A True Copy? Documents and the Production of Legality in the Bombay Inam Commission

Abstract

This essay examines the role of Indian-language documentation in the production of legality in colonial western India, focusing on the workings of the Bombay Inam Commission (1852-1863). It situates legal validation of claims to tax-free land revenue within the broader process of securing, organizing, classifying, and registering Marathi- and Persian-language documents. Combating the effects of rain, dirt, and pests on old state records often sold as “waste paper,” the Inam Commission deployed material interventions to secure a legal archive for verifying individual claims to property. While such evidence weighed heavily in the evaluation of the testimony and corroborating documents of an individual claimant’s case-file, questions of writing also shaped the legal reasoning of the Commission. Inquiries about any given document’s conformity to or deviation from conventional style figured prominently in judgments about its authenticity. The scribe Sayyid Usman’s investigation in 1856 of a date in a Persian document attributed to the Mysore ruler Tipu Sultan struggled to establish the parameters of conventional style against the plurality of entangled regimes of property. I argue that a material approach to writing allows us to better understand the imperfect and dispersed production of legal truth in imperial settings.

As he copied down one of many old land grants piling up in a crowded government office in colonial India, the Persian scribe Sayyid Usman may have reached for a sheet of paper that survives today in the Pune branch of the Maharashtra State Archives. In the center of the page are three circles, one of which contains the beginnings of a seal of a judge named Karimuddin Khan. Surrounding the outlines of seals are disconnected lines of text: a Bismillah invocation, a reference to the signature of an unnamed Nawab, and a notice that a document had been written on January 14, 1857 and was received in the office of Hearn Saheb, one of the copyist's supervisors. More mysterious are several inscriptions of the word “self-rule” (swarajya), both in the Modi script of Marathi and the Nastaliq script of Persian and Arabic, in the bottom right corner. The Marathi phrase “God's displeasure will occur (iswarachi avkrupa hoil),” appears in both scripts at the top left. Was the copyist doodling and perhaps daydreaming of independence, mere months before the 1857 rebellion against British rule? Or did he stumble across these words in a royal order, swarajya commonly referring to the homelands of the pre-colonial Maratha state? Or might his clumsy attempts at writing “self-rule” in Modi (suraj, sadaj, swarajya) suggest that they were merely convenient locutions for practicing handwriting? Or were they the ephemeral leavings of not one, but multiple hands of clerks responsible for decoding, copying, stamping, and filing documents acquired by the British colonial state? Although the itinerary of this particular scrap of paper may prove elusive, questions about writing at the margins were central to the British imperial state's encounter with documentation in South Asian languages.¹

It is now a scholarly commonplace that early modern empires relied on the control and circulation of written records, and the British Empire in South Asia was no exception.² Even prior to its transformation from joint-stock corporation to territorial state, the British East India Company organized long-distance trade, managed its employees, and established diplomatic relations with Indian rulers through the exchange of accounts, treaties, and correspondence.³ By the late eighteenth and early nineteenth centuries, the Company began to develop a more robust administrative capacity by establishing central bureaucratic offices and absorbing existing local administrative structures in the process of conquest and negotiation with Indian rulers.⁴ The Company's bureaucratic apparatus emerged in tandem with its turn toward land surveying, revenue collection, and legal adjudication, areas of governance in which written information about local practice was indispensable.⁵ In perhaps the most influential recent contribution to this literature, Bhavani Raman explores the ways in which the Company “cutcherry” was shaped by tensions between “continuous writing and discretion,” which manifested in recurring anxieties about forgery, counterfeit, and concealment.⁶ Scribal knowledge of conventional writing and embodied practices of mnemonic recall and recitation were integral to the Company state's efforts to decode Indian-language documents, yet their inextricability from forms of attestation associated with local custom and authority was incompatible with new evidentiary paradigms. Beyond the control of scribal labor and reformation of scribal praxis, documentation itself became a site of colonial governmentality, in part through the creation of new material technologies of writing like the file, the stamp paper, and the public register.⁷

Scholarly investigations of colonial documentation in South Asia have been powerfully shaped by a somewhat older set of debates about power, agency, and the production of knowledge about South Asia. Whereas earlier arguments emphasized the powerful ruptures produced by colonial discourse about Indian society, subsequent investigations revealed the critical role played by Indian intermediaries in gathering and deciphering information and, in some cases, honing the assumptions and methodologies of key fields of knowledge.⁸ The classic figure of the “native” informant has been richly complicated by social, cultural, and intellectual histories of the scribal classes whose expertise was redirected toward the twin imperatives of maximizing revenue yields and facilitating a market-driven legal regime of property.⁹ Conversely, the colonial bureaucracy itself no longer appears to be an epistemic monolith, but rather a fallible construction of both British and Indian intelligence-gathering prone to “information panics” at the seams and borders of state control.¹⁰ In relation to this historiography, this essay suggests that the colonial production of legal knowledge equally grappled with the persistence of diverse South Asian languages, scripts, and forms of writing. Persian and regional languages persisted in both official administration and everyday communication, albeit within a new linguistic economy of prestige structured by the dominance of English.¹¹ Furthermore, numerous types of Indian-language documentation remained legitimate instruments of legal and political discourse.¹² This essay proposes that attention to the sometimes irresolvable questions posed by the content, form, and materiality of Indian-language documents will produce new insights into how knowledge was made at the seams of colonial power and “native” expertise.

