Volume 17 - Issue 25, Dec. 9 - 22, 2000
India's National Magazine
from the publishers of THE HINDU
Table of Contents
Temple desecration in pre-modern India
The historical experience of temple desecration in pre-modern India - and, at a more general level, contested history revolving round Indo-Muslim rulers and states - has become a sensitive mass political issue in contemporary India. The demolition of the Babri Masjid, on December 6, 1992, by storm-troopers of the Sangh Parivar, and the train of communal violence and 'ill-fare' this vandalism brought to different regions of the country, propelled the issue to national centre-stage. The ideologues of the Hindu Right have, through a manipulation of pre-modern history and a tendentious use of source material and historical data, built up a dangerously plausible picture of fanaticism, vandalism and villainy on the part of the Indo-Muslim conquerors and rule rs. Part of the ideological and political argument of the Hindu Right is the assertion that for about five centuries from the thirteenth, Indo-Muslim states were driven by a 'theology of iconoclasm' - not to mention fanaticism, lust for plunder, and uncompromising hatred of Hindu religion and places of worship. In this illuminating and nuanced essay on temple desecration and Indo-Muslim states, which Frontline offers its readers in two parts, the historian Richard M. Eaton presents important new insights and meticulously substantiated conclusions on what happened or is likely to have happened in pre-modern India.
- Editor, Frontline
IN recent years, especially in the wake of the destruction of the Babri Mosque in 1992, much public discussion has arisen over the political status of South Asian temples and mosques, and in particular the issue of temples desecrated or replaced by mosqu es in the pre-British period. While Hindu nationalists like Sita Ram Goel have endeavoured to document a pattern of wholesale temple destruction by Muslims in this period,1 few professional historians have engaged the issue, even though it is a properly historical one.
Figure 1 : Silver coin of Ali Mardan (ca. 1208-1213), commemorating the conquest of Bengal by the newly-established Delhi Sultanate, in May 1204. Courtesy of G.S. Farid, Asiatic Society, Calcutta.
This essay aims to examine the available evidence with a view to asking,
* What temples were in fact desecrated in India's pre-modern history?
* When, and by whom?
* How, and for what purpose?
* And above all, what might any of this say about the relationship between religion and politics in pre-modern India? This is a timely topic, since many in India today are looking to the past to justify or condemn public policy with respect to religious monuments.
FRAMING THE ISSUE
Much of the contemporary evidence on temple desecration cited by Hindu nationalists is found in Persian materials translated and published during the British occupation of India. Especially influential has been the eight-volume History of India as Tol d by its Own Historians, first published in 1849 and edited by Sir Henry M. Elliot, who oversaw the bulk of the translations, with the help of John Dowson. But Elliot, keen to contrast what he understood as the justice and efficiency of British rule with the cruelty and despotism of the Muslim rulers who had preceded that rule, was anything but sympathetic to the "Muhammadan" period of Indian history. As he wrote in the book's original preface:
Figure 2: Stone sculpture of Narasimha I (1238-64), raja of the Eastern Ganga dynasty, standing with a sword in his belt (right). The king is leaning over a priest and worshipping the dynastic state-deity, Lord Jagannath, flanked on the right by a Siv a linga and on the left by an image of Durga. Originally in the Surya temple at Konarak, Orissa, ca. 1250; presently in the National Museum, New Delhi. Courtesy of John C. Huntington.
