A Boy, a Pot and Thirty Wards — How a Tamil village chose its leaders by lot, and rewrote its own rules in 921
Blossomed, Not Assembled · Volume IV of VI · Parts Sixteen to Nineteen
Earlier volumes argued that India cannot copy its way to development, that its inheritance must be sifted before it is trusted, and that Kauṭilya wrote his rules on wealth, tax and force for officials he expected to be tempted.
← Previous: III · The Honey on the Tongue · Series hub · Next: V · Who May Drink From the Tank? →
In about 921, a Tamil village chose its committees by lot from a pot, barred men who hid their accounts, and then rewrote its own rules. The wall at Uttaramērūr does not prove ancient democracy. It shows something more useful, rules that were corrected.
In Brief: For readers who have seen the Uttaramērūr inscriptions shared as proof of "ancient Indian democracy" and want to know what the stones actually say. The article reads the rules of about 919 and 921 CE line by line: the pot, the boy, the thirty wards, the accounts rule, the banned relatives. It shows what changed in two years and what cannot be known. It then follows the village assembly to Ambedkar's refusal in 1948 and the panchayats of 1992, and asks what "developed" should mean for a country whose institutions must outlive their founders.
Two inscriptions at Uttaramērūr, cut about 919 and 921 CE, show a Tamil village writing rules for choosing its committees and then rewriting them, and they change what viksit should mean, from a nation opening like a flower to institutions that later hands keep, use and correct.
The article reads the rules as rules: a lottery from a pot, one year in office, accounts before a man could stand again, and a ban that reached his relatives. It finds a franchise no modern republic could accept and mechanisms any republic could use. It finds one recorded revision, not a civilisation's habit. It follows the village assembly to Ambedkar's refusal in 1948 and to the panchayats of 1992, where the form has arrived and much of the money and power has not. What the wall leaves open is who drew the water.
Opening — The Boy Who Could Not Know Whose Name He Held
In the fourteenth year of the Cōḻa king Parāntaka I, about 921 CE, the assembly of a Tamil village called Uttaramērūr met to choose its committees for the year.² The village was divided into thirty wards. Each ward had written the names of its eligible men on palm-leaf tickets and tied them into a bundle. The bundles went into a clay pot.
We know what happened next because the village cut the procedure into the wall of its Vaikuṇṭha Perumāḷ temple. The eldest of the temple priests present was to stand, lift the pot and look upwards, "so as to be seen by all people".
A boy was then to take out a ticket, "any young boy standing close, who does not know what is inside". The ticket went to the officer whom Venkayya's translation calls the arbitrator, the madhyastha. He was to take it "on the palm of his hand with the five fingers open". Then the name was read aloud.¹
The wall gives the steps, not their reasons, so the reasons that follow are my reading. The pot is lifted, I think, so that no one can say it was tampered with on the floor. The rule asks for a boy who does not know what is inside, and I infer the point is that he cannot favour a name he cannot know. The open palm, I take it, is there so that no one can swap one leaf for another.
The rules guard three different things: the draw itself, who may stand, and what a man does once he holds office.
The lottery is not what makes the wall worth a whole volume, though. About two years earlier, in the twelfth year of the same king, the same assembly had written a different set of rules, and that earlier set survives on the same temple.³ The village wrote its procedure down, lived with it, and wrote it again.
What happens to an institution when the people who run it write its rules where everyone can read them, and then change them?
The answer begins with a muddy, practical thing, because one of the committees chosen from that pot existed to look after the village tank.
Part Sixteen — Why Would a Village Need a Committee Just for Its Tank?
Because a tank does not stay a tank by itself. Left alone, it silts up, its bank breaks, and in time it stops holding water.
In Tamil, a tank is an ēri. It is not a lake that nature made. It is an earthen bank, the bund, built across a gentle slope so that rainwater collects behind it in a wide, shallow pool. Openings called sluices are cut through the bund. Through them the water is let out into channels, and the channels carry it to the fields below. Much of the Tamil country depends on rain that falls in a few months of the year. The tank is how a village keeps the monsoon for the dry months.
A tank is built once and kept every year
Every monsoon carries soil into the tank, and the soil settles on the bed as silt. Each year of silt means less water held. Water, cattle and burrowing animals wear at the bund. Sluices jam. Channels choke with weeds. And the water must be shared: whose field gets it first, how much, and what happens in a dry year when there is not enough.
None of that work is done once. A tank needs people who will dig out the silt before the rain, mend the bund, clear the channels and settle the quarrels, year after year. It also needs money set aside for that work, money that cannot quietly be spent on something else.
Tamil kings and villages knew this, and wrote it down. An inscription from the reign of an earlier Cōḻa king, Āditya I, records land declared to be erippatti, "tank-land", whose income was set apart "for (the maintenance of) the tank" at a village called Nerkunram. It ends with a curse on anyone who takes that income away: "He who reduces the erippatti shall also enter the hell lower than the seventh hell."⁴ The curse tells us what the danger was. The tank's money was the money most likely to be taken.
Infrastructure is maintenance, and maintenance is an institution.
The Uttaramērūr rules create a tank committee, the ēri-vāriyam, and say how its members were chosen. They do not say what it did. V. Venkayya, the epigraphist who first translated the rules in 1908, guessed that the committee "was probably entrusted with the annual removal of silt, occasional repairs" and similar work.¹ His guess is reasonable. It is still a guess, and I will keep it marked as one.
