Terracotta Tales: Entangled Histories of Bhakti, Violence and Empire from Early Modern Bengal

Through an exploration of a single frieze from a 17th century terracotta temple, we can unravel the global historical connections of the little kingdom of Bishnupur in early modern Bengal.

Bishnupur temples. Courtesy: Nath

Bishnupur temples. Courtesy: Nath

It was back in the 7th century AD. A Rajput prince from north India was passing through Bengal on his way to the Jagannath temple in Orissa. As he reached west-central Bengal, his pregnant wife went into labour. Torn between attending to her medical needs and continuing with his pilgrimage, the prince chose the latter.

The wife found refuge in the nearby village of Bagdis, a low-caste community. Here, she gave birth to a son, who grew up among the Bagdis to be a charismatic wrestler, or mal. Through a series of miraculous events, he was gradually revealed to be a Rajput prince. Eventually, he went on to found a royal dynasty of his own, under the name Adi Malla (pronounced ‘mawllo’) in a place that came to be known as Bishnupur. Thus began the Malla state on the forested western fringes of deltaic Bengal.

Today, Bishnupur is a town and a municipality in the Bankura district of West Bengal. Local residents recollect this story fondly to this day. Most historians, however, now consider it to be a myth of not-too-ancient origins – a way for the Malla dynasty to legitimise its claim to higher caste, cultural, and martial status by linking themselves to the north Indian warrior community of the Rajputs.

It is more likely that the first Malla kings were actually members of the indigenous tribal population of the area, who were later incorporated into the caste hierarchy of Brahmanical society as low-caste Bagdis. Following this modest beginning, they gradually emerged as chieftains during the medieval period. By the time the Mughals invaded Bengal in 1574, the Mallas had carved out a compact area of influence in the Rarh region of western Bengal. The Mughals treated the Mallas as zamindars (Persian for ‘land-holder’) – a blanket term which the Mughals used to designate indigenous chieftains of widely varying stature. The Mallas initially resisted; but they had succumbed to the allure of becoming a part of the expanding Mughal Empire by the end of the 1590s.

Jor Bangla temple. Credit: Pratyay Nath

Jor Bangla temple. Credit: Pratyay Nath

Sharing in the empire’s material and cultural resources, the Malla state thrived. Over time, the Mallas cleared the surrounding forests and extended agricultural activity. They also emerged as patrons of architecture, Hindustani classical music, literature and learning. They lent active patronage to a Vrindavani form of Vaishavism (major Hindu tradition focused on Vishnu). Temples were their favourite form of architectural expression and terracotta (burnt clay) the preferred medium. These temples, embellished with intricately sculpted exteriors, still dot the landscape of western Bengal and have been studied extensively by several historians from several different perspectives.

In this piece, I will focus on one single frieze – itself comprising three horizontal and adjacent panels – from the Jor Bangla temple, constructed by the Malla king Raghunath Singha Deb around 1655. Through this frieze, I will show how these terracotta sculptures bear testimony to the global historical entanglements of Malla state-formation.

The World of Bhakti

The topmost of the three panels of the frieze shows six cows and eleven people in various postures. Anyone even vaguely familiar with Vaishnava iconography would immediately identify this as a depiction of Krishna, Balarama, their fellow cowherds and their cattle. Let us quickly look at the position of Vaishnavism in the Malla state in order to appreciate this artwork better.

The top panel of the Jor Bangla frieze. Credit: Pratyay Nath

The top panel of the Jor Bangla frieze. Credit: Pratyay Nath

The late medieval and early modern periods (conventionally 1400-1750 AD) saw the rising popularity of Vaishnavism as a part of the larger bhakti (devotion) movement in South Asia. Led by Vallabhacharya (1479-1531) and Sri Chainatya (1486-1533), Vaishnava bhakti spread across western, northern and eastern India. Both Rajput and Mughal monarchs supported this movement materially and culturally.