The common people must have been plunged into the lowest depths of wretchedness and despondency. The few glimpses we have, even among the short Extracts in this single volume, of Hindus slain for disputing with Muhammadans, of general prohibitions agains t processions, worship, and ablutions, and of other intolerant measures, of idols mutilated, of temples razed, of forcible conversions and marriages, of proscriptions and confiscations, of murders and massacres, and of the sensuality and drunkenness of t he tyrants who enjoined them, show us that this picture is not overcharged....2
With the advent of British power, on the other hand, "a more stirring and eventful era of India's History commences ... when the full light of European truth and discernment begins to shed its beams upon the obscurity of the past."3 Noting the far greater benefits that Englishmen had brought to Indians in a mere half century than Muslims had brought in five centuries, Elliot expressed the hope that his published translations "will make our native subjects more sensible of the immense advantag es accruing to them under the mildness and the equity of our rule."4
Elliot's motives for delegitimising the Indo-Muslim rulers who had preceded English rule are thus quite clear. Writing in 1931 on the pernicious influence that the colonial understanding of pre-modern Indian history had on subsequent generations, Mohamma d Habib remarked: "The peaceful Indian Mussalman, descended beyond doubt from Hindu ancestors, was dressed up in the garb of a foreign barbarian, as a breaker of temples, and an eater of beef, and declared to be a military colonist in the land where he h ad lived for about thirty or forty centuries.... The result of it is seen in the communalistic atmosphere of India today."5
Figure 3:Vijayanagara, A.D. 1430. Devarajapuram copper plate inscription showing the signature of Vijayanagara's state-deity Virupaksha, certifying a grant of land to Brahmins made by King Devaraya II (1425-1446). Virupaksha's signature, at the botto m, is in Kannada script, while the rest appears in Sanskrit. Collection of the R.S.R. Archaeological Museum, Rajahmundry, A.P. Photo by Phillip B. Wagoner.
Although penned many years ago, these words are relevant in the context of current controversies over the history of temple desecration in India. For it has been through selective translations of pre-modern Persian chronicles, together with a selective u se of epigraphic data, that Hindu nationalists have sought to find the sort of irrefutable evidence that would demonstrate a persistent pattern of villainy and fanaticism on the part of pre-modern Indo-Muslim conquerors and rulers. One of Goel's chapters is even entitled "From the Horse's Mouth." In reality, however, every scrap of evidence in this controversial matter requires careful scrutiny.
TEMPLE DESECRATION BEFORE INDO-MUSLIM STATES
It is well known that, during the two centuries before 1192, which was when an indigenous Indo-Muslim state and community first appeared in north India, Persianised Turks systematically raided and looted major urban centres of South Asia, sacking temples and hauling immense loads of movable property to power bases in eastern Afghanistan. The pattern commenced in 986, when the Ghaznavid Sultan Sabuktigin (reign 977-997) attacked and defeated the Hindu Shahi raja who controlled the region between Kabul an d northwest Punjab. According to Abu Nasr `Utbi, the personal secretary to the sultan's son, Sabuktigin "marched out towards Lamghan (located to the immediate east of Kabul), which is a city celebrated for its great strength and abounding in wealth. He c onquered it and set fire to the places in its vicinity which were inhabited by infidels, and demolishing the idol-temples, he established Islam in them".6
|Figure 4:Image of Durga seized from the Chalukyas by Rajendra I, Chola king (1012-1044), and taken to his capital. Photo: American Institute of Indian Studies, Varanasi. Courtesy of Richard H. Davis.||Figure 5:The Tamil inscription at the base of this sculpture, seized by the imperial Cholas in 1045 from their Chalukya enemies, reads: "This is the door guardian brought by Lord Vijayarajendradeva after burning (the Chalukya capital) Kalyanapuram." I nstitut Francaise d'Indologie, Pondicherry. Courtesy of Richard H. Davis.|
Linking religious conversion with conquest - with conquest serving to facilitate conversion, and conversion serving to legitimise conquest - `Utbi's brief notice established a rhetorical trope that many subsequent Indo-Muslim chroniclers would repeat.