A tank is a commons
A tank is also a shared resource. Water that one field takes, another field cannot have. Elinor Ostrom, who won the Nobel prize in economics for studying such commons, found that the ones which lasted tended to share certain features.²⁹ The users were clearly defined. The rules fitted local conditions. Someone watched how the resource was used. Penalties started small and grew. And the users could change the rules themselves.³⁰
I use her findings as a lens, not as a claim that a Cōḻa village knew them. Through that lens, the Uttaramērūr rules say a great deal about the committee: who sits, for how long, who checks the money, how the rules change. They say nothing about the water. Who drew first in a dry year? Who carried the silt? Keep those two questions in mind. The wall never answers them.
The old custom, revived as a scheme
The Tamil country had a name for the shared labour that kept a tank alive, kuḍimarāmattu, roughly "repair by the people of the place". Villagers gave work, or a share of the cost, to desilt and mend the tank they drew from.
On 13 March 2017, during a severe drought, the Tamil Nadu government relaunched the custom as a ₹100 crore scheme under the same name. The launch took place at Manimangalam, in Kanchipuram district, the same district as Uttaramērūr.⁵ The first phase covered 1,519 tanks for about ₹100 crore, and a second phase planned 2,065 tanks for ₹332 crore.⁶ On paper, the work was to be done by water users' associations of local farmers.
What do we know about whether it worked?
Less than we should. The fullest study I found is by R. Sakthivadivel and Manisha Shah for the IWMI-Tata Water Policy Programme in 2018. It looked at thirty tanks in eight districts, and its authors say it was not externally peer-reviewed.⁶ Its findings hold for those thirty tanks, not for the whole scheme. In them, the authors found real gains in some villages, especially where the demand came from farmers and local leaders were trusted.
They also found that in many of the sampled places the farmers' association was "just used as a dummy to satisfy the scheme requirements". Local engineers often chose a contractor first and formed an association around him. The authors report that "inadequate monitoring during implementation by PWD officials has led to haphazard desilting". They warn that "the sustainability of the implemented work is questionable if proper maintenance is not done".⁶
Newspaper reports from 2018 carry farmers' complaints of unfinished desilting and of works handed to ruling-party men, which the irrigation department denied.⁷ I could find no audit by the Comptroller and Auditor General devoted to the scheme, and no evaluation that followed the same tanks for several years. That gap is not proof that the scheme failed. It means the question that matters most is still unanswered. Did the scheme rebuild a standing body that keeps the tank after the money stops, or did it pay for a round of digging?
There is one test from the old wall that the new scheme can be held to. Did any rule require an association to show the farmers its accounts before it could take the next contract? The study I read does not mention one. I do not know that no such rule existed. I know I could not find it, and a millennium-old village thought it worth carving in stone.
Part Seventeen — Hide Your Accounts, and Your Uncle Cannot Stand Either
The Uttaramērūr rules are often shared online as proof of "ancient Indian democracy". They are more interesting than that, and less flattering. They should be read the way one reads any set of rules: who may stand, who may not, how the choice is made, and what stops cheating.
Uttaramērūr in the tenth century was a brahmadeya, a village granted to Brahmin landholders, and the inscriptions call it a caturvēdimaṅgalam, a settlement of Brahmins learned in the four Vedas.³ It was governed by its assembly, the sabhā. Other Tamil settlements had other bodies. A village that was not a brahmadeya had its ūr, and a few trading towns had a merchants' body, the nagaram.³¹ So these are the rules of one sabhā, not of every Tamil village.
The fuller rules are those of the fourteenth year. The Tamil texts of both inscriptions have a modern critical edition by the DHARMA project.² ³ For the English I quote V. Venkayya's translation of 1908, the first published.¹
Who was allowed into the pot?
The village had thirty wards. The Tamil word usually given for a ward is kuḍumbu.⁸ The people of each ward met and put forward names. To be eligible, the later rules say:¹
"He must own more than a quarter (vēli) of tax-paying land." A vēli is a Tamil land measure, about 2.5 hectares by one historian's estimate. A quarter vēli was therefore roughly an acre and a half.³¹
He must live "in a house built on his own site".
"His age must be below 70 and above 35."
"He must know the Mantrabrāhmaṇa", the collection of Vedic verses and ritual texts, and know it well enough to teach it.
He must be "well conversant with business" and "virtuous", with "honest earnings" and a "pure" mind.
There was a second door for learning. A man who owned only one-eighth of a vēli could still have his name put in, if he had learnt one Veda and one of the four bhāṣyas, the classical commentaries.¹ ⁸ Learning could stand in for half the land. Nothing could stand in for being a man of the right birth in the right village.
Who was kept out?
The disqualifications are where the rules show what the village feared. A man could not stand if he had served on any committee in the previous three years.¹ That is rotation. He could not stand if he had served on a committee and "has not submitted his accounts". Neither could "all his relations specified below".¹
The wall then names the relatives one by one: cousins through the mother's sisters, the mother's brother, uncles on the father's side, in-laws, the man's own father and son.¹ ⁸ A man barred for hiding the accounts could not sit on the committee through a cousin or a brother-in-law. The rule-makers, I think, had met that trick before. The cost fell on men who had done nothing wrong themselves. Every rule against nepotism faces the same tension, between stopping office passing to a proxy and punishing a man for his relative's failure.
