Two image

Homo Hierarchicus: The Origins of an Idea

In the Beginning

The Portuguese have been credited with the initial use of the term casta to refer to the social order of India, although according to at least one source the first use of the word was applied only to the lowest Indian classes in contradistinction to their overlords.1 In the early sixteenth century, the traveller Duarte Barbosa reported some features of a caste order after extended stays in India, in particular on the basis of his stay in the great kingdom of Vijayanagara. He wrote that there were “three classes of Heathen, each one of which has a very distinct rule of its own, and also their customs differ much one from the other.”2 The first, and principal, class included the king, “the great Lords, the knights and fighting men.” The second class were “Bramenes, who are priests and rulers of their houses of worship.”3 Brahmans were accorded a position of respect, and endogamy within caste units was said to be a general custom, as too the application of sanctions within castes to maintain caste discipline. Barbosa expressed no surprise or shock about Indian society, and although he reported the position of Brahmans as exalted and that of untouchables as inferior, he did not mention the varna system of the four ideal caste groups, nor did he moralize about caste one way or the other.4 Subsequent European writings about India add little to his account, and frequently comment even less on things like caste. Jean Baptiste Tavernier, a French merchant who traveled to India frequently in the middle of the seventeenth-century, reported on the reign of Aurangzeb as well as on commercial relations between India and Europe, and made some mention of Hindu beliefs, rituals, and customs. But caste was barely mentioned, except as the report of the priests who say there are seventy-two castes, though “these may be reduced to four principal [castes], from which all others derive their origin.”5

The Mughal rulers of India were aware of a formal varna scheme that could be used to encompass in textual terms the entirety of the Indian social order. Abu’l Fazl, the author of the late sixteenth-century gazetteer and administrative codebook of Akbar’s court, the A’in-i-Akbari, noted that the four varnas—the brahmans or priests, the ksatriyas or warriors, the vaisyas or merchants and agriculturalists, and the sudras or laborers and servants—had been born from the primordial body of Brahma, and that all subsequent subdivisions of caste groupings were due to intermarriage among these four original categories. But the A’in-i-Akbari spent far more time delineating the kin-based social categories that actually made up the local social order than it did commenting on caste—predictably, no doubt, given the concern of the book to systematize forms of revenue collection and local government under Mughal rule. And under direct Mughal rule, the most salient titles conferring status were those that signified a relationship to or an honor derived from the Mughal court, such as Mansabdar, Zamindar, or Bahadur.6

It was not that caste was not there—the earliest writings report Brahmans and untouchables, and touch on caste concerns in marriage practices and religious matters, among other things—but rather that it did not seem particularly striking, important, or fixed. This was not only because the European world was itself mired in hierarchical forms but also because caste did not present itself as either dominant or especially clear. When it did appear as a phenomenon that was even remotely coherent, it did so as the explanation of Brahman “informants” about the Hindu order of things, and was presented at a level of abstraction that made its formal structure appear exotic and not especially relevant. One of the first British reports of this nature came in a treatise by Alexander Dow, an officer in the East India Company’s army who had studied Persian and published a translation of a standard history of India by Firishtah (The History of Hindostan) in 1768. In his introduction to the translation, he wrote about matters he thought would complement the history, ranging from the nature of Mughal government to the character of Hindu customs, manners, and beliefs. Dow relied on the tutelage of a Brahman pandit in Banaras and adopted a textualist and Brahmanic view of Indian society. Dow was critical of Mughal ignorance of Hinduism and disregard for Brahmans, though he tells the story of how Abu’l Faz’l’s brother studied Sanskrit and the Vedas under false pretenses—an illustration of Akbar’s famed tolerance of and interest in these matters. Dow was clearly impressed by the erudition of the Brahmans with whom he conversed, and was aware of the shortcomings in his own scholarly capacity to write about them, since he had studied Persian rather than Sanskrit.

Dow reported the existence and elegance of the Vedas, and dismissed some rampant reports about Hindu society, such as that sati was widely observed, noting that it had, “for the most part, fallen into desuetude in India; nor was it ever reckoned a religious duty, as has been very erroneously supposed in the West.” But he wrote only perfunctorily about social life, which consumed a mere seven pages of text.7 His section on “caste” is only a page long. He wrote: “The Hindoos have, from all antiquity, been divided into four great tribes, each of which comprehends a variety of inferior casts. These tribes do not intermarry, eat, drink, or in any manner associate with one another, except when they worship at the temple of Jagganat in Orissa. . . . The first, and most noble tribe, are the Brahmans, who alone can officiate in the priesthood like the Levites among the Jews.” Dow then described the other three varnas, which he called tribes, by noting their formal associations: “Sittri-s” or Ksatriyas, “ought to be military men; but they frequently follow other professions,” “Bise-s” or Vaisyas “are for the most part, merchants, bankers, and bunias or shop-keepers,” and “Sudders . . . ought to be menial servants, and they are incapable to raise themselves to any superior rank.” According to Dow, all those who are excommunicated from the four tribes are, with their posterity, “for ever shut out from the society of every body in the nation,” becoming members of the Harri cast, presumably the “untouchables.” The threat of excommunication was so severe that any intermixture of blood was prevented. This, however, was all there was to it—an account clearly based on his Brahman pundit’s exposition of the basic outlines of the Dharma Sastras.

Missionary Views

The first extensive, and in the early years of the nineteenth century the most influential, European account of caste united textual formulations with empirical observation. It appeared in the text published under the name of the Abbé Dubois in 1816, Description of the Character, Manners, and Customs of the People of India, and of Their Institutions, Religious and Civil.8 I say “under the name” because the scholar Sylvia Murr has recently discovered that Dubois’s original manuscript was in large part based on an obscure manuscript written by Père Coeurdoux in the 1760s. Dubois himself went to India to escape the “ravages” of the French revolution.9 He presented “his” text to Mark Wilks—the British resident in Mysore and author of an extensive history of Mysore—in 1806. Wilks forwarded the text to the Madras government the following year, with an enthusiastic endorsement, after studying it for some time and using it for his own study of Mysore history.10 He wrote:

The Manuscript of the Abbé Dubois on Indian Casts . . . contains the most correct, comprehensive, and minute account extant in any European language of the Customs and Manners of the Hindus. Of the general utility of a work of this nature, I conclude that no doubt can be entertained. Every Englishman residing in India is interested in the knowledge of those peculiarities in the Indian casts which may enable him to conduct with the natives the ordinary intercourse of civility or business without offending their prejudices. These prejudices are chiefly known to Europeans as insulated facts, and a work which should enable us to generalize our knowledge, by unfolding the sources from which those prejudices are derived, would, as a manual for the younger servants of the Company, in particular, be productive of public advantages, on which it seems to be quite superfluous to enlarge.11

Madras bought the copyright for 2,000 star pagodas, which Dubois asked to be put in government paper, the interest from which served him for some years as a kind of regular pension. The purchase of this manuscript was reported by the Madras government to the Board of Directors of the East India Company in 1807 as “an arrangement . . . of great public importance.”12 Lord William Bentinck, the governor of Madras, explained the importance of Dubois’s work thus:

The result of my own observation during my residence in India is that the Europeans generally know little or nothing of the customs and manners of the Hindus. . . . We do not, we cannot, associate with the natives. We cannot see them in their houses and with their families. We are necessarily very much confined to our houses by the heat; all our wants and business which would create a greater intercourse with the natives is done for us, and we are in fact strangers in the land. . . . I am of opinion that, in a political point of view, the information which the work of the Abbé Dubois has to impart might be of the greatest benefit in aiding the servants of the Government in conducting themselves more in unison with the customs of the natives.13

Dubois performed an anthropological service to the British rulers of India, doing so in part because as a French Jesuit missionary he was thought to be able to cross social worlds far more readily than the imperial British themselves. But, as was true with all missionary perspectives, social worlds were crossed in order to convert souls, a social fact that led to very strong views on the subject of caste.

