravidian languages. But “the Kondh” is itself a monolithic label that conceals enormous internal diversity. The Dongria Kondh of the Niyamgiri hills and the Kutia Kondh of Belghar in Kandhamal are both classified as PVTGs — both are Kondh sub-groups, but they live in different terrain, practise different agricultural systems, and have different social structures. Treating them as interchangeable is like treating a mobile banking app and an air traffic control system as the same software because they both run on Linux.

The Kondh governance system is built around the mutha — simultaneously a clan unit and a territorial unit. There are approximately fifty muthas, each encompassing a cluster of villages. Members of a mutha consider themselves related through common descent; marriage within the mutha is strictly prohibited. This is not just a social rule. It is a system design that forces inter-community connection: you must marry outside your mutha, which means every family has kinship ties to other muthas, which means political alliances and economic networks are woven through the marriage system itself.

At the village level, the Kondh operate through a council called the Gudi, composed of elders (mukhiyas). The secular head is the Saanta, supported by the Mondal (administrative chief at mutha level), the Bismajhi (deputy headman), and the Barik (village messenger). But alongside this secular governance runs a parallel structure of religious specialists: the Jani (full-time ritual specialist), the Dishari (astrologer and diviner), and among the Dongria Kondh, the beju and bejuni (male and female priests).

This dual secular-sacred authority is not ornamental. It is a checks-and-balances system. The secular head makes administrative decisions; the religious specialist determines whether those decisions are in alignment with the community’s spiritual obligations. When the Dongria Kondh rejected Vedanta’s mine, it was not a political calculation in the way the Indian state understands politics. It was a community deciding that allowing their sacred mountain to be destroyed was incompatible with their obligations to Niyam Raja, the deity who resides on the peak. The secular and sacred were not separate domains to be weighed against each other. They were the same domain.

The Kondh system of dispute resolution is worth studying closely because it reveals a governance logic that the modern state has never been able to replicate. When a dispute arises, one party approaches a mukhiya. The mukhiya instructs the barik to invite all community members to a conflict resolution meeting. A council of elders convenes, both parties present their cases, and a collective decision is reached. The penalty structure is distinctive: both parties contribute to a fine, with the accused paying more than the victim. This emphasis on shared responsibility fosters communal unity rather than adversarial division. The system’s effectiveness rests on the community’s trust and respect for elders — a resource that no legislation can create and no budget can purchase.

The Kondh relationship with surrounding non-tribal polities was marked by autonomy punctuated by violent intervention. The British campaign to suppress Meriah sacrifice (a practice of human sacrifice to the earth goddess Dharani Penu) in the 1830s-40s was perceived by the Kondh as a direct attack on their religious faith. Under leaders like Dora Bisoi and Chakra Bisoi, the Kondh mounted sustained resistance from 1837 to 1856. When Captain Macpherson rescued 170 Meriah victims, the Kondh assembled before his camp and demanded them back, forcing Macpherson to release them. British revenue collection of up to 50% of land revenue through coercion compounded resentment. The Kondh were not passive recipients of colonial rule. They fought. They lost, eventually, but they fought on their own terms, defending their own governance logic.

The Saora, the Juang, and the Bonda: Three Kinds of Isolation

The Eastern Ghats produce not just geographic isolation but civilisational divergence. Three communities — the Saora (Gajapati), the Juang (Keonjhar), and the Bonda (Malkangiri) — illustrate how the same broad ecology of forest and hills can produce radically different governance architectures.

The Saora are a Munda (Austroasiatic) ethnic group inhabiting the Eastern Ghats of southern Odisha — Ganjam, Gajapati, Rayagada, Koraput. Their internal division tells its own story: the Lanjia Saora (Hill Saora) live in the hills, maintain traditional customs, practise shifting cultivation, and are classified as a PVTG. The Sudha Saora (Plain Saora) live in the plains, have integrated more deeply into mainstream society, and practise settled agriculture. Same community, split by altitude into two different worlds.

The Saora organise around the birinda — an extended family or lineage group tracing descent from a common ancestor of four to five generations. Village territories, especially hill-slope cultivation land, are distributed on the basis of birinda membership. The governance structure features a dual leadership that is almost constitutional in its separation of powers: the Gomango (secular head) exercises administrative and judicial authority, while the Buyya (sacred head) presides over all religious ceremonies and spiritual matters. Both positions are normally hereditary. Neither can function without the other. It is as if someone designed a village government with an executive branch and a spiritual branch and gave them co-equal authority by design.