Then come the moral bars. The rules exclude a man against whom incest or the first four of the "five great sins" of the law books were recorded, and a man outcast "for association (with low people)" until he had performed expiation.¹ They exclude the "foolhardy", a man who had stolen "property of others", and "one who has ridden on an ass and one who has committed forgery".¹ Riding an ass, I take it, means being paraded on one as a punishment. Some of these bars lasted "to the end of their lives", even after expiation.¹
One more line matters, because it is often misquoted. "When one who is on the committees is found guilty of (any) offence he shall be removed (at once)."¹ That is removal for wrongdoing. It is not a recall, in which voters vote a member out. The wall does not say who judged the offence.
How the committees were made
The later rules describe a draw in two stages. Each ward's names were tied into one bundle with a covering ticket. The bundles went into the pot, one was drawn, and from it a single ticket. Venkayya's translation itself marks with a question mark whether the boy drew the second time as well.¹ The arbitrator received the ticket with his palm open and read it out. One man was chosen in this way for each of the thirty wards.
From those thirty, the men who had already served on the garden and tank committees, and who were "advanced in learning" and "advanced in age", formed the annual committee of twelve. The Tamil name usually given is saṃvatsara-vāriyam, vāriyam meaning a committee.⁸ Twelve more formed the garden committee and the last six the tank committee. A further draw of thirty filled two smaller committees of six each, one called the pañcavāra committee and one for gold.¹ Every member was to "hold office for full three hundred and sixty days and (then) retire".¹
The village accountant, whom Venkayya calls the "arbitrator", had his own rule. Only a man "who possesses honest earnings shall write the accounts (of the village)". He had to hand his accounts to "the great men of the big committee", and be declared honest, before he could hold that post again.¹ The procedure was meant to last "as long as the moon and the sun (endure)".¹
A royal officer sat with the assembly while the rules were fixed, "in accordance with the (royal) command". The wall names him, Karañjai Koṇḍaya-kramavitta-bhaṭṭan, also called Sōmāśiperumān.¹
A franchise no republic could adopt
Before anything here is praised, one fact has to stand at the front. This was an assembly of landholding Brahmin men in a Brahmin village. Every eligibility rule, from land to house to Vedic learning, kept out the people of that settlement who were not landholding Brahmin men: women, tenants, labourers, artisans, and every caste that did not own land there. The rules turned existing privilege into formal eligibility. No modern republic could adopt them, and none should.
The mechanisms are a different matter. Take away the franchise and four working parts are left:
selection by lot, which takes ambition and campaigning out of office
rotation, one year in office and three years out, which stops a man from settling in
accounts rendered as a condition of standing again
a ban on named relatives, which closes the easiest way around the other rules.
Why choose by lot and not by vote?
Venkayya, writing in 1908, called "casting lots" a method "followed by all primitive communities".¹ Ancient Athens, which Europe has long treated as the cradle of democracy, would have disagreed. Aristotle, describing the constitutions of his day, wrote that "the appointment of magistrates by lot is thought to be democratical, and the election of them oligarchical".¹⁰ An election rewards the man with the most friends, money or fame. A lottery among the qualified gives every qualified man the same chance and makes bribing the voters pointless.
Uttaramērūr's lottery was democratic only inside a very small room. Inside that room, it did what Aristotle said a lottery does. It made office hard to buy. Outside the room, it could change nothing. No lottery, however fair, can correct an injustice in the list of names it draws from.
Why the accounts rule matters most
Of the four working parts, the accounts rule is the one that lets an institution learn. A committee spends money. The next committee, and the assembly behind it, can only judge that spending if it is written down and shown. A man who will not show it cannot stand again, and nor can his family. The account is how the village could find out what its last tank committee did with the tank's money. Engineers call this feedback. Without it, a committee can fail for years and nobody can prove it.
An account is not proof that an institution worked, though. It is where judging it begins. A tank can have honest accounts and still fail to hold water. And an account is useful only if someone can read it, question it and act on what it shows. At Uttaramērūr, the accountant's books went to "the great men of the big committee", a few men from the same small circle.¹
Why carve rules on a wall?
Writing a rule where everyone can see it is itself a technology of government. It fixes memory, so the rule does not drift each time an elder retells it. It limits what a powerful man can later claim the rule said. And it makes a change visible, because a rule that was never recorded cannot be seen to have changed.
The last point is the hinge of this volume. Because both sets of rules were cut in stone, we can still lay them side by side.
What the wall does not say
Some popular claims about Uttaramērūr go beyond the stone, and they should be set aside. The inscriptions are not the "oldest constitution" in any sense a lawyer would use. They are rules for choosing committees in one village assembly, made under a king.
They do not provide for voters to recall members. They do not ban offenders' families for seven generations. They name particular relatives. Voters did not write their preferred candidate on a palm leaf in secret. The people of a ward put forward eligible names, and chance chose among them. A careful summary on Wikipedia says a man found guilty of an offence was "removed immediately", which is close to the wall. Retold as a recall vote, it is not.⁹
Was Uttaramērūr a democracy, then? Not in any sense a modern reader means by the word. It was a self-governing assembly of a privileged few, with unusually careful rules against its own members' greed.
Part Eighteen — Why Did the Same Village Rewrite Its Rules Within Two Years?
The wall does not say why. It does let us see exactly what changed, and that is rarer than it sounds for a document more than a thousand years old.
The earlier rules, from the twelfth year, about 919 CE, cover the same ground: thirty wards, tickets in a pot, a young boy to draw them, annual, garden and tank committees, a year in office.¹ ³ Put side by side with the rules of about 921, five differences stand out.
