Sumit Guha

Speaking Historically: The Changing Voices of Historical Narration in Western India, 1400–1900

In recent decades, historians have increasingly turned to consider the institutional practices, epistemological assumptions, and silent exclusions that shape the discipline.¹ This reflexive movement has led to the identification of difference and disjunction in the formerly seamless lineage from classical Greece to twentieth-century Europe, and has encouraged a search for alternate and parallel forms of historical knowledge. It has prompted a reconsideration of neglected hermeneutic issues that a generation of historians had disregarded in the frenzy of scientism.² The entire process is itself rooted in larger changes in the world of learning, by which knowledge has become "retrospective and critical, more concerned with the reconstruction of the past, of the practices in which people made knowledges—which is to say, historical."³

Historical problematics themselves emerge historically. So do the meta-systems that distinguish "native traditions" from "scientific history"—a distinction suddenly imposed in much of the colonial world at the end of the nineteenth century. I will address both these points through a study of the changing deployment of the historical narrative in Marathi—a major language of western India whose literary traditions go back to the twelfth century C. E., early in what Sheldon Pollock has called the vernacular millennium.⁴ This article traces the socio-political structures that generated and marked this narrative genre from the sixteenth century onward. I will show how it emerged and how it changed: first in the eighteenth century with the rise to subcontinental prominence of a Marathi-speaking elite, and again with the establishment of British colonial rule a century later.⁵ By analyzing the changing response to these narratives through the past five centuries, we can gain a new perspective on the generative relations between structures of power, on the one hand, and historical traditions and practices, on the other.⁶ But before that, let me begin by situating modern evaluations of pre-colonial texts within the comparative global setting that began with the Enlightenment discovery and ranking of "civilizations."

Speaking very broadly, the historiography of India has hitherto been marked by two opposed positions: either that Indic civilization lacked any capacity for rational history, or that it had always possessed a distinguishably historical tradition. This issue really gained significance with the worldwide spread of Western power in the nineteenth century. Western historical practice took institutional shape in the century after the French Revolution, an epoch marked by global imperialisms that were driven by metropolitan, and opposed by emergent, nationalisms. As François Furet describes it, history now became "the genealogical tree of European nations and of the civilization they bore."⁷ To be a nation, and not (as Winston Churchill famously described India) "merely a geographical expression," it was necessary to have a historical consciousness. As Prasenjit Duara has noted, "national history secures for the contested and contingent nation the false unity of the self-same, national subject evolving through time."⁸ Securing this was especially important in the non-Western world, and nationalists were eager to recover and represent their own history. Youssef Choueiri writes of North Africa that "Tunisian, Algerian and, later on, Moroccan historians began to articulate an image of the long-forgotten past in national terms. They used the structure of French scholarship and turned its prototypes upside down."⁹

But how should this recovery of "their own" history proceed? There were two obvious routes: one was to affirm or assume that equally valid historical knowledge had long existed within indigenous scholarly traditions, and draw on these to develop a usable national past. The other was to defeat Western imperialism by deploying its own scientific-historical apparatus to build a more satisfactory national history. In colonized India, state control excluded the first strategy from the education system that took shape in the mid-nineteenth century, but it was vigorously pursued in the larger public sphere of printed and oral polemic. An early and immensely influential example of this is Satyartha Prakasa by the founder of the Arya Samaj, a major modernist Hindu organization. In this book, Dayananda Sarasvati drew on the ancient Hindu traditions of scriptural exegesis as the source of true knowledge to establish his theories and refute Western ideas on (for example) the early history of India.¹⁰

Few colonial intellectuals had the self-confidence to be so dismissive of Western learning, and so a significant number sought to defeat Western historians with Western methods. A leading figure in this enterprise was V. K. Rajwade (1864–1926), a nationalist who forsook all gainful employment and devoted almost forty years of his life to locating, editing, and publishing primary sources for the history of the Marathas.¹¹ Rajwade came to the discipline of history in the 1880s—a time when (as Bonnie Smith has shown), historical study in the West had moved from the public and declamatory lecture hall into the closed professional seminar that produced factual narratives based on archival sources. It was this practice that was seen to transform history into "knowledge and secular truth."¹² Rajwade desired, above all, to displace colonial history by a more rigorous application of its own methods. He published his first volume of documents in 1898. Meanwhile, in a more directly confrontational vein, Rajwade's militant contemporary Savarkar sought to mark the fiftieth anniversary of the major anticolonial uprising still known to the Anglophone academy as "the Indian Mutiny" with a new history of The Indian War of Independence. Banned before its appearance, it was printed in France, and stray copies circulated illegally in India. While unabashedly propagandist in tone and rhetoric, it drew most of its information from nineteenth-century British sources read "against the grain," and supplemented by a few similarly authentic Marathi documents.¹³ The colonial official Valentine Chirol described the book as a "very remarkable history of the Mutiny, combining considerable research with the grossest perversion of facts and great literary power with the most savage hatred."¹⁴

After India became independent in 1947, nationalist history unself-consciously used the methods of Western historiography to develop a dominant national narrative engaged in a running polemic with "imperialist" history.¹⁵ A sharp critique of both these "elite" positions marked Ranajit Guha's launching of the Subaltern Studies project in the early 1980s. But the Subaltern school was still generally content to assume that true historical understanding came from the West. The tools used to recover subaltern struggles from the disdainful neglect or discursive appropriation of the elite were drawn first from Marxism, then structuralism, and (later) postmodernism.¹⁶ Ranajit Guha himself found no Indian historical writing worthy of note until the Bengali-language Rajabali issued from the colonial College of Fort William in Calcutta in 1808. Until then, it would appear, a "hallowed and hoary sense of the past had been reproducing itself in Puranic discourse for a thousand years or so until as recently as the eighteenth century."¹⁷

Ranajit Guha can hardly have been ignorant of the existence of histories firmly located in dated time and written in Persian, but it would seem that these were not "Indian" enough for him. Only Bengali and English writings appear in his "Indian" historiography. This, although Persian histories were read all over South, West, and Central Asia, while at the time of writing the first "Indian" history, the Rajabali, could have had no readers outside the narrowly provincial Bengali literati community and a handful of trainee English civil servants reluctantly completing the curriculum in "native languages." But radical historians have not been the sole adherents of this position. The conservative R. C. Majumdar—who saw India's unity as mainly grounded in a shared Hindu culture—had long accepted the absence of historical writing in ancient India.¹⁸ Another leading historian of modern India, Sumit Sarkar is more aware that the pre-colonial was not a tabula rasa conveniently available for the Orientalizing fantasies of imperialism, and that

[p]re-colonial India, with its very long traditions of written culture, produced numerous texts of recognizable historical intent or value: Puranic king-lists, dynastic chronicles, histories of castes and religious sects, biographies of holy men, genealogies of prominent families. As elsewhere, there were evident links between the quantum of such texts or documents and levels of organized, bureaucratic power ... historical accounts became much more numerous under the Delhi sultanate and the Mughal empire.¹⁹

The suggested link between bureaucracy and textual production is none that I shall explore.