In revisiting the problem of paperwork in South Asian languages, this essay borrows some insights from the burgeoning field of cultures of documentation in the Islamicate and Persianate worlds.¹³ Rather than assume a lacunae of “original” documentation in the centuries prior to colonial rule, or mine those that have survived for morsels of “useful” information, it has been argued that we must carefully examine how the formal conventions of writing structured the very social, political, and legal realities to which they referred.¹⁴ Nandini Chatterjee has made a persuasive case for the merits of this approach in her study of a landlord family's relationship to the everyday Mughal state. By identifying stable terminologies of rights and office-holding across their multilingual archive, she is able to illustrate “the percolation of a vocabulary of governance” across multiple layers of authority.¹⁵ Attention to the entire life cycle of documentation—from composition and production to circulation and consumption to storage and disposal—promises to shed light on how state and corporate entities, families and individuals managed their entitlements and obligations across periods of historical change. In line with this new approach, the aim of this essay is not to offer another account of the anxieties of colonial bureaucracy about “native” corruption, but rather to understand how the form and materiality of Indian-language documents became key sites of intervention in the production of legal truth under the British colonial state.

This article examines the role of Persian- and Marathi-language documents in producing legality within the Bombay Inam Commission (IC) of nineteenth-century western India. As a special tribunal established by administrative fiat for the adjudication of titles to land, the Commission wielded the full powers of a civil court, from the interrogation of witnesses and collection of evidence to the issuance of legal judgments. I argue that its reproduction, registration, and validation of documents together constituted a procedure of legal truth-making. In the first section, I explore how the Commission worked to extract documents from material cycles of disuse, decay, disposal, and reuse and insert them into new processes of transportation, storage, and security. Once a document was secured, as I explore in the second section, its reproduction into an individual's case file, and corroboration against existing registers and village- and district-level accounts, determined whether or not it might become evidence in support of a claim to property. But a document's status as a potential source of legality also depended on whether or not it was “genuine,” which could only be evaluated with reference to its conventional style (chal). The third section examines the working practices of Sayyid Usman, a Persian scribe (munshi) in the Southern Maratha branch office, and follows his investigation of a confounding date on a document (sanad) of the Mysore ruler Tipu Sultan. The Commission's investigation revealed that it was impossible to verify the conventionality—much less the authenticity—of a document without recourse to multiple networks of imperfect knowledge and experience tied to overlapping regimes of writing.

Managing Documents in the Bombay IC

The Bombay IC was inspired by the discovery that a disproportionate share of the land revenue of southern Maharashtra and northern Karnataka was alienated through tax-free grants known as inʿam. Land revenue had been an animating concern of the Company government since its acquisition of the diwani, or right to collect taxes on agricultural lands, in Bengal, Bihar, and Orissa in 1765. Taxation not only added to the Company's financial resources, but also allowed it to reaffirm its territorial sovereignty and extend its authority over the socioeconomic activities of its new Indian subjects, principally through the establishment of a legal regime of property.¹⁶ The Company's new focus on governing property helped to shift its relationship to Indian-language documentation. As Nandini Chatterjee and Leonard Hodges have argued in their essay for this issue, European trading corporations had long sought out grants from Indian rulers within a “transactional and performative field” of competitive state-formation.¹⁷ But as Company revenue collectors and surveyors began to investigate the “origins” of Indian proprietary rights, they increasingly assumed the authority to determine which documentation met the standards of legal fact. In the western Deccan, these early investigations into the documentary foundations of rights to property culminated in the Bombay IC.

Initial inquiries in the mid-1840s led to the Bombay IC's formal establishment by the Governor General's Act No. XI of 1852 and its extension across the Bombay Presidency.¹⁸ Beyond circulating general notices for landholders to present their titles, the Bombay IC empowered local officials to seize any written materials that appeared to be “public” records and therefore potential evidence in its investigations. Their efforts resulted in the investigation and adjudication of numerous titles, the expropriation of hundreds of thousands of rupees in revenue, and the formation of a new archive of landed right and local administration.¹⁹ In a seminal article, Prachi Deshpande sheds light on the Bombay IC as “the story of the disciplining of Modi documents and their scribes.”²⁰ She shows that its prosecution of bribery, concealment, and forgery worked to distinguish “genuine” from “fraudulent” claims, which, in turn, it supported by sorting “useful” from “useless” documentation. Drawing on new materials, including case files containing the testimonies (kaifiyats) and copies of the supporting Marathi and Persian documentation of individual inʿam claimants as well as internal Marathi correspondence, this essay's argument is different. It suggests that the Bombay IC’s animating concern was to secure the “passage” from written document to individual claim to legal judgment; that is to say, it was concerned with how to make documents legal.²¹ The road from “papereality” to legal truth was littered with material barriers to interpretation.

Reports of IC staff who coordinated with district revenue collectors (mamlatdar), sub-district collectors, and their clerks and peons to search private homes for records reveal patterns of material decay and reuse. Sub-assistant Commissioner Rangrao Bhimaji on November 24, 1857 reported that over the course of 12 days at six stations in the Dharwar collectorate, he had gathered 230 bodies of material through “desultory conversations” as well as surprise visits backed up by armed guards. Generally, records were “carefully locked up in large boxes or thrown in corners, nitches or the upper part of a House.” He was surprised at the “high state of preservation” of accounts in the house of a record-keeper in Ranebennur that been stuffed “in a corner of the house, in a bamboo basket…just above the fire place and the basket having been first used for keeping chilies.” Another house disclosed “bamboo baskets covered with leather containing old duftures, but as no care at all was taken for their preservation from the attacks of insects, they were one heterogeneous mass of rubbish, dirt, and minute pieces of paper.” Only 17 daftars were not “rendered entirely useless by the attacks of white ants.”²² Conversely, suspicions of tampering were raised when fifty or sixty bundles (rumal) of accounts on the shelves of a former administrator for Gujarat and the Konkan, now based in Pune's Shaniwar Peth, were “quite free from dust or cobwebs, and the knots on the ends of the napkins and on the bits of twine round the several bundles of accounts looking as if tied that morning.” Additional rumals were discovered in “a recess in the wall of a room on the second floor, and from a large box in an adjoining apartment” as well as a 15-foot “sleeping apartment.”²³