Age. The earlier rules wanted men "below the age of sixty and above thirty". The later ones say "below 70 and above 35".¹
Learning. The earlier rules asked for a man "learned in the Vedas and śāstras" and "conversant with business". The later ones name a text, the Mantrabrāhmaṇa, demand that he can teach it, and add the one-eighth vēli route for a scholar with little land.¹
The ban on relatives. The earlier rules barred "close relations of the great men (just) retired from the committees", all of them.¹ The later rules bar the relatives only of men who had failed to submit their accounts, and they list those relatives by name.¹
New bars and new duties. The later rules add the list of sins and crimes, the lifelong bars, removal for an offence, the rules for the accountant, and the two-stage draw before the priests.¹
A different royal officer. In the twelfth year, the wall names the officer as Tattanūr Mūvēntavēḷāṉ.³ That is all the stone says about him. Venkayya, reading the title, called him "one of his Sudra officers".¹ That caste label is Venkayya's reading, not the wall's. In the fourteenth year the officer was Karañjai Koṇḍaya-kramavitta-bhaṭṭan, whose name, in my reading, suggests a Brahmin scholar.¹
Even the stone was mended. One syllable of the second officer's name "appears to be written over an erasure", Venkayya noted.¹ That was a mason's slip, not a rule revised.
What we know, what we can guess, what we cannot know
The evidence needs to be kept in separate boxes.
What we know. Two sets of rules, two years apart, cover the same ground differently, in the same village, on the walls of the same temple.¹ ³
What we can reasonably infer. The later rules aim the ban at a specific failure, unaccounted money, where the earlier rules struck at a whole group of recent office-holders and their families. Venkayya saw in both texts "the implied indignation against the committee members who had just then vacated office". He went further. They "must have embezzled communal funds and would not submit themselves to any sort of scrutiny".¹
He read the narrower ban of 921 as better judgment, after a first rule that was harsh.¹ I think his reading is the simplest one. It is still his inference, and now mine.
What we cannot know from the inscriptions. Why the rules changed. The later rules never mention the earlier ones. The change could have come from a quarrel between families, a new royal officer with different ideas, or a king's preference, as easily as from the experience of a rule that did not work.
What would settle it. A record of the dispute, the embezzlement or the shortage of candidates that came before the revision. As far as I know, none survives.
So the wall is evidence of revision, not proof of learning. Learning would need a problem seen, a change made because of it, and a better result after. The wall shows only the change.
Does this prove India had a habit of reform?
No. It proves one well-recorded revision in one village. Once is not a pattern. The temple walls at Uttaramērūr carry many other inscriptions, and I have not read them for this purpose. A historian who found the same assembly revising its rules again, decades later, would turn one case into the beginning of a pattern. Until then, the honest claim is narrow. On this soil, at least once, people wrote rules for an institution, lived with them, and rewrote them.
The lotus meets the wall
The series took its title from a word. Viksit, the Hindi word in "Viksit Bharat", comes from the Sanskrit vikasita, which the great Sanskrit dictionary of Monier Monier-Williams glosses as "opened, expanded, blown", the ordinary word for a flower that has opened, above all the lotus.¹¹ The first volume of this series tested an image built on that word. A lotus that opens has not had petals fixed onto it. What appears when it opens was folded in the bud.
Read as a claim about a nation, the image said that India would develop by opening out what it already holds, fed from outside, and not by having another country's shape fixed onto it. Its opposite was a country assembled from parts made elsewhere.
I held to that opposition when this series began. The wall at Uttaramērūr does not fit it. The village institution was not a custom that grew by itself. It was drafted clause by clause, with an eligibility test, a list of exclusions and a procedure for drawing lots. Within two years it was drafted again. Nothing on that wall opened from a bud. Hands made it, and hands remade it.
Then there is the lotus itself. It roots in the mud of shallow wetlands, in ponds, lakes, marshes and floodplains.¹² In the Tamil country, where few rivers run all year, I think much of the still water a lotus could grow in is water that people have held back on purpose. I have no count to prove it. If I am right, a lotus on this soil often grows in a tank someone dug, and a committee kept clean.
Design, contract and revision are not the same thing
Other Indian institutions were also made, not grown, but in different ways. Kauṭilya's revenue system was designed. His treatise names forty ways an official can steal and a response to each.¹³ The endowment recorded in a cave at Nāsik was contracted. Money was deposited with named guilds of weavers, at fixed interest, on stated terms.¹⁴ Only Uttaramērūr shows, on the evidence of its own walls, an arrangement that was revised. None of the three is an essence unfolding from inside. Each is people building an arrangement on purpose and writing it down.
What a Harvard idea of reform has to do with a Cōḻa village
Three development economists, Matt Andrews, Lant Pritchett and Michael Woolcock, argue that the capability of a state is not inherited from any civilisation. It is built. Their book of 2017 calls the method "problem-driven iterative adaptation": name a local problem, try a solution, watch what happens, adjust, and try again.¹⁵ On their account, a country's ancient texts help its officials about as much as its cuisine does.
I think they are right about the method. Their idea is a modern one, and I use it here as a comparison that helps us see the wall, not to claim the village practised it early. We do not know how the second set of rules was argued out. The authors also say reform needs room to move, which they call the "change space". It needs authority, acceptance by those affected, and the ability to do the work.¹⁶ In 921 the authority came from a royal letter, the acceptance from the assembly, and the ability from men who could read, count and keep accounts.
