Jump to ContentJump to Main Navigation
Indian DevelopmentSelected Regional Perspectives$

Jean Drèze and Amartya Sen

Print publication date: 1997

Print ISBN-13: 9780198292043

Published to Oxford Scholarship Online: September 2011

DOI: 10.1093/acprof:oso/9780198292043.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 03 December 2021

Uttar Pradesh: The Burden of Inertia

Uttar Pradesh: The Burden of Inertia

(p.33) 2 Uttar Pradesh: The Burden of Inertia*
Indian Development

Jean Drèze (Contributor Webpage)

Haris Gazdar

Oxford University Press

Abstract and Keywords

This chapter discusses the problem of economic and social backwardness in Uttar Pradesh and its causal antecedents. Among these are the disastrous functioning of public services in rural areas, the persistence of widespread illiteracy, and the suppression of women's agency in society. This chapter also talks about the social and political circumstances underlying these diverse failures.

Keywords:   Uttar Pradesh, economic backwardness, social backwardness, public services, rural areas, illiteracy, suppression of women's agency

1. Introduction

Why Uttar Pradesh?

One of the lessons emerging from a wide range of recent development experiences is that public action can play a powerful role in promoting essential aspects of the quality of life, even at an early stage of development. The literature on this subject includes a number of instructive case studies, such as that of Kerala appearing in this book. In this paper, we investigate an example of the other side of the same coin, i.e. the penalties of inaction.

Uttar Pradesh can also be seen as a case study of development in a region of India that currently lags behind much of the rest of the country in terms of a number of important aspects of well-being and social progress. The region is characterized inter alia by exceptionally high levels of mortality, fertility, morbidity, undernutrition, illiteracy, and social inequality, and a slow pace of poverty decline.1 (p.34) This region consists of four large, adjacent states of north India: Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh. There are important differences between these four states, and it would be a mistake to consider them as a single entity. But the causes of backwardness in these different states, nevertheless, appear to have much in common, and comparative investigations have pointed to many broad similarities in the social, cultural, and even political make-up of these states.2 A case study may help to understand the causes of retarded development in this region.

Aside from supplying a useful case study, Uttar Pradesh deserves a good deal of attention in its own right. In fact, a large part of India's total population—one could even say of the world population—is involved: at the time of the 1991 census, Uttar Pradesh had a population of 139 million, larger than that of any country in the world other than China, India, Indonesia, Brazil, Russia, and the United States.3 Understanding what goes on in Uttar Pradesh, and what can be done about it, is in some ways just as important as to study Bangladesh or Ethiopia.

The plan of the chapter is as follows. After some methodological clarifications in the remainder of this section, we turn in section 2 to the available evidence on living conditions in Uttar Pradesh. We also discuss the respective contributions of slow economic growth and ineffective public action to the persistence of endemic deprivation in this region. The issue of ineffective public action is illustrated in section 3 with a case study of basic education in rural areas. section 4 presents further comments on the causes of educational backwardness in Uttar Pradesh. section 5 explores the general phenomenon of public inertia in this region, and some of its social and political roots. Concluding remarks are presented in section 6.

(p.35) Methodological Clarifications

Before we begin, three methodological clarifications are in order. First, some of the analysis presented in this paper (particularly in sections 2 and 3) is essentially comparative in nature—it deals with how Uttar Pradesh has done in different fields in comparison with other states or regions of India. The presentation of comparative data will gain in clarity if we choose one ‘reference region’ with which Uttar Pradesh is consistently compared. We have decided against taking India itself as the reference region, simply because the four north Indian states mentioned earlier have a large weight in the all-India average (e.g. they account for 40 per cent of the total population). Since the motivation of this paper is partly to study Uttar Pradesh as a case of this collection of states, a comparison with a region of which these four states are a large component would not be particularly helpful. Specifically, such a comparison would tend to hide the sharp contrasts that exist between these states and the rest of India. As an alternative, we have adopted ‘South India’ (defined as the union of Andhra Pradesh, Karnataka, Kerala, and Tamil Nadu) as the reference region.

This reference region may be considered as a relatively ‘advanced’ part of India, in terms of many of the indicators used in this paper. The advanced character of this region, however, should not be exaggerated. The largest of the four south Indian states, Andhra Pradesh, is in some important respects (e.g. literacy levels) much closer to the large north Indian states than to Kerala or Tamil Nadu. The state of Kerala, of course, has exceptional achievements in some fields, but it only accounts for 15 per cent of the total population of south India. Basic calculations of indicators such as life expectancy, literacy, domestic product, etc., suggest that south India is often not very different from the rest of India outside Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh, In any case, none of what follows depends on south India being ‘representative’ of India outside the large northern states—we simply propose it as a helpful benchmark. From time to time, we will also refer to the specific contrast between Uttar Pradesh and Kerala, which is particularly relevant to the concerns of this book.

The ‘South India’ figures presented in this paper have been calculated as weighted averages of the state figures. The weights used are different for different indicators. In most cases, we have (p.36) been able to use the weights relevant to the respective exercises (e.g. population aged 7 and above for the 7 + literacy rate, male population for the female—male ratio, number of births for the infant mortality rate, etc.). In a few cases, some approximation has been necessary, because the exact weights were not available. But the margin of error involved in the approximations is, in every case, small enough to be ignored for our purposes.

The second clarification concerns the reference period used in the presentation of different indicators. Comparisons between different states, and related statistical exercises used in this paper, are sometimes sensitive to the choice of reference period. In order to have an impartial rule in this respect, the most recent year for which the relevant information could be obtained has been chosen as the reference year in each case (except when there was a specific reason for focusing on a different year). In a few cases, we have had to use 1981 census data, because the corresponding information from the 1991 census had not been released at the time of writing.

Finally, this paper makes considerable use of field-based observations, derived inter alia from village studies as well as from our own field work, to take the analysis beyond what emerges from, more aggregative secondary data alone.4 This type of information will be useful, for instance, in understanding patterns of gender relations, the functioning of public institutions, and the nature of rural politics. These field observations are not, of course, representative of the state in a statistical sense. This qualification will have to be borne in mind as we go along, but it does not prevent field-based investigations from providing valuable insights into different aspects of the economy and society of Uttar Pradesh.

(p.37) 2. Living Conditions in Uttar Pradesh

2.1 Background

Uttar Pradesh is primarily an agricultural state, with a high proportion (just above 80 per cent) of its population living in rural areas, and primarily engaged in the agrarian economy. Although urbanization and non-agricultural employment have been increasing over time, conditions of production in agriculture and the distribution of agrarian assets (particularly cultivable land) remain the most important determinants of the material condition of the population.

Two recent developments in the region's agrarian history can be regarded as significant turning points. The first came with the reforms of land revenue and property rights that followed India's independence, generally known as ‘zamindari abolition’. These reforms abolished the role of private intermediaries in the land revenue system, and led to a clearer definition of private property rights in land. The structure of land ownership has remained, more or less, the same since then. These early reforms coincided with the post-independence adoption of social and economic development as official goals of public policy. The second development was the spread, in the nineteen sixties and seventies, of modern agricultural practices in western Uttar Pradesh, and their subsequent diffusion to other regions of the state.

Although neither of these two episodes has been particularly dramatic (compared, for instance, with land reforms and productivity growth in other developing regions, including parts of India), they do define the broad parameters of change in the economic circumstances of the bulk of the population. The land reforms limited the powers of large feudal landlords, and gave ownership rights to a vast majority of tenant farmers who previously did not own land.5 The reforms did not, however, eradicate landlessness, nor did they prevent the persistence of massive inequalities of land ownership in the state. The land ownership structure has changed (p.38) little in Uttar Pradesh in the forty years since the abolition of zamindari.6

(p.39) Technological change led, over time, to a significant expansion of private agricultural incomes, and also laid the basis for some diversification of economic activity.7 In the absence of major redistributive programmes, however, the gradual expansion of private incomes only led to a slow decline in conventional indicators of poverty. This, combined with the fact that other bases of improvement in human well-being (such as efficient public services and widespread literacy) were severely neglected over the same period, resulted in comparatively limited achievements in terms of the elimination of endemic deprivation. Some relevant evidence is presented in Table 1, and discussed in further detail below.

Table 1. India, Uttar Pradesh and South India: Selected Indicators


Uttar Pradesh

South India

Population, 1991a (million)







Life expectancy at birth, 1990–2 (years)









Death rate, age 0–4, 1991 (per 1,000)









Literacy rate, age 7+, 1991 (%)









Average per-capita consumer expenditure, 1987–8 (Rs/month at 1970–1 prices)









Head-count ratio, 1987–8 (percentage of the population below the poverty line)









(Note.a) In brackets, the proportion of the population living in rural areas (percentage).

Source. Drèze and Sen (1995), Statistical Appendix, based on data presented in Nanda (1992), Government of India (1993a), Tendulkar et al. (1993), and derived from the 1991 census, the National Sample Survey, and the Sample Registration System. The life expectancy figures are unpublished estimates supplied by the Office of the Registrar-General. The figures for ‘South India’ have been calculated as weighted averages of the relevant state-specific figures.

2.2 Basic Demographic Indicators

The demographic evidence on survival chances in Uttar Pradesh provides a helpful starting point for our enquiry. As discussed in the companion volume (Drèze and Sen, 1995), the central goal of development can be seen as the expansion of human ‘capabilities’.8 Further, there is an obvious case for considering survival as one of the basic capabilities of primary interest. Some mortality and longevity indicators for Uttar Pradesh are given in Table 2, along with the corresponding figures for South India.

Table 2. Demographic and Health Indicators

Uttar Pradesha

South India

States doing ‘worse’ than Uttar Pradesh

Life expectancy at birth, 1990–2 (years)


54.6 (14)


Madhya Pradesh, 53.5


56.8 (13)


Madhya Pradesh, 54.1

Orissa, 55.9

Other mortality-related indicators

Crude death rate, 1992

(per 1,000)

12.8 (14)


Madhya Pradesh, 12.9

Under-five mortality rate,


141 (15)



Estimated maternal

mortality rate, 1982–6

(per 100,000 live births)

931 (14)


Rajasthan, 938

Fertility indicators

Total fertility rate, 1991

5.1 (15)



Crude birth rate, 1990–2

35.8 (15)



Female—male ratio, 1991

Females per 1,000 males

879 (14)


Haryana, 865

Notes. (a) In brackets, Uttar Pradesh's rank among 15 ‘major states’ for which the relevant data are available: Andhra Pradesh, Bihar, Gujarat, Haryana, Himachal Pradesh, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Orissa, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh, West Bengal. The ranking is based on arranging the states in increasing order of the relevant indicator, except in the case of life expectancy and female—male ratio (decreasing order).

(b) Probability of dying before age 5.

Sources. Life expectancy: unpublished estimates (based on SRS data) supplied by the Office of the Registrar-General. Crude death rate: Sample Registration System 1992, Statement 26, p. 30. Under-five mortality rate: International Institute for Population Sciences (1994a), Table 34, p. 85. Maternal mortality rate: Mari Bhat et al. (1992), Table 4. Total fertility rate: SRS data presented in Government of India (1993a), pp. 20 and 40. Crude birth rate: Sample Registration Bulletin, January 1994, pp. 22–9 and 46–53. Female—male ratio: Nanda (1992), p. 13, based on 1991 census data. The figures for South India have been calculated as weighted averages of the relevant state-specific figures.

As this table indicates, life in Uttar Pradesh is short and precarious. Female life expectancy, for instance, is still below 55 years, and the under-five mortality rate is as high as 141 per thousand. To put things in perspective, these figures are not very different from, say, the corresponding estimates for sub-Saharan Africa (53 years and 160 per thousand, respectively).9 Among all major Indian states, Uttar Pradesh has the highest under-five mortality rate, the second-highest crude death rate, and the third-lowest life expectancy figure.10 The (p.40) number of maternal deaths per 100,000 live births in Uttar Pradesh was estimated to be as high as 931 in the mid-eighties. Only five countries in the world, among those for which official figures are available, had higher estimated maternal mortality rates at that time: Somalia, Bhutan, Ghana, Gambia, and Congo.11

Survival indicators in Uttar Pradesh are quite dismal not only in comparative international terms, but also in relation to what has been achieved in the more advanced Indian states. For instance, a new-born girl can expect to live 20 years longer if she is born in Kerala rather than in Uttar Pradesh. And the probability that she will die before the age of one is more than six times as high in Uttar Pradesh as in Kerala (Government of India, 1993a, p. 31). Even the demographic gap between Uttar Pradesh and South India as a whole is quite striking (Table 2).

The demographic evidence on child survival is consistent with independent evidence on child nutrition. According to the recent National Family Health Survey, Uttar Pradesh comes second to Bihar (among India's major states) in terms of the incidence of undernutrition among children below the age of five.12

Aside from low survival chances, the demographic characteristics of Uttar Pradesh include high fertility rates. In fact, according to the latest available figures (Table 2), Uttar Pradesh has the highest birth rate among all Indian states, as well as the highest fertility rate. Uttar Pradesh has made comparatively little progress so far in terms of the ‘demographic transition’ from high to low levels of mortality and fertility.