With their gradual obsolescence, documents became “waste paper” subject to new cycles of production and consumption. In the residence of the former Peshwa's head record-keeper Chintopant Deshmukh, IC officials observed that “the roomals of accounts were all heaped together in a lumber room under the tiles . . . and that some old papers not relating to the Government mahals etc have been mixed with them, the whole heap being apparently regarded or at least treated as waste paper.”²⁴ IC correspondence attests to a healthy trade in “waste paper” among the “grocers and paper makers” of the city of Pune.²⁵ In one instance relayed by Assistant Commissioner J.A. Cowper on November 26, 1852, the mamlatdar's head guard “stopped a lad in the bazar conveying a bundle of what seemed to be old accounts, and enquired what he was doing with them. The reply was that they had just been purchased from a grocer.” The accounts led them to a series of shopkeepers trading in paper, including one Devchand Seth who had purchased 130 seers of accounts for 24 rupees that had either been reused to pack up grocery items or resold to a “maker of fireworks and taboots.” Similar transactions suggest that the market rate was only 2–3 annas for a seer, or about 2 pounds of paper.²⁶ When Cowper's clerk espied a statement of village land and the diary of the mamlatdar of Vijaydurg from 1768 to 1769 in a “snuff shop,” he returned with “instructions to purchase at a higher price than the market rate for waste paper.” Within the small bundle that he purchased were five documents that according to a clerk “employed in the Peishwa's Dufter for 23 years between A.D. 1789 and 1817” contained the handwriting of the aforementioned head record-keeper.²⁷ “Public” documents could comfortably settle into the “private” homes of state officials, but following their deformation into “waste paper,” they might easily proceed into the economic exchanges of the bazaar.

Even if a bundle of documents had not been reused or discarded, it was likely to become an object of concern in relation to processes of transportation, storage, and security. Discovered documents, including any loose papers, were rebundled in cloth, sealed, numbered, loaded into carts, covered with tarpaulins, and finally transported to the record room of an IC branch office.²⁸ In one case, “it took some eight or ten men more than a day” to lower 1,000 bundles from the upper story of a district administrator's house. Their subsequent carriage to Pune required sixty bullocks as well as the manpower of several indentured laborers.²⁹ Fearing that disgruntled government servants would secrete valuable records, Inam commissioners secured them in their findspots with sealed locks; however, in Pune, the commissioners were not the only ones knocking on doors.³⁰ Allegedly in conjunction with the agitations of “disaffected persons,” it was reported in April 1852 that policemen were importuning individuals and even breaking into homes to obstruct the IC’s efforts.³¹ Lock and seal were available to local judicial officials too. When an assistant commissioner called at the home of the widow Mhalsabai Temburnikar, she pleaded that “her duftur room was locked up the key being given in to ‘the Hoozoor’ and that the Nazirs seal (of attachment) was affixed to the door,” as certain papers had been attached by a judge in connection with her suit against a kinsman. The commissioner agreed to leave but first pasted a slip of paper (chithi) on a wooden, padlocked, and purportedly empty box in the outer courtyard until he could return with the court officer. On this occasion, he examined a couple of rumals in the record-room and locked it with his own seal but was alarmed to learn that the chithi had been torn off the box.³² The security measures of colonial record-keeping coped with the sheer facticity of the physical document through the application of inks, seals, stamps, strings, locks, keys, and indeed more papers.

Understanding how clerical labor was mobilized around the implements of record management requires delving into the operations of an IC branch office. Three years after the IC's formal establishment, in May 1855 Assistant Commissioner A.J. Etheridge proposed to “remodel” the Southern Maratha branch at Belgaon. His goal was to impose a clearer “chain of responsibility” focused on strict registration of documents.³³ According to the organization chart that he generated (see Table 1), it was the “especial duty” of Shankar Vyankatesh, assistant to the head scribe, “to keep a correct register of all kyfyuts taken, and to catalogue them by villages, a book for each talook being kept for this purpose, to be examined and reported upon by the Chitness every month, for the information of the Assistant Inam Commissioner.”³⁴ Etheridge believed that if kaifiyats had been systematically registered from the beginning, clerks would neither waste time in counting them nor be able to escape punishment for any errors or omissions. Clerks in the Marathi department also had to keep “inward and outward registers of letters” sent to the department head; special “barnishee carcoons” looked after “general letter books and registers of papers relating to claims of heirship”; and “a diary of work performed per day by each carcoon” was maintained.³⁵ Added to these several registers was the labor of arranging, cataloguing, and impressing 970 rumals of documents with a numerical stamp and a circular IC stamp at an estimated cost of 1,962 rupees.³⁶ Indeed managing documents was a costly business. According to a March 1859 monthly estimate, the branch office's basic supplies included several different kinds of paper; lamp-oil, sand, gum, cotton for cord, ball thread, wax cloth, needles, salt for white ants, leather bags, “mats for the carcoons to sit on,” charcoal for drying damp papers, “Marathee ink,” vinegar for English ink, turpentine and lamp black for stamping ink, bamboo backs for files, and of course, sealing wax.³⁷

Table 1. Names, Titles, and Salaries of Clerical Staff of Southern Maratha IC Branch Office

Source. PA, English Records, list 1, rumal 39, file 2A, no. 442, ff. 78–81. Not listed for reasons of space are seven peons (pattawala) who were employed to guard the branch office day and night and four more who were to attend on the assistant commissioner. Of the seven, three men, including the head peon (naik), were to watch the office during its working hours from 1000 to 1700 hours, when they would be relieved by others for the night watch. In addition, they were responsible for sweeping and cleaning the office and delivering any messages.

