Conceptual illustration of water storage and fields; not a historical site reconstruction

Water, work and the agrarian landscape

Eri Tanks and Water Management

Behind the still surface of an eri lies a moving history: rainfall gathered, water released, labour organised and rights negotiated.

My Tamil Nadu Editorial DeskReviewed 17 November 2026
Editorial visual · My Tamil Nadu editorial team · Original project SVG; schematic, not documentary evidence
LandscapeStorage, channels and cultivated land
Close readingUkkal inscription, SII III no. 10
InstitutionsAssemblies, patrons and workers
Key distinctionA dated repair is not a foundation date

Orientation

Overview

Medieval records, colonial changes and modern comparisons

An eri is often translated as an irrigation tank: a reservoir that stores water for use beyond the moment when it arrives. On a map it may appear as a single blue shape. Historically, that outline is only the centre of a much larger arrangement. The land feeding it, the outlet releasing water and the fields receiving it belong in the same story. So do the people whose work kept those connections open.

This guide follows those connections through physical features and written records. A medieval land sale at Ukkal offers an unusually concrete starting point; academic studies then widen the view to village institutions and colonial changes. Present-day research helps explain how tanks function, but it cannot establish what every tank looked like a thousand years ago. The main water and everyday-life guide places this closer study beside agriculture, settlement and household work.

Read the landscape

How the parts of an eri connect

The water surface is only one part of the system. Follow supply, storage and the two different routes out.

1

Catchment & supply

Rainfall runoff or a supply channel brings water towards the tank.

2

Stored water

A bund holds back water. Storage changes with inflow, use, seepage and evaporation.

3A

Sluice → field channel

A controlled outlet releases irrigation water towards the cultivated area.

3B

Surplus outlet → downstream

A separate overflow route carries excess water away when the storage threshold is reached.

Conceptual relationships only. This is not a surveyed tank, a construction plan or a claim that every historical eri had the same layout. A downstream connection need not lead to another tank.

Diagram background: FAO: water-harvesting structures and water sources and availability. These explain physical relationships; the historical evidence is discussed below.

Chronology

A careful timeline

These entries place crop evidence, recorded institutions and later studies in time. They do not establish a single origin date for farming or irrigation.

  1. A dispersed written record

    The Pudukkottai archive describes sluice inscriptions spanning the sixth to sixteenth centuries; this is not a single date of invention.[7]

  2. Ukkal’s tank boats

    SII III no. 10 records a land transaction supporting two boats assigned to a village tank.[1]

  3. Changing administrative claims

    Colonial discussions of village maintenance accompanied expanding state claims over irrigation.[9]

  4. Tanks within connected landscapes

    Recent research examines cascades, channels, groundwater and institutions together.[5][6]

Historical reading

Follow the water beyond the tank

A catchment is the surrounding area from which water reaches a tank. An inlet or supply channel may bring that flow into storage. The bund is the embankment retaining it. A sluice provides a controlled outlet towards irrigation channels; a surplus outlet carries excess water away. These features have different jobs, even when a small diagram places them close together. Their dimensions and arrangement must be established separately at each site.

The ayacut is the area served by an irrigation work. A field within it still depends on the route from outlet to plot. A visible expanse of stored water therefore tells us little, by itself, about whether a particular cultivator received water when needed. The distinction between storage and delivery is the reason the diagram extends into the fields instead of stopping at the bund.

Modern observations at Mailam record damage or obstruction across several parts of this chain: supply channels, sluices, surplus structures and field channels. The study also describes a shift towards groundwater use. Those observations show why a tank cannot be assessed solely by its surface area. They are evidence about the surveyed landscape, rather than a condition report on all Tamil Nadu tanks.[6]

Historical reading

Different landscapes, different connections

Tanks might collect local runoff or receive water through a wider supply network. Some operate within cascades, where water passing from an upper tank contributes to a lower one. The 2023 TNAU–NABARD study examined 315 cascades containing 2,013 tanks across fifteen districts. Its scale makes an important point without requiring every tank to fit one diagram: connections between reservoirs can be central to their performance.[5]

A cascade also changes the historical questions. Were linked works established together, or connected through later alterations? Did different settlements share responsibility for the intervening channel? Could an upstream change affect downstream cultivation? A present-day map can identify relationships to investigate; it cannot supply their dates. Nor does an orderly sequence of water bodies prove that one ruler planned the whole watershed.