Alongside these formal offices, the Kudan (shaman) holds significant authority as a spiritual diagnostician — combining the functions of priest, prophet, and healer. Unlike the Gomango and Buyya, whose positions are inherited, the Kudan’s role is achieved through perceived spiritual gifts. This is a meritocratic appointment within an otherwise hereditary system — a different API for a different function.

The Saora land system, called Garajang Andruku (village community ownership), creates nested layers of ownership: the village owns the territory; birinda groups hold customary rights over specific tracts; individual cultivators hold usufruct rights contingent on active cultivation. This is not private property. It is also not state property. It is a complex system of overlapping claims that colonial and post-colonial property law has never been able to accommodate. The Lanjia Saora are also known for their distinctive wall paintings (ittalan or idalon), created under the direction of the Kudan to propitiate spirits and heal the sick — art as governance, aesthetics as medicine.

The Juang of Keonjhar district are one of the most ancient tribal groups in Odisha, classified as a PVTG. They speak the Juang language (Mundari family) and organise their villages into four pirhs: Satkhand, Jharkhand, Kathua, and Rebena. The territory in Keonjhar where they predominantly live is called Juang Pirh — the land itself is named after them.

The Juang’s most distinctive institution is the majang (or mandaghar) — a rectangular house standing conspicuously in the centre of every village, functioning as community house, courthouse, guesthouse, cooperative store, musical instrument repository, ritual venue, cultural centre, and what one might only call a museum. It is the single most multifunctional governance institution I have encountered in any of this research. Unmarried boys and girls become members. The barabhai (village elders) adjudicate disputes here. Visitors are housed here. Communal grain stores are kept here. Village instruments are stored here. Ceremonies take place here. Young people learn traditional skills, cultural values, and social responsibilities through experiential learning within its walls.

The majang is not just a building. It is an institutional design that solves multiple problems simultaneously: education (youth learn by living together), social integration (inter-family unity through shared living), cultural preservation (storytelling, song, craft), labour organisation (community projects), conflict resolution (youth learn to negotiate differences in close quarters), and security (communal vigilance). It is, in effect, a community operating system running in a single physical space. No modern development programme has produced an institution this efficient or this comprehensive.

Most Hill Juang villages have uni-clan composition — a single clan per village — and village exogamy is the rule. This means the clan-village identity overlap is nearly complete, and the marriage network is the mechanism that connects villages across the pirh. The social architecture and the governance architecture are the same thing.

The Bonda (also Bondo, or Remo in their own language) of Malkangiri district represent the extreme case. Approximately 12,000 people confined to about 130 square kilometres of hill area in the Khairput block, near the tri-junction of Odisha, Chhattisgarh, and Andhra Pradesh. Genetic studies link them to the first wave of human migration out of Africa. They are, in a very literal sense, among the oldest continuous populations on earth.

Bonda villages are traditionally autonomous. The Naik (village chief) presides over council meetings at the Sindibor — a megalithic stone platform at the centre of the village that serves simultaneously as council chamber and shrine. The architecture is the argument: political authority and spiritual authority occupy the same physical space because they are not conceived as separate domains.

The Bonda practise a unique blend of settled and shifting cultivation, growing over 20 crop varieties while preserving rich biodiversity. Their farming system has been recognised as an effective indigenous response to climate change, maintaining soil health through diversified cropping and fallowing. They have matriarchal elements — women have significant authority in family and social matters, and Bonda women often marry men eight to ten years younger than them.

The Bonda’s relationship with the outside world has been one of fierce, deliberate isolation. The British described them as “stubborn” — a characterisation the tribe appears to embrace. They have been “targets” of development schemes for fifty years through the Bonda Development Agency, established in 1976-77. The results are a monument to the failure of monolithic development: 33% literacy (against a state average of 73%), 79% undernutrition among Upper Bonda, 7.8% malaria prevalence. Villages like Barguda lack schools, anganwadi centres, and basic road connectivity. Children must trek two kilometres over steep hills for education.

The paradox deserves to be stated plainly: the Bonda have a dedicated development agency, live in a fully Scheduled district, are designated as a PVTG, and are protected by every constitutional provision the Indian state has designed for tribal welfare. After fifty years of this protection, their development indicators remain among the worst in India. The institutional architecture is elaborate. The results are catastrophic. Something in the architecture is fundamentally misaligned with the reality it is supposed to serve.