By moving documents from the market economy of the urban bazaar to the inscriptional economy of the government office, the IC aimed to verify their “authenticity.” In his scheme for alphabetic stamping of the records held at the Southern Maratha branch, sub-assistant commissioner Welsh enumerated two measures:

1st The corroboration of accounts collected from individuals, by more authentic records

2nd The prevention of the danger of a future fabrication of “ancient” accounts.

Each of the above measures involves a thorough arrangement of the records…and without the systematic arrangement of the records I fear there can be no corroboration of the collection Duftur. At present both authentic and unauthentic records run great risk of becoming hopelessly confounded; and in view, too, to the prevention of the danger of a future fabrication of “ancient” accounts, the systematic arrangement of the records is equally indispensable, for after the lapse of a few years, it is not an uncommon thing for accounts even of catalogued dufturs to disappear or to look most suspiciously modern if not properly stamped.³⁸

Welsh's invocation of the specter of fabrication reminds us of the Commission's disciplinary function. But prevention of scribal forgery and concealment was subsidiary to directing scribal labor toward “corroboration,” or legal validation, which entailed rendering the content and form of documents more “authentic” and therefore suitable for the adjudication of claims. While the IC would repeatedly attempt to displace its failures onto the figure of the “native informant,” I will show in the remainder of this essay that its proceedings as a civil court hinged more on its approach to Indian-language documentation as a problematic source of legal truth.

Files, Registers, and the Making of Legality

Even as the IC worked to monopolize any and all government records, it used the records in its possession to rule on the validity of individual claims to property. In doing so, it inherited basic assumptions about the fallibility of Indian-language documentation's representation of the legal terms of inʿam. In the pre-colonial western Deccan, the Perso-Arabic term inʿam came to designate a broad category of tax-free land tenure granted by a sovereign ruler. In theory, an inʿam was permanent and hereditable; however, in practice, it was partible and fully alienable by sale, inheritance, or government sequestration, a fact that British officials seized upon in their early investigations of Deccan land tenures. Eluded in his search for proof of “original” rights to property, W.H. Sykes, the statistical reporter to the Bombay government, found and translated documents issued under the collective authority of village headmen and assemblies of notables. If the state had “the lordship of the soil,” he doubted, “could it possibly have sanctioned such absolute and unrestrained disposal of it…It is idle, therefore, to talk of hereditary occupancy.”³⁹ Reports of the extensive alienability of inʿam seemed to confirm suspicions about the legality of the core terms of inʿam title deeds. The Madras governor Thomas Munro in a February 1, 1822 minute asserted, “The terms employed in such documents—‘for ever,’—‘from generation to generation,’—or in Hindu grants—‘while the sun and moon endure,’—are mere forms of expression, and are never supposed, either by the donor or the receiver, to convey the durability which they imply.” Unable to repose confidence in the documents presented by claimants to inʿam, the IC aimed to reconstruct the history of ownership of any given inʿam claim based on evidence extracted principally from the records of the so-called Peshwa's Daftar at Pune.

Acquired by the East India Company during the Third Anglo–Maratha War (1817–18), the Peshwa's Daftar principally referred to the daily registers (rozkirds) of the Maratha Empire's ministerial government at Pune, which contain entries for all manner of state orders, accounts, and correspondence dating to the first quarter of the eighteenth century.⁴⁰ The Pune registers have a complex internal logic that requires further study; for our purposes, it should be noted that they reflect the systemization of a longer tradition of legal attestation, registration, and record-keeping in the early modern western Deccan.⁴¹ While attestation of written conveyances for interpersonal transactions tended to be left in the hands of village- and district-level record-keepers and judicial assemblies, there was a very robust tradition of state registration of orders, including royal orders and grants of land.⁴² In his study of Sultanate-era documentation, the historian G.H. Khare emphasized that injunctions to copying, sometimes accompanied by a warning against demanding a new order every year, proliferated in orders pertaining to hereditable property.⁴³ The formulaic structure of multilingual, multi-handed registration notes, often featuring the term “bar,” on the versos of Sultanate farmans also indicates the emergence of a state registration procedure and terminology, one that persisted after the Maratha conquest of the western Deccan. The survival of the Pune registers allowed the IC to corroborate or undermine the documentary evidence provided by inʿam claimants. “When its registry had been duly effected,” as Commissioner Alfred Thomas Etheridge put it in his 1871 report, “the sunnud or deed of grant might be looked upon as a valid instrument.” As “the pivot upon which the whole machinery worked,” the Pune registers were the bedrock of the Commission's working legal archive.⁴⁴