The likeness ends at the village gate, though. The economists describe reform by officials who are allowed to fail and learn. Uttaramērūr shows rule-writing under a king's supervision, by a privileged assembly, for its own members. And a republic adds a limit the village never had. Trying, failing and adjusting must happen inside citizens' rights. No experiment may suspend them.
What "blossomed" should mean now
I give up the opposition I started with, then, and I would rather say so plainly than hide it. The choice for India was never between blossoming and assembly. Every institution that works was assembled by someone. The real choice is between two kinds of assembly. One is a copy that nobody on the spot can argue with, amend or own. The other is an assembly that the people who inherit it keep testing, maintaining and rebuilding until it works in their hands.
Assembly comes first. Blossoming is what happens when an institution is reproduced, used and corrected by people who did not build it.
Uttaramērūr shows the correcting, by the same generation, inside two years. It does not show the reproducing, across generations, by people who did not draft the rules. A thing is not viksit, on this reading, because it grew without hands. It is viksit when what hands built is kept, used and changed by hands that came later.
Part Nineteen — Ambedkar Called the Village "a Sink of Localism". Why Did India Build Panchayats Anyway?
Because Ambedkar was right about the village, and the Republic still needed local government. The answer it found took four decades and is not finished.
On 4 November 1948, moving the Draft Constitution in the Constituent Assembly, B. R. Ambedkar answered members who wanted the Republic built on village panchayats. He quoted the British administrator Charles Metcalfe, who had praised Indian village communities as "little republics having nearly everything that they want within themselves", surviving every change of ruler. Ambedkar granted the point and refused its value.¹⁸
"No doubt the village communities have lasted where nothing else lasts. … But mere survival has no value. … I hold that these village republics have been the ruination of India. … What is the village but a sink of localism, a den of ignorance, narrow-mindedness and communalism? I am glad that the Draft Constitution has discarded the village and adopted the individual as its unit."¹⁷ ¹⁸
Uttaramērūr fits his words. Its assembly was a body of landholding Brahmins, and it would not have admitted a man of Ambedkar's birth to the pot. His objection has to be conceded, not managed. The village as a political unit, with a franchise of birth and land, deserved to be refused.
Didn't Ambedkar reject village government altogether?
He rejected the village as the unit of the Constitution. He did not block local government. Eighteen days after his speech, on 22 November 1948, K. Santhanam moved a new article: "The State shall take steps to organise village panchayats and endow them with such powers and authority as may be necessary to enable them to function as units of self-government." Ambedkar replied, "Sir, I accept the amendment," and later, "I have nothing more to add." The Assembly adopted it.¹⁹ It became Article 40.²⁰ Accepting the amendment did not mean he withdrew his critique. The village of Article 40 is not sovereign. It sits inside the Constitution, below its rights.
The true sequence has three steps. First came the refusal. The individual, not the village, became the Constitution's unit. Then came a weak instruction. Article 40 sits among the Directive Principles, which by the Constitution's own terms no court can enforce.²⁰ For four decades the village panchayat existed in the Constitution as an aspiration, and states did with it what they chose.
Parliament later wrote down what had gone wrong. The 73rd Amendment's own Statement of Objects says panchayats had not become viable and responsive bodies. Among the reasons it gives are the "absence of regular elections" and the "lack of financial resources".²¹
The panchayat of 1992 was built on Ambedkar's unit
In 1992 Parliament passed the 73rd and 74th Amendments.²¹ ²² They gave village panchayats and urban municipalities a place in the Constitution and required regular elections. They reserved seats for Scheduled Castes and Scheduled Tribes in proportion to their population, and reserved "not less than one-third" of seats for women.²¹ The Eleventh Schedule listed twenty-nine subjects, from drinking water to minor irrigation, that states could hand to panchayats.²¹
The Republic did not return to the village Ambedkar rejected. It rebuilt a local body on the unit he chose, the individual citizen, with a vote for every adult. It then wrote in, from the start, the people the old assemblies had kept out. The panchayat is a new construction, not a restored Cōḻa body. It shares a few methods with the old assembly and almost nothing of its franchise or its source of authority.
The Act also created the Gram Sabha, a body of every person on the electoral roll of the village.²¹ It is the old assembly's counterpart with the opposite door. The sabhā at Uttaramērūr admitted a few landholding men. The Gram Sabha admits every adult on the roll. In principle, any voter there can question what the panchayat did with its money. What it may actually do, the Constitution leaves to state law.²¹
Have panchayats worked?
The form has arrived. The power has arrived only in part. Article 243G says a state legislature "may" endow panchayats with powers.²¹ Listing twenty-nine subjects in a schedule is not handing them over. The test is whether a panchayat has the funds, the functions, and the functionaries, the staff who carry out its decisions. A study for the Ministry of Panchayati Raj, released in 2025, scored the states out of 100. The national average for functions was about 29, down from about 35 in 2013–14.³² A panchayat with reserved seats and no budget is a form waiting for its capability.
