Administration, Extraction and Outsourcing

Despite all its turbulence, Malwa was known to be an exceptionally fertile and prosperous province, remarkably free of the disastrous famines that periodically afflicted other parts of the Mughal empire.Footnote ¹¹ The Ā’īn described it as a temperate country with parts as pretty as fairyland, endowed with a very long sacred and secular history, including that of an alchemic paras stone having once been discovered there, although unfortunately lost due to human frailty. Ujjain, the provincial capital, abounded in Hindu places of worship and produced contemporary miracles for Abul Fazl to report. On his way to the Deccan March 1599 CE (month Farwardin, regnal year 46), Abul Fazl apparently witnessed the Shipra River, on which Ujjain was situated, flowing with streams of milk. Not all was necessarily milk and honey in this province, however, since the Ā’īn also noted that this was a highly militarised populace – apparently there was nobody, including the peasants and grain dealers, who were without weapons.Footnote ¹² Getting taxes out of such a populace, then, was a venture that was inseparable from military operations.

For purposes of administration and taxation, the sūba of Malwa was divided into twelve sarkārs, of which Mandu was one.Footnote ¹³ Unlike in several of the other sarkārs, the caste of the zamīndārs of Mandu was not noted, being too varied for enumeration.Footnote ¹⁴ The total revenues collected from the province rose rapidly from the time of its conquest by the Mughals until the mid-seventeenth century, then dipping, but not precipitously, during the battle of succession leading to Aurangzeb’s coronation and during the Deccan wars. But our aim here is not to assess the province’s contribution to a gross national (or imperial) product, but to show how the intertwining of regulation and outsourcing, related to the main military-cum-administrative spine of the regime, proved fecund with entrepreneurial possibilities for our protagonists.

In order to understand the dynamics of this family’s multiple entrepreneurial ventures, we should begin with a quick review of what we know about the dispensation of tax-related institutions, offices and paperwork in Mughal India. Experts of Mughal history may wish to skip this section, although a quick skim would help clarify how I see the structural features described relating to the specific dynamics of the family’s fortunes.

The Mughals had a highly systematic view of the rural order that they presided over, and one that was relevant principally because it financed the imperial military–administrative hierarchy. The nobles, who in the official view were mere officers of state, were designated mansabdārs. Abul Fazl’s Ā’īn had no section on ‘nobility’ as such; mansabdārs appeared in the ‘book’ or section on the army.Footnote ¹⁵ In a clear expression of the military ethos of the Mughal regime, all mansabdārs, no matter whether they were provincial governors or heads of treasury, were designated military officers; this was only partly fiction. The mansabdārs’s pay was graded accorded to their rank, and Abul Fazl provided a rather complex table outlining what was due to mansabdārs on forty-nine grades, ranging from the middling 600 to the princely 10,000.Footnote ¹⁶ The implications of the table were fervently debated by historians for several decades from the beginning of the twentieth century, until poring over pay scales in Persian manuscripts went out of academic fashion.

As we all know, however, pay scales, then as now, are highly contentious texts. They declare and record entitlements, but frame these with so many caveats and conditions that it takes an expert to decode them. And of course, they change, or their interpretation changes, fairly frequently. Already in Abul Fazl’s table, there wasn’t just one salary figure corresponding to one rank; for each rank several species of animals, each of multiple grades, were listed, together with the corresponding costs of maintenance of each such animal. Each rank also included three salary points – first, second and third class. The mansabdārī system of ranks was introduced in that form and under that name by Emperor Akbar; the idea was that commanders recruit and supply an agreed number of equipped soldiers and the equipment, including horses and other animals. At its simplest, the mansabdār’s rank corresponded to the number of equipped horsemen he was obliged to supply, so a mansabdār with the rank of 500 was obliged to supply 500 horsemen on demand.

However, already in Akbar’s reign and by the time of the completion of the Ā’īn-i Akbarī (1592), it was recognised that not all mansabdārs could match the number of their recruits to their rank. The Ā’īn provided three columns for every mansab rank, those who could supply the full number of horsemen being of the of the first class of that rank, those who could bring at least half that number or more were second class, and others, third class.Footnote ¹⁷ This table in the Ā’īn has been the source of a fair amount of confusion and debate, because it does not help us trace the emergence of the ‘two-part’ mansab rank, which most scholars agree had already taken place in Akbar’s reign, and possibly, before the Ā’īn was completed. In the full-blown Mughal mansabdārī system, the difference between the mansabdār’s rank and the number of troopers he supplied (and was paid for) was formalised into a two-part rank – the zāt or personal rank, and sawār, or the number of troopers to be supplied.Footnote ¹⁸ In the two subsequent reigns – Shah Jahan’s and Aurangzeb’s – a further complication was introduced in that the equipage of the horsemen were taken into account – with different rates of pay being disbursed depending on whether the horsemen in question were equipped with single horses, or with two or three (very useful for forced marches).

This was a system of salary calculation comprised of multiple caveats which can be summarised as follows. A mansabdār’s rank was split into two parts, the personal rank or ‘zāt’, which entailed a certain allowance associated with the rank itself, regardless of any specific services, and the other part consisting of the number of horsemen or ‘sawār’ they were supposed to equip, maintain and provide. The sawār rank was associated with a certain rate of pay per horseman. Since a mansabdār’s contingent might include some one-horse soldiers and some two- or three-horse soldiers, his sawār pay would correspond to a formula that took into account the different rates payable for these differently equipped soldiers.

For all such intricate formulae for calculating salaries, most of the mansabdārī pay was in fact not disbursed in cash, but allocated as the right to take the taxes of designated regions. These allocations were known as jāgīrs, and the right-holder, the jāgīrdār, was the mansabdār seen from the vantage point of the tax office. Barring khālṣa lands, whose taxes were reserved for the emperor, the rest of the empire was parcelled out in jāgīrs. To make things even more complicated, a mansabdār received jāgīrs scattered all over the empire, which, moreover, where frequently changed – all to prevent them from slipping back to the kings they had once been, or aspiring to sovereignty. The value of the jāgīr naturally depended on the taxes that could be collected from the villages – mauzaʿs – within the jāgir. Some parts of the Mughal empire had been formally measured and assessed for their tax-paying capacity; these parts were called z̤ābtī. In others, a round figure of tribute had been imposed on the clearly unsubjugated landlords. In all cases, the official revenue demand of any unit of taxation was termed jamaʿ (from the Arabic word for collection), later jamaʿdāmī – referring to the collection of copper coin, dām, in which the taxes were paid. The officials assigning jāgīrs simply had to match up the recorded tax-paying capacity of certain regions with the mansabdār’s salary, applying fractions if needed.

