Preamble

1Of the two largest groups of underprivileged minorities in India, the Dalits (ex-Untouchables, present Scheduled Castes) are well known, while tribals are almost unheard of outside the country. While Dalits form an inextricable element of Indian society and are both socially and geographically present everywhere, the tribals live on the margins of civilization. The two groups probably share a common origin, and both are branded by their low status relative to the dominant Hindu society. They have received special attention from the Indian authorities, in line with the Gandhian ideal, and benefit from most governmental efforts to combat poverty. At quite an early stage, their development became politicized, as a Dalit or tribal vote during elections carries the same weight as that of any other Indian citizen. The fiftieth anniversary of Indian Independence, and of the death of Gandhi, seems a good time to assess the state of development of these “poorest of the poor.”

Methodology and techniques

2The documents that follow present possibly the first ever exploration of the results of the Census of 1991, with the help of figures and figures, forming an outline for a mini-atlas. For the first time, the data of the Census of India are available on diskette ; nevertheless, the quality of the information files presented several problems, which necessitated a long and difficult process of verification. Moreover it was not possible to conduct a census in some areas of India, or only partially so, due to security difficulties and/or conflict. In addition, quite a few regions have undergone a revision of administrative boundaries, so that the data do not correspond with those of earlier censuses. We have tried to minimize these shortcomings by various means, so as to provide at least a glimpse of the main trends, even at the cost of some imprecision.

3Although the statistical material available is extensive, it does not as yet match the wealth of Western censuses, and this often forced us to approach our topics indirectly. On the other hand, the cumbrous nature of some methods of treatment used has made it impossible to analyse all the available information within the framework of the present article. We have therefore selected those results which seemed most interesting and most effectively underlined the basic issues concerning our study, while complementing, as far as possible, already published information. This consideration has led us, for example, to analyse rural rather than urban data—a choice that is more than justified by the fact that 74.3 per cent of all Indians, 81.3 per cent of Scheduled Castes and 92.6 per cent of Scheduled Tribes live in rural areas— some of the highest rates in the world, especially for such a densely-populated country. This exercise is less a purely academic one than it might seem, in view of the fact that the Indian authorities in charge of development for the minorities seriously lack the means and methods of investigation to undertake even a basic survey of existing conditions. The treatments utilized here also aim to make the best use, at as little a cost as possible, of the only source of information that is available for the whole (or almost the whole) of India, the population census. The reader will notice that it has not been possible to employ a uniform data scale for the comparison of figures, since the difference between the populations of the two groups varies too much to allow effective common graphic representation. In some cases, we have retained an older mode of calculation that is less precise, but the only way to enable comparison with earlier census figures. We were also unable to reproduce all the figures and analysis documents utilized by us, which would have allowed the reader to check the findings presented and make an individual analysis. The large-scale tables would have taken up too much space and were too difficult to print. We hope the figures will be sufficient.

Topics and figures

4The documents presented here attempt to investigate, from the viewpoint of spatial distribution, a few phenomena which seem to us to be good indicators of the present situation of the underprivileged minorities. While we have limited our investigations to rural populations, which are the most representative of the Indian population as a whole, and the least studied, we are aware that the urban side of the situation would provide an important complement to the study.

5The rural population of India in 1991, and changes in it since 1961 (Figures 13.1 to 13.3), forms a kind of introduction, since it provides the framework for most of the “scheduled populations” and will serve as a basis for comparison. Then follows the spatial distribution of the two minorities, in absolute figures and as a proportion of the total population (Figures 13.4 to 13.7). Their demographic growth is explored through the increase in their numbers from 1961 to 1991 (Figures 13.8 and 13.9). Literacy rates, viewed in relation to the sex ratio of the literate population, (Figures 13.10 to 13.12) give a measure of the effectiveness of Indian education policy. Figures 13.13 to 13.15 introduce a new category of data in the Indian census : children below 7 years of age. Figures 13.16a and 13.16b together analyse changes in the sex ratio of the rural population on a state-by-state basis, from 1901 to 1991, while Figure 13.17 attempts to construct a general typology of sex ratios, based on a multivariate visual matrix analysis. Lastly Figures 13.18 to 13.20 present the results of a detailed matrix analysis of 5 groups of occupational categories : 1) cultivators + animal husbandry + forestry ; 2) agricultural labourers ; 3) craftsmen + construction workers ; 4) industrial workers ; 5) services.

