Bhangarh Fort: Haunted Ruins Beside a Living Village

The first thing you meet at Bhangarh Fort is not a ruin. It is a board.

Dark blue, white-bordered, the white Devanagari weathered but entirely legible. Government of India across the top, with the Ashoka emblem set between the two words. Under that, Archaeological Survey of India, Bhangarh. Then three numbered rules.

The first is the one the world knows. Entering within the boundary of Bhangarh before sunrise and after sunset is prohibited. That sentence has been photographed a few hundred thousand times. It appears in every listicle, every travel vlog, every documentary segment about India’s most haunted place, and it is almost always presented the same way: as the moment a national government admitted in writing that something here is wrong.

The second rule states that bringing livestock of any kind inside the boundary is an offence in law.

The third states that the kewda trees within the boundary are the property of the Archaeological Survey of India, and that damaging them in any manner is forbidden.

Then a closing note that legal action will follow any breach of the above, and a signature line. By order, the Superintending Archaeologist.

Read it twice. There is nothing on that board about ghosts. No spirits, no curse, no princess, no sorcerer, no warning about what befalls people who linger. Three clauses, and two of them concern goats and trees.

A rule against grazing gets written where there are animals to graze. A rule protecting one named species of tree gets written where somebody has a reason to cut it. Neither clause makes any sense in an empty place. Both of them assume neighbours, close by, with livestock and an interest in timber, and a habit of coming and going across ground the state would prefer they left alone.

There are neighbours. Bhangarh is an inhabited village. India’s 2011 census records it under location code 073334, with 1,945 people in 286 households across a revenue area of 1,294 hectares.

The ruins sit beside all of that: five gates, a bazaar street, the Gopinath temple on its high plinth, the collapsed upper storeys of the palace stacked against the hillside, and beyond the wall a working settlement in the Aravalli foothills.

The story told about this place, in every language it gets told in, requires that none of that be there. It requires an emptiness. No one dares live at Bhangarh. No roof will stand. Nobody has returned in four hundred years.

Somebody has to be standing in the way of that story, and roughly two thousand people are.

The Place: What Is Actually at Bhangarh Fort

Bhangarh Fort sits in a shallow valley on the eastern flank of the Aravallis, in Rajgarh tehsil, about eighty-three kilometres from Alwar town and rather less from Dausa. Hills close the site on three sides. The fourth opens onto cultivated ground.

This is the edge of the Sariska Tiger Reserve, and the boundary matters in a practical way. The land beyond the monument is not empty countryside but managed forest and grazing land, and the animals that live in it, nilgai, wild boar, langur, the occasional leopard, do not observe the perimeter.

Approach from the east and the site arranges itself along one axis. Everything the town needed was strung along a single street that runs from the outer gates towards the hill, and the higher you go the more important the buildings become, until the ground steepens and the palace takes over.

Walk it and the order is legible. Past the main gate the ground is dense with temples: Hanuman, Ganesh, Keshav Rai, Mangla Devi, Someshwar, and Gopinath standing on a raised plinth some fourteen feet high, carved in yellow stone. All of them are Nagara in style. The shrines to Hanuman and Shiv Mahadev were built as cenotaphs rather than as temples, in Jhiri marble, which is an unusual choice and nobody has explained it. Beside the temple complex is the Purohitji Ki Haveli, the head priest’s house.

Then the street proper. Jauhari Bazar runs between rows of shop plots whose divisions are still readable in the stonework, one narrow unit after another, and the Nachan Ki Haveli sits along it, named for the dancers who worked there. Beyond that the residential quarters, then the palace at the far limit of the walls, reported as seven storeys and now standing at four.

Four subsidiary gates carry names that give away the geography of the trade the town lived on: Lahori, Ajmeri, Phulbari, Delhi. Outside the walls there is a Muslim tomb, recorded in the 1880 Rajputana gazetteer as belonging to a son of one Hari Singh, which is the sort of detail that survives in an official volume and appears in almost nothing written since.

The village occupies the flat land outside the walls of Bhangarh Fort. Its revenue area runs to roughly thirteen hundred hectares, which is a great deal more ground than the monument covers, and most of it is farmed or grazed. The two things sit against each other with a line drawn between them, and that line is the entire administrative fact of the place.