While this sounded very transparent on paper, it was already officially recognised in Akbar’s time that the official tax demand was hugely inflated – a sort of imperial flourish than a reliable source of information on how much money the mansabdār-jāgīrdār might actually expect to receive from his jāgīrs. By Emperor Shah Jahan’s reign, jāgīrs were assigned taking a ‘month-ratio’ into account, that is, a jāgīr’s worth was taken to be equal to that fraction of the year, and so expressed as a twelfth. If, on the other hand, shortfalls were significant and unexpected, a takhfīf (from the Arabic for ‘lightening’) was applied to compensate.Footnote ¹⁹ The infamous jāgīrdārī crisis beset the Mughal empire during Aurangzeb’s reign, because as more and more elites were incorporated into the empire with mansabs, there were fewer and fewer fully paying jāgīrs to go around. Collection in the newly conquered regions naturally depended on the extent of subjugation, which in turn depended on the military capacity of the jāgīrdār in question. Imperial surveys, in such a context, could be conventional hyperbole geared to be displays of power, rather than the collation of precise information.Footnote ²⁰ However, the vast stores of documents across household archives, recording the minutiae of transactions, offer a different kind of source base, not just for correcting the imperial statistics, but for understanding how the system really functioned. Tax lay on the seams of the Mughal empire, where warfare met rule-referenced negotiation.

Taxmen in the Countryside

Whatever the dues of a certain mansabdār-jāgīrdār, there had to be an administrative structure in place to actually collect a share of what the peasants in the allocated jāgīr produced. And here is where some confusion begins to enter the picture, because earlier historians produced descriptions of ‘provincial administration’ that offered a picture of a centrally appointed hierarchy, of which the fiscal and administrative wings were, moreover, separately managed. The fiscal side consisted of the provincial diwān, answering directly to the imperial diwān, and managing a staff of tax-recorders and collectors, and a regularly replenished store of paperwork, which was transmitted to the centre. Notably, this highly coherent picture was based primarily on manuals – the great Ā’īn-i Akbarī itself, and also the later, mainly eighteenth-century dastūr al-ʿamals.Footnote ²¹ On the other hand, from the 1960s, historians, who combined their knowledge of manuals with detailed study of administrative documents similar in nature to our archive, began to emphasise the key role of the jāgīrdārs, and the fact that they had to appoint their own tax officials.Footnote ²² The trouble of course is that no manual of governance actually states how this was to be done: who appointed and controlled whom. We are therefore left with a blurred image, consisting of multiple alternatives. We could imagine a centrally appointed tax bureaucracy, always in situ, which would pass on the collected tax revenues to the current jāgīrdār with minimal or no intervention from him; judging even from contemporary accounts, we can be sure that this did not happen. We could also speculate on the possibility of a completely noble household-centred tax collection structure that entered and left regions together with the noble. This too, seems implausible, because we certainly have records of the post of the provincial diwān, including a very well-known account produced by just such a diwān, albeit one who was deprived of access to the diwānī records himself.Footnote ²³ The notion of complete administrative portability also does not match the picture of long-term entrenchment at the very lowest levels, in posts such as chaudhrīs and qānūngōs, as presented by our archive and all comparable collections. The tension, then, would have been somewhere at the middle of the spectrum, around the region of the ʿamalguzār – the equivalent of the still-crucial post of the district magistrate – where overlapping structures wove into each other.

In his classic work, Irfan Habib focussed on an imperial system that extended from the centre down to the villages, with standard administrative personnel whose offices evolved over time. However, Habib combined this picture of uniformity, best realised in the khālṣa or crown lands, with the variability of the jāgirdār’s mobile tax-office. Habib seems to suggests that in non-crown lands (that is, up to four-fifths of the empire), there was no permanent tax administration except the men that each new jāgīrdār brought in, their job titles mirroring offices in the khālṣa, and their work following imperial regulations. The only offices that were indeed local and unchanging (with jāgīr assignments) in all this were those of the chaudhrī and qānūngō,Footnote ²⁴ which places the protagonists of this book at the crux of the Mughal apparatus.

More recently, Faruqui has proposed a more interlaced picture, based on his studies of the households of Mughal princes, who were also the highest-ranking nobles. Faruqui shows examples where the princes, especially Aurangzeb and his sons, incorporated lower-ranking nobles, also holding significant administrative offices, into their household and loyalty network, and conversely, how long-standing members of the princely household were appointed to administrative posts from which they would be in a position to assist the prince in managing and milking their jāgīrs. We have already seen in the Chapter 2 how the petty locally based Sisodia Rajput noble, Amar Singh, was recruited by Dara Shukoh into his household to assist with the collection of taxes when Dhar was transferred from the jāgīr of Murad Baksh into his, in 1657. So Faruqui’s suggestion of an ethnic–confessional divide, such that the only local men in the picture were the village-based Hindu officials (i.e. the chaudhrī and qānūngō),Footnote ²⁵ possibly needs modification. Petty Rajput mansabdārs of local roots could equally be recruited to service the tax-collecting needs of a jāgīr grantee. Overall, however, all these findings point towards the crucial administrative role of the noble’s household, and its interlacing, or indeed co-formation with the Mughal state, which was ‘patrimonial-bureaucratic’ not just at the imperial centre, but also at its lower levels.Footnote ²⁶ The current archive offers us an unusually detailed view into the instantiation of such a system, especially through the crucial function of tax extraction, and from the vantage point of its lowest-level functionaries.

The Mughal Vision of the Agrarian System

Although it is the everyday Mughal state and its functionaries that is our main interest in this book, it is still useful to start with Abul Fazl, like everybody else. We will then proceed to the manuals that began proliferating towards the end of the seventeenth century, in order to trace an outline of the personnel structure that was meant to structure tax-extraction in the countryside. Abul Fazl called the main imperial tax official in the district: the ʿamalguzār, alternatively known as the ʿāmil.Footnote ²⁷ This man, said Abul Fazl, was meant to be a friend to the peasant, advancing him loans on soft terms when needed and collecting taxes flexibly, through a variety of measures, in cash as well as kind, depending on the area under cultivation and the value of the crops. Assessing taxes payable was a major part of his duties. He was supposed to deal with every agriculturist’s case on an individual basis, be aware of past assessments and correct any errors therein, and make adjustments for changes in usage of land, for example, for animal husbandry rather than cultivation of crops. He was to keep stock of tax-free grants of land and check for the accuracy of land boundaries, work on abolishing illegal taxes and submit a monthly report on the general conditions of the district. In addition, he was supposed to keep an eye on law and order in the area, apprehending and punishing thieves and highwaymen, and even taking on the role of the kotwāl, or police officer, if required.Footnote ²⁸

