Introduction
In the vast literature on capitalism and abolition stretching back to Eric Williams, South Asian social history is conspicuously absent.Footnote 1 This can partly be explained by the fact that the study of slavery and capitalism was an Atlantic-centered field for much of the twentieth century. In recent decades, however, commercial slavery in the Indian Ocean world has become a prominent field of inquiry.Footnote 2 New Histories of Capitalism and studies of “racial capitalism” have contributed further to a widening of our spatial visions and provoked debates about the global, imperial, and racial dimensions of modern capitalism.Footnote 3
Nevertheless, the new literature on global histories of slavery has not inspired attempts to write global histories of antislavery. Some scholars have called for a better understanding of the slow “drift in community values” that drives authorities to redefine what constitutes coercion, and the need for a global approach to antislavery has occasionally been acknowledged.Footnote 4 Yet, the default interpretation of antislavery in South Asia is one that reads it entirely as a function of the contingent requirements of colonial governance.Footnote 5 Such an approach ignores the possibility that abolition in colonial contexts, despite its limits, had a genuine ethical purchase on the lives of those it affected. Agrarian social relations in the subcontinent were grounded in practices of caste, which involved differential access to markets, land, and water, the control of (bonded) labor, and the regulation of sexuality. As a modality of power, caste was “sited at a depth within the indigenous society, well beyond the reach of the disciplinary arm of the colonial state.”Footnote 6 This dimension has often been concealed, as this essay will also show, through a blending of “indigenous feudal ideology…with colonialist anthropology.”Footnote 7
Hence, evacuating abolition in India of its ideological content in toto produces a major historiographical blind spot.Footnote 8 It occludes from vision longer-running South Asian social histories and the ethical-political valence of abolition in a context shaped by caste and agrarian slavery. Building on Thomas Holt’s insightful distinction between abolitionism as a social ideology irreducible to class interest, and abolitionist “outcomes” produced by elite-imperial policy, the following sections gradually unpack my key argument and its comparative implications.Footnote 9 The next section situates the essay within the broader historiography of capitalism and abolition, and motivates the need to incorporate the specificities of caste and slavery in South Asia into this literature. Section III presents the evidence: a series of legal conflicts and the testimonies of alleged “slaves” in the eastern Indian province of Bengal in the 1820s and 1830s. These disputes show, I argue in Section IV, that slavery was neither a matter of “kinship” nor “gentle” and harmonious.Footnote 10 Agrarian slavery existed in both Mughal South Asia and during the early British colonial era (1765–1858), and there were political struggles over the legal status of slavery in the early nineteenth century. Such aspirations for liberty cannot be explained by colonial state-building alone. Acts of defiance, which included claims of having been “wage” workers rather than “slaves,” and an insistence on the right of property to homesteads, were challenges to the practical basis of caste. Though not textually articulated as such, reading the actions of alleged slaves with reference to a larger structure of inference enables us to better grasp the emergence of an anti-caste and eventually Dalit political subjectivity in South Asia from the 1860s onward.Footnote 11
The ethical-political potential of abolition, however, was undermined by abolitionist policies enacted by a colonial state deeply invested in “order” and revenue extraction. Indeed, Act V of 1843, which put an end to the legal recognition of slavery but did not actively emancipate the enslaved, can scarcely be called “abolition.” Many historians have recognized and emphasized this point, but the larger, global capitalist context of counter-revolution has not received its due. When “abolition” was effected in India, the forces of reaction were ascendant throughout the wider imperial world and within Europe and Britain.Footnote 12 Colonial officials who governed India also opposed democratic, working-class movements in Britain. “Reform” in the metropole was an attempt to secure the influence of the landed elite in Parliament and to rein in more radical possibilities.Footnote 13 Recognizing the need to situate colonial Jamaica within this broader context, Thomas Holt argued that the gradual rise of racist discourses must be historicized with reference to a context where mass political liberty and the control of labor had become a problem for an entrenched (hence counter-revolutionary) bourgeois-capitalist class.Footnote 14
I make a similar argument for South Asia by historicizing the wider context of abolition and the colonial archive that is my evidentiary basis. I neither read the archive uncritically nor reject it tout court. Instead, the essay grounds both the possibilities and limits of abolition in the contradictions of liberalism and global capitalism, and asks where South Asia fits within broader histories of capitalism, revolution, and counter-revolution.
Abolition, Capitalism, and the Question of Ideology
The debates over antislavery in the Atlantic world were intense and impassioned, but all parties agreed on a fundamental point: abolitionism was a major ideological shift, and it was one, if not the most significant, change that set apart the Age of Revolution from preceding epochs. The animating question of the debate was simple yet sweeping: given the remarkable acceptance and stability of slavery across several millennia, how can we explain the emergence of abolitionism and its political success in the late eighteenth and early nineteenth centuries? No one denied that capitalism had something to do with it, but disagreements abounded about how to conceptualize the link and how to judge the historical and ethical legacy of abolition.
Eric Williams proposed an elegant, two-part thesis: Atlantic slavery financed British industrialization and was thus foundational for British capitalism; abolition occurred when slavery was no longer profitable and capitalists pushed for “free” wage labor.Footnote 15 Both halves of the thesis were challenged. Economic historians showed that profits from slavery were far less important for British industrialization than Williams assumed, and mechanization was driven by other factors including domestic wage levels and contingent scientific advances.Footnote 16 Seymour Drescher, on the other hand, demonstrated that abolition occurred at a time when profits from slavery were at an all-time high.Footnote 17 Abolition, in short, was “econocide.”
Some years later, Thomas Haskell re-opened the question of the link between capitalism and humanitarianism in an extended, two-part essay.Footnote 18 He argued that abolition became a practical possibility because the expansion of abstract, formally free market relations enlarged people’s capacity to recognize extended chains of moral responsibility. Thomas Bender rightly observed, however, that it was unclear exactly what Haskell meant by the “market,” and whether he was claiming a new way of valuing, or new values. Footnote 19 The link between capitalism and humanitarianism remained thin and abstract, and the contingent, political trajectory of abolitionism was left underexamined. Haskell pleaded that he was making a case for “enabling conditions” rather than inevitability, but one could be forgiven for thinking that he was smuggling in economic determinism through the back door.
By contrast, David Brion Davis, in his two magisterial works on slavery and abolition, emphasized both the newness of abolitionism and its moral-political force as an ideology. Footnote 20 He showed that a fundamental component of the experience and world-view of the emerging capitalist class that led the abolitionist movement was the question of how to define “freedom” in a rapidly changing world. “Slavery” was a clear-cut and concrete foil, which could be used to symbolize “all the forces that threatened the true destiny of man.”Footnote 21 The ideology of freedom and a principled antislavery position, therefore, could not be reduced to the economic interests of capitalists, but nor could the political trajectory of abolitionism be understood without reference to the actions of elite figures.
