Chapter VII

— Land Revenue

The Imperial Gazetteer of India, Vol. IV, The India Office, 1907 · from page 204 · about 79 minutes

Read by machine, not yet checked by a person. It may hold small errors.

A subject more remote from the interests of the average Englishman than the land revenue of India can scarcely be conceived. In order to appreciate it in any way he must translate himself into entirely new surroundings. He must leave behind him his limited conception of a land where the majority of the people live in big cities, where every one is dependent on the town for subsistence and for luxuries, where urban interests and urban wealth everywhere predominate. He must look instead upon a vast country where the immense majority of the people live by agriculture, where towns are exceedingly few in number, and where town industries are comparatively insignificant. In India the whole outlook is agricultural. The normal landscape of a prosperous Indian district is represented by some wide alluvial plain or slightly rolling upland where almost every square yard is under the plough. Hedges and walls are, as a rule, unknown; at harvest time the waving fields of rice, wheat, or other food-grains reach almost without interruption to the horizon; and in the intervals between the harvests an equally uninterrupted stretch of shimmering heat-baked soil fades away into the haze. Every morning during the cultivation season the Indian peasant—the most frugal and patient in the world—goes out to his fields with his cattle to work his well or to plough his land: and every evening he returns to his thatched cottage by the trees or his mud-built house in the closely packed village. His wants, such as they are, are met by the local artisans and menials, whom he rewards with customary contributions from his harvest. From this out-turn every class is provided for: it is to this that the Government looks for its revenue, the landlord (if there be one) for his rent, the grain merchant for his profits, and the labourers and artisans for their wages. To all these persons—and collectively they represent nearly the whole of India—the harvest is the centre of interest, and to most of them the ‘state of the crops’ is no mere formal topic of conversation but the all-absorbing question of life.

To appreciate the question of Indian land revenue, therefore, the Englishman must put himself into a social environment entirely different from that with which he is acquainted at home. His political and economic preconceptions will also require considerable readjustment. The place, for instance, which is occupied by the land tax in English finance is very small, the receipts amounting to less than 1 per cent. of the total public income ; whereas in India the land revenue provides more than 23 per cent. of the whole income of the state and nearly 39 per cent. of the revenue proper. The system, moreover, of tenures and titles to land in the British Isles is so unique, and there is so little in England corresponding to the peasant holdings or to the ‘cadastre’ of continental nations, that an English inquirer approaches the land problems of India from a standpoint even more remote than that of the inhabitants of the greater part of the rest of Europe. He is apt also, if he pushes his investigations far, to be appalled by the amount of official literature to be studied on the subject and the esoteric character of the language in which much of it is written. Indian histories, too, while recording the names of celebrated soldiers and diplomatists, seldom mention those of men who, like Shore, Duncan, Munro, Thomason, and Wingate, have by their labours in the department of Indian land revenue profoundly influenced the welfare of many generations throughout vast tracts of country, and whose work must always retain for Indian administrators a great deal both of interest and importance.

It is at the same time most difficult to satisfy the wants of an outside inquirer by presenting in limited space a conspectus of Indian land revenue administration which shall be at once illuminating and accurate. To every general statement there are numerous exceptions ; and a pronouncement which is true as regards one Province or District can seldom be accepted without modification for other parts of the country. There are no doubt certain principles and methods which in a general way underlie the various systems in force, and it is the object of the following pages to present these as clearly as possible ; but in the perusal of what is here written it cannot be too carefully borne in mind that the actual details of revenue work are characterized by a variety and intricacy which no general description, much less an account as succinct as the present must be, can in any way hope to reproduce.

The land revenue of modern India is a form of public income derived from the immemorial custom of the country. In its primary form the land revenue was that portion of the cultivator's native rule. grain-heap which the state annexed for the public use, and this crude method of realizing the bulk of the state income appears to have been practically the only method in force throughout the greater part of India until the sixteenth century of our era. In that century the Mughal power was finally established in the north of India and began gradually to extend its influence southward into the Peninsula. Under its firm and centralized administration the system of land revenue underwent a series of reforms, chiefly connected with the names of the emperor Akbar, who ruled from A.D. 1556 to A.D. 1605, and his Hindu minister, Todar Mal. Cash payments, for instance, were to a large extent substituted for those in kind ; the cash rates were, when possible, fixed for a period of years instead of being subject to annual alteration ; and a more or less uniform system of revenue accounts was established, the terminology of which has now permeated every part of India1. But with the break-up of the Mughal empire and the increasing independence of the outlying provincial governors, the practice of leasing the revenue of large tracts of country became common ; the check on the farmers of revenue became less and less careful ; and although the memory of the Mughal assessments did not altogether die out, the collection of land revenue by the later Muhammadan governors, and by the Marāthā and Sikh powers which grew up alongside of them, became practically little more than a disorganized scramble for the greatest amount of income which could be wrung from the land.

As the several Provinces came under British control, their assessments were gradually reduced to order, the systems selected being at first tentatively adopted according to the varying circumstances of the different tracts and becoming more and more crystallized as time went on. The process was carried out with no little deliberation. In Bengal, which was acquired in 1765, a settled system was introduced by Lord Cornwallis in 1793. The greater part of the Madras Presidency was taken over in 1792–1801, and although the principles of the present system were introduced by Sir Thomas Munro and others early in the nineteenth century, the actual basis of the existing arrangements dates from as late as 1855. In the Agra Province, which was for the most part annexed in 1801 and 1803, the present system, though started in 1822, was first put on a working basis by the agency of Mr. R. M. Bird in 1833 and owes much of its subsequent success to the exertions of Mr. Thomason between 1843 and 1853. So too in Bombay the tracts which fell under British dominion in 1818 were not taken systematically in hand till 1835, nor were they assessed on any fixed system prior to the Joint Report published in 1847 by Mr. Goldsmid, Captain Wingate, and Captain Davidson. A number of different systems were thus gradually evolved on lines which were for the most part mutually independent. The Supreme Government has, as a rule, wisely avoided the enforcement of unnecessary uniformity, so that the differences, both in principle and in method, among the various revenue systems in India are by no means inconsiderable. It is usual, however, to differentiate them roughly on broad lines according to the status of the person from whom the revenue is actually demanded. Where the revenue is imposed on an individual or community owning an estate, and occupying a position identical with, or analogous to, that of a landlord, the assessment is known as ‘zamīndāri’; and where the revenue is imposed on individuals who are the actual occupants, or are accepted as representing the actual occupants of holdings, the assessment is known as ‘ryotwāri1.’ The former of these systems prevails throughout Northern and Central India, that is to say in Bengal, the United Provinces, the Punjab, and the Central Provinces; and the latter in Bombay, Madras, Assam, and Burma. About 53 per cent. of the land revenue assessment of British India is zamīndāri and 47 per cent. ryotwāri.

In the Native States the assessments generally partake of the character of those prevalent in the adjoining portions of British territory. As a rule, the methods adopted are more primitive and the assessments higher than in British Districts, but the general tendency is for the revenue systems of the Native States to become gradually assimilated to those of British India. The same may be said of the outlying areas under British administration, such as Upper Burma, parts of Assam, and the Baluchistān Districts, where the revenue arrangements are still in a somewhat incomplete state and have not yet been brought into line with those of the older Provinces. In the following pages, therefore, which profess to indicate the general principles of the Indian land revenue administration, it has not been thought necessary to refer again to the systems at present in force in Native States, or in Districts of the nature above described.

It will be found convenient to note the general features of the Indian land revenue administration, whether zamīndāri or ryotwāri, under three heads which represent the processes actually followed: namely, (1) the preparation of the cadastral record, (2) the assessment of the revenue, and (3) the collection of the revenue so assessed. The first two processes are known collectively in most Provinces of India as the ‘Settlement’ of the land revenue, and the officer who carries them out is known as the ‘Settlement Officer.’ The duties of the assessing staff entail a minute local inspection from village to village through large tracts of country, and there are few officers of Government who are thrown more into contact with the people than the Settlement Officers, or have greater opportunities for understanding their wants and feelings.

I. The Cadastral Record

An essential preliminary to the assessment of land in India, as in other civilized countries, is the preparation of a cadastral map. It is true that in Bengal, where the revenue was permanently assessed in 1793, the present assessment rests on information obtained without the aid of a survey; but this defect has given rise to inconveniences both fiscal and administrative, and it was found necessary in 1892 to introduce a cadastral survey and record in the Bihār Districts of that Province in order to regulate the relations of landlord and tenant. In Provinces outside Bengal the existing assessments are based almost without exception on a field to field survey. The map is prepared in some instances by the scientific staff of the Survey department; in others the skeleton data alone are provided scientifically and the rest of the work is plotted by a local staff. A separate map is usually prepared for each ‘village,’ the village in India being a tract of land corresponding in some ways to the ‘parish’ in Great Britain. The ultimate unit of assessment in most Provinces is the holding, and it is usual to map as they stand the fields of which the holding is composed. In Bombay, however (and to a lesser degree in Madras also), the actual holdings are to a large extent grouped into small ‘survey numbers’ with practically immutable boundaries which are solidly and carefully marked out upon the ground; and in Bombay this fact constitutes so characteristic a feature of the revenue system that the occupant who holds a survey number on the condition of paying the revenue assessed on it is said to hold on the ‘survey tenure.’