Since, however achieving all this would have been impossible for an outside appointee without the benefit of local knowledge, Abul Fazl indicated the range of extra-local and local officials who could assist the amalguzār. These included specially appointed survey officers (jarībkash and paimanda), and clerks (bitikchī and karkūn), all of whom appeared to accompany him. The amalguzār’s entourage also incorporated eminent villagers. All procedures of measuring, collecting and recording, while apparently administrative, included contractual elements. It began with the surveyors – the amalguzār was encouraged to take z̤amānat (surety) from these officials – no doubt to ensure that they did not cheat. From the kalantar – literally biggest man (i.e. village head man) – too, a muchalka (agreement) was to be taken that no lands would be hidden. Then, as the official called karkūn kept writing the sawānih-i z̤abtī (record of assessment), village officials called muqaddam and patwārī would write along. It is not clear what ham-qalam bāshand implies, since it could mean both keeping their own records and writing on the same document. But it is clear that mutual checking, validation and signing to indicate such validation was an essential part of the process. At the end of the assessment process, the amalguzār would compare the documents, seal them with his own seal, and give a copy to the bitikchī. This detailed record – called khasra – would be sent by the bitikchī to the court, but an abstract would be entered into the village accounts, which would be reverified by the karkūn and the patwarī, and then forwarded to the imperial centre.

When it came to actual collection of the taxes, the official called bitikchī, a constant companion of the amalguzār, appeared to be the accountant who would interact with the villagers paying the taxes and the pargana and villager record keepers, the muqaddam and patwārī, most intensively. Bitikchī and karkūn may have been alternative ways of referring to the same accountant always accompanying the amalguzār, because Abul Fazl assigns the same tasks to one and then the other, and only defines the post of bitikchī. In any case the bitikchī/karkūn would maintain a roznāmcha (day-ledger) of collections and periodically send these ledgers to the court. It is in connection with the bitikchī’s office that the tasks of a host of village officials, and the papers they would regularly generate, are slightly elaborated: the qānūngō would maintain and supply the muwāzana dah-sāla (ten-year rent records),Footnote ²⁹ the patwārī would also maintain rent rolls called taujīh, according to which the muqaddam would collect the taxes. Copies of the patwārī’s rent rolls had to be supplied to the bitikchī for cross-checking with his own collection records, along with the sarkhat or yad-dāshti documents that had been given to the cultivators, perhaps as a receipt. Periodically, the bitikchī would send his ledgers, countersigned by the treasurer and the ʿāmīl, to the court.Footnote ³⁰ The muqaddam would retain, for his pains, the collections of one biswa per bīgha of land, that is, 1/20 of the collections. But the muqaddams were constantly suspect: Abul Fazl warned the amalguzār not to allow the muqaddams to make their own tax assessments, lest they oppress the peasants.

Abul Fazl’s description left a lot unclear about these several officials, their exact duties, the length of their office-holding, and their mutual hierarchy. Among other things, there is no mention of the office of chaudhrī – which, in combination with qānūngō, kept our protagonists in business over so many generations. Things are both clarified and complicated with reference to the proliferating manuals of governance and related enumeration and documentation – the dastūr al-ʿamal, munshāts and siyāqnāmas – that became very popular in the early eighteenth century, many bearing dedicative titles naming the Emperor Alamgir. Taken together, these overlapping genres offered further details by providing model forms of appointment to various posts, including the lower-ranking ones that did not quite make the cut with Abul Fazl.

Dastūr al-ʿamal – Rules for Doing Things

Despite their abundance, the dastūrs are difficult sources. One source of difficulty is the variety in nomenclature for what appears to be the same posts; while some of the differences arise from regional conventions, in other cases we can only speculate why one title is used rather than another. The village-level tax-collecting officer whom Abul Fazl called muqaddam, for example, appears to be the same as the chaudhrī of Hindustan (northern and central India), the desāī of Gujarat and some parts of the Marathi-writing areas and the dēśmukh or munīwar of Dakhin (western and southern India). At the neatest conceptualisation, the muqaddam/chaudhrī/dēśmukh/desāī was a man co-opted by the regime from among several village headmen and, by his standing, inevitably a zamīndār.Footnote ³¹ The Sikh empire, which retained and elaborated the Mughal administrative posts and terms, and was most closely observed in vivo by the British, had one or more muqaddams in every village, but only some of them were chaudhrīs whose hereditary job was to collect taxes, taking a share of the collections as commission. But the extent of overlap between (even) these north Indian terms – chaudhrī, muqaddam – remains murky, because contemporary documents reveal that both chaudhrī and muqaddam could also be termed panch, thus also associating this position with that of the legendary five elders of village governance.Footnote ³²

The other, and rather serious difficulty arises from what we can surmise about the context of production of these dastūr al-ʿamal manuals. Some of the most elaborate of these, frequently used by historians, were clearly produced under British pressure and patronage. In fact, the form was so popular with eighteenth-century Company officials, that there were British dastūr al-ʿamals produced, to explain the workings of the undeniably novel Company courts in familiar Persianate idiom, perhaps to the Indian personnel of these courts.Footnote ³³ And while accessing the Mughal system via the accounts of carried-over personnel is a logical idea, there are risks of ignoring the possible interests such people would have had in representing things in a certain way. And in this connection, it is rather crucial that, in elaborating the office of the chaudhrī, these eighteenth-century manuals were commenting on the status of zamīndār – perhaps the single most contentious policy issue of eighteenth-century colonial India.

Before we take a look at these manuals then, we have to take note of the other position that brought rewards and tax-free lands to our protagonists – that of the qānūngō. Mohan Das had been a qānūngō, and his grandson, Purshottam Das, himself styled chaudhrī, appeared to work in constant conjunction with individuals called Paras Ram and Parmanand, both qānūngōs. The qānūngō, if we remember our Abul Fazl, was the first point of contact in the district for the imperial bitikchī – supplying him with ten years worth of revenue records on demand. Unlike British observers of the eighteenth and nineteenth century, historians are generally of the opinion that the classic Mughal system consisted of two entirely distinct offices. The qānūngō was a hereditary accountant and record keeper, drawn predominantly from literacy- and numeracy-oriented castes, such as Kāyasth or Brahmin,Footnote ³⁴ and therefore both institutionally and sociologically distinct from the martial zamīndār who worked as chaudhrī. The fact that distinguished qānūngōs acquired extensive grants of land, and even mansabs, building up huge zamīndārīs in the process,Footnote ³⁵ is seen as an unintended outcome, and actual Mughal grants appointing individuals simultaneously to the posts of chaudhrī and qānūngō as a symptom of imperial decline. Indeed, historians may be echoing a colonial commonplace: British colonial officials in the early nineteenth century were convinced that, with the decline of previously centralised imperial power, there was an epidemic of land-grabbing, facilitated by the connivance of lower-level record keepers, especially qānungōs,Footnote ³⁶ or worse still, by the clubbing of official positions with portfolio investment.Footnote ³⁷ But what did the manuals say?