2.3 Educational Levels

Empirical analyses of the determinants of demographic outcomes in India have consistently brought out the crucial role of literacy in (p.41) (p.42) reducing mortality and fertility rates.13 In particular, the four states that are commonly identified as lagging behind the rest of the country in terms of the demographic transition (Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh) also turn out to be the four states with the lowest literacy levels.14 The 1991 census indicates that, for persons aged 7 and above, the literacy rate in these four states ranges from 38 per cent in Bihar to 44 per cent in Madhya Pradesh as against 59 per cent in South India (with an even larger north—south gap in the younger age-groups).

The poor educational record of these four states is particularly striking in the case of female literacy. In Uttar Pradesh, only one woman out of four in. the 7+ age group was able to read and write in 1991 (see Table 3). Further, aggregate literacy figures tend to hide sharp variations between different regions and population groups, implying extremely low achievements for the most disadvantaged groups at a more disaggregated level. While the 7+ female literacy rate in Uttar Pradesh as a whole was 25 per cent in 1991, the figure goes down to 19 per cent for rural areas, 11 per cent for the scheduled castes, 8 per cent for scheduled castes in rural areas, and 8 per cent for the whole rural population in the most educationally backward districts. Currently-available data from the 1991 census do not permit further disaggregation, but the 1981 census figures suggest that, in Uttar Pradesh, female literacy remains close to zero for large sections of the society. For instance, the crude female literacy rate among scheduled castes in rural areas in 1981 was below 1.5 per cent in 18 out of Uttar Pradesh's 56 districts, and below 2.5 per cent in a majority of districts.15

Table 3. Uttar Pradesh: Educational Achievements and Participationa



Literacy rate, age 7+, 1991 (%)


52 (63)

19 (41)


70 (83)

50 (68)

Rural and urban combined

56 (68)

25 (49)

Literacy rate, age 7+, 1991: scheduled castes in rural areas (%)

39 (49)

8 (28)

Literacy rate, age 10–14, 1987–8 (%)


68 (84)

39 (72)


76 (90)

69 (85)

Proportion of children aged 12–14 who have never been enrolled in a school, 1986–7 (%)


27 (14)

68 (28)


19 (7)

39 (22)

Proportion of rural children attending school, 1987–8 (%)

Age 5–9

45 (75)

28 (68)

Age 10–14

64 (74)

31 (58)

(Note. a) In brackets, the corresponding South India figures (calculated as weighted averages of the relevant state-specific figures).

Sources. Compiled from Nanda (1992, 1993), Tyagi (1993), Sengupta (1991), Visaria et al. (1993), based on census and National Sample Survey data.

It might be added that, despite its far-reaching individual and social significance, literacy alone is not a momentous achievement in terms of the amount of learning involved. If we consider more demanding criteria of educational attainment, such as the completion of primary or secondary education, the achievement rates are (p.43) correspondingly lower. For instance, in 1992–3, only half of all literate males in Uttar Pradesh, and 40 per cent of literate females, had completed the cycle of eight years of schooling involved in the primary and middle stages.16 Many children in Uttar Pradesh, if they are literate at all, acquire this skill on the basis of a fleeting passage through the educational system.17

(p.44) As far as the specific issue of literacy is concerned, the most sobering feature of the educational situation in Uttar Pradesh is the persistence of high levels of illiteracy in the younger age groups. It is not just that a lot of adults are illiterate, pulling down the average literacy rate, with most people in the younger age groups being literate.18 Even in the younger age groups, illiteracy is endemic, especially in rural areas. In the late eighties, the incidence of illiteracy in the 10–14 age group was as high as 32 per cent for rural males and 61 per cent for rural females, and more than two-thirds of all rural girls in the 12–14 age group had never been to school (see Table 3). Uttar Pradesh is nowhere near the realization of the constitutional goal of free and compulsory education for all children up to the age of 14, which was supposed to have been reached by 1960.

2.4 Gender Inequality and Female Deprivation

The persistence of endemic illiteracy is not Uttar Pradesh's most distinctive social failure. In fact, the other large north Indian states (Bihar, Madhya Pradesh, and Rajasthan) fare no better than Uttar Pradesh in that respect. There is another field, however, in which Uttar Pradesh seems to fare worse than most, if not all, other Indian states—that of gender equality.

One basic indicator of the disadvantaged position of women in Uttar Pradesh is the female—male ratio in the population (see Table 4). In 1991, the number of females per 1,000 males in Uttar Pradesh was as low as 879. One Indian state has an even lower female-male ratio: Haryana, bordering on western Uttar Pradesh, where there were only 865 females per 1,000 males in 1991. However, disaggregated figures show that the ‘epicentre’ of the problem of low female—male ratios is not in Haryana but in western Uttar Pradesh. That region, which has more than one-third (p.45) of the population of the entire state and nearly three times the population of Haryana, has a female—male ratio of only 0.84.

Table 4. Female Disadvantage

Uttar Pradesh

South India

Females per 1,000 males (1991)

All ages



Age 0–6



Gender bias in survival

Ratio of female to male death rates (1991)

Age 0–4



Age 5–14



Age 15–34



Maternal mortality rate, per 100,000

live births (1982–6)



Gender gap in life expectancy (1990–2)

Female—male difference in years



Sources. Calculated from Mari Bhat et al. (1992), Nanda (1992), Government of India (1993a), Visaria et al. (1993), and Drèze and Sen (1995), Statistical Appendix—all based on census, National Sample Survey, and Sample Registration System data.

Some brief international comparisons may help to put this extraordinary number in perspective. The only countries with a female—male ratio lower than 0.84, among all those listed in Human Development Report 1994 (pp. 146–7), are the following: Kuwait (0.76), Bahrain (0.73), Qatar (0.60), and United Arab Emirates (0.48). These exceptionally low female—male ratios are overwhelmingly attributable to male in-migration. If we exclude cases of exceptional male in-migration, the country with the lowest female—male ratio in the world is Pakistan, with 92 females per 100 males. This is considerably above Uttar Pradesh (which, incidentally, has a much larger total population than Pakistan), not to speak of western Uttar Pradesh. Uttar Pradesh is not just a setter of world records when it comes to the female deficit in the population, it is virtually in a league of its own.

The main cause of Uttar Pradesh's low female—male ratio is the considerable female disadvantage in survival from birth until the mid-thirties (Table 4). For the 0–4 age group, female death rates in (p.46) Uttar Pradesh are 16 per cent higher than male death rates, in contrast with the typical pattern of strong female advantage in that age group, which applies even in South India.19 The female disadvantage in childhood is especially influential, since mortality rates tend to be particularly high in the younger age groups. Further, the link between excess female mortality in childhood and parental neglect of female children in this region of India is well documented.20 Uttar Pradesh's low female—male ratio is a tangible reflection of anti-female discrimination.

The effects of female disadvantage in child survival are enhanced by even greater gender disparity in death rates between the ages of 15 and 35. This is in contrast to South India, where in the same age group the gender gap changes in favour of females (Table 4). Much of the excess female mortality in this age group in Uttar Pradesh reflects the combined effects of high maternal mortality and high fertility. The average number of births per woman is about twice as high in Uttar Pradesh as in South India, and the risk of maternal death from a particular birth is almost three times as high (see Tables 2 and 4). Anti-female discrimination in. infancy and childhood, combined with high, levels of fertility and maternal mortality, imply that female life expectancy at birth in Uttar Pradesh is 2.2 years below the corresponding figure for males—in contrast with a three-year advantage of females over males in South India.

Before concluding on this issue, it is worth mentioning that the female—male ratio in Uttar Pradesh is not only low, it has also been steadily declining since the beginning of this century—from 0.94 in 1901 to 0.88 in 1991. It is difficult to explain this steady decline of the female—male ratio in Uttar Pradesh without invoking the persistence, and possible accentuation, of unequal gender relations.21 (p.47) The general process of modernization and development seems to have done very little, so far, to reduce gender inequality in Uttar Pradesh.

2.5 A Question of Poverty?

Poverty and Well-being

It might be tempting to think that the main cause of Uttar Pradesh's low achievements in terms of survival chances, child nutrition, fertility decline, basic education, gender equality, and related aspects of well-being, lies in high levels of poverty.22 There is, however, little evidence to support this hypothesis. Of course, there is plenty of poverty in Uttar Pradesh; in 1987–8, almost half of the population was estimated to live below the ‘poverty line’.23 But the incidence of poverty is also high in India as a whole. In fact, poverty indicators for Uttar Pradesh and India have been quite close to each other in most years for which the relevant data are available, as Fig. 1 illustrates.24 Differences in poverty levels between Uttar Pradesh and India as a whole cannot explain why Uttar Pradesh does so much worse than average in terms of a wide range of indicators of well-being.

Uttar Pradesh: The Burden of Inertia

FIG. 1. Trends in Head-count Index of Rural Poverty in Uttar Pradesh and India, 1958–91

Source. Unpublished figures calculated by Dr Gaurav Datt, World Bank, based on National Sample Survey data.

To prevent a possible misunderstanding, we should emphasize that we have no intention of diminishing the significance of high poverty levels in Uttar Pradesh, either as an indicator of material deprivation in that state, or as a cause of other kinds of deprivations. Obviously, the low level of incomes in Uttar Pradesh is a major (p.48) constraint on individual and social opportunities. The point is that, in this particular respect, Uttar Pradesh is not very different from India as a whole, so that the causes of Uttar Pradesh's extraordinary backwardness in. terms of basic social achievements (such as child survival and elementary education) have to be sought elsewhere.

The same point applies, with even greater force, when we compare Uttar Pradesh and Kerala. Both states, according to the available estimates, had similar levels of poverty (as measured by the head-count ratio) in 1987–8, the latest year for which state-specific estimates of the head-count ratio are available.25 They are, however, (p.49) poles apart in the scales of literacy and mortality indicators. The proportion of illiterates among females aged 7 and above, for instance, is more than five times as high in Uttar Pradesh as in Kerala (75 per cent and 14 per cent, respectively), and a similar observation applies to, say, the infant mortality rate (98 per thousand in Uttar Pradesh, compared with 17 per thousand in Kerala).26 Even in a comparison between Uttar Pradesh and South India, much the same pattern applies, with South India doing much better than Uttar Pradesh in terms of literacy and child survival, despite similar levels of poverty. As will be argued further on, these astonishing contrasts have much to do with the nature of public action in the respective states.

East and West

The hypothesis that material poverty is not the main cause of Uttar Pradesh's social failures receives some support from a consideration of regional contrasts within the state. The main clue in this respect comes from a comparison of eastern and western Uttar Pradesh. These two regions, with roughly equal population sizes, account for about three-quarters of the total population of the state. The western region, which has enjoyed significant economic growth during the last three decades, is now considerably more prosperous than the eastern one. The head-count ratio of rural poverty in western Uttar Pradesh, for instance, was only 26 per cent in 1987–8, compared with 43 per cent in eastern Uttar Pradesh (see Table 5). Similarly, real wages in western Uttar Pradesh appear to be about twice as high as in the eastern region.

Table 5. Uttar Pradesh: Regional Contrasts


Share of total UP population, 1991 (%)

Child mortality rate, 1981a

Female—male ratio, 1991

Estimated rural birth rate, 1988–90(per 1,000)

Literacy rate, age 7+, 1991 (%)

Incidence of rural poverty, 1987–8b

Index of real wages for male agricultural labourers,1989–92c




















































All Regions










South India









(Notes. a) Probability of a new-born child dying before age 2 (multiplied by 1,000). This is considered to be the most reliable among the district-level estimates of infant and child mortality that have been calculated from 1981 census data (see Government of India, 1988, p. 2; and Government of India, 1989, p. 2). As recommended in these publications, we have used the ‘graduated estimates’, whenever available.

(b) Head-count ratio (percentage of the population below the poverty line). Note that these estimates are not directly comparable with those presented in Table 1, since they are based on a lower ‘poverty line’.

(c) In Rs/day at 1970–1 prices.

Sources. Birth rates: Swamy and Sinha (1994), pp. 84–5, and Sample Registration Bulletin, July 1994, pp. 15–20; real wages: calculated by Bipul Chattopadhyay (Institute of Economic Growth, Delhi) based on data published in Agricultural Wages of India, various issues; child mortality rates: calculated from 1981 census data presented in Government of India (1988), pp. 196–210, and Nuna (1990), p. 119, using estimates of birth rates in Mari Bhat (1995) for constructing appropriate weights; incidence of poverty, 1987–8: Drèze and Srinivasan (1995), based on a special tabulation of the National Sample Survey (43rd round); female—male ratios and literacy rates: calculated from Nanda (1992), pp. 13, 210–14, and 294–306. For each indicator, this table presents the latest available figures at the time of writing.

In spite of this economic advantage, western Uttar Pradesh fares no better than eastern Uttar Pradesh in terms of the available indicators of well-being and social advancement. This applies, for instance, to mortality and fertility levels. In 1981 (the latest year for which region-specific mortality data are available), western Uttar Pradesh had a higher child mortality rate than eastern Uttar Pradesh, with a particularly large east—west gap for female children (see Table 5). Western Uttar Pradesh also has considerably higher fertility levels.27 Similarly, western Uttar Pradesh has failed to take advantage of its comparative prosperity to achieve any kind of lead (p.50) (p.51) in the field of literacy and education. And as we saw earlier, gender inequality seems to be more extreme in western Uttar Pradesh than in any other part of the state.