Other scholars anticipated Sarkar's reluctance to make the Indian historical tradition coeval with colonial rule. But (with rare exceptions) they also tended to seek indigenous historical scholarship, not in Persian or the modern vernaculars, but in the ancient high language of Sanskrit (sometimes supplemented by the Buddhist sacred language, Pali). The Sanskritist A. K. Warder complained for, example, that not only had nobody written a comprehensive overview of Indian historical thought but that the object itself "is even widely supposed not to exist." He added that "[p]ractically all existing 'Indian history' written in English can in fact be adequately characterised as 'imperialist' historiography. "²⁰ In 1966, V. S. Pathak published a scintillating study of five "historical" poems from the Sanskrit tradition arguing (a decade before Hayden White) that the poets had winnowed factual elements from royal lives in order to emplot a narrative of royal triumph consummated by union with a queen who embodied the glory of the kingdom.²¹

Warder published his massively learned survey in 1972. It was then cited by Romila Thapar in a 1974 lecture in which she argued that the people of ancient India had a special kind of historical tradition. This was one in which "the past was not recorded as a succession of political events, for the legitimation of political authority was more important, and it was to this that the historical tradition gave precedence" via the transmission of genealogical lists. Since she also views many genealogies as fabrications, it is clear her definition of "historical" is far wider than Majumdar's, and this enables her to find "history" in Sanskrit texts where Majumdar could not discern it.²² But this is achieved at the cost of blurring, if not obliterating, the distinction between historical evidence and historical narrative, cult and chronicle. Meanwhile, largely ignored by the Anglophone academy, Indian scholars of the different regional languages worked on recovering and editing their respective literary and historical legacies in institutions such as the Bangiya Sahitya Parisat, the Bharata Itihasa Samshodhana Mandala, the Maharashtra Sahitya Parishad, the Rajasthan Oriental Institute, and many others.

The issue of the validity of non-Western narrations of the past gained a new prominence in the wake of Edward Said's celebrated attack on the imperial foundations of the Western academy in Orientalism (1978). The 1970s saw scholars launching important efforts to study historical narratives in living vernaculars. In 1979, James Siegel published a subtle study of several long poems from the Atjeh region of Indonesia, characterizing these as the "historical thought of a Sumatran people." But as he describes them, the "narrative of the epics cannot be equated with 'what happened' even though some of the events described actually did occur. Atjehnese themselves did not distinguish epics on historical topics from those based on mythical themes." It would appear, in fact, that these texts can be distinguished from "fantasies derived from Indian or Malay sources" only because the personages and transactions also appear in Western sources.²³ This point was sharply made by Shelly Errington in 1979, when he argued that viewing Malay hikayat as quasi-historical was intrinsically wrong, and amounted to an effort to force them into a Western genre where they did not belong.²⁴A different argument was made by Virginia Matheson in 1986. She proposed that several Malay narratives of the eighteenth century argued obliquely over the ranking of regional centers of power, and can be read "as mirror images or responses to each other."²⁵ This suggests that the target audiences of these texts possessed the background knowledge needed to interpret them: knowledge that modern scholars have to reconstruct from non-Malay sources. Matheson's article can then be seen as an exercise in "ethnohistory"; defined by Nicholas Dirks in 1987 as "the reconstruction of an indigenous discourse about the past."²⁶

Dirks himself attempted this for an eighteenth-century South Indian kingdom in a book where the oral testimony of contemporary informants had to be grounded in the evidence of colonial archives. Partly inspired by Dirks, Phillip Wagoner went further in his study of the Telugu Rayavacakamu, establishing it as a seventeenth-century text, rather than the sixteenth-century official dispatch that it purported to be. Viewing the text as an "organically coherent literary whole," Wagoner showed how its author's knowledge of the recent past was deployed as "part of an ideological argument for the political legitimacy of the Madurai Nayaka regime" (that came to prominence after the fall of the imperial city of Vijayanagara in 1565).²⁷

Meanwhile, a Hindu nationalist claim to the site of a sixteenth-century mosque in Ayodhya grew ever more strident, despite an overwhelming scholarly consensus on its lack of any real historical basis. A mob demolished the mosque on December 6, 1992, and that sparked Hindu-Muslim violence across India and Bangladesh.²⁸ Indian historians were confronted with a powerful challenge to their disciplinary claim to produce authentic knowledge of the past. The issue was on the minds of scholars gathered for a conference in 1997, and Daud Ali, the editor of their proceedings, pressingly raised the question: "can we write a history of conceptions of the past, or a history of regimes of historicity in South Asia?"²⁹ A bold answer was presented in an essay by Sanjay Subrahmanyam that prefigured a co-authored book that appeared in 2001.³⁰

At the core of this important monograph are skilled and subtle analyses of Western and Indian accounts of two celebrated eighteenth-century battles in South India. While clearly indebted to Siegel, Errington, Matheson, Dirks, and Wagoner, among others, the authors are bolder than their precursors, and seek to "recover as history a significant body of literature from late medieval and early modern South India." This is to be done by reading texts exclusively through South Indian "society's visions and understandings of the past" as extant through the fifteenth to eighteenth centuries. The producers of this history are identified as karanams—a select elite of officials knowing more than one language, reading different scripts and "with access to" Persian and Sanskrit. One may infer that this social stratum is assumed to be the real bearer of South Indian society's "visions and understandings." The authors' putatively total envelopment in a pre-colonial "South Indian" cultural perspective, of course, prevents them from following Siegel in using external referents to winnow "historical topics" from "mythical themes." Yet as anyone who dips into the literary corpus of pre-colonial India knows, every narrative set in the past tense can hardly be deemed historical in either content or intent. The authors are aware of this problem and resolve it by an ex cathedra decree: "Readers or listeners at home in a culture have a natural sensitivity to texture. They know when the past is being treated in a factual manner." Epistemologically speaking, this is hardly a convincing position. (Where does one get a domiciliary certificate for an eighteenth-century culture?) The essentialist approach of the authors is also exemplified in their reification of both "genre" and "community": "we propose that history is written in the dominant literary genre of a particular community ... If purana is the pre-eminent literary form, history will be written as purana; if kavya [epic poem] dominates, we will find history as kavya; if prose chronicles come to the fore, they too will serve history."³¹ A brilliant exercise in comparative philology, Textures of Time repays reading for the several subtle literary analyses that it contains, rather than for the shaky theoretical scaffolding that encumbers them.

The above-mentioned Hindu nationalist campaign to destroy the Ayodhya mosque was defended, in part, by recourse to an esoteric "unrecorded" history.³² Partha Chatterjee recently reflected approvingly on how social pressures have led some historians to redefine the discipline by not "excluding popular practices of memory from its list of approved practices" and "incorporating within itself an appropriate analytic of the popular."³³ Yet more recently, a mob attacked and damaged an important archive in Pune, India, because an American scholar who (they alleged) had written disrespectfully of the seventeenth-century monarch Shivaji had used that collection.³⁴ How are disputes between "indigenous discourses about the past" or "popular practices of memory" to be resolved in the present? Who decides who else is "at home" in a culture? Moving the epistemological goalposts is not a fruitful strategy for dealing with differences.

Perhaps the history of the Marathi bakhars suggests an answer. These narratives emerged in settings where the historian did not stand forth as a judge: rather, he (rarely, she) spoke as a plaintiff, a defendant, and sometimes a witness. Narratives were often produced in judicial disputes over heritable property. So they contested alternative stories and sought to ground themselves in witness testimony, documentary evidence, and the "common knowledge" of the local community. That community itself was usually religiously, culturally, and linguistically diverse. The assembly of worthies that acted for it often included rich and poor, high and low caste, Hindu and Muslim alike. The members of the judicial assembly scrutinized documents and interrogated witnesses. Sometimes, lack of evidence forced them to leave an issue unresolved. The award in one such case stated that the matter could be resolved only if a specific original document mentioned by several witnesses was found, but not otherwise.³⁵ I propose that the larger setting for these discourses was the interaction of expanding royal power and resilient local society in the early modern era. This setting led peasant proprietors and great nobles alike to develop coherent narratives of local pasts. I thus seek to ground a genre (the Marathi bakhar) in the social history of Western India, and to show how the effort to generate a "Maratha identity" itself impacted the genre. Unlike Textures of Time, my object is to historicize both genre and community, rather than attempt a literary analysis of the hundreds of Marathi texts that are the subject of this essay.