In the late eighteenth and early nineteenth centuries, when the East India Company was expanding its claims on the revenues of the various parts of the now-crumbling Mughal empire, there was something of a minor Orientalist craze for acquiring guides that would explain just how Mughal administration and its Indo-Persian paperwork functioned, and would open the caves of treasure that lurked just beyond reach. An internally diverse genre of Persian-language manuals began to pour into the collections of specific individuals, particularly, those involved with the supervision of the perpetually suspect Indian administrative officials.

One such collector was James Grant, the first British sarishtadār (record keeper), appointed in Bengal in 1786.Footnote ³⁸ Grant’s contribution to the India Office’s and, ultimately, the British Library’s collection of Persian manuscripts, many identifiable by the large Persian-language seal he used, was decidedly functional in scope. It included a dastūr al-ʿamal which has been frequently cited by historians of the Mughal empire.Footnote ³⁹ This text purported to have been composed in the third year of Aurangzeb’s reign, under his orders. All this was rendered rather untrustworthy by the fact that the unnamed author got the Hijri date for the regnal year wrong, taking it to be AH1065 rather than AH1070, as no near-contemporary would. This quite-likely late-eighteenth-century composition offered detailed instructions on enumeration and calculation in the context of tax-collection, using various locations in northern India to Afghanistan as realistic locales for the various sample revenue accounts that should be written, and the associated procedures that had to be followed. For all its detail, it did not actually descend to officers at the village level and their functions, and instead offered generalities about the khidmat-i diwāngirī (office of diwan), for example, that all officers associated with it, down to the zamīndārs, should be well adjusted to each other, collect and forward all paperwork generated by ͑amils to the office of the diwān, so that the rights of the Lord were respected and nobody was oppressed.Footnote ⁴⁰

There are several dastūr al-ʿamals, in James Grant’s and several other late-eighteenth-century collections, all with various imperial dedications.Footnote ⁴¹ The contents can vary widely: some are only light-touch surveys of provinces and districts therein, without any mention of offices, posts or procedures and united only by a neo-imperial interest in administration.Footnote ⁴² The codices in which these dastūrs are bound contain a second kind of manual, called khulāṣat al-siyāq, which elaborated further on the method of writing the cryptic accountancy numbers which were the stock-in-trade of tax officials.Footnote ⁴³ These numerals were logographs based on abbreviations of Arabic words for numbers, and were predictably used in documents that needed to record exact numbers, such as tax records, or indeed, transactional documents. Naturally, cracking the code for such numerals became a serious concern for the East India Company.Footnote ⁴⁴ The manuals that we may think of as ‘Mughal’ were very likely the product of that colonial effort to access the Ali Baba’s cave of Indian taxes.

Dictionaries were a third kind of guide into this intricate world, once again produced in abundance during the long colonial transition.Footnote ⁴⁵ In the early nineteenth century, Khwaja Yasin, resident of Daha, in Karnal district (about seventy miles from Delhi), wrote a dictionary for the use of British officials. Khwaja Yasin’s short Persian-language dictionary was focussed on administrative terminology and, in particular, on Indian words – lafẓ-i Hindī – as well as regional usages. For example, it offered the word chākarān but explained that it belonged to the muhāwara (idiom or usage) of Bengal.Footnote ⁴⁶ It also included variations of word meanings, offering a much larger range of offices for the administrative term ͑āmil, than that envisaged by Abul Fazal. According to Yasin, the term could be applied to anybody who held a position of trust (amānī), which could well include officials of law courts.Footnote ⁴⁷ With regards to chaudhrī, he had an interesting sociological explanation preceding the administrative one: ‘chaudhrī sardār-i firqa rā guyand … har kasī ke az zamīndār̄an muaʿtamad alaīhū bāshad ū-rā az sarkār khitāb-i chaudhrāī dehand (the leader of a sect is called chaudhrī … all those among the zamīndārs who are reliable are given the title of chaudhrāī by the government)’. The gloss continued, in a rather ungrammatical sentence: ‘Working as the helper of the ḥākim in everything, his task is mālguzārī (collecting taxes). The other, smaller zamīndārs were given a z̤āminī (bond) by the ḥākim, contracting them to pay the taxes. And the custom was that everyone of the z̤āminī mālguzārs took 5 per cent (of the collections) as the right of nānkār’. Towards the end of the definition, Khwaja Yasin noted that all groups had chaudhrīs, even the artisans, for example.Footnote ⁴⁸ This definition, which combined social eminence, reliability from point of view of the regime, and one or more contracts, offered a satisfactorily rounded view of a post that derived from status and was formalised by agreement.

The last type of manual written by Indian authors that we should consider here are forms of appointment orders. This was indisputably based on an older genre and shaded into the broader area of Persian-language formularies, or munshāts. These, too, were extremely popular among Company Orientalists, and were among the first Persian texts to be translated to English. One such formulary of appointment orders was produced in the early eighteenth century, by an individual who was surmised by the historian John F. Richards to have been a junior official in a central or provincial revenue department during Aurangzeb’s reign. This text describes the posts of chaudhrī and qānūngō in tandem, thus: ‘at this time, according to the exalted order, falān (so-and-so) is appointed to the khidmat (service/office) of chaudhrāī and qānūngōi of pargana so-and-so due to the transfer of so-and-so upon payment of (amount) pīshkash (tribute)’. The ‘so-and-so’ form of anonymising documents points to a modular standard, one intended to be copied without much introspection. The officer thus appointed was then instructed to both maintain and despatch tax records to the provincial diwān, as well as maintain law and order and refrain from illegal extractions.Footnote ⁴⁹ Based on his extensive research, especially into the administrative system in Golconda, Richards deemed such conflation of posts exceptional,Footnote ⁵⁰ but there are further such conflations suggested even within this manual – the military and fiscal posts of faujdār and amīn (referred to as amānat) are co-awarded in a single model appointment form.Footnote ⁵¹ At least by the early eighteenth century, then, the writers of formularies expected such conflations. In case of the chaudhrī-qānūngō, the demand of pīshkash (which we have also come across in Chapter 2) identified such appointments as part of a process of attrition and incorporation of rural notables, rather than unilateral imperial appointment.Footnote ⁵²