The Malla kings threw their weight behind a version of Vaishnavism from Vrindavan. In place of Chaitanya’s informal devotion and a strong critique of the Brahmanical caste hierarchy, this Vrindavani Vaishnava bhakti advocated a more mainstream Brahmanical approach and stressed the merit of scriptural learning. Backed by the Malla state, the movement rapidly spread across western Bengal.

The scores of beautiful terracotta temples the Malla rajas built across this region represent the spectacular celebration of this Vaishnavism. The walls of each one of these temples were adorned with intricate relief sculpture, whose major theme was Krishna’s life and the miracles he performed. For instance, several friezes – such as this one from Jor Bangla temple – shows his various avataras (incarnations). The present panel, hence, is a part of this much wider history of the rise and spread of Vaishnava bhakti as well as its state sponsored patronage across early modern South Asia.

Connections between this panel and a wider history of Vaishnava aesthetic grammar can be teased out not only through examining the contents of the depictions but also the means used to depict these narratives.  For instance, the depiction of cattle herds as a reference to Krishna’s life as a young cowherd in Vrindavana is a very common motif in Vaishnava iconography. A Rajasthani painting from the 18th century can be cited as an example of this.

In another example, the figure of Balarama, Krishna’s brother, gives us more clues. In panel below, he stands to the immediate right of the series of the three cows in the middle. He stands on one leg, while the other is folded in a way that the foot rests on the thigh. The figure is bent in three places – neck, waist and knee. He holds a horn in his left hand and a plough in his right. The upper body seems to be naked, while a dhuti covers the lower.

Balarama in the Jor Bangla frieze. Credit: Pratyay Nath

Balarama in the Jor Bangla frieze. Credit: Pratyay Nath

Drawing upon Nilakanth Joshi’s work on the historical evolution of the iconography of Balarama, we can point out that this pose was called trinata or ‘bent in three’. It had been considered a well-established and time-honoured way of representing Balarama – as well as other divinities – since the Kushana period (1st-2nd centuries AD). The horn in his hand is also an accompaniment that became increasingly distinctive of his representations in the early modern period. Two paintings – one from mid-19th century Rajasthan (fig. 8) and the other from late-19th century Bengal (fig. 9) – further illustrate that such artistic conventions were continued in subsequent times. Both these paintings show Balarama on the left and Krishna on the right. Both the figures stand in the trinata posture.

Finally, in one of the most recent incarnations of this iconography, the divine brothers stand as mirror images of each other in an iconic painting by Jamini Roy (1887-1972). Incidentally, Roy hailed from the district of Bankura as well. His birthplace, Beliatore, is less than 40 km away from Bishnupur.

‘Krishna and Balarama’ by Jamini Roy, 1930s.Credit: Pratyay Nath/

‘Krishna and Balarama’ by Jamini Roy, 1930s. Courtesy: Pratyay Nath/

What all this means is that through its use of elements from a standardised form of Vaishnava iconography and visual grammar, the Jor Bangla panel entangles the history of Bishnupur temple architecture with a wider world of Vaishnava aesthetics prevalent in South Asia across space and time.

The World of Violence

The middle panel of the Jor Bangla frieze. Credit: Pratyay Nath

The middle panel of the Jor Bangla frieze. Credit: Pratyay Nath

The middle panel of the frieze shows three long and shallow boats with dragon heads. Each of them has several people sitting and rowing the boat with pedals. There are also several people standing. They seem to be aiming at figures that resemble matchlocks (a 16th-17th century version of today’s rifle) with both hands. They are clad in what appear to be short coats and trousers. Everybody wears a sort of headgear that resembles the European hat.

Compared to the familiar cowherd figures of the first panel, these figures are more difficult to identify. Let us begin our investigation with the boats. As Jean Deloche argues, boats used in South Asia would rarely have replicas of dragon heads adorning them, although heads of other animals and birds were common. On the other hand, dragon-headed boats have historically been quite common on the rivers across Southeast Asia. From here, their use also spread to China. In medieval times, China used to import large numbers of dragon-boats from Indonesia and other parts of Southeast Asia for festivals and racing tournaments (fig. 12). These dragon-boats are also popular in our own times, as the vehicle of various boat-races organised across the world (fig. 13), especially in the USA and Canada.