The manuscript was sent to London, where it was eventually translated into English and published in 1816, without benefit of the fact that at the same time Dubois was engaged in making extensive changes to the text. Copies had remained in Madras, where they were examined by a group of scholars at the College of Fort St. George, in particular A. D. Campbell, secretary of the Madras Board and a noted scholar of Dravidian languages, who had recommended revisions. The college had also attained a dictionary and a work on Hindu ethics from Dubois, but returned these as of little interest, noting “that the most useful part of the former [materials] had been compressed in the Abbé’s work on castes.”14 Dubois apparently did make major changes. He “expunged the whole of the 8th and 9th chapters of his work as too incorrect for publication, and in lieu thereof has substituted a new chapter containing about 400 pages of close writing. He has also forwarded about 300 pages of the same kind, being additions and new matter which he wishes to insert in different parts of his work.”15 Campbell further noted that “The Board in submitting this letter have directed me to explain to the Government that the alterations, additions and corrections of the Author, though considerable, appear to them the least which the work requires. There is nothing perhaps of more importance to the Hindoo community than that their distinctions of cast should be well understood by the civil officers of the government in the interior of the country, yet there is no subject at present on which it is so difficult to procure correct information.”16 The exact nature of Campbell’s concerns is unclear; what does come through from the correspondence is the fact that although Dubois’s treatise was clearly considered inadequate in some respects and deeply flawed in others, it nevertheless continued to receive the support and approbation of Company servants in Madras well after it was submitted and, judging from Murr’s account, more than fifty years after some of the material was written. The British were aware that despite their growing expertise in matters concerning land revenue issues on the one hand and linguistic and grammatical issues on the other, they still knew painfully little about local social life and customs. And given Bentinck’s description of British interactions with the colonized populations of India, it was unlikely they could do better than Dubois’s book, flawed, and apparently dated, though it was.

Many years later, in a prefatory note to the revised edition, the noted Orientalist Max Müller gave testimony to the importance of Dubois’s text by noting that it contained

the views of an eye-witness, of a man singularly free from prejudice and of a scholar with sufficient knowledge, if not of Sanskrit, yet of Tamil, both literary and spoken, to be able to enter into the views of the natives, to understand their manners and customs, and to make allowance for many of their superstitious opinions and practices, as mere corruptions of an originally far more rational and intelligent form of religion and philosophy. Few men who were real scholars have hitherto undertaken to tell us what they saw of India and its inhabitants during a lifelong residence in the country, and in spite of the great opportunities that India offers to intelligent and observant travellers, we know far less of the actual life of India than of that of Greece and Rome.17

While confessing the extraordinary limits of Orientalist knowledge in India, Müller made clear both the prestige of Dubois’s text and his supposition that he was a man free of prejudice, if firmly located in the eighteenth rather than the nineteenth centuries. But although Dubois was neither English nor a colonial servant, he was a Christian missionary, and his other writing on India, which was at the time far more controversial, concerned his observations on the potential position of Christianity in India. In 1823 he published a pamphlet entitled “Letters on the State of Christianity in India; in which the conversion of the Hindoos is considered as impracticable, to which is added a vindication of the Hindoos, male and female, in answer to a severe attack made upon both by the Reverend *****.”18 He wrote that “The Indians are a people so peculiarly circumstanced that I consider it next to impossible to make among them real and sincere Christians; the force of the prejudices and customs among them is known by all.”19

Dubois was not unimpressed by the intellectual merits of Hinduism, but was extremely critical of Brahmans, both because he saw them as the chief impediment to Christianization and because he viewed them as given over to sensual pleasures and preoccupations, the “empire of the senses.” According to the Rev. James Hough, who took issue with Dubois, “his arguments are founded upon the bad character of the Hindoos, but especially of the Brahmins—upon the extensive influence of the latter over all other castes of Hindoos—upon the nature of their superstitions and the inveteracy of their prejudices.”20 And Dubois was particularly critical of the efforts of early missionaries to use pageantry and festival to attract Hindus, as the net effect had been to interest only the “outcasts, or persons left without resources and without connexions in society; or among stupid and quite helpless fellows.”21 As we shall see, much early “colonial” ethnography was in fact written by missionaries, who observed Indian society more closely than did British officials, but experienced it in relation to their primary concern with Christian conversion. Dubois was one of the first such missionary ethnographers and, as Bentinck duly noted, a great authority at a time when the colonial administration knew so little.

Dubois began his book by describing the “caste system” in India, referring to the varna system as outlined in the Dharma Sastras. He then noted that there were many prejudices against caste, and persons who believe that “caste is not only useless to the body politic, it is also ridiculous, and even calculated to bring trouble and disorder on the people.” But he went on to say: “For my part, having lived many years on friendly terms with the Hindus, I have been able to study their national life and character closely, and I have arrived at a quite opposite decision on this subject of caste. I believe caste division to be in many respects the chef-d’oeuvre, the happiest effort, of Hindu legislation. I am persuaded that it is simply and solely due to the distribution of the people into castes that India did not lapse into a state of barbarism, and that she preserved and perfected the arts and sciences of civilization whilst most other nations of the earth remained in a state of barbarism.” His high opinion of caste was linked to his very low opinion of Hindu morality: “We can picture what would become of the Hindus if they were not kept within the bounds of duty by the rules and penalties of caste, by looking at the position of the Pariahs, or outcastes of India, who, checked by no moral restraint, abandon themselves to their natural propensities. . . . For my own part, being perfectly familiar with this class, and acquainted with its natural predilections and sentiments, I am persuaded that a nation of Pariahs left to themselves would speedily become worse than the hordes of cannibals who wander in the vast waste of Africa, and would soon take to devouring each other.” If Dubois had negative views of Brahmans, they were nothing to what he took the “outcaste Pariahs” to be. Without a Brahmanically ordered caste system, Dubois felt that India would “necessarily fall into a state of hopeless anarchy, and, before the present generation disappeared, this nation, so polished under present conditions, would have to be reckoned amongst the most uncivilized of the world.”22

Dubois’s chapter on “Pariahs” demonstrated his “upper-caste” contempt for them, as well as for those—including Europeans—who lived in any proximity to them. Thus it was that Dubois believed that a Christian mission was destined to fail if it could convert only “untouchables” and have no success at all with Brahmans. He had nothing but admiration for the great early seventeenth-century Jesuit missionary Roberto de Nobili, who took on Brahman customs and modes of life, and managed to convert a number of upper-caste Hindus to Christianity, in part because he did not insist, as many other missionaries did, that they break caste mores. But once the Catholic Church challenged this form of conversion, leading to the famous controversy about the so-called “Malabar Rites,” the Jesuits had to fall in line with stated Church policy that religious conversion should lead to social conversion, as well.23 And through his own experience, Dubois despaired even of the possibility of religious conversion, taking a far dimmer view than did de Nobili.