Historians of the Mughal empire have used these manuals intensively and in conjunction with sources they consider entirely distinct, if useful, namely, accounts of East India Company officials looking to understand the system. In fact, although separated by language and authorial identity, the Persian and English accounts were both products of the same process of colonial investigation into the Mughal system, and in using them, it is important to be aware that we are following a well-trodden path of colonial discovery. Thus the famous ‘Amini Report’ of Bengal, produced in 1778, under direct orders of Warren Hastings, was part of that effort to settle the revenues of Bengal (i.e., establish how much taxes were to be paid and by whom). It classified the taxes and the tax-payers, in the process defining a zamīndār as the ‘superior of a district, of which … he collects the rent for which he pays a revenue to the government’, but also, rather confusingly given what we have seen before, as superior to a chaudhrī.Footnote ⁵³ This may have been Bengal-specific usage or an error, but the Amini Report did offer useful definitions of the zamīndārs’ rights, such as the nānkār, the proceeds of land set apart revenue-free to support him.Footnote ⁵⁴ It also described the types of documents and subordinate offices a zamīndārī required. Around the same time, there was a report produced on the post of the qānūngō, with an aim to assessing the usefulness of this office. This report included long lists of the kind of documents a qānūngō was supposed to maintain, including rent rolls and copies of qabūliyats (deeds of acceptance entered into by zamīndārs).Footnote ⁵⁵ No rent roll documents have been found in the reconstructed archives of the protagonists of this book. Nor do we see the papers related to zamīndārī administration in Bengal in the early nineteenth century. We have to provisionally conclude, therefore, that either the regional traditions were significantly different or the kinds of documents considered essential to preserve in early colonial Bengal were different from those in Mughal Malwa.Footnote ⁵⁶

Despite their detail, and hence their usefulness in opening up the underbelly of zamīndārī, what these British-commissioned manuals and surveys lacked was a sense of politics that so characterises the documents in our collection; the constant negotiation between village strongmen and imperial nobles, and all who passed in between. It was that negotiation that contextualises the contractually toned documents, as well as the pleadings and receipt of royal grace. In decapitating the Mughal and post-Mughal polities, the Company Raj, despite its name and ceremonial, aimed to reduce systematic jostling to routinised administration. And in working through colonial-era Persian materials, historians may have been distracted by the language and misled into taking colonial ambition for description of precolonial reality.

Contractual Incorporation: the Reality of Governance

The Mughals had such a vision of perfect order themselves. As we have seen, the chaudhrī and qānūngō were supposed to be locally co-opted officials, their roles clearly distinguished: military ability made a good chaudhrī and accountancy skills made a good qānūngō. The offices also mapped on to what are now seen as caste-specific skills. In reality, however, things were more blurred, and precisely because of the necessity of collaboration between the chaudhrī and the qānūngō. Whether or not this was a symptom of imperial decline, such overlap in posts was also the case with our family. But it was not just Purshottam Das’s family that concentrated these posts among kinsmen, or even in the same hands.

In the neighbouring sarkār of Chanderi, a family of qānūngōs had built up substantial fortunes, based on initial direct contact with a prince and the imperial court and sustained reaffirmation of their position and perquisites by repeated imperial orders. They gained control over several villages, tax-free, as ͑inām, bestowed their names on some of these villages and established marts blessed by the imperial court.Footnote ⁵⁷ In comparison, Mohan Das’s family dealt in a smaller territory, only a district or two, as opposed to an entire sarkār.

As we know from John Richards’s detailed work of Mughal administration in Golconda using the largest extant collection of Mughal administrative documents, the chaudhrī (dēśmukh in the Deccan) gave an undertaking or qabūliyat whereby he promised to collect and deposit an agreed sum of money in the provincial treasury. Such agreements were made every two or three years, and in some cases, especially during times of transition, involved the payment of an enormous ‘fee’ for the privilege, whether to the locally deputed mansabdār (noble) or to the imperial court directly.Footnote ⁵⁸ From the point of view of the regime, this process of give and take and mutual attrition was part of incorporating the local aristocracy into the state apparatus; from the point of view of zamīndārs-turned-chaudhrīs, it was a risky investment, whose success depended on their military as well as diplomatic skills, for they had to extract the taxes engaged for (plus make a profit) from the peasants, while maintaining some kind of alignment with the bureaucratic ideal of objective surveys, transparent and accurate records and bloodless tax-collection. Maintaining that alignment depended a great deal on collaboration with the other official named by Abul Fazl, the qānūngō; it was therefore most convenient for the zamīndār-contractor when he, or his family members, could hold both offices at once.

The association of the rural warlords with the tax-extracting state is well documented first and foremost in a type of document called qaul qarār, of which we have fifteen in our collection, all from the National Archives of India, except two, which are from the family’s household in Dhar. They range in date from 1626 to 1726, which could point to this being a Mughal form abandoned by the Marathas, or it could simply be a function of the sparseness of surviving documents for the family after the second quarter of the eighteenth century.

The earliest qaul in the collection, dated 1626, began the Persian section thus: ‘Qaul qarār patta-yi ijāra-yi mauzaʿ Mohanpur fī ʿamal-i pargana Dhar sarkār Mandu gīrad [?] az qabūliyat-i Purshottam Das … ’ (A qaul qarār document of lease of the village of Mohannpur in the administration of district Dhar, sarkār Mandu, was taken by the agreement of Purshottam Das …). The document detailed the amount of māl-jihāt (land tax) and sāyirjāt (other taxes) that Purshottam Das would pay for the three years of 1034 to 1036: 46 rupees in the first year, and 52 rupees per year thereafter. The amounts of money payable are written in the siyāq numbers that we touched upon earlier in this chapter, as well as Hindi numerals. The calendar used is a combination of Islamic months (13 Zu al-qaʿda, in this case) and Faslī years.Footnote ⁵⁹ The document bore two seals, the more legible one saying ‘Banda (slave/slave of) Sundar Das jīū’ in Persian script. The identity of Sundar Das is not clear: he may have been an ancestor whom members of the family venerated in their personal seals, or he may have been a revenue official, possibly also a relative. The Hindi version, in the bottom half of the document, followed the Persian text closely but not entirely. While the first four lines followed the Persian so closely that nearly all the words were identical although written in the Nagri, rather than Perso-Arabic script, the next three lines added elaborations not provided in the Persian section. It also ended with instructions in a different formula. The Persian portion ended with ‘mublagh mazkūr rā sāl besāl faṣl befaṣl bi lā ͑ujar jawāb gūyand’ (he will answer for the said amount from year to year and season to season, without excuse), whereas the Hindi added a bit of declarative flourish and a pat on the back ‘sahī hamāra kaul bōle hai, khātir jama kar kamāī karṇā. Bidi’ (A true/valid kaul has been uttered. Earn with a settled mind).Footnote ⁶⁰ The shikasta of the Persian section is particularly difficult to read but the words are distinguished; the Hindi, on the other hand, is even harder because of the absence of word breaks and the incursion of graphemes now associated with other languages, such as the character for ‘ja’ in Marathi/Modi and Gujarati. The use of the retroflex nasal in the imperative verb endings, as in karṇā indicates a Rajasthani phonology, which accords well with the Malwi/Rangri later observed in use in the region.