It is, hence, probable that the boats portrayed in the panel originated in Southeast Asia. The question, then, becomes – how did kings and sculptors of Bishnupur in western Bengal get to know about these Southeast Asian vessels? In order to answer this question – and also identify the people on these boats – we now have to make a slight detour.

Re-routing through Kerala

In 1498, a Portuguese adventurer named Vasco da Gama reached Calicat in modern Kerala with four ships. His sea-voyage was a part of a series of naval expeditions undertaken by the kingdoms of Portugal and Spain in the 15th century. The kingdoms were searching for potential Christian allies who would help them wage war against the Ottoman Turks, who were conquering their way from West Asia into Europe around this time. They were also motivated by a search for gold and an urge to establish direct maritime commercial contact with the spice-producing regions of Asia. Following da Gama’s landfall, the Portuguese unleashed a wave of naval violence on the existing maritime networks of the Indian Ocean. Over the course of the 16th century, they managed to seize several ports and carve out a maritime empire for themselves.

Meanwhile, many Portuguese soldiers and traders fled the control of the Portuguese crown and the ambit of official maritime enterprise centred on Goa to seek fortune individually. Coastal Bengal turned into one of the most popular refuge spots for these Portuguese deserters. Starting in the mid-16th century, they settled there in large numbers, especially in the Chattagram-Sandip area of curent-day Bangladesh. Some traded and some preached, while a very large number of them took up employment with various local polities as mercenary soldiers, especially as matchlockmen. In time, they also fought in the armies of the Mughal Empire and the Arakan Kingdom of coastal northern Burma.

In the early-17th century, these Portuguese renegades – in alliance with the Arakanese – started rampant piracy and slave-raids in coastal Bengal. Portuguese and Arakanese war-boats – carrying cannons and matchlockmen – would scour the low-lying areas of the delta, loot villages and carry off thousands of people as slaves. The Arakanese would retain some of these slaves themselves and sell off the rest to the Dutch trading company, VOC. The VOC needed slave labour to work their nutmeg plantations in Southeast Asia. These incursions left a deep scar on the collective psyche of Bengali society.

Could it be that the middle panel of the frieze under consideration is a reference to these naval raids? The costumes of the men on board, which fit the description of contemporary Portuguese attire, certainly support such a proposition. The fact that all the men who are standing are aiming their guns makes it evident that this is a scene of war.

Such an argument also helps us explain the presence of the dragon-boats. As Deloche argues, such boats might have been commonly used in Arakan – located as it was in the western part of mainland Southeast Asia. It is possible that during their naval raids in coastal Bengal in the 17th century, the Portuguese and the Arakanese used some dragon-boats. Even if such raiders did not actually reach Bishnupur, it is still highly probable that tales of their devastation and descriptions of their seemingly peculiar boats travelled far.

Gunmen on dragon-boat. Detail, middle panel. Credit: Pratyay Nath

Gunmen on dragon-boat. Detail, middle panel. Credit: Pratyay Nath

In that case, this middle panel connects Malla state-formation with at least three global processes. First, through the figures of the pirates or mercenaries, it makes a reference to the history of Portuguese and Arakanese naval attacks, slave-raids, and piracy. This makes it a part of a trans-continental history involving Portuguese trade and piracy in the Indian Ocean, Arakan state-formation and territorial claims in Bengal, Dutch settlements in Southeast Asia, and a global trade in slaves and spices.
Second, the portrayal of the matchlocks links this frieze with a history of a relatively new military technology that had disseminated across South Asia over the 16th century. This technology brought about momentous changes in military affairs in vast parts of the world, including early modern Bengal.

Finally, the depiction of the dragon-headed longboats connects this panel with the larger history of navigation of Southeast Asia and China as well as the trans-regional transmission of technology. It also reflects the deep ties and close contact that Bengal historically shared with both these regions.