All this does not mean that living conditions in general are no better in western than in eastern Uttar Pradesh. The fact that the incidence of poverty, based on conventional measures, is much lower in the former region, is an achievement of major importance. But the absence of any outstanding achievement in western Uttar Pradesh in many crucial fields, despite a significant lead in terms of income-based indicators, points once again to the importance of other neglected bases of social progress, such as public involvement in the fields of basic education and health care as well as women's effective participation in society and politics.

2.6 The Role of Public Services

The preceding observations suggest that the primary failure of economic development in Uttar Pradesh does not relate so much to the low level of private incomes (a problem which the state shares with much of the rest of India) as to the transformation of private incomes into well-being achievements. One of the relevant determinants of these achievements is the reach and functioning of public services. In this respect, the record of Uttar Pradesh is extraordinarily poor.

Some preliminary indications of this failure are given in Tables 6 and 7, which also include corresponding figures for South India.28 (p.52) (p.53) The figures speak for themselves. Whether we look at health care provisions, or at educational facilities, or at the public distribution system, or indeed at almost any essential public services for which relevant data are available, Uttar Pradesh stands out as a case of resilient governmental inertia as far as public provisioning is concerned. Here again, the contrast with Kerala is particularly striking, but even the contrast with South India is quite startling.

Table 6. Selected Indicators Relating to Public Services

Uttar Pradesh

South India



Percentage of recent births (1992–3)

preceded by

Tetanus vaccine




Antenatal checkup




Proportion of births taking place in medical institutions, 1991 (%)




Proportion of children aged 12–23 months who have received some vaccination, 1992–3 (%)




Proportion of villages with medical facilities, 1981 (%)




Number of hospital beds per million persons, 1991





Proportion of rural settlements of 300 persons or more having primary school, 1986 (%)




Proportion of primary schools held in ‘open space’, 1986a (%)




Proportion of primary schools with only one or two teachers, 1986 (%)




Proportion of rural children aged 12–14 who have ever been enrolled in a school, 1986–7 (%)









Other services

Proportion of the rural population receiving subsidized cereals from the public distribution system, 1986–7




Per-capita supply of food grains through the public distribution system, 1986–7 (kg/year)




Proportion of rural households with electricity connection, 1991




(Note. a) Not including schools held in ‘tents’, ‘thatched huts’, or ‘kachcha buildings’.

Source. Compiled from Government of India (1992a, 1993a, 1993b), Tyagi (1993), Visaria et al. (1993), International Institute for Population Sciences (1994a, 1994b), Parikh (1994), and the District Census Handbooks of the 1981 census. The original sources are: Census of India 1981, Census of India 1991, National Sample Survey, National Family Health Survey, Fifth All-India Educational Survey, Sample Registration System. For further details on sources, and for state-specific figures, see Drèze and Sen (1995), Statistical Appendix.

Table 7. Percentage of Villages with Selected Public Amenities, 1992–3

Uttar Pradesh

South India


Health facilities

Any health facility




Primary health centre or sub-centre




Trained birth attendant




Mobile health unit




Other institutions





Fair price shop




Cooperative society




Mahila mandal




Youth club




Source. International Institute for Population Sciences (1994a, 1994b), Tables 11.3 and 11.4, based on the National Family Health Survey 1992–3.

Interestingly, these contrasts cannot be plausibly explained in terms of differences in levels of government expenditure alone. The proportions of government expenditure allocated to health and education, for instance, are similar in Uttar Pradesh and South India. In absolute terms, per-capita government expenditure on education in Uttar Pradesh is only 23 per cent below the corresponding figure for South India, with a similar proportionate gap in the case of health.29 (p.54) The restricted scope and quality of public services in Uttar Pradesh, in comparison with South India, seems to have less to do with the level of government expenditure than with distorted patterns of social spending as well as with the defective functioning of the services in question. We will return to this issue, with special reference to the low effectiveness of the schooling system in Uttar Pradesh. The state of health services is no less alarming, especially in rural areas, as numerous studies have documented.30

One particular aspect of the dismal functioning of health services in Uttar Pradesh is worth specific mention, since it relates closely to other themes of this paper, including the central role of gender inequality and women's oppression in Uttar Pradesh's social failures. This concerns the large-scale displacement of health care services in rural Uttar Pradesh by family planning campaigns, focusing mainly on female sterilization, and often involving heavy-handed methods. (p.55) A few testimonies, taken from recent studies of health care services in the state, may be worth citing on this point:

‘the rampages of the family planning programme are particularly devastating … [the] preoccupation with attaining of the given family planning targets has had devastating effects on the other health activities’ (Budakoti, 1988, pp. 153–4);

‘the sterilization target achievement has the highest priority or rather the single priority’ (Maurya, 1989, p. 167);

‘under [health and nutrition education programmes] the ANMs/FHWs [Auxiliary Nurse Midwives and Family Health Workers] were motivating the people only for adoption of family planning practices to achieve their targets’ (Ashok Kumar, 1990, p. 70);

‘as from the highest level to the lowest everybody is asking regarding sterilization targets … health workers under the pretext of work of motivating sterilization cases neglect other work’ (Shah, 1989, p. 120);

‘in [rural and tribal] blocks it was observed that the visits of the medical and para-medical staff were irregular as most of the time they were busy with family planning and other campaigns’ (Krishnamurthy and Nadkarni, 1983, p. 51);

‘during the main months for family planning campaigns (usually December to March) virtually all energies of maternal and child health staff may be directed towards those ends [i.e. family planning targets]’ (Jeffery et al., 1989, p. 216);

‘in every scheme relating to the welfare of the rural poor, the beneficiaries are asked to practice family planning by undergoing vasectomy, or to sponsor vasectomy cases in order to avail of the benefits provided by schemes’ (H.N. Singh, 1993, p. 35).31

It is important to note that the neglect of public services in Uttar Pradesh is not confined to specific programmes, such as those for (p.56) which data are presented in Tables 6 and 7. Rather, it is a case of comprehensive failure of social provisions in a wide range of fields, including basic education, land reform, child immunization, public distribution, maternal health, social security, public works, environmental protection, anti-poverty programmes, among others. There are few exceptions to this pattern. Nor is it easy to cite any example of a successful or innovative public programme relating to the promotion of human well-being on a widespread basis. We will return to this issue, and to the social and political roots of these failures, in section 5.

2.7 Women's Agency

The issue of gender inequality in Uttar Pradesh was introduced in section 2.3, mainly from the point of view of its consequences for women's well-being. Another crucial aspect of unequal gender relations in Uttar Pradesh is the suppression of women's agency in society, with extensive implications not only for the well-being of women but also for economic development and social progress in the society as a whole. The social importance of women's agency is discussed in the last chapter in this volume with specific reference to demographic outcomes, but it is relevant in other fields as well, including those of political and social change.32

In Uttar Pradesh, extreme social restrictions on women's freedom of movement and activities suppress women's agency by physically confining them to the domain of the household, and even diminish their ability to act effectively within that domain.33 Low female participation in education is one aspect of this general pattern of women's limited interaction with the outside world. Similarly, less than half of all married women in Uttar Pradesh have regular exposure to any mass medium (compared to nearly 80 per cent in both Tamil Nadu and in Kerala).34 Women's limited opportunities to acquire education (p.57) and information (whether through schooling, social interaction, mass media, or other means) is bound to affect their ability to play an informed role in the family and society.

Another indication of the restricted agency roles of women in Uttar Pradesh is the low level of female labour-force participation. Female labour-force participation is an important indicator of gender relations in at least two respects. First, there is much evidence that gender inequality within the family (e.g. the survival disadvantage of girls vis-à-vis boys) tends to be lower when adult women have wider opportunities for gainful employment.35 Second, participation in gainful employment is one indication of the general participation of women in society, outside the narrow confines of domestic work.

The 1991 census counts only 8 per cent of women in Uttar Pradesh as ‘main workers’ (persons involved in economically productive employment for at least half of the year).36 The corresponding figures for India and South India are 16 and 24 per cent, respectively. Among all Indian states, only Punjab and Haryana have lower female labour-force participation rates than Uttar Pradesh (4 and 6 per cent, respectively). As with female—male ratios, female labour-force participation rates are even lower in western Uttar Pradesh (2.5 per cent) than in Punjab or Haryana.37 The fact that the talents and initiative of most women in Uttar Pradesh are overwhelmingly focused on domestic work represents a colossal suppression of their potential contributions in other fields.

Some of the connections between women's agency and social progress are discussed in the companion volume (Drèze and Sen, 1995) on the basis of broad comparisons between Indian states. It is worth noting that even within Uttar Pradesh, some important regional patterns seem to have much to do with gender relations and women's agency. The contrast between the hills and the plains is particularly relevant here. One important feature of the Himalayan region of Uttar Pradesh (consisting of eight districts in the northern part of the state) is a high level of female labour-force participation (p.58) —higher, in fact, than the average for India and even South India. Interestingly, this is also the only region of Uttar Pradesh where the female—male ratio is above unity, and where female children have a survival advantage over male children (see Table 8). The Himalayan region is also a relatively ‘progressive’ region of Uttar Pradesh in many other respects (e.g. it has considerably lower rates of mortality, fertility, and illiteracy), and it is quite plausible that these achievements partly reflect a more active and equal participation of women in the society as a whole.38

Table 8. Gender-Mated Indicators for UP Regions, 1981


Female—male ratio

Ratio of female q(2) to male q(2)a

Female labour force participation rateb (%)

Crude female literacy rate (%)

Percentage of girls aged 5–14 who are attending school

Total fertility rate

All ages

Age 0–9









































All Regions








South India








Notes. (a) q(2) is the probability of a new-born child dying before age 2 (as in Table 5).

(b) Proportion of ‘main workers’ in the total female population.

Sources. Calculated from district-level data published in Nuna (1990), Nanda (1992, pp. 294–306), Government of India (1989), Census of India 1981, Primary Census Abstract, Part II–B, Census of India 1981, Part IV–A, Social and Cultural Tables, Tables C-l, C-2, and C-4, and Census of India 1981, General Economic Tables, Part II. Each regional figure is a weighted average of the relevant district figures, with the size of the relevant population group in each district being taken as the relevant weight. In the case of male (female) q(2), the relevant weight for each district is the number of male (female) births; in the absence of district-level data on the number of births by gender, we have used the total number of births in the district as weight.

We end on this subject by noting that, like female—male ratios, female labour-force participation rates in Uttar Pradesh appear to have decreased rather than increased since the beginning of the century (there may, in fact, be a causal connection between these two distinct trends). According to census data, the ratio of female to male participation in the labour force was 0.43 in 1901, but only 0.13 in 1991. Changes of definitions and survey techniques may have contributed to this apparent decline, but are unlikely to account for the whole of it (especially since the tendency, in recent decades at least, has been to promote a more inclusive definition of female labour-force participation). Here again, there is little evidence of economic development having done very much for gender equality in Uttar Pradesh.

2.8 Discussion

Uttar Pradesh is one of India's most backward states, as far as the living conditions of the population are concerned. The proximate causes of that backwardness include a high level of poverty, as measured by conventional indices such as the head-count ratio. In that particular respect, however, Uttar Pradesh does not really stand out among Indian states. In order to explain, say, the exceptionally high levels of mortality and fertility in Uttar Pradesh, we have to take note of other social failures that have intensified the deprivations associated with low income levels. The contrasts between (p.59) (p.60) Uttar Pradesh and South India point to three social failures of deep significance: low levels of education, the restricted role of women in society, and the poor functioning of public services.39

These three particular failures (which are, of course, interrelated) do not exhaust the range of relevant influences, but they certainly play a central role in the persistence of endemic deprivation in Uttar Pradesh. Consider, for instance, the problem of high infant mortality. In Uttar Pradesh, a large proportion of infant deaths are due to tetanus and diarrhoea.40 In both cases, cheap and effective means are available for preventing or curing the disease (vaccination, in the case of tetanus, and oral rehydration, in the case of diarrhoea). If so many children continue to die of tetanus and diarrhoea in Uttar Pradesh, it is not primarily because their parents are too poor to do something about it.41 More influential reasons arc: (1) inadequate public provisions for primary health care (including child immunization), especially in rural areas; (2) low educational levels, leading inter alia to a poor understanding of the causes and possible prevention of these elementary diseases, and to ineffective utilization of the services that are available;42 and (3) the suppression of women's (p.61) informed agency in the family and society, including in matters relating to child survival.43

Underlying the specific failures relating to public services, elementary education, and women's agency, is a deeper failure to achieve the kind of basic social change that facilitates progress in these fields. One aspect of the ‘inertia’ that accounts for slow social progress in Uttar Pradesh is the apathy of the state, but an equally important factor is the failure of civil society to challenge oppressive patterns of caste, class, and gender relations. The society of Uttar Pradesh remains steeped in traditional inequalities, which makes it that much harder to achieve widespread literacy, to run efficient public services, or to promote the agency of women in social and political matters. For instance, the conservative nature of gender relations makes it very difficult for women to work as teachers or doctors in rural areas, even if they have the required qualifications.44 Similarly, as will be discussed further on, the resilience of caste and gender inequalities is a strong obstacle to the spread of basic education. And the highly divided nature of the rural society in Uttar Pradesh has seriously constrained the scope for collaborative public action (e.g. the provision of local public services) at the village level. In that sense, social change — or lack or it — occupies centre-stage in the story of Uttar Pradesh's past failures and possible future achievements.