Qauls were the most persistently bilingual and bi-scribal of all document types encountered in this family’s archives – typically, with Persian in the upper half of the document and Hindi in the lower. Examining similarly bilingual legal and administrative documents from sixteenth-century south-western India, Sumit Guha has suggested that these were the result of greater state penetration into the countryside and its resources. Even in the seventeenth century, after self-conscious efforts in the Maratha empire to create a Sanskrit-derived administrative and diplomatic vocabulary, the Marathi used in the substantive portion of the documents continued to abound in vernacularised Persian words. As Guha suggests, such language use attests to a specific kind of plurilingual skill set possessed by administrative functionaries, who knew some Persian administrative terms and formulae by heart and found it easiest to use these in the Marathi/Moḍi script when writing out formal orders that were dictated by kings and commanders. Such scribes may not have been able to write the Perso-Arabic script, at all, or very well.Footnote ⁶¹ If this was indeed so, then this is very similar to the social context that produced the Judaeo-Arabic documents of the Cairo Genizah, and possibly the Judaeo-Persian documents of the Afghan Genizah, in which, incidentally, the writers demonstrate at least some facility in writing the Arabic script, too.Footnote ⁶²

The villages and places named in these documents map out the area of operations and influence – dā’ira – for Purshottam Das, his descendants and his associates. Ten of these qauls are for individual mauzaʿs – Mohanpur (twice), Nalawada (three times), Antrai, Nekpur (twice), Ekalduna and one unidentifiableFootnote ⁶³ – for sums that range from 35 rupees (Mohanpur, 1654) to nearly 900 rupees (Nalawada, 1664). The villages are arranged in a tight circle around the city of Dhar, and I do not have information to judge why one of these villages would be viewed as so much more valuable to a contractor for revenue than the others. There are also qauls of higher value and for larger regions – Purshottam Das took on ijāra the entire neighbouring pargana of Jamli in 1659, for what appears to be a paltry sum of 901 rupees.Footnote ⁶⁴ In 1655, that is, in the same year that he was confirmed in his chaudhrī-ship of pargana Dhar, Purshottam Das joined several others in a collective qaul for the entire pargana of Dhar, for the amount of 89,501 rupees,Footnote ⁶⁵ which offered a discount in case of locust attacks; in 1659, he was party to another collective qaul, worth 109,571 rupees.Footnote ⁶⁶ In two of these three district qauls, the seal is that of Nawazish Khan, the long-time jāgirdār of the area, and the marginal notes are in the bold hand associated with his parvānas. Clearly, at some point, this jāgirdār felt it most convenient to deal with the entire district rather than for one village at a time; Purshottam Das, the recently elevated patriarch of the family, was ready to step up to provide the lead. The two collective qaul documents are very useful for mapping out a range of associates for the core line of the family – we see Purshottam Das chaudhrī in the lead, followed by his most long-standing partner, Paras Ram qānūngō, and a host of others who are later glossed as the muqaddamān and ijāradarān of the pargana. Many of these were Purshottam Das’s relatives, near and far: there was his cousin Suraj Bhan in both cases, and in the second there were also Tilok Chand and Kanwal Das, more distant cousins with whom he had disputed over hereditary entitlements.

Thus although the qaul of 1626, just discussed, was a contract with an individual, for one village and bilingual, this is not true of all the qauls in the collection. Seven of them, that is, nearly half the number, are only in Persian. There does not appear to be any identifiable pattern to the use of languages – Persian-only qauls appear throughout the date range, for individual as well as collective qauls.

Since the usage of documents changes their meaning, it is unproductive to offer a single definition of what a qaul was. However, this was a prolific documentary form that shared linguistic space with Persian, Indic languages and Arabic, each language-zone introducing its own phonetic and semantic variations. To take some examples, in the Marathi (and Persian)-writing areas, kaul or kaulnāma indicated a document of assurance, which was renamed abhayapatra in the Sanskritising later seventeenth century.Footnote ⁶⁷ Possibly first produced in the fourteenth century, some of these kaulnāmas were bilingual (Marathi and Persian), but the majority in Persianised Marathi continued to be produced well into the colonial period and were used to record a range of dealings which required assurances.Footnote ⁶⁸ Meanwhile in the Telugu- and Tamil-writing region, and especially the warrior-state of Pudokottai, which came into existence in the seventeenth century, ‘kaval’ indicated protection, while the king assumed ‘patikkaval’ rights which entailed the right/duty to protect a ‘pati’ or place.Footnote ⁶⁹ In the nineteenth century, Indian Ocean mariners referred to safe-conduct passes as qawls, and used these in combination with flags of various regimes that might afford them protection against capture at sea.Footnote ⁷⁰

Thirteen of the fifteen documents in our collection all begin with the words Qaul qarār patta-yi ijāra; the four words making up this titular phrase presenting a wonderful example of linguistic hybridisation across Arabic, Persian and Hindi through administrative practice. The word qaul, means ‘speech’ or ‘utterance’ in Arabic, and derives from the Arabic root q-w-l, which is related to a spectrum of meanings related to the speech act. Conventionally, all hadith began with ‘Qala .. (He said/reported)’. Qarār was another common Arabic word, meaning a resolution or settlement, derived from the root q-r-r, which in turn generates a range of words associated with the sense of stability, finality, settling. A word derived from that complex is iqrār, which is a technical legal term in Islamic law, implying a binding declaration, which counts as the best evidence in an Islamic tribunal. Together, then, qaul and qarār referred to an Arabic vocabulary, and implied an utterance that settled matters, ironically, encoded in writing in Persian and Hindi.

But the next two words took the mixing of vocabularies and concepts several stages further: paṭṭā, with its characteristic retroflex consonants is a Sanskrit word that bears a range of meanings from tablet or plate for writing on, a document written on that plate, or a diadem (similarly indicating title). Interchangeable with patra, the word is very similar to the Persian nāma, in indicating textual genres that ranged from personal letters to royal edicts and official reports. In Persian documents that use the word paṭṭā itself (as in the one we have just seen), the retroflex sound, characteristic of the Indic soundscape is not indicated, and indeed, this is the orthographic pattern well into the nineteenth century. In colonial India, paṭṭās were ubiquitous, and referred to a range of title-deeds pertaining to property and entitlements related to land, but also more abstract rights, such as employment. In his dictionary, Khwaja Yasin produced a definition of the term, explaining that it was a lafẓ-i Hindī, and provided alongside a model which defined the rights and responsibilities of a revenue-paying zamīndār.Footnote ⁷¹ The states of Marwar from the nineteenth century compiled registers of such deeds in paṭṭā-bahīs which are still a major object of interest for landowners in those areas.