Let us return to Sumit Sarkar's useful suggestion that the production of such texts is connected to the exertion of bureaucratic power. Much of this essay will be devoted to an examination of it. But we must note that different classes of historical record appeared through time. The ancient king-lists, for example, began to be compiled at least two millennia ago, and were recopied, miscopied, and edited for centuries after.³⁶ Furthermore, alternative and divergent king-lists coexisted without any effort at comparison and authentication. It is also worth considering why the various genres were not uniformly produced in the various regional languages that took shape between the Khalji sultanate (1290–1325) and the Mughal Empire (ca. 1550–1750). "Bureaucracy" or at any rate administration by specially trained literati was, for example, strongly present in greater Rajasthan or we could not have the detailed local information on (for instance) tax assessments, crops, irrigation sources, and castes found in works such as Mumhta Nainsi's Vigat from the 1660s.³⁷ But the narrative portions of the Rajasthan texts are firmly centered in the genealogical mode, worked into a frame constituted by genealogical family and its appanages. I may in passing remark that Nainsi's compilation certainly recognizes that the different clan and lineage histories he recounts were occurring simultaneously, if unconnectedly, in diverse locales. The link between them was that all the protagonists were Rajputs (though this term occurs infrequently). Nainsi also signals differences between what he deems established facts and more doubtful reports. His account of the famous Sisodiya clan begins by saying (my term) "factually" that they were "originally" called Gaihalot, and then adds that "one report" said that their first dominion was in the South—in Nasik Trambak.³⁸ Nainsi's conscious adoption of the genealogical mode was, I suggest, related to the genealogical sources of legitimacy and clan-based structures of authority in the Rajasthan macro-region—a structure that ultimately circumscribed the power of literati officials like him.

And so to the real topic of this essay, the Marathi bakhar. This was a prolific genre, with some 200 or more works in this class known to be extant and seventy in print.³⁹ I shall begin by considering the meaning of the term and then propose that narratives of this type had their beginnings in two related sources: lawsuits tried by the local community and inquests by incoming royal authority. I shall then move on to show how such narratives mutated as they became enmeshed in the Maratha imperial project of the eighteenth century before changing once again when their dominion was subverted by the incoming British.⁴⁰ Let us begin with the term itself. Rajaram Hervadkar is the leading editor and analyst of the genre. He suggests that the term itself arose by metathesis from the Arabic khabr (news report). Another authority, Bapuji Sankpal, on the other hand, has argued that the two words were used differently, and that as both terms were in circulation simultaneously, one could not have arisen by metathesis from the other.⁴¹ The real significance of naming practice should be sought not in etymology but in usage. Whatever its origins, the term was certainly well established in official usage by the 1670s, when the Maratha sovereign Shivaji ordered the preparation of a thematic Sanskrit thesaurus that would reduce the excessive use of "Yavani" (Islamic) terms of statecraft. Bakhair is listed in a chapter dealing with specialized terms used by recordkeepers and accountants (lekhanvarga) of the Rajvyavaharkosa, and given the Sanskrit equivalent akhyayika.⁴²Akhyayika itself is defined in a classical Sanskrit thesaurus as uplabdhartha—obtained or received knowledge—in meaning therefore, exactly the same as the South Asian usage of khabar.⁴³ It is preceded by equivalents for statistical statements (terij and tala), and followed by a Sanskrit substitute for ferist (list). This placement in the section that offers equivalents for standard tax-office terms suggests that the bakhar itself was something that the divan—finance minister—often had to address. But the divani daftar (loosely, the fisc) was an institution found in most parts of the subcontinent from at least the fifteenth century onward; its existence would not by itself explain why so many narratives of the bakhar genre came to be composed in Maharashtra rather than elsewhere.

I have invoked Indo-Persian administrative terminology, and it is therefore necessary to address the possibility that the genre was essentially borrowed from the Persianate literati who staffed the uppermost levels of the administration under the sultans of southern India and (later) the Mughal emperors. It is almost certain that some writers of bakhars would have known of or read Persian histories, but this knowledge would not by itself generate Marathi historical narratives unless the regional setting was such as to prompt such writing. We should otherwise expect this genre to be found all over the subcontinent in various languages, which is not the case. So this article will seek the impulses and the settings that led literati to generate these texts within certain identifiable conventions and in a specific regional language.

Let me now turn to a setting where people have long had the incentive to produce credible narratives of contested pasts: this was (and is) the court of law. Officials of the fisc often presided over the judicial gatherings that decided important property claims.⁴⁴ Long disputes over patrimonies were the stuff of everyday life in Maharashtra, and, as the historical origins of the claim had to be explained, bakhars were also written as a consequence. We have an early case (1610) where the term bakhar is used in the proceedings to mean simply a statement of fact (in this case, the location of the boundary).⁴⁵ Two hundred years later, a century-long dispute over the hereditary astrologer's office of twelve villages was settled in favor of the Pavaskar family; one of the latter's friends then composed a congratulatory Marathi poem. The poem tells us that at one stage Sridhar Joshi took his plaint to the judge at the provincial governor's court. The local landlord sent his testimony in the form of a bakhar, which however, was scrutinized and rejected by the judge. This did not deter the false claimant from again raising his claim, this time by bribing an important person at the court of the king Sambhaji (ruled 1680–1689) and presenting another slanted bakhar.⁴⁶ A bakhar was clearly seen as something that was (or should be) factual. It was also literary, as Sankpal has pointed out.⁴⁷ But implicit in his analysis is the assumption that this literariness moves these narratives away from being "real" histories. Does this "literary" quality weaken my suggestion that the pre-history of the bakhar is to be found in the written pleas filed before judicial authorities? I would suggest that the resort to softening and omission of truth, to ornamented narrative and rhetorical device, is characteristic of the forensic process, and these qualities may be seen as an important source of many features of the bakhar genre itself. These traits would not, nonetheless, detract from the effort at verisimilitude made in the text, or turn it into a pure work of the imagination.

Protracted legal disputes that dragged from forum to forum and were reopened with each change of regime were common, and literati in Maharashtra had frequently to prepare testimonies in such cases.⁴⁸ The relation between the maintenance of family chronicles and the defense of patrimonial claims is elegantly illustrated in a book recently published by A. R. Kulkarni, doyen of Maratha historians. The volume contains both the chronicle or memorandum book (1627–28 to 1696–97) of the Jedhe family and a karina that explains the history of their patrimony. The chronicle itself is a list of dates together with brief notes of important events: battles, coronations, the births and deaths of important people, and so on. This information was then drawn on to write a karina—an explanatory history of how they acquired and held their lands.⁴⁹

But gentlefolk like the Jedhes and courtly literati were not the only authors of historical narratives. Take the following summary of evidence from a suit begun in the early eighteenth century and resolved in 1785. The plaintiff was a poor goldsmith whose grandfather had been a village lord (khot). I cite a summary of the plaintiff's written and oral evidence prepared by a clerk sent by the chief justice.

You [the plaintiff] wrote out a plea the substance of which, as supplemented by additional evidence, is that the khoti [lordship] of the aforementioned village is anciently ours, but we cannot tell which of our forefathers secured this patrimony. During the Bijapur reign Krishnaset son of Yesaset held the lordship. Krishnaset's brother was Sonaset; upon his death his wife became a sati. The immolation site in the village still has a masonry marker. After this Bhagset son of Krishnaset held the office, as did Dadset son of Bhagshet. After them it was held by Krishnaset and Lamset, sons of Dadset. Lamset died, leaving a son Sonset, my father. Krishnaset fled the village due to the turmoil caused by the invasion of Naroji Ghorpade, and went to Rangana, lived there five or six years and died. His son Dadset lived at Satara and worked as a goldsmith to support himself, and my father Sonshet came back to our village. Fifty or sixty years have gone by since then; owing to poverty he could not fulfill the functions of the landlord's office.⁵⁰

We may notice how the narrative is sought to be authenticated by a genealogy linking the claimant to the original holder, to locally well-known events such as the invasion of Naroji Ghorpade, and to a still-extant masonry memorial to a widow-immolation.