The women's reservation shows both halves at once. By the ministry's count in 2025, India has over 15 lakh elected women in panchayats, about 46.6 per cent of all elected members.²³ That is presence. Power is a separate question. A committee of the Ministry of Panchayati Raj, reporting in February 2025, studied the "Pradhan Pati", also called the sarpanch-pati, the husband or male relative who attends meetings and signs papers for an elected woman. It noted that the term "does not only mean husbands, but includes male relatives exercising real political power".²³
Where women governed, the results were measurable. Raghabendra Chattopadhyay and Esther Duflo studied 265 village councils in West Bengal and Rajasthan, where seats for council heads had been reserved for women by rotation, close to a natural experiment. Women in both states complained more often than men about drinking water. Councils led by women invested more in water.²⁴ A woman in the chair changed what the village built.
Later work by Lori Beaman and colleagues found that after ten years of reservation, women were more likely to stand for and win seats that were not reserved.³³ In villages that had twice had a woman as head, the gap between the hopes parents held for daughters and for sons narrowed, and girls' schooling caught up with boys'.³⁴
Gender is not the only line, though. A Dalit woman in the chair can face obstruction that a woman of the dominant caste does not. In 2020, in Cuddalore district of Tamil Nadu, a Dalit woman panchayat president said she had been made to sit on the floor at meetings while others sat on chairs. Police registered a case under the law against atrocities on Scheduled Castes.³⁵ Being elected, being present and being able to govern are three different things.
The direction is the one this series argues for. The arrival is not yet. A body that does not control its money, and cannot be made to show its accounts, does not stay a body for long.
The Republic's own remembered tradition
By the definition this volume reached, some of India's most important inheritances are not ancient at all. The Constitution was assembled from Indian argument, British statute and foreign models. The government's official text of 2024 counts 106 amendments, the last made in 2023.²⁵ The count proves only that it can be changed. Some amendments opened it to more people, and some bent it.
The Election Commission, set up on 25 January 1950, has run national elections through many changes of government.²⁶ The money between the Union and the states has been renegotiated by Finance Commissions since the first reported in 1952.²⁷ Direct transfer of benefits into bank accounts began in 43 districts on 1 January 2013 and is now routine.²⁸
Each was assembled. Each has outlived, or is outliving, the people who began it. Each is being corrected in use. By a modern analogy, and the analogy is mine, they are the Republic's smṛti, the "remembered" tradition that in classical India stood beside the revealed texts. Classical smṛti was a body of authoritative writing, and these are institutions. The likeness lies in the remembering and the revising. An account of India's inheritance that stopped at 1947 would lose the argument to the museum-keeper on dates alone.
Ending — Who Drank? The Question the Wall Never Answers
There is one thing the walls at Uttaramērūr do not record, and it is the thing that mattered most to most of the people who lived near that tank.
The rules say who could sit on the tank committee, six men a year, drawn from landholding Brahmin families who had passed every test of land, learning, age and conduct. They do not say who could draw water from the tank. They do not say whose fields the sluice fed first in a dry year. They do not say who carried the silt out of the bed each summer, basket by basket, on their heads.¹
The labourers, tenants and families who lived beyond the Brahmin streets do not appear in these rules at all. Their absence is not evidence that they were denied water. The wall is silent, and silence is not proof. It is evidence of who designed the rules, and of who had no say when the rules were rewritten.
The boy in the temple could draw a name without knowing whose it was. That guarded the pot against favour. It could not put one new name into the pot.
A tank can be well designed, well kept and corrected in use, and still be kept for some and not for others. A thousand years after that pot was lifted in the temple, at another tank, in another language, people would ask out loud the question the wall leaves blank. Who may drink?
Words to keep
bhāṣya, a classical commentary on a sacred text. Here, one of the four commentaries a learned candidate could know.
brahmadeya, a village granted by a king to Brahmin landholders, governed by their own assembly.
caturvēdimaṅgalam, a Brahmin settlement whose members were learned in the four Vedas. Uttaramērūr's formal name used it.
ēri, a tank, an earthen bund across a slope that holds rainwater for irrigation.
ēri-vāriyam, the tank committee of a village assembly.
erippatti (ēri-patti), "tank-land", land whose income was set apart for keeping a tank in repair.
kuḍavōlai, "pot-leaf", the palm-leaf ticket put into a pot for the drawing of lots.
kuḍimarāmattu, repair work by the people of the place, especially shared labour to desilt and mend a tank.
kuḍumbu, a ward of the village. Uttaramērūr had thirty.
madhyastha, the arbitrator or accountant who received the drawn ticket and kept the village accounts.
Mantrabrāhmaṇa, a Vedic text of verses and ritual instructions. The later rules required a candidate to know it well enough to teach it.
nagaram, a body of merchants governing a trading town in the Cōḻa period.
sabhā, the assembly of a Brahmin village.
saṃvatsara-vāriyam, the annual committee, the senior committee of the assembly.
smṛti, "what is remembered", the body of tradition that stood beside the revealed texts.
ūr, the assembly of a village that was not a brahmadeya.
vāriyam, a committee of the village assembly.
vēli, a Tamil measure of land, about 2.5 hectares by one estimate. A quarter vēli was roughly an acre and a half.
vikasita, Sanskrit for "opened, expanded, blown", the word behind viksit.