Notwithstanding such rhetoric, however, invasions of India by Sabuktigin and his more famous son Mahmud of Ghazni (r. 998-1030) appear to have been undertaken for material reasons. Based in Afghanistan and never seeking permanent dominion in India, the e arlier Ghaznavid rulers raided and looted Indian cities, including their richly endowed temples loaded with movable wealth, with a view to financing their larger political objectives far to the west, in Khurasan.7 The predatory nature of these raids was also structurally integral to the Ghaznavid political economy: their army was a permanent, professional one built around an elite corps of mounted archers who, as slaves, were purchased, equipped, and paid with cash derived from regular infusi ons of war booty taken alike from Hindu cities in India and Muslim cities in Iran. For example, Mahmud's plunder of the Iranian city of Ray, in 1029, brought him 500,000 dinars' worth of jewels, 260,000 dinars in coined money, and over 30,000 dinars' wor th of gold and silver vessels. India, however, possessed far more wealth than the more sparsely populated Iranian plateau. Mahmud's 1026 raid on Somnath alone brought in twenty million dinars' worth of spoil.8
The dynamics of north Indian politics changed dramatically, however, when the Ghurids, a dynasty of Tajik (eastern Iranian) origins, arrived from central Afghanistan toward the end of the twelfth century. Sweeping aside the Ghaznavids, Ghurid conquerors and their Turkish slave generals ushered in a new sort of state quite unlike that of the foreign-based Ghaznavids (see Figure 1). Aspiring to imperial dominion over the whole of north India from a base in the middle of the Indo-Gangetic plain, the new De lhi Sultanate (1206-1526) signalled the first attempt to build an indigenous Muslim state and society in north India. With respect to religious policy, we can identify two principal components to this project: (a) state patronage of an India-based Sufi o rder, the Chishtis, and (b) a policy of selective temple desecration that aimed not, as earlier, to finance distant military operations on the Iranian Plateau, but to delegitimise and extirpate defeated Indian ruling houses.
The first of these policies was based on a conception of religion and politics well summarised by the Deccani court-poet `Abd al-Malik `Isami. Writing in 1350, `Isami observed that
Figure 6:Silver coin of Sultan Iltutmish (1210-1235). Courtesy of G.S. Farid, Asiatic Society, Calcutta.
the existence of the world is bound up closely with that of the men of faith. In every country, there is a man of piety who keeps it going and well. Although there might be a monarch in every country, yet it is actually under the protection of a fakir (S ufi shaikh).9
Sufis, in other words, were understood as the "real" sovereigns of Indo-Muslim states. Among all South Asian Sufi orders, moreover, the Chishtis were the most closely identified with the political fortunes of Indo-Muslim states, and especially with the p lanting of such states in parts of South Asia never previously touched by Islamic rule. The pattern began in the first half of the fourteenth century, when that order's rise to prominence among Delhi's urban populace coincided with the political expansio n of the imperial Tughluqs.10 By effectively injecting a legitimising "substance" into a new body politic at the moment of its birth, the patronage of Chishti shaikhs by governors in Tughluq provinces, or by independent rulers succeeding to po wer in those provinces, contributed positively to the process of Indo-Muslim state-building.
Equally important to this process was its negative counterpart: the sweeping away of all prior political authority in newly conquered and annexed territories. When such authority was vested in a ruler whose own legitimacy was associated with a royal temp le - typically one that housed an image of a ruling dynasty's state-deity, or rastra-devata (usually Vishnu or Siva) - that temple was normally looted, redefined, or destroyed, any of which would have had the effect of detaching a defeated raja fr om the most prominent manifestation of his former legitimacy. Temples that were not so identified, or temples formerly so identified but abandoned by their royal patrons and thereby rendered politically irrelevant, were normally left unharmed. Such was t he case, for example, with the famous temples at Khajuraho south of the Middle Gangetic Plain, which appear to have been abandoned by their Candella royal patrons before Turkish armies reached the area in the early thirteenth century.
Figure 7:Site of the Svayambhusiva temple at Warangal, demolished in 1323 by the Tughluq prince Ulugh Khan. Photo: John Henry Rice.