Prior to consultation of registers or any other corroborating evidence, the legal validation process began with the assembly of the claimant's case file. Organized by village, sub-district, and district, each file contained the following components: a heading with a serial number, the IC stamp, and the last year in which the land claimed appeared in revenue accounts; a table containing the name and patronymic of the inʿam claimant, the amount of land claimed, and often, the type of grant under which the claim was made; the kaifiyat of the claimant; copies of the documents presented in support of the claim; and a family tree. The kaifiyat consisted of answers to sixteen standard questions to be “put viva voce” to the claimant, mainly concerning the past and present condition of the lands in question and the terms by which it was granted.⁴⁵ Most importantly, the sixteenth question instructed the claimant to “produce whatever documents you may have in proof of what you have stated in the above 15 answers; and mention if there are in the records of government any papers which you have wish to have examined.”⁴⁶ The compilation of the kaifiyat, and indeed of the entire case file, was the work of a complex and multilingual village bureaucracy. In this sense, they are comparable to the pattas of the village katcheris of the Dutch Vereneedigne Oostindische Compagnie in Sri Lanka and Bengal that Alicia Schrikker and Byapti Sur have analyzed for this issue.⁴⁷ Sitting before the mamlatdar and at least one copyist, village proprietors were to declare any supplementary information that might explain discrepancies between their testimony and corroborating documents. For example, the Brahman mendicant Bhau Dixit Valvade of Karhad submitted a decree (adnyapatra) of the nobleman Jagjivan Parshuram Pratinidhi to support his claim to an endowment of 10 bighas of land. But, as he explained, “Pratinidhi originally informed the village of an inʿam of ten bighas of land, but eleven bighas of land fell into our lap. And we do not know how one more bigha of land than [what is listed in] the document was acquired. And [though] the decree is for ten bighas, we have enjoyed eleven bighas of land from the beginning.”⁴⁸ Hence the correspondence or lack thereof between claimants’ testimonies and the written contents of their documents and associated records was key to the adjudication of rights to property in the IC.

Copies of supporting documents were appended with notes stating the names of the copyist and the “comparer,” who compared the copies to the originals, as well as any errors or ambiguities in the transcriptions. IC writers used notes to reveal the difficulties of reproducing legal documents and shield themselves from accusations of misconduct. But these notes were also material to the process of legal validation. For example, in a case involving copies of two documents issued by Peshwa Balaji Bajirao to Govind Khanderao Chitnis in 1745–46 for 15 bighas in the village Wanwadi, Commissioner J.A. Cowper was led by unnamed circumstances “to suspect the genuineness of the documents of which these purport to be exact copies.” Upon calling for the originals, he realized that the name of the village was “introduced by the erasure of letters previously written, the erasure being a most palpable one, and being so clumsily made that in each case the paper of the sunnud has been partially destroyed.”⁴⁹ The copyist and comparer, Rango Konher and Vitthal Vyankatesh, had not called attention to the erasure; rather, they had simply noted the striking out of a matra line above the same word. Whereas the copyist stated that he had shown the original to the comparer, the latter claimed that he had not seen the original, inspected the copy on its own, and was only “told” of the stricken matra line. Cowper assessed this to be “an imperfect method of proving copies” and a “culpable act of omission,” although not a “substantive offence.”⁵⁰ In addition, he decided that the name “Ghorpadi” had been altered to “Wanwadi,” and, notwithstanding this alteration, that the documents “appeared to be genuine as their registry is forthcoming in the Paishwa's state diaries.” Neither the same diaries nor “authentic land registries of the village” had any record of land granted to Chitnis in Wanwadi.⁵¹ In sum, proving a copy to be “true,” if done perfectly, might aid in the authentication of original documentation, but ultimately, legal validity rested more on the concurrence between document and register.

Following the assembly of the case file, the consultation of registers and relevant government fiscal records was the step in the IC dispositif that drove an individual claim to property toward judgment of its legal validity. A January 1860 report on Vitthalrao Honap, deshmukh of Kadewalit and Shrigonda, suggests the register's power to invalidate, even if a claim was based on seemingly authentic documents. The report indicated that due to the transfer of these districts from the Peshwa to the Nizam in 1760, it was not possible to compare the information contained in the seventeen documents offered by the claimant with village records. It elaborated, “The registry of those documents…which are alleged to have been issued by the Peishwa, is not discoverable in the Hoozoor diary forthcoming in the office. Some of them, are letters alleged to have been issued from the Chitnees Departments, and are not endorsed with the word “Bar” (registered). There is, however no reason to doubt the authenticity of the documents recorded above, which have been produced in original.” Regardless of the authenticity of the documents supposedly issued by the Peshwa, the lack of corroborating register entries or registration marks made it impossible to view the claimant's lands as legitimate “wuttun holdings.”⁵² Sometimes former government records yielded mixed evidence. The claim of one Kusappa to a deshmukh watan in three villages in the district Chimmalgi was not reflected in the deductions section of a revenue estimate from 1767–78; however, lower-level “mahal accounts” did make the relevant deductions in several years.⁵³ In addition, an individual-wise abstract of the Peshwa's diaries, known as ghadni, contained for the year 1774/5 “a registry of orders” issued to the claimant's forebear Nago Ram.⁵⁴ In conjunction with information suggesting the holding's resumption in 1808–9, this patchy history of ownership was not sufficient to substantiate the claimant's title.

Hitherto I have suggested that the passage from written document to law encompassed an assemblage of heterogeneous elements, ranging from “waste paper” to the case file to the state register. If a document's “authenticity” allowed an individual claim to move forward, only the conformity of document and register could secure judgment. But what made a document seem authentic? Beyond the appearance of deliberate alteration, case reports only occasionally remarked on the form and content of documents. “The general appearance” of a 1703–4 mahzar submitted by the deshmukh Kusappa was “against its genuineness. The style of writing and composition does not appear to be ancient as the date of the paper would seem to indicate and the several seals on it look decidedly suspicious.”⁵⁵ Such remarks on “style of writing” along with seals, dates, paper quality, and other features betray the extent to which formal analysis of documents was a consistent preoccupation in the Commission's legal reasoning. Indeed the perceived conventionality and, therefore, “genuineness” of any discovered written artefact was integral to its transformation into a legal document. In the following section, we will explore how IC clerks verified the style of documents, and how verification of style across different languages and regimes of writing traversed distinct, but overlapping networks of expertise.