Dubois made a separation between caste as a kind of civil institution and Hinduism as the religious basis of it. Although he wrote of caste in approving, sometimes even glowing, terms, he made no secret of his revulsion to Hinduism as a religion. He argued that “caste regulations counteract to a great extent the evil effects which would otherwise be produced on the national character by a religion that encourages the most unlicensed depravity of morals, as well in the decorations of its temples as in its dogmas and ritual.”24 And yet caste was not unconnected to religion, since he suggested that it was the genius of caste to protect India—and by implication Hinduism—from foreign influence. “The Hindus have often passed beneath the yoke of foreign invaders, whose religions, laws, and customs have been very different from their own; yet all efforts to impose foreign institutions on the people of India have been futile, and foreign occupation has never dealt more than a feeble blow against Indian custom. Above all, and before all, it was the caste system that protected them.”25 Thus caste protected Hindus from a fall into barbarism but was also the reason why Christian missions were so woefully unlikely to succeed.

For Dubois, highest praise was reserved for the ancient Hindu lawgivers or “legislators,” who conceived the caste system and organized it in terms of sufficient sanction and conviction to prevent the possibility of systematic breakdown for millennia to come. So impressed was he that he not only explained the failure of Christianity through caste, he also recommended against substantial interference in Indian affairs on the part of the colonial rulers.

It is in the nature of the Hindus to cling to their civil and religious institutions, to their old customs and habits. . . . Let us leave them their cherished laws and prejudices, since no human effort will persuade them to give them up, even in their own interests, and let us not risk making the gentlest and most submissive people in the world furious and indomitable by thwarting them. Let us take care lest we bring about, by some hasty or imprudent course of action, catastrophes which would reduce the country to a state of anarchy, desolation, and ultimate ruin, for, in my humble opinion, the day when the Government attempts to interfere with any of the more important religious and civil usages of the Hindus will be the last of its existence as a political power.26

Dubois’s writings attained extraordinary authority for the British rulers for a number of reasons, not least because he justified British opinion that interference in social and religious customs should be kept to an absolute minimum. Even though he was a Christian missionary, he thus advocated Company policy designed precisely to keep missionaries out of India. For years, in fact, the Company prohibited as much missionary activity as it could and tried to interfere in social and religious matters as little as possible. When William Carey and John Thomas arrived in Bengal in 1793, they were not only the first Baptist missionaries in India, they were the first English clergymen who went to India as missionaries with official Company sanction. Even so, they were not encouraged to proselytize, but expected rather to restrict missionary activities to educational and linguistic pursuits, to set up schools and translate the Bible. Indeed, in the eighteenth and early nineteenth centuries, the Company actively involved itself in local religious activities, it regulated endowments for temples and other religious institutions and compelled “untouchables” and Christian converts from the lower castes to assist in Hindu festivals. In addition, European soldiers were posted in attendance at various religious rituals to bolster the appearance of Company pomp and circumstance.27 It was only because of the efflorescence of evangelical Christianity in early nineteenth-century Britain and the rise to political power of evangelical Christians such as Charles Grant and William Wilberforce that the East India Company was forced to change its policy and allow missionary work in India. After 1813, licenses were regularly granted for missionary work, and after 1833, the requirement for licensing was lifted altogether.28 But even with the growing rapprochement between missionaries and Company officials, there were still tensions between the two and major differences in attitudes toward India and opinions about how much the British government should intervene. After the Great Rebellion of 1857, official British opinion once again veered away sharply from the encouragement of any active intervention, expressly assuring Indian subjects that they would not be coerced to change their beliefs or customs. Accordingly, missionaries became even more concerned about the missed opportunities of empire and the general failure of British crown rule to take Christianization as its primary mission.

If Dubois was peculiar among missionaries in his high regard for caste and his discouragement of official intervention, he nevertheless held an opinion that became a convention among missionaries—that caste was the chief impediment to conversion. Carey and his colleagues operated the Serampore schools for some years before realizing that caste was a potential problem. Soon, however, Carey’s colleagues wrote that they saw caste as “a prison, far stronger than any which the civil tyrannies of the world have erected; a prison which immures many innocent beings.” In some despair, Carey wrote that “All are bound to their present state by caste, in breaking whose chains a man must endure to be renounced and abhorred by his wife, children and friends. Every tie that twines around the heart of a husband, father and neighbor must be torn and broken, ere a man can give himself to Christ.”29 Missionaries concluded that caste was “the most cursed invention of the devil that ever existed,” and began to insist on the renunciation of caste as a clear sign that converts had renounced idolatry, as well. That the missionaries had little success except among the lower castes fueled their resentment. And thus it was that missionaries came to accept Dubois’s characterization of caste as a system that brooked neither individual dissent nor freedom of movement. Given the preoccupation with, and history of, conversion, missionaries came to hold a special contempt for caste; their ascription to it of a totalizing power was not seen in other contemporaneous British writings.

Missionaries who felt the need to move gradually, as well as those who maintained some interest in converting Indians of upper-caste backgrounds, tended to argue that caste was principally a civil rather than a religious institution, not unlike other systems of social rank and privilege. The principal missionary society in Madras wrote to its constituents in 1809 that it “does not countenance the adherence of the Christian converts to any former religious restrictions which are not consistent with their Christian liberty, yet it cannot be in the power or wish of the Society to abolish all distinctions of ranks and degrees in India.”30 Bishop Heber, who set important mission policies in the 1820s, subscribed to this view as well, as did the “Tanjore missionaries” who argued that “caste had existed as a purely civil institution before the coming of the Brahmans, who had made of it something sacred and immutable.”31 But a report commissioned by the bishop of Madras in 1845 suggested instead that caste “distinctions are unquestionably religious distinctions, originating in, and maintained by, the operation of Hindu idolatry.” As Duncan Forrester has observed, “They defended their negative approach as being based on a new and far more profound knowledge of Hinduism and Indian society than had been available to Bishop Heber and the early missionaries.”32 By the middle of the century, missionaries came overwhelmingly to agree that caste was an unmitigated evil. In a resolution made by the Madras Missionary Conference of 1850, it was stated that caste “is one of the greatest obstacles to the progress of the Gospel in India . . . whatever it may have been in its origin, it is now adopted as an essential part of the Hindu religion.”33 Such opinion had ramifications for missionary policy, as missionaries had different views on whether caste should be tolerated in the Church and in its educational institutions, or whether the breaking of caste scruple around issues such as interdining be considered requirements for inclusion within the Christian fold (or for acceptance into Christian institutions). It was also significant in that missionaries wrote far more about caste and other social matters than did any others during the first half of the nineteenth century. Missionaries were obsessed with caste.