Sources
V. Venkayya, "Two Inscriptions at Uttaramallūr", in John Marshall (ed.), Archaeological Survey of India, Annual Report 1904–05 (Calcutta: Superintendent of Government Printing, 1908), pp. 131–145. Inscription A (year 12) edited and translated at pp. 136–139, Inscription B (year 14) at pp. 139–145. Scan and searchable text: https://archive.org/details/in.ernet.dli.2015.35485
Emmanuel Francis (ed.), "Uttiramērūr, Vaikuṇṭha-Perumāḷ, time of Parakesarivarman, year 14", DHARMA project (ERC no. 809994), draft edition 2024, with bibliography and dating (920/921 CE after Venkayya, 921 CE after Mahalingam 1989): https://erc-dharma.github.io/tfa-uttiramerur-epigraphy/texts/htmloutput/DHARMA_INSUttiramerur00002.html
DHARMA project, "Uttiramērūr, Vaikuṇṭha-Perumāḷ, time of Parakesarivarman, year 12" (918/919 CE after Venkayya, 919 CE after Mahalingam 1989), with summary naming Tattanūr Mūvēntavēḷāṉ and the three committees: https://erc-dharma.github.io/tfa-uttiramerur-epigraphy/texts/htmloutput/DHARMA_INSUttiramerur00001.html
South Indian Inscriptions, vol. 3, no. 93, Nerkunram, 24th year of Rājakesarivarman (Āditya I), English translation as reproduced at: https://www.whatisindia.com/inscriptions/south_indian_inscriptions/volume_3/no_93_94_rajakesarivarman_aditya_i.html
Press Trust of India, "TN revives traditional participatory water resource mgmt", Business Standard, 13 March 2017: https://www.business-standard.com/article/pti-stories/tn-revives-traditional-participatory-water-resource-mgmt-117031300450_1.html
R. Sakthivadivel and Manisha Shah, "Will Kudimaramathu Make Communities Think Tanks Again? A study of tanks in transit, coping mechanism of communities and government action", IWMI-Tata Water Policy Research Highlight 06 (2018): https://cgspace.cgiar.org/rest/bitstreams/b8e1a3e0-8dd6-417c-aba3-df4742a23a13/retrieve
R. Lenin, "Kudimaramathu fails ryots", Deccan Chronicle, 30 August 2018: https://www.deccanchronicle.com/nation/current-affairs/300818/kudimaramathu-fails-ryots.html
K. A. Nilakanta Sastri, Studies in Cōḷa History and Administration (Madras: University of Madras, 1932), ch. VI and Appendix II, pp. 131–175 (Tamil terms and the list of relatives, consulted through secondary accounts, see How this was written): https://archive.org/download/K.A.NilakantaSastriBooks
"Uthiramerur", Wikipedia (quoted as an example of a careful popular summary of the rules, not as an authority), accessed October 2026: https://en.wikipedia.org/wiki/Uthiramerur
Aristotle, Politics, Book IV, ch. 9 (1294b), trans. Benjamin Jowett: https://web.mit.edu/classics/Aristotle/politics.4.four.html
Monier Monier-Williams, A Sanskrit-English Dictionary (Oxford, 1899), s.v. vi-kas, vikasita.
National Parks Board, Singapore, Flora & Fauna Web, "Nelumbo nucifera" (habitat): https://www.nparks.gov.sg/florafaunaweb/flora/2/2/2257
Kauṭilya, Arthaśāstra 2.8, ed. and trans. R. P. Kangle (Bombay: University of Bombay, 1960–65).
É. Senart, "The Inscriptions in the Caves at Nasik", Epigraphia Indica vol. 8 (1905–06), Nasik no. 12.
Matt Andrews, Lant Pritchett and Michael Woolcock, Building State Capability: Evidence, Analysis, Action (Oxford University Press, 2017), Introduction p. 5 and ch. 7 (open access): https://library.oapen.org/bitstream/20.500.12657/31857/1/624551.pdf
Duncan Green, "Building State Capability: Review of an important (and practical) new book", Building State Capability programme, Harvard Kennedy School, 5 April 2017: https://bsc.hks.harvard.edu/2017/04/05/building-state-capability-review-of-an-important-and-practical-new-book
B. R. Ambedkar, Constituent Assembly Debates, vol. VII, 4 November 1948, the passage ending "adopted the individual as its unit": https://www.constitutionofindia.net/debate_paragraphs/357708/
Constituent Assembly Debates, a member's speech reproducing Ambedkar's words of 4 November 1948 on Metcalfe's "little republics": https://www.constitutionofindia.net/?p=134542
Constituent Assembly Debates, vol. VII, 22 November 1948, K. Santhanam's amendment inserting article 31-A and Ambedkar's acceptance: https://www.constitutionofindia.net/debates/22-nov-1948/
"Article 40: Organisation of village panchayats", Constitution of India and draft history, Centre for Law and Policy Research: https://www.constitutionofindia.net/articles/article-40-organisation-of-village-panchayats/
The Constitution (Seventy-third Amendment) Act, 1992, with Statement of Objects and Reasons, Articles 243(b) and 243A (Gram Sabha), 243D, 243G and the Eleventh Schedule, Legislative Department, Government of India: https://cdnbbsr.s3waas.gov.in/s380537a945c7aaa788ccfcdf1b99b5d8f/uploads/2023/03/2023030272.pdf
Ministry of Housing and Urban Affairs, Government of India, "The Constitution (74th Amendment) Act, 1992" (explanatory note on the Act, with the one-third reservation for women in municipalities): https://mohua.gov.in/upload/uploadfiles/files/74th_CAA13.pdf
Ministry of Panchayati Raj, Report of the Advisory Committee on Transforming Women's Representation and Role in Panchayati Raj Institutions: Efforts to Eliminate Proxy Participation, vol. I (chair: Sushil Kumar), 5 February 2025: https://cdnbbsr.s3waas.gov.in/s316026d60ff9b54410b3435b403afd226/uploads/2025/09/202509241727574699.pdf