The cadastral map having been completed, a field register is usually prepared to correspond with it, and from this field register is prepared the ledger of holdings which corresponds to the ‘matrice cadastrale’ of Europe. This record is primarily a fiscal record, the object of which is to show from whom the assessment of each holding or field or number is to be realized and the amount to be realized in each case. It is not sufficient for this purpose to detail the names of the persons in possession, for there may be several parties in possession of the same land, holding various classes of rights—one being in the position of over-proprietor, another in that of underproprietor, another as mortgagee, another as tenant, another as sub-tenant, and so forth; and it is necessary to specify which of these parties is responsible for the revenue or (to use the technical phrase) with which of them the settlement is to be made. The principles upon which this point should be determined were the subject of prolonged discussion in the various Provinces at the commencement of British rule, and their determination involved an immense amount of inquiry into the complex and ill-defined tenures then found to be in existence. Owing to the interest excited by these inquiries the decision of revenue questions in India for many years entailed a minute examination of ancient tenures, and it is only of late years—since the old questions have been set at rest—that the study of the technique of Indian land revenue has to a certain extent been separable from the historical investigation of the origins of existing rights in land.

When the country was first taken over, the peasant who cultivated the land was found in some cases to be a temporary occupant only: in others he was recognized by custom as having certain hereditary rights. In some cases he paid the revenue direct to Government, in others the payment was made through an intermediary. This intermediary was in some cases a village headman, whose only authority over the peasant lay in the right to collect revenue; in other cases he was a large revenue contractor appointed by the ruler of the day, or an official to whom the revenue had been assigned, or some hereditary chief whose position had been wholly or partially recognized by the Government in power, or a community of hereditary cultivators which had survived the disintegrating influence of time. The degree too in which any of these intermediaries interfered with the land or was in his turn interfered with by Government varied greatly. In some cases he could oust cultivators, in others he could not. In some cases he could break up waste or alienate land, in others not. In some cases the Government could remove him during his life or confiscate his rights on his death; in others he and his heirs were left in undisturbed enjoyment. In some cases he was the only intermediary between the cultivator and the Government; in others he was one of a long chain of intermediaries, each of whom was allowed, or had usurped, special claims.

Soon after annexation, therefore, it had in each Province to be determined whether the revenue should be taken direct from the cultivating peasant, and, if not, from which class of intermediaries it should be realized. In determining this point the Governments of the day were led partly by motives of convenience, and partly by the presumption that one or other of the parties must stand to the soil in a position resembling more or less that of proprietor, and that the party so standing was caeteris paribus entitled to be held responsible for the revenue. Accordingly, in Bengal, in parts of Madras and of the United Provinces, and in the Central Provinces, the highest class of intermediaries, standing next below the Government, were generally accepted as the payers of the revenue. In the rest of the United Provinces and in the Punjab a somewhat similar position was accorded to the representatives of the village communities, the several members of which were accepted as proprietors of their several shares or holdings. In Bombay, Assam, and Burma, on the other hand, and in the greater part of Madras, the intermediaries, where they existed, were very generally disregarded and the assessment was collected from the actual occupant of the soil. In cases, such as are found more particularly in Northern India, where two classes of proprietors, superior and inferior, were recognized, the revenue was imposed on one party or the other as circumstances prompted; but the management of the land generally was left to the inferior proprietor, while the superior was guaranteed as compensation an allowance by deduction from the Government demand. In one way or another a decision was thus ultimately arrived at in each Province that the parties who possessed some specified class of right should be held responsible for the revenue, and the cadastral record necessarily included the names of the persons understood to be in possession of that particular class of right1.

The cadastral record of a District which is drawn up at the time of assessment is thus primarily a fiscal record prepared in order to enable the Government to identify the persons who are under obligations to itself. But from what has been said above it will be seen how closely the matters connected with the preparation of this record are bound up with the general issues of the land-tenure question throughout India. It has therefore been usual to supplement entries in this record, either by additional entries or by a separate record, in such a way as to show to a greater or less extent the existing rights in, and encumbrances on, the land. In the older ryotwāri Provinces, where rights in the soil other than those possessed by the occupants who pay the revenue are not especially prominent, the record has, it is true, retained to a large extent a purely fiscal character. But although the Madras records still concern themselves only with the entries of the responsible revenue payers, arrangements have recently been made in Bombay for starting a supplementary register for the record of mortgages, sub-tenancies, and the like. In Bengal, although the preparation of a complete record of tenants' rights has hitherto been mainly confined to Bihār and to certain temporarily settled tracts elsewhere, the revenue roll is supplemented by registers in which mortgagees and certain classes of tenure-holders and farmers are entitled to record their rights. In Northern India, the Central Provinces, and the ryotwāri Provinces of Assam and Burma the cadastral record is of a much more complete character, setting forth all classes of proprietary or tenant right, and all the more common forms of farm and mortgage; and the preparation of the record has hitherto constituted a very important part of the Settlement Officer's duties. In the greater part of British India therefore the cadastral record forms a valuable compendium of the existing titles in, and encumbrances on, each parcel of the soil; and in most of the Provinces where it is prepared in the form of a record of rights, this record is given by law a presumptive force, being held to be correct until the contrary is proved.

A map and cadastral record has of necessity to be prepared when a District is brought under assessment; and, for fiscal purposes, the changes which occur between one assessment and another must to some extent be reproduced in the record. In the ryotwāri areas, where the number of revenue-payers is large, and where the entries have as a rule been confined to a record of revenue-payers, the registers are usually corrected with some care from year to year. But in the zamīndāri Provinces, where the fiscal objects of the register were comparatively limited and it consisted generally of a record of rights in which the Government was not directly interested, there was in former times a tendency to leave the register untouched between one assessment and another, that is to say, for periods varying from twenty to thirty years. This neglect entailed grave inconveniences in two directions. It became necessary at each new assessment to prepare the record again from the beginning at a considerable expenditure of time and trouble both to Government and to the people; and the record itself, after a few years from the date of its completion, no longer represented with any accuracy the existing state of rights in the land. Arrangements have accordingly been made during the last twenty years under which the records are now in most Provinces revised either annually or at short intervals in such a way as to maintain intact and up to date the accuracy of the initial record; the staff to which the work is entrusted has been very greatly improved; the parties interested are as a rule required under legal penalties to report all changes in rights; and in more than one Province the revised editions of the record thus maintained by local agency are given the same presumptive authority at law as the initial record itself. There is therefore gradually being built up in India a very extensive and complete system of registration of title by public entry. In the case of the more important transfers of proprietary right the revenue record is supplemented by the registration of assurances which is effected under the general Registration law; but for the vast majority of persons connected with the land in India the revenue record forms the only evidence of title, and a large number of transfers are attested in no other way than by entry in this record. There has thus, out of the fiscal necessities of the Government, been developed the nucleus of a system of public transfer, which, as Sir Henry Maine has said, ‘is now the system of the whole civilized world, except England and the countries under the influence of English jurisprudence’; and the revenue records over a large part of India combine, like the ‘cadastre’ of the Continent, the functions of a fiscal memorandum with those of a proprietary register.

II. The Assessment of the Land Revenue

As already noted, the original custom of native governments throughout the greater part of India was to take as land revenue a share of the produce in kind. Revenue is indeed still levied in this manner in many of the Native States, and there are large tracts of India where rents are taken by landlords in no other way. The grave inconveniences of such a system of revenue collection—the large staff of officials required; the troublesome storage of grain; and the openings for peculation, harassment, and oppression—can however be easily understood. The system was, therefore, under native rule gradually subjected to various modifications. Sometimes the presence of a government official on the threshing-floor was dispensed with by having the out-turn of the standing crop and the government share estimated beforehand. Sometimes the grain was not actually removed by the government, but was compulsorily sold to the landholder and the value realized by the government in cash. And sometimes, especially when the produce was not easy to divide into shares, the standing crop was assessed at cash rates per bīgha1. The great reforms of Akbar's time carried the process a step farther—a calculation of the average cash value of the state's share of each crop for a series of years, was worked out per bīgha, and the landholders were given an opportunity of paying their revenue in cash at bīgha rates, which were fixed for a definite period, without fear of enhancement. A great deal of the revenue, especially in the outlying provinces, was still realized in kind, but the accounts were all kept in the cash equivalents; and as time went on, it became customary, while maintaining cash bīgha rates as a standard, to lease the collection for villages and other larger areas to middlemen who paid to the state fixed lump sums per annum. The rates on which these lump assessments were based were, however, gradually lost sight of or manipulated for the benefit of the contractors, and the contract assessments themselves were gradually forced up as the necessities of the government increased. After the advent of British rule the principle of a contract assessment in cash was maintained, and the revenue was assessed in this form either on the individual fields, as in the South, or on the larger estates or villages, as in Northern India. At the same time the old system was revived, by which the contract was fixed for a series of years instead of being liable to capricious alteration. The occupant in Southern India might, it is true, throw up his holding and so escape the assessment imposed upon it, whereas no such option rested with the landowners in the North; but the principle of assessment was in either case the same : namely, that the revenue should be realized in the form of a fixed annual payment in cash, the amount of which should be subject to no alteration during a prescribed term of years. And this fixed assessment in cash may be looked on as the standard characteristic of the land revenue of modern India.