Introduction

6In 1991 Scheduled Caste and Scheduled Tribe communities numbered 138 and 68 million individuals, respectively, which is almost one quarter of the 839 million inhabitants of the Indian Union (16.5 per cent and 8.1 per cent, totalling 24.6 per cent).

7Dalits and tribals probably share a common origin, dating back to the first inhabitants of the sub-continent. Under pressure from Indo-European invaders coming from the north-west, in order to avoid extinction (probably by assimilation rather than extermination, since the Hindus have demonstrated a far higher level of tolerance than Western societies) indigenous populations probably withdrew gradually to more inaccessible regions : dense forests, hills, etc. A frequent occurrence, especially in Asia. Their descendants are the tribals who are still to be found in forested and mountainous regions. One hypothesis concerning Dalit origins is that they were fractions of indigenous population that were gradually integrated into Hindu village communities and consigned to the lowest rank in the socio-religious hierarchy of the caste system—the outcastes. The figures showing the distribution of the two groups (13.4 to 13.7) confirm this line of reasoning, indicating a complementary character of their respective spatial distribution.

8While the Dalits are well-known in the West for their low social and economic status, little is known about the tribal groups, although they are no less interesting than other tribal societies studied by ethnologists elsewhere in the world (Herrenschmidt, 1974). Over the last thirty years, their situation has deteriorated rapidly, and they now have the dubious distinction of being numbered amongst those traditional societies in danger of cultural extinction. Their refuge on the margins of the more fertile lands first occupied by the Hindu civilization said to be of Indo European origin, and their space-consuming farming techniques, akin to those of the Neolithic, make them vulnerable to the pressures generated in an India which is in the midst of demographic transition. Ill-prepared to take advantage of the legal measures that are supposed to provide them assistance and protection, they helplessly watch as the forests and lands that are the core of their culture and the basis of their economic system are taken away from them (by massive deforestation, construction of big dams, etc.). Their standard of living has declined drastically and in some areas is even lower than that of the Dalits. This deterioration is exacerbated by a progressive erosion of their culture and traditional life-styles that used to maintain community solidarity.

Figures 13.1 and 13.2 : distribution of rural population

9The rural population of India consists of 623 million individuals, at an average density of 219 inhabitants per square kilometre, the most densely-populated regions being the ancient centres of Indian civilization. The first of these is the Indo-Gangetic plain in the north, with 270 million inhabitants, 220 of whom live in rural areas, and with an overall average density of 513 inhabitants per square kilometre (450 in rural areas). The oldest nucleus of population can be found on the borders of Bihar and Uttar Pradesh, between Patna and Varanasi, the mythic and historical heart of Hindu civilization, where there are average rural densities ranging from 800 to as much as 1,100 inhabitants per square kilometre. 52 million rural and 7 million urban inhabitants live there in an area of 67,000 square kilometres. A more recent nucleus is centred in Calcutta, where a rural population of 41 million occupies an area of 51,000 square kilometres, with average rural densities of 930 inhabitants per square kilometre. The total population, including that of Calcutta, amounts to 59 million. Densities in the rest of the plain vary between the national average of 219 and 600 inhabitants per square kilometre, with a strong north-south gradient and a more gradual east-west one (towards Rajasthan and the Punjab). These high densities stop as soon as one moves away from the irrigated alluvial plain.