Caste and Early Colonial Historiography

It is not that other British commentators were entirely unimpressed by the salience of caste or its significance for a whole series of civilizational diagnoses. Late-eighteenth- and early-nineteenth-century British writings on India, however, tended to say relatively little about caste and to be formulaic at best, long after Alexander Dow. Most British writing on India in the eighteenth century concerned military matters, and reflected a painstaking and painful history of conquest, negotiation, alliance, deception, and warfare. The great hero of the century was Clive, whose military successes in Bengal and Madras had established the basis for Company control over vast sections of India’s most fertile lands. But even after the successes of Plassey and Arcot, the history of colonial conquest was highly fraught. The Mysore rulers Haidar Ali and Tipu Sultan had to be engaged four times before their final defeat in 1799, before which many commentators noted the possibility of a very different political future for the subcontinent. And the Marathas were a potent force until the military and diplomatic successes of the British in 1818. In writings about warfare and military and political intrigue, and in basic concerns about conquest and control, caste figured as of little significance. The belief that the British, like the Mughals and the Sultanates before them, had merely walked into India, facing only minimal resistance because of caste divisions, could hardly have been generated, let alone sustained, at times of open and regular military engagement.

Even when they were no longer explicitly engaged with matters of conquest, the British were concerned about land revenue and the growing need to provide regular income from agriculture to supplement the mercantile profits of international trade; this also mitigated a preoccupation with caste in favor of the village. Thomas Munro, Mark Wilks, and Charles Metcalfe all wrote eloquently about the importance of the village community. Metcalfe’s words are perhaps the most quoted: “The village communities are little republics, having nearly every thing they can want within themselves, and almost independent of any foreign relations. . . . Dynasty after dynasty tumbles down; revolution succeeds to revolution; . . . but the village community remains the same.”34 Thomas Munro, the architect of the Madras ryotwari settlement (which, unlike Cornwallis’s zamindari settlement in Bengal, was made with individual cultivators), wrote in a report from Anantapur of May 15, 1806, that “Every village, with its twelve Ayangadees as they are called, is a kind of little republic, with the Potail at the head of it; and India is a mass of such republics. The inhabitants, during war, look chiefly to their own Potail. They give themselves no trouble about the breaking up and division of kingdoms; while the village remains entire, they care not to what power it is transferred: wherever it goes the internal management remains unaltered; the Potail is still the collector and magistrate, and head farmer. From the age of Menu until this day the settlements have been made with or through the Potails.”35 The Fifth Report of 1812 quoted Munro liberally in an endorsement of the view that village government had been in place “from time immemorial,” though it also reflects the debate over land tenure in Madras as to whether individual cultivators or the headman of the village should be the agent of revenue settlements, about which much more in a later chapter.36 Charles Metcalfe and Mountstuart Elphinstone elaborated the idea in relation to the abiding belief in the evanescent but extractive presence of Oriental despotisms. Elphinstone wrote that “These communities contain in miniature all the materials of a State within themselves, and are almost sufficient to protect their members, if all governments are withdrawn”;37 and Metcalfe elaborated his comment quoted above by noting that the village communities have “nearly everything that they want within themselves, and [are] almost independent of any foreign relations.”

The focus on the village was part of the early colonial preoccupation with questions of property, landholding, and revenue collection. In late eighteenth-century Bengal, the views of John Shore, Philip Francis, and Charles Cornwallis had all led, after extensive debate, to a permanent settlement with zamindars, or landlords. This led to a massive reorganization of local power meant simultaneously to coopt the chiefs and magnates of late Mughal rule and install a loyal and landed aristocracy to serve a physiocratic vision for a new India.38 But as the immediate concerns of conquest gave way to the need for more revenue, Thomas Munro and others made their political careers by arguing for a completely different organization of revenue collection in the countryside. The colonial concern to know India began with the desire to understand local forms of landholding and agrarian management, and voluminous statistics and narratives both reflected this concern and fueled continuous arguments about the best ways to rule India and collect revenue. Suspicion of local magnates turned into a paternalist romanticism around the figure of the rural yeoman, and thus the colonial interest in local history and social organization crystallized first around images of gentry landlords, village republics, and sturdy yeomen. There were many different views, but virtually all colonial commentators were impressed by the integrity, and relative autonomy, of the village.39

Although by far the greatest number of early colonial records concerned questions of revenue and property, the Company, and numerous Company servants, were interested in discovering a more broadly based context in which to situate polemical debates over property, rent, and agrarian structure. Colin Mackenzie, who played a particularly important role in the rescuing of south India’s precolonial history (and he will play an important role in the story ahead), tried to distance himself from revenue debates, and committed himself to the collection of local texts while he engaged in his cartographic and surveying activities in the subcontinent between 1784 and his death in 1821. Significantly for our story here, he encountered very little concerning “caste” in his vast collection of local texts, traditions, and histories, nor did he comment frequently about it. In Mackenzie’s initial project of collecting representative texts, histories of places (particularly temples) and political families (and lineages) predominated. The south Indian landscape that he traversed was dotted with temples that served as convenient reference points for trigonometrical surveying and general route maps, due to the tall gopuram towers built over the gateways into temple structures that often served as centers for marketing and defense in addition to worship. Every temple had a history that inscribed the significance of its deity and the ground of the deities’ worship with a special past of miracle and power. The south Indian landscape had also been controlled by myriad little kingdoms, ranging widely in size, each with a family history for the chief or king. Thus the set of local tracts collected by Mackenzie contain literally hundreds of accounts of one lineage headman after another who managed to become a little king through a combination of strategies and successes. Frequently, of course, these stories were told as bids to become recognized as zamindars, if not as arguments per se for a zamindari-like settlement in the south.

Mackenzie’s preoccupation with local chiefs and kings was in part the result of his clear recognition of the political landscape of late precolonial peninsular India; it was also in part a reflection of the more general recognition—both military and economic—of the need to understand the native aristocracy, its immediate past, and its claims to local authority. When Mackenzie began his survey of Mysore after the defeat of Tipu Sultan in 1799, the general assumption among most East India Company officials was that a revenue settlement with the local lords or zamindars, along the lines of the 1793 Permanent Settlement in Bengal, would be the most suitable form of local governance and revenue collection for Madras presidency. Thus Mackenzie’s archival concern with the political history of the Deccan made a great deal of sense for early colonial administration because of its emphasis on the pasts and pedigrees of the potential landlords of a zamindari revenue settlement. As the consensus around the need for a ryotwari settlement grew, interest among many of Mackenzie’s old friends in the results of his labors diminished.