The solid advantages connected with fixity of the demand, the opportunities it afforded for the extension and improvement of agriculture, and the immense relief which it usually provided after the harassing variations of pre-annexation days, have made the Indian Government somewhat chary of allowing departures from the principle of a fixed revenue. In some tracts, indeed, as in Lower Burma, the assessment was imposed mainly by means of a fluctuating cash rate on the area cropped ; and in certain special tracts elsewhere, especially those affected by river inundation or by irrigation from tanks or canals, assessments of a wholly or partially fluctuating nature were introduced ; but as a general rule the system of fixed assessments was for many years very jealously maintained. It has always been recognized that a fixed demand, however light in itself, is apt to press hard on small holders in bad years in areas where the produce varies considerably from harvest to harvest; but it is only of late that, owing to the gradual improvement of the village agency and the increased accuracy of agricultural statistics, it has been found possible to modify to any adequate extent the rigidity of fixed assessments by apportioning the demand of each year more closely to the out-turn from which that demand has to be paid. It is now not uncommonly the custom, as will be further noted below, to postpone or remit in bad seasons a portion or the whole of the fixed revenue in tracts affected by agricultural distress; and there is also a growing tendency in some Provinces to substitute for the fixed demand formerly in use some system of fluctuating assessment which usually takes the form of acreage rates on the areas sown or matured in each harvest, the rates so imposed being, like the lump assessments above described, unchangeable for a period of years. The assessment of fluctuating revenue has no doubt the disadvantage that it can only be carried out by a fairly well-trained staff, and that it subjects the landholders to a certain amount of annoyance and extortion at the hands of the lower grades of government officials; but it has also the great merit of apportioning the demand as nearly as possible to the varying out-turn of the land. Before extending the system to any new tract, its merits and demerits are closely weighed in connexion with the special characteristics of the tract under consideration; but, as above noted, the general tendency at the present day is to make a more liberal use of the fluctuating principle than had hitherto been customary.

It being premised, then, that the revenue, whether fixed or otherwise, is levied by means of a cash demand on each unit assessed, the next step is to note the considerations affecting the incidence of the cash demand thus imposed. Under native rule the assessments were usually represented either directly or indirectly as a certain fraction of the gross produce. What the fraction was and on what system it was based, it is impossible in any general way to record. The ancient textbook known as the Laws of Manu allows the king to take in kind one-twelfth to one-eighth, or if necessary one-fourth1; and these fractions, though scarcely to be accepted as serious historical data, are interesting as showing what the Brāhman authors of a pre-Muhammadan treatise looked upon as suitable shares for land revenue. The cash assessments of Akbar, according to the contemporary Ain-i-Akbarī, represented a third and, according to later authorities, a quarter of the produce2. The Hindu Rājās in the greater part of Madras, according to Sir Thomas Munro, took shares varying from two to three-fifths, and the Marāthā rulers in Western India are said to have exacted not less than one-half3. Similar figures can be quoted on similar authority for other parts of India; and though they may not have been based on strictly mathematical data and may not have been applicable as they stand to the whole area of the tracts in question, yet they serve to illustrate the kind of standard to which the native revenues of pre-annexation days tended to approximate1. The question owes its interest mainly to the fact that (although the net 'assets' form in some ways a better ground of comparison) it has been customary to compare the incidence of the native and British assessments by contrasting the shares of the gross produce usually taken by either; and in this connexion the revenue totals have of late years been sometimes contrasted with estimates of the value of the gross produce. The Famine Commission of 1878–80 made a rough calculation which shows that the revenue of the various Provinces ranged from about 4 to 8 per cent. of the gross produce2. If the average out-turns accepted in 1878 be applied to the areas given by the Famine Commission of 1898, the total land revenue of India may be said to represent 5½ per cent. of the gross produce. Further and more accurate calculations made by the Famine Commission of 1900–1 show that in the Central Provinces the incidence of the present land revenue is less than 4 per cent. of the average value of the produce; that in Berār it is about 7 per cent., in Ajmer 10 per cent., in the greater part of the Punjab 7 per cent., in the Deccan something over 7 per cent., and in Gujarāt 20 per cent. Calculations recently made for Madras show that the land revenue and water-rates of the ryotwāri areas of that Presidency (excluding Malabar and South Kanara) fall at an average of about 10 per cent. on the gross produce. Even allowing for a margin of error in these figures, it is therefore clear that the Indian Government now takes a very much lower share of the gross produce than was customary in pre-British days.

It is however only for statistical purposes that the present land revenue is attempted to be shown in terms of the gross produce. Economists have long ago recognized the injustice (in other than backward tracts) of land revenue assessments fixed at a uniform fraction of the gross produce, and the assessment of the revenue on this principle is now practically as obsolete in British India as in most of the civilized states of Europe. Except in Bombay, where the assessment is not fixed in terms of the produce at all, the revenue throughout India is assessed so as to represent a share not of the gross, but of the net produce. The meaning of the term ‘ net produce ’ or ‘ net assets ’ as employed for the purposes of assessment varies, it is true, in different parts of India1. In Northern India and in the Central Provinces it represents the rent, when rent is paid, or that portion of the gross produce which would, if the land were rented, be taken by the landlord; and in these Provinces it is therefore practically the equivalent of the actual or hypothetical rental. In Madras and Lower Burma, on the other hand, where Government deals as a rule direct with the cultivator, the net produce is the difference between the assumed value of the gross produce and a very liberal estimate of the cost incurred in raising and disposing of the crop (see p. 219). Speaking generally, therefore, the ‘ net assets ’ represent a higher share of the produce in Burma and Madras, where no middleman intervenes between the peasant proprietor and the Government, than they do in Northern India. In the two former Provinces they are the cultivating ‘ assets,’ in the latter the proprietary ‘ assets2.’ But the principle throughout is that the revenue should be based on facts or calculations representing not the gross but the net production of the land to the payer, and it is in the correct ascertainment of this net production that the crucial feature of the assessment from a statistical point of view consists. It has been thought advisable therefore to record briefly the standard methods of procedure adopted in the larger Provinces for the ascertainment of the net incomings on which the assessment is imposed.

In the United Provinces the object aimed at is the ascertainment of the actual renting value of the land to be assessed. The chief guide adopted in the Province of Agra (and the system in Oudh is on somewhat similar lines) is the recorded cash rental of lands under tenants, and the rents so recorded are accepted as the basis for the assessment of such lands unless they are found by inquiry to be unstable, grossly inadequate, or fraudulent. In order to test the recorded rents, the area to be assessed is divided into circles and the soils in each circle are classified; the incidence of the recorded rents for each class of soil (after eliminating fraudulent, inadequate, privileged, nominal, fancy, and other abnormal rents) is then ascertained and compared with the rates of rent for each soil which are generally admitted to be in force. The results of the comparison are further tested by an examination of the statistics regarding population, crops, size of holdings, and so forth, and ultimately a standard set of rent rates is worked out. These standard rates are, with some modifications, utilized for the valuation of lands other than those which pay cash rents, that is, for determining the rental which such lands could fairly pay if so rented. By combining the rental valuations for rented and non-rented lands, a figure is obtained which represents the net ‘assets’ on which the assessment is levied.

In the Central Provinces, also, the actual cash rents are accepted as the net ‘assets,’ but as the rents are, generally speaking, revised by the Government at the time of assessment, the question resolves itself into the ascertainment of a suitable cash rental. To accomplish this, the comparative value of the various soils is ascertained, and the result is recorded in terms of a common unit, known as the ‘soil unit.’ The incidence of the existing rent on this unit in each village is then checked by a comparison with the incidence in other villages and by an examination of past enhancements and the rise of prices; and on these considerations is based a standard unit for each group of villages. This unit is then modified for each village according to local circumstances; and when the village unit is applied to the various soil-areas contained in each holding of the village, the result represents for each holding, and for the aggregate of holdings, the standard rental which could fairly be demanded. This is modified where necessary in each holding with reference to the present rent and any other special circumstances; and the sum of the rents actually fixed represents the amount which the landlord is legally entitled to claim, and constitutes the accepted ‘assets’ for the purpose of revenue assessment.

In the Punjab, where fair specimens of cash rents are not common and where only half the area is under tenants, the calculation of the ‘assets’ is somewhat more difficult. The area to be assessed is divided into circles and the cultivation in each circle is classified, either according to the soil or the mode of irrigation, or both. The area of crops grown in each class is ascertained, an average rate of out-turn is estimated, and the value of the result is worked out by applying certain accepted average prices. The figure thus arrived at represents the value of the gross out-turn. From this (or from the produce before valuation) are deducted the items, such as fodder crops, payments to village menials, &c., which are in practice deducted before division of the grain-heap; to the remainder is applied a percentage representing the average rate of grain rent recorded, and the result is (subject to certain minor tests) accepted as the value of the net ‘assets.’

In Madras the process is up to a certain point similar to that employed in the Punjab, though differing in details. The soils are carefully classified and grouped, and for each of the classes and sub-classes a ‘grain value’ is adopted, which represents the normal gross produce per acre of the chief food-grains. The gross produce is then converted into money by the application of a commutation rate, representing something less than the average prices of the preceding twenty years. From this sum a deduction of 10 to 27 per cent. is made for merchants’ profits and distance from markets, and another deduction of from 6¼ to 25 per cent. to allow for vicissitudes of seasons and unprofitable patches of soil. From the amount as thus reduced is subtracted the estimated cost of cultivation as fixed for each class of soil: that is to say, the cost of cattle, implements, and seed, the wages of labour, and the cost of transplanting and manure; and the balance after subtracting these is taken (subject to further reductions made for relative inferiority by grading villages and irrigation sources) as the net produce on which the assessment is to be based.

In Lower Burma the system followed is similar to, but simpler than, that of Madras. Homogeneous tracts are framed, soils are classified, out-turns are arrived at, and local harvest prices are applied to the result, so as to evolve the value of the gross produce per acre on each class of soil. From this is deducted the average cost of cultivation, including hire of cattle, cost of seed, depreciation of implements and farming stock, and the hire of such labour as is employed to supplement that of the cultivator and his family.