The hereditary office (the plaint continues) was then usurped for many years. When challenged, the new occupant also claimed legitimate possession, and the court scrutinized documents on both sides. The displaced claimant had to explain the adverse possession of the incumbent. The judge sought sworn testimony from various villagers, and each of them declared what he had himself seen as well as what he had learned from his father or other older persons. Documents were also challenged on occasion. So, for example, eleven villagers testified that they had heard that one Keso Tanadeva had forcibly taken the old landlord and important villagers to the temple and extorted deeds transferring ownership to him. However, three others stated that they only knew that the deeds had been written and attested, but they did not know the process involved.⁵¹ The obvious effect was to question the value of the documents produced as evidence.

The significance of refuting contrary evidence was well understood. So, for example, in preparation for a lawsuit, the Sitole family compiled a memorandum showing what old documents they possessed from the time of Malik Ambar (early 1600s) to the present. Interestingly, it also included lines of rebuttal for documents held by the other side: "When the Nizamshahi [dynasty] fell, the Idilshahs took over and the district came under Bijapur. At that time the two brothers did not go to greet the new ruler, whereupon he was angered and conferred the deshmukhi [hereditary office] on the Ghatge family [who might therefore have records too]. Then Vithoji Naik [Sitole] went to Bijapur, got the Ghatges removed and secured a royal order in his favor. We have the farman [decree, which would supersede other documents held by the Ghatges]on (special) green paper."⁵² Similarly, when the Jagdales and Yadavas feuded over an estate, the former took pains to compile a set of testimonials from many neighboring landholders. These stated that the estate had been awarded by the then king without an assembly of local worthies attesting to its true ownership. The Yadavas, on the other hand, sought to explain this away by reference to various incidents in the strenuous campaign against the Mughal Empire having prevented this being done.⁵³

Not just great nobles but even small-town literati maintained memorandum books listing noteworthy events. So a Brahman resident of Cheul listed sacrifices, ceremonies, the names of Brahmans who attended and officiated, the installation of statues, gifts, and fees given as well as major political events in chronological order.⁵⁴ As the rights to officiate and receive fees were often hotly contested between Brahman lineages, this information was of more than passing significance. Similarly, as many tax-free assignments were held by Brahmans, suspensions of and inquiries into them were important. But other information could also be swept into his memorandum book. Two extracts:

Sake 1714 Paridhavi samvatsara Bhadrapad dark half [of the lunar month], 14th day, [tax-free] land-grants in the territory of Janjira Kulaba were taken over by the government ... Sake 1723 Jyeshta dark half 12th day, a huge fish floated up near the shore of Naganva harbor. It was thirty or thirty-one cubits long and as thick as the full height of an adult man. All the Prabhus and other folk went to look at it.

Documents and narratives would ultimately be tested in an assembly of regional or village notables (majlis, gotsabha, samasta dehijan). Literally thousands of such gatherings are mentioned in the records of medieval western India. It is important, I think, to note that people of diverse communities and persuasions would be present. Contestants could not, therefore, appeal to esoteric or sectarian knowledge, but turned rather to the "common knowledge" or "common sense" of a diverse local community, narrated in a language that everyone could understand. I have already mentioned the local assembly that met in 1610 to resolve the boundary dispute between two villages of Pune district. It was headed by the hereditary Islamic judge Kazi Abdulla and his deputy, Kazi Ismail. Also present were Mir Isiph, deputy commandant of the fort of Kondana, Nur Miya Vali, custodian of an important Sufi shrine. But the gathering also included two Brahman officials, the heads of the merchants of three market towns, and a long list of village headmen and notables from the nearby villages.⁵⁵ Thus Islamic scholars and officials, Brahman bureaucrats, merchants, local gentry, and substantial peasants were all present. All these would have to be convinced, and would then authenticate the award with their seals, signatures, or marks. Most proceedings were held in the open or in public places. People of many classes would thus periodically gather to hear, debate, and renew local knowledge and common sense. Some of them might have to stand forth as witnesses, plaintiffs, or respondents. In a largely nonliterate society, such meetings would be important in shaping and transmitting local and regional history. It is therefore not surprising that the English surgeon Thomas Coats observed of Maharashtrian peasants in the early nineteenth century: "they are fond of conversation, discuss the merits of agriculture, the characters of their neighbours and every thing that relates to the concerns of the community, and many of them are not without a tolerable knowledge of the leading events of the history of their country."⁵⁶

Like royal orders, grants, and other documents, the decrees resulting from such assemblies were firmly located in dated time, often with the Salivahana Saka alongside the solar Islamic (suhur) year; sometimes, a regnal year would be added as well. That all these dating systems referred to the same neutral commonplace linear time went without saying. Furthermore, implicit in such cross-dating is the idea that time has a single flow, and that different markers are arbitrarily placed. There was no single eschatological event that changed time, nor did it move to some eventual end of days. In that sense, this time was more secular than that of Jean-Baptiste Bossuet's universal history (for example), where human destiny was irretrievably altered by the incarnation of Jesus.⁵⁷

But the law court was not the only time and place where local and family history was rehearsed, contested, and remade. Many bakhars frequently also present themselves as responses from knowledgeable subordinates to a question from a superior authority. This rhetorical device may also give us a clue to the origins of this hybrid genre of writing: as Sumit Sarkar suggested, it stems from an increasingly bureaucratic regime in which successive rulers needed to know their domains, and began by demanding testimony from local officials. Such inquests would be common wherever royal authority had weakened powerful intermediary lords and penetrated to the localities. This is suggested in one of the very earliest, if not earliest, bakhars extant—the Mahikavatici bakhar.⁵⁸ It records that after the kingdom was conquered by Nagarsya:

Samvat 1228 that is ... Salivahana Saka 1163 the 7th day of the bright half of the month Magha, Tuesday ... then the people of the land gathered. The district gentry, their assistants, the village elders, the cleverer village headmen, tax officers, merchants, Brahmans—all communities brought their records, papers and charters and all came into the royal presence.⁵⁹

It is unlikely this happened at either of the dates given, but it was common some centuries later. As Rajwade pointed out, this manuscript comprises not one but six distinct works by at least two authors, subsequently copied and recopied over perhaps two hundred years into a single manuscript. Indeed, it is in many ways an excellent example of a transitional bakhar text/texts. The first part begins with the heading: "Lines of descent–lineage origins–the separation of classes–exposition." The author describes how the goddess of speech animated him, and then refers to a Sanskrit Bhavisyottara Purana as his source and re-tells some traditional cosmological myths. We learn of families descended from the Sun, the Moon, or the Serpent Sesa, and names and affiliations of the kings who ruled from the city of Paithan. It ends by explaining the multiplicity of castes by proper and improper marriages, and affirms its source to be yet another purana.⁶⁰ Rajwade dates this poem to the sixteenth century, and that certainly seems plausible on linguistic grounds.

The next major segment of the manuscript is in prose, and brings us into the familiar territory of the bakhar. The author makes a ferocious effort at verisimilitude and plunges us into precise time and specified place: "In the year of Vikramarka 1125 that is ... Salivahana Sake 1060, ... the bright half of the month of Magha, 5th day, Monday, from Anahilavadapattana near Champaner."⁶¹ The narrative then plunges into a brisk if sometimes inconsistent history of land grants and land seizures clearly intended to explain and justify contemporary claims to land and estates. Another ten pages describe how disputes over estates or honors were settled under the Navait Muslim rulers and their successors, the sultans of Gujarat. Finally, portions of a story found earlier in the text are recopied in order to explain the ancient roots and noble descent of a set of families not previously mentioned. I could go on. In short, this is a composite document whose segmented structure exhibits the different narrative modes that went into constituting the more integrated narratives produced from the seventeenth century onward.