From my earlier work, I was aware of the prevalence of texts that concerned kings and temples. However, when I first turned to the Mackenzie collection as a repository for early ethnographic knowledge about southern India relatively uncontaminated by the official interest in land tenure, I was surprised to find very few caste histories.40 There were some general texts about castes, as also some curious lists of caste groups that resembled Borges’s Chinese encyclopedia more than later ethnographic surveys. But there were only a few specific caste histories. Those that did exist seemed of uncertain textual genre, hastily put together from the chance concerns and remarks of local subcaste headmen. Although Mackenzie occasionally mentioned the need to collect texts with information about caste, I found systematic material about caste only in his statistical and cartographic collections, as also in some of his drawings. In the statistical tables called caneeshamaris, “the population of the districts by castes, families, and villages” was carefully counted and presented by local public officials.41 Some of these tables were transcribed on his actual maps of Mysore and the Ceded Districts. Here, the compilations of population data under caste headings seemed to have the same indexical function for the map as the delineations of field types and irrigation sources. These lists were highly particularistic and idiosyncratic. Though Brahmans were usually at the head, the lists were neither highly formalized nor easy to compare across districts or regions. I was surprised by the discovery that my interest in finding early (and little mediated) texts on caste turned up so little.

Only when I turned to Mackenzie’s drawings did it seem that I had finally found caste. One of Mackenzie’s largest portfolios has eighty-two drawings depicting different groups in the northern Deccan drawn during the early years of the nineteenth century, labeled as drawings of “costume.”42 Costume was the key sign and objective focus of ethnographic difference. This emphasis on costume was in part a reflection of the fact that clothes in India (as in England) were important markers of hierarchy and difference. It was surely also because of the lack of any clear sense of what a pictorial survey of the castes and tribes would be like, as well as, perhaps, because of the influence of the cult of the picturesque, which was preoccupied with the colorful and exotic aspects of the Indian social order. The castes and groups that found their way into Mackenzie’s portfolio reveal a very particular ethnographic sensibility. There were portraits of the ancient kings of Vijayanagara, royal Darbar scenes, court servants and soldiers, and of court officials. Both in the absence of any kind of systematic and autonomous sense of a “caste system” and in the concentration of attention on characters who reflected the political landscape of the eighteenth-century Deccan—the same characters who figured in most of Mackenzie’s local texts—we can see major differences between Mackenzie’s vision of India’s ethnography and the ethnography that became canonized in the late nineteenth century.

Despite Mackenzie’s years of collecting historical materials, he never prepared a historical synthesis of his own, let alone a catalogue of his collection. And the great Orientalists—Sir William Jones, Nathaniel Halhed, Henry Thomas Colebrooke, and so on—never actually wrote histories or systematic accounts of India, either. The first such history was written instead by James Mill, a Benthamite journalist who secured lifetime employment with the East India Company soon after publishing his magisterial, and voluminous, history of British India in 1817.43 In the House of Commons, Thomas Macaulay declared it to be the greatest work to appear in English since Gibbon’s Decline and Fall of the Roman Empire. Mill’s text, unsurpassed as a general British history—and as a canonic text for the training of East India Company servants—for the rest of the century, was in part an argument for Utilitarian principles, in part a challenge to Orientalists such as Sir William Jones.44 As suggested by Javed Majeed, it “shaped a theoretical basis for the liberal programme to emancipate India from its own culture.”45 An attack on Orientalist knowledge in several senses, it held praise of India’s civilizational greatness, even if lodged firmly in the past, accountable for blocking serious attention to the need for progress and modernization. Mill had written earlier about the problems of “Oriental rhetoric on the riches of India,” criticizing Sir William Jones’s susceptibility to the “idea of Eastern wonders,” by which he meant wonders both cultural and economic.46 Throughout his history, Mill systematically debunked the claims of Orientalist scholars that there was anything of merit in India’s past: “Rude nations seem to derive a peculiar gratification from pretensions to a remote antiquity. As a boastful and turgid vanity distinguishes remarkably the oriental nations they have in most instances carried their claims extravagantly high.” Mill cautioned his reader that “the legendary tales of the Hindus have hitherto, among European inquirers, been regarded with particular respect”; but “because, without a knowledge of them, much of what has been written in Europe concerning the people of India cannot be understood,” he proceeds reluctantly to relate the mythological origins of Indian civilization.47 He made clear, however, his contempt for India’s early cultural heritage, and for its government and law in particular. As a further claim for his own authority against that of the Orientalists, he argued in his preface that his lack of knowledge of any Indian language was no disadvantage in his quest, and that his lack of firsthand experience of India—he never journeyed there—rendered his capacity to evaluate the myriad writings on and testimonies of India with the dispassionate objectivity necessary for an adequate historical account.

Mill argued his case polemically, for he was concerned that if the British nation, and government, “conceived the Hindus to be a people of high civilization, while they have in reality made but a few of the earliest steps in the progress to civilization, it is impossible that in many of the measures pursued for the government of that people, the mark aimed at should not have been wrong.” And he was clear who the culprit for this misconception had been: “It was unfortunate that a mind so pure, so warm in the pursuit of truth, and so devoted to oriental learning, as that of Sir William Jones, should have adopted the hypothesis of a high state of civilization in the principal countries of Asia.” The claim of civilization had weighty consequence, then as now. Had this been mere scholasticism, Mill’s alarm would have been less severe, but as “Sir William was actuated by the virtuous design of exalting the Hindus in the eyes of their European masters; and thence ameliorating the temper of the government,” Mill was clear that he was attacking both a general cast of mind and a set of governmental policies.48 In particular, as a good Utilitarian, Mill was concerned about Jones’s influence on legal policy, for the Company had sought to disrupt legal precedent as little as possible in the matter of personal law. But Mill was also concerned about the Company’s general commitment to the maintenance of the status quo wherever possible, whether in matters of local government, or social policy: “We have already seen, in reviewing the Hindu form of government, that despotism, in one of the simplest and least artificial shapes, was established in Hindustan, and confirmed by laws of Divine authority. We have seen likewise, that by the division of the people into castes, and the prejudices which the detestable views of the Brahmans raised to separate them, a degrading and pernicious system of subordination was established among the hindus, and that the vices of such a system were there carried to a more destructive height than among any other people.”49 In keeping with this general position, the “Anglicists”—of whom Thomas B. Macaulay, who had authored condemnations of Indian literature and learning in defense of his advocacy of the expansion of English education, was the other prime example—would have the Company overturn as much as possible of the weight of the past in order to prepare the way for new modernizing policies and institutions.