We will return to this in section 5, Before that, the next two sections present a more detailed empirical investigation of some of the issues we have been concerned with so far, with special reference to primary education.

(p.62) 3. Schooling in Uttar Pradesh: A Field Investigation

3.1 The Setting

In this section, we report the findings of an informal field investigation of the functioning of primary schools in rural Uttar Pradesh. The enquiries we have carried out (in February–March 1994) essentially consist of unannounced visits to primary schools, supplemented by detailed discussions with local residents, teachers, officials, and activists. Altogether, we have visited 16 villages in four different districts: Moradabad, Rae Bareli, Pratapgarh, and Banda. In each district, we had made prior contact with a local person, who accompanied us to different villages around the place where he or she lived.

This method obviously falls short of guaranteeing a representative sample of villages and schools, and the findings reported below should be read in that light. If we think that they are worth reporting, it is because this informal investigation brings out striking regularities in the functioning of primary schools in different areas, which cannot reasonably be attributed to chance.45

Our survey involved unannounced visits to all government schools in the sample villages. The majority of these schools were primary schools, although some sample villages also had a ‘middle’ or secondary school.46 In this section, the term ‘sample schools’ specifically refers to the government primary schools of the sample villages. All the sample villages, except two, had at least one government primary school. Some of them also had a private school. Although we collected a good deal of indirect information on these private schools, we only visited a few.

The sample villages are spread over four of the five ‘regions’ of Uttar Pradesh, namely Western (Moradabad), Central (Rae Bareli), Eastern (Pratapgarh), and Southern (Banda).47 Rae Bareli (p.63) and Pratapgarh districts, however, are adjacent to one another. In terms of agro-climatic zones and social conditions, then, our surveys effectively covered three distinct regions of the state: Moradabad in western Uttar Pradesh, a ‘Green Revolution’ district with a relatively dynamic economy; Rae Bareli and Pratapgarh on the central—eastern border, formerly a stronghold of Thakur zamindars (whose presence is still felt), where agricultural growth is of more recent origin and non-agricultural activities are quite limited; and the eastern part of Banda in southern Uttar Pradesh, where an overwhelmingly scheduled-caste and adivasi population still lives in the oppressive shadow of powerful ‘Dadu’ Brahmin landlords.

In Moradabad district, our investigation focused on Palanpur, a village familiar to one of us, and the surrounding villages. On Palanpur itself, a good deal of useful information is available from earlier household surveys, covering the 1957–94 period.48 Some use of that supplementary information will be made in this section and the next one.

3.2 Accessibility of Primary Schools

As was mentioned earlier, all but two of the sample villages had at least one government primary school. The two villages without a government primary school (one in Pratapgarh, the other in Banda) were relatively small.49 In both cases, local residents cited tangible political factors as being responsible for the absence of a school. In the Pratapgarh village, we were told that local Thakur landlords, who wielded a great deal of power, had obstructed the provision of a primary school. In Banda, we visited a scheduled-caste village where a school officially sanctioned under a scheduled-caste development scheme had failed to materialize. Instead, the school was built in a neighbouring high-caste settlement within the same administrative area. These two exceptions, while affecting a relatively small number of persons, are of some political significance, in so far as they illustrate the absence of a well-accepted social consensus on the need to universalize primary education in Uttar (p.64) Pradesh, and the lack of effective political organization among disadvantaged groups (on which more further on).

Even these two villages, however, were within walking distance of government primary schools in neighbouring villages or settlements. In this respect, the sample villages were not atypical of rural Uttar Pradesh as a whole, where the proportion of the population living more than two kilometres away from a primary school is only around 2 per cent (Tyagi, 1993, p. 54). In other words, the existence and accessibility of schools does not seem to be the main cause of persistent educational backwardness in this region (even though much scope remains for improving the schooling infrastructure, as will be discussed shortly). As we move from schools to schooling, however, the educational situation in rural Uttar Pradesh appears in an extremely poor light.

3.3 Physical Condition of Schools

The first thing that struck us on approaching most of the sample schools was the dilapidated condition of the buildings, and, linked to that, the poor utilization of whatever facilities were available. Not a single one of the schools we visited had full use of the building. Buildings were not usable due to prolonged decay, pending repairs, incomplete construction, and lack of maintenance. In one village, the school building was being used by the local landlord as a cattle shed. In another, the headmaster (who was from outside the village) had made part of the school buildings his residence.

In several villages, we found evidence of school buildings falling into disrepair due to lack of maintenance, and then being abandoned in favour of new school buildings rather than repaired. In some of these cases, even the new building was not usable.50 The most commonly-used part of the school building was the veranda, where children of all grades were often huddled together. This space provides some shelter from the elements, but is open on three sides (p.65) and hardly conducive to concentrated study. In many cases we found teachers and children sitting under trees near the school building.

The facilities available in most of the sample schools were minimal. Typically, the school building was completely bare, except for the occasional table and chair used by the headmaster. In several cases, we heard that whatever furniture had existed earlier had been appropriated by the village headman, the school teacher, or other influential local residents. Widely-used teaching material was limited to notebooks, slates, and some basic textbooks.

The pathetic physical condition of the sample schools is in sharp contrast with the claims recently made by the Government of India in connection with the expansion of schooling infrastructure under Operation Blackboard. This scheme was initiated in 1987–8 with the stated objective of providing all primary schools with some essential facilities, including ‘(i) a building comprising at least two reasonably large all-weather rooms with a verandah and separate toilets for boys and girls, (ii) at least two teachers in every primary school, as far as possible one of them a woman, and (iii) essential teaching learning equipment including blackboards, maps, charts, toys, and equipment for work experience’ (Government of India, 1994a, p. 31). Recent reports of the Department of Education state that Operation Blackboard had been ‘implemented’ in 91 per cent of the country's primary schools by the end of 1992–3, rising to 99.9 per cent by the end of 1993–4.51 In contrast with these cheerful statements, there is no evidence of Operation Blackboard having had much practical impact in the sample schools.52

3.4 Teacher Attendance and Teaching Practices

Teacher absenteeism was endemic in the sample schools. By all accounts, this is the most basic problem of the schooling system in the sample villages.

(p.66) One indication of the magnitude of this problem is the simple fact that two-thirds of the teachers in the sample schools were absent, for one reason or another, at the time of our unannounced visit. In other words, the attendance rate of teachers in these schools is only around one-third.53 Only two of the fifteen schools we visited had full attendance of teachers at the time of our visit In some cases, particularly when the school had only one teacher, the absence of the teacher(s) implied that the school remained closed for the day. On several occasions, we found that a school had been closed without prior notice for the day, or a large part of it, because the teacher(s) had decided to engage in some other activity.

Another aspect of the problem of teacher absenteeism is that most teachers come late and leave early. We rarely found a school to be open on time in the morning, or after 12.30 in the afternoon, when the lunch break is supposed to begin. Schools rarely reconvene after this break. In effect, therefore, the school day lasts for under three hours on average (when the school opens at all).54

The consequences of low teacher attendance are all the more serious given that the official number of days of teaching in a year is quite low in the first place — about 220 according to government figures.55 Combined with a teacher attendance rate of about one-third, this implies that the actual number of full teaching days per teacher per year in the sample schools may be as low as 75 or so.

Further, we found that teachers actually performed very little teaching even when they were present. In fact, no active teaching was taking place in any of the fifteen sample schools at the time of our visit.56 At best, the teacher(s) had given exercises to the pupils. (p.67) When they were present at all, teachers in the sample schools were found to be engaged in one or more of the following activities: supervising children; playing cards; talking with each other; talking with visitors (other than ourselves!); reading comics; preparing rolls for the forthcoming election of the management committee of a local credit cooperative.

In effect, the primary schools we visited were little more than child-minding centres (that too, working only occasionally). In most cases (even when more than one teacher was present), we found that all the children had been assembled together in one place, irrespective of age or grade, and were only expected to maintain a semblance of order. Some of them were working on exercises given by the teacher, and others were teaching themselves or each other. But many more were just playing or passing time. Supervision took one of the following forms: watching the children from a desk or chair; asking one senior child to maintain order; letting the children look after themselves. In many schools, the ambience was nothing short of chaotic.

These direct observations were amply confirmed by informal conversations with parents and local residents. Shirking and absenteeism on the part of school teachers was widely perceived as the fundamental problem of government schools in all the sample villages. The quality of whatever teaching does take place is also a matter of widespread popular concern.

3.5 Female Teachers

None of the sample schools had a female teacher. Even in the few schools that had separate primary sections for girls, we did not (p.68) encounter any female teacher. We were told that female teachers typically work in larger agglomerations, or in villages close to main roads.

This explanation is, indeed, the most plausible way of reconciling our field observation with official statistics, which indicate that 18 per cent of all primary-school teachers in Uttar Pradesh are female (Government of India, 1994a, p. 289). The distribution of female teachers, it appears, is highly uneven, with the teaching staff of primary schools being almost exclusively male in large parts of the state, especially the less accessible ones.57

The uneven distribution of female teachers in Uttar Pradesh is only one aspect of the issue of gender imbalance in teaching. Even the average of 18 per cent for the state as a whole is low; the corresponding figures for South India and Kerala are 39 per cent and 66 per cent, respectively.58 In fact, among all Indian states, Uttar Pradesh has the lowest proportion of female teachers. Further, the proportion of female teachers in Uttar Pradesh has been declining in recent years, in spite of the supposed emphasis of official policy on a rapid expansion of female involvement in primary-school teaching.59

The low number and declining proportion of female teachers in Uttar Pradesh, and their particularly low presence in rural areas, relate to the restrictive and unequal nature of gender relations in that state. As was discussed earlier, women's labour-force participation in Uttar Pradesh is extremely low, and female access to the public domain is also very limited. The low involvement of women in teaching is one reflection of this general suppression of women's agency in society. Further, the patriarchal environment makes it very difficult for female teachers to reside in (or commute to) rural areas on their own, and nor can a married female teacher expect her husband to follow her if she is posted at some distance from his own place of work.

(p.69) The absence of female teachers in a large majority of rural schools in Uttar Pradesh may well be a serious constraint on the expansion of primary education, and particularly of female education. There is, indeed, some evidence that parents often have greater confidence in sending their daughters to school if the school has some female teachers.60 The presence of female teachers in most schools is also important in so far as schooling is as much a socialization experience as a process of formal learning. The virtual exclusion of women from teaching positions in Uttar Pradesh diminishes the quality and diversity of that socialization experience, for girls and boys.61

3.6 Enrolment and Attendance

In all the sample schools, the number of pupils actually present at the time of our visit was well below — often as much as 50 per cent below — the number of children officially enrolled. Teachers also report that attendance levels vary a great deal over the year.62 While there are many cases of children not turning up for long periods (and often eventually dropping out), the most common pattern is one of erratic attendance, with children dropping in and out of school according to circumstances.

One particular aspect of this pattern is a high level of child absenteeism during periods of high activity in the agricultural cycle. Pupil absenteeism, for instance, was said to be widespread at harvest time. It might be added that public examinations are usually held in late April, which is the time of the wheat harvest in Uttar Pradesh. This basic conflict between the schooling and agricultural calendars (p.70) is a telling indication of the lack of sensitivity of official policy to the needs of the rural poor.63

Teachers, in general, did not think that actively encouraging school attendance was part of their responsibility. This attitude contrasts with the accepted practice in schools run by voluntary organizations, which often employ people to bring children. It also contrasts with accounts of how schools functioned in the past in the sample villages. According to local residents, the teachers used to consider the monitoring and promotion of school attendance as an ordinary part of their duty.

The proportion of female children among attending children was, on average, about one-third (with a similar proportion for enrolled children). This is quite close to the ratio of female enrolment to total primary-level enrolment in Uttar Pradesh as a whole (39 per cent).64 In some of the middle schools, the proportion of girls was higher. One reason for this is that boys are disproportionately enrolled in private schools, which are more important at the middle than at the primary level.

School fees in government schools were nominal: around 10 paise per month in class 1, rising to around 1.50 rupees per month in class 5. Very often school fees were in arrears. In a number of schools, it was standard practice for teachers to continue the registration of a pupil even if his or her fees were overdue, if necessary by paying the fees out of their own pockets. The motives for doing so were not altogether altruistic. In the sample schools, monitoring of teachers by district educational authorities is primarily based on periodic inspection of the school register.65 In order to avoid being transferred, a teacher has to ensure that enrolment does not fall below the official norm. By paying school fees in the names of (p.71) children who have actually dropped out (or have never been enrolled in the first place), teachers are able to maintain inflated registers and to reduce the chances of being transferred.66

3.7 Private Schools and Implicit Privatization

Little information exists on private schooling facilities in rural Uttar Pradesh. According to official statistics, less than 3 per cent of all primary schools in rural Uttar Pradesh are managed by private institutions.67 These statistics, however, only cover recognized private schools (‘aided’ and ‘unaided’), and field investigations — including our own — suggest that unrecognized private schools account for a much larger share of all private schools.