The Santhal: The Most Documented Democracy

Travel north to Mayurbhanj, and the governance architecture changes entirely. This is Santhal country, and the Santhals have built what may be the most elaborate traditional democratic system of any tribal community in India.

The Manjhi-Pargana system operates at four levels. At the village level, seven officials govern: the Manjhi (village headman and primary worshipper), the Paranik (deputy headman), the Jog Manjhi (youth supervisor who oversees the atur, the youth dormitory), the Jog Paranik (assistant youth supervisor), the Godet (messenger), the Naeke (village priest), and the Kudam Naeke (assistant priest who worships forests and nature spirits). Seven positions, each with defined scope, covering secular administration, youth oversight, communication, and religious functions. This is not an ad hoc arrangement. It is a designed system with clear role separation.

Above the village, a Pargana of fifteen to twenty villages forms an intramural organisation, headed by the Parganat. Above that, the Deshmanjhi serves as a higher authority. And above all of these sits the Lo Bir — the democratic assembly that is the most remarkable institution in the entire Santhal governance architecture.

The Lo Bir was convened during the annual communal hunt in January. After the hunt, a people’s assembly was held at night. All pending issues were discussed. Everyone could speak. Even the poorest person could bring forward complaints. And critically — even complaints against the parganas, deshmanjhis, or manjhis themselves could be raised and adjudicated. This was a popular accountability mechanism operating in a society that had no written constitution, no formal judiciary, and no concept of “rights” in the Western legal sense. The leaders were hereditary, but they were answerable to the assembled people. This is a form of democratic governance that predates British parliamentary democracy by centuries.

The Santhal judicial system has four tiers: Manjhi Baisi (village court), Mapanjhi Baisi (inter-village court), Pargana Baisi (pargana court), and the Lo Bir Baisi (supreme assembly). Four tiers of courts, each with defined jurisdiction, each with escalation procedures. It is a judicial architecture that any systems designer would recognise as a well-structured hierarchy with proper error handling.

The Santhal land system ties political office to land allocation. When a new village is established, the rulers sit together and divide land in defined proportions: the Manjhi receives four portions, the Paranik three, the Jog Manjhi two, and each of the remaining four officials receives one. This creates material support for governance — the officials are compensated for their work through land grants — while maintaining a proportional hierarchy that reflects responsibility.

Each Santhal village maintains a sacred grove (Jaher or Jaheera) on the outskirts, regarded as the abode of all Santhal deities. In front of the Manjhi’s house, the Manjhithan serves as a sacred place and seat of Manjhi-haram, the founding ancestor deity. The sacred grove system represents a form of community-managed conservation that, in Odisha’s tribal areas as a whole, has produced at least 11,000 documented sacred groves — a conservation infrastructure built and maintained without a single government programme.

The British recognised the power of this system. After the Santhal Rebellion of 1855-56 (the Hool) — one of the largest tribal uprisings against the East India Company, driven by zamindari exploitation — the British response was to create the separate Santhal Parganas administrative district in present-day Jharkhand. They did not dismantle the Manjhi-Pargana system. They recognised it as a functioning governance structure and built administrative boundaries around it. This is an extraordinary concession from a colonial power that generally assumed its own institutions were superior.

The Bhuiyan: Lords of the Soil

One more system deserves attention, because it inverts every assumption about the relationship between tribal communities and the state.

The Bhuiyan (also Bhuiya, Bhuyan) are found across Keonjhar, Sundargarh, Jharsuguda, Mayurbhanj, Sambalpur, and Angul. Their name derives from the Sanskrit bhumi — land. They call themselves the autochthons, the original inhabitants. The territory where they predominate in Keonjhar is called Bhuyan Pirh after them, just as the Juang territory is called Juang Pirh. The land is named after the people, not the other way around.

The Bhuiyan governance system (the pirha, similar in name but different in structure from the Juang system) operates through village-level pradhans and inter-village pirha panchayats led by sardars. But the Bhuiyan’s most distinctive feature is not their internal governance. It is their relationship with the non-tribal state.

The Bhuiyans had king-making authority over the Raja of Keonjhar. This was not ceremonial. According to Bhuiyan tradition, they originally stole a boy from the royal family and made him their king, offering him the right to rule and punish wrongdoers, but simultaneously reserving the right to dethrone him if he became oppressive. In the coronation ceremony, the Bhuiyans played a vital role without which the coronation was not considered complete. Chiefs who held higher positions in the administrative machinery had no right to exercise any authority until they had received the tilak (token of investiture) from their Bhuiyan vassals. The rulers of the Keonjhar state were dependent on Bhuiyan support for their political legitimacy.