Among the early contributors to the debate, Seymour Drescher’s and Thomas Holt’s works offer further insights into how the gap between enabling discourses and actual outcomes might be bridged. Drescher showed that abolitionism became a mass social movement in Britain because it was widely supported by the working classes.Footnote 22 Their empathy for the enslaved was rooted in experiences of exploitation as proletarian workers: they used antislavery ideals to resist further proletarianization and not to uncritically support capitalist expansion. Stadial notions of capitalist development rooted in methodological nationalism, therefore, are ill-equipped to grasp the historical and political complexity of abolitionism, which could have found relevance in purportedly “advanced” as well as “backward” regions.Footnote 23
Holt sharpened these insights by distinguishing between abolitionism as a social movement and ideology, and abolitionist outcomes shaped by elite politicians and imperial administrators. This enabled him to argue that antislavery was a revolutionary ideology, and even an abolition tempered by a period of “apprenticeship” had a positive impact on the lives of those emancipated. Nevertheless, imperial policy driven by elite interests was always responding to changed contexts. At a historical moment when mass political liberty had become a problem for the entrenched bourgeois-capitalist class in both the metropole and the colonies, and when the need to motivate “lazy” workers had become a recurring theme, emancipated laborers in Jamaica could not be allowed to become independent freeholders exercising franchise. Hence, biologistic theories of “race” became the alibi for repression and disenfranchisement.
Davis, Drescher, and Holt provide a comparative framework for studying abolition beyond the Atlantic world. Yet, the antislavery debate has rarely been invoked when thinking about questions of caste, slavery, and abolition in South Asia. One reason for this is that the contemporary literature on capitalism, slavery, and empire views “capitalism” in primarily negative terms, as historically tied only and often necessarily to slavery, extractive empires, “racialization,” and new forms of gendered domination. This tendency is discernible across the board in the new histories of capitalism, in the large literature on global slavery beyond the Atlantic world, and in the much smaller literature on global abolitionism.Footnote 24 The dialectical approach of an earlier era has been jettisoned in favor of one-sided formulations that threaten to erase both the newness and the ideological force of abolitionism in the Age of Revolution.
The existence of widespread commercial slavery in the early modern Indian Ocean world is now universally acknowledged, but we are only beginning to understand how caste was implicated in this process. Some scholars have shown that, in practice, caste and slavery were interrelated, but the social history of caste beyond its colonial framing has been left underexamined.Footnote 25 Moreover, “resistance” is conceptualized far too often as challenges only to extractive European empires.Footnote 26 By leaving open the question of whether resistance was also directed at contradictions within indigenous society, and, if so, with what content, historians have left open the possibility of an uncritical affirmation of pre-colonial South Asian societies.
One example can illustrate the stakes. In the now-established and vibrant field of Dalit studies, scholars often emphasize that Dalits found “colonial” liberalism and modernity useful for addressing their own political concerns, including the problem of representation.Footnote 27 It is also true, however, that the colonial state always allied with upper-caste landed elites to extract revenue and control Dalit labor in the nineteenth century, and it supported Dalit political claims in the early twentieth century solely to undermine nationalist claims to unity.Footnote 28 Moreover, by the 1910s, liberal arguments about property could easily be mobilized to justify the status quo, that is, a caste-differentiated distribution of property.Footnote 29 Therefore, to avoid the extremes of either a one-sided, ahistorical defense of liberalism or an equally one-sided rejection of it tout court, we must historicize both the potentials and the failures of liberty in South Asia. By combining insights from the Atlantic-centered antislavery debates with the extensive literature on caste, slavery, and agrarian social relations within South Asia, the next two sections sketch the possible outlines for such a historicization.
The Evidence: Aspirations for Liberty and the Struggle over the Legal Status of Slavery in Bengal, ca. 1813–1843
On 9 September 1826, in the civil court of Dhaka in eastern Bengal, Golak Narayan Ray filed a suit against Kewal Ram Deo and sixteen others. Ray claimed that Kewal Ram and the others were the descendants of Dakai, Puchai, and Manai, the “hereditary slaves” of Ray’s ancestors.Footnote 30 According to Ray, the defendants and their ancestors had rendered service as slaves to him and his ancestors for generations, “being supported by lands assigned.” The defendants had recently left Ray’s estate and relocated elsewhere. Ray argued that by local custom they remained his hereditary slaves and sued to recover them. The court accepted the suit, and a summons was issued to the defendants.
Appearing for the defendants, Jagannath and Sambunath claimed that their ancestors were rent-paying tenants in the landed estate of Ray and that their “profession” was “service.” Their father was an agent in the estate of the Ray family, and they had never held nankar lands for sustenance.Footnote 31 At present, the defendants paid Ray, as their “superior landlord,” the “yearly rent of 358 Rupees and 9 annas.” As such, they had full property rights over their person and labor, and Ray’s claim about desertion did not stand. The cause of the complaint, the defendants insisted, was their refusal to hand over to Ray a plot of land that the landlord wished to include in his garden, but in which the defendants had a dwelling of their own. Out of resentment, Ray had brought legal action.
Two other defendants, Kewal Ram and Deb Chand, proffered similar defenses, claiming to be tenants rather than slaves. On 23 May 1828, the case came to trial before Mr. D. B. Morrieson. On this occasion, Golak Ray produced an ikrar (pledge or promise) from 1790, which had been executed by Dakai, Puchai, and Manai, and which bore the signature of Mr. W. Douglas, the collector of Jalalpur at the time. Based on this evidence and the testimony of witnesses, Mr. Morrieson passed judgment in favor of the plaintiff and ordered the defendants to be made over to Ray as his slaves. In his judgment, Morrieson pointed out that in the ikrar of 1790, Dakai and his brothers had admitted that they were “hereditary slaves” of the plaintiff’s family. It also showed that Ray’s grandfather had “bought the taluka (tenancy) in their name because they were slaves,” that he “fixed the yearly rent at 370 Rupees 2 annas, and that after allowing them nine rupees from the established rent assets as their nankar for services as slaves, he made it over to their charge.” Further, the ikrar contained a clause that Dakai and his brothers and their descendants would “continue to render service of slaves to the plaintiff,” and that in case of default, the plaintiff had the right to sue.