Mill’s (and Macaulay’s) critiques of Indian society were very like, in tone and content, those of the missionary Charles Grant, despite the marked political differences between the two camps of Utilitarians and Evangelicals. As early as 1796, Grant had written his “Observations on the Asiatic Subjects of Great Britain,” to make his case against prevailing Orientalist policies of respect for Indian custom, religion, and law that had been promulgated by Warren Hastings.50 He began his diatribe by asking, “Are we bound for ever to preserve all the enormities in the Hindoo system? Have we become the guardians of every monstrous principle and practice which it contains?” Like Mill, Grant believed that “the true cure of darkness is the introduction of light. The Hindoos err because they are ignorant; and their errors have never fairly been laid before them. The communication of our light and knowledge to them, would prove the best remedy for their disorders.”51 Although Grant’s conviction was that Anglicization and Christianization would bring in this light, Mill was concerned with a more general notion of modernization, one that critiqued even some aspects of English institutions. Both held, however, that the religion of the Hindus was an abomination, and believed Brahmans and Brahmanism to be responsible for social depravity. Mill wrote that “by a system of priestcraft, built upon the most enormous and tormenting superstition that ever harassed and degraded any portion of mankind, their minds were enchained more intolerably than their bodies; in short that, despotism and priestcraft taken together, the Hindus, in mind and body, were the most enslaved portion of the human race.”52 Mill’s view of caste followed, accordingly, as a prime example of an Indian institution predicated on priestcraft and adapted to despotism.

Mill’s idea of caste, however, was entirely textualist. Despite his virulent critique of the Orientalists, on the subject of caste he completely conceded their authority, though he maintained a missionary-like disdain for the institution. Like most colonial commentators in the early nineteenth century, Mill’s view of caste derived from Jones’s 1794 published translation of The Laws of Manu (Manu Dharma Sastras).53 Jones achieved his early reputation from his prodigious accomplishments in Oriental literatures and languages, and eventually proposed, on the basis of his linguistic studies, the historical kinship of Sanskrit and European languages. Jones went to India as a jurist, and among his many other achievements played a critical role in the translation, explication, and advocacy of classical canons of Indian law. His translation of Manu was the last of his legal contributions, as he died shortly after its publication, but the text became important for reasons well beyond its place in the delineation of personal law. Manu concerned such topics as the social obligations and duties of the various castes (varna) as well as of individuals at different stages of life (asrama), the proper forms of kingship, the nature of social and sexual relations between men and women of different castes, ritual practices of many kinds but mostly those connected with life-cycle transitions and domestic affairs, as well as procedures for the adjudication of different kinds of everyday quarrels and disputes. The text is about dharma, which means duty as well as law, religion as well as practice. According to Wendy Doniger and Bardwell Smith, “By the early centuries of the Common Era, Manu had become, and remained, the standard source of authority in the orthodox tradition for that centrepiece of Hinduism, varnasrama-dharma (social and religious duties tied to class and stage of life).”54 The text was the subject of nine separate commentaries, which suggests its importance over the years. Sanskritists have debated the significance of the text for actual legal practice, and it reads more like a synthetic compilation than a code of law. Whatever its historical status, however, most scholars today agree that it took on unprecedented status as an “applied” legal document only under early British rule.55 I would argue further that the canonic importance of this text for understanding the foundational nature of Indian society was an even more significant break with the past; it encapsulated British attempts to codify not just law but also social relations in a single, orthodox “Hindu”—and therefore necessarily “Brahmanic”—register. From Jones and Mill to Dumont and Marriott, Manu has taken on a general anthropological significance it could never have had before, with enormous consequences for the refashioning of basic assumptions about both religion and society.

The Manu text was self-evidently the compilation of Brahman scholars; it could hardly have been otherwise. But the canonization of the text in colonial thought has both rendered caste by definition Brahmanic and opened the “Hindu” social world to charges of the kind made by both Grant and Mill, namely, that caste society was under the exclusive domination of Brahmans who reserved for themselves not only pride of place in the caste hierarchy but such perquisites as the right to receive rather than give gifts and general exemptions from corporal punishment even if found guilty of serious crimes. Along with a text from the Rg Veda that gives a canonical origin story for the caste system (the Brahman being born from the head of Purusa, the Ksatriya from the arm, the Vaisya from the thigh, and the Sudra from the feet), the Manu text has been trotted out for the last two hundred years as the classical statement of the caste system. It provides both an originary account of the four varnas, and an explanation, through the process of intermarriage and miscegenation, for the generation of the myriad actual caste groups, or jatis, reported by every ethnographer of Indian society. Thus the text has been seen as both prescriptive and descriptive, and the two functions have frequently collapsed into one. Given the history of colonial and anthropological textualization, the text has had a life far outside its own textual confines, and certainly beyond a narrow group of scholastic Brahman jurists. For Mill, who eschewed as much as possible reference to the very texts canonized by Orientalist scholarship, the Manu text seemed unavoidably central in his account of caste. Ironically, he relied on the work of Sir William Jones to mount his own devastating critique of Jones.

Mill wrote that “On the division of the people, and the privileges or disadvantages annexed to the several castes, the whole frame of Hindu society so much depends, that it is an object of primary importance.”56 Interspersed with disparaging functionalist asides about the requirements and susceptibilities of “rude” societies, the chapter on caste reiterated a fairly standard account of the four varnas, with the law of Manu as the primary authority. Although Mill conceded that the division of society into four classes or castes represented the first step in civilization—a step taken by the Egyptians but not by the Arabs—he was contemptuous of the relative primitiveness of the system, as also its reliance on the superstitious power of the priestly caste. He wrote:

As the greater part of life among the Hindus is engrossed by the performance of an infinite and burdensome ritual, which extends to almost every hour of the day, and every function of nature and society, the Brahmans, who are the sole judges and directors in these complicated and endless duties, are rendered the uncontrollable masters of human life. Thus elevated in power and privileges, the ceremonial of society is no less remarkably in their favour. They are so much superior to the king, that the meanest Brahman would account himself polluted by eating with him, and death itself would appear to him less dreadful than the degradation of permitting his daughter to unite herself in marriage with his sovereign.57

And yet, with no trace of contradiction, Mill’s next chapter is on government, which he opened by noting that “After the division of the people into ranks and occupations, the great circumstance by which their condition, character, and operations are determined, is the political establishment, the system of actions by which the social order is preserved. Among the Hindus, according to the Asiatic model, the government was monarchical, and, with the usual exception of religion and its ministers, absolute.”58 Here Mill proposed the well-worn theory of Oriental despotism, quoting Manu to suggest the divinity of the king. By his acceptance of this account, and his use of it to describe Indian society in its entirety, Mill provided his readers with a view of caste that confused commensal and conjugal regulation with the total social order, and denigrated the role of the king, and the status of political life, in a way perfectly consonant with British interest in justifying their rule. Mill’s Utilitarian critique of colonial misrule shared far more with those he attacked than with any who would assert either the ideological or institutional importance of political rule in the period before British conquest.59 In the end, the chapter merely rehearsed his view of the rudeness of Hindu society and polity rather than the limits of the textual version of the position of Brahmans in society.

