• 17 Nov, 2026

Mahasthangarh, on the west bank of the Karatoya north of Bogura, is the earliest city-site yet discovered in Bangladesh: the ramparted remains of Pundranagara, capital of ancient Pundravardhana, named in a Mauryan-era inscription of the 3rd century BCE and still a living place of pilgrimage.

On the west bank of the Karatoya river, about 11 kilometres north of Bogura, a vast rectangle of grass-grown brick ramparts encloses fields, ponds and village lanes. From the road it looks like a green fortress wall running to the horizon. This is Mahasthangarh, the earliest urban archaeological site so far discovered in Bangladesh, and the remains of Pundranagara, the capital of the ancient territory of Pundravardhana. Cities in Bengal have risen and fallen for well over two thousand years, but none yet excavated reaches as far back as this one: a small limestone slab found here in 1931 carries the oldest written record known from Bengal, and the settlement it names was already a seat of government in the age of the Mauryan Empire.

Mahasthangarh stands on the raised red soil at the edge of the Barind tract, some 15 to 25 metres above the surrounding floodplain. Beside it flows the Karatoya, today a modest stream but once one of the great rivers of Bengal. The site is protected by the Department of Archaeology, which calls Mahasthan "the earliest city-site of Bangladesh". Around the citadel, roughly a hundred ancient mounds spread across the countryside, and on the citadel's south-eastern mound a saint's tomb still draws devotees every day. Few places in Bangladesh hold so many of the country's ages in a single view.

Pundranagara: The Oldest City in Bangladesh

The Pundras were among the peoples of early Bengal, and their country, Pundravardhana, occupied much of the north. Its capital appears in the sources as Pundranagara or, in Gupta and later usage, Paundravardhanapura, "the city of Pundravardhana". According to Banglapedia, the national encyclopedia, Pundranagara served as the administrative headquarters of Pundravardhana from the Mauryan period to the end of Pala rule, and perhaps into the Sena period; under the Guptas it was the capital of the Pundravardhana bhukti, its Bengal province.

That the ruins at Mahasthan are that city was first argued in the nineteenth century. Francis Buchanan Hamilton located and visited the site in 1808, and in 1879 Alexander Cunningham identified it as the capital of Pundravardhana, matching the ground to the account of the Chinese pilgrim Xuanzang, who had travelled through Pundravardhana in the seventh century CE and described a flourishing capital with monasteries around it. The proof fell into place in 1931, when the Brahmi inscription was unearthed inside the citadel: it names the city directly, in its Prakrit form "Pudanagala". The medieval Sanskrit text Karatoya Mahatya also treats Mahasthan and Pundranagara as the same place.

The Mahasthan Brahmi inscription is a fragment of limestone just 4.4 by 5.7 centimetres, bearing six lines of Prakrit written in the Brahmi script. Deciphered by D. R. Bhandarkar, it records an imperial order addressed to the mahamatra, the high official stationed at Pundranagara: grain and oil were to be stored and given out, and money in cowrie shells advanced, so that the people of the area could be carried through a time of calamity, most likely a flood or famine. Scholars generally date the slab to the third century BCE and treat it as a decree of the Mauryan Empire, possibly of the emperor Ashoka's time, though a minority of scholars have argued from its letter forms for a somewhat later date. Banglapedia sums up its weight: it is the earliest evidence of Mauryan authority in Pundravardhana, and the first definite evidence of urbanisation in Bengal. A food-relief order on a scrap of stone is, in effect, the first page of Bengal's written history.

The Citadel and Its Ramparts

The citadel, the fortified heart of the city, is rectangular in plan, measuring roughly 1.52 kilometres from north to south and 1.37 kilometres from east to west, an area of about 185 hectares. Its ramparts, rebuilt in brick over earlier earthen defences, still stand in long stretches; when excavations began they rose 11 to 13 metres above the surrounding country. Geoarchaeological work cited by UNESCO suggests the first fortification here was a mud wall raised around the fourth century BCE. A deep moat guarded the southern, western and part of the northern sides, while the Karatoya itself protected the east. Gateways pierced the walls at the four sides: Kata Duar in the north, Dorab Shah Toran in the east, Burir Fatak in the south and Tamra Dawaza in the west, with bastions such as Munir Ghon along the circuit. At the north-eastern corner a flight of steps called Jahajghata, "the ship landing", runs down towards the river, a reminder that this was once a working river port.