Read that again. The tribal community did not serve the king. The king served at the pleasure of the tribal community. The Bhuiyans functioned as an organised militia, holding lands on conditions of military service and maintaining themselves in readiness to either oppose their raja or fight for him. This is a form of popular sovereignty — the right of the people to install and remove their ruler — that political theorists usually associate with Enlightenment philosophy. The Bhuiyans were practising it while Locke was still a child.

When the British tried to interfere with this system — installing their own candidates as rajas, overriding the Bhuiyan king-making prerogative — the Bhuiyans revolted. The 1867 uprising was the biggest tribal rebellion in 19th-century Odisha, triggered directly by British interference in the Bhuiyan practice of crowning or rejecting the king of their choice. The second phase, led by Dharanidhar Naik in 1891-1895, represented sustained armed resistance. The pirha system was not a quaint cultural artefact. It was a political institution capable of organising military resistance in defence of its prerogatives.

What They Share

The diversity is real and irreducible. The Gond garh system is federal. The Kondh mutha system is decentralised. The Santhal Manjhi-Pargana system is hierarchical with democratic accountability. The Juang system centres on the majang institution. The Bonda system is village-autonomous. The Bhuiyan system includes king-making authority over non-tribal rulers. These are genuinely different architectures, built for different terrains, different economies, different population densities, and different relationships with neighbouring powers.

But underneath the diversity, there are structural commonalities that define what it means to be a tribal community in Odisha — commonalities that set these 62 communities apart from the dominant social order of the plains.

First: land is a relationship, not a commodity. Every community studied here practised communal land tenure in some form. The village, the clan, the birinda, the mutha — some collective body held rights over territory. Individual families held usufruct rights contingent on active cultivation. You could use the land. You could not sell it, mortgage it, or transfer it to an outsider. The Saora Garajang Andruku system, the Gond clan-based tenure, the Santhal office-linked land allocation, the Kondh mutha-territory overlap — all express the same principle: the land does not belong to a person. The person belongs to the land.

This is not a primitive precursor to property law. It is a fundamentally different operating system for human-land relations. And when colonial property law — which requires individual, documented, alienable title — was imposed on top of it, the incompatibility was total. The Indian Forest Acts of 1865, 1878, and 1927 progressively criminalised shifting cultivation, the agricultural practice that most tribal communities depended on. Hill slopes customarily used for podu were taken over by the state as forest lands or revenue wastelands. Communities that had been custodians of the forest for centuries were reclassified as “encroachers” overnight. The very unit of governance that had maintained the forests — the clan, the village council, the birinda — was rendered legally invisible.

Second: governance is dual — secular and sacred — by design. The Saora have the Gomango (secular) and Buyya (sacred) as co-equal heads. The Kondh have the Saanta/Mondal (secular) and Jani/Dishari (sacred). The Bonda have the Naik (secular) and Sisa (sacred). The Santhal have the Manjhi (secular, but also a worshipper) and the Naeke (sacred). This is not a coincidence across unrelated cultures. It is a design pattern: separating the power to make practical decisions from the authority to determine spiritual alignment prevents any single individual from monopolising power. The sacred specialist cannot make administrative decisions. The secular head cannot override spiritual obligations. Both are necessary for the community to function.

The modern state has no equivalent. When the Indian government acquires tribal land for mining, it offers monetary compensation — a transaction that is legible only within the secular domain. It has no mechanism for addressing the spiritual dimension: the fact that the mountain is a deity, the grove is the residence of ancestors, the river is a living being with whom the community has obligations. The monolithic API does not have fields for these data types.

Third: the dormitory institution is a governance technology. Three of the seven major communities maintain or maintained formal youth dormitory systems — the Gond ghotul, the Juang majang, and the Santhal atur. These are not cultural curiosities. They are institutional designs that solve the problem of social reproduction: how do you transmit knowledge, values, skills, and governance capacity across generations in a society without formal schools, without written texts, without a professional teaching class? You build a physical space where young people live together, learn by doing, absorb the community’s norms through daily practice, and develop the conflict resolution skills they will need as future leaders. It is a technology for producing citizens, built into the architecture of the village itself.