It would be wrong to explain this phenomenon by appealing to an essentialised "theology of iconoclasm" felt to be intrinsic to the Islamic religion. It is true that contemporary Persian sources routinely condemned idolatry (but-parasti) on religio us grounds. But it is also true that attacks on images patronised by enemy kings had been, from about the sixth century A.D. on, thoroughly integrated into Indian political behaviour. With their lushly sculpted imagery vividly displaying the mutual inter dependence of kings and gods and the commingling of divine and human kingship, royal temple complexes of the early medieval period were thoroughly and pre-eminently political institutions. It was here that, after the sixth century, human kingship was est ablished, contested, and revitalised. Above all, the central icon housed in a royal temple's "womb-chamber," and inhabited by the state-deity of the temple's royal patron, expressed the shared sovereignty of king and deity (see Figures 2 and 3).
Moreover, notwithstanding that temple priests endowed a royal temple's deity with attributes of transcendent and universal power, that same deity was also understood as having a very special relationship, indeed a sovereign relationship, with the particu lar geographical site in which its temple complex was located. As revealed in temple narratives, even the physical removal of an image from its original site could not break the link between deity and geography.11 "A divine power," writes Davi d Shulman, "is felt to be present naturally on the spot."12 The bonding between king, god, temple, and land in early medieval India is well illustrated in a passage from the Brhatsamhita, a sixth century text: "If a Siva linga, i mage, or temple breaks apart, moves, sweats, cries, speaks, or otherwise acts with no apparent cause, this warns of the destruction of the king and his territory."13 In short, from about the sixth century on, images and temples associated with dynastic authority were considered politically vulnerable.
Given these perceived connections between temples, images, and their royal patrons, it is hardly surprising that, as Richard H. Davis has recently shown,14 early medieval Indian history abounds in instances of temple desecration that occurred amidst inter-dynastic conflicts. In 642 A.D., according to local tradition, the Pallava king Narasimhavarman I looted the image of Ganesha from the Chalukyan capital of Vatapi. Fifty years later armies of those same Chalukyas invaded north India and brou ght back to the Deccan what appear to be images of Ganga and Yamuna, looted from defeated powers there. In the eighth century Bengali troops sought revenge on king Lalitaditya by destroying what they thought was the image of Vishnu Vaikuntha, the state-d eity of Lalitaditya's kingdom in Kashmir.
In the early ninth century, the Rashtrakuta king Govinda III invaded and occupied Kanchipuram, which so intimidated the king of Sri Lanka that he sent Govinda several (probably Buddhist) images that had represented the Sinhala state, and which the Rashtr akuta king then installed in a Saiva temple in his capital. About the same time, the Pandyan king Srimara Srivallabha also invaded Sri Lanka and took back to his capital a golden Buddha image that had been installed in the kingdom's Jewel Palace. In the early tenth century, the Pratihara king Herambapala seized a solid gold image of Vishnu Vaikuntha when he defeated the Sahi king of Kangra. By the mid-tenth century, the same image was seized from the Pratiharas by the Candella king Yasovarman and instal led in the Lakshmana temple of Khajuraho.