Finally, ijāra. This Arabic word, implying lease, was the only unambiguously legal word in the quartet of terms that made up the name of this document. Today known worldwide as a device used to achieve sharī ͑a-compliant mortgages, there is an extensive jurisprudential literature associated with ijāra, much of which concerns the risks of usury or ribaʿ, which is prohibited in Islamic law. In the Mughal context, it was most commonly associated with revenue farming, that is, a contract undertaken to collect taxes in an area and pay in a fixed sum – whether to the imperial tax office or a jāgirdār. Although it appears that the method was frowned upon, it was still widespread.Footnote ⁷² Colonial surveys in northern India in the early nineteenth century tried to distinguish the real ownership rights of the zamindār, from a mere contractor or mustājir (one who had taken on an ijāra contract), but found that things were often too tangled to separate.Footnote ⁷³ In the qaul qarār paṭṭā-yi ijāra documents, we see direct record of that complex transactional bundle that combined local power with tax contracting.

The protracted late-eighteenth-century process of colonial ethnography and archival hunts uncovered a vast amount of documentation related to rural landholding, including several qauls. Some of these were from the Northern Circars (or the northern sarkārs) which were ceded to the English East India Company in 1765 by the embarrassed Mughal Emperor Shah Alam, out of the area controlled by the provincial governor-turned-ruler of the Hyderabad state, Nizam ul-Mulk Asaf Jah, and only gradually wrested by the British from the latter’s control.Footnote ⁷⁴ The detritus of the colonial machinery-in-formation was analysed by an Orientalist named Charles Frances Greville, who had never been to India but had very strong views about its governance.Footnote ⁷⁵ Greville advocated careful scanning of sanads (documents) of various zamīndārs and an effective land registry, believing, like the James Grant he quoted approvingly, that much revenue was being lost to the state otherwise.Footnote ⁷⁶

Among these crucial documents were ‘cowles’ that, according to Greville, used to be given to zamīndārs every year by the Mughal district revenue officer, the ͑āmil, based on a rent roll prepared by the great accountant Todar Mal in 1592. Naturally, that rent roll had to be locally updated with more recent information about hustobood (hast-o-būd, that which is and was), based on the daul or more specific documents in the possession of the zamīndārs.Footnote ⁷⁷ Our qaul, then, would become the next year’s daul – and form part of the kāghaz-i khām (rough papers) in Hindavi that Emperor Aurangzeb supposedly demanded be acquired from every village, translated and their contents or summary entered into the official Persian rent roll or tumār, all in the interest of more transparent and efficient revenue administration.Footnote ⁷⁸ Believing in the truth of its findings, the Company government continued to hand out ‘cowles’ to zamindārs well into the nineteenth century, giving Mughal terminology and forms an extended lease of life. That formal extension, however, flattened out a political process into an image of free-market contracting. Fortunately, in our collection, we have documents that reveal to us the multiple negotiations that surrounded the contracts of the qaul qarār in Mughal-era Malwa.

Conflicts and Resolution

The mansabdār-jagīrdār Nawazish Khan had developed a symbiotic relationship with our zamindār family, as we have seen in Chapter 2. The noble was able to order Purshottam Das to look for camel thieves, run up to court for unstated emergencies and send along valuable supplies, in turn confirming their titles and intervening in family disputes to sort out the errant relative Suraj Bhan. But when it came to taxes, his own salary depended on it, and there was no room for a soft touch. In a peremptory parvāna dated 13 Zu al-qaʿda 1073 AH (18 June 1663), Nawazish Khan noted that Purshottam Das chaudhrī and Paras Ram qānūngō, both of Dhar, had sent in a petition about the collection of taxes. Whatever the duo had requested, it was summarily swept aside; it is necessary, said Nawazish Khan, that whatever had been written about (i.e., agreed), taxes in the qabūliyat should collected, and no excuses were to be made. In fact, such efforts should be made that the district was cleared of arrears (bī bāqī shavad). The same message was repeated in summary on the right margin of the document, in a characteristic bold hand we see in many of Nawazish Khan’s parvānas, perhaps his own.

We can only imagine that Purshottam Das and his work partner backed down in the face of this dismissal. But sometimes the demands of superiors did become too much to bear. Two documents in our family collection point to such an episode. These documents, which were collected from two different repositories – the National Archives of India and the family’s mansion in Dhar – point towards outright conflict between three crucial officials whom we can recognise from the dastūr al-ʿamals – faujdār, amīn and karōrī – and Purshottam Das, by this time chaudhrī of several parganas, including Dhar, Nalcha and Hindola. One of the documents offers the year for a key incident, which, if we take to be a Faslī year, as the NAI archivist has done, is 1669; the conflicts described in the other document are either the same one or associated and similar. In any case, they together point to the complex nature of overlaps that we have been attempting to reconstruct, and are particularly illuminating because they indicate conflict.

The first document is an account or record of a series of incidents, made in a general maḥz̤ar-nāma form, beginning with the formula ‘bāiʿs-i tahrīr-i īn sutūr’ (the reason for writing these words), and appears to be a recitation of antecedents backing up a petition, the crux of which was alleged inability to pay the agreed taxes. Purshottam Das chaudhrī and Parmanand, describing themselves as mutasaddiān (officials) of parganas Dhar and Nalcha, said the position was this: because of the taʿādī and sitam (oppression) of the faujdār Ahmad Beg, amīn Barbek and karōrī Nimatullah, they had been forced to take their complaints and leave their districts and go to the exalted court on 10 Muharram 1072 (1661 CE) whereby these three officials were replaced by men more acceptable to themselves. However, through counter-lobbying, the displaced faujdār attempted to have the areas allocated to himself in jāgīr. Although this effort was foiled by the protests of Purshottam Das and company, they returned to their villages to find that the imperial tax officials had collected taxes according to the harsh assessments of the previous officials, causing havoc and depopulation. As such, Purshottam Das and his associates claimed inability to pay the taxes and requested a fresh assessment so that the charges could be handed over to the karōrī.Footnote ⁷⁹

The second document is a parvāna, issued by a noble unidentifiable by the seal, but whose scribe had very similar handwriting to the person writing several of the parvānas issued by Nawazish Khan. In this undated document, the noble assured Purshottam Das, referred to as chaudhrī of the parganas Jamli and Hindola, that his ʿarzdāsht had reached its destination. Purshottam Das’s complaint, about the displacement (bi-jā shudan) of the mardam-i mahājan and khushbāsh (worthy and well off) of qasba Sultanpur, due to the bī-sulūkī (bad manners) of the karōrī of that area, had been noted. Purshottam Das was directed to pass on the enclosed note, accompanying the parvāna, to the offending karōrī, and work together with him to conciliate and soothe the locals, so that they returned to their places. The noble, annoyed by the dislocation caused by the left-footed karōri, encouraged him and everybody else to thereafter engage themselves in the improvement of agriculture to ensure increase in tax collections.Footnote ⁸⁰