When Mountstuart Elphinstone published his two-volume history of India in 1842, he felt the need to explain why he would write a history so soon after Mill’s had seemed to set the standard for any such work and make all subsequent efforts seem redundant. Elphinstone justified his work principally on the grounds of his Indian experience, which might, he believed, “sometimes lead to different conclusions.”60 Elphinstone had been appointed to the Bengal Civil Service as early as 1796, had later been a resident in Poona before playing a prominent part in the final Maratha war, and was appointed the first governor of Bombay, a position he held from 1819 to 1827. But despite all his time in India, his anthropology was also based predominantly on the Manu Dharma Sastra. In his opening sections, entitled “State of the Hindus at the Time of Menu’s Code” and “Changes since Menu, and State of the Hindus in Later Times,” he reproduced a textual view of caste and early Indian society, writing for pages about the four varnas, the complex rules and formulations about the separation and mixing of castes, and the consequent proliferation of the myriad jatis that would later become the recognizeable caste units of contemporary ethnography. Nevertheless, Elphinstone also wrote about changes in caste, suggesting that many Brahmans had taken a worldly turn—a view not surprising for one who was the resident to the Peshwas in the last days of Maratha rule. Elphinstone also noted, more charitably than either Grant or Mill, that the institution of caste, “though it exercises a most pernicious influence on the progress of the nation, has by no means so great an effect in obstructing the enterprise of individuals as European writers are apt to suppose. There is, indeed, scarcely any part of the world where changes of condition are so sudden and so striking as in India.”61 And Elphinstone granted India the status of a major world civilization, although some of the qualifications in his admiration had to do with caste, or rather with a system in which, as he noted, the “priests, as they rose into consequence, began to combine and act in concert: that they invented the genealogy of casts, and other fables, to support the existing institutions.”62 Elphinstone’s history differed markedly from Mill’s, and he wrote with the detail and passion of one directly involved in Indian affairs. And yet it is striking how little changed was the ethnographic account, if only because, as he wrote like Bentinck before him, of the limits imposed on Englishmen, who “have less opportunity than might be expected of forming opinions of native character.”63 Once again, the sacred text was held to be reliable and important because of both the extreme variability of the Indian situation and the limited knowledge of the English about the social lives of Indians.

Elphinstone’s account was far more nuanced than Mill’s, at least in part because he sought to use his understanding of Indian history and society as the justification for the fashioning of new systems of rule and revenue collection, whereas Mill was content with nothing less than a complete break with the past.64 Although as a practical matter Elphinstone did not always disagree with the reformist recommendations of the Utilitarians, he was opposed to the Utilitarian spirit, and argued repeatedly, in broad agreement with Thomas Munro, Charles Metcalfe, and John Malcolm, for the importance of establishing governmental policy in harmony with Indian social, if not political, institutions. Thus for Elphinstone the break with Orientalist knowledge was never complete, though the preoccupations often differed markedly. But Elphinstone’s history never took on the importance of Mill’s text, and it is significant that H. H. Wilson, an accomplished Orientalist and the first Boden Professor of Sanskrit at Oxford, decided to edit Mill’s History, using this as the pretext to make myriad editorial emendations for the publication of Mill’s fifth edition, as late as 1858. Even after the Great Rebellion, Mill’s history was canonic.

Wilson, who made his initial reputation preparing a catalogue of the Mackenzie collection, soon took on the task of translating and editing classical Sanskritic texts, but he never lost his interest in history or in adapting Orientalist scholarship to questions of concern to colonial rule. His reedition (and “continuation”) of Mill’s text was a grudging acknowledgement of the extraordinary influence of Mill and his views. Nevertheless, Wilson was scathing about Mill’s ignorance of India, and his preface made it clear that the reedition was less an endorsement of Mill than an attempt to use Mill’s own status to disturb official opinion. He wrote:

Considered merely in a literary capacity, the description of the Hindus in the History of British India, is open to censure for its obvious unfairness and injustice; but in the effects which it is likely to exercise upon the connexion between the people of England and the people of India, it is chargeable with more than literary demerit: its tendency is evil; it is calculated to destroy all sympathy between the rulers and the ruled; to preoccupy the minds of those who issue annually from Great Britain, to monopolize the posts of honour and power in Hindustan, with an unfounded aversion toward those over whom they exercise that power. . . . There is reason to fear that these consequences are not imaginary, and that a harsh and illiberal spirit has of late years prevailed in the conduct and councils of the rising service in India, which owes its origin to impressions imbibed in early life from the History of Mr. Mill.65

And yet, despite these obvious differences between Wilson the Orientalist and Mill the Anglicist, it is extraordinary how similar their views were about caste. Wilson corrected Mill’s notion that Brahmans were primarily priests, or that their high status depended upon a relationship to the priesthood, and made numerous other corrections and revisions to Mill’s early chapters on the Hindus. And yet he relied on the very same textual source, the Manu Dharma Sastra, to provide the corrected evidence for an understanding of caste and Hindu society.

Caste and the Orientalists

The Orientalists lost the force of their influence soon after the opening decades of the nineteenth century, battered by the combined attacks of the Evangelicals and the Anglicists, not to mention the development of new administrative knowledge that began to develop an empirical density of its own from as early as the 1790s. They nevertheless left an important mark, not least in the canonization of certain texts as the basis on which empirical observations and Anglicist judgments about Indian society would be made.

The last major Orientalist contribution to fundamental debates concerning the character of Indian civilization and the nature of Indian society can be seen in the writings of Max Müller, whose general popularity is in retrospect perhaps less important than the fact that he strongly influenced certain Indian social reformers and nationalists, most significantly Gandhi. Müller wrote an essay on caste in 1858, just after the Great Rebellion, or in colonial reference, the “Sepoy Mutiny.”66 Müller’s essay on caste sought specifically to clarify the terms of the discussion around caste that had been vastly exacerbated by the rebellion; as he put it, “Among the causes assigned for the Sepoy mutiny, caste has been made the most prominent.” The rebellion led to passionate debate about caste, a debate that licensed missionary denunciation of the Company’s toleration of it, on the one hand, and prompted severe criticisms of missionaries for their role in alarming Indian subjects about British intentions to make them lose caste altogether, on the other.67 The debate over caste was joined by Müller with his characteristic belief that textual authority should have pride of place in official knowledge about India. He carefully distinguished himself from those missionaries who advocated an attack on caste as retribution for the revolt, while he also asserted in different ways his Christian convictions and credentials. He was clear that neither India nor regard for Indian civilization should suffer from the effects of the revolt: “Whatever the truth may be about the diabolical atrocities which are said to have been committed against women and children, a grievous wrong has been done to the people of India by making them responsible for crimes committed or said to have been committed by a few escaped convicts and raving fanatics; and . . . it will be long before the impression once created can be effaced, and before the inhabitants of India are treated again as men, and not as monsters.”68 Müller asked the fundamental question about the religious or social status of caste, and noted that there was no easy answer: “Now, if we ask the Hindus whether their laws of caste are part of their religion, some will answer that they are, others that they are not.” In characteristic Orientalist fashion, he advised resort to the texts themselves: “We are able to consult the very authorities to which the Hindus appeal, and we can form an opinion with greater impartiality than the Brahmans themselves.”69 Müller made clear his concern that religious beliefs—determined by textual standards—be respected in all events. But he also made clear that his study of the texts suggested that the caste of the Vedas and of later—degraded—periods were altogether different.