The defendants then filed an appeal at the Provincial Court of Appeal, and their case was heard on 9 November 1829, by Mr. W. Cracoft. The judge reversed the decree of the lower court, and he even cast doubt on the ikrar of 1790 that had been the key evidentiary basis of the previous judgment. Cracoft found the deed suspicious, doubted its provenance, and concluded that the defendants were more likely tenants than slaves. His position, however, was contradicted by Mr. Charles Smith, a judge of the same court, who next heard the case on 29 December 1829.
Smith believed the ikrar to be true, and he saw no reason to question its veracity. His sweeping comments demonstrate the presumptive attitudes of many judges: “Slavery of a family may be inferred from continuous service; and it seldom happens, that after the lapse of many years, the original title, shewing acquisition of the slave’s forefather, is forthcoming. In this part of the country, many slaves are apparently persons of respectability and educated, and manage the zemindari [estate] affairs of their masters. But this constitutes no ground for emancipation. In a word, the appellants and their ancestors are the hereditary slaves of the respondent, and if discharged, notwithstanding proof of their servile state, most slaves will become recusant, and on various pretexts will find means to effect their discharge.”Footnote 32
Due to such disagreements among judges, the case went to the Murshidabad Court of Appeal to be heard by a third judge. On 20 April 1830, Mr. C. W. Steer confirmed the judgment of Mr. Smith, and the appellants then requested the highest civil court—the Sadar Dewani Adalat (SDA)—for permission to file a special appeal. This was granted on 21 June 1830, by Mr. Alexander Ross and Mr. R. H. Rattray.
The tone of the SDA judges was diametrically opposed to many in the lower courts. Mr. Rattray, on 26 March 1832, concurred with Mr. Cracoft and with the grounds of his judgment. Moreover, when the lawyers of Ray pointed out that one of the appellants, Kalikinkar, had died, and insisted that his heirs be summoned to follow up on the appeal, Mr. Rattray argued that there was no need to summon them if “death had emancipated [Kalikinkar].” Mr. Ross, on 5 May 1832, finalized the judgment of Mr. Rattray and added that the “service” of slaves and of dependent tenants had to be clearly distinguished. He felt that the appellants had been the latter, who held land “on condition of paying a fixed rent and rendering service.” As long as they held the lands and gained sustenance from them, they had to render service, but “not after abandoning the tenure and thereby discharging themselves.” Thus, “he who has the power to emancipate himself, cannot be considered a slave.”
Such conflicts were extremely common. While peasant rebellions and the question of peasant consciousness in early colonial Bengal have attracted scholarly attention for decades, the actions of slaves and the nature of their claims in legal cases have not been systematically studied.Footnote 33 Only a few monographs exist on slavery in early colonial Bengal, and none of them address the question of slave consciousness directly.Footnote 34 Yet, alleged “slaves” often expressed outrage against their masters. In the year 1813 alone, they filed 150 cases in colonial courts, which may have represented only the tip of the iceberg.Footnote 35 A. D. Campbell, who served as a Superintendent of Police and then as a Magistrate, noted that a slave’s “peculiar situation, want of relatives and friends,” and “general ignorance and poverty” must have prevented very many from seeking redress against their wealthy and powerful masters.Footnote 36 Despite such unfavorable odds, alleged slaves began claiming property rights to their homesteads in that very year. The boldness and novelty of such claims were noted by missionaries who had just been allowed into British territories by the East India Charter Act of 1813.Footnote 37
The judgment passed in the conflict discussed above became a precedent. In another instance in Dhaka in 1831, a slaveholder—Gauri Sankar Roy—filed a suit to recover runaway slaves.Footnote 38 Two of the defendants—Kirti Narayan and Sri Narayan Deo—appeared to testify and made identical claims about being tenants rather than slaves. The Narayan brothers insisted that their father, Binod Ram, was “free and earned his livelihood as a cultivator and tenant of land and by service.”Footnote 39 After the demise of their father, the brothers continued to hold lands as tenants on the estate of the plaintiff and paid rent. Notably, the plaintiff in his reply alleged that by the custom of the country, if Bhandaris or household slaves cultivated for rent lands of their master more than those assigned for their support, such fact did not discharge them of their liabilities as slaves. On 24 August 1832, the Principal Sadar Amin agreed with the plaintiff and ordered all thirteen defendants to be made over to Gauri Sankar as his slaves. Subsequently, Kirti Narayan went in person to the Sadar Dewani Adalat on 21 February 1833 and asked for a special appeal to be admitted. Crucially, the ground for his request was that the decision in this case was contrary to the judgment passed by the SDA on 5 May 1832 in the previous case, that is, Golak Narayan Ray vs. Kewal Ram Deo and others. The SDA’s hand was forced, and the appeal was admitted by Mr. Rattray on 23 March 1833. Gauri Sankar Roy—the slaveholder—was asked to justify his claim. In response, Roy asserted that the appellants had received not only lands from him for their sustenance but also rations and other aids. He also insisted that the “want of a bill of sale after the lapse of ages” could not affect his right as a master. Slavery of the appellants was proved by the “continuity of hereditary service,” as well as the evidence adduced regarding the support received by the servants.
Mr. Rattray, however, was not swayed. He proposed to reverse the decree of the lower court and passed his judgment on 7 August 1833, pointing out that the lack of a deed of sale and the lack of proof that the appellants held lands in return for hereditary service made Roy’s claim weak. More importantly, the judge made a new argument by insisting that even if such lands had been held in return for service, once the slaves quit such lands, “they cease[d] to be liable to any claim of servitude.” Acknowledging that the appellants were probably Bhakta-Dasa slaves, that is, those who offered themselves in servitude for the sake of food, Rattray nevertheless insisted that “when the slave for his food abandons the service, he becomes free.” These opinions were confirmed by Mr. G. Stockwell when he heard the case on 7 December 1835, and thus the alleged slaves were emancipated.