It will be seen from these brief abstracts that the calculation of the net ‘assets’ or net produce is (as land valuations always must be) based to a certain degree on hypotheses. The uncertainty is reduced to a minimum in the United Provinces, where the recorded cash rental is largely followed; but where there are only rents in kind, the calculation involves certain assumptions regarding the out-turn and the value of the produce which must always be subject to error, and in Provinces like Madras and Lower Burma, where the rents are too few to be employed as standards, another source of possible error is introduced in the shape of the deduction for costs of cultivation. These uncertainties are accepted as inevitable, and the process of land assessment nowhere professes to be an exact science; but in order to be on the safe side it has been usual— at least during the last half-century—to prepare the various estimates with considerable caution, and the prevailing tone in the calculations on which the assessments are based is one of extreme—and perhaps from a purely statistical point of view excessive—moderation.

As has been already noted, the state in India has from time immemorial held itself entitled to appropriate a share of the produce of the soil for the public use. When a native ruler took a share of the gross produce, it will generally be found that the share taken varied according to the soil, the position, the manner of cultivation, and other considerations; this being so, the cost of production was in reality an important factor in the determination of the varying shares taken by the state, and these shares, though expressed in terms of the gross produce, were thus often the equivalent of a fairly uniform share of the net produce. What share of the net produce was at any given time and place absorbed by the revenue was a matter which depended largely on the necessities of the ruler, and regarding which nothing like statistical information of a general character is forthcoming. The maxim of a Muhammadan lawyer is sometimes quoted to the effect that 'there shall be left for every man who cultivates his lands as much as he requires for his own support till the next crop be reaped, and that of his family, and for seed. This much shall be left to him: what remains is land tax and shall go to the public treasury1.' In other words, the revenue of the Mughal period was accepted as in theory absorbing the whole of the economic rent, and to a large extent this was the actual practice prevailing in India before British annexation2. The history of the growth of property in India shows that in the days immediately preceding British rule the intermediary between the cultivator and the Government was seldom permitted to retain in his hands any considerable portion of the net 'assets' beyond what was requisite to cover the costs of collection. Under the native assessments, and the earlier British assessments which were based upon them, land had little or no value in the market, sales and mortgages were exceedingly rare, and the intermediaries of various kinds, and even the actual cultivators, very often abandoned their rights in the soil rather than accept responsibility for the payment of the revenue. In 1793 the previous practice in Bengal was declared by the Government to have been that ‘after deducting the expenses of collection, ten-elevenths were usually considered as the right of the public and the remainder the share of the landholder,’ and it was ostensibly on this basis that the permanent assessment of Bengal was made in the same year.

In Bengal, where the assessment itself has remained unchanged, its relation to the rental has now fallen, owing to the rise of prices and the development of the country, from over 90 per cent. to something less than 25 per cent. In the other Provinces where the assessment is subject to periodical revision, it has been the policy of the Government to reduce gradually the proportion borne by the assessment to the net ‘assets,’ and this is nowhere better exemplified than in the Province of Agra, where the standards of assessment at various periods during the last century have been:—

TBL-026 · FOLIO 259Diplomatic CSV · Tidy CSV
In 181290 per cent. of the net ‘assets.’
182280„„„
183272½„„„
184966⅔„„„
185550„„„

In 1885 a further reduction was made by an alteration in the definition of the ‘net assets’: these had previously been held to represent the average annual ‘assets’ which the proprietors were likely to obtain during the coming term of settlement, but since 1885 they have been restricted to the actual ‘assets’ at the time of settlement. And the effect of recent rent legislation is to cause a still further reduction, for by obstructing enhancements it preserves in the hands of the tenants a part of the increased income which would otherwise become assessable to Government revenue. Another notable example of the gradual reduction of the proportion claimed by Government is furnished by the case of Orissa. There the standard was in 1822 declared to be 83·3 per cent. of the ‘assets’: in 1833 it was lowered to 70–75 per cent.: in 1840 to 65 per cent., with a permissive reduction to 60 per cent.: while at the resettlement lately concluded it has been brought down to 54 per cent.

Economists must recognize that whatever arithmetical standards may be adopted for guidance in such matters, it is impossible to carry out in actual practice an assessment of revenue in terms of the ‘assets’ which can be represented by precise mathematical fractions, and the Indian Governments have not tied themselves down by legislation to the adoption of any definite share of the ‘assets’ as their limit of revenue demand. They have, however, in their instructions to the assessing officers, laid down certain approximate standards, and in practice a certain amount of variation from those standards is permitted. In the United Provinces the normal standard is 50 per cent., but deviations for good reasons are allowed, assessments of over 55 or less than 45 per cent. alone requiring special sanction ; and the actual share taken averages less than 50 per cent. in the Province of Agra and less than 47 per cent. in Oudh. In the Central Provinces, which have been for a shorter period under British rule, the standard prescribed is from 50 to 60 per cent., assessments up to 65 per cent. being allowed only where the existing revenue already exceeds this proportion of the newly ascertained net ‘assets’ and has been paid without difficulty : and the average percentages taken in the Districts recently settled vary from 47 to 59. In the Punjab 50 per cent. is regarded as the maximum, and in most Districts the assessments are much below this. In Madras also the prescribed maximum is 50 per cent., and the rates actually taken on the net produce, depreciated for settlement purposes in a manner highly favourable to the ryot (page 219), are on the whole considerably below this figure. In Lower Burma, again, 50 per cent. is nominally the maximum, but for the present a provisional standard of 25 per cent. is accepted. For India as a whole, therefore, it is a fair generalization to say that the standard share of the calculated net ‘assets’ or produce to be taken by the Government for public use is approximately one-half ; but the instances above given will show not only that the calculations of the net produce are usually very moderate, but also that the share actually taken is more frequently below than above one-half1.

In the last few paragraphs an attempt has been made to describe in some detail the statistical basis of the assessments in India, as it is chiefly on this basis that the proposals of assessing officers are checked by their superiors, and as it constitutes the only fair ground of comparison between the assessments of one tract and those of another. It would, however, be a mistake to suppose that the demand on the individual revenue-payer represents a rigid adherence to a standard share of the net produce calculated on purely statistical data. It is unusual to apply, as they stand, the rates obtained by the statistical calculation of the net ‘assets.’ The continuity of the assessment, for instance, has to some extent to be maintained by a comparison with the existing revenue. The selling value of land, though it is not accepted in India, as in some other countries, as the basis of the land tax, is recognized as a useful supplementary test. In Provinces where rents are not sufficiently common to be themselves accepted as the net ‘assets,’ the existing rentals are often used for comparison with the calculated ‘assets.’ Other practical considerations are also brought under review, such as the size of the holdings, the general condition of the people, the increase or decrease of population, the ease or difficulty with which the past revenue has been collected, and so forth. On these and similar considerations a set of rates, separate from and generally lower than those deducible from the net ‘assets’ data, is prescribed for application to the various soils and circles under assessment.

In Provinces like Lower Burma, where the soils are fairly uniform and the assessment is practically fluctuating, these rates are then applied as they stand: but in the zamīndāri Provinces of Northern India there are further safeguards in their application. The rates are sanctioned for a tract or group of estates so as to give a certain revenue for that tract or group; and the assessing officer, in distributing the sum over the estates, is at liberty within certain limits to exceed or go below the figure which each estate would pay according to the strict rates. The process of determining the assessment is in fact, as Thomason described it, ‘not one of arithmetical calculation but of judgement and sound discretion.’ When it comes to the distribution of the revenue over the holdings within the village or estate, the individual landholders are further protected from a mere doctrinaire application of rates by the fact that the internal distribution of the assessment is as a rule left by the Government to the landholders to carry out by private arrangement1. In one way or another therefore considerable play is given to the working of local considerations and to the judgement and discretion of the assessing officer in the actual application of the standard rates, and it would be a mistake to imagine that the revenue is imposed in any part of India with slavish adherence to purely arithmetical data.

The principle of adhering to what are known as ‘general considerations’ rather than to statistical calculations of the net produce, though it also lies at the root of the Central Provinces system described on page 218 above, constitutes the distinguishing and essential feature of the assessments of Bombay. Under the Bombay system the comparative values of the different ‘survey numbers’ are first ascertained by a minute and careful soil classification, the results of which are expressed in fractions of a rupee, sixteen annas representing generally the best soil. The area to be assessed is then divided into homogeneous groups according to physical and economic characteristics; and for each group a careful review is made of the past revenue history, prices, selling and letting and mortgage value of land, vicissitudes of season, and every other relevant fact indicating the incidence of the previous assessment and the economic conditions of the tract. The general result to be attained by the revision of assessment being decided on, rates are determined which, when applied to each field by means of the classification, will bring about that result, higher rates being imposed on those groups which have the greater advantages and lower rates on the less favourably situated groups. The rates being applied automatically to the various fields according to their relative valuation, an assessment is arrived at for each field, and this constitutes the demand for the ensuing period of settlement.

The revenue assessable, whether on the basis of the net ‘assets’ or otherwise, is in most parts of India subject to certain deductions, the more important of which may be here noticed: namely, (1) deductions to avoid sudden enhancement, (2) deductions to protect private improvements, and (3) deductions by way of assignment.