River irrigation, tanks and wells should not be arranged as successive stages through which the entire region passed. Research in Karur and Tiruchirappalli distinguishes villages dominated by canals from others combining tanks, wells and rainfed cultivation. That modern regional contrast cautions against treating the Kaveri delta as the model for every Tamil agricultural landscape.[4]

Historical reading

Why a king’s name is only part of the date

Royal support mattered. IGNOU’s account of medieval urbanisation associates Rajendra I with the Cholaganga tank and its canal network, while discussing irrigation in the growth of settlements. Large projects could express authority as well as supply agriculture. They belong beside the political history of the Cholas, but they do not explain every small reservoir, channel or later repair.[8]

Dating a tank means separating several events that are easily collapsed in a short heritage label. The first impounding of water, an enlarged bund, a replacement sluice, a grant for upkeep and a modern rebuilding may belong to different centuries. A dated inscription can securely anchor the transaction it records without dating every adjacent stone. A ruler’s name preserved in a place-name is a useful clue to investigate, not an excavation result.

The archaeological water-management study hosted by Tamil Virtual Academy documents inscriptions on different hydraulic features and notes the loss or displacement of some examples. Its material directs attention to individual structures and their contexts. If an inscribed stone has moved, its connection to the present outlet needs checking before its date is used for that outlet.[10]

Historical reading

An inscription close-up: two boats at Ukkal

South Indian Inscriptions III, no. 10, published by Hultzsch in 1899 and digitised by DHARMA, comes from the mandapa’s north wall in Ukkal’s Vishnu temple. It gives Rajendra-Choladeva’s fourth regnal year; we retain that date without a calendar conversion.[1]

In paraphrase, the assembly sold land to support the maintenance of two boats assigned to the village tank by a royal servant. The deed describes the plot through a sluice channel, neighbouring land and other boundaries. It also records payment and the sale’s tax-free status. This is a documented arrangement for continuing support, not simply praise of a benefactor.[1]

What does that establish? Boats formed part of the tank’s recorded arrangements, and land income was committed to their upkeep. It does not establish their dimensions, design or daily journeys, or demonstrate an inland commercial ferry service. The published wording does not describe their operating routine. Those missing details should remain missing rather than being supplied by an attractive reconstruction.[1]

Historical reading

Who could speak for the village?

A second Ukkal inscription, SII III no. 11, includes members concerned with tank supervision among an assembly making an assignment for temple purposes. The edition dates it by the sixteenth year of a king titled Parakesarivarman; its note treats identification of the ruler as uncertain. The passage shows tank responsibility within a wider institutional setting. It does not provide a complete constitution or prove that every resident participated.[2]

Noboru Karashima distinguishes the ur of an ordinary village, the sabha of a Brahmana settlement, the nagaram of a town and the nadu of a locality. These names should not be translated into one uniform democratic village council. His study also warns that older accounts gave disproportionate attention to the election of sabha members and neglected other aspects of rural society.[3]

For a history of water, the practical question is who could turn a decision into an obligation. The donor, the body accepting a grant and the people performing the work need not have been the same group. An inscription naming the first two may leave the third largely anonymous. Reading it critically means preserving that unequal visibility without pretending we can recover every missing voice.

Historical reading

The work after construction

A foundation story has a clear beginning; maintenance has a recurring calendar. Waterways accumulate material, structures wear and changes upstream affect what arrives downstream. Keeping a system usable calls for decisions about labour, resources and timing. These continuing demands are easy to overlook when the surviving monument is a handsome sluice pillar or an inscription celebrating a gift.