Fourth: forests are governance infrastructure, not natural resources. The 11,000 sacred groves documented across Odisha’s tribal areas — called Jahera by the Santhals, Pat by the Bathudi and Gond and Saora, Thakurani in general Odisha usage — are not parks. They are governance infrastructure. They operate under strict community-enforced rules: no tree cutting, no collection from the forest floor, no hunting. Violation invites punishment from presiding deities. These rules have maintained forest cover and biodiversity for centuries without a single Forest Department officer, a single government notification, or a single conservation budget. The Dongria Kondh regard the entire Niyamgiri hill range as sacred, a living deity. When the state proposes to mine such a landscape, it is not proposing to extract minerals from “land.” It is proposing to dismantle a governance system in which the forest is both the governed territory and the governing authority.

Fifth: oral tradition is the constitutional framework. None of these governance systems are written down in the communities’ own records. The garh system, the mutha system, the Manjhi-Pargana system, the pirha system — all operate through oral tradition, collective memory, and lived practice. The “constitution” is not a document. It is the accumulated memory of how disputes were resolved, how land was allocated, how leaders were chosen, and how the sacred geography was maintained. When colonial administrators demanded written records as proof of land rights, they were demanding that an oral constitutional framework translate itself into a written one — a translation that necessarily lost the relational, contextual, and mutable qualities that made the oral system work. The Forest Rights Act of 2006 attempts to address this by recognising customary rights. But the process of claim-filing, evidence-gathering, and committee verification still operates on documentary logic. In the CAG audit of FRA implementation in Odisha, 87.5% of Forest Rights Committee members reported that they never received the forest and revenue maps they needed to verify claims. The system designed to recognise oral rights still runs on documentary infrastructure.

Thirteen Degrees of Vulnerability

Within the 62 communities, thirteen have been designated as Particularly Vulnerable Tribal Groups — PVTGs. Odisha has more PVTGs than any other state in India. The designation is based on three criteria: pre-agricultural level of technology, extremely low literacy, and declining or stagnant population growth.

The numbers are worth confronting directly.

The Kutia Kondh of Belghar in Kandhamal have a literacy rate of 4.98%. The Dongria Kondh: 6.35%. The Didayi of Malkangiri: 6.87%. For every hundred Kutia Kondh adults, fewer than five can read. The state average is 72.87%. The gap is not a “development deficit.” It is an abyss.

The health data compounds the picture. Malaria prevalence among the Kutia Kondh is 25% — one in four people. Among the Dongria Kondh, 13.8%. Among the Paudi Bhuyan, 11%. Undernutrition among Upper Bonda children reaches 79%. Anaemia among women in some PVTG communities ranges from 70% to 92%. The Bonda have an HBV (Hepatitis B) prevalence as high as 14.18%.

The total PVTG population in Odisha is approximately 7.73 lakh (Census 2011). The largest group, the Saora, numbers 5.35 lakh. The smallest, the Birhor of Sundargarh and Sambalpur, numbers 596 people. Five hundred and ninety-six. An entire people who could fit in a single auditorium.

Each PVTG has a dedicated development agency — the Bonda Development Agency since 1976, the Juang Development Agency, the Dongria Kondh Development Agency, the Kutia Kondh Development Agency. These agencies have operated for decades. They implement programmes in education, healthcare, skill training. They spend budgets. They file reports. And the numbers barely move.

This is where the microservices metaphor becomes not just an analytical framework but a diagnosis. A development agency staffed by non-tribal bureaucrats, headquartered at a block office, implementing standardised schemes designed in Bhubaneswar or Delhi, operating in communities with 5-33% literacy and distinct cultural systems, using a language (Odia or Hindi) that may not be the community’s first language, applying frameworks (individual land title, formal schooling, modern healthcare) that may be structurally incompatible with the community’s own institutions — this is a monolithic API call to a service that speaks a different protocol entirely. The request is well-formed by the caller’s standards. It is unintelligible to the receiver. The connection times out. The caller logs an error (“scheme implemented, outcomes below target”) and moves on.

What would a microservices approach look like? It would start by reading the documentation of each service — understanding the Bonda’s Sindibor council system before designing a governance intervention, understanding the Juang’s majang before building a school, understanding the Saora’s birinda land system before filing FRA claims. It would design interfaces that respect each community’s internal protocol. It would allow each service to scale, evolve, and fail independently rather than coupling all thirteen PVTGs to the same development pipeline. It would, in short, treat these communities as the distinct civilisations they are rather than as thirteen instances of the same problem.

This is not what happens. What happens is PM-JANMAN: one scheme, Rs 24,104 crore, covering all 75 PVTGs across India. One API for seventy-five different systems. The same allocation formula for the 596 Birhor of Sundargarh and the 5.35 lakh Saora of Gajapati. The monolith ships on schedule. The microservices crash silently.