In the early eleventh century, the Chola king Rajendra I furnished his capital with images he had seized from several prominent neighbouring kings: Durga and Ganesha images from the Chalukyas; Bhairava, Bhairavi, and Kali images from the Kalingas of Oris sa; a Nandi image from the Eastern Chalukyas; and a bronze Siva image from the Palas of Bengal (see Figure 4). In the mid-eleventh century, the Chola king Rajadhiraja defeated the Chalukyas and plundered Kalyani, taking a large black stone door guardian to his capital in Thanjavur, where it was displayed to his subjects as a trophy of war (see Figure 5). In the late eleventh century, the Kashmiri king Harsha even raised the plundering of temples to an institutionalised activity; and in the late twelfth and early thirteenth century, while Turkish rulers were establishing themselves in north India, kings of the Paramara dynasty attacked and plundered Jain temples in Gujarat.15
This pattern continued after the Turkish conquest of India. In the 1460s, Kapilendra, the founder of the Suryavamshi Gajapati dynasty in Orissa, sacked both Saiva and Vaishnava temples in the Cauvery delta in the course of wars of conquest in the Tamil c ountry.16 Somewhat later, in 1514, Krishnadevaraya looted an image of Balakrishna from Udayagiri, which he had defeated and annexed to his growing Vijayanagara state. Six years later he acquired control over Pandharpur, where he seems to have looted the Vittala image and carried it back to Vijayanagara, with the apparent purpose of ritually incorporating this area into his kingdom.17
Although the dominant pattern here was one of looting royal temples and carrying off images of state-deities, we also hear of Hindu kings engaging in the destruction of the royal temples of their political adversaries. In the early tenth century, the Ras htrakuta monarch Indra III not only destroyed the temple of Kalapriya (at Kalpa near the Yamuna River), patronised by the Rashtrakutas' deadly enemies, the Pratiharas, but also took special delight in recording the fact.18
IMPERIALISM OF THE DELHI SULTANATE, 1192-1394
In short, it is clear that temples had been the natural sites for the contestation of kingly authority well before the coming of Muslim Turks to India. Not surprisingly, Turkish invaders, when attempting to plant their own rule in early medieval India, f ollowed and continued established patterns. The table and the corresponding maps in this essay by no means give the complete picture of temple desecration after the establishment of Turkish power in Upper India. Undoubtedly some temples were desecrated b ut the facts in the matter were never recorded, or the facts were recorded but the records themselves no longer survive. Conversely, later Indo-Muslim chroniclers, seeking to glorify the religious zeal of earlier Muslim rulers, sometimes attributed acts of temple desecration to such rulers even when no contemporary evidence supports the claims. As a result, we shall never know the precise number of temples desecrated in Indian history.
Nonetheless, by relying strictly on evidence found in contemporary or near-contemporary epigraphic and literary sources spanning a period of more than five centuries (1192-1729), one may identify eighty instances of temple desecration whose historicity a ppears reasonably certain. Although this figure falls well short of the 60,000 claimed by some Hindu nationalists,19 a review of these data suggests several broad patterns.
First, acts of temple desecration were nearly invariably carried out by military officers or ruling authorities; that is, such acts that we know about were undertaken by the state. Second, the chronology and geography of the data indicate that acts of te mple desecration typically occurred on the cutting edge of a moving military frontier. From Ajmer in Rajasthan, the former capital of the defeated Cahamana Rajputs - also, significantly, the wellspring of Chishti piety - the post-1192 pattern of temple d esecration moved swiftly down the Gangetic Plain as Turkish military forces sought to extirpate local ruling houses in the late twelfth and early thirteenth century (see Table and Map 1: nos. 1-9). In Bihar, this included the targeting of Buddhist monast ic establishments at Odantapuri, Vikramasila, and Nalanda. Detached from a Buddhist laity, these establishments had by this time become dependent on the patronage of local royal authorities, with whom they were identified. In the 1230s, Iltutmish carried the Delhi Sultanate's authority into Malwa (nos. 10-11), and by the onset of the fourteenth century the Khalji sultans had opened up a corridor through eastern Rajasthan into Gujarat (nos. 12-14, 16-17).
Delhi's initial raids on peninsular India, on which Khalji rulers embarked between 1295 and the early decades of the fourteenth century (nos. 15, 18-19), appear to have been driven not by a goal of annexation but by the Sultanate's need for wealth with w hich to defend north India from Mongol attacks. In 1247, Balban, the future sultan of Delhi, had recommended raiding Indian states for precisely this purpose.20 For a short time, then, peninsular India stood in the same relation to the North - namely, as a source of plunder for financing distant military operations - as north India had stood in relation to Afghanistan three centuries earlier, in the days of Sabuktigin and Mahmud of Ghazni. After 1320, however, a new north Indian dynasty, the Tughluqs, sought permanent dominion in the Deccan, which the future Sultan Muhammad bin Tughluq established by uprooting royally patronised temples in western Andhra, most prominently the Svayambhusiva complex in the centre of the Kakatiyas' capital city of Warangal (nos. 20-22. See Figure 7). Somewhat later Sultan Firuz Tughluq did the same in Orissa (no. 23).