10The coast of the peninsula forms an almost continuous belt of high population densities, including the eastern deltas, Kerala and central Tamil Nadu : 90 million inhabitants in all, with average rural densities of 440 inhabitants per square kilometre, reaching peaks of nearly 1,500 in Kerala, an important spice-coast since antiquity.

Figure 13.1 : Rural population. Numbers

Figure 13.2 : Rural population. Overall rural densities

11The interior of the peninsula, in contrast, is much less densely-populated. Nevertheless, the undeniable effects of the huge water projects set up since Independence can be observed. The courses of big rivers, such as the Kaveri. the Krishna, the Godavari and, to a lesser extent, the Mahanadi, can easily be traced from the patches of higher densities (150 to 360) that mark them. Elsewhere densities of less than 150 indicate either arid zones (the Thar desert in the north-west, Rayalaseema in the south) or forested areas inhabited by tribals. It may be noted that rural densities fall below 45 inhabitants per square kilometre in only 44 districts. The extensive tribal area in the north-east along the Brahmaputra is characterized by lower densities along the periphery, corresponding to mountainous areas, in contrast to the more densely-populated valley (an average of 250 inhabitants per square kilometre).

12Migrants who have come more or less illegally from teeming Bangladesh are the reason for this, as well as the social tensions, which have blighted Assam for the last twenty years or so.

13Two divergent phenomena stand out : the increase in population on the north-western and north-eastern frontiers ; and the stabilization in the population of the high-density regions of the peninsula. Practically no coastal district shows more than the national average of 1.87 per cent annual increase over 30 years, and most of them lie below 1.5 per cent. This also applies to the two states of Tamil Nadu and Kerala that have the highest population densities in the south, levelling out at 1.34 per cent. The latter experienced a drop in growth, from 2.22 per cent, between 1961 and 1971, to 0.35 per cent, between 1981 and 1991. This result can be linked with the high education level of the population (Figures 13.10-13.12). The only southern regions with growth rates higher than the national average are those affected by recent agricultural irrigation programmes : the middle and upper Krishna and Godavari basins. The Kaveri valley, running from Karnataka through Tamil Nadu, seems to have reached a density limit. This makes it easier to understand the intensity of the conflicts between these two states over the use of Kaveri waters.

14The observation regarding saturation levels also applies to some regions in the north, more precisely to the central Indo-Gangetic plain : 1.6 per cent increase between 1961 and 1991 in the Punjab, 1.14 per cent between 1981 and 1991 ; 1.87 per cent between 1961 and 1991 in Uttar Pradesh. The most notable exception is clearly Rajasthan, where the Thar desert has seen its population, admittedly small, increase by more than 3.5 per cent annually, due to the Rajasthan canal. In 30 years, the population of this region has grown from 6.3 to 14.9 million. In Bihar, increases are moderate (1.9 per cent) but still enough to cause concern in one of the poorest and most densely-populated states of India (almost 500 inhabitants per square kilometre). Birth control measures have had little impact on illiterate populations subject to quasi-feudal social structures. The tribal north-east has also experienced a sustained growth of 2.5 per cent per annum, the highest in the still sparsely-populated mountain areas (Nagaland 3.57 per cent).

Figure 13.3 : Increase in rural population 1961-1991

15The spatial distribution of these two groups accurately reflects their situation relative to Hindu society. Dalits, who fulfil important social and economic functions,1 are found everywhere except where Hindu society itself is absent, i.e. in the tribal areas ! The tribals are confined to certain specific locations in a few areas, around which their concentration progressively lessens. A more or less significant tribal “background” exists elsewhere, but is practically absent from the Hindu “cow-belt” of the Indo-Gangetic plain. It may be noted that Scheduled Castes are numerous in delta regions and more generally in areas of intensive cultivation, where a large labour-force is required. The figures indicating density should be compared with those showing occupation (Figures 13.20-13.22).

16There is a striking degree of complementarity in the spatial distribution of the two minorities : very few tribals in the Indo-Gangetic plain and the southeast of the peninsula, where a large number of Scheduled Castes are to be found. Conversely, the strong concentrations of tribals in the north-west of the peninsula (borders of Maharashtra and Gujarat), Orissa and Madhya Pradesh, correspond to pockets empty of Scheduled Castes. Hence the appeal of a hypothesis which argues for the tribal origin of Scheduled Castes who were slowly integrated into Hindu society. But are they the only ones ? Perhaps the extraordinary complexities of local caste systems reflect ancient tribal roots ? Only a more detailed study at subregional and village levels would throw enough light to confirm or disprove this hypothesis.

17The growth rate over 30 years of these two minorities is significantly higher than that of the rest of the rural population : 2.29 per cent and 2.84 per cent, respectively, as opposed to 1.94 per cent for the rural population as a whole, and 1.76 per cent if we consider only the non-minority population. The figures—which were difficult to compile because of the many changes in administrative boundaries which have taken place in the course of the period under consideration, and necessitated the estimation of figures in order to avoid too many statistical gaps—reveal interesting trends. Firstly, they provide overall images that are very different from the growth of the population as a whole (Figure 13.3) : while this increases mostly in the north, the main growth among Scheduled Castes is in the centre of the country, on the borders of Maharashtra and Madhya Pradesh. It extends west towards the Mumbai region, south towards Bangalore, and follows the Krishna and Godavari valleys eastwards. Clearly, the greatest increase in the population of Scheduled Castes corresponds to areas where, formerly, they were less common. It also corresponds to agricultural water projects ; so it is logical for it to appear also in western Rajasthan. On the other hand, areas with a high density of Scheduled Castes (Figure 13.6), such as Uttar Pradesh and Tamil Nadu, show little increase.

Figures 13.4 and 13.5 : Rural populations—Dalits and Tribals

Figure 13.6 : Rural population 1991. Proportion of Dalits

Figure 13.7 : Rural population 1991. Proportion of Tribals

18Scheduled Tribes have undergone a development that is spatially similar to that of the Scheduled Castes, focusing on the centre of the peninsula, but on the Telengana Plateau in Andhra Pradesh. It extends very noticeably up to the southern borders of Karnataka and into southern Maharashtra. This configuration is due to political reasons rather than to geographical ones. In Andhra, significant increases are due to the inclusion of the Lambada group in the list of Scheduled Tribes in this state just before the legislative elections of 1977. This change of status led to massive emigration from Maharashtra, where the group was not scheduled.

19The literacy level among Dalits (Figure 13.10) is very low, 33.25 per cent, and very uneven : 390 literate women for every 1 000 literate men. This sex inequality is linked closely with literacy levels. The higher the level of literacy, the greater the equality between the sexes. Its spatial distribution is very similar to that of the overall population (Figure 13.12) with the exception of two groups. Group B, with a very low level of literacy but a relatively lesser degree of inequality between the sexes, may indicate official efforts in the uplift of the Dalits. Group C1, which shows a marked degree of sex inequality, includes few Dalits.

20The literacy level among the tribal population (Figure 13.11) is even lower than that of the Dalits, with 27.45 per cent, but slightly more balanced, with 406 literate women for every 1 000 literate men. Similar geographical patterns are evident in Figures 13.11 and 13.12, but figure 13.11 indicates a higher level of literacy in the tribal states of the north-east, and in eastern Maharashtra. A group with a very high sex inequality, B1, can be found in the Gangetic valley, where there are fewer tribals.

21The literacy level among the rural population as a whole (Figure 13.12) is higher, 44.69 per cent, and less uneven : almost one literate woman for every two literate men. Group A1, with the highest literacy level, shows 938 literate women for every 1 000 literate men. Groups A2, A3, A4, and A5 form successive circles around A1 in the south-western states, in the east, and in the far north. The B groups show a high level of sex inequality with a low literacy level, with the exception of B1. They are found mainly in the northern half of the country, as well as in the west of Andhra Pradesh and of Orissa. All three figures display the very low level and great sex inequality of literacy in Rajasthan.

22Minorities have a higher birth-rate than the rest of the population, since they possess a proportion of children that is significantly higher than the average for rural India which is 18.76 per cent : 19.92 per cent for Scheduled Castes, and 20.36 per cent for Scheduled Tribes. Yet again, the north-south asymmetry is noticeable, both for the overall population and for Scheduled Castes. Even though this disparity is less evident among tribals, the north-central region does have a higher proportion of children than other tribal regions. Again, it should not be forgotten that the size of the populations concerned in these areas is often very small. Areas of low and high values should be compared with the literacy figures, for the expected correlation coefficients are striking. The strong demographic dynamism of the central Deccan is confirmed, fostered by high birth rates and migration.

Figure 13.8 : Dalits. Growth 1961-1991

Figure 13.9 : Scheduled tribes. Growth 1961-1991

Figures 13.10 : Rural population 1991. Dalits : literacy

Figure 13.11 : Rural population 1991. Tribals: literacy

Figure 13.12 : Rural population 1991. Overall literacy

Figures 13.13 : Rural population 1991. Proportion of children below 7

Figure 13.14 : Rural population 1991. Dalits: proportion of Children below 7

Figure 13.15 : Rural population 1991. Tribals: proportion of children below 7

Figure 13.16a : Rural population 1991. Sex ratio* : changes 1901-1991 by state

Figure 13.16b : Rural population 1991. Sex ratio* : changes 1901-1991. Map of profiles in figure 16a

23The sex ratio (number of females per 1 000 males) in India has shrunk from 972 in 1901 to 927 in 1991, in a slow decline which has shown a tendency to accelerate since 1961. This acceleration is more marked in rural India where, from 979 at the beginning of the century, it was still 963 in 1961 but has fallen to 938 over the last 30 years. In urban areas, starting from 910, because of the presence of migrants, it fell steeply to reach 831 in 1941, then rose considerably to 894 in 1991.

24Diagram 13.16a shows changes in the sex ratios in Indian states from 1901 to 1991, in comparison with that of India as a whole. The line at the centre of each profile, sinking slightly, represents the country’s overall development. Black indicates values that are higher than the Indian average, and grey those that are lower. The states have been grouped according to similarities in their profiles. Type A includes profiles in which the sex ratio has increased over time, Type B those that have dropped. These are subdivided into three main families : profiles that have remained continuously above the average (black), those which have crossed the average or fallen below it, and those which have remained below average.

25The average sex ratio of the rural population was 939 females for every 1 000 males in 1991 (Figure 13.17). The northern half of India has a smaller proportion of females than the southern half. Tribals show a more balanced sex ratio, 976, as against 937 (other castes) and 926 (Scheduled Castes). In the heart of the tribal zone of Madhya Pradesh and Orissa, and in Arunachal Pradesh, the ratio exceeds 1 000. On the northern and western margins of the central zone, and in the north-east apart from Arunachal Pradesh, it drops to 960. In mixed zones (Tribals / Scheduled Castes) it falls below 950. Elsewhere the sex ratio among tribals tends to be almost always higher than that of other social groups. It appears as if tribal societies may have more equal female/male mortality rates.

26The sex ratio among Dalits is the lowest, at 926. However, it approaches equitable levels, at 995, on the eastern coast. It is average, 951, in the southern half of India. On the other hand it is very low, 863, in the northern half. Here we find the same patterns indicated in Figures 13.10 (literacy), and 13.14 (proportion of children below 7), demonstrating how closely demography and education are linked. The exception of Andhra Pradesh may be noted, where there is a very low literacy rate, but one that is relatively egalitarian compared with that of the north, accompanied by a near average sex ratio.