The Story of a Sanad

Sayyid Usman, son of Sayyid Jaʿfar, was the lone Persian munshi in the Southern Maratha branch office of the IC. Apart from his patronymic, Sayyid Usman's origins remain obscure; however, we know that he worked for the IC from 1846 until at least the end of 1860. By his own admission, he had taken only 15–20 days of leave in over a decade of service before receiving two and a half months for personal reasons in 1859.⁵⁶ He was “responsible, conjointly with the Carcoon employed, for the correct translation of all Persian Sunnuds and other documents.”⁵⁷ Although the Commission may have employed several Persian writers across the Presidency, the standard procedure was that any document with Persian writing “should be forwarded to the Hoozoor with the Kyfeeut for the purpose of having the Persian writing copied.”⁵⁸ Sayyid Usman reported directly to the assistant Inam commissioner, but his income was comparable to any writer hired on a permanent basis and supervised by a head clerk (huzur karkun), fluctuating between 12, 15, and 16 rupees per month.⁵⁹ Based on his surviving work product, it is probable that he knew some Marathi and Urdu in addition to Persian.

Sayyid Usman deployed a cautious and transparent palaeographic strategy in his transcriptions of Persian documents. Rather than offer speculative readings, he added brief annotations (sharh) in red ink wherever he was unable to produce reliable transcriptions: “One letter has gone (ek huruf rafta ast)”; “In this place, the letters have gone (dar in-ja huruf rafta and)”; “One letter is not becoming clear (ek huruf wazih nashawad).”⁶⁰ Correspondence with IC officials indicates that he was accountable for any perceived gaps, errors, or variations in the Persian documents that crossed his desk. Commissioner M.F. Gordon on September 18, 1855 requested that he produce a detailed reported outlining whether the year 1172 in a Persian document was a Hijri, Fasli, or regnal date, and why the name of the sender was not included. More generally, he asked whether or not the document was written according to the conventional style (chali-pramane).⁶¹ Sayyid Usman replied that 1172 was a Hijri year, and that the sender, based on the writing in the seal, was the nawab of Sawanur. In response to the thornier final question of style, he admitted that he could not make sense of the presence of a smaller seal bearing the later date 1176, but still affirmed that its contents and the discipline of its writing (lihinyachi shista) were acceptable.⁶² Because the orthopraxis of writing had become central to colonial legal reproduction and validation, minor discrepancies in content or form could raise major concerns within the colonial bureaucratic apparatus. For those who staffed this apparatus, the devil was in the details.

The IC’s inquiry in the spring of 1856 into the date of a Persian document (sanad) issued by the Mysore ruler Tipu Sultan clarifies the stakes of the question of “correct” or conventional writing. The occasion of the inquiry was an exchange between Sayyid Usman and Commissioner Gordon in which the scribe had stated that the number “5121” appearing in the sanad should be read as “1215,” which would correspond to the year 1787–88 CE. When Gordon pressed him, citing another document containing the number “5112,” he insisted that both ought to be rendered as “1215.” Gordon demanded that he commit himself to a written statement, in which he elaborated that “according to the practice of the Muhammad era, there is a practice of reading and writing in reverse.” He stated that when asked about the figure “5121,” he had simply reported in line with this practice. As to the second number “5112,” he pleaded, “That year according to the number must be fixed at 1215. Otherwise, in that— meaning, in that year—no other meaning is forthcoming.”⁶³ Hard pressed to adduce any other reason for the brute facts on the paper staring back at him, he resorted to an account of a recognizable convention of reading and writing numbers in Persian documents “in reverse.” That is, numbers in Persian documents are customarily written from left to right, even though Persian employs the right-to-left Arabic script.

The immediate problem with Sayyid Usman's explanation was that when applied to these two different numbers, it would yield two different dates, a result that he could not explain to his supervisor's satisfaction. Gordon on April 21, 1856 surmised that the explanation “if not chargeable with an intention to mislead, has evinced a degree of ignorance which is very discreditable to him after a service of so many years in his present situation.”⁶⁴ The officials were so dismayed that they considered re-assessing his qualifications and docking his pay at his annual performance review.⁶⁵ Before coming to a decision, they instructed him to clarify his position, either based on his own understanding or with reference to an almanac (jantri) or other book (kitab) that might contain evidence of this practice.⁶⁶ Unfortunately, as Sayyid Usman noted in his reply, there was no such text to which he could refer; instead, he attributed his reading of the “5121” as the year 1215 to what people had told him and the existence of additional documents with this number.⁶⁷ Refusing to accept this response, the commissioners warned that his salary would be withheld if he could not come up with a more persuasive account of the dates of the documents by the first of the month.

In the absence of complementary examples or reference texts, Sayyid Usman struggled to contextualize the writing of the sanad of Tipu Sultan. On May 12, he took the initiative to “go and bring evidence from wherever it might be found. If it pleases the government, and I am so ordered, I will go to Srirangapatna, Mysore, etc. and bring back evidence.”⁶⁸ In the meantime, his superiors wrote to the collector of Belgaon district, requesting that he seek an explanation from “a Persian munshi who might have information.”⁶⁹ They also asked the judge of Dharwar district to consult with a “maulvi,” who, based on subsequent communication, we understand to be the mufti, or Muslim legal interpreter, attached to the district court. When the mufti could not produce a satisfactory response, the case was referred to the qazis of Dharwar.⁷⁰ This appeal to Muslim jurists to explicate the peculiarities of a Persian sanad reflects the long-standing role of qazis and muftis in authenticating legal documents.⁷¹ Under colonial rule, as Elizabeth Lhost has explored, the qazis and muftis continued to play this role, even as the Company attempted to restrict their activities to adjudicating the affairs of Muslim populations, especially those pertaining to areas of Islamic law deemed to be “religious.”⁷² Despite efforts to monitor and standardize the qualifications of those appointed to qazi offices, there was significant variation in the credentials and abilities of individual qazis stationed across the Bombay Presidency. It was reported 7 years before the affair of Tipu Sultan's sanad that only one of a highly diverse group of fifty-eight qazis working in Dharwar district had passed the official government exam.⁷³