THE GROWTH OF REGIONAL SULTANATES, 1394-1600
From the late fourteenth century, after the tide of Tughluq imperialism had receded from Gujarat and the Deccan, newly emerging successor states sought to expand their own political frontiers in those areas. This, too, is reflected in instances of temple desecration, as the ex-Tughluq governor of Gujarat and his successors consolidated their authority there (see Map 2: nos. 25-26, 31-32, 34-35, 38-39, 42), or as the Delhi empire's successors in the South, the Bahmani sultans, challenged Vijayanagara's c laims to dominate the Raichur doab and the Tamil coast (nos. 33, 41). The pattern was repeated in Kashmir by Sultan Sikandar (nos. 27-30), and in the mid-fifteenth century when the independent sultanate of Malwa contested renewed Rajput power in eastern Rajasthan after Delhi's authority there had waned (nos. 36-37).
In the early sixteenth century, when the Lodi dynasty of Afghans sought to reassert Delhi's sovereignty over neighbouring Rajput houses, we again find instances of temple desecration (nos. 43-45). So do we in the late sixteenth and early seventeenth cent uries, when the Bahmani Kingdom's principal successor states, Bijapur and Golconda, challenged the territorial sovereignty of Orissan kings (nos. 55, 59; Maps 2 and 3), of Vijayanagara (no. 47), and of the latter's successor states - especially in the so uthern Andhra country (nos. 50-51, 53-54, 60-61; Maps 2 and 3).
Unlike the Deccan, where Indo-Muslim states had been expanding at the expense of non-Muslim states, in north India the Mughals under Babur, Humayun, and Akbar - that is, between 1526 and 1605 - grew mainly at the expense of defeated Afghans. As non-Hindu s, the latter had never shared sovereignty with deities patronised in royal temples, which probably explains the absence of firm evidence of temple desecration by any of the early Mughals, in Ayodhya or elsewhere. The notion that Babur's officer Mir Baqi destroyed a temple dedicated to Rama's birthplace at Ayodhya and then got the emperor's sanction to build a mosque on the site - the Babri Masjid - was elaborated in 1936 by S.K. Banerji. However, the author offered no evidence that there had ever been a temple at this site, much less that it had been destroyed by Mir Baqi. The mosque's inscription records only that Babur had ordered the construction of the mosque, which was built by Mir Baqi and was described as "the place of descent of celestial bein gs" (mahbit-i qudsiyan). This commonplace rhetorical flourish can hardly be construed as referring to Rama, especially since it is the mosque itself that is so described, and not the site or any earlier structure on the site.21
However, whenever Mughal armies pushed beyond the frontiers of territories formerly ruled by the Delhi sultans and sought to annex the domains of Hindu rulers, we again find instances of temple desecration. In 1661 the governor of Bengal, Mir Jumla, sack ed the temples of the neighbouring raja of Cooch Bihar, who had been harassing the northern frontiers of Mughal territory (no. 64; Map 3). The next year, with a view to annexing Assam to the imperial domain, the governor pushed far up the Brahmaputra val ley and desecrated temples of the Ahom rajas, replacing the principal one at Garhgaon with a mosque (nos. 65-66).
All of these instances of temple desecration occurred in the context of military conflicts when Indo-Muslim states expanded into the domains of non-Muslim rulers. Contemporary chroniclers and inscriptions left by the victors leave no doubt that field com manders, governors, or sultans viewed the desecration of royal temples as a normal means of decoupling a former Hindu king's legitimate authority from his former kingdom, and more specifically, of decoupling that former king from the image of the state-d eity that was publicly understood as protecting the king and his kingdom. This was accomplished in one of several ways. Most typically, temples considered essential to the constitution of enemy authority were destroyed. Occasionally, temples were convert ed into mosques, which more visibly conflated the disestablishment of former sovereignty with the establishment of a new one.