The Parallel Civilisation

Here is the structural argument, stripped to its core.

Odisha contains a parallel civilisation. Not a marginal population. Not a “vulnerable section” to be uplifted. A civilisation — or more accurately, dozens of civilisations — with their own governance systems, their own land relations, their own languages, their own spiritual geographies, and their own relationships with power. Nearly one in four people in Odisha belongs to this parallel civilisation. Nearly half the state’s territory is constitutionally designated as their domain.

The Indian state has built an extraordinary institutional architecture to protect them: the Fifth Schedule (since 1950), the Tribes Advisory Council, the Governor’s regulatory powers, the Land Transfer Regulation (1956), PESA (1996, rules still not finalised after twenty-seven years), the Forest Rights Act (2006), the Prevention of Atrocities Act (1989), 21 Integrated Tribal Development Agencies, 13 PVTG micro-projects, the Tribal Sub-Plan, PM-JANMAN, the National Commission for Scheduled Tribes. Layer upon layer of constitutional provisions, legislative enactments, institutional bodies, and funding schemes.

And yet: PVTG literacy ranges from 4.98% to 41.75% against a state average of 73%. Conviction rate for atrocities against STs in Odisha from 2019 to 2021: 0%. Approximately 8.4 lakh acres of tribal land alienated with only 6.75% restored. PESA rules unfinished after twenty-seven years. 43% of land acquisitions in Scheduled Areas conducted without gram sabha consent. The FRA rejection rate for Other Traditional Forest Dwellers: 99.77%.

The gap between institutional architecture and lived reality is not a bug. It is the system working as designed — or rather, working as the incentive structures dictate. The constitutional protections exist to demonstrate that the state cares. They do not exist to ensure that tribal communities have power. The difference between caring and power is the difference between a monitoring dashboard and an admin console. The dashboard shows you what is happening. The admin console lets you change it. The tribal protection framework is an elaborate dashboard. Nobody has given these communities the admin console.

This is not a story that the Jagannath framework — the framework explored in The Lord of the Blue Mountain — can accommodate. Jagannath’s genius is syncretic absorption: pulling diverse traditions into one tent, one Mahaprasad, one Rath Yatra. But the tribal communities of Odisha were not absorbed. The Sabara Dalapati walks at the front of Jagannath’s chariot, yes — a ritual acknowledgment of tribal priority. But the Dongria Kondh do not worship Jagannath. The Bonda’s animistic beliefs centre on nature worship and ancestral spirits. The Santhal’s Jaheera is the abode of Santhal deities, not Vaishnava ones. The tribal communities that make up nearly a quarter of Odisha’s population operate, in spiritual terms, outside the Jagannath system entirely. If Jagannath is the organising principle of Odia collective consciousness, then a quarter of Odisha has a different consciousness altogether.

This is not a story that Delhi’s Odisha fully captured either. The PESA betrayal documented in that series — the twenty-seven-year non-implementation, the 126 land acquisitions without gram sabha consent, the Samatha judgment circumvented through state PSU proxies — is the policy layer of a deeper structural problem. The policy fails not because policymakers are malicious (though some are) or incompetent (though some are). It fails because it is monolithic policy designed for a microservices reality. You cannot write one API specification for 62 different systems and expect it to work. The specification will inevitably reflect the assumptions and priorities of the system designer — in this case, a state apparatus rooted in coastal, urban, Hindu-caste-dominant Odisha — and those assumptions will be structurally incompatible with the governance logic of communities that operate on fundamentally different principles.

The first step toward a different outcome is the step this chapter has attempted: seeing the 62 communities as 62 different systems, each with its own architecture, its own logic, and its own API. The Gond garh is not the Kondh mutha is not the Santhal Pargana is not the Juang pirh is not the Bonda village council. They share structural principles — communal land, dual authority, oral governance, sacred geography — but they implement those principles differently, in different terrain, with different histories, at different scales. Any intervention that ignores this diversity — that treats “tribal” as one word, one budget line, one policy framework — is doomed to the same failure that fifty years of development agencies have already produced.

The question is not whether the Indian state should protect tribal communities. It should. The question is whether the Indian state is architecturally capable of protecting 62 different civilisations simultaneously using a monolithic governance framework designed for a different kind of society entirely. The evidence from seven decades of constitutional promise and institutional failure suggests the answer.

Sources

Source Research

The raw research that informs this series.