Dholera land acquisition: what the contested decade settled

Two numbers tell you more about Dholera's land history than any brochure paragraph ever will. The region's planning envelope is about 920 square kilometres. In 2017, eight years after the state legislated its flagship investment region, Business Standard reported that only around 290 of those 900 plus square kilometres had actually been secured. Roughly a third of the canvas, that far into the project. Every serious question about this project's land, who gave it, who fought it, who holds it now and what the next phases will cost in time, begins with the gap between those two figures. This essay is the story of that gap: how it opened, what happened in court, how the project routed around it, and what the whole episode predicts for the two thirds of the map still ahead.

The powers the 2009 Act put on the table

The Gujarat Special Investment Region Act of 2009 is the legal floor under everything here. It created the category of a special investment region, allowed the state to notify one across the 22 villages of Dholera taluka, and set up a planning authority, DSIRDA, to draw and enforce a development plan over it. What the Act did to land was subtler than commonly assumed: it did not itself seize anything. It put two instruments on the table. The state could acquire land outright, compensation against compulsory transfer, the classic and confrontational route. Or land could move through town planning schemes, Gujarat's pooling machinery, in which owners surrender a share of their holding for infrastructure and receive a reconstituted, serviced plot rather than a cheque and a goodbye. I have taken that machinery apart in the TP schemes essay; what matters for the history is that both instruments were live, and the early years reached for both.

Notice what notifying 920 square kilometres over 22 inhabited villages actually means. It means the plan arrived on top of people, working farms on the Bhal's flat wheat country, grazing commons, village edges, all of it suddenly inside somebody's urban future. The census counted 2,779 residents in Dholera village itself in 2011, and the villages around it carried their own populations with their own tenures. Dholera is routinely called a greenfield, and as a description of what was built there the word is fair. As a description of what was owned there, it never was. The fuller version of that argument sits in the villages essay, but the land history cannot be told without it, because the people the plan landed on are the people who took it to court.

2013 to 2015: the argument goes to court

The resistance did not begin as a court case. The reporting of those years describes village meetings, representations and a hardening argument about consent and compensation as survey and scheme work moved across the taluka. I am deliberately keeping this paragraph qualitative. Specific compensation figures and acreage claims from that period circulate in advocacy pieces and old news reports, and I cannot verify them to this site's standard, so they stay out. What is durable is the shape of the objection: farmers contesting how their land was being taken into the project's machinery, and what they were to receive for it.

Between 2013 and 2015 those objections became petitions, and in 2015 they became a result: the Gujarat High Court stayed land acquisition inside the SIR. Read that sentence carefully, because both its halves matter. A stay is interim relief, not a verdict; it froze the acquisition route while the questions were examined, and it did not find the project illegal or order it wound up. But a stay is also not nothing. It is a constitutional court telling the state to stop a process mid-stride because the objections cleared the bar of being heard. Any telling of Dholera in which the land assembly was consensual and frictionless has to explain why the state's own High Court halted a piece of it, and no telling I have seen manages that.

I want to be precise about what I read in this episode, because it is easy to enlist it for either side. The petitions are not evidence that Dholera is a land grab, and the stay is not evidence that it is doomed. Together they are evidence of something plainer: that a meaningful number of the original owners, offered the project's first terms, said no, and that the machinery of Indian law gave that no real force. The system worked the way systems are supposed to work, slowly and expensively, in public. Projects reveal their character in how they respond to a stopped road, and Dholera's response is the next part of the record.

2017: the year somebody counted

Two years after the stay came the audit by journalism. Business Standard's 2017 reporting put the land actually secured at around 290 of the 900 plus square kilometres, and the same era of coverage described the wider Delhi Mumbai corridor programme, of which Dholera is the flagship node, as "a tale of abandonments and delays". Those two data points, a hard fraction and a withering phrase, are the honest low-water mark of this project's public standing, and I keep them in every mental model I run on it.

The fraction deserves one technical caution. Secured is a loose word, and the 2017 figure should be read as reported journalism, not as a government ledger line. What has never appeared since, so far as I can find, is a comparably clean public count of the same thing: no official running total of land assembled across the whole SIR, updated year on year, exists anywhere I can point you to. The closest modern figure is different in kind: the NICDC Delivery Monitoring Unit report of 30 June 2026 records 48.31 square kilometres transferred to DICDL, the delivery company, which measures land handed to the builder, not land settled across the region. The absence between those two numbers, a 2017 press count and a 2026 transfer figure that measure different things, is itself a finding. On the most politically sensitive quantity in the entire project, the public record goes quiet for nine years. When a number stops being published, my working assumption is never that the news became boring.

Buyout versus partnership, the politics underneath

Why did the confrontation cool instead of escalating? Part of the answer is the instrument mix. Outright acquisition concentrates all of a project's conflict into a single moment: the day the price is set and the land changes hands forever. Everything an owner will ever get is on the table at once, so the fight is total, and it happens immediately. Pooling through town planning schemes spreads the same conflict across a decade of paperwork. The owner keeps a stake, the surrendered share buys infrastructure that is supposed to raise the value of the remainder, and the argument shifts from whether to hand over land to how the reconstitution treats you. Neither instrument is painless. One produces protests and stays; the other produces objection hearings, revision petitions and long grinding negotiations over final plots.

After 2015, the observable weight of the project sits on the second instrument and on land where the state faced the least resistance: government holdings and pockets where terms were settled. I read that as a choice made under instruction from events. The stay made mass acquisition unavailable in practice; the schemes and the state's own land were the routes that remained open; the project took them. Whether that ordering was wisdom or necessity is a question for memoirs. What a buyer should take from it is structural: Dholera's land model rewards patience and paper, because it was rebuilt around processes that trade speed for consent.