The river explains the city. In its prime the Karatoya carried the waters of the Teesta southwards towards the Ganges, and James Rennell's map of 1767 still shows Pundranagara's successor settlement on a major north–south navigation route. A Mughal-era chronicle claimed the river had once been three times the size of the Ganges, and the Karatoya Mahatya celebrates it as a sacred stream of the Puranas. Then, in the catastrophic floods of 1787, the Teesta tore away eastwards to join the Brahmaputra, and the Karatoya, starved of its Himalayan water, slowly silted into the small river that now slips past the ramparts. It was the river's desertion, as much as any conquest, that ended the city's career as a port and capital.

What the Excavations Found

Systematic excavation began in 1928–29 under K. N. Dikshit of the Archaeological Survey of India, around Jahajghata, Munir Ghon and Bairagir Bhita. Work resumed in 1934–36 at Bairagir Bhita and Govinda Bhita, and in the 1960s around the shrine area, Parasuramer Prasad, Mankalir Dhap, Jiyat Kunda and the northern rampart. Mahasthan and the dozens of mounds around it had been declared protected monuments in the early 1920s, but the digging since has shown how much survives: in the eastern rampart area, excavators have counted eighteen successive construction layers before reaching virgin soil some 17 metres down.

Read as a sequence, the layers give Mahasthangarh its seven broad periods. The earliest, pre-Mauryan levels hold Northern Black Polished Ware, rouletted ware and black-and-red pottery, mud houses with mud floors and cooking hearths. The Mauryan period adds roof tiles, cast copper coins, bronze mirrors and lamps, ring stones and terracotta plaques. Post-Mauryan (Shunga–Kushan) and Kushan–Gupta levels follow, then the Gupta and late Gupta centuries, when substantial brick temples rose in and around the city. The Pala period, from the eighth century, was the city's most flourishing phase, when Buddhist monasteries multiplied outside the walls in the same imperial world that raised Somapura Mahavihara at Paharpur, west across the Barind. The final period is the Muslim age: a fifteen-domed mosque of the Sultanate centuries at Mankalir Dhap, a single-domed mosque of the Mughal emperor Farrukhsiyar's time, Chinese celadon and glazed wares. The small finds match the span: silver punch-marked coins running from the fourth century BCE onwards, a fifth-century stone Buddha from Vasu Bihar, bronze images of the tenth and eleventh centuries, an Arabic inscription of 1300–01 recording the tomb of a naval officer, and a Persian inscription of 1718–19 from Farrukhsiyar's mosque.

Since 1993 the work has been carried on by a Bangladesh–France joint mission, set up under a 1992 cultural agreement between the two governments. The French and Bangladeshi teams brought radiocarbon dating and fine stratigraphy to the eastern rampart and, from 2000, to the shrine area on the western side. The mission is still running: press reports of its 33rd season, in late 2025, describe work at Bairagir Bhita uncovering a tight cluster of eight wells, among them the first brick-lined well ever found at Mahasthangarh, in what the excavators read as a commercial or food-supply quarter of the ancient city. After nearly a century of digging, Pundranagara is still producing firsts.

The Sites Around the Citadel

The walled city was only the core. Suburbs spread over the countryside on every side except the east, where the river ran, and around a hundred mounds within a radius of about nine kilometres are known, many still unexcavated.

Just outside the north-eastern corner of the citadel, across from the site museum and 185 metres from Jahajghata, stands Govinda Bhita, where the exposed bases of two temples overlook a bend of the Karatoya. The remains found here span an extraordinary range, from the third century BCE to the fifteenth century CE, and the complex was rebuilt repeatedly in the Gupta and medieval centuries. Inside the citadel, the named mounds tell the same story of long reuse. Bairagir Bhita preserves temple bases constructed and reconstructed in four periods between the fourth and eleventh centuries, with two carved sandstone pillars. Parasuramer Prasad, the "palace of Parasuram", named for the last Hindu king of local legend, has yielded an eighth-century Pala stone Vishnupatta, glazed sherds of the fifteenth and sixteenth centuries above it, and two East India Company coins of 1835 and 1853. Mankalir Dhap combines terracotta plaques and bronze images with the base of the fifteen-domed Sultanate mosque, and Khodar Pathar Bhita has produced stones carved with Buddha figures. The Jiyat Kunda, a well inside the walls, lives in folklore as a pool whose water could restore dead soldiers to life, a power it lost, the story says, on the day the saint Balkhi's kite dropped a piece of beef into it.