27The sex ratio of other groups (neither Tribal nor Dalit) stands between these two, at 937 females for every 1 000 males. It is very high on the west coast, as are literacy levels, and a low proportion of children below 7. The sex ratio is also high in north-western Gujarat, but does not show any correlation with literacy or fertility. Central Maharashtra and Karnataka, like southern Gujarat, show an average level at 956.

Figure 13.17 : Rural population 1991. Typology of sex ratios

28The north, the Himalayan foothills, and Rajasthan stand out here again as a result of their low figure, of 892, which unfortunately correlates closely with the lowest and least egalitarian literacy rates. In other areas where Scheduled Castes predominate, the average sex ratio is a little higher than that of the Scheduled Castes. In tribal areas, on the other hand, it is distinctly lower than that of the tribals and slightly lower than that of the Scheduled Castes.

Occupations

29In the 1991 Census, occupations were divided into nine categories. We have maintained a distinction between cultivators and agricultural labourers, rather than clubbing them together in the same category, “agriculture”, since cultivators represent 48.4 per cent of the rural working population, and agricultural labourers, 31.6 per cent. We have also kept as a single group “Forestry, animal husbandry and fishing”, activities that are characteristic of some tribes, and represent 2.2 per cent of the working population. Household industry includes artisans and construction, and employs 2.2 per cent of the working force, while Industry includes industries other than household industry and mining, and counts for 5 per cent of the working force. Services include commerce, transport, communication and other services ; it represents 10.5 per cent of the working population.

Figure 13.18 : Rural population. Typology of occupations by social category

30Classifying occupations according to social category (Figure 13.18) underlines the tribals special profile. They are mostly engaged in agricultural activities. 57 per cent of them are cultivators, and 33 per cent agricultural labourers. Only 2.5 per cent work in industry, 4 per cent in services, 2 per cent in forest and fishery, and 1 per cent are artisans. The Scheduled Castes are primarily agricultural labourers, 55.1 per cent, and less often cultivators, 29.4 per cent. They also work in industry, 6.8 per cent, and household industry, 2.2 per cent. Other castes are to be found in all the categories except agricultural labour, which represents only 24.8 per cent of their occupations.

31A strong dichotomy can be observed between areas engaged mainly in agriculture and those engaged in a variety of occupations (Figure 13.19). The former correspond to regions where Scheduled Castes are relatively numerous, except for the Punjab, Haryana and northern Uttar Pradesh, as well as south-eastern Madhya Pradesh. In the north cultivators are more numerous, in the south and the eastern Ganges valley agricultural labourers predominate. Where tribals are numerous, in central India, Scheduled Castes are mainly engaged in household industry. They are employed in the service sector in the tribal areas of the north-east. Elsewhere, where they are less numerous, they show a mixed profile. And in the western Ghats, they are often engaged in forestry.

32The figure of tribal occupations (Figure 13.20) presents a less clear picture than those of the Scheduled Castes and other social groups. The heart of the central tribal zone is predominantly agricultural, and there too there are more cultivators in the north and agricultural labourers in the south, but this is less clearly marked. Multiple activities including household industry are found in the south-eastern part of India. The western coast is mainly occupied with forestry and fishing. The western parts of Maharashtra, Gujarat and Rajasthan show a predominance of industry. Services predominate in the Ganges basin and the north-east.

33For other social groups (Figure 13.21), agriculture is dominant throughout the central part of India from the north to the south,. The same dichotomy between the north (cultivators) and the south plus Bihar (agricultural labourers) appears again here. On the western coast, industry is the main occupation, while on the eastern coast household industry predominates. The north-east (Tribal states) and the north-west (Punjab, Haryana and northern Uttar Pradesh) stand out because of the large number of people employed in the service sector. The Western Ghats, and also the heart of Assam, have significant populations involved in forestry.

Figure 13.19 : Rural population. Dalits : general typology of occupations

Figurel3.20 : Rural population. Tribals: general typology of occupations

Figure 13.21 : Rural population. Other castes : general typology of occupations