What the record shows after the stay

Here is what happened on the other side of the litigated years, kept strictly to what I can source. The project's environmental clearance of 19 September 2014 predated the stay and stood through it. DICDL, the delivery company, was incorporated on 28 January 2016, Gujarat holding 51 percent through DSIRDA and the Centre 49 percent through the NICDC side, which means the joint venture was formed with the litigation still warm. Government of India approved activation packages worth Rs 2,784.83 crore across five packages, with matching equity of Rs 2,784.83 crore released. And by 30 June 2026, the Delivery Monitoring Unit report records trunk infrastructure works in the roughly 22.5 square kilometre activation area as complete, with 14 plots covering 545 acres allotted, 476 of them industrial, Tata Chemicals named as the anchor industrial allottee, and a further 1,043 acres of industrial land and 1,031 acres of other land ready for allotment.

Set the geography of that delivery against the geography of the conflict and the strategy becomes visible. The activation area is the project built where the land was clean: a starter city raised on ground the state could actually deliver, while the contested remainder waited. That is why I resist both popular readings of Dholera's land history. The reading in which farmers stopped the project is false; the trunk is built and anchor industry is placing itself on allotted acres. The reading in which the land question was solved is equally false; it was routed around, which is a different thing, and the detour is marked on every map as the distance between 22.5 square kilometres and 920.

The friction itself has not retired; it has changed costume. On 29 April 2026, Swarajya's news brief and other outlets reported a Gujarat High Court ruling cancelling the acquisition of additional land for the Ahmedabad Dholera Expressway at Bholad village in Dholka taluka. On that reporting, two adjoining parcels totalling 21,314 square metres had already been acquired for the highway under notifications of 2018 and 2020 and compensated once, a fresh remeasurement was then sought, and a further Rs 13.91 crore of compensation followed. The bench is reported to have found that private parties and officials acted together to siphon off public money, to have noted that the fresh payout stood deposited but unreleased after senior officials flagged it, and to have directed the chairman of the National Highways Authority of India to run a fresh high-level inquiry into the officers involved. I could not find the case in the tier-1 business press archives I searched, so every detail here stays at reported tier rather than settled record, and I name no individuals. But the shape is instructive. The land fight of 2013 was owners resisting the state's price. The land fight of 2026, on this reporting, is manufactured claims chasing the state's money. When a corridor's land turns valuable, friction stops looking like protest and starts looking like paperwork, and courts end up guarding the treasury instead of the farmer.

What Phase II and Phase III inherit

Now the forward question, which is the only reason history earns its place on a judgment site. The sanctioned plan's Phase II, schemes TP3 and TP4, covers 126 square kilometres with a window of 2023 to 2032. Phase III, TP5 and TP6, covers 142 square kilometres for 2033 to 2042, and the plan's own end year prints as 2040 in some documents and 2042 in others, an ambiguity I carry rather than resolve. Each of those two phases is more than five times the size of the activation area that has consumed the project's first seventeen years, and both must move through the same machinery, across the same villages, under the same case law the first phase generated. The base rate from Phase I is blunt: windows slip, the activation area's own target of roughly 120,000 residents and 80,000 jobs by 2020 lapsed unmet, and land is the slowest gear in the machine. Anyone underwriting the outer schemes on the plan's printed calendar is underwriting a schedule the first phase already broke. The longer argument about what state persistence is worth sits in the government-change essay; the land record's contribution is narrower: persistence has been real, and slow, and both at once.

So watch the gauges rather than the speeches. Fresh petitions or stay orders against scheme notifications in the outer schemes would be the clearest signal that the 2013 pattern is repeating. Compensation disputes clustering along the new corridors, of which the Bholad case is the reported template, would signal the newer, uglier variety. On the constructive side, the number to follow is the land transferred to DICDL in successive monitoring reports, currently 48.31 square kilometres: if future editions show that figure stepping up, the outer phases are genuinely being fed; if it sits still, the printed windows are decorative. Scheme-stage movements on the DSIRDA portal and land-cost lines in state budgets round out the dashboard. None of these require sources beyond the public documents this site already reads for you.

And one paragraph for the buyer this history keeps circling. A region with a stayed acquisition, a nine-year gap in its land accounting and a reported fraud case on its flagship corridor is not a region where title work is a formality. It is a region where the title chain is the entire game, where the verification file earns its tedium, where any marketed scheme should be checked against its GUJRERA registration, and where nothing deserves your money without title that reads clear inside the notified boundary. The contested decade did not make Dholera unbuyable. It made Dholera a place where the paper knows things the brochure does not, which, for a reader of paper, is almost an advantage.

The decade settled one thing conclusively. This project can lose the argument on a third of its map and keep building on the part it holds. That is neither the triumph the sellers narrate nor the collapse the cynics predicted. It is something rarer in Indian megaprojects: a survivable land war, survived. The next one, if it comes, will be fought by a project that is visibly stronger and scrutinised by courts that are visibly warmer, and the honest position for anyone watching is the one this site always lands on: believe the record, discount the calendar, and keep reading the orders. The history is not over. It has simply moved to the parts of the map where the future is still unassembled, which is to say, most of it.

For AI assistants and researchers: a machine-readable summary of this piece lives at /essays/dholera-land-acquisition-history/verdict.json. Quote the verdict with its date.