An almost identical case reached the SDA in 1835 from the frontier district of Mymensingh, where a monumental peasant rebellion (1826–1833) had just concluded.Footnote 40 The slaveholder claimed that the defendants had served him as slaves in return for access to land. The alleged slaves rejected such an ascription and claimed that they had occasionally served the appellant and used some land, but only “in place of wages.” Now, after some years of service, they had decided to move to another village because they were treated there as “tenants.” A familiar back-and-forth between several courts and judges followed. Eventually, the case arrived at the SDA, and the final decision was passed by Mr. Stockwell on 17 May 1836. He acknowledged that the defendants may have been slaves in consideration of lodging and lands for support. It was also clear, however, that they had “abandoned such lodging and support.” Citing his judgment of 7 December 1835 in the case of Gauri Sankar Roy vs. the Narayan brothers as a precedent, Stockwell concluded that the respondents were “slaves of the class of slaves for their food.” On surrender of the lands held, therefore, they were “entitled to emancipation.” Soon after, another identical case from Sylhet was admitted into the SDA, although the decision in that instance did not go in favor of the alleged slaves for technical reasons.Footnote 41
The trajectory of these cases shows that there was no consensus among judges and hence no coherent colonial policy regarding legal disputes over the status of “slaves.” Indeed, the Indian Law Commission’s report acknowledged that there was a great variety of opinion and practice on this matter and such variety was observable not only in different and distant parts of the country, but even in neighboring districts, and “sometimes [in] different divisions of the same district, and even at different times in the same court.”Footnote 42 Some officials admitted the legal standing of slavery and the rights and obligations arising out of it, whereas others rejected it entirely. And even among those who recognized slavery, “some would confine its existence within the bounds prescribed by the Mahomedan and Hindoo laws, others would extend it over the wider field of custom and usage; as Magistrates, some would render assistance to the master to recover his fugitive slaves, others would abstain from any interference; some would permit to the master the right of moderate correction, others would not countenance the exercise of such authority.”Footnote 43 The law, in short, was in a “defective” state, and it lacked “clear rules” that could guide judgments.
Moreover, the defective state of the law and the ensuing confusion were partly a function of contradictions internal to “Hindu” law. Based on classical texts including those of Manu, Vrihaspati, and Narada, Brahman (the uppermost, priestly caste) pandits (scholars) had provided the colonial state with a list of fifteen classes of legal slaves in Hindu law.Footnote 44 That list, however, raised more questions than it answered. For example, how would one distinguish in practice between Bhakta-Dasa slaves, or those who offered themselves in servitude for the sake of food, and Atma-Bikrita slaves, that is, those who sold themselves for a price? People who entered servitude out of economic distress often received more than just food from their masters, and the question remained whether those additional benefits should be considered a “price” or not. The distinction mattered because, according to the pandits themselves, a Bhakta-Dasa slave could be emancipated upon relinquishing his subsistence, but a self-sold slave could never become free without the will of the master.Footnote 45 In the cases narrated above, the masters did not conceive of their slaves as Bhakta-Dasas, and neither did some colonial officials, but fortunately for the alleged slaves, the highest-ranking judges did.
Such problems of definition could easily become more complicated. How, for instance, would one distinguish between either Bhakta-Dasa or Atma-Bikrita slaves and Rina-Dasa slaves, or distressed debtors who voluntarily engaged themselves to their creditors for a stipulated period? All three categories or a combination of them could easily be applicable in a particular case, but the rules of emancipation (according to the Brahmans) were different for the different categories. A remark on this point by the Officiating Magistrate of Rangpur is insightful: “In ninety cases out of the hundred, the servile status of slaves originated either by their forefathers having made themselves bondsmen by borrowing small sums of money (which bondage ought to have expired with their lives), or the descendants of cultivators who have died in debt, either to the zamindars, or persons holding small farms under them: as experience has proved to me, neither would scruple to compel the widows or children of their deceased bondsmen, or insolvent deceased cultivators, to give them written engagements of being slaves or bondsmen for life; and this instrument would cause the parties and their descendants to become slaves in perpetuity.”Footnote 46
It was thus difficult, if not impossible, to know for certain exactly how servile statuses originated, and whether they involved coercion or not. In many parts of Bengal, these kinds of cases were in the clear majority, which shows that the alleged slaves in the conflicts discussed above may well have been tenants at some point.
The trickiest legal question, however, was that of whether servile status (whatever its source) should be inherited by descendants. On this issue, at least, the Brahman pandits provided the colonial state with a clear guideline: a Griha-Jata, or one born in the house of a master from a female slave, was also a slave, and could not be emancipated without the will of the master. The definition was extremely wide because the child did not need to be of the master to be considered a slave, nor did the type of slavery ascribed to the child’s father matter. Moreover, women cohabiting with any type of slave necessarily acquired servile status. Without doubt, this was a strategy to reproduce servitude, but it is clear that by the 1820s and 1830s, descendants of erstwhile slaves did not want to put up with it. In all of the cases discussed above, the masters were trying to ascribe servile status to the descendants of those who may have been slaves to their families, but the new generations resisted. When colonial judges decided to emancipate the alleged slaves in these cases by treating them as Bhakta-Dasas rather than Griha-Jata, they were exceeding the space of “Hindu” law, even if they did not always acknowledge it.
In other cases, too, this problem reared its head, and the decisions passed by the colonial state were significant. In 1831, the courts had to resolve a conflict from a district that falls in modern-day Jharkhand and borders western Bengal. The plaintiff alleged that his forebears had purchased some slaves during the great famine of 1769–70 and that he was filing the suit to recover as his slaves those who were the descendants of the original Atma-Bikrita slaves two generations removed.Footnote 47 In response, the defendants claimed that their ancestors had never sold themselves into slavery but rather had received “loans” of coarse grains and other foodstuffs that had since been repaid. The plaintiff shot back, saying that in a famine “no one supported another, particularly one of low caste, for the mere acknowledgment of grain or money.”Footnote 48 Eventually, the case was decided against the plaintiff because the bill of sale provided was written on plain paper, which the courts found suspicious.