Then the kewda. Pandanus, the screwpine, grows here in thickets, all spined leaves and stilt roots, the male plants producing cream-coloured flower spikes inside protective bracts. This is not decorative scrub. Kewda flowers are distilled into one of the most valuable attars in Indian perfumery, a honeyed, hyacinth-sweet oil, and the hydrosol left over from the distillation flavours biryani and ras malai in kitchens across northern India. The flowers are picked at dawn because they lose their scent within hours of opening.

A stand of kewda is a standing crop. That is why the third clause exists.

One correction is worth making. There is a published Archaeological Survey of India inventory for Bhangarh, including a brief history and a site plan. It is not a full excavation report, but it is enough to show that the site has been formally surveyed and described; saying that no archaeological survey exists would be too strong.

The History of Bhangarh Fort

The Archaeological Survey of India’s published inventory says Raja Bhagwan Das founded the township in the later half of the sixteenth century and that it later became the capital of Madho Singh, brother of Man Singh I. Rajasthan Tourism gives a different shorthand, calling it a seventeenth-century fort built by Madho Singh. The precise year 1573 is repeated widely online, but the official sources I found do not establish it, so a date that exact should be treated cautiously.

The town worked. Bhangarh Fort sat on routes running through the Aravalli passes, it had a market street, temples and a court. Rajasthan Tourism says the township once contained more than 9,000 houses, a claim best read as official tradition rather than a verified population count.

Madho Singh’s son Chhatra Singh followed him. Then Chhatra Singh’s son Ajab Singh built a new fort several kilometres away on a hilltop, called it Ajabgarh, and moved his court into it.

That single decision is the beginning of the end, and there is nothing remarkable about it. When a Rajput ruler moved his seat, the old capital lost its patronage, its administration and the trade that depended on both. It happened across Rajasthan. Amber itself emptied when Jaipur was built.

The later decline is clearer than its final cause.

The Chalisa famine of 1783–84 followed severe monsoon failure across northern India and took its name from the Vikram Samvat year 1840. It affected the Delhi territories, the Doab, eastern Punjab, Kashmir and Rajputana. Some later estimates put the death toll as high as eleven million, but reliable contemporary totals do not exist. The famine coincided with the Laki eruption in Iceland; research has proposed volcanic cooling as one contributor to wider climate disruption, but it does not establish a simple causal chain from Laki to Bhangarh.

It is tempting to make the Chalisa the clean historical answer to the curse, but the site-specific evidence does not let us go that far. Bhangarh had already been declining, and the surviving official descriptions do not prove that famine alone emptied it. The honest conclusion is narrower: drought, famine and the loss of courtly patronage offer a plausible material history of abandonment without requiring a curse.

Bhangarh Fort passed eventually into the care of the Archaeological Survey of India, which is why there is a boundary, a ticket, a guard and a board. And in the last two decades it has become one of the most visited ruins in Rajasthan, on the strength of something that has nothing to do with any of the above.

The Legend

Two stories are told locally, and both of them are on the Rajasthan state tourism website, which is worth holding on to. That page calls the temples a sublime example of seventeenth-century workmanship and, a few lines further down, tells you the place is cursed. The same department publishes both sentences.

The first belongs to Baba Balau Nath, a sadhu who was performing tapasya on this ground before there was a town on it. When the raja came to build, the ascetic gave his consent with a condition attached: no structure was to stand taller than his own dwelling, and if the shadow of any building fell across it, the town would be finished. Generations later Ajab Singh raised the palace, the shadow reached, and the condition came due. Balau Nath is said to lie in a small samadhi on the site still.

The shape of that story is worth noticing before it gets flattened. It is not a haunting and there is no ghost in it. It is a boon granted with a condition, broken by a descendant of the man who accepted it, and the consequence is not a curse in the horror-film sense but the maturing of an agreement. That structure runs deep in Indic narrative, and it carries an argument about ascetic authority outranking royal ambition. A raja may build. He may not build over a man doing tapasya.

The second story is the one that travels.