In most of the areas we visited, the network of government primary schools (classified as ‘local body’ schools in official statistics) was supplemented with formal and informal schools operated by private individuals or institutions. Most of these schools were unrecognized, and therefore unaided. Roughly speaking, private schools were of two types:68 (1) profit-oriented schools managed on commercial principles, and (2) non-profit educational institutions run by voluntary organizations such as religious missions, development agencies, community institutions, and charitable trusts.

In the sample villages, the first type was more common than the second, except in Banda district where there was an active network (p.72) of schools run by voluntary organizations. The discussion in this section therefore concentrates on profit-oriented private schools. As far as non-profit schooling institutions are concerned, we will only note in passing that these institutions, while few in number, are often well-run and well-attended. The high attendance and low drop-out rates in these schools, where teachers and managers often have a genuine commitment to the promotion of basic education (especially among disadvantaged groups), demonstrates that low attendance in government schools has more to do with the abysmally low standards of teaching and management in these schools than with any lack of interest in education on the part of children and their parents. There are indeed cases where, in the same village, a well-run philanthropic school is packed with enthusiastic children (girls as well as boys) while the local government school exudes a familiar atmosphere of desertion, apathy, and decay.69

Turning to profit-oriented private schools, one obvious characteristic of these schools is that they charge substantial fees. The actual level of fees varies a good deal between different schools and regions. In the sample villages, the average fee in private primary schools was of the order of 15 rupees per child per month. This is roughly equivalent to three kilograms of wheat, or to the daily wage of an agricultural labourer in central or eastern Uttar Pradesh (in Moradabad district, wages are a little higher, but so are school fees).

Private-school teachers are poorly qualified and poorly trained. It is generally agreed that government teachers are more competent. But private-school teachers, unlike government teachers, turn up for work and do their job. They have a strong incentive to do so, since they might lose their job if attendance declines due to poor teaching standards. In short, teachers in private schools are accountable to parents in a way that simply does not apply in the case of government teachers. This is so in spite of private-school teachers earning much lower salaries than government teachers. In private schools, teachers are often employed on a part-time basis for less than 500 rupees per month, compared with a starting salary of 2,200 (p.73) rupees per month for a government teacher (also effectively part-time!).

School attendance in private schools is significantly male-dominated, for two reasons. First, many parents are more reluctant to pay school fees for female than for male children. Second, attending a private school often involves commuting to a different village, something which female children are not easily allowed to do.70 The pro-male bias of private schooling is one reason why female children are the first victims of the poor functioning of government schools, as will be discussed further on.

A crucial problem faced by private schools in rural areas is that it is hard for many of them to obtain official recognition (maanyata) from the government. Without maanyata, private schools cannot issue recognized primary-school certificates. The standard way of dealing with this problem of non-recognition is the following: children are taught in private schools until grade 5, when they are transferred to government schools for the sole purpose of obtaining a certificate. Most private schools establish informal links with local government schools in order to implement this procedure. This system is advantageous not only to private schools, whose existence depends on it, but also to government teachers, who are able to maintain inflated rolls as a result (since the children in question are officially enrolled in government schools). Inflated rolls ensure, as was mentioned earlier, that government teachers are not transferred to other schools due to low enrolment.

The system of ‘fifth-grade transfer’ described in the preceding paragraph is only one way in which public schooling in Uttar Pradesh has been, in effect, partly privatized. Another form of implicit privatization observed in some of the sample villages is the practice of government teachers hiring private teachers to mind the children while they do something else (this is viable for the government teachers, given the large salary differentials between government and private teachers). Some government teachers also give private tuitions in return for fees (during or outside their normal working hours), sometimes to the same children whom they are supposed to teach at school.

(p.74) 3.8 Failure of Public Schooling

The picture emerging from this field investigation is quite bleak, to put it mildly. Teaching and management standards in government schools are extremely poor, and play a major part in the persistence of low attendance levels. Even those children who do attend government schools receive very little education, due to high rates of teacher absenteeism and shirking as well as to crude teaching methods.

Our own observations about the dearth of teaching in the sample schools were amply corroborated by local residents, who tend to be highly critical of the functioning of the schooling system. Most local residents also take the view that teaching standards in government schools have significantly deteriorated during the last two or three decades. Specifically, it is widely agreed that the extent of teacher absenteeism and shirking has dramatically increased over this period. What is discouraging is not just that so little teaching goes on in government schools, but also that there is no sign of any improvement in that respect. On the contrary, the system has, by all accounts, decayed over time. Further, pupil-teacher ratios in Uttar Pradesh have risen at an alarming rate in recent years, e.g. by almost fifty per cent between 1981–2 and 1992–3 (Tyagi, 1993, p. 84). In that sense, the current evolution of government schooling in rural Uttar Pradesh seems to be characterized by a remarkable decline in effective quantity as well as in quality.

This finding may seem hard to square with the fact that literacy rates, after all, keep increasing even in Uttar Pradesh (as elsewhere in India). An important part of the answer may lie in the recent expansion of private schooling facilities, described as ‘mushroom growth’ by some experts.71 This development, itself partly a response to the decay of the public schooling system, may have made an important contribution to the continued expansion of educational achievements in Uttar Pradesh. It is also likely that the demand for primary education has rapidly increased in Uttar Pradesh in recent years (due to rising incomes, higher levels of parental literacy, and (p.75) related factors), inducing a larger proportion of parents to send their children to school despite the declining quality of teaching in government schools. The fact remains that, had the system of public schooling in rural areas expanded in quantity and quality, instead of stagnating, educational achievements in Uttar Pradesh would now be much higher than they actually are. The slow but steady improvements that have taken place in overall indicators of literacy and education in recent years have occurred in spite of the persistent inadequacy of public schooling at the village level, rather than as a result of positive government policies.

4. Public Policy and Schooling Decisions

As the field investigation discussed in the preceding section brings out, the failures of schooling in rural Uttar Pradesh are fairly extensive, and their roots are deep. Transforming the educational situation is not just a question of increasing public spending on education, or of accelerating the quantitative expansion of educational facilities, or of introducing ad hoc ‘schemes’ to supplement the basic schooling system, or of undertaking a short-term ‘campaign’ for total literacy. Such initiatives would certainly be useful (and have been lacking in Uttar Pradesh, in comparison with many other Indian states), but the primary issue is to ensure that every village in the state has a well-equipped, well-staffed, well-functioning, and well-attended primary school. This section discusses a few basic issues relating to that essential goal.

At a general level, it may be useful to distinguish between the broad issues of provision and utilization of schooling facilities. While different authors have tended to put different emphases on each, a successful expansion of educational achievements obviously depends both on the adequate provision (and functioning) of schooling facilities and on the widespread utilization of these facilities. Both aspects are considered in this section.

4.1 Provision and Accountability

Our field investigation suggests that the physical supply of schooling infrastructure is no longer the main constraint on educational (p.76) expansion at the primary level, in the sense that the vast majority of the rural population lives within short distance of a primary school This diagnosis is corroborated by secondary data. According to the Fifth All-India Educational Survey, in 1986, 89 per cent of the rural population in Uttar Pradesh lived within 1 km of a primary school, and 98 per cent lived within 2 km.72

Considerable scope remains, however, for expanding and improving the schooling infrastructure in rural Uttar Pradesh. For instance, effective access of young girls to primary schools may depend on these schools being located within the village or hamlet where they live, given that it may be socially unacceptable for them to wander outside the village.73 In 1986, less than half of all rural settlements of 300 persons or more in Uttar Pradesh had a primary school, compared with 77 per cent in India as a whole and 88 per cent in South India. In that respect, Uttar Pradesh was the worst-performer among all Indian states.74 Similarly, effective access of all female children to primary education may depend on a strong presence of female teachers in the local schools. And in that respect, too, as we saw earlier, Uttar Pradesh is at the rock-bottom of the scale among all Indian states.

The reach of the schooling system, thus, is not quite universal, despite a major expansion of physical infrastructure in the post-independence period. We have also noted how the facilities available in the sample schools are, in most cases, extremely poor. Having said this, the most striking weakness of the schooling system in rural Uttar Pradesh is not so much the deficiency of physical infrastructure as the poor functioning of the existing facilities. The (p.77) specific problem of endemic teacher absenteeism and shirking, which emerged again and again in the course of our investigation, plays a central part in that failure. This is by far the most important issue of education policy in Uttar Pradesh today.75

The main constraint in the sample villages is not that there are too few teachers (though this is certainly the case, too), but that the appointed teachers spend an alarmingly small part of their time teaching. Many of them regard teaching as part-time secondary employment, which they combine with other — often more valued — economic pursuits. Even when they are present at the school, with no opportunity to pursue other productive activities, they have little motivation to engage in active teaching.

There is a sharp contrast, in this respect, between government and private schools. In private schools, there is virtually no problem of teacher absenteeism or shirking. Unlike government teachers (whose earnings are not linked to performance), private-school teachers have strong incentives to work hard, since failure to do so easily leads to dismissal. Indeed, in a private school the manager or headmaster is accountable to the parents, who pay fees and expect tangible teaching services in return.76 A shirking teacher, therefore, cannot expect much protection or indulgence from his or her superiors. There is a chain of accountability that stretches from the providers of the service to its ultimate recipients.

In government schools, this chain of accountability is extremely weak. Notionally it exists in the form of the school inspection system, whose agents (the government-employed school inspectors) may be thought of as acting on behalf of the users. The links, however, are fragile and tenuous. Users of the service (i.e. parents and the community at large) have little direct control over the activities of the school teachers or inspectors. The only control they have is through their political representatives who can approach the (p.78) district educational authorities. In practice this amounts to very little, since basic education is low on the political agenda, and also because the holders of political power are more interested in using the schooling establishment as a means of extracting and dispensing public resources for their own advantage than in promoting the cause of widespread literacy.

In the absence of any effective control by the actual users, teachers are only subject to routine formal supervision from the government machinery. This official supervision mechanism, too, is quite ineffective, and it is worth considering exactly where it breaks down. One part of the story, which has already been mentioned, is that school teachers have permanent jobs. Given that it is virtually impossible to fire a shirking teacher, the official disciplining device consists of transfers. In practice, teachers are transferred only if the number of children in the school falls below stipulated norms. Correspondingly, the ‘inspection’ system is almost exclusively based on sporadic review of the school registers by school inspectors.

There is evidence that government school teachers do take the threat of transfer seriously enough to ensure that the required enrolment levels are maintained, at least on paper. Their response to this threat, however, is not less shirking, but manipulation of records to avoid sanction. This, as we have seen, takes the form of fictitiously enrolling children who have effectively dropped out (or who have never attended school in the first place), if necessary by paying fees on their behalf. Another method is to register private-school pupils in government schools. The phenomenon of widespread absenteeism and shirking of government teachers in the sample villages is ample testimony of the inadequacy of the official monitoring system.

We conclude on this subject with a few supplementary remarks. First, informal observations suggest that teacher absenteeism and shirking are more common in village schools, particularly those with few teachers, than in schools situated in towns or large villages. Schools with a comparatively large number of teachers tend to be organized on hierarchical lines, with greater scope for supervision and peer pressure, and less opportunity for collusive shirking. Larger schools are also more visible, and therefore more exposed to public scrutiny. Another possible factor is that, among schools situated in larger agglomerations, there is a modicum of competition for child enrolment, making it harder for a school to get away with non-provision (p.79) provision of teaching services. It is in the small and relatively isolated village schools, where one or two teachers are expected to act as their own supervisors, that the problem of inadequate work incentives seems to be most acute. A corollary of this observation is that the strengthening of teaching staff in small village schools may have an important role to play in improving teaching practices.

Second, teacher absenteeism is facilitated by the existence of a significant nexus between the local elites and the public schooling system. Many teachers have close relations with the local elite, not only because of their relatively high earnings by local standards, but also because political connections were often instrumental in getting them appointed in the first place. There is a problem of ‘adverse selection’ here, as high salaries attract candidates with privileged political or social connections.77 The crucial role played by school teachers at the time of elections, census enumerations, and related events is another common basis of expedient alliances with the local elite. Based on these alliances, many teachers have ample political protection, further helping them to evade scrutiny and avoid sanctions.

Third, to be fair to the teachers, it should be mentioned that their professional environment is in many ways extremely demotivating,78 While their salaries are quite high by local standards, and while many of them have privileged connections, teachers seem to enjoy little general social esteem, and their status in the government hierarchy is low. More importantly, the working conditions in village schools (including the frequent need to cope with multi-grade teaching) make the task of conscientious teaching highly challenging. In these circumstances, it is not surprising that even teachers who are otherwise strongly motivated find it hard to maintain high teaching standards.

Finally, it is difficult to see how the problem of lack of accountability of the schooling establishment can be effectively addressed without the active involvement of concerned parents and local communities. The current emphasis of public policy on the (p.80) ‘decentralization of school management’, based on panchayati raj institutions, can be seen as giving some recognition to that need. But there has been far more promise than action, as far as decentralization is concerned. As things stand, village-level and even block-level representative institutions have virtually no control over any significant decisions relating to school management. Further, given the current political links between the schooling establishment and the rural elite, formal decentralization cannot be expected to achieve very much unless it goes hand in hand with more active political mobilization of disadvantaged groups.79 A genuine transformation can only be achieved if schooling is taken up as a major political issue by popular organizations and social movements.