The rebuked karōrī may have been the Nimatullah against whom Purshottam Das had complained, and the incidents as well as their dates are close enough for us to speculate that the two documents were related to a single protracted episode of conflict, over the appropriate mode of tax-collection, between imperial officials and our zamīndār family. But the nature of Purshottam Das’s complaint points to more entangled relationships; here, the collusion of the faujdār with the amīn and karōrī appears to have produced an oppressive local clique. Even when effectively challenged, such officials, or at least those with military ranks, could attempt to cling to those rural revenues through alternative ways, and not necessarily for the benefit of the regime – the effort of the faujdār to get the area allocated in his jāgīr is telling. However, large zamīndārs such as Purshottam Das and his associates preferred certain imperial officers over others. Also, there were multiple routes for expressing that preference – from approaching the mansabdār-jāgīrdār through petitions, up to a direct personal petition to the imperial court.

Tax Contracts and Law

The majority of documents explicitly related to the collection of taxes are contractual without making any explicit references to a body of law. However, it is also within this subset of documents that we have the few explicit references to ‘law’ as an abstract entity. These references occur in a protracted set of complex transactions, where a tax contract is underwritten by Purshottam Das, and subsequently transferred to him as guarantor.

In 1662, Purshottam Das and his associate Paras Ram paid 1,600 rupees to a servant of the jāgīrdār Wazir Khan, as surety for Hari Ram chaudhrī.Footnote ⁸¹ It is not clear whether their willingness to do so was due to family connections or an astute business move; in any case, this was not an isolated instance, for in 1666, Purshottam Das again stood surety for chaudhrī Kanpil of Amjhera, which was the neighbouring district to the west of Dhar, which, as we know, was a Rāthoḍ Rajput base with which the family had been involved.Footnote ⁸² This time, the jāgirdār reminded Purshottam Das that Kanpil had defaulted.

Defaulting was serious business, and the bonds taken and given were in dead earnest, for in the same year, 1666, a copy of an entry in a register – an yād-dāsht – noted that Hari Ram chaudhrī’s dastūr was confiscated in favour of Purshottam Das.Footnote ⁸³ The resulting order, a parvāna under the seal of Wazir Khan, made clear reference to bodies of law – according to sharīʿa and ʿurf, it said, Purshottam Das was the rightful owner of the dastūr, which Hari Ram had forfeited by not being able to pay up in time.Footnote ⁸⁴

At the time of this order, Wazir Khan was the ṣūbadār of Malwa, having been appointed to that position after the death of the previous ṣūbadār, Najabat Khan.Footnote ⁸⁵ In 1664, Najabat Khan had stepped in to prevent Suraj Bhan being deprived of his share of inheritance due to his unpopularity with his kinsmen. Now in 1666, Wazir Khan had to decide that Hari Ram chaudhrī’s defaulting and absconding was a serious enough failure to deprive him of his dastūr – customary rights.

It does not seem to be a coincidence that when it came a question of potential deprivation of inherited rights – of office-holding and associated rewards – the ṣūbadār involved himself in making the decision. ʿUrf was the traditional classical Islamic term for custom. Islamic legal scholars were generally sympathetic to ʿurf provided it did not conflict with core doctrines; it is striking that this explicit reference to legal adjustments with local custom occurs, not in a document issued by a qāz̤ī, but an order issued by a noble concerned about the source of his salary, and dealing with complex outsourcing of tax-collection under his banner. Was the stray reference to ʿurf in this document a relatively unusual effort by a perhaps exceptionally sharīʿa-literate mansabdār to translate the ubiquitous dastūr into a technical language recognisable within the wider Islamic world? If so, who was that translation for the benefit of? We are inevitably left with some questions that we cannot as yet answer.

Conclusion: Tax and War

A flush of recent literature on scribes serving various precolonial regimes reveals a spectrum of social statuses and functionaries, united by the common skills of writing and record keeping: the arts of inshā and siyāq. Collectively, the professional group was deemed munshīs in Indo-Persian vocabulary. At the highest end of the spectrum were ministers and diplomats who, while proud of the powers of the pen, were not entirely averse to or immune from wielding the sword when called upon to do so by their kings.Footnote ⁸⁶ Such men sported Persian titles, savoured Persian poetry and were fully au courant with the Indo-Persianate courtly culture or adab. At the lower end of the spectrum were village-level scribes, who dealt more in numbers and vernacular languages and scripts. In north India or Mughal lands, these men, generally of the Kāyasth or Khatrī caste groupings, acquired a culturally shallow level of Persianisation, with lesser investment in Persian classical poetry and more in legal and tax jargon – the vocabulary of which tended to be shared across multiple languages.Footnote ⁸⁷ When various areas came under British control, these lower-level functionaries typified the information bottlenecks that came to be denounced as sources of corruption, whereas, of course, they were also the conduits that made the colonial regime possible by making money and information flow, by whatever means. In Bengal, the post of the qānūngō was abolished in 1793, when the Governor-General’s council decided that embodied local knowledge was redundant if the amount of tax due was agreed through written contracts.Footnote ⁸⁸ Thus, while popular lore depicted munshīs (and qānūngōs among them) as essentially a bunch of pen-pushing scribes with their secret bundles of crucial information that could make or break tough men and honest soldiers,Footnote ⁸⁹ in fact, information, landholding and force may not have been such distinct areas of expertise. Multitasking could extend to religion, too: a future imperial ṣadr – supervisor of charities and so of charity-enjoying ʿulāma and qāz̤īs – made his mark as the secretary of a Mughal princess.Footnote ⁹⁰

The story of qānūngōs-cum-chaudhrīs that this book tells is meant to query the assumed line between the pen and the sword and propose that munshīs and thākurs did not necessarily have to be completely different kinds of people in Mughal India. There were a variety of routes to acquiring visibility and, in select cases, utility in the Mughal regime; in the countryside this almost necessarily included the ability to manage violence. Tax-collection, however systematised on paper, was only one step away from minor warfare; combinations of skills therefore made good entrepreneurial sense, at least in some cases. On the other hand, successful office-holding led to acquisition of lands and tax privileges, pulling the proto-bureaucrat into the militarised world of landlordism. The Mughal regime ruled by riding this process, and the story of our family of landlords illustrates how. What is feared across the world today as ‘state capture’Footnote ⁹¹ was actually how the state operated in early modern South Asia.