Ratnavati was the princess of Bhangarh, said to be the most beautiful woman in Rajasthan, and on her eighteenth birthday the marriage offers began arriving from the ruling houses. In the town lived Singhia, a tantrik, who wanted her and knew the match was impossible. He saw her maid in the market buying scented oil and worked on the bottle so that the princess would surrender herself on touching it. Ratnavati saw what he had done. She poured the oil out onto the ground, or in other tellings smashed the bottle against a boulder, and the boulder rolled down on Singhia and crushed him. Dying, he cursed the town so that everyone in it would die and none of them be reborn. Within a year Bhangarh was at war with Ajabgarh, and the princess was killed. A chhatri on the hillside is pointed out as Singhia’s.

Two things get lost when this is retold in English. The word tantrik here is doing folk work, not theological work. It denotes a stock antagonist of Rajasthani village storytelling, a man who uses ritual power for private ends, and it is not a description of Tantra as a religious tradition any more than a witch in an English folk tale describes Wicca. And in several local versions Ratnavati is not a passive target at all. She recognises the enchantment because she has learning of her own, and she defeats him with it. The retellings that reach international audiences tend to leave her as a beauty who got lucky.

Around these two stories a much newer layer has grown, and it is almost entirely in English and almost entirely online. That no roof will stand at Bhangarh Fort, and that any built collapses. That anyone entering after dark does not come out. That the fort was abandoned in a single night. That the smell of attar drifts through the ruins, which is a nice touch given what grows there.

And at the centre of every modern account of Bhangarh Fort, quoted in the first paragraph, that board.

Is Bhangarh Fort Haunted? What the Evidence Explains

Start with the document, since everything else rests on it.

The board is an ASI notice on a centrally protected monument, and every clause on it is practical. Restricting access outside daylight is an ordinary safety and conservation measure for an unlit ruin with uneven stonework, open drops and no night-time visitor infrastructure. The Ancient Monuments and Archaeological Sites and Remains Act gives the Survey power to regulate access and protect the site.

The claim endlessly repeated in English—that Bhangarh is the only monument in India with such a restriction, and that the state has therefore conceded something—is not supported by the notice or by ASI policy. Many heritage sites use fixed daylight visiting hours. What is unusual at Bhangarh Fort is not the rule. It is that the rule gets read as testimony.

And nobody reads past clause one. There is no version of the paranormal argument that engages with the livestock provision or the kewda provision, because both of them describe a working relationship between a settlement and a piece of ground, which is precisely the thing the ghost story cannot accommodate.

You can watch the reading narrow, and it has a timeline. Around 2010 the English text in circulation rendered the second and third clauses together as a prohibition on shepherds and woodcutters. That is a poor translation and a good summary, because at least it contains people. By the current crop of travel pages, clause one is quoted and the other two have gone. And in a 2025 piece in the Times of India the board is not quoted at all. It appears only as a thing the Survey was forced to erect because of paranormal activity, a motive that exists nowhere on it. The same article does reach for a village, and names Ajabgarh, several kilometres away on a different hill, whose people are said to avoid living near the fort. The village at the fort’s feet does not appear. Fifteen years, three stages, and at each one the document says a little less about anybody who lives here.

Then the abandonment, which the History covers and which needs no supernatural supplement. Add to it the ordinary physics of ruin. Domestic buildings in a Rajasthani hill town were roofed with timber and mud, and timber and mud are the first things to go once nobody is maintaining them. The temples are still standing with their superstructures because they are dressed stone, built to a different specification for a different purpose. A site where the houses are roofless and the temples are not is not a site where roofs refuse to stand. It is a site where two building traditions decayed at their normal, wildly different rates.

The princess is harder to test, and the honest answer is that I cannot find her. There is no contemporary record of a Ratnavati of Bhangarh, no documented war between Bhangarh and Ajabgarh in the year the legend requires, and the tellings themselves disagree about whether she was a princess or a queen, whether she poured the oil or threw the bottle, and whether Singhia was crushed at Bhangarh or she died at Ajabgarh. That degree of internal disagreement is normal for oral tradition and fatal for a historical claim.

The newer English-language accretions fail faster. Abandoned in a single night is contradicted by a decline traceable across more than a century. Nobody who enters returns is contradicted by ordinary daytime tourism. Annual visitor figures of around two hundred thousand are repeated widely, but I could not trace them to official data, so I have not used them as fact.