Despite his manifest approval of Vedic civilization and his increasing contempt for later developments, Müller recommended against major governmental intervention, arguing instead for a gradualist approach. His recommendation seems somewhat contradictory, though it was motivated in large part by a general sense of political caution: “It is now perceived that it will never answer to keep India mainly by military force, and that the eloquent but irritating speeches of Indian reformers must prove very expensive to the tax-paying public of England. India can never be held or governed profitably without the good-will of the natives, and in any new measures that are to be adopted it will be necessary to listen to what they have to say, and to reason with them as we should reason with men quite capable of appreciating the force of an argument. There ought to be no idea of converting the Hindus by force, or of doing violence to their religious feelings.”70 Müller revealed a combination of admiration and sensitivity, while at the same time he held the view that Indian problems were the result of degradation and corruption from the Vedic ideal, rather than related to colonial rule. Although few British administrators were compelled by such understandings, even when they accepted Müller’s advice to respect, or at least attempt not to outrage, Indian opinion, Müller’s views became influential among many Indian subjects. Most notably, his general views about Indian civilization had great significance for the development of Gandhi’s thought. Gandhi followed Müller in identifying the soul of Indian civilization as that of the Vedic age, and the distortions of later history as beginning in the time of Manu. But Gandhi could never agree with Müller, as did most other social reformers, that the social and the religious could be separated.

Despite the bitter reaction in Britain to the rebellion, and the outcry against caste among missionaries and the general public, the British government, which assumed direct rule in 1858, was obviously concerned to do nothing further to threaten the continuation of its rule over the subcontinent. In the queen-empress’s proclamation of the establishment of direct British authority over India, it was said that “We disclaim alike the right and the desire to impose our convictions on our subjects. We declare it to be our royal will and pleasure that none be in anywise favoured, none molested or disquieted, by reason of their religious faith or observances, but that all shall alike enjoy the equal and impartial protection of the law; and we do strictly charge and enjoin that all those who may be in authority under us that they abstain from all interference with the religious belief or worship of all of our subjects, on pain of our highest displeasure.”71 Although Orientalist opinion ceased from this point on to have major influence on British policy makers, a pragmatic desire to avoid further complications from such intervention, or perceptions thereof, overrode missionary wishes to use the moment to strike a fatal blow against both caste and Hinduism. And since there was general agreement among missionaries and officials alike that caste was simultaneously religious and social—given among other things the shared assumption about the perverse character of Hinduism as a religion—this also meant that the government would seek to disturb caste sensitivities, whatever they were and however they were sanctioned, as little as possible. Ironically, the most difficult challenges to the policy of nonintervention came from Indian social reformers, who throughout the nineteenth century sought to mobilize support from government and private citizens alike for major assaults on caste, in relation to issues that emerged from the treatment of women in upper castes to treatment of lower castes and in particular “untouchable” groups. In this endeavor, missionaries clearly played an important role, though what began as a concern with the impediments to conversion soon became used by others to resist both conversion and the embarrassment that accrued from missionary criticism.

If caste never succumbed to an analytical dualism that allowed the easy separation of the social and the religious, it was in part because the ideological underpinnings of separate religious and social (or political) domains had only developed—however uncertainly—in Europe from the middle of the eighteenth century, and were still imposed in ways that made little sense in Indian society, let alone in the colonial contexts that such deliberations inherently took place. England’s own secularist self-representations were irrevocably tied to Christian assumptions and ideology.72 In India, colonial rulers saw caste as the quintessential form of civil society, simultaneously responsible for India’s political weakness and a symptom of the overdevelopment of its religious preoccupations. Missionaries and officials both viewed caste, and Hinduism, from a position in which Christianity was heralded not just as the true religion but as one that allowed for genuine separations between the political and social on the one side and the religious on the other. It is true that there was both disagreement about and confusion over what caste really was—whether it was a convenient or at the very least necessary institution for empire or an impediment not just to conversion but to the moral justifications of empire, whether it should be attacked or ignored until the nationalist cause would allow it to be taken care of outside of the imperial glare, and whether it referred primarily to the textual varna scale of four orders or the empirical muddle of myriad jati groups. By 1858 there was nevertheless general recognition that caste was the foundational fact of Indian society, fundamental both to Hinduism (as Hinduism was to it) and to the Indian subcontinent as a civilizational region. Caste emerged, stronger than ever, from the legacy of Orientalist forms of knowledge.

As the Orientalists faded away, and as missionaries lost out to the imperatives of empire, British officials increasingly felt the need to find other means to answer the disturbing questions raised by their rule and the revolt against it. If questions of conquest and then revenue collection dominated the formation of official knowledge in the years between Plassey and the rebellion, questions of order and the maintenance of rule took pride of place for the next century. The last half of the nineteenth century witnessed the development of a new kind of curiosity about and knowledge of the Indian social world, exhibited first in the manuals and gazetteers that began to encode official local knowledge, then in the materials that developed around the census, which led to Risley’s great ambition for an ethnographic survey of all of India. During this same period, missionaries continued to play a role, contesting official policies of nonintervention and continuing their critique of caste and religion. But the critique of caste that was heard loudest now came from a very different place, mobilized by Indian critics and activists as varied as Rammohun Roy and Dayananda Saraswati, M. G. Ranade and G. K. Gokhale, J. G. Phule and Rabindranath Tagore, and, into the next century, M. K. Gandhi, B. R. Ambedkar, and E. V. Ramaswami Naicker. The importance of caste in the census led to increasing focus on the recognition of caste categories by the official apparatuses of government, but the critique of caste began, like the not unrelated questions around the position of women, either to disappear or be seen as a domestic issue that should be addressed only after self-rule had been instituted. Caste itself did not disappear. Instead, it seemed stronger than ever, and the massive proliferation of vernacular texts concerning caste (especially for “backward” castes) in the first two decades of the twentieth century confirmed the transformation even as they provided a principal mechanism for the mobilization of new political identities and strategies. As a result, caste continued to embarrass and to enliven debates over tradition and modernity, the relationship of civil society to religion, and the place of politics in Indian culture and the development of nationalist ideology. In moments of civilizational assertion, caste could be seen as something that had united India as a nation many years before the arrival of the British, and in moments of civilizational embarrassment caste could be held accountable for the ease of the British conquest of India itself. It was in this context that both the rise of official fascination with the centrality of caste, on the one hand, and reformist critiques of caste, on the other, would unfold.