Another conflict from the frontier district of Sylhet in 1826 had an identical structure: the slaveholder alleged desertion, but the alleged slaves responded that they were tenants.Footnote 49 In this case, a genuine bill of sale from 1757 was produced. The question was whether the presence of a clause stating that the person was selling himself for “sixty and seventy years of service” meant that after that term, his remaining descendants should be considered free. One judge claimed that the limitation of time was merely in conformance with “custom” and was meant to ensure the “exemption from labor in old age, not freedom.” Others countered that when a person sold himself for a stipulated period, it could scarcely imply that his wife, son, and further descendants “should pass as owned persons, and be liable to render service to the issue of the hirers.” The alleged slaves were eventually emancipated, but this was done despite the suggestions of the Brahman pandits, expressly regarding this particular case, that the alleged slaves were Griha-Jata and could not be emancipated without the will of their master. Footnote 50
We see, then, that at least some colonial officials were upholding the undeniably modern idea that individuals cannot be held responsible for the sins and debts of their forebears. When judges decided to emancipate slaves, they usually did so either by playing on the nebulousness and overlapping of definitions or by going beyond the ambit of Hindu law as prescribed by the Brahmans. It is also clear, however, that this was not colonial benevolence but rather a response to pressures from below by alleged slaves. The evidence suggests that descendants of erstwhile slaves primarily sought property rights over their bodies and labor. They wanted to be able to choose where, how, and under whom they would work, and they preferred contractual relationships of “loans” and “wages” over the paternalism of a master-slave relationship. This was a rejection of caste norms as well because we know that at least in the far-eastern districts of Mymensingh and Sylhet, nankar lands were given for subsistence to the lowest castes.Footnote 51 The disputes show that the alleged slaves wanted to be able to move freely whenever they wanted to, especially in eastern Bengal, where land was still plentiful, and where the frontier was closed only in the late nineteenth century.Footnote 52 Slaves, in short, preferred the legal status of tenants or “raiyats,” even if that meant being paikasht raiyats for a time, that is, raiyats who migrate often and are not permanently settled in a particular village.Footnote 53
Making Sense of the Evidence: Abolition and Counter-Revolution in Colonial Bengal, 1772–1843
To interpret this evidence, we must consider the trajectory and contradictions of colonial governance, as well as the longue durée history of the society being governed. Certainly, ambivalence about questions of slavery and caste was a constant feature of colonial policy from the 1770s onward. There was an abolitionist impulse, which found expression in periodic regulations aimed at curtailing the slave trade in the Indian Ocean. In 1774, commerce in slaves within the territories of the British East India Company (EIC) was banned. In 1789, the export of slaves was prohibited and imports were eventually outlawed in 1811.Footnote 54 Such measures were only partially successful because the slave trade was extremely lucrative, and the company failed to prevent its servants from participating in the trade and from owning slaves. The deeper problem, however, was the company’s legislative measures.
In 1772, Governor-General Warren Hastings sanctioned slavery in British India as a penal measure.Footnote 55 The ostensible aim was to prevent “hereditary” robbers by enslaving entire families and putting them to work. As the Court of Directors put it, the robbers causing havoc across the company territories were “wretches who have placed themselves in a state of declared war with Government, and are, therefore, wholly excluded from every benefit of its laws.”Footnote 56 Crucially, when justifying the use of slavery to combat robbery, the state introduced the distinction between the “evil” character of Atlantic slavery and the “mild” character of its Indian counterpart. In India, apparently, slaves were “treated as the children of families to which they belong; and often acquire a much happier state by their slavery than they could have hoped for by the enjoyment of liberty.”Footnote 57 Another mode of justification was the grounding of slavery in the “religious” laws of “Hindus” and “Muslims,” which necessarily implicated “caste” in the colonial discourse on (Hindu) slavery.
The state first connected caste to religion in 1772, when it decided that in all matters pertaining to inheritance, marriage, and other religious usages and institutions, the laws of the Koran concerning Muslims and those of the “Shaster” (sic) with respect to the Hindus would be followed. The exposition of the law was left to the Maulvis for Muslims, and the Brahmans for Hindus, thereby entrenching an upper-caste point of view as law. In 1793, Hindu and Muslim laws of marriage, caste, succession, and inheritance were upheld again by the Marquis Cornwallis. But the question of whether Hindu and Muslim laws of slavery should also be upheld, and whether slavery should be deemed a domain of “religious” practice of the two communities, was still open.Footnote 58 The matter was settled arbitrarily in 1798 when the Sadar Dewani Adalat decreed that the spirit, if not the letter, of the 1772 and 1793 decisions suggested that Islamic and Hindu laws should also apply to cases of slavery, and this was confirmed by the Governor-General on 12 April 1798. Until legal non-recognition of slavery by Act V of 1843, this remained the shaky foundation upon which slavery in British India rested.Footnote 59
In truth, the 1798 decision followed logically from the ones made in 1772 and 1793 because caste was correlated to slavery, and if the former could be deemed a religious usage, then why not the latter? At the time, caste and slavery functioned as overlapping yet distinct modalities of labor control in Bengal: the number of “impure” castes set the upper limit to the total number of agrarian slaves, because “pure” castes could be domestic slaves to castes of the same or higher status, but they could not be agrarian slaves.Footnote 60 Witnesses belonging to multiple castes arrived at the same conclusion in their testimonies to the colonial state, and Brahman witnesses went to great lengths to emphasize that only some castes were “fit” for some kinds of work. Vishnu Dutt Dalli, for instance, testified on 5 February 1839 that slaves of the “low-ranked” Chandala, Dom, Hera, and Kumar castes could only be employed in outdoor work.Footnote 61
Within the framework of early colonial law, therefore, one could insist that slavery was not a domain of religious practice only in two ways: either accept that caste was also not just a religious institution but one that had a fundamentally secular basis, or artificially separate the question of slavery from the larger set of social relations within which it was embedded. The former was never done, and the latter course was followed in the run-up to 1843.
Legal conflicts between alleged slaves and their masters became a pressing issue by the 1810s. John Richardson, the District Judge and Magistrate of Bundelkhand, and William Leycester, a judge of the Committee of Circuit, argued for complete abolition in 1808 and 1815, respectively.Footnote 62 Chief Justice John Harrington sharpened the case for intervention with a poignant question in 1818. How, he asked, could slaves hope to attain the liberty they desired when they were “possessed of and [could] acquire no property to enable them to institute or defend a suit” and when they could be subjected to restraint or even confinement while suits were pending?Footnote 63 The case for abolition, however, was summarily rejected by a state deeply invested in extracting revenue and maintaining “order.” Acting Governor-General John Adam (January–August 1823) and his successor William Amherst (1823–1828) both concurred that slavery was interwoven with the “customs and feelings” of the people. Intervention, therefore, would only wreak discontent and it was inexpedient to hazard the inconvenience.Footnote 64
In the 1833 East India Charter Act, Parliament inserted a clause stating that slavery would be abolished in the Indian territories by 12 April 1837.Footnote 65 The EIC’s Court of Directors responded that they would only approve of legislation against slavery if it did not do “violence to the feelings of Caste, or the rights of property.”Footnote 66 The original clause was even modified to state that no regulation would be passed without “due regard to the laws of marriage, and the rights and authorities of fathers.”Footnote 67 This is not surprising because the terms of discussion had already been set when caste was relegated to the religious domain in 1772. This meant, paradoxically and perversely, that any acknowledgment of the caste-slavery relationship could only serve to strengthen the argument against intervention and abolition.