In consequence of the long term for which settlements are usually made, it is inevitable that new assessments, however accurately and fairly fixed in view of increased resources, must involve in some individual cases a considerable increase in the Government demand and a corresponding sudden decrease in the share of the revenue-payer. The hardship thus occasioned has been usually recognized, and arrangements have in most Provinces been made to mitigate it, either by limiting the degree of enhancement or by making the enhancements progressive over a series of years, or by both means. The present condition of the question is described as follows in a recent state paper published by the Government of India1:—

‘The mitigation of a large enhancement by spreading its imposition over a term of years has been a recognized feature in the settlement procedure of Upper India for a long time past, but has not till recently been brought systematically into practice. In 1895 the Government of India, with the concurrence of the Secretary of State, drew general attention to the advisability of making larger use of progressive enhancements. In the North-Western Provinces very complete effect has already been given to this principle. Similar rules have recently been used in the resettlement of the Seoni District in the Central Provinces, and the expediency will now be considered of prescribing it for general guidance in those Provinces. The rules on this subject contained in the Bengal Settlement Code are of particular application to ryots and tenure-holders; but they admit the use of progressive assessments in the Orissa settlements, though they lay down no definite scheme of progression, and, as a matter of fact, progressive assessments were most liberally granted in those settlements at a loss to the state of nearly 8 lakhs of rupees. In the Punjab, the use of progressive assessments has been discouraged on the ground that, though an appropriate means of easing an enhancement to a large landholder, they are not suitable to the circumstances of the petty proprietors who hold a very large proportion of the land in that Province. Large increases in the demand have been commonly avoided by under-assessment. But it seems open to question whether an expedient which has proved serviceable in other parts of India might not be usefully adopted in the Punjab, and the point will be considered, though the effect of progressive assessments in this Province would be to raise, not to lower, the Government revenue. Turning now to ryotwāri settlements, a rule of the Madras Settlement Code limits to 25 per cent. the enhancement which may be imposed at once, the balance being imposed by annual instalments, each not exceeding 12½ per cent. on the original assessment. This gives a ryot six years in which to accommodate himself to the doubling of his assessment. In the Bombay Presidency also the levy of substantial enhancements is distributed over a term of years, and the maximum enhancement may not exceed double the former amount. Deviations from these rules have, however, apparently been permitted. The procedure of ryotwāri settlements renders it difficult for an assessing officer to pay close regard to the circumstances of individuals in framing his proposals; and there is, therefore, the greater need of general rules to obviate hardships in particular cases, even if it be conceded that men who cultivate their own land can support a heavier percentage enhancement than those who subsist upon rental receipts. The question is one that calls for, and will receive, further consideration.’

The practice of exempting from assessment the increased income due to such improvements as have been made by private enterprise is one which seldom finds a place in forms of taxation other than that relating to land, and even in respect of land it was a principle only imperfectly recognized by the native rulers of India. But, although other forms of profits are duly taxed as they arise, the principle that land improvements should be exempted, either permanently or temporarily, from taxation has now been accepted by the Indian Government, and the degree to which the exemption is carried in the various Provinces will be understood from the following extract from the state paper above quoted, which explains that the principle in question is now provided for by definite rules

‘culminating, in the case of the Bombay Presidency, in legal enactments which secure to the cultivator in perpetuity the whole of the profit arising not only from such irrigation works as private wells or tanks but from the minor improvements which would count for an increase in assessment under a system of reclassification of the soil. The Madras ryots have a recognized right to enjoy for ever the fruit of their improvements, and the exemption of wells, irrigation channels, and tanks which are private property is provided for by executive orders. Minor improvements are also protected, as in Bombay, by the permanent recognition of a land classification once fairly effected. In zamīndāri Provinces, where the revenue is temporarily assessed on estates as a whole, and not on each particular plot of land composing them, the state has not similarly surrendered its right to all share in improvements in which the capacity of the soil plays a part with the industry or outlay of the cultivator. But the principle followed has been that additional assessments should not be imposed on these grounds until the private labour or capital expended upon them has had time to reap a remunerative return. In the Punjab and Bengal the term of exemption has been fixed, without reference to the term of settlement, at twenty years for masonry wells, five years for canal distributaries, and ten years for other irrigation works. In the North-Western Provinces and the Central Provinces, irrigation works not constructed by Government are freed for the term of settlement next following their construction, the average period of exemption being forty-five years in the former and thirty years in the latter Provinces1.

There remains a further form of deduction from the land revenue receipts which, unlike the two above described, is peculiarly oriental in its origin : namely, that due to the practice of assigning or alienating to individuals the right of Government to collect its demand from the land or of permitting individuals to hold their land wholly or partially free of revenue. As each Province was taken over from the preceding native rulers an immense number of such grants were found in existence, and an immense amount of time was spent in the early days of British rule in determining to what extent and on what terms such grants should be continued. New grants of this character are still sometimes made ; but the majority of those now in existence are an inheritance from native rule. The grants owe their existence to many causes, but they mostly have their origin in service of some kind done to Government or the community ; and although the service is often of an intangible nature or has now become ancient history, the maintenance of these grants is recognized as a fair charge on the state.

There are per contra certain additions to the land revenue demand, known comprehensively as ‘cesses,’ which, though not classed as land revenue, are, like the centimes additionnels in France, usually treated of along with it. These cesses are of two main classes : namely, (1) the local rates which are levied for certain local objects, such as roads, schools, dispensaries, and the like, which fall within the purview of the local boards ; and (2) the sums payable for the remuneration of village officers, such as the headman, the accountant, and the watchman.

The local rates in the zamīndāri Provinces are assessed on the rental, and the rate in force in Bengal is 6 1/4 per cent., in the Punjab 5.2 per cent., and in the Province of Agra 6 per cent. ; but in the last-named area two-fifths of the proceeds are devoted to the maintenance of a village watch which is usually provided for by a separate contribution. In the ryotwāri Provinces the cess is taken on the revenue ; and the rate in Lower Burma is 10 per cent., in Assam 8.3 per cent., and in Madras and Bombay 6 1/4 per cent.

The village officers were under native rule usually remunerated by grants of land, by payments from the grain-heap at harvest, or by fees collected direct from the people. The watchmen in the Central Provinces and Bombay are still maintained on these lines, but elsewhere they are supported wholly or partly by the proceeds of a cess to which in some Provinces non-agriculturists contribute. The headman, one of whose chief duties is the collection of the revenue, receives a remuneration, which in Northern and Central India takes the form of a commission, generally amounting to 5 per cent., on the revenue which he collects. The accountant, whose duties have been of late years considerably enlarged by the increased attention now paid to the maintenance of agricultural records, is in some Provinces supported with the help of grants from the public revenues, but till 1906 a cess provided at least part of his salary.

The cesses are levied under the authority of various legal enactments and can, subject to the same legislative authority, be altered from time to time at the will of the Government. In determining the land revenue, on the other hand, the Government is accustomed to announce that the amounts or rates imposed will not be altered till the expiry of a certain specified period. The first attempt to fix an assessment for a period extending over more than one year appears to have been that represented by the decennial settlement which was introduced towards the end of the reign of the emperor Akbar. On the expiry of this decennial settlement the acreage rates on which it was based do not seem to have been expressly altered, and indeed these rates appear to have been accepted as the standard or normal assessment for many years afterwards ; but they were merely in force 'until further notice,' and in practice the demand of the later Mughal rulers, though based on the old rates or the lump assessments which they involved, was constantly altered by varying degrees of accretion, either by all-round enhancements, or by manipulation of the unit of measurement, or by the addition of cesses. Anything in the shape of a special contract or privileged assessment was looked on as liable to alteration on the death of the grantor or the grantee, and the ordinary assessments were usually open to modification each year according to the caprice of the ruler or his revenue-farmers.

This system was continued by the British Government in Bengal for some years after the assumption of the revenue administration of that Province in 1765 ; but the vexatious character of these annual assessments was soon recognized, and in 1772 the settlement of Bengal and Bihār was concluded, mostly with revenue-farmers, for a period of five years. A good deal of discussion followed as to the sufficiency of the available information for a long-term settlement and other points ; but ultimately in 1786 the Court of Directors prescribed the commencement of a ten-year assessment, and ordered that on its completion another report should be made which would enable them to decide as to the necessity or otherwise of any further changes.

The decennial settlement was accordingly commenced, and, under Lord Cornwallis's orders, it was announced that the assessment then demanded would be fixed in perpetuity if the Court of Directors agreed. The Directors, after careful consideration, concluded that there was not much hope of improving the information on which the settlement was based and that the evils of the previous system could only be avoided by a permanent arrangement. They hoped also that, by making the revenue demand permanent, they would induce the landlords to make improvements and in many indirect ways benefit the country. Accordingly the assessment was in 1793 declared to be perpetual, and the settlement then made, with some subsequent additions, constitutes what is known as the Permanent Settlement of Bengal. In 1795 the permanent system was extended to the Benares districts (now in the United Provinces), and in 1802 to certain portions of the Madras Presidency. Under these arrangements about five-sixths of the present Province of Bengal, one-eighth of Assam, one-tenth of the United Provinces, and a quarter of Madras, representing in all about one-fifth of the area of British India, are permanently settled. The land revenue throughout this area has now, like the English land tax, none of the characteristics of taxation, and may be said to be at the present day nothing more than a rentcharge, the burden of which has long ago been discounted by the reduced selling price of the land which it affects.

When the greater part of the Province of Agra came under British rule in 1803, it was at first proposed that, after two or three experimental settlements for short terms, the permanent assessment should be extended to this Province also ; but the proposal was ultimately negatived. Thenceforth the system adopted in the older Provinces was to continue the existing settlements for short terms, or without specification of term, until a long-term settlement could be introduced ; while in newly annexed Provinces the short-term settlements necessary in the days immediately following annexation were as soon as possible succeeded by settlements for long periods. It is now therefore the practice in the tracts outside the scope of the permanent settlement to assess each District or other similar unit of area for a period of years ; and as the assessments were not originally introduced at a single date throughout India or throughout each Province, the result is that the dates for the reassessment of the various Districts are not simultaneous, but range, as it were, in a rough kind of échelon through each recurring period of time. The commonest term of settlement at the present day is one of thirty years—a period first introduced in Bombay in 1838, and thence extended to Madras and the Agra Province, where it has been the standard period for the last half-century. The same principle was followed in an extension of the Orissa settlement in 1867, and in confirming most of the settlements made in the Central Provinces between 1860 and 1870. But it never came into use in the Punjab, where in the greater part of the Province the shorter term of twenty years has been the recognized rule. The general question of the term of settlement was examined in 1895, when it was decided that thirty years should continue to be the ordinary term in Madras, Bombay, and the United Provinces ; that in the Central Provinces twenty years should be the general rule ; and that in the Punjab also twenty years should be the usual term, thirty years being admitted in some cases. In backward tracts, such as Burma and Assam, and in exceptional circumstances, such as exist in Sind, shorter terms are permitted. The reasons for this differentiation are thus explained by the Government of India :—

‘ Where the land is fully cultivated, rents fair, and agricultural production not liable to violent oscillations, it is sufficient if the demands of Government are readjusted once in thirty years, i. e. once in the lifetime of each generation. Where the opposite conditions prevail ; where there are much waste land, low rents, and a fluctuating cultivation ; or again where there is a rapid development of resources owing to the construction of roads, railways, or canals, to an increase of population, or to a rise in prices ; the postponement of resettlement for so long a period is both injurious to the people, who are unequal to the strain of a sharp enhancement, and unjust to the general taxpayer, who is temporarily deprived of the additional revenue to which he has a legitimate claim1.’

The standard period, therefore, to which the greater part of India is gradually tending to conform may be looked upon as thirty years.

The assessments of India are thus divided into two main classes according to the period for which they run: namely, permanent and temporary. Of the total land revenue of India in 1900–1 about 16 per cent. belongs to the former and 84 per cent. to the latter category. The respective merits of the two systems have been a good deal discussed; and suggestions, chiefly of an irresponsible character, have from time to time been put forward for altering the status quo, either by abrogating the permanent settlement in the tracts where it is in force, or by extending it to tracts where it has not hitherto been adopted. On the one hand, it is contended that the Government has lost and is losing an enormous revenue owing to the permanent settlement; that if a temporary settlement were substituted for it, the Government would now be entitled to a revenue of at least 9 crores of rupees per annum from the permanently settled tracts, or double the existing demand; that the indirect benefits claimed for the permanent assessment far from compensate for this direct and material loss; that the relative incidence of the revenue, even if approximately fair in 1793, has become grossly unequal in the course of the last hundred years; that the rest of India has to be taxed more heavily in order that Bengali landlords may continue to enjoy a purely unearned increment; that if India had been under representative government, the permanent settlement would long ago have been abolished; and that the state is justified in rescinding on behalf of the general taxpayer an obligation which has proved to be contrary to the public weal. On the other side, it is urged that if a permanent settlement were extended to the temporarily settled tracts, the expense and harassment of the present assessment operations would be avoided; that there would be no temptation to abandon cultivation on the approach of a revision of settlement in order to reduce the ostensible ‘assets’; that the accumulation and investment of capital would be directly encouraged; that the people would lead a fuller and more contented life; and that the indirect benefit thus accruing in the future would more than compensate for the immediate loss of revenue.

It is not now usual to consider either of these suggestions as within the bounds of practical politics, but the latter suggestion, namely, the extension of the permanent system to temporarily settled areas, was under serious discussion as lately as in 1861–83. The general principle that a permanent settlement was advisable throughout India was indeed accepted by the Secretary of State, Sir Charles Wood, in 1862, and orders were given by him for its gradual introduction. But the practical difficulties connected with the proposal soon became apparent; the possibilities of increased cultivation were found to be much larger than were at first anticipated, the failings of the system in Bengal became more prominent, the unproductive use of rent by the majority of landlords excited greater comment, and the rise in the prices of produce and the fall in the value of silver gradually assumed greater dimensions. So that the scheme, after being shelved for many years, was finally negatived by Lord Kimberley in 1883.

Although, however, the proposal for a general permanent assessment was quashed, the discussions to which the proposal gave rise were fruitful in other ways. An effort was made to arrive at a scheme by which, while retaining the principle that assessments were liable to revision, the public might escape as far as possible the expense and harassment which the long inquiries preliminary to assessment had hitherto entailed; and a plan was discussed by which, when land had once been given a fair and equitable assessment, the initial revenue thus arrived at should not be altered in future except in order to meet (1) an extension of cultivation, (2) an increase of produce due to improvements made by the state, and (3) a rise of prices. The object of the plan was to avoid the detailed and troublesome inquiries which had hitherto preceded the fixing of the assessment, and, while avoiding the evils of a permanent settlement, to adopt what has been described as a 'permanent system of settlement.' But the practical objections to the adoption in its entirety of a uniform scheme of this nature were found to be considerable. No provision was made, for instance, for the taxation of such improvements as were only indirectly due to the action of the state, and which resulted from the increase of population, the development of the country, the introduction of new staples, and so forth. The guidance of price statistics was also found to be partial and fallacious; different villages would be differently affected by general variations in price; and in the Provinces where the revenue was based on the cash rental, it was found unsafe to assume that the rentals varied directly with the prices of produce.

The scheme in question is not therefore accepted at the present day as the sole basis of assessment in any Province; but considerable advances have recently been made in other ways towards attaining the object which the scheme was devised to secure, namely, the reduction of the tedious and harassing investigations hitherto entailed by a revision of assessment. In most parts of India it was formerly the custom at each revision of assessment to prepare a fresh survey of the land, a fresh classification of the soils, and in some Provinces a fresh record of the rights of the landholders; the old measurements, classifications, and records of thirty or twenty years before being set aside as inaccurate or obsolete. As the surveys have improved in quality, however, it has been found possible to do away to a large extent with the practice of complete revision. In Bombay a classification of soil made for the second time, or once approved as final, is by law accepted as incapable of revision. In Madras, although no pledge has been given against future revisions of classification, the principle has been accepted that the existing classification, if found to be in the main equitable, shall not, on resettlement, be disturbed. So too in Northern India the great improvements effected in the land records during the last twenty years have enabled the Government to arrange that, when a map and record have once been accepted as sufficiently accurate, they shall be kept up to date and thus be at once available for use at a new assessment. By these means, and by the simplification of the methods of assessment, the period spent over the settlement of a District, which in Northern India used to be six or eight years, has now been reduced to an average of about four years, or even less, and is being still further reduced as the new arrangements come gradually into force1.

Minds accustomed only to European systems are apt to start with a prejudice against a form of revenue which may absorb half the rent, and to advocate its immediate conversion into taxation of the standard type. A reference, however, to the chapter on Indian rents in this Gazetteer2 will show that historically the existence of the land revenue is anterior, and not subsequent, to that of private rents; that it finds its origin and present basis in the immemorial right of the Crown in India to a share of the produce of all land within the state; and that rent, and rent-yielding property in land, owe their existence to the policy, adopted during the last hundred years, of gradually reducing the incidence of the revenue; so that, as a matter of history, the right of the state to the rent is superior to that of the private landholder. Moreover, in view of the fact that a land revenue of the Indian type is practically unknown to European readers, it is advisable to note that in comparison with most known forms of taxation it has merits which should not be overlooked. It is, for instance, the only considerable class of state revenue which (obtained as it is from profits on the staple industry of the country) can be raised without enhancing prices or diminishing general consumption. It approximates more than most forms of taxation to the ‘single tax’ which has, rightly or wrongly, been advocated by some theorists as the ideal type of assessment, and effects, in a measure, that ‘nationalization of the land’ which is not without strenuous supporters. It falls on that part of the produce which goes (or which under ordinary circumstances would go) not to the cultivator, but to the intermediary rent receiver. Even in respect of the latter, it is when permanently fixed a charge the burden of which has long been discounted, and when not fixed in perpetuity it approximates to a charge of this nature in so far as it is fixed for long terms. It is scarcely to be imagined, therefore, that a form of revenue which has these merits, and has besides the great advantage of having been from time immemorial known and accepted by the people in India, should be exchanged in any wholesale manner for other sources of taxation which would as a rule involve more unpopular and inquisitorial methods. That the proportion of the land revenue to the total income of the state is, however, gradually diminishing with the general development of the country will be evident from the following figures :—

TBL-027 · FOLIO 272Diplomatic CSV · Tidy CSV
Proportion borne by the1870–1.1880–1.1890–1.1900–1.
land revenue to :—p. c.p. c.p. c.p. c.
The gross income of the state .39·029·528·023·2
The taxation proper .43·941·640·838·6

III. The Collection of the Land Revenue

The land revenue under native rule was, even more than at the present day, the mainstay of the state finances, and its collection, which (as will be observed below) was often a very serious undertaking, was entrusted to the highest class of officials, and was looked upon as the central function of Government. To this day it is usual throughout the greater part of India to speak of the officer who represents the civil administration of each District, not as the Magistrate, or the Prefect, or the Administrator, but as the ‘Collector’: and the officers in magisterial charge of the subdivisions of a District are similarly known in most parts of India by native titles which indicate their connexion with the recovery of revenue. So too the minor administrative divisions of the country have for the most part been determined from the fiscal point of view, and the chief representatives of the several villages or groups of villages for purposes of police and general administration are in a large part of India the headmen whose primary duty is the collection of the revenue from the individual villagers. These facts may serve to explain how it is that the business of collecting the land revenue, though far less serious than it used to be, still occupies a much larger place in Indian administration than is usually allotted to similar duties in Europe.

Owing partly to the general lack of capital among the agricultural population, the land revenue is seldom recovered by a single annual payment. The usual custom is for a portion to be paid after each harvest, the dates and amounts being fixed to meet the local circumstances of each tract. The dates are usually fixed so that each revenue-payer should have time to realize his produce or his rent before he is called upon to pay the revenue, and yet should be required to pay the revenue as soon as possible after the rent or produce has been realized.

For the recovery of sums not paid by due date the Government has extensive powers conferred by law. In the disturbed period of native rule preceding British occupation the collection of the land revenue was sometimes a hazardous business, large bodies of troops being not infrequently employed and villages in some cases having to be coerced by the aid of artillery. In the early days of British occupation, while the people were still turbulent and the assessments severe and uneven, the most stringent measures were frequently adopted, and processes such as the sale of a defaulting estate or holding to an outsider were employed with a frequency which would not now be tolerated and which furnished at the time a grave cause of discontent among the people. At the present day, although the collection of revenue has for the most part been reduced to a matter of routine, the Government has very properly, in view of the importance of the demand and the vast number of individuals assessed, prescribed that the land revenue shall be regarded as a first charge on the land, and has retained for itself very ample powers of recovery. The procedure in the several Provinces differs somewhat in detail, but the general features are very similar, and it is the almost invariable rule that a defaulter, before being actively proceeded against, should be presented with a writ of demand. On this failing to secure the required payment, it is usual either to proceed against the movable property of the defaulter by attachment followed by sale, or to treat in a similar manner the immovable property which is the subject of the default or other immovable property held by the defaulter, or occasionally to arrest and confine the person of a wilful and stubborn defaulter for a limited time, or (in Northern India) to transfer the defaulter's land to his co-sharers for a period, or to take over the management of the property in default for a limited term. In some Provinces the order in which the various processes shall successively be tried has been laid down ; in others this is left to the discretion of the executive authorities. But, as a rule, the writ of demand or the threat of attachment serves in itself to bring about without further trouble the payment of such sums as are in arrear, and the more serious procedure of sale is seldom resorted to. Where too, as in the greater part of Northern India, the co-sharers of a village are jointly and severally responsible for the revenue, it is becoming more and more uncommon to realize arrears at the ultimate charge of any but the sharer actually in default.

There is one exception to the general system of recovery sketched above : namely, the procedure established in the permanently settled areas of Bengal. The declaration of the permanent settlement in 1793 was very soon followed by regulations prescribing that, if the whole or part of an instalment of revenue was left unpaid at sunset on the day prescribed for payment, the estate concerned should immediately become liable to sale. The original severity of the rule has been gradually modified and the need for its exercise has gradually decreased as the incidence of the assessment became lighter, but the Bengal procedure is still unique in this respect that it applies to the recovery of arrears no other process than that of sale.

In addition to the gradual amelioration of the actual procedure of recovery, there is a further point connected with the collection of the land revenue in which considerable progress has been made of late years. It has already been noticed that in the greater part of India the assessment is a fixed cash demand, representing the revenue fairly payable on an average of a series of seasons. When the fixed assessments were first introduced, the memory in most Provinces of the evils connected with the variable assessments of native rule was still strong, and for many years the fixity of the demand introduced by the British was looked on as an axiom of revenue administration with which it was unwise to interfere. There lay at the root of the system the idea that the landholder in good seasons lays by enough to meet his revenue in bad seasons, and within certain limits this supposition was reasonable. Experience, however, gradually showed that among the smaller landholders and in tracts subject to great vicissitudes of season this was seldom the case, and that even in the richer tracts the hypothesis broke down under the stress of a serious scarcity. In areas, therefore, which are under a fixed assessment, it has been customary of late years in most Provinces to grant with greater freedom, as necessity arises, a postponement of the whole or part of the demand on a harvest, and even, in cases where ultimate recovery would entail real hardship, to allow an absolute remission of the demand. The detailed arrangements under which suspension or remission of land revenue is granted differ considerably in the different Provinces1; but the general aim in all parts of India is the same, and the practice of modifying the fixed collections by reasonable suspensions and remissions is now recognized in most Provinces. A notable example of this tendency was exhibited during the recent famine. In the years 1899–1900 and 1900–1 the collection of Rs. 2,06,37,000 of revenue was suspended in the Districts affected by famine, and at the conclusion of the famine in 1902 no less than Rs. 1,98,23,000 was entirely remitted.

The available figures for the land revenue returns of the Mughal empire are incomplete and not altogether trustworthy, but from the latest investigations of historical students the following appears to be the best approximation we can make to the income from this source of revenue at various periods2 :—

TBL-028 · FOLIO 276Diplomatic CSV · Tidy CSV
Date.Approximate relation of area of portion dealt with to present British India.Total land revenue in rupees, excluding Afghān Provinces.Land revenue of the areas corresponding roughly to the Punjab, the United Provinces, and Bengal.
Rs.Rs.
Akbar, a.d. 1594Three-fifths13,21,36,8317,85,89,108
„ a.d. 1605Three-fifths17,45,00,000*...
Shāh Jahān, a.d. 1628Three-fifths17,50,00,000*...
„ a.d. 1648Three-fourths21,15,00,00012,20,00,000
Aurangzeb, a.d. 1661Three-fourths26,35,12,75216,18,05,020
„ „ 1665 (?)Three-fourths22,06,70,500†11,44,35,500†
„ „ 1695Three-fourths20,12,76,42511,88,01,650
„ „ 1697Four-fifths38,30,39,552‡12,87,69,050‡
„ „ 1700 (?)Four-fifths33,68,16,58412,42,42,674
„ „ 1707Four-fifths29,77,70,87613,15,69,338
Muhammad Shāh, a.d. 1720 (?)Four-fifths29,40,76,776‖10,71,01,533‖
Land revenue demand of 1900–1.All26,25,45,45912,85,36,508

In making comparisons between the Mughal revenue-roll and that of the present day, several considerations have to be borne in mind. The peasants were in Mughal times liable to support a very large militia, the cost of which has been estimated at almost as much as the revenue itself; but this liability appears to have soon fallen into desuetude or to have been enforced only in time of war. The Mughal collections, too, were far more elastic than the British collections have hitherto been, for in good seasons the cultivator was subjected to surcharges of various kinds, and in bad years the demand (which was a severe one) was seldom collected in full. In any case the important point to remember is that the area under cultivation and the population were both much less in Akbar’s time than now, while the purchasing power of the rupee was then twice or thrice as great as that of the corresponding coin at the present day. In the Punjab, the United Provinces, and Bengal, for instance, the assessment of Akbar was calculated to form the equivalent of some 4·8 million tons of wheat, while the present assessment of the same Provinces corresponds approximately to 1·9 million tons1.

The development of the land revenue of British India since the Mutiny is shown by the following figures :—

TBL-029 · FOLIO 277Diplomatic CSV · Tidy CSV
1856-7.1870-1.1880-1.1890-1.1900-1.
Land revenue receipts in lakhs of rupees .17,3019,9621,9124,0526,25

The only territorial expansion of fiscal importance during these years was that represented by the annexation of Upper Burma in 1886; and the increase is mainly due to extension of cultivation and rise of prices, for, as the foregoing pages have shown, the Government share of the net produce is a diminishing rather than an increasing factor.

The general incidence of the land revenue demand for the year 1900–1 in the several Provinces is shown in the following table :—

TBL-030 · FOLIO 277Diplomatic CSV · Tidy CSV
Province.Land revenue assessed per head of population.Incidence per acre of revenue in fully assessed area *.
For total area.For cultivated area.
Rs.a.p.Rs.a.p.Rs.a.p.
Upper Burma .2861251144
Lower Burma .2120204222
Assam .1130133126
Bengal .08110740127
Agra .155121202
Oudh .1361181151
Ajmer .01310711311
Punjab .146089115
Sind .2142253281
Bombay .1041071910
Central Provinces .012110460101
Berār .29101031311
Madras .1951222411
Coorg .20812111411

In India, as a whole, the land revenue in 1900–1 came to Rs. 1–3–7 per head of the population, and the incidence in the fully assessed lands was for the total area R. 0–12–9 per acre, and for the cultivated area Rs. 1–7–8 per acre.

The above sketch is intended to present the outlines of a system of taxation which, as already noted, must of necessity be strange to the ordinary English reader. The Indian land tax is a heritage from preceding native rule, and the present state of its administration is the result of a natural growth which has proceeded on different lines in the different Provinces. Starting with a fairly detailed knowledge of native practice and a few axioms of orthodox political economy, the officers who have built up the present revenue administration of India have independently arrived at results which will, it is believed, compare very favourably with those reached in the contemporary systems of Continental Europe. The Indian arrangements are no doubt still in many respects defective. Some of them are in the experimental stage, and experiments in land revenue, like all experiments connected with land, require long periods in which to mature. There are several questions in which finality cannot yet be said to have been attained, but the whole procedure is by slow degrees developing on broader and more liberal lines than heretofore. To secure an adequate land revenue for the state with the least possible injury to the agricultural classes—this is in its widest form the problem which is every year in one or other of its details taxing the ingenuity and enthusiasm of a large number of the officers of the Government, and it is not unreasonable to expect that as time goes on the problem will be brought nearer and nearer to a satisfactory solution.

Bibliography

Abul Fazl.—Ain-i-Akbarī. Translated from the original Persian : vol. i by H. Blochmann, 1873; vols. ii and iii by Col. H. S. Jarrett, 1891 and 1894 (Calcutta).

The Fifth Report from the Select Committee on the Affairs of the East India Company. Two volumes (London, 1812. Reprinted at Madras, 1883).

Selections from the Revenue Records of the North-Western Provinces, 1818–20 (Calcutta, 1866).

Major-General Sir Thomas Munro.—Selections from his Minutes and other Official Writings, by Sir A. J. Arbuthnot. Two volumes (London, 1881).

The Survey and Settlement Manual, Bombay Presidency (Bombay, 1882).

Report of the Indian Famine Commission, 1880 (London, 1880).

Report of the Indian Famine Commission, 1898 (Simla, 1898).

Report of the Indian Famine Commission, 1901 (Calcutta, 1901).

Statistical Atlas of India (Calcutta, 1895).

Systems of Land Tenure in various Countries—Cobden Club Papers. The article on India is by Sir George Campbell. (London, 1881.)

B. H. Baden-Powell.—The Land Systems of British India. Three volumes (Oxford, 1892).

B. H. Baden-Powell.—A Short Account of the Land Revenue and its Administration in British India, with a Sketch of the Land Tenures (Oxford, 1894).

Indian Land Revenue Policy. A Collection of Papers issued by the Government of India in 1902, the first of which is ‘An authoritative exposition of the principles on which the Land Revenue Administration in India has been based in the past, and by which it will be guided in the future.’ (Calcutta, 1902.)

R. C. Dutt.—Open Letters to Lord Curzon on Famines and Land Assessments in India, to which the above is a reply (London, 1900).

Notes

  1. 1. A somewhat similar system was carried out in the Bombay Deccan by Malik Ambar, the great minister of Ahmadnagar, who died in A.D. 1626. See Baden-Powell's Land Systems of British India, iii. 205.
  2. 1. This is perhaps the most practical distinction between zamīndāri and ryotwāri; other characteristics have from time to time been looked upon as differentiating the two systems, such as the treatment of waste land, the annual revision of the demand, joint and several responsibility, the rights of Government in the soil, and so forth—but it is unnecessary to confuse the reader with discussion of these points. The distinction has its historic origin in the varying degrees in which, in different parts of the country, tribal occupation of territory has superseded the rights of the ruler, or full proprietary right has been granted to the individual. Under zamīndāri tenure the land is held as independent property. Under ryotwāri tenure, it is held of the Crown in a right of occupancy which is, under British rule, both heritable and transferable. Even in the distinction made in the text there is an element of inconsistency, for as time goes on there is a tendency to treat the many small self-cultivating landlords in the village communities of Northern India more and more individually, after the ryotwāri pattern.
  3. 1. The persons thus recognized as proprietors (and also certain classes of tenants) were in almost all parts of India permitted to exercise free rights of transfer by sale, mortgage, gift, &c.; but an extravagant use of these rights tends in many Provinces to the disintegration of the larger properties and the transfer of land from the agricultural to the non-agricultural classes. Efforts have been made, from 1862 onwards, to preserve historic estates from the break-up with which they are thus threatened by means of Acts permitting temporary management by Government, transmission by primogeniture, and other remedies; and legislation of this kind is now in force in Sind, in various parts of Bombay, in Oudh, the Punjab, Bundelkhand, Ajmer, Chotā Nāgpur, and in the zamīndāri areas of Madras. Steps have also been recently taken to curtail the right of transfer to money-lenders and other non-agriculturists in the Punjab and in Bundelkhand: and similar restrictions have been imposed on certain classes of tenants in the Central Provinces and on the new occupants of certain vacant areas in Bombay.
  4. 1. A bīgha is, generally speaking, five-eighths of an acre.
  5. 1. Mānava-Dharma-Shāstra, vii. 130, x. 118–20.
  6. 2. Ain-i-Akbarī, Jarrett, ii. 63; Grant's Political Survey of the Northern Circārs, 5th Report, Madras ed., ii. 165.
  7. 3. Arbuthnot's Munro, i. 248; Bombay Government letter 281 of March 30, 1901; Memo. para. 1, cf. Harington's Analysis, iii. 303, 324, 346.
  8. 1. Such rates must be interpreted as indicating the relation of the revenue over a whole village or tract to the produce of that village or tract, and not as implying necessarily a uniform rate over all the holdings or fields. Where grain-rents prevail, the rates for the fields or holdings usually differ according to the soil, class of cultivation, &c., and the revenue of native sovereigns must as a rule have been differentiated ultimately in the same way. One of the difficulties connected with the existing descriptions of Akbar's settlement lies in the fact that the exaction of a uniform share of the gross produce from every cultivated acre throughout the empire is an idea almost inconceivable to any one acquainted with Indian rents in kind.
  9. 2. Page 112, Part II of Commission's Report.
  10. 1. The expression ‘ net produce ’ is that ordinarily employed in ryotwāri, and ‘ net assets ’ in zamīndāri, Provinces.
  11. 2. Technically speaking there are three main bases on which the land can be assessed: namely, the selling value, the renting value, and the value of the net produce. The first is believed to be the system generally followed in America, the second in England, and the third on the Continent of Europe. Although the second and third systems are thus technically distinct, it is convenient in dealing with Indian arrangements to look on them as variations of the same principle, especially as the two must in practice be to some extent combined whenever the rent is taken in grain.
  12. 1. Report on the settlement of the Begum Sumroo's Pergunnahs, 1840. North-Western Provinces Revenue Reporter, vol. ii, 1874.
  13. 2. See, for instance, Holt Mackenzie's Memo. of July 1, 1819, paragraphs 317 and 348 (Selections from Revenue Records, North-Western Provinces, 1818–20, pp. 75 and 80); evidence of James Mill (quoted, ibid. 1822–33, p. 391); Munro's Minute of December 31, 1824 (Arbuthnot's Munro, i, p. 247). Cf. Fifth Report of 1812, p. 18 (Madras edition).
  14. 1. And this in its turn is in bad years liable to total or partial remission under the arrangements described on page 237 below.
  15. 1. To avoid misapprehension it may be well to note that the revenue is not in any Province of India assessed by apportionment as it is in France: that is to say, it is nowhere the custom to fix in a lump sum the revenue of the Province and to distribute it periodically over the Districts and villages in accordance with the local valuations. The fact that the settlements are for long terms and that the assessments in Northern India are first determined in the lump for groups of estates and are subsequently distributed over the estates and holdings gives the assessments, at least in the zamīndāri [p. 224] Provinces, something of the appearance of what the French call an ‘impôt de répartition’; but the object of the tax is to assess each unit (viz. in Northern India the estate and in ryotwāri Provinces the survey number or field) at a certain proportion of its value, so that, although for convenience the data are considered by groups or Districts, the Indian land revenue should apparently be classed as a ‘rated’ tax or an ‘impôt de quotité’ and not as an ‘impôt de répartition.’
  16. 1. Resolution of the Governor-General-in-Council, No. 1, dated Jan. 16, 1902, paragraph 34.
  17. 1. Resolution of the Governor-General-in-Council, No. 1, dated Jan. 16, 1902, paragraph 20.
  18. 1. Resolution of the Governor-General-in-Council, No. 1, dated Jan. 16, 1902, paragraph 18.
  19. 1. The District is a fair unit to consider in this connexion, as the operations may usually be said to commence and close in the various parts of a District at the same time, and the interval between the commencing and closing represents roughly the period during which each landholder in the District is liable to periodical visitation and annoyance. The landholder is not affected by settlements proceeding in other Districts; and when it is said that the Madras survey, for instance, occupied from 1855 to 1896, this relates to the survey of many Districts and does not in the least imply that each landholder was subject to survey operations for forty years. (Even the total length of time taken to survey the Madras Presidency compares not unfavourably with that taken over the French Cadastral Survey, which lasted from 1807 to 1850.)
  20. 2. Vol. III, chap. ix.
  21. 1. In Madras, where it is the custom to assess fields capable of irrigation at high rates, the annual granting of remissions for failure or excess of water is part of the ordinary revenue routine.
  22. 2. See especially Thomas's Revenue Resources of the Moghal Empire (1871), Hunter's Indian Empire, 3rd edition, pp. 352–7, and Jadunath Sirkar's India of Aurangzib (Calcutta, 1901).
  23. *. Total for whole empire: Provincial details not available.
  24. †. Omitting Bengal.
  25. ‡. Omitting Oudh.
  26. ‖. Omitting Orissa.
  27. 1. The above calculation presumes that wheat sold in Akbar’s time for 3⅓ Akbari maunds or 1⅔ modern standard maunds to the rupee, and now sells for ·4 maunds to the rupee—see Prinsep’s Useful Tables, i. 77; Ain-i- [p. 239] Akbarī (Blochmann), i. 62. Figures are unfortunately not available for a proper comparison of the cultivated area in the same Provinces at the two periods under consideration, but there can be no reasonable doubt that cultivation has increased considerably since a.d. 1594.
  28. *. That is, after excluding the area which is held free of revenue or at privileged rates, under the arrangements described on p. 227 above.