But let us return to the territorial inquest as another source of the bakhar genre. It is doubtful if the Raja Nagarsya indeed ordered such an inquest, but later records of such inquiries survive—as, for example, when the Maratha king Shivaji conquered the districts of Cheul and Rairi in the early 1650s. He summoned the hereditary gentry-administrators and chiefs of the merchants to the fort of Prabalgad, and ordered them to provide a complete account of the tax collections under the preceding regime. "Thereupon Appaji Tavji, hereditary district accountant [deshkulkarni] said that under the Nizamshahs and Adilshahs we would make a lump-sum assessment and did not keep accounts of the [actual] payments made by the peasants." He was punished, and arrangements made for local accounts to be kept in the future.⁶² Historical narratives were important in bargaining over tax rates—thus on the reestablishment of Maratha power in Satara district after 1708, local hereditary officials were able to produce a detailed memorandum of the tax rates fixed at the coronation of Chatrapati Shivaji in 1674, as well as the changes made under his successor Sambhaji.⁶³ The corresponding officials in the Anjanvel subdivision of the Konkan maintained an even more detailed account, last updated for the British in 1824. Its real focus was on tax arrangements, but included historical notes such as those excerpted below:

Kanada Ramraja Chhatrapati by caste [jati] a Lingayat Vani—in his time there was no fort in the subdivision; then Pavar of the Maratha caste built a fort at Gudhe in Valambe area and set up the Gudhe province about 600 years ago. The province was depopulated; taxes were not collected by measurement; instead the cultivated area was estimated annually and a sum levied after negotiation with the peasants ... This system continued for about 275 years, and then Patshahi rule began.
    Ali Idilshah and Bahri Nizamshah, who were brothers, killed the Kanada Ramraja at the camp of Anegondi, and Ali Idilshah enthroned himself at Bijapur in the year salas tisa[in?] maya [193?] Sake 1424[1501–2 CE] ...
    In the manner detailed above the Adilshahs ruled; in that period the land tax was not assessed on the basis of a half-share from each cultivator; the demand was adjusted depending on the state of the cultivation—this was the system during the entire rule of the Padshahs.
    Shivaji Bhosle lived in the vicinity of Chandi-Chandavar [sic; Senji and Thanjavur]; Bhosle and Jadhavarao were related by marriage; a family quarrel ensued, and Bhosle rose in rebellion from 1574 [CE 1652]—he continued for 22 years. Then in 1596 [1674 CE] Shivaji Maharaj Chhatrapati was crowned at Raigad, and he sent generals with an army to occupy the Konkan. Sivaji Maharaj conducted affairs for 22 years before his enthronement and for six after it.⁶⁴

Thus the discussion of tax arrangements under different regimes would also involve a historical narrative about them. But it is only in the 1650s that the historical notes gain a degree of accuracy. I believe that this reflects the new demands of Shivaji's administration, which led officials quickly to collect then current oral traditions to lend authority to their description of previous tax regimes, and to keep regular notes thereafter.

But the judicial case and the administrative inquest were not the only templates for the bakhar. These factual narratives also came to serve the identity of the Marathas as a rising regional elite. The above-cited officials of Anjanvel certainly saw the coronation of the Maratha sovereign Shivaji in 1674 as an epoch-making event. That act was the culmination of a campaign that both drew on and reinforced local and regional patriotism. Shivaji began raising a local militia in the 1640s and seized some crucial strongholds in western Maharashtra from his nominal overlord, the Adilshahi sultan. The sultan's ministers sent threatening letters warning the local gentry that their heads would be struck off if they connived with him. Shivaji reassured a wavering supporter in a message that subtly highlighted his own rootedness in the locality: "Rohidesvara, the original presiding deity of your valley, who exists in self-created form next to the sendri tree on the plateau at the crest of your peak, has assured me success."⁶⁵ Shivaji succeeded in founding an independent kingdom that survived a full-scale Mughal assault. The introductory chapter to Ramacandra Pant's treatise on statecraft was written in 1717: "This is the kingdom that was assailed by a powerful enemy like [the Mughal emperor] Aurangzeb and still beat him back. Aurangzeb focused all his forces and expended all his resources against it, but failed to conquer; bootless, disheartened he took himself off to the land of death. And Aurangzeb was master of fifty-four kingdoms; endowed with wealth and soldiers second to none, of whom it was famously said 'either the lord of Delhi or the lord of the world' [are equal in wealth]."⁶⁶ Dynastic identity thus merged with and built on regional pride, and the protean bakhar was turned to yet another narrative: the construction of a common glorious past.

Not surprisingly, therefore, one of the most famous early bakhars—that detailing the life and deeds of Shivaji, was written in 1694, during the hard days of the struggle with the Mughal emperor Aurangzeb, whose armies had overrun most of Maharashtra. Its author Krishnaji Anant Sabhasad framed text as an answer to King Rajaram's question: "My father achieved great feats of arms, confronted four empires—such was his valor. That being so, how has Aurangzeb been able to come and wreck fort after fort? You are an old and knowledgeable servant of the state—write an account of his [Shivaji's] deeds [caritra] from the very beginning." Significantly enough, the text, while opening with an invocation of the deeds of the founder-hero, ends by speaking in terms of a regional patria ruled by a Maratha king: "He proclaimed his dominance from the Narmada river up to Ramesvara [traditional southern extremity of India] and captured the land. Adalshai, Kutubshai, Nizamshai, Moglai—these four empires as well as the twenty-two empires from beyond the seas were kept in check and a new empire founded; a Maratha Padshah seated himself on the throne as a Chhatrapati [sovereign monarch]." It is perhaps not without significance that, while the work opens by describing itself as a caritra (biography), the closing sentences refer to it as a bakhar.⁶⁷ The Maratha identity had been foregrounded from at least the 1670s, when a dependent of the newly crowned Chatrapati wrote a Sanskrit epic on his career. It begins with Shivaji's grandfather, Malavarman, who is emphatically described as a Maratha king ruling in Maharashtra.⁶⁸ The identity was also extensively invoked by Shivaji's sons and successors during the darkest periods of the war with the Mughals.⁶⁹

As the Marathi-speaking gentry who had long written or dictated narrative histories for consumption in the regional political arena expanded their horizons to a dominance over the peninsula and ultimately, the entire Indian subcontinent, a fresh layer of macro-narratives was added to the local and particular histories that continued to be compiled. So, in 1813, we find a manuscript that applies the same term to what is clearly seen as a Maratha national history: Svarajya mhanon Maharashtaraje ajruyi bakhar—a literal translation would be "the authentic bakhar of our kingdom, that is, of the Maharashta kingdom."⁷⁰ The change first identified by Prachi Deshpande had occurred, and we now also need to "view the bakhars as texts that were closely linked to the spread and exercise of Maratha power in the eighteenth century and which played an important part in its practice and representation."⁷¹ The qualification that I would add to this formulation of the link between Maratha power and the bakhar genre is that only some select macro-bakhars (to coin a gross neologism) worked primarily with that goal, and a large number continued to build narratives centered on lesser lineages and petty disputes, as well as the lives of worthy and holy personages and their doings.

Deshpande would exclude "prose hagiographies" from the bakhar genre; but the distinction of sacred and secular was not made in the texts themselves—after all, the Sabhasad Bakhar credits dreams, omens, and prophecies.⁷² Supernatural intervention was accepted and invoked in the prosaic world of the law court: if written and oral evidence failed, the disputants could offer to undergo an ordeal. Religious foundations could also get involved in property disputes that required secular adjudication. Perhaps this even led to a change in the tone of some hagiographies themselves. A full-length hagiographic bakhar—the life of the seventeenth- century holy man Samartha Ramdas written in 1793 by his follower Hanumantsvami—opens with a narration of sources and an admission of their incompleteness.