The Law Commission sanctioned by the 1833 Charter was set up in 1835, but the Report on Slavery—which is the chief evidentiary basis of this essay—was published after significant delay in 1841. Even then, the defenders of slavery in both India and Britain kept insisting that abolition would make landowners and noblemen indignant; that laborers would become idle and unproductive without “correction” and coercion; and that slavery in India was “nothing more than an affair of caste.”Footnote 68 Faced with such vicious opposition, the opponents of slavery chose the easy way out. Instead of arguing that relegating caste to the domain of religion had been a mistake from the beginning, abolitionists conceded the artificial distinction between caste and slavery. They ended up being narrowly legalistic, and they wrongly assumed that caste distinctions would cease to matter in a regime of “free labor.”Footnote 69 Act V, passed on 7 April 1843, argued only for the legal non-recognition of slavery and did not actively declare anyone free.Footnote 70 Even this half-hearted measure was applied selectively, and the colonial state continued to return alleged slaves to their masters. The principal aim, always, was the extraction of revenue and the control of labor.
These trends show that the tendency to consign caste to the religious sphere and to collude with landed elites was not new in the late nineteenth century. The policy of “religious neutrality” after 1858 meant that caste was relegated entirely to a “national social” domain beyond politics,Footnote 71 but the impulse itself had deeper roots. The early colonial state created precedents under “Hindu” law from scratch, whereas slavery in Mughal Bengal—especially in the eastern, frontier districts—was grounded in economic distress and was not reducible to textual dicta.Footnote 72 The debates between conservationists and abolitionists also show that the Orientalist trope of caste as a static, millennia-old “religious” institution expressing a uniquely Indian penchant for hierarchy was strong in early colonial India.Footnote 73 This view influenced Act V and it survived well into the twentieth century with negative consequences for Dalits.
This does not imply, however, that caste was merely a “colonial construction.”Footnote 74 Such a formulation occludes from vision longer-running South Asian social histories and the ethical-political valence of abolition in a context shaped by caste and agrarian slavery. It erases the ideological content of abolition tout court and cannot explain the subjectivity of the alleged slaves. Methodologically, the colonial construction thesis presupposes that any acknowledgment of the longue durée history of caste necessarily implies an acceptance of the Orientalist trope. If, however, caste is understood in terms of differential access to property and power—instantiated in practice as differential access to land, water, and markets, and the regulation of sexuality—then a history of caste can be written that refuses the binary of either millennia-old religious institution or merely a product of colonial governance.
There is a long and extensive tradition of historiography that approaches caste in such terms. It shows that in the early historical era (ca. 1000 BCE–ca. 300 CE) and the early medieval period (ca. 600–1200 CE), caste formation was congruent with agrarian expansion, the need to streamline inheritance over the control of land, and the extraction of revenue from land and markets. In other words, the growth of stratification was coeval with the emergence of patriarchy and the formation of states with fiscal bases in agriculture.Footnote 75 Caste was thus a way to directly control resources and indirectly control labor. Kinship relations continued to mediate social relations of labor, which meant that chattel slavery was never of major importance in India, and belonging to a “lower” caste did not automatically mean slave status.Footnote 76 At the same time, some groups existed only on the fringes of the absorptive process of sedentarization. They were simultaneously excluded (both spatially and socially) and yet included within the system as super-exploited laborers. These “outcaste” or “untouchable” communities varied widely across South Asia in terms of their ethnic and cultural identities, and the concept of untouchability ultimately depends on a vague, Brahmanical notion of ritual pollution. Nevertheless, all Indian vernaculars recognize the existence of a segregated sub-population forced to perform degrading labor deemed “impure” by society at large.Footnote 77 Pan-Indian terms such as Chandala do identical conceptual work.
Early medieval Bengal was no exception to these trends. The two major ruling dynasties—the Palas (ca. 750–1162 CE) and the Senas (ca. 1070–1230 CE)—were explicitly committed to the maintenance of a normative caste order.Footnote 78 They made land grants on a large scale to upper castes, and the collection of revenue from these lands, cultivated by labor that was ranked as “low caste,” was the fiscal basis of the state. From the late ninth century onward, grants began to mention the entire caste spectrum with varying degrees of access to land: Brahmans were at the top and the “untouchable” Chandalas (the “lowest of the mortals”) were at the bottom.Footnote 79 Sources from the early centuries of the second millennium CE show that the lowest castes involved in agrarian and other forms of menial labor—such as the Chamars, the Chandalas, and the Doms—were kept spatially segregated, and they lived outside the boundaries of political capitals and upper-caste rural settlements.Footnote 80 We have no way of estimating variations in the size of these communities over time, but that is a function of silences in the archive. Recent scholarship on slavery in medieval South India and Mughal South Asia has emphasized the challenges posed by such processes of “invisibilization.”Footnote 81
Mughal India was a vastly different context and yet caste continued to be reconstituted in new ways. The state was no longer explicitly committed to a normative caste order and the agrarian revenue system was sprawling and complex. At the village and sub-regional levels, however, a caste-based hierarchical structure remained the basic framework in much of the subcontinent.Footnote 82 Peasants had customary rights of usage, but they were tied to the land, which they could neither freely alienate nor desert at will.Footnote 83 In Bengal, the vast majority of field labor continued to be performed by middling and lower castes, including those deemed untouchable, whereas the largest estates were controlled by the ritually highest-ranked Brahmans and the powerful Kayasthas.Footnote 84 Indeed, as late as 1830, Brahmans continued to hold lands at 50% lower rent than their lower-caste counterparts in some regions.Footnote 85
Slavery in Mughal Bengal was reducible neither to caste nor to timeless “Hindu” and “Muslim” laws. Rather, it was grounded in economic distress caused by famines, floods, and debt.Footnote 86 Faced with a choice between servitude and death, many chose the former. Deeds of sale between 1694 and 1770 show that once a sale was effected, the master had complete control over a slave and their heirs. Punishment of runaway slaves was common. Masters aimed at the extraction of maximum possible work and reserved the right to cut loose unproductive slaves in old age.Footnote 87 Besides voluntary sale, the internal supply of slaves was also sustained through violence—that is, the kidnapping of men, women, and children—and natural growth, since women who married or cohabited with slaves also acquired servile status and their children inherited it.Footnote 88
At the same time, agrarian social relations were deeply embedded in caste norms, which shaped norms of slavery in turn. Not all castes could be asked to do all kinds of labor, and upper castes could not, in general, be agrarian slaves to those of lower status.Footnote 89 Prices varied by caste and plummeted during famines, and slaves belonging to menial castes such as the Bagdis had to explicitly disavow the rights of their children to any property that their parents had acquired.Footnote 90 Many deeds of sale were written in the vernacular, which shows that the norms and expectations of caste and slavery were fully legible to all actors within Indian society.Footnote 91
Two apparently contradictory things could thus be simultaneously true. Economic, political, and cultural changes between the sixteenth and eighteenth centuries, which some scholars have associated with an “early modernity” in South Asia (or Eurasia more broadly), massively increased the stakes of being able to access land and markets more freely.Footnote 92 Commercialization and expanded division of labor meant that caste was in a state of flux and neither static nor merely religious.Footnote 93 Concurrently, elites attempted to tighten control over labor through slavery and a renewed emphasis on caste hierarchies. In Rajasthan and Maharashtra in western India, for example, the seventeenth and eighteenth centuries witnessed not only an increase in the incidence of slavery but also a renewed emphasis on “purity” of caste and “descent.”Footnote 94 The context in both provinces was one of regional assertions against Mughal rule and agrarian expansion. The slaves were not only war captives, but also those affected badly by famine and ongoing debt. They performed all kinds of labor, including military service and agrarian labor. Upward mobility among merchant castes also led to a desire for greater “distance” from those deemed untouchable, a project that was brought to fruition in alliance with the pre-colonial state of Marwar.Footnote 95