The responses that the Commission received to its queries about the sanad reveal the extent to which no individual, however experienced or well-trained, could claim absolute expertise in all modes of writing legal documents. Perhaps least helpful was the unnamed group of qazis, who simply replied that “they can give no explanation of it without the sunnuds.” Likewise, the munshi of the district court “could give no solution.” The British judge himself could only speculate that “the peculiarity of the two dates having exactly the same figures” might reflect “some Mahommedan combination of figures meaning some peculiar year.”⁷⁴ Apparently perplexed by the Commission's references to the dates in question as belonging to the “Mahommedan” or “Mahomedee” era, the mufti Ghiyas ud-Din, in a May 23 report penned by his clerk Tulja Khan and bearing his own signature, declared, “That year is not at all Muhammadi because at that time, Muhammad himself did not exist.” He further suggested that there would be a 7-year difference in the conversion of the 1215 date, depending upon whether it belonged to the Hijri and Fasli calendars.⁷⁵ In another report from the same day, he added, “Among the Muslim people, there is no tradition of applying numbers backwards and forwards.”⁷⁶ Across all of these different responses, what was consistent was the difficulty of associating a practice of writing, and a mode of reckoning time, with any particular Islamic state or Muslim community on the basis of a single document.

Against the assumption of a uniform Islamic temporality, the investigation revealed several competing explanations of Tipu Sultan's “Muhammadi” era. In response to the Commission's queries, the Persian munshi employed by the collector of Belgaon advised that whereas the Hijri era was reckoned from the year of Muhammad's flight from Mecca to Medina, Tipu calculated the “Muhammadi year” of his documents from the year of Muhammad's assumption of the title of Prophet, which he stated to be a difference of 14 years. Based on information gleaned from his father, who had worked for the sub-collector of Hubli in the 1830s, he also noted that the documents included a Shaka date in “Hindvi” and confirmed that numbers were written from right to left, contrary to prevailing convention.⁷⁷ In the meantime, our own munshi Sayyid Usman wrote to a Mysore-based acquaintance, Sayyid Haider Mutaʿalli, who stated that “Muhammadi” referred to a solar calendar beginning in the year of Muhammad's birth, and commonly used in the revenue department of Tipu Sultan. In Persian documents, he explained, “it was customary to write the year inversely, the units being placed on the left . . . but the figures would not be inverted indiscriminately, as asserted by the Moonshee [Sayyid Usman] in his deposition.” The Marathi takidpatras issued by Tipu's officers included both “Muhammadi” and “Samvutsur” years, the latter referring to the 60-year cycles (samvatsar) of Jupiter.⁷⁸ The “concordance” of these calendars in successive batches of documentation ultimately enabled the Commission to confirm the date of the sanad in question.⁷⁹

The twists and turns of the Commission's inquiry into the date of a single document suggest both continuities and ruptures in the creative adaptation of Persianate cultures of writing, including under the innovative Mysore ruler Tipu Sultan. In the cultural politics of today's Karnataka, his supposed choice of Persian as the official “language of administration” in 1792 is sometimes branded as “anti-Hindu” or “anti-Kannada” in line with colonial stereotypes about the supposed religious zeal of “Tipu the Tiger.”⁸⁰ Recent scholarship on Tipu's administration and library, including his own writings, have yielded a more complex image of the ruler. It seems incontestable that he deployed new and forceful Islamicate concepts of sovereignty—most notably, sarkar-i khudadadi (God-given government)—yet he also sponsored translations from Kannada, English, French, and other languages, and patronized Hindu shrines.⁸¹ In line with his initial refusal to include the name of the Mughal emperor on his coinage, he also broke with existing Persianate practices of state-based computational writing by introducing new weights, measures, and of course his own Islamic calendar, known as the “Mauludi calendar” after the spiritual birth (maulud) of the Prophet. It appears that this calendar was based on the Shaka era, with entirely new names for its months drawing on the divinity of the Prophet, the prestige of Tipu's dynasty, and Abjad numerology.⁸² Finally, the survival of Marathi takidpatras reflects an administration in which Persian, Marathi, and Kannada all played significant roles. According to a 1796 hukmnama, four Marathi writers and two Kannada writers in the treasury were to keep accounts in these languages.⁸³ On the other hand, Tipu encouraged a shift away from south Indian palm-leaf record-keeping to paper, perhaps indicating an attempt to reduce the authority of local Kannada-speaking accountants.⁸⁴ Further research is required on the precise functions of different languages and forms of writing in the Mysore state, but it is clear that there was no single “language of administration.”