4.2 Poverty and the Demand for Schooling

Poverty is often assumed to be the main reason why many Indian parents do not send their children to school. One version of this story, which is particularly popular in official circles, is that poor parents cannot afford to send their children to school because their labour makes a crucial contribution to the household economy.

In the absence of firm empirical evidence on this issue, it is difficult to assess the actual importance of poverty as an obstacle to widespread schooling.80 Literacy rates and school attendance certainly do go down as one considers progressively poorer sections of the population.81 What is far from clear, however, is to what (p.81) extent this correlation reflects an income effect as such, rather than the influence of some other variables that are themselves correlated with income. Examples of possibly relevant variables include the literacy of parents, the quality of available schooling facilities, and the social support which different sections of the population receive in pursuit of their educational aspirations.82 What looks like a problem of poverty may often turn out, on closer examination, to have more to do with other factors.

To illustrate this point, consider literacy rates in different per-capita expenditure groups in rural and urban areas. According to National Sample Survey data, in 1986–7 the literacy rate in the second-highest quintile of the per-capita expenditure (PCE) scale in rural areas was 48 per cent. The literacy rate in the lowest quintile of the PCE scale in urban areas was a little higher (50 per cent), even though the average PCE in that group was only half as high as the average PCE in the former group. To put it another way, everyone in the first group is above the official ‘poverty line’, while no-one in the second group is, and yet the literacy rates in the two groups are very similar. One plausible reason for this contrast is the better access of the urban population to schooling facilities.

Recent empirical investigations tend to confirm that the importance of poverty as a cause of persistent illiteracy has often been exaggerated, and other influences underplayed. In a recent analysis of inter-district variations in child labour, for instance, Labenne (1995) finds that poverty has relatively little explanatory power after controlling for adult literacy, schooling facilities, caste, and gender. Maharatna (1995), based on a study of child activities in rural West Bengal, argues that the opportunity cost of child labour is quite low even among very poor rural households. S. Sinha (1995), drawing on many years of experience with child labourers in rural Andhra Pradesh, also argues that ‘parents do want their children to be educated and poverty as a limiting factor is highly overrated’ (p. 40).83

(p.82) Our own field investigation lends some support to this diagnosis. In government schools, fees are extremely low (almost symbolic, at the primary level), and we found no evidence of non-fee payments being demanded from parents by school authorities. There are some cash costs in the form of items such as clean clothes, slates, and books, but even these are not high given the highly informal mode of operation of small village schools (uniforms, for instance, are not required). The opportunity cost of children's time is certainly an important consideration for some parents, but school hours are short, and schooling can be combined with a substantial contribution to the household economy at other times.

The point is that the willingness of parents to bear these costs, such as they are, and to coax their children into going to school, may depend crucially on the quality of the schooling services they obtain in return.84 The importance of material deprivation as a factor of non-attendance, therefore, has to be evaluated in the light of the poor functioning of the schooling system in many areas. In this connection, it is worth recalling the strong popular appeal of well-functioning primary schools managed by voluntary organizations, even among economically deprived groups (see section 3.7). These positive experiences indicate the possibility of achieving widespread literacy in the younger age groups even when a large part of the population is still quite poor.85 This is not to deny that poverty makes it harder to send one's children to school, or that this disincentive requires specific attention in public policy, in addition to the issue of school management and teaching practices.

4.3 Educational Achievements and Social Inequality

As we saw earlier, one feature of the educational situation in Uttar Pradesh is the existence of large disparities in literacy achievements (p.83) between different regions and social groups. The problem of low average literacy rates is compounded by large inequalities, reflected in abysmally low literacy rates for the most disadvantaged sections of the population. Scheduled-caste women, for instance, remain almost entirely illiterate in a majority of districts (see section 2.3).

Similar disparities can be observed at the village level. To illustrate, Table 9 presents some information on literacy rates and school attendance in the village of Palanpur (Moradabad district) in 1993.86 As this table indicates, gender and caste-based inequalities in educational achievements can be extremely large even within a single village. It is remarkable, for instance, that the female literacy rate in Palanpur varies from zero among the Jatabs (Palanpur's main scheduled caste, also known as Chamars) to 100 per cent among the Kayasths (who have a long tradition of involvement in clerical occupations and consider schooling as an essential part of every child's upbringing).87

Table 9. Literacy and School Attendance in Palanpur, 1993

Caste/ community

Number of persons

Percentage of literates among persons aged 7 and above

Percentage attending school among children aged 6–10a













































Notes. (a) All schools (including private schools and government schools in nearby villages) are taken into account in these attendance figures.

(b) There were no Kayasth children between the ages of 6 and 10 in 1993.

Source. Household survey carried out by the authors. Except for the ‘others’ group, different castes/communities have been listed in rough descending order of status in the local social hierarchy. The Jatabs are Palanpur's main ‘scheduled caste’.

These contrasts suggest that there is more to the problem of educational backwardness in Uttar Pradesh than the failure of the state to provide adequate schooling facilities. Different sections of the population in Palanpur have access to similar facilities, and yet their educational achievements show enormous variations. These contrasts may relate, to some extent, to differences of income between different groups, but this is only a small part of the story. Gender differences in literacy rates, for instance, are sharp in all income groups (and even if they were confined to low-income groups, one would still have to explain why girls are discriminated against within those groups). Similarly, caste-based differences in educational achievements are statistically significant even after controlling for differences in income levels.88 (p.84)

The female disadvantage in basic education links with some deep-rooted features of gender relations in Uttar Pradesh.89 The gender division of labour, which relegates most adult women (including those with relatively good education) to domestic work, diminishes the perceived ‘returns’ of investment in female education. The prevailing norms of village exogamy and patrilocal post-marital residence imply that these returns (and other benefits of female education) flow primarily to a daughter's future in-laws rather than to her parents. And marriage transaction patterns may act as a tangible disincentive against female education, given that an educated daughter is expected to marry a more educated man, often (p.85) implying higher dowry payments.90 The remarkably backward state of female education in Uttar Pradesh fits in a tight web of mutually-reinforcing gender inequalities and patriarchal practices.

The reasons why caste continues to be an important determinant of educational achievements and school attendance, independently of income and parental education, are less obvious. It is hard to ascertain, for instance, to what extent the schooling system discriminates against children of disadvantaged castes. Judging from our field investigation, blatant forms of caste-based discrimination (e.g. denying access to schooling facilities to certain castes, or requiring children of different castes to sit separately at school) have by and large disappeared, and nor is there much evidence in other studies of the survival of such practices.91 More subtle forms of discrimination, however, seem to remain quite widespread. Some examples include (1) discrimination against scheduled-caste settlements in the location of schools, (2) teachers refusing to touch scheduled-caste children, (3) children from particular castes being special targets of verbal abuse and physical punishment by the teachers, (4) low-caste children being frequently beaten by higher-caste classmates.92

(p.86) There are other ways in which caste is likely to remain an important determinant of educational achievements, even in the absence of discriminatory practices in the schooling system. The ability of parents to assess the personal and social value of education depends, among other things, on the information they have at their disposal. If their entire reference group is largely untouched by the experience of being educated, that information maybe quite limited. Even if some aspects of the general value of education (e.g. the economic returns to male education) are relatively obvious, they may be perceived as alien to certain groups, and relevant only to ‘others’.93 In Palanpur, for instance, it is taken for granted that Kayasth children go to school, and the benefits which Kayasth households have been able to derive from their high education levels are widely understood, but that experience is not necessarily considered by (say) Jatab parents as relevant for themselves. Jatabs and Kayasths are seen to have different roles in society, and the perceived value of education is contingent on these assumed roles.

It is tempting to conclude from this discussion that educational expansion in Uttar Pradesh depends crucially on a transformation of social attitudes and practices that have little to do with the provision of schooling facilities, or even with public policy in general. This assessment has to be qualified in at least two ways.

First, educational backwardness in Uttar Pradesh primarily reflects the combined effects of (1) the state's failure to provide adequate schooling facilities, and (2) social norms and practices that have been detrimental to the widespread utilization of available facilities. Consider, for instance, the problem of endemic female illiteracy in Palanpur. Negative attitudes to female education (themselves related, as was discussed earlier, to the patriarchal nature of gender relations in Uttar Pradesh) certainly have had much influence here. That influence, however, has been magnified by the non-functioning of the (p.87) village school over long periods of time. Many parents have dealt with the failure of the local school by sending their boys to school in neighbouring villages, but very few have done the same for girls. In fact, in 1983–4 only three girls were studying outside the village. All three of them were Kayasth; other parents simply considered it socially unacceptable to allow their daughters to wander outside the village. The failure of the state to provide adequate schooling facilities in Palanpur, therefore, is also a crucial part of the story. Similar remarks apply to the promotion of education among disadvantaged castes.

Second, it would be a mistake to consider that public policy is concerned with the provision side alone. The official goal of public policy, ever since independence, has been a rapid move towards universal education until the age of 14. If that goal cannot be achieved simply by expanding schooling facilities, the state has a responsibility to address other relevant constraints (including, for instance, the conservatism of social attitudes towards female education). Schooling decisions are, ultimately, private decisions taken at the household level. But these decisions are responsive to social norms and external incentives that can often be decisively influenced through government policy and public action.94

4.4 Education and Politics

The challenge of educational reform in Uttar Pradesh is extremely exacting. In the absence of any kind of accountability to the public, the system of government schools has been comprehensively corrupted. The failure of the schooling system, in combination with persistent inequalities of class, caste, and gender, has kept even the most elementary educational achievements out of reach of large sections of the population.

The most remarkable feature of the educational situation in Uttar (p.88) Pradesh is that this state of affairs seems to be passively accepted by the general public. At the village level, we found that most parents were extremely critical of the schooling system, but had little sense of (or faith in) the possibility of making organized demands for change. In some cases, this inertia was quite extreme. In Palanpur, for instance, the village school was virtually non-functional for as long as ten years, between 1983 and 1993, due to systematic absenteeism on the part of the local teacher (who also happened to be the son of the headman, a prosperous high-caste landowner). This blatant abuse did not trigger any organized protest, or any serious discussion in the village panchayat.95

A similar pattern of inertia applies at the state level. For one thing, there are no signs of the government of Uttar Pradesh taking any bold initiatives in the field of basic education, in spite of the alarming nature of the current situation. On the contrary, it is easy to cite many examples of continued indifference. One of the most telling symptoms in this respect is the sustained decline of real per-capita public expenditure on education in recent years — by almost 20 per cent between 1991–2 and 1993–4.96 The number of primary-school teachers per capita has also steadily gone down in recent years (see Fig. 2), further aggravating the spiralling decline of teacher—pupil ratios in the eighties.97 Similarly, the state government has taken little interest in the Total Literacy Campaign, even after the considerable potential of that campaign had been well demonstrated in several other states.98 The under-utilization of large grants earmarked for the promotion of elementary education (received from international agencies as well as the central government) is yet another symptomatic indication of the low priority given to basic education by the state government. Here again, official neglect has (p.89) provoked little challenge from opposition parties, interest groups, the media, or the general public.

Uttar Pradesh: The Burden of Inertia

FIG. 2. Number of Teachers per 10,000 Persons in Uttar Pradesh.

Source. Calculated from annual reports of the Department of Education, Ministry of Human Resource Development, New Delhi.

The privileged background of most political leaders, the undemocratic nature of village politics, and the role of the schooling system as a means of patronage, all play a part in this continued neglect of the obvious need for educational reform in Uttar Pradesh. This neglect also fits in a general pattern of lack of responsiveness of Uttar Pradesh politics to the basic needs of the citizens; we will come back to that in the next section. In the specific context of basic education, it is worth adding that the absence of any strong political demand for educational reform in Uttar Pradesh illustrates the elementary — yet widely overlooked — fact that literacy is a crucial tool of effective participation in democratic politics. Widespread illiteracy makes it that much harder for disadvantaged groups to ensure that their needs receive due attention in public debates and (p.90) political contests, and the apathy of the leadership towards these needs, in turn, is responsible for the persistence of widespread illiteracy.99

This self-sustaining circle is a central feature of Uttar Pradesh politics, and also has much relevance in interpreting Indian politics in general. On the positive side, these observations suggest that the expansion of basic education, and the political campaigns and social movements that might lead to that expansion, have a central role to play in the transformation of Indian politics. In particular, the value of basic education as a tool of political empowerment deserves much more attention from political and social leaders than it has received so far.

5. The Burden of Inertia

5.1 Aspects of Public Inertia

The crisis of rural schooling is not an isolated example of the dismal functioning of public services in Uttar Pradesh. As we saw in section 2, much of the backwardness of the state in terms of demographic transition, health indicators, educational achievements, and gender inequality can be plausibly linked to similar failures of state intervention and public action in a wide range of domains.