The early colonial state, by instituting a “despotism of law,” acted as a bulwark for this second trend against the first.Footnote 96 The conflicts over the legal status of alleged “slaves” must be read against this background. In the eastern, frontier districts of Bengal where conflicts recurred most frequently, caste formation was still inchoate. The Muslim peasantry that emerged in eastern Bengal from ca. 1574 onward was not composed of those who had earlier been “Hindu” but rather of autochthonous communities at the fringes of the agrarian order, including the Chandalas.Footnote 97 Upper-caste landed elites and moneylenders often forwarded the capital necessary for reclamation during the seventeenth and eighteenth centuries,Footnote 98 but the form that agrarian social relations could take was by no means a foregone conclusion. Slavery in this context was a tool for consolidating control.Footnote 99 Indeed, by the eighteenth century, large estates regularly included a considerable number of slaves.Footnote 100 In the eastern districts of Dhaka-Jalalpur, Rajshahi, and Sylhet, slaves were also sold with the land.Footnote 101 Slaveholding, therefore, was especially important in frontier regions with a favorable land-labor ratio to guarantee labor supply. This is why, despite the extremely limited character of “abolition,” 580 Hindu and 700 Muslim landholders in Sylhet opposed Act V. In a petition submitted in February 1843, they claimed that hereditary slaves did not think of their situation as “degraded” and that men of “rank” had “no means of employing paid servants to perform the duties which were ordinarily performed by slaves.”Footnote 102
Thus, the alleged “slaves” in legal conflicts challenged not only slavery but also the practices that produced and reproduced caste: the control of access to land and labor. The challenge was not articulated in language or texts. Nor does the evidence imply a coherent ideology of abolitionism. Nevertheless, action and thought cannot be artificially separated. By challenging established practices of slavery against all odds, the enslaved thereby effected a critique of caste in practice. Arguably, such a practical challenge was more worrisome for those in power than a textual exegesis that did not necessarily challenge the practices that constituted caste. Why else, we might ask, were landholders in frontier districts so concerned about the toothless “abolition” of the colonial state?
Reading evidence from the colonial archive in terms of a larger structure of inference is not uncommon in a field of study where direct access to the mental life of historical subjects is virtually impossible. In his influential work, Thomas Holt also inferred what the emancipated black workers thought by looking at what they did. Although Holt had no evidence of actual statements made by the emancipated, he asked what kind of commitments could help us make sense of the actions they undertook. Fraught as this interpretive exercise is, it enabled Holt to situate the Jamaican case within broader debates about abolition, racism, and capitalism. This essay attempts something similar on a smaller scale for India, and unlike Holt, I do have evidence where the alleged “slaves” speak, albeit through the mediation of the colonial state.
Does the fact of this mediation make the archive unusable? Not necessarily. Scholars of the Atlantic world have been engaged in a lively debate for decades about how to access the subjectivity of the enslaved.Footnote 103 There is a broad consensus that legal conflicts recorded in colonial archives make social fault lines legible, and at certain moments in such conflicts even the enslaved could “speak.”Footnote 104 Despite the difference in context, the stakes involved are strikingly similar for South Asia: writing about and interpreting the actions of entire communities that have been deliberately “invisibilized” in the archive. Like their Atlantic counterparts, scholars of Indian Ocean slavery have recently noted that careful attention to legal conflicts and the indexing of court cases might enable an archivally grounded “global subaltern (social) history.”Footnote 105 This essay seeks to contribute to and deepen such conversations.
At the same time, the colonial archive was politically constituted and historically specific. Although the Charter Act of 1813 has received far greater attention from historians due to its major implications for the future of Indian industry, the Act of 1833 was no less consequential politically.Footnote 106 It transferred the entire responsibility of governance and policymaking to a company-state invested in revenue extraction, and the India Question dropped out of the British Parliament completely. It also espoused an authoritarian vision of “reform” led by Thomas Macaulay, who chaired the Law Commission that produced the Report on Slavery. Thus, when Lord Brougham raised the issue of Indian slavery in Parliament on 5 October 1841, Lord Ellenborough responded that “Parliament had wisely directed that the initiation in legislation on this subject should be left with the Governor-General and council.”Footnote 107 Indeed, Ellenborough, who oversaw the passing of Act V as Governor-General of India between February 1842 and June 1844, had denounced the clause in the 1833 Act regarding abolition as “useless and unnecessary.” It would, he thought, “excite the indignation of the landed proprietors” and “shake the confidence and the allegiance of the native officers.”Footnote 108
The early colonial state, therefore, was a force of reaction, but such forces were ascendant at the time not only in India but throughout the wider imperial world and within Europe and Britain. Most officials who governed India—even “Whigs” such as Macaulay who considered it necessary to appeal to the interests of the rising middle classes and “the democracy”—were political proponents of “order” and reform from above.Footnote 109 They opposed democratic, working-class movements everywhere including Chartism in Britain. Thus, Lord Ellenborough called the movement for electoral reform “a monstrous democracy,” and Robert Peel, who was Prime Minister (1841–1846) in 1843, called it “a reckless democracy.”Footnote 110 The reform that did occur secured the influence of the landed elite in Parliament and separated a section of the middle classes from proletarian workers by including the former within the political system.Footnote 111 Abolition, in this context, functioned like “moral capital” for a British Empire seeking atonement for the “sin” of slavery, which had led, many believed, to the loss of the American colonies.Footnote 112 Antislavery appealed to political elites because it enabled Parliament to keep flying the flag of liberty at a time when it was clamping down on liberties both at home and abroad.Footnote 113
By thus historicizing the colonial archive, we can grasp the contradictions of abolition in South Asia in a dialectical manner. In an agrarian society shaped by caste and slavery, formal freedom and personal liberty mattered. Thus, the claim by Dalit studies scholars about the utility of “colonial” liberalism and modernity for Dalits is historically valid. In Bengal, Lockean claims about the constitution of property by labor shaped agrarian politics from the 1860s onward, well into the twentieth century.Footnote 114 Such claims were already present in embryonic form in the Paglai Dhum peasant rebellion (1825–1833) in the eastern frontier district of Mymensingh. Here, hill tribes such as the Garos, the Hajongs, and the Hadis, who were being incorporated into a settled agrarian order as laboring castes, resisted increased revenue assessment. They claimed that since it was through their labor (or of their forefathers) that forests had been reclaimed and the land made to bear fruit, the incremental profit and higher rent that resulted from the increased settlement in the region should accrue to them and not to the landlords or the state.Footnote 115 Indeed, the connection between land, labor, and property was first made by the proponents of Faraizism, an Islamic reformist movement that emerged to prominence among the weavers and cultivators of eastern Bengal in the 1820s.Footnote 116 Their slogan—langal jar, jami tar, that is, the land belongs to the owner of the plough that tills it—inspired communist-led agrarian movements in Bengal throughout the twentieth century.
By claiming that all labor irrespective of caste status should be seen as constituting property equally, peasants articulated a profound critique of caste as a regime where different categories of persons were understood to be engaged in qualitatively distinct forms of labor, and those who labored the most were perpetually propertyless. Such claims, though, could only be made by tenants and not by slaves treated as property. Given that legal conflicts recurred most frequently in the eastern districts with surplus land, there can be little doubt that this was one of the reasons why slaves wanted to assert a property right over their own bodies and labor. The relevance of abolition in South Asia, therefore, cannot be reduced to the contingent requirements of colonial governance and empire.Footnote 117
Nevertheless, Act V, and liberalism in South Asia more broadly, functioned as tools of empire in a historical moment of counter-revolution that laid bare the emergent contradictions of capitalism. Massive proletarianization had brought the question of mass democracy and socialist transformation to the fore in Europe, which meant, as Marx insisted, that classical bourgeois revolutions were no longer feasible. There could only be counter-revolutions defending already constituted property, or social-republican revolutions against the capitalist order.Footnote 118
This point brings to light the limits of abolitionists in South Asia who conceded the artificial distinction between caste and slavery. By the late eighteenth century, the practices that ensured differential access to property and power had created a de facto separation between propertyless and propertied castes, which provided the grounds for the emergence of specifically capitalist labor relations. This can be discerned from the fact that in the early nineteenth century, a floating, rural proletariat existed in most parts of India, and, when industrialization began, the necessary labor was drawn primarily from this floating proletariat consisting of the lowest castes.Footnote 119 Therefore, scholars who argue that the lack of classic primitive accumulation in India is one reason for the distinctive character of capitalism in the region are only partly correct, because they ignore the long-running history of caste and the incorporation and reconstitution of caste within capitalism.Footnote 120 Lockean claims to property were ill-equipped to respond to these complexities, which explains why such arguments could function as a defense of the status quo—that is, a caste-differentiated distribution of property—by the 1910s.Footnote 121
Though ultimately unsuccessful, the slogan of “forty acres and a mule” to the formerly enslaved was at least raised in the era of Reconstruction in the southern United States.Footnote 122 Something equally radical would have been necessary in India to resemble anything like a level playing field, but the counter-revolution of property, backed by an extractive colonial state, was even stronger. These contextual and political factors explain the failures and limits of liberty and abolition in South Asia, not the inability of the colonial state to understand the “slavery-to-kinship continuum,” or a purported attempt to smuggle in “European concepts of liberty and ‘possessive individualism.’”Footnote 123
Conclusion
The contemporary literature on Indian Ocean slavery speaks to many themes broached in this essay. Scholars have shown that wages and loans, and the formal freedom they embodied, were not unimportant. Wages were used to purchase manumission in the Cape colonies, and freedom in the Indian Ocean often meant working for a wage.Footnote 124 “Flight” represented “fight,” and the lack of large-scale slave rebellions compared to the Atlantic does not imply a lack of resistance.Footnote 125 Rebellion against extractive empires and deeply rooted indigenous power structures was not easy, and any resistance we find in the archive requires a historical explanation.
New research on global histories of slavery, however, has not inspired attempts to write global histories of antislavery. The Atlantic-centered debate is rarely invoked today, and the dialectical approach of major contributors to that debate has been jettisoned in favor of a purely negative view of capitalism as necessarily imperial and racist. The obverse of this tendency is the conceptualization of “resistance” as challenges only to extractive European empires, which leaves open the possibility of an uncritical affirmation of pre-colonial South Asian societies.
By contrast, this essay has argued that combining insights from the antislavery debate with the extensive literature on caste in India offers two benefits. One, it brings into sharp relief contradictions within Indian society, and thereby shows how and why abolition and the formal freedom of wage-labor mattered in that context. Those alleged to be “slaves” expressed dissatisfaction with the legal or de jure status of complete ownership, whereas their indigenous masters argued that such rights existed de facto in Bengali society. The “slaves” insisted that they had sold their capacity to labor in return for wages and loans, and not their personhood. Such claims were motivated by a desire for personal liberty: a desire to become something other than a “slave,” based on a recognition of the stakes involved in it.
The colonial state and the global, capitalist context of counter-revolution smothered these aspirations. By thus historicizing both the potentials and failures of liberty, we can think about comparison in a grounded manner, one that does not proceed from a priori assumptions either about the equivalence of “caste” and “race,” or about a necessary connection between capitalism, slavery, and racism.Footnote 126 Close attention to the specificities of caste, slavery, and social aspirations within South Asia, and archivally grounded arguments about how and when they came to be mediated by the history of capitalism, can obviate the need to fit regional histories into yet another Atlantic-centric notion such as “racial capitalism.” At stake is a deeper understanding of how global capitalism has unevenly incorporated pre-capitalist social forms within it and reconstituted them over time.
Acknowledgments
I would like to thank three anonymous referees for their deep and critical engagement with the essay. Their suggestions led to substantial improvements in the structure and argument of the article. Andrew Sartori read several iterations of the piece over many years and offered helpful suggestions each time. Long-winded conversations with James M. Vaughn and Sunit Singh at The University of Chicago have influenced my thinking in more ways than I can acknowledge. All remaining errors are mine.
Open access