The World of Empire

The third and lowest panel (Fig. 15) of the frieze depicts a group of people – mostly men – who look remarkably different from those in the other two. First, each of them is wearing a kameez (long flowing upper-body tunic), pajama (pants typically worn with kameez), pointed shoes, kamar bandh (belt) and headgear. Their postures vary greatly, from carrying falcons and guns to riding horses. They immediately remind us of figures we have seen in Mughal miniature paintings. In one such mid-17th century painting, for instance, the fifth Mughal Emperor Shah Jahan (r. 1628-1658) stands on the globe. His attire is identical to that of the people in our third panel. This typical Mughal attire itself evolved over the 16th century as a fusion between Persian and Rajput attires.

Given that the Malla rajas were junior allies of the Mughal Empire, we can safely conjecture that this panel depicts a group of Mughal aristocrats. But by the early-17th century, Malla rajas themselves had entered the ranks of this aristocracy. Over the 17th and 18th centuries, they absorbed various elements of Mughal culture through a process of mutual cultural exchange. They must have looked up to their senior imperial allies for developing a stately sartorial language. It must have been as a part of this process that they started dressing up as per the Mughal aristocrats’ normative guidelines. Kumkum Chatterjee has also pointed out that the Malla elite usually dressed the Mughal way in public, especially while carrying out courtly business. The figures in the third panel, hence, could as much represent members of the Mughal elite as that of the Malla aristocracy. In the long run, the difference between the two blurred away completely, as far as public attire was concerned.

Four portraits. Mughal miniature painting, early-17th century. Credit: Pratyay Nath/ pinterest

Four portraits. Mughal miniature painting, early-17th century. Courtesy: Pratyay Nath/ pinterest

Once again, what is being depicted here is as important as how it is being depicted. In this respect, it is important to note that the posture and positioning of the figures depicted in the panel bear the unmistakable imprint of Mughal miniature painting. This can be seen in the posture of the figures, the things they hold, their alignment and so on. A look at a few Mughal miniature paintings makes this clear. One early-17th century painting from a Mughal muraqqa (album) is a perfect example. It shows four imperial mansabdars (commanders). Like the figures of the panel, they are not a part of any narrative; they are simply stand-alone figures. In both cases, these figures serve the purpose of modern ethnographic exhibits. Their function is not to tell a tale; it is to represent the normative ways of a Mughal aristocrat. Their principal importance lies not in whether they represent actual historical characters, but in the way they embody a normative prototype.

Finally, let us focus our attention on a single figure – the man in the extreme right of the panel. A close look would reveal that he is reclining against a bolster, has his legs crossed and is smoking a hookah. Once again, this bears strong similarity with figures from contemporary paintings from various other parts of South Asia. In one example, the Pahari painter Nainsukh shows his patron Raja Balwant Singh of Jasrota in a somewhat similar posture. Balwant was not quite dissimilar to the Malla rulers in being a king of modest standing from modern Himachal Pradesh. In this painting, he sits on a throne, reclining against a bolster and smoking a hookah. He is viewing a painting presented to him by Nainsukh, who stands behind him in a deferential posture. A couple of courtiers are sitting in front of the Raja, while three musicians are performing music.

This is a representation of the Raja as a patron of the arts – a quality that, as Rosalind O’Hanlon points out, increasingly came to define Mughal elite masculinity in course of the 17th century. In order to be a true Mughal aristocrat, bravery and loyalty were not enough; one also needed to demonstrate cultural refinement. Different zamindars, who were in service of the Mughal Empire, absorbed this idea – and the various accompaniments of a courtly context of such artistic patronage, like the hookah, the bolster and so on – as a part of this cultural exchange.