At Ukkal, the provision for boat upkeep makes continuation visible in the record itself. Elsewhere, the most useful evidence may be a repair grant, an account or a survey identifying an altered channel. None should be reduced to a ceremonial list of generous patrons. The promised resource, the institution receiving it and the task it supported together explain how maintenance was meant to happen.

An account of collective work must also ask how obligations were distributed. Did the same people supply labour and receive irrigated land? Were workers paid, compensated through rights, or compelled? Where the cited record does not say, this guide leaves those questions open. “Community participation” is too broad an answer to substitute for the actual terms of an arrangement.

Historical reading

Stored water and unequal access

Karashima’s epigraphic approach studies changing landholding through groups of inscriptions rather than selecting one convenient example. Records can preserve taxation, boundaries, disputes and cultivation arrangements as well as donations. This gives irrigation history a social geography: the tank mattered within a landscape of differentiated rights, rather than an undivided village estate.[3]

The modern Tiruchi district study makes inequality more directly visible. In its tank-and-well villages, it describes land concentrated among intermediate castes and substantial agricultural wage labour by Scheduled Caste workers, alongside continuing discrimination. These are findings from particular surveyed communities. They should not be projected unchanged into Chola-period Ukkal, but they do show why a shared water body cannot be assumed to create equal social standing.[4]

A field near an outlet and one at the far end of a channel also occupy different physical positions. That does not by itself reveal the rules of allocation. Historical reconstruction needs both the route water could take and evidence for who was entitled to use it. Keeping those questions together makes room for cooperation and conflict within the same account.

Historical reading

Kudimaramat and the colonial search for tradition

David Mosse’s historical study examines how colonial officials made village maintenance, commonly discussed as kudimaramat, part of their irrigation administration. Nineteenth-century inquiries sought evidence of customary collective labour while the state expanded its own authority over water. His argument challenges the assumption that official descriptions merely recorded an intact, autonomous village institution inherited from a remote past.[9]

This changes how a colonial report should be read. Its description of an old custom may also be an argument about who ought to pay or work in the present. Official anxiety about disrepair can reveal administrative priorities without offering a neutral photograph of the preceding centuries. The vocabulary of tradition belongs to the history being investigated.

That does not erase local knowledge or the reality of shared work. It means identifying which practice, place and period a claim concerns. A medieval endowment, a nineteenth-century maintenance demand and a contemporary association can all concern the same water landscape while operating through different rights and expectations. Continuity in the reservoir’s name need not mean continuity in its government.

Historical reading

Seeing a tank as a historical landscape

The British Library’s EAP1293/1 collection documents sluice inscriptions in Pudukkottai district across sixty visited sites. Its catalogue places technology beside changes in cultivation, settlement and social organisation. This is a valuable route into local evidence because it directs attention away from famous capitals and towards structures embedded in working landscapes. The date span describes the collected material, not a construction sequence for every recorded tank.[7]

A useful site record would bring together the inscription’s location, a transcription, a translation and a plan showing its relation to the waterworks. Photographs can preserve details that later repairs obscure; older maps may locate channels no longer evident on the ground. These forms of documentation answer different questions. A good historical explanation makes their relationship visible instead of letting a single date stand for the whole landscape. A boundary named in a deed might be identifiable on a later map, for example, but that correspondence needs to be demonstrated through location and context rather than a similar sounding name.

Start with the water and everyday-life hub for the wider agricultural setting. The archaeology guide explains how material remains and written records are evaluated, while the Chola and colonial histories supply political context. For an individual tank, the most revealing story may be a modest one: a changed outlet, a recurring maintenance obligation or a disputed boundary that explains how cultivation was sustained.

Human scale

People and institutions

Named rulers are only one part of the history. These groups and individuals help show how evidence and social life connect.

Cultivators and agricultural workers

Access to land, irrigation and wages could place people in very different relationships to the same tank.

Assemblies and donors

Transactions could connect local institutions with royal servants and other patrons.

Recorders and researchers

Inscriptions, surveys and fieldwork preserve different parts of a changing landscape.