Although the main focus of our fieldwork was on the schooling system, we were also able to form a judgement on the functioning of other public programmes, such as primary health care, nutritional interventions, poverty alleviations schemes, and public works in the sample villages. Direct observation as well as extensive discussions with local residents strongly suggest that the standards of operation of most of these programmes in the sample villages are no higher than those of the schooling system. To illustrate, most of the sample villages had no semblance of functioning public health services (even (p.91) in cases where a health centre had been officially set up or sanctioned). Similarly, none of the sample villages showed any sign of serious activity under the ‘anganwadi’ programme (officially known as the Integrated Child Development Scheme), which is supposed to have wide rural coverage; many residents, in fact, had not even heard of the programme, even in cases where an anganwadi officially existed in their village.100

The failure of public intervention in Uttar Pradesh extends well beyond the provision of public goods and services. Most of the major developmental and redistributive programmes have gone in and out of fashion without making much of an impact. Whether these were state government programmes (such as agrarian reforms and panchayati raj) or central government schemes for which the state government acts primarily as an implementation agency (e.g. the Integrated Rural Development Programme or the Jawahar Rozgar Yojana), the pattern has consistently been one of inadequate commitment, ineffective implementation, and insignificant results.

Beyond the abolition of the zamindari system which took place all over India soon after independence, no serious effort at agrarian reforms ever made any headway in Uttar Pradesh. Even the enactment of basic land ceiling laws took the best part of a decade after the abolition of zamindari. Land ceilings, when they were finally enacted, were higher than in any other state, and numerous loopholes — both in the law and in its application — enabled the landlords to retain much of their initial holdings.101 Voluntary programmes of land distribution such as Bhoodan and Gramdan also had very limited results.102 Even land which was assigned to the village community (known as gram sabha land) as opposed to individual landlords was appropriated in many places by vested political interests.103

Even for programmes such as the public distribution system (PDS), the Integrated Rural Development Programme (IRDP), and (p.92) the Integrated Child Development Scheme (ICDS), where the role of the state government is essentially to implement transfers of various kinds to target groups (as opposed to directly redistributing assets between different classes), the gap between promise and delivery has been wide indeed. Uttar Pradesh has the lowest per-capita supply of food grains through the PDS among all major states (S. Jha, 1994). Field-based studies of IRDP, the main national anti-poverty programme, have produced much evidence of systematic abuse and corruption.104Antyodaya’ village programmes have suffered more or less the same fate (Vais, 1982). The ICDS programme in Uttar Pradesh is described by one of its own former Directors as ‘a complete write-off’ and a ‘bakwas (nonsensical) scheme’.105

This comprehensive failure of public services, and of development-oriented interventions in general, can be seen to have two mutually-reinforcing roots: (1) the state's low commitment to broad-based development and social equity, and (2) the failure of civil society to challenge that apathy, and more generally to promote social needs and the interests of disadvantaged groups. The crisis of rural schooling, discussed in the preceding section, provides a good illustration of these issues. The poor condition of schooling facilities, and the absence of an adequate number of teachers (especially female teachers), reflect the low commitment of state authorities to the goal of universalization of basic education. The chaotic functioning of schools (reflected, for instance, in a wide gap between stipulated and actual teaching hours) is also the responsibility of the state, but the failure of village communities to discipline school teachers plays a part in that outcome as well. Further, entrenched social inequalities (including oppressive caste (p.93) and gender relations) hinder collective action for improved schooling facilities, and also restrict the utilization of whatever schooling is actually supplied.

5.2 State Apathy and Public Accountability

It may be argued that state apathy in the field of social policy is an all-India phenomenon, and that there is nothing special about Uttar Pradesh in this respect. This is not quite accurate. Many other states, in fact, have a much better record not only of implementing social programmes sponsored by the central government, but also of taking important initiatives in this field. Health care in Kerala, social security in Tamil Nadu, employment guarantee in Maharashtra, panchayati raj in Karnataka, drought relief in Gujarat, primary education in Himachal Pradesh, land reform in West Bengal, are some examples.106 These initiatives have not been uniformly successful, but the point of interest here is that they indicate some responsiveness of state policy to developmental concerns and social needs. In Uttar Pradesh, however, there are no important examples of such initiatives, and the most consistent feature of state policy in these fields is one of resilient inertia.

The problem of low commitment to social needs in terms of government policy is amplified by ineffective implementation at the local level. Schools do exist but teachers are frequently absent; health care facilities are provided, but are used to promote female sterilization; poverty alleviation programmes are launched, but end up being used as instruments of patronage by the rural elite. The sanctioning of programmes and resources at the state level is meaningless if implementation bears no resemblance to stated policy objectives.

As was discussed in the previous section with reference to school education, the absence of accountability in the public sector plays a major part in this implementation failure. The breakdown of formal monitoring procedures (e.g. the school inspection system) is one aspect of the problem. In the absence of any credible threat of sanction (even in the relatively weak form of transfer rather than (p.94) dismissal), teachers and other government employees have little incentive to do their duty. While public-spirited individual initiatives can make a positive difference, the main issue is that of institutional corruption.107

The formal monitoring procedures, however, are quite similar all over India, and it is not clear why they should be particularly ineffective in Uttar Pradesh (and its neighbours) in comparison with other states. One relevant consideration here is that, aside from formal monitoring procedures, the informed vigilance and articulated demands of the public also have a role to play in ensuring the proper functioning of local public services. The fact that a village teacher shirks, for instance, is much easier to observe for the residents of that village than for a government inspector. If the concerned residents are able to organize and have means of putting pressure on the village teacher (either directly or through government institutions), their vigilance can be an effective disciplining device. Similarly, if the users of local public services are aware of their entitlements and resolved to defend them, it is that much harder for the local government doctor to appropriate the furniture of the village health centre, for the ration-shop manager to sell her supplies on the black market, for the headman to sell the village trees for his own benefit, and for anyone to steal the electricity wires.108 This brings us to the issue of local democracy and village politics.

5.3 Village Institutions and Local Governance

In pre-independence Uttar Pradesh, the institutional basis of local governance largely derived from the network of social and economic relations associated with zamindari and jajmani. The powerful zamindars dealt with higher levels of political authority, and sometimes also played a role in matters of collective interest at the (p.95) village level. The jajmani system defined patron-client relations pertaining not only to private transactions but also to some public goods and services. While the services of non-agricultural castes such as carpenters, smiths, barbers, etc., were privately consumed, some castes were responsible for services of a more public nature such as sweeping, sanitation, drainage, and street maintenance.109 Even schooling was largely organized on the basis of traditional caste obligations (in this case involving Brahmin teachers) in many villages.110 This system of customary obligations was, of course, highly unequal and extremely unjust.111

The abolition of zamindars after independence undermined the basis of their political authority and economic power. These reforms, and the development of market relations, also hastened the decline of patron-client relations. While many traditional activities became more and more market-based, a similar transition could not be expected in the case of activities with a large element of public consumption. This led to some erosion of public provisions that had been previously organized on the basis of customary obligations and patron—client relationships.

A well-documented example of this phenomenon concerns the decay of collective irrigation facilities after the abolition of zamindari. Many studies have noted how the latter development led to the collapse of traditional arrangements for the creation and maintenance of village tanks and related irrigation facilities in different parts of India.112 Similarly, there is some evidence of a deterioration (p.96) in collective arrangements for the supply of cleaning, drainage, and sanitary services based on customary obligations in the early post-independence period.113

The collapse of the pre-independence social order after zamindari abolition, and the replacement of the hierarchical system of customary obligations by more impersonal labour relations, were important steps in the emancipation of the labouring classes. These positive developments were not accompanied, however, by a corresponding adaptation of collective arrangements for the provision of collective goods and services.114 Nor was state intervention successful, in Uttar Pradesh, in providing a sound basis for such arrangements. The challenge of creating an effective system of participatory local governance remains largely unmet.

5.4 Accountability, Democracy and Factionalism

In Uttar Pradesh as elsewhere in India, local government bodies (Panchayati Raj Institutions) are now constituted on the basis of elections. To the extent that government employees are ultimately answerable to political representatives, it may be argued that the public has an opportunity to exercise some control on the quality of local public services and related matters. There are, in fact, interesting cases where an improved practice of local democracy (based on administrative decentralization and/or political organization) is considered to have led to some progress in the quality of development programmes. It has been reported, for instance, that democratic decentralization in Karnataka in the late eighties led to a major improvement in the performance of village teachers and health workers.115 Similarly, the comparatively successful implementation (p.97) of land reform and poverty alleviation programmes in West Bengal has built on systematic political activism at the local level.116 The experience of Village Development Boards in Nagaland is another example of positive achievements of local democratic institutions.117

In Uttar Pradesh, however, the decentralized Panchayati Raj Institutions have failed to provide anything like an effective basis of local democracy and accountability. Elections have not been held on a regular basis, and when they are held, they are dominated by factional rivalries at the expense of social concerns. Privileged groups (usually high-caste landlords) have exercised a tight control on local government institutions,118 and used them to their private advantage at the expense of public needs. This has been a consistent pattern from the early years of zamindari abolition when decentralization was first implemented.

Decentralization was perceived as a problematic issue from the very start. Those familiar with rural inequalities warned that devolution of political power might well result in the enhanced tyranny of dominant elite groups.119 As a matter of fact, it soon became clear that political power at the village level remained with the propertied classes. Contrary to common expectations based on an idealized view of harmonious village coexistence, the introduction of new elected bodies led to exacerbated tensions in the early years.120 Some of these tensions were due to the assertion by the newly-enfranchized poor of their rights vis-à-vis the dominant elites made up of erstwhile zamindars and their allies.121 A number of (p.98) studies, however, show that following an initial period of ferment starting around zamindari abolition, the situation settled, often to the advantage of the landlords.122 Although economic and political changes led to some relative decline of the old elite groups of high-caste landlords, these groups continued to dominate village politics, and captured the new Panchayati Raj Institutions by means foul or fair. These institutions rapidly became instruments of elite power rather than popular control.123

While political changes at the time of zamindari abolition undoubtedly gave new freedoms to the poorest sections in Uttar Pradesh, there was also much continuity in the forms and bases of political mobilization. The jajmani system of patronage and dependence was a system of factional alignment par excellence, and factions remained the basis for political mobilization even after the abolition of zamindari and the disappearance of jajmani.124 Factional strife has been an influential part of political life in much of India, but this has been particularly so in Uttar Pradesh. Not only have political parties been organized around factional lines at all levels, even within parties factional struggle has been one of the primary preoccupations.125

At the village level, factions are typically multi-caste and multiclass coalitions, often organized around alignments based on social (p.99) and economic rivalry within the village and led by the main landowning families.126 Caste, or more precisely sob-caste kinship groups, have acted as natural units of local factional mobilization. Inter-faction rivalries are often little more than economic and social competition between families of similar stature, which can be considerably enhanced by securing access to state resources.127 Thus, the appropriation of public resources and the dispensation of patronage play an important role in sustaining and nurturing factional politics. Developmental or welfare programmes, in particular, are often regarded by faction leaders as useful channels for the recruitment and reward of supporters.128 A headman, for instance, is more likely to use his influence to protect a delinquent headmaster, and thereby gain a useful ally, than to take the side of the parents, who include enemies as well as supporters.129

It is difficult to think of political systems where factions do not exist, or play no important role.130 What is remarkable about Uttar Pradesh politics, however, is that for nearly two decades after independence, factions remained the only significant basis of mobilization, and even nominal reference to ideology or policy were nearly absent from the mainstream. Human deprivation and social inequality did not figure as important political concerns at all.131 In this regard, political developments in Uttar Pradesh differed markedly (p.100) from, say, West Bengal, where class organizations have been important early on, and also from Maharashtra and the south Indian states, where ‘low-caste’ organizations have played an important role in shaping political life.132 In the late sixties, when a challenge to caste-based social inequality did finally emerge on the political agenda in Uttar Pradesh, it was led by the relatively well-off ‘middle’ castes, and factional politics continued to exercise a predominant influence.

One of the most harmful aspects of factional politics has been that rewards for political allegiance are made in the form of access to public resources such as public-sector employment (including teaching positions in village schools), government subsidies (e.g. through IRDP loans), and building contracts (including those for school buildings). Moreover, this system of patronage-based governance is not simply a localized phenomenon, it has corrupted political institutions at all levels. Leading political parties have played a critical role in the development of this perverted system of governance.133 In these circumstances, it would be naive to expect state action to promote social opportunities on a wide basis, or the electoral process to act as a sound instrument of accountability.

Factionalism in Uttar Pradesh can be regarded both as a sign of social conservatism and as an obstacle to social change. The political economy of factionalism and patronage can be seen as a manifestation of the success of the rural elites in marginalizing the concerns of the poor and the disadvantaged from the political agenda. Under these conditions, the immense potential of democratic politics as a basis for social change has been largely wasted so far.134 The factional basis of local politics has also suppressed the emergence of other types of mobilization, including class-based coalitions, anti-caste movements, and women's organizations. The failure of civil society in Uttar Pradesh to rise beyond faction-based politics has played an important role in slowing down social change and preserving the traditional balance of political power.

(p.101) 5.5 Political Participation and Social Inequality

Social and political life in Uttar Pradesh seems to be marked by two related features: pervasive inequality and resilient conservatism. Traditional patterns of caste and gender relations continue to exert a strong influence (with, for instance, women's social role remaining overwhelmingly confined to domestic work). Political mobilization still follows old patterns dominated by factionalism and patronage. And extreme inequalities of political power continue to exclude large sections of the population from effective participation in the democratic process.