Bishwambhar Ray (Chhabi Biswas) in Jalsaghar (1958). Credit: Movie Screenshot

Bishwambhar Ray (Chhabi Biswas) in Jalsaghar (1958). Credit: Movie Screenshot

It is striking how the legacy of these norms has lingered on till our own times. In 1958, Satyajit Ray produced his landmark movie Jalsaghar (The Music Room) based on a short story by Tarashankar Bandyopadhyay. The protagonist Bishwambhar Ray – played masterfully by Chhabi Biswas – was an early twentieth century-Bengali zamindar with dwindling fortunes. Here too, Mughal norms informed the way in which Ray etched the character of Bishwambhar. A cultivated man brought up in a highly Mughal-ised aristocratic family of rural Bengal, Bishwambhar was shown as a man fond of horse-riding and being an unwavering patron of the arts. In several scenes of the movie, he sat in mehfils along with other aristocrats and enjoyed musical or dance recitals, reminding us of the above painting by Nainsukh. His usual posture on such occasions closely approximated the reclining figure from the Jor Bangla panel as well as the depiction of Balwant Singh.

Our panel, then, is a commentary on the ongoing process of Mughal-isation of the Malla state – something that had strong parallels in other parts of South Asia as well. Through the stylised figures of the Mughal-ised aristocrats, it lays down the normative activities of the imperial elite of the Mughal Empire, of which the Malla rajas were a part since the close of the 16th century. Dressing up in a certain way was expected of them, as was expertise in falconry, hunting, firing a gun, horseback-riding and a cultivated appreciation of the arts. By referring to these diverse cultural practices, the panel establishes a connection with yet another world – that of a trans-regional Persianised imperial culture that spread across early modern Iran, Central Asia, and South Asia.


The complete frieze. Credit: Pratyay Nath

The complete frieze. Credit: Pratyay Nath

Several points emerge from the above discussion on the frieze. Let us conclude by highlighting three. First, these terracotta temples bear testimony to the religious tolerance of Mughal imperialism – something generations of historians have pointed out in other contexts. The Mughals happened to be Sunni Muslims. Yet the Malla rajas faced no opposition from their imperial masters when it came to executing their massive temple-building project. In fact, it was under the protection and with the support of the Mughal state that the Mallas celebrated their faith in built space. This presents yet another piece of strong evidence against the portrayal of Mughal rule as a dark phase of Indian history, one where blood-thirsty Muslim tyrants went about ambushing unsuspecting Hindus and destroying their temples.

For the Malla state, these temples were as much centres for the public celebration of the state cult – Vrindavani Vaishavism – as a public exhibition of their ruling ideology and an indicator of the political and cultural processes that were shaping them. It is in this light that we need to understand the adjacent positions of the panels narrating the Mallas’ close association with Vrindavani Vaishnavism, the gradual process of their Mughalisation and their negotiations with emergenging European power in the region. In the process, the sculptures capture the multiple worlds at the conjunction of which the Mallas built their state. They also signify the burgeoning connectivity that early modernity brought to the world in its wake.

Above everything else, what these terracotta friezes represent is a spectacular celebration of plurality. One indicator of this plurality is the presence of European figures – a people traditionally regarded as mlechchha or yavana by Brahmanical social norms – among them. The other indicator is the presence of the Mughal – or Mughalised Malla – figures, which highlight the close association of the Malla state with the Sunni Muslim empire. It may seem surprising today, but the Malla rajas found absolutely no contradiction in having images of both these people sculpted on the walls of their houses of worship. Clearly, religious pride was far less fragile back then compared to what it is now in many parts of the world. In this spirit, these temples are wonderful specimens of the delightful early modern world, one that was remarkably plural, eclectic, and inclusive – sometimes more so than our own societies today.

Pratyay Nath is an assistant professor of History at Ashoka University. 


    Analysis on the communal harmony and co- optation of certain cultures and blend into a tolerant non- oppressive society. This also lends to the fact that D D Kosambi analysis of ancient rock carvings to study socio-economic and cultural integration can be extended to many parts of India. This explains the continuation of brahminism (through modified forms such as vaishnavism) by mergers and re- mergers of local culture integration. A meticulous protective shield to continue brahmin hegemony can be visualised by propagation of bhakti by the rulers patronising brahmin saints.

  • Unikrishna Panniker

    Interesting and informative article but the attempt to make the Mughals look tolerant is not convincing as a internet search quickly proves.