The one claim I would leave standing is the smell of attar in the ruins. There are kewda thickets on the site. Kewda is one of the most diffusive florals in the world, and it is picked at dawn precisely because it broadcasts and then fades. If people report perfume at Bhangarh Fort, the likeliest explanation is that they are smelling a plant.

Which leaves the government, and the split running through it. One department of the state sells the curse. A body of the union government posts a board about grazing. Two arms of the same establishment, one trading on the story and one managing the livestock, and the visiting public has concluded that the second is quietly endorsing the first.

My Thoughts

Here is the thing I keep turning over. A haunting needs an empty stage.

That is not a rhetorical flourish, it is a structural requirement. A ghost story cannot survive occupancy. The moment you put people in a place, doing ordinary things in it on an ordinary Tuesday, the uncanny drains out, because the uncanny is a property of absence and the absence has to be total. One woman hanging washing is enough to end it. So the story does what stories do when reality obstructs them. It edits.

At Bhangarh Fort the editing has been remarkably complete. Search in English and you will find hundreds of thousands of words on this place in which no living person appears except as a tourist, a guide or a guard. The village is not disputed or argued with. It is simply not mentioned. It has been removed from the frame the way a photographer waits for pedestrians to clear.

And I want to be careful about what I am accusing anyone of, because it is not lying. Almost nobody writing about Bhangarh Fort knows the village is there. They have never looked, because the story they arrived with did not have a slot for it, and a thing with no slot does not register as missing. That is a more troubling mechanism than dishonesty. Dishonesty can be corrected.

What sharpens it for me is that the sign has been telling anyone who reads it, for years, in plain language, that this is a populated agricultural landscape. Livestock and a valuable crop. Those two clauses are the only lines on that board that describe a relationship between human beings and this ground. Clause one describes no relationship at all. It gives a time. And clause one is the only one anybody has ever quoted.

So the world has been photographing the evidence and cropping out the part that contradicts it, for fifteen years, without meaning to.

There is a harder edge to this, and it is about who pays. The night restriction does not only exclude visitors. It applies to the boundary, and the people whose fields and animals are on the other side of that boundary live under it too. The grazing clause makes an offence of an activity that people in the Aravallis have practised on this ground for longer than the fort has stood. The kewda clause asserts that a commercially valuable crop growing on the edge of a farming village is the property of the union government. Every one of those is defensible as heritage management. All three also happen to transfer the use of a piece of land from the people beside it to an institution and, downstream, to an industry.

That tourism industry is real, even if its scale is poorly documented. Visitors arrive at a village of under two thousand to see a place they have been told is empty. Some of that money reaches the village. I do not know how much, and neither does anyone writing in English, because nobody has asked.

The legend, when you strip the enchanted oil out of it, is a story about a town that ended. The Chalisa is a story about towns that ended across large parts of northern India, and about monsoon failure, hunger and uncertain climate connections. The second story is larger, better attested and more frightening in every way that matters, and it lost, comprehensively, to a curse.

I think it lost because a curse has a villain and a famine does not. You can point at Singhia. You cannot point at a failed monsoon. And a story you can point at is a story that lets you stop, whereas the Chalisa keeps going, out past Bhangarh and across northern India and into questions about food, water and who was governing at the time, none of which fit on a board.

The state can bolt a notice to the ground that stops you walking in after sunset, and it can prosecute you for grazing a goat, and it can claim the trees. None of that has kept out the only thing here that has actually displaced anybody.

So what would you have to write on a sign, and who would you have to be to sign it, to stop two thousand people being edited out of their own village?

Visiting Bhangarh Fort

Bhangarh is reached by road in Alwar district, between Jaipur and Alwar; Dausa is the most useful nearby rail connection. Drive times vary and public transport is limited, so arrange the return journey before leaving. Many visitors pair the trip with Chand Baori at Abhaneri or with Sariska.