The form of desecration that showed the greatest continuity with pre-Turkish practice was the seizure of the image of a defeated king's state-deity and its abduction to the victor's capital as a trophy of war. In February 1299, for example, Ulugh Khan sa cked Gujarat's famous temple of Somnath and sent its largest image to Sultan `Ala al-Din Khalji's court in Delhi (no. 16; Map 1). When Firuz Tughluq invaded Orissa in 1359 and learned that the region's most important temple was that of Jagannath located inside the raja's fortress in Puri, he carried off the stone image of the god and installed it in Delhi "in an ignominious position" (no. 23). In 1518, when the court in Delhi came to suspect the loyalty of a tributary Rajput chieftain in Gwalior, Sultan Ibrahim Lodi marched to the famous fortress, stormed it, and seized a brass image of Nandi evidently situated adjacent to the chieftain's Siva temple. The sultan brought it back to Delhi and installed it in the city's Baghdad Gate (no. 46; Map 2).
Similarly, in 1579, when Golconda's army led by Murahari Rao was campaigning south of the Krishna River, Rao annexed the entire region to Qutb Shahi domains and sacked the popular Ahobilam temple, whose ruby-studded image he brought back to Golconda and presented to his sultan as a war trophy (no. 51). Although the Ahobilam temple had only local appeal, it had close associations with prior sovereign authority since it had been patronised and even visited by the powerful and most famous king of Vijayanag ara, Krishnadevaraya. The temple's political significance, and hence the necessity of desecrating it, would have been well understood by Murahari Rao, himself a Marathi Brahmin.22
In each of these instances, the deity's image, taken as war trophy to the capital city of the victorious sultan, became radically detached from its former context and in the process was transformed from a living to a dead image. However, sacked images we re not invariably abducted to the victor's capital. In 1556, the Gajapati raja of Orissa had entered into a pact with the Mughal emperor Akbar, the distant adversary of the sultan of Bengal, Sulaiman Karrani. The raja had also given refuge to Sulaiman's more proximate adversary, Ibrahim Sur, and offered to assist the latter in his ambitions to conquer Bengal and overthrow the Karrani dynasty. As Sulaiman could hardly have tolerated such threats to his stability, he sent an army into Orissa which went st raight to the Gajapati kingdom's state temple of Jagannath and looted its images. But here the goal was not annexation but only punishment, which might explain why the Gajapati state images were not carried back to the Bengali capital as trophies of war. 23
Whatever form they took, acts of temple desecration were never directed at the people, but at the enemy king and the image that incarnated and displayed his state-deity. A contemporary account of a 1661 Mughal campaign in Cooch Bihar, which resulted in t he annexation of the region, states that the chief judge of Mughal Bengal was ordered to confiscate the treasure of the defeated raja, Bhim Narayan, and to destroy the image of the state-deity. But the judge himself issued orders against harming the gene ral population, warning that if any soldiers were caught touching the property of the common people, their hands, ears, or noses would be removed.24 In short, in newly annexed areas formerly ruled by non-Muslims, as in the case of Cooch Bihar, Mughal officers took appropriate measures to secure the support of the common people, who after all created the material wealth upon which the entire imperial edifice rested.
If the idea of conquest became manifest in the desecration of temples and images associated with former enemies - itself an established tradition in pre-Turkish Indian practice - what happened once the land and the subjects of those same enemies were int egrated into an Indo-Muslim state? This question, together with the pattern of temple desecration under the imperial Mughals, will be taken up in the second of this two-part essay.
Richard M. Eaton teaches South Asian history at the University of Arizona, U.S., and is the author of several books on pre-modern India, including The Rise of Islam and the Bengal Frontier, 1204-1760 (New Delhi: Oxford University Press, 1997) and Essays on Islam and Indian History (New Delhi: Oxford University Press, 2000).