Historical geography

Connected places

Explore the landscapes and settlements discussed in this guide.

Visual reading

A schematic water landscape

These project-created visuals support orientation. They are not presented as photographs of an excavation, monument, person, or event.

Schematic landscape showing water storage and fields; not a reconstruction of a particular historical place
A connected landscapeOriginal diagrammatic illustration. The position of the water, channels and fields is conceptual and does not identify a site or period.My Tamil Nadu editorial team · Original project SVG; schematic, not documentary evidence

Questions

Frequently asked questions

Were all Tamil Nadu tanks built by the Cholas?

No. A tank’s history must be established individually, distinguishing foundations, enlargements, repairs and later administration.

Does a tank committee mean the whole village governed equally?

No. The institution, membership and rights need to be identified from the evidence; the word community does not establish equal participation.

Does a sluice inscription date the entire tank?

It dates the event or claim recorded, subject to the reading. The relationship between the inscribed feature and other construction phases requires separate investigation.

Editorial trust

Sources and further reading

Citations in the article point to this full list. Source type and review date are shown so readers can evaluate the evidence.

  1. 1
    South Indian Inscriptions III, no. 10: Ukkal, north wall of the mandapa (Hultzsch, 1899)Archaeological Survey of India; DHARMA digital edition by Emmanuel FrancisPublished primary inscription; Tamil edition, English translation and apparatus checked · Reviewed 17 November 2026
  2. 2
    South Indian Inscriptions III, no. 11: Ukkal (Hultzsch, 1899)Archaeological Survey of India; DHARMA digital editionPublished primary inscription; translation checked; regnal identification uncertain · Reviewed 17 November 2026
  3. 3
    Epigraphical Study of Ancient and Medieval Villages in the Tamil Country (2011)Noboru Karashima; Review of Agrarian StudiesAcademic epigraphic study; full article consulted · Reviewed 17 November 2026
  4. 4
    Designing Collective Action: Problems of Local Water Management in Tiruchi District, India (2011)Review of Agrarian StudiesAcademic longitudinal village study; methods and irrigation discussion checked · Reviewed 17 November 2026
  5. 5
    Rainfall and Tank Storage: Relooking the Tank Performance with Time Tested Tank Storage Pattern Using Tank Cascade Approach in Tamil Nadu (2023)Tamil Nadu Agricultural University / NABARDInstitutional field research; executive summary and study scope checked · Reviewed 17 November 2026
  6. 6
    Socio-hydrology and sustainable tank management: empirical case from a Mailam tank cascade, Tamil Nadu, India (2025)Frontiers in WaterPeer-reviewed contemporary case study; methods and infrastructure observations checked · Reviewed 17 November 2026
  7. 7
    EAP1293/1: Sluice Inscriptions from Pudukottai District, Tamil Nadu, IndiaBritish Library Endangered Archives ProgrammeInstitutional collection catalogue; indexed description checked · Reviewed 17 November 2026
  8. 8
    Urbanisation in Medieval India I: irrigation and settlement discussionIndira Gandhi National Open UniversityUniversity teaching publication; indexed irrigation passage checked · Reviewed 17 November 2026
  9. 9
    Colonial and contemporary ideologies of community management: the case of tank irrigation development in south India (1999)David Mosse; Modern Asian Studies / SOAS repositoryPeer-reviewed historical anthropology; indexed colonial discussion checked; direct PDF fetch unavailable · Reviewed 17 November 2026
  10. 10
    Marabucar Nir Melanmai / Traditional Water ManagementTamil Virtual Academy digital libraryInstitutionally hosted archaeological study; indexed sluice-inscription discussion and appendix checked · Reviewed 17 November 2026

Editorial review: Reviewed 17 November 2026 using accessible excavation records, institutional summaries and scholarly research. Some full publications were unavailable; abstracts and indexed excerpts support only the specific claims discussed, while the publications also provide further reading. Dates and interpretations retain their stated uncertainties.