Entry is permitted only during daylight; the ASI notice prohibits access before sunrise and after sunset. Published hours and fees vary across travel sites and can change, so confirm them locally or with Rajasthan Tourism or the ASI Jaipur Circle before setting out. Allow roughly ninety minutes to two hours. The layout follows a single main axis and can be visited independently, though local guides are available.

November to February is the season. Summer here is punishing and the ruins offer almost no shade. Early morning is both the coolest and the emptiest, and weekends bring substantial domestic crowds.

The palace is stacked in levels against the hillside and reached by old stone stairs and ramps that are uneven throughout. There are unfenced drops and open wells across the site. Wear proper shoes, carry water and keep children close on the upper levels. Shade and visitor facilities are limited.

Two notes on conduct, and the second matters more than it usually does. This is a protected monument, so nothing gets climbed, carved, removed or flown over without permission. And this is somebody’s village. The people you pass on the road are not part of the attraction, and their houses are not photo opportunities. What lies beyond the perimeter is a working farm, not atmosphere. If you buy a drink or a snack, buy it there rather than bringing it from Jaipur.

Staying overnight in Jaipur? Compare places to stay in Jaipur. If you book through this link, Shin Stories may earn a small commission at no extra cost to you.

Before You Go

This is Location 14 of twenty in Season 1 of The Dark Atlas. The companion episode takes the board one clause at a time and will appear in the Spotify feed when released. Continue through the archive with the Highgate Vampire and Lawang Sewu, or explore every mapped location in The Dark Atlas. Follow the podcast and Location 15 arrives without you going looking for it.

Sources

Some book links below are affiliate links. As an Amazon Associate Shin Stories earns from qualifying purchases, at no extra cost to you.

Books and Academic Works

Thordarson, Thorvaldur, and Stephen Self. “Atmospheric and environmental effects of the 1783–1784 Laki eruption: A review and reassessment.” Journal of Geophysical Research: Atmospheres 108, no. D1 (2003).

Imperial Gazetteer of India, vol. III, “Economic,” section on the history of Indian famines. The standard nineteenth-century summary of the Chalisa, and the source most later accounts trace back to.

Rajputana Gazetteer (1880), pp. 289–90. The earliest official description of the site consulted here, and the source for the tomb outside the gate.

Ramanujan, A. K. Folktales from India: A Selection of Oral Tales from Twenty-Two Languages. New York: Pantheon, 1991. Useful on the boon-with-condition structure that governs the Balau Nath story.

Official and Institutional Sources

The Archaeological Survey of India notice board at Bhangarh. The primary document for this article, transcribed and translated from the Hindi in a published photograph rather than from the English versions in circulation. The photograph itself is of unestablished authorship, in circulation since at least June 2012, and is not reproduced here.

Archaeological Survey of India, Ancient Site (Bhangarh). Published inventory and site plan; the basis for the corrected founding chronology and the description of the surviving complex.

Government of India, Ancient Monuments and Archaeological Sites and Remains Act, 1958, as amended. The statutory framework for access and protection at centrally protected monuments.

Rajasthan Tourism, official Bhangarh Fort page. The state government’s publication of both legends, cited as evidence of what the state promotes rather than as evidence for the legends, and one side of the conflicting official chronology.

Census of India 2011, Alwar Primary Census Abstract. Bhangarh village, location code 073334: 1,945 residents, 286 households and 1,294 hectares.

Journalism and Long-form Reporting

“Ghost of Rajasthan: Exploring the haunting legends of Bhangarh Fort.” The Times of India, 2025. Cited specifically for what it does not contain: the board is never quoted, and the only village named is Ajabgarh.

Coverage of Bhangarh in English is otherwise thin and heavily recycled, and much of what ranks is machine-generated. Where any of it disagrees with the board or the census, the board and the census win.

Further Reading

Davis, Mike. Late Victorian Holocausts: El Niño Famines and the Making of the Third World. London: Verso, 2001. Concerns later famines than the Chalisa, but remains a major account of how drought, governance and mortality interact on the subcontinent.

Parveen, Wajeda, and Anrukati Sharma. “Bhangarh Fort: a case study of dark tourism.” In Dynamics of Commerce and Management. Archers & Elevators Publishing House, 2014. A minor volume, and as far as I can establish the only academic treatment of Bhangarh as a tourism phenomenon.