About six kilometres west of the citadel, in the village that still bears its name, Vasu Bihar (Narapatir Dhap) exposes the bases of two monasteries and a temple. Cunningham identified it with the Po-shi-po monastery that Xuanzang described a few miles west of the capital, and its excavated bronzes and stone sculpture belong mainly to the Pala centuries. Three kilometres south of the citadel, at Gokul village, the Gokul Medh is a terraced podium of 172 rectangular blind cells, dated to the sixth or seventh century. Local tradition calls it Behula's bridal chamber, tying the brick terraces to the Behula and Lakshindar story of the Manasa legends.

The Mahasthangarh Museum

The site museum stands beside the citadel, in front of Govinda Bhita, and was established in 1967, according to Banglapedia's history of the Department of Archaeology. It came late: most of what the early excavations found had already left for the Indian Museum and the Ashutosh Museum in Kolkata, with some pieces going to the Varendra Research Museum in Rajshahi. What the Mahasthangarh museum displays today is drawn from the site and the wider Varendra region: stone, bronze and terracotta sculpture of Hindu and Buddhist deities ranging from the Shunga to the Sena period, punch-marked and later coins, inscriptions, terracotta plaques, weapons and ornaments. For a visitor it is the necessary first stop before the walk out onto the ramparts.

A Living Sacred Landscape

Mahasthan means "a place of great sanctity", and the name is not a fossil. On the citadel's south-eastern mound stands the mazar of Shah Sultan Balkhi Mahisawar, the Muslim saint whom legend credits with bringing Islam to Pundravardhana. The story, told and retold in Bengali ballads, says he came from Balkh riding a boat shaped like a fish, mahi in Persian, hence Mahisawar, "the fish-rider"; that he asked the Hindu king Parasuram for just enough land to spread a prayer mat, and that the mat miraculously grew until war followed, ending with the king's fall. Historians treat the tale as legend, and even the saint's century is disputed in the traditions, but the shrine is firmly documented in the Mughal period: in 1685, under Aurangzeb, the dargah held rent-free land by imperial sanad, the gateway called Burhi ka Darwaza was built for it, and in 1719 one Khodadil raised the single-domed mosque beside the tomb, in the reign of Farrukhsiyar. That mosque is still in use, and devotees still climb the mound daily to make mannat, vows, at the tomb.

Hindu devotion runs alongside. Shila Devi's Ghat, on the Karatoya below the citadel, is named for Parasuram's daughter in the same cycle of legends, and pilgrims have bathed in the river at Mahasthan for centuries, drawn by the sanctity the Karatoya Mahatya proclaims. Each year, on the last Thursday of the Bengali month of Baishakh, the Baishakhi Mela fills the ground around the shrine: the Daily Star has described Muslim devotees gathering at the mosque and Hindu devotees at their own spots among the mounds, in one of the oldest continuing fairs in northern Bengal. Archaeological site and pilgrimage centre occupy the same ground here, which is both Mahasthangarh's distinction and its management headache.

Why Mahasthangarh Matters

Mahasthangarh matters first as evidence. It is the reference site for the beginning of urban life in the Bengal delta: no other place in Bangladesh offers a stratified sequence running from pre-Mauryan pottery to Mughal glazed ware in a single fortified landscape, and its little Brahmi slab anchors Bengal's history to a dated, readable document earlier than any other. It matters, too, as the first of Bengal's capitals. Pundranagara was governing a province when later capitals such as Sonargaon were not yet imagined, and the line of Bengal's urban history runs from these ramparts forward.

Recognition has come in stages. Mahasthangarh was declared the Cultural Capital of SAARC for 2016–17, announced in November 2016 by the SAARC Cultural Centre and inaugurated in January 2017, with a year of festivals on the site. In 2023 Bangladesh placed the "Cultural Landscape of Mahasthan and Karatoya River" on UNESCO's Tentative List, the inventory from which future World Heritage nominations are drawn. It is worth being precise: Mahasthangarh is not yet a World Heritage Site, as Somapura Mahavihara has been since 1985, and tentative listing is a candidacy, not an inscription. The site also carries warnings. In 2010 the Global Heritage Fund placed Mahasthangarh among twelve sites worldwide "on the verge" of irreparable loss, citing looting, encroachment and poor drainage, and Bangladeshi newspapers have repeatedly reported brick-robbing and building on protected mounds.

Yet the story is not closed. The joint excavation still returns each season, the museum still receives the finds, and the fair still gathers on the last Thursday of Baishakh. A city that began as a Mauryan provincial capital, outlasted the river that made it, and turned into a shared shrine of two faiths is not a ruin so much as a very long continuity, and Mahasthangarh remains the place where Bangladesh's urban past begins.