For his work in cracking the date of the sanad, the intrepid munshi Sayyid Usman received “a certificate on the part of government, acknowledging that he has rendered important public service.” In addition, his pay was bumped from 12 to 16 rupees per month.⁸⁵ Yet his was not a straightforward story of a “native” expert vindicated. Solving the riddle of the sanad entailed passage between several nodes in the apparatus of colonial governance: Sayyid Usman himself and his supervisors in the Commission; a British district judge, the legal interpreter of his court, and one or more qazis in his district; a tax-collector, his munshi, and by proxy, his father, former assistant to a sub-collector; and finally, an acquaintance in Mysore whose professional experience eludes us. Moreover, the various elements under consideration—the appearance of dates in different sanads, the mode of reading and writing dates within the Muslim community, the calendars used under Tipu Sultan—were heterogeneous to each other. They introduced new premises at different scales of generality even as they worked to move the inquiry forward. Each of these premises further extended a chain of reasoning about the conventionality of a specific document, one of many that undergirded colonial truth-making about the legality of Indian-language documentation. Far from yielding a unitary agent of colonial knowledge-production, the story of the sanad discloses the many material disjunctures and epistemic displacements engendered by colonial bureaucracy in action.

Conclusion

The events of the 1857 rebellion cast a pall on the deliberations of the Bombay IC. Even if there was no clear evidence that rebels had been motivated by the Commission's operations, the widespread disaffection of rural magnates and land-owning peasants challenged the presumption that the expropriation of titles to alienated revenue would ultimately confer a benefit on Indian subjects by enshrining a legal right to private property. Moreover, the Commission's painstaking inquiries—with their passage from document to file to register and back again—had created a slow-moving bureaucratic Goliath. The Summary Settlement Acts of 1863 superseded the adjudication of further inʿam claims in favor of a more efficient compromise whereby title-holders would pay a quit-rent in exchange for retaining the remainder of the revenue. The conclusion of the Bombay IC signalled the end of the colonial state's long engagement with the historical conventions of Indian-language documentation. Part of this disengagement must be attributed to the shift in colonial administration toward English, and away from Persian and regional languages. But it was also a matter of changing methodologies and instruments for making documentation answerable to the requirements of legal truth-making. Whereas the IC’s proceedings turned on difficult and sometimes irresolvable questions about the materiality of documents, about their concordance with the remnants of a pre-colonial ecology of documentation, and about their conventional style, new procedures for registration and validation rendered such questions moot.

Undoubtedly many of the record-keeping procedures that organized the routine work of Commission writers foreshadow the British colonial state's drive toward legal and bureaucratic rationalization in the aftermath of the 1857 rebellion. The persistence of instruments like the barnishi register indicates the extent of the colonial state's investment in writing as a means of administrative efficiency and disciplinary power. But if the IC was a laboratory for testing more advanced techniques of colonial state formation, its experimental design was unable to control for all variables impacting the documents that it examined. Exposure to rain, insects, debris, and the cycles of the marketplace left their effects on the materiality of the document, whose appearance might then work against its “authenticity.” Sheer human error and malfeasance—even so slight as a scribe's failure to recognize the alteration of a single pen-stroke—further contributed to the fallibility of the documentary record. It was at the limits of the colonial state's epistemic control that it was forced to more directly engage with the written world of pre-colonial documentary convention and procedure. In the absence of a clear point of origination of property rights, legal reasoning about inʿam claims became entangled in a chain of references embedded in multiple kinds of documentation, both within and outside the confines of the individual case file. Try as it might to attribute uncertainties about the finer points of writing to the incompetence of Indian intermediaries, considerations of conventional practice, community norms, and the layered protocols of successive state regimes all factored into the process of making a document legible and indeed legal.

The passage from legal claim to documentary proof to proprietary right included a critical detour through the Pune registers and the multitudes of village and district revenue accounts that the IC accumulated. Hence the story of the IC is also partly the story of the formation of a colonial archive. It was impossible to reconstruct histories of ownership purely on the basis of documents presented by inʿam claimants, prompting the introduction of new facts from information extracted from registers. The document's appearance in the record of the Maratha state—even through a mere registration mark—imbued it with a different evidentiary reality. Through material interventions ranging from bundling, transportation, and storage to a record room to the discarding of “useless” papers to the stamping of “useful” records, the colonial state fashioned a legal archive out of the remnants of a pre-colonial bureaucracy. Long after the conclusion of the IC’s work, its instrumental limning of the complex architecture of the “Peshwa's daftar” and construction of a new edifice out of the local records that it confiscated shaped the parameters of historical scholarship for generations. Initially questions of land tenure and property rights—the very ones that animated the IC—and then ones of social relations and economic organization defined scholarly understanding of imperial state-formation in the Deccan across Sultanate, Maratha, and British rule. But our access to the historical realities of these paper empires, as I have attempted to show in this article, can only be made richer by a closer study of the forms of writing and modes of document management and validation that they shared, over and against the occlusions of colonial power.

Precisely because IC proceedings did not occur in an insular institutional environment, they are not entirely legible as a transitional stage in the maturation of the colonial state or the colonial archive. Rather, I have suggested that the boundaries of state administration of documents were porous, and not merely because of the scribal malfeasance that so preoccupied the IC. The discovery of numerous accounts in the residences of the Peshwa's record-keepers, and in the stacks of “waste paper” sold in the grocery shops of Pune city, betrays a more extensive chain of linkages among state, household, and bazaar. Further research into the spatial loci and material practices of record-keeping prior to the colonial state's imposition of a program of legal rationalization and archival management in the second half of the nineteenth century is much needed. Only then will it be possible to understand how it is that a seemingly straightforward inquiry into a date on a Persian sanad produced a series of epistemic displacements among the relevant, but limited knowledge of a munshi, a qazi, a tax-collector, and even a casual acquaintance. Neither the official gaze nor the “native” expert alone can piece together the puzzle of colonial knowledge. Rather, it becomes incumbent on us to investigate how knowledge was materially encoded and decoded across interlocking networks of legal, religious, and political authority.