I had [earlier] written a life of the Svami. You asked me to communicate it to you. It has been expanded on the basis of books, and on what the company of his contemporary disciples wrote down and preserved, and also certain stories that I have myself heard. This is communicated to you. Who indeed can tell you everything about the Svami? His deeds are infinite, his qualities infinite and so too his devotion. What is knowable I have written.⁷³

By 1805, the British began visibly to overshadow the Maratha empire, and in 1817–1818, Pune and Nagpur—the two major cities of Maharashtra— came under British military occupation. The early British regime sought eagerly for bakhars: a manuscript found in the fort of Rairi (in 1803?) was translated and found its way into the diplomatic and military records of the Bombay government secretariat.⁷⁴Bakhars were certainly used by Grant Duff, the first colonial historian of the Marathas.⁷⁵ They continued to be written well after the imposition of colonial rule and the beginnings of Western education, and they differed little from the developed bakhars of the eighteenth century. The loss of political power now seems to have sustained efforts at recording the past. One Krishnaji Sohoni served the Maratha regime in its last decades; he subsequently became an ascetic and lived in the holy town of Vaijnatha, where his narrative of the period when the Brahman ministers were ascendant (c. 1720–1818) became sufficiently popular for at least two persons to compile it into history. Sohoni and presumably most of his auditors were Brahmans, and their historical assessment is summed up in the disillusioned epilogue to the story:

God granted a boon that Brahmans should rule the earth for a hundred years in the Dark Age. This is in the Puranas and the words of [the ancient seer] Vyasa. So it happened. In Sake 1635 [1712–1713] the Pesvas' government came into being; in Sake 1739 the kingdom fell. Until the time of Savai Madhavrao the kingdom was dominant. All its enemies lay underfoot with folded hands. Then, after Bajirao Raghunath came to the throne [1796], there were no more campaigns and expeditions, and the ruler saw no need to maintain armies. And so the present state of things has come about. ⁷⁶

Early colonial efforts to settle land tenure and limit tax exemptions also stimulated the writing of additional histories by landed families justifying their claims. Furthermore, as recently demonstrated in a penetrating article by Wagoner, early colonial narratives and reports were underpinned by the paleographic and linguistic work of traditional scholars, who often also wrote short histories.⁷⁷ Selections from this information were then seamlessly assimilated into the master narratives constructed in the colonial reports that provided the templates for most colonial history. We can see this happening by (for example) comparing the manuscript reports of Richard Jenkins's assistant Vinayakrao and Jenkins's own Report on the Territories of the Rajah of Nagpore.⁷⁸ The latter, in turn, was used in a multitude of colonial histories through the next century.

The responses to colonial history soon moved beyond oblique attacks such as that made in Sohoni's bakhar. Nilakantharao Kirtane published the first modernist Marathi critique of British historian Grant Duff's History in 1867. The first "modern" Marathi historical novel also appeared that year.⁷⁹ By 1890, Rajwade had decided that solidly grounding history in the latest methods of source criticism would enable India to recapture its own history. He commented that efforts like Kirtane's were hamstrung by a lack of dependable contemporary sources. Rajwade was scathing in his critique of the bakhars that were slowly coming into the world of print, and devoted several pages of the introduction to his first volume of sources (published in 1898) to a denunciation of their exaggerations, anachronisms, chronological confusions, and love for fabulous stories. The truly important sources, he held, were contemporary letters, dispatches, and other documents, and he used a few of these to demonstrate triumphantly how Grant Duff had been completely wrong in his account of the movements of the Maratha armies in 1754.⁸⁰

And so Rajwade launched the enterprise of defeating colonial knowledge by crushing it with a superior body of primary, contemporary sources. At the same time, another response to colonial dominance came in the form of the historical novel and play. As Prachi Deshpande has pointed out, novelists and dramatists drew heavily on historical documents to lend verisimilitude to their works.⁸¹ Marathi historical novels multiplied parallel to the new public interest in history. The major novelist Harinarayana Apte declared that the two pillars of the historical novel were oral traditions (dantkatha) and (authentic) histories (itihasa).⁸² Such novels clearly inherited much of the color and anachronistic spirit of the traditional bakhars. Meanwhile, Rajwade struggled with the difficulties of writing satisfying historical narratives in a rigorously documented mode. In his last years, he took a more sympathetic view of the bakhars—especially the Mahikavatici Bakhar, which he edited and published in 1924. In a lengthy introduction, he bends over backward to provide an authentic factual basis for the diverse and contradictory narratives contained in it; and in his conclusion he casts two personages from it as heroes—Nayakorao, the warrior, and Kesavacarya, the Brahman preceptor.⁸³ It is clear that he projects them as fifteenth-century forerunners of a Hindu nationalist movement that he desired in the colonized present. So the pioneer of the strict construction of verifiable history from contemporary sources found its results ultimately unsatisfying, and returned to narrative modes far closer to those of the bakhars he had formerly denounced. But the rich archive that his generation recovered and preserved still stands, echoing with voices that yet contest the hegemonic claims of nations and communities (and the property claims of their neighbors).

This article had two objects: to survey the history of modern historical practice in a postcolonial nation-state, and to carry out a more detailed examination of the structures that generated historical texts in the under-researched pre-colonial era. Through this juxtaposition of current and past historical practices, I have attempted to show how we modern historians can situate the work of our obscurer predecessors without resorting to relativism. Relativists would simply declare that all forms of memory are valid, and then return to the safety of their own commonsense version of it. If I have succeeded, this essay uncovers the setting that generated other historical practices, and thus enables us to compare them with our own. The past is a foreign country. But we can hope to know why they did things differently there.

The writing of this essay has been made possible by a fellowship from the American Council of Learned Societies. An early draft was read at the Annual Meeting of the American Historical Association, San Francisco, January 6, 2002. I am indebted to Nancy Hewitt, Ronald Inden, Steven Lawson, Vijay Pinch, and Phillip Wagoner for comments on earlier drafts. I am deeply indebted to Prachi Deshpande for permission to read and cite her unpublished thesis and for many invaluable comments and suggestions. This essay has also benefited from several close readings by Indrani Chatterjee, the comments of six anonymous reviewers, and the labors of the American Historical Review's editorial staff. The usual disclaimer holds.
I have translated all the Indian-language texts cited herein. With one exception, all citations are from either the Modi or the Nagari script: when rendering into Roman, I have followed the appropriate convention but omitted diacriticals. This convention has also been followed for proper names, barring a few that have established Roman renderings ("Shivaji," for example).

Sumit Guha is a professor in the Department of History, Rutgers University. His recent books are Environment and Ethnicity in India, c.1200–1981 (1999) and Health and Population in India (2001).This article is part of his current study of the politics of language and memory in India through the past millennium.

Notes

¹ Bonnie G. Smith, The Gender of History: Men, Women and Historical Practice (Cambridge Mass., 1998); Dipesh Chakrabarty, Provincializing Europe: Postcolonial Thought and Historical Difference (Princeton, N.J., 2000), esp. chap. 4; Turning Points in Historiography: A Cross-Cultural Perspective, Q. Edward Wang and George G. Iggers, eds. (Rochester, N.Y., 2002); Peter G. Bietenholz, Historia and Fabula: Myths and Legends in Historical Thought from Antiquity to the Modern Age (Leiden, 1994); Glen W. Bowersock, Fiction as History: Nero to Julian (Berkeley, Calif., 1994); History and the Present, Partha Chatterjee and Anjan Ghosh, eds. (Delhi, 2002).\.

² Donald R. Kelley, Fortunes of History: Historical Inquiry from Herder to Huizinga (New Haven, Conn., 2003), 337–38.

³ Ronald Inden, "Introduction," Querying the Medieval: Texts and Practices in South Asia, Inden et al., eds. (New York, 2000), 4.

⁴ Sheldon Pollock, "India in the Vernacular Millennium: Literary Culture and Polity 1000–1500," Daedalus 127, no. 3 (Summer 1998): 41–74.

⁵ For an analysis of how the rise of Maratha power bore on linguistic usage, see Sumit Guha, "Transitions and Translations: Regional Power and Vernacular Identity in the Dakhan c. 1500–1800," Comparative Studies of South Asia, Africa and the Middle East 24, no. 2 (Fall 2004).

⁶ The relevant literature is to be found in at least sixteen South Asian languages plus English. It is obviously impossible for me to do more than give a few examples. As the ultimate focus of this article is on Marathi-language writings from western India, most of my examples will be drawn from there. To avoid confusion, let me clarify that "Marathi" is the language; "Maratha" is the ethnonym, and "Maharashtra" is the name of the Indian state predominantly inhabited by speakers of Marathi.

⁷ François Furet, "The Birth of History in France," In the Workshop of History, Jonathan Mandelbaum, trans. (Chicago, 1984), 98.

⁸ Prasenjit Duara, Rescuing History from the Nation (Chicago, 1995), 4.

⁹ Youssef M. Choueiri, Modern Arab Historiography: Historical Discourse and the Nation State, rev. edn. (London, 2003), 71. Also see David C. Gordon, Self-Determination and History in the Third World (Princeton, N.J., 1971). Even the matchless historical tradition of China had to be recast. What China needed, the leading intellectual Hu Shih wrote in 1914, was: "1. the principle of inductive reasoning; 2. a sense of historical perspective; 3. the concept of progress." Cited in Min-Chih Chou, Hu Shih and Intellectual Choice in Modern China (Ann Arbor, Mich., 1984), 168. See also Q. Edward Wang, "China's Search for a National History," in Wang and Iggers, Turning Points in Historiography, esp. the telling quote from Teng Shih, 196.

¹⁰ Dayanand Sarasvati, Satyartha Prakasa (1884; 34th rpt., Ajmer, 1998). This method is deployed throughout his work: for example, "when the Veda says this, why would intelligent people adopt the delusions of some foreigner?" 237.

¹¹ See Laksamanasastri Josi's introduction to his edition of Rajavade Lekhasangraha (1958; rpt. edn:, Delhi, 1992), 1–13.

¹² Smith, Gender of History, 103–29, 128.

¹³ Vinayak D. Savarkar, The Indian War of Independence (1908; first Indian edition, Bombay, 1947).

¹⁴ Valentine Chirol, Indian Unrest, cited in Bhaskar G. Deshpande, Krantisurya Savarkar, 2nd edn. (Pune, 2002), 110.

¹⁵ Sumit Sarkar, "The Many Worlds of Indian History," in Writing Social History (1990; rpt. edn., Delhi, 1999), 38–39.

¹⁶ Ranajit Guha, "On Some Aspects of the Historiography of Colonial India," introduction to Subaltern Studies I (Delhi, 1982), 1–8.

¹⁷ Ranajit Guha, An Indian Historiography of India: A Nineteenth-Century Agenda and Its Implications (Calcutta, 1987), 11–12, 33–35. The Purana(s) are massive Sanskrit compendia of traditional lore, including myths, king-lists, and legends.

¹⁸ Majumdar confessed himself unable to explain what he viewed as the lack of historical writing in India. "This fact seems to be more of an accident than the result of any definite cause or causes." Ramesh C. Majumdar, "Ideas of History in Sanskrit Literature," in Historians of India, Pakistan and Ceylon, Cyril H. Philips, ed. (London, 1961), 27.

¹⁹ Sarkar, "The Many Worlds of Indian History," 6.

²⁰ Anthony K. Warder, An Introduction to Indian Historiography (Bombay, 1972), viii-ix.

²¹ Vishwambhar S. Pathak, Ancient Historians of India: A Study in Historical Biographies (Bombay, 1966).

²² Romila Thapar "Ideology and the Interpretation of Early Indian History," read at Cornell University, 1974, published in Interpreting Early India (1989; rpt. edn., Delhi, 1999), 20.

²³ James Siegel, Shadow and Sound: The Historical Thought of a Sumatran People (Chicago, 1979), 209–10, and footnotes. Emphasis added.

²⁴ Shelly Errington, "Some Comments on Style in the Meanings of the Past," Journal of Asian Studies 38, no. 2 (February 1979): 231–33.

²⁵ Virginia Matheson, "Strategies of Survival: The Malay Royal Line of Lingga Riau," Journal of Southeast Asian Studies 17, no. 1 (March 1986): 34.

²⁶ Nicholas B. Dirks, The Hollow Crown: The Ethnohistory of an Indian Kingdom (1987; new edn., Cambridge, 2000), 58–59.

²⁷ Phillip B. Wagoner, Tidings of the King: A Translation and Ethnohistorical Analysis of the Rayavacakamu (Honolulu, 1993), 10.

²⁸ Propagated by pamphlets like Ramjanmabhumi ka Raktaranjit Itihasa (Bloody History of the Birthplace of Ram), cited in Sarkar, Writing Social History, 2.

²⁹ Daud Ali "Introduction," Invoking the Past, Ali, ed. (Delhi, 1999), 4.

³⁰ Sanjay Subrahmanyam, "Recovering Babel: Polyglot Histories from the Eighteenth-Century Tamil Country," in Ali, Invoking the Past, 280–321; V. Narayana Rao, David Shulman, and Sanjay Subrahmanyam, Textures of Time: Writing History in South India 1600–1800 (New Delhi, 2001; New York, 2003).

³¹ Textures of Time, 2–5, emphasis original; for karanams, see esp. 19–20, 115–18.

³² The confident assertion of a special unrecorded history of the "Hindu" community is found in BJP's White Paper on Ayodhya and the Rama Temple Movement (introduction by L.K. Advani), no place, dated April 1993. "There are unrecorded traditions of many military expeditions by the Hindus between 1528 and 1707 for repossessing the Janmabhoomi," 20, emphasis added.

³³ Partha Chatterjee, "Introduction: History and the Present," in Chatterjee and Ghosh, History and the Present, 17.

³⁴ "A Taste of Bamiyan," Outlook, January 19, 2004 (weekly from New Delhi; on the web at www.Outlookindia.com), 18–19.

³⁵ The complete award includes a critical summary of the pleas and a scrutiny of documents produced on both sides. It dates from 1773–1774. Those attesting the award include Muslims, Brahmans, local artisans, tradesmen, and low-caste village servants. Bharata Itihasa Samshodhana Mandala Varsika Itivritta Sake 1835 [1912–1913]: 360–72.

³⁶ Ludo Rocher, The Puranas (Wiesbaden, 1986), 123.

³⁷ See, for example, the detailed statistical statements of Jodhpur in Mumhta Nainsi-ri Likhi Marwar-ra Parganan-ri Vigat, Narayanasingh Bhati, ed., 3 vols. (Jodhpur, 1968–1974), 1: 145–70, 186–372. But these scribes would be more numerous and less accomplished than the elite polyglot karanams envisaged in Textures of Time. A valuable annotated translation of significant portions of the Rajasthani narratives has recently appeared: Richard D. Saran and Philip M. Ziegler, The Mertiyo Rathors of Merto, Rajasthan: Select Translations bearing on the History of a Rajput Family, 1462–1660, 2 vols. (Ann Arbor, Mich. 2001). They note that the narrative portions of Nainsi's work are devoted almost entirely to clan histories; 11.

³⁸ Badriprasad Sakariya, ed., Mumhta Nainsiri Khyat, 4 vols. (Jodhpur, 1984–1994), 1: 1.

³⁹ Rajaram V. Hervadkar, Marathi Bakhar, rev. enl. edn. (Pune, 1975); Bapuji Sankpal, Bakhar Vangmaya: Udgama ani Vikas (Pune, 1982); Marathi Vangmayaca Itihasa, Ramacandra S. Jog, ed. 6 parts (1963–1993; rpt. edn., Pune, 1999), 3: 490–571. Most of the background information in this essay is drawn from these sources. Many interesting glimpses of the London manuscript collection are to be had from James F. Blumhardt and S. G. Kanhere, Catalogue of the Marathi Manuscripts in the India Office Library (London, 1950).

⁴⁰ I take this idea from the outstanding work of Prachi Deshpande, "Narratives of Pride: History and Regional Identity in Maharashtra, India c. 1870–1960," (Ph.D. dissertation, Tufts University, 2002), 47–55; and "Writing Regional Consciousness: Maratha History and Regional Identity in Modern Maharashtra," in Region, Culture and Politics in Maharashtra, Rajendra Vora and Anne Feldhaus, eds. (Delhi, forthcoming).

⁴¹ Sankpal, Bakharvangmaya, 19–27; Hervadkar, Marathi Bakhar, 1–5.

⁴² Dhundhiraja, Rajavyavaharakosa, ms. composed c. 1676; Kailash N. Sane, ed., printed in Sivacaritrapradipa, Gangadharrao V. Mujumdar and Dattatreya V. Apte, eds. (Pune, 1925), 169.

⁴³ Amarasimha, Amarakosah, annotated by Visvanatha Jha (Delhi, 1999), 1: 58.

⁴⁴ Verdicts from judicial assemblies often ended with the warning: "whoever violates this will be a wrong-doer before the community and punishable by the Diwan."

⁴⁵ Datto V. Potdar and Gangadharrao N. Mujumdar, eds., Sivacaritrasahitya 2, (Pune, 1930), 113–18.

⁴⁶ This interesting manuscript is in the British Library, London, Mss Mar A-7 pt. 1, chap. 6, verses 40–41, 48: "tenen svamate lihili bakhar."

⁴⁷ Sankpal, Bakhar, 70.

⁴⁸ For an excellent account with many documents attached, see Vitthala T. Gune, The Judicial System of the Marathas (Pune, 1953).

⁴⁹ Sakavali means "sequence of years." Ananta R. Kulkarni, ed., Jedhe Shakavali-Karina (Pune, 1999); Karina, an Arabic loan-word, is derived from qrn—to join or connect; it is here applied to an explanatory history. John T. Platts, A Dictionary of Urdu, Classical Hindi and English (1884; rpt. edn., Delhi, 1997).

⁵⁰ Maharashtra State Archives, Pune, Parasnis transcripts, vol. 12, fols. 22–23.

⁵¹ Parasnis transcripts, vol. 12, fol. 87.

⁵² Selections from the Peshwa Daftar, Govind S. Sardesai, ed., 46 vols. (Bombay, 1931–1935), 31: 67.

⁵³ Rajwade, Marathyancya Itihasancin Sadhanen, 22 vols. (various places, 1898–1918), Vol. 15 (Dhule, 1912); reprinted as volume 2 in the new series (Dhule, 2002), 2: 86–89; Appasaheb Pavar, ed., Tarabaikalina Kagadpatre, 2 vols. (Kolhapur, 1969), 1: 189–203. I owe this idea to Prachi Deshpande (personal communication).

⁵⁴ Mandala Itivritta Sake 1835, 43–45.

⁵⁵ Sivacaritrasahitya 2, Potdar and Mujumdar, eds., 113–18.

⁵⁶ Thomas Coats, "Account of the Township of Lony," in Transactions of the Literary Society of Bombay 3 (1823): 192. He adds, "The cultivators ... form almost the whole of the population of the township ... A school is established in the town, where reading, writing and arithmetic are taught; but it is only attended by the children of brahmins, shopkeepers and patails [headmen], so that scarcely any of the cultivators can either read or write."

⁵⁷ Jean-Baptiste Bossuet, Discourse on Universal History [1681], Elborg Foster, trans. (Chicago, 1976).

⁵⁸ One manuscript was edited and printed by V. K. Rajwade: Mahikavatici Bakhar (1924; rpt. edn., Pune, 1999). Others exist: a slightly variant one is British Library, London, Add Mss 26494 B.

⁵⁹ Mahikavatici Bakhar, text: 41. The two dates do not match.

⁶⁰ Mahikavatici Bakhar, text: 1–31.

⁶¹ Mahikavatici Bakhar, text: 32. The dates do not match. Rajwade dated this part of the text to 1448 CE.

⁶² Peshwa Daftar, 31: 21.

⁶³ Peshwa Daftar, 31: 88–90.

⁶⁴ Mandala Itivritta Sake 1835, 319–23. The first Arabic year is garbled, and neither of the events mentioned occurred in 1501–1502. The later dates are quite accurate. The Arabic words spell out the Suhur year.

⁶⁵ Rajwade, Marathyancya Itihasacin Sadhanen, new series, 2: 242.

⁶⁶ Ramacandra Pant, Ajnyapatra, S. N. Banhatti, ed. (1961; rpt. edn., Nagpur, 1986), 59.

⁶⁷ Krishnaji Anant, Sabhasad Bakhar, Bhimrao Kulkarni, ed. (Pune, 1987), text: 1, 100.

⁶⁸ Paramananda, Sivabharatam, introduction by Vinayakraya Apte (Pune, 1930), Chap. 1, verses 42–43.

⁶⁹ Setumadhavrao Pagdi, Hindvi Svarajya ani Mogal (Pune, 1966), 17–18.

⁷⁰ British Library, London, Mss Mar D.31, fol. 126.

⁷¹ This most important insight is from Deshpande, "Narratives of Pride," 19–20.

⁷² Sabhasad Bakhar, 2.

⁷³ Printed in Ramdas ani Ramdasi 27 (1958–1959): 239.

⁷⁴ "An Account of Shahaji and His Son Shivaji," in George W. Forrest, ed., Selections from the Letters, Despatches, and Other State Papers in the Bombay Secretariat: Maratha Series (Bombay, 1885), 1–22.

⁷⁵ James Grant Duff, A History of the Mahrattas, 3 vols. (London, 1826). The Marathi translation appeared in 1829–1830 and became a required schoolbook, reprinted eleven times by 1838. Jog, ed. Marathi Vangmayaca Itihasa, 4: 553–54.

⁷⁶ Pesvyanci Bakhar, Bhimrao Kulkarni, ed. (Pune, 1987), 146.

⁷⁷ Phillip B. Wagoner, "Precolonial Intellectuals and the Production of Colonial Knowledge," Comparative Studies in Society and History 43 (October 2003): 783–814; also Nicholas B. Dirks, "Colonial Histories and Native Informants: Biography of an Archive," in Orientalism and the Postcolonial Predicament, Carol A. Breckenridge and Peter van der Veer, eds. (Philadelphia, 1993), 279–313.

⁷⁸ Richard Jenkins, Report on the Territories of the Rajah of Nagpore (1830; new edn. with modernized spelling, Nagpur, 1923), acknowledges his British assistants, but never mentions Vinayakrao. Hundreds of folios of Vinayakrao's Modi script field-notes, tour diaries, and transcriptions from 1808 to 1823 are in the British Library, London, Marathi Mss D.35 to D.46.

⁷⁹ Marathi Vangmayaca Itihasa Jog, ed., 4: 556.

⁸⁰ Rajwade, Itihasacarya Vi. Ka. Rajwade: Samagra Sahitya, 10 vols., M.B. Shaha, ed. (Dhule, 1998), 10: 1–8, 24–25.

⁸¹ Deshpande, "Narratives of Pride," 116–44.

⁸² I rely here on Jog, ed., Marathi Vangmayaca Itihasa, Pt. 5, 1: 448–49; Deshpande, "Narratives of Pride," 124, cites the popular novelist V. V. Hadap: he claimed to use his imagination only where "historical evidence had failed in its task" (Deshpande's translation).

⁸³ Mahikavatici Bakhar, introduction, esp. 109–10.

Content in the History Cooperative database is intended for personal, noncommercial use only. You may not reproduce, publish, distribute, transmit, participate in the transfer or sale of, modify, create derivative works from, display, or in any way exploit the History Cooperative database in whole or in part without the written permission of the copyright holder.