Raghabendra Chattopadhyay and Esther Duflo, "Women as Policy Makers: Evidence from a Randomized Policy Experiment in India", Econometrica 72:5 (2004), pp. 1409–1443: https://econpapers.repec.org/RePEc:ecm:emetrp:v:72:y:2004:i:5:p:1409-1443 (full text: https://www.povertyactionlab.org/sites/default/files/publications/65_Duflo_Women_as_Policy_Makers.pdf)
Legislative Department, Ministry of Law and Justice, The Constitution of India [as on 1st May, 2024], preface ("all the amendments up to the Constitution (One Hundred and Sixth Amendment) Act, 2023"): https://lddashboard.legislative.gov.in/sites/default/files/coi/COI_2024.pdf
Office of the District Election Officer, Kolkata South, Government of West Bengal, "National Voters' Day" ("to mark the foundation of the Election Commission of India, i.e. 25th January 1950"): https://deokolkatasouth.wb.gov.in/nvd
Finance Commission of India, "Previous Commission Reports" (First Finance Commission report, 1952): https://fincomindia.nic.in/ and Report of the First Finance Commission, 1952 (chair K. C. Neogy), GIPE digital library: https://dspace.gipe.ac.in/xmlui/handle/10973/41353
Press Information Bureau, press note on Direct Benefit Transfer ("will be rolled out from 1 January 2013 in 43 identified districts"), December 2012, as hosted by the Department of Financial Services: https://financialservices.gov.in/sites/default/files/order-notices/PIB%2520Press%2520Note.pdf
Elinor Ostrom, "Beyond Markets and States: Polycentric Governance of Complex Economic Systems" (Nobel prize lecture), American Economic Review 100:3 (2010), pp. 641–672, drawing on her Governing the Commons (Cambridge University Press, 1990): https://www.aeaweb.org/articles?id=10.1257/aer.100.3.641
Michael Cox, Gwen Arnold and Sergio Villamayor Tomás, "A Review of Design Principles for Community-based Natural Resource Management", Ecology and Society 15:4 (2010), art. 38 (Ostrom's principles of boundaries, local fit, collective choice, monitoring and graduated sanctions): https://www.ecologyandsociety.org/vol15/iss4/art38/main.html
Y. Subbarayalu, interviewed by Parvathi Menon, "Reconstructing Chola Agrarian History", Review of Agrarian Studies 13:1 (2023), on the sabhā of brahmadeya villages, the ūr of non-Brahmin villages, the nagaram, and the vēli as about 2.5 hectares: https://ras.org.in/reconstructing_chola_agrarian_history
V. N. Alok (Indian Institute of Public Administration), Status of Devolution to Panchayats in States: An Indicative Evidence Based Ranking, summary, study for the Ministry of Panchayati Raj (2024, released February 2025), Table 1 and p. 27 (functions sub-index 29.18, down from 35.34 in 2013–14): https://static.pib.gov.in/WriteReadData/specificdocs/documents/2025/feb/doc2025213501601.pdf
Lori Beaman, Raghabendra Chattopadhyay, Esther Duflo, Rohini Pande and Petia Topalova, "Powerful Women: Does Exposure Reduce Bias?", Quarterly Journal of Economics 124:4 (2009), pp. 1497–1540: https://www.nber.org/papers/w14198
Lori Beaman, Esther Duflo, Rohini Pande and Petia Topalova, "Female Leadership Raises Aspirations and Educational Attainment for Girls: A Policy Experiment in India", Science 335:6068 (2012), pp. 582–586: https://dspace.mit.edu/handle/1721.1/82636
"Woman Dalit panchayat chief in TN forced to sit on floor, barred from hoisting flag", Onmanorama, October 2020 (Therku Thittai, Cuddalore district): https://www.onmanorama.com/news/india/2020/10/11/woman-dalit-panchayat-chief-in-tn-forced-to-sit-on-floor-barred-.html
How this was written
I read V. Venkayya's 1908 translation of both Uttaramērūr inscriptions in the searchable scan of the Archaeological Survey report, and checked dates, temple, officers and committee names against the DHARMA project's modern editions of the Tamil text, which online give only the opening lines. The Constituent Assembly passages, the 73rd Amendment Act, the 2018 IWMI study, the 2025 ministry report, the 2024 devolution study and the papers by Chattopadhyay, Duflo, Beaman and colleagues were read at the links given.
I could not open the relevant pages of K. A. Nilakanta Sastri's 1932 study. The Tamil name of the ward, the one Veda and one bhāṣya clause, and the full list of barred relatives come through his work as summarised by others, and agree with the parts of Venkayya I could read. Ostrom's 1990 book and the full text of her 2010 lecture were not opened. Her principles come through a 2010 review of them. Wikipedia is cited once, only as an example of a popular summary.
My own readings, marked in the text: the purposes of the lifted pot, the boy and the open palm, "ridden on an ass" as a punishment, the Brahmin reading of the second officer's name, the guess that much still water in the Tamil country is held back by people, the redefinition of "blossoming", and the smṛti analogy. Ostrom's principles are used as a lens, not as a claim about what the village knew.
Three findings would change the conclusion: a record explaining why the rules changed in 921, later inscriptions showing the same assembly revising its rules again, or evidence that the tank committee governed who could draw water.
No imagined person appears in this volume.
Comments