To some extent, these problems apply elsewhere in India as well. In many other states, however, popular movements have had some success in fostering social change and in altering the balance of political power. In West Bengal and Kerala (discussed elsewhere in this book), political mobilization along class lines has led to far-reaching reforms concerned with land tenure and participatory democracy. Kerala further stands out for other social pursuits based on popular movements, such as resistance to caste discrimination and the universalization of basic education. Besides Kerala, Tamil Nadu, Kamataka, and Maharashtra have seen important socio-cultural movements against caste oppression, which have also taken up the issues of educational deprivation and political marginalization.135 Social and political movements of this type have been comparatively weak — though not altogether absent — in Uttar Pradesh, or for that matter in large parts of the northern region.

It is beyond the scope of this paper to ascertain the ultimate roots of social and political inertia in Uttar Pradesh. Our primary aim has been to highlight the role of public inertia (involving the state as well as civil society) as a cause of persistent human deprivation in the state. Identifying the roots of this inertia would require detailed historical enquiry. In this section, a more modest attempt is made to understand how certain features of class, caste, and gender relations in Uttar Pradesh have made it that much harder to achieve rapid social change in the post-independence period.

(p.102) Agrarian Structure and Agrarian Politics

In the post-independence period, agrarian politics in Uttar Pradesh have revolved around producer interests (such as input subsidies and procurement prices), and the leadership of agrarian movements has remained firmly in the hands of relatively prosperous landowners. After zamindari abolition, there was little political impetus for further land reform, whether in the form of land redistribution or tenancy reform. Zamindari abolition weakened the position of ‘feudal’ landlords, but strengthened the class of smaller landlords and peasants. Rapid technological change in agriculture, starting in the late sixties, further enhanced the economic position of landowning farmers.

Another relevant feature of the agrarian structure in Uttar Pradesh is that it does not include a large class of landless labourers. Although land ownership in Uttar Pradesh is highly unequal, a relatively small proportion of the population is entirely landless. Relatedly, Uttar Pradesh has the second-lowest proportion of agricultural labourers in the rural male labour force among all major states, and the second-highest proportion of rural male workers reporting cultivation as their primary occupation (Nanda, 1992). This factor, too, has contributed to the predominance of producer interests and prosperous farmers in agrarian politics. In contrast with Kerala and West Bengal, where a significant proportion of the rural population had no direct stake in the existing structure of property rights, the potential constituency for class-based mobilization geared to radical agrarian reform and related goals has been a relatively small one in Uttar Pradesh. There is strong evidence, in fact, to suggest that land reforms under zamindari abolition were consciously designed to prevent the possibility of class-based political mobilization.136 Landowners, even poor ones with marginal holdings, found themselves supporting rich farmer-led mobilization on producer interests, rather than challenging, alongside the landless, the highly unequal distribution of land.

It is also worth noting that even among the ‘farmers movements’, those based in Uttar Pradesh (mainly in the western part of the state), have been known for their hostility to demands relating to (p.103) wage labour, women's emancipation, and economic equality. In contrast, the main farmers' organization in Maharashtra (which, like its counterpart in western Uttar Pradesh, is also primarily concerned with producer issues such as input and output prices) has given serious consideration to these concerns.137 In Uttar Pradesh, the economic interests of groups other than prosperous farmers have neither been incorporated in the farmers' movements, nor had enough political weight to achieve much independent expression.

Caste and Class

Another distinguishing feature of the agrarian structure of Uttar Pradesh is the dominant position of certain ‘high’ castes with combined privileges of land ownership and ritual status.138 Specifically, dominant landowners in Uttar Pradesh frequently belong to high-ranked castes with a martial tradition, commonly identified as Kshatriya, Thakur, or Rajput.139 This is in contrast with much of south India and Maharashtra, where dominant landowners often belong to castes that rank much lower in the ritual hierarchy (on this, see Srinivas, 1962). The conjunction of temporal power and ritual authority in Uttar Pradesh has made it that much harder to challenge the prevailing inequalities of caste and class.

The emergence of strong anti-caste movements, in particular, would have been quite difficult in such circumstances. While anti-caste and ‘anti-high-caste’ movements have a long and significant history in many parts of India, such movements are of relatively recent origin in Uttar Pradesh.140 In Maharashtra and much of (p.104) south India, where the dominant landowning groups themselves often belonged to low-ranked castes in terms of ritual status, sections of the rural elite were initiators and active participants in these movements.141 By contrast, the landowning martial castes of Uttar Pradesh (particularly in the central and eastern regions), far from initiating any challenge to the caste hierarchy, have actively repressed such initiatives.

The situation has been somewhat different in parts of western Uttar Pradesh, where the agrarian economy is dominated by Jats, who are not highly ranked in the ritual caste hierarchy. Anti-Brahmin rhetoric was liberally employed by Charan Singh in the sixties to build an alliance of ‘backward’ castes under the leadership of Jat landowners. It is important to note, however, that a wide gulf (in terms of both economic and ritual status) separates landowning castes such as the Jats from dispossessed ‘untouchable’ castes such as the Chamars. This gulf came to be reflected in the hostility of the peasant movements to the interests of scheduled-caste agricultural labourers.142

Recent years have seen a growing mobilization of scheduled castes in Uttar Pradesh politics, notably under the leadership of the Bahujan Samaj Party (BSP).143 Thus far, this movement has not gone much beyond electoral coalition-building and other relatively narrow objectives such as caste-based reservation of public-sector employment. It remains to be seen whether these new coalitions are sustainable, and whether they are capable of putting the needs of disadvantaged groups on the political agenda.

(p.105) Resilient Patriarchy

Many scholars have noted that the patriarchal culture of north India is particularly strong among dominant landowning communities, such as the Rajputs and Jats. Some notable features of this culture, many of which can be plausibly related to the material circumstances of agrarian or landowning communities, include the practices of patriliny and patrilocality, a strong emphasis on the ideology of the joint family, and a pronounced gender-based division of labour.144 Among the martial castes, these patriarchal practices are compounded by an obsession with ‘honour’, the preservation of which depends partly on women adhering to conservative norms of behaviour ranging from seclusion to sati.145 The traditions and values of these martial castes have had a strong influence on other dominant landowning castes, reinforcing their patriarchal culture.

Gender relations among the labouring classes have tended to be less unequal, partly due to the influence of much higher rates of female labour-force participation.146 As we saw earlier, however, in Uttar Pradesh these classes are relatively unimportant in numerical terms, compared with other states. This is one reason why gender relations in Uttar Pradesh have been overwhelmingly influenced by the fiercely patriarchal practices of the propertied classes.

These links between caste, class, and gender relations in Uttar Pradesh have been a factor not only of extreme gender inequalities but also of great resilience in these unequal patterns. In the absence of any organized opposition to the prevailing caste hierarchy, for instance, disadvantaged castes in Uttar Pradesh have often attempted to elevate their status by emulating rather than challenging the culture (p.106) of the higher castes, including (and perhaps especially) their patriarchal practices.147 The introduction of restrictions against widow remarriage among low-ranked but upwardly-mobile castes, and the spread of dowry among communities which used to practice bride-price, are two well-documented trends that can be interpreted along these lines.148 While the phenomenon of Sanskritization is not confined to Uttar Pradesh or north India, it may have been particularly detrimental to gender equality in that region, where the castes considered as ‘role models’ (mainly the landowning martial castes, rather than the Brahmins as in much of south India) have an extremely patriarchal tradition.

Recent patterns of change in female—male ratios in Uttar Pradesh make interesting reading in this light. As discussed in the companion volume (Drèze and Sen, 1995, chapter 7), the female—male ratio in Uttar Pradesh has steadily declined since the beginning of this century, from 0.94 in 1901 to 0.88 in 1991. The last figure is quite typical of the female—male ratios that were already found among the martial and dominant landowning castes of Uttar Pradesh in 1901 (e.g. 0.89 among Rajputs, 0.85 among jats, 0.80 among Gujjars), and the overall decline of the female—male ratio in Uttar Pradesh has largely taken the form of female—male ratios among other groups declining to values around or even below 0.9 from much higher levels. These trends are quite consistent with the notion that the patriarchal culture of the dominant castes has been spreading to other groups in recent decades.

A related issue of some importance is that of women's freedom of action and participation in gainful employment and non-domestic activities. As we noted in section 2, female labour-force participation rates in Uttar Pradesh (outside the Himalayan region) are exceptionally low even by Indian standards, and so is the participation of women in non-domestic activities in general. This is one aspect of the dominant patriarchal culture, which has not diminished in the least—and may have spread—in recent decades. Restrictions on (p.107) women's freedom of movement and activities have been an especially important factor of inertia in gender relations, in so far as they have made it extremely difficult for women to act collectively.149 As women's movements elsewhere in India have amply demonstrated, organized action is essential to challenge patriarchal norms, since individual rebellion is all too easily repressed.

Inequality and Repression

The preceding discussion may help to understand the resilient nature of social inequalities in Uttar Pradesh, and of the resulting disparities of political power. It is important to emphasize that the main issue is not just economic inequality in the conventional sense of the term (e.g. inequality of income or land ownership). In fact, there is little evidence of economic inequality being particularly high in Uttar Pradesh compared with other Indian states.150 What is more relevant is the particular form which social and economic inequalities have taken, and the mutually-reinforcing nature of different types of inequality (relating in particular to class, caste, and gender).

The high concentration of power and privileges deriving from the combined effects of inequalities based on class, caste, and gender has made for an environment that is extremely hostile to social change and broad-based political participation. One particular symptom of this problem is the brutal repression of social and political movements geared to the emancipation of disadvantaged (p.108) groups. Attempts by women to claim their property rights, or by agricultural labourers to claim higher wages, or by members of the scheduled castes to resist high-caste oppression have often been met with violence, rape, and murder.151 The much-discussed ‘criminalization of politics’ in Uttar Pradesh fits in this general pattern of violence-based social control.152 One plausible reading of the recent social history of Uttar Pradesh is that the prevailing patterns of class, caste, and gender inequality have made it particularly difficult for disadvantaged groups to cope with repressive violence.

6. Concluding Remarks

In drawing lessons from Uttar Pradesh's development experience, we must avoid the trap of regarding this state either as typical of India as a whole, or as a special case of little significance for other states. Much of India is very different from Uttar Pradesh, as we have discussed with particular reference to the contrast with South India. At the same time, large parts of India are like Uttar Pradesh in some important respects. In particular, the social failures that have hindered Uttar Pradesh's development (e.g. widespread illiteracy, pervasive inequality, endemic corruption, and the suppression of women's agency in society) are not confined to that state alone. These failures are, in many cases, particularly prominent in Uttar Pradesh, where their human consequences also come into sharper focus, but this feature enhances rather than diminishes the relevance of this regional perspective on Indian development. It is on this understanding that we conclude with some general observations on the interpretation and wider significance of the case study presented in this paper.

(p.109) To put first things first, we should restate the basic fact that Uttar Pradesh is one of the regions in India that lags behind as far as many basic indicators of development are concerned. If this point is worth reiterating, it is because Uttar Pradesh has not usually been seen in that way, at least not until recently. As we saw in section 2, Uttar Pradesh does not differ very much from India as a whole in terms of conventional poverty indicators, and given the overwhelming influence of these indicators in policy debates, the state has not been the object of any special concern in development planning. Further, Uttar Pradesh is often seen as one of the ‘progressive’ Green Revolution regions that have — it is often argued — received too much rather than too little official attention and government support (e.g. in the form of agricultural credit and infrastructural investment). This assessment, too, is based on neglecting important indicators of human deprivation in the state, such as mortality and illiteracy. Interestingly, it is the widespread concern with high fertility in the northern region, including Uttar Pradesh, which eventually led to greater attention being paid to ‘social development’ in the so-called BIMARU states (Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh).

Second, as far as the causes of endemic deprivation in Uttar Pradesh are concerned, low per-capita incomes and slow economic growth are only part of the story. Indeed, poor economic performance in terms of these and other standard indicators is a problem which Uttar Pradesh shares with many other Indian states. The issue of economic performance is, of course, very serious, but there are many other failures to address in transforming living conditions in Uttar Pradesh. Among the relevant failures are aborted land reforms, the displacement of health care services by family-planning programmes, the decay of the public schooling system, the widespread corruption of poverty alleviation programmes, the suppression of women's informed agency in society, and the fragile basis of local democracy. Underlying these diverse problems is a basic failure of public action — whether of a collaborative or adversarial type — to focus on the promotion of social needs, particularly those of disadvantaged sections of the population.

Third, this case study of Uttar Pradesh makes particularly interesting reading in the light of Kerala's development achievements, discussed in Ramachandran's contribution to this volume. Just as Kerala's experience illustrates what can be achieved through (p.110) determined public action at an early stage of development, Uttar Pradesh illustrates the penalties of inertia. Even the specific fields of action which have been so central to Kerala's achievements (e.g. land reform, gender relations, basic education, local democracy) turn out to be remarkably similar to those where public inertia in Uttar Pradesh has exacted such a heavy price. This particular contrast is of major significance in understanding the divergent development performances of different Indian states and their practical implications.

Fourth, the persistence of widespread illiteracy in Uttar Pradesh (even in the younger age groups, and particularly among females) is an issue of great importance on its own, both as an aspect of human deprivation in that state and as a cause of other kinds of deprivation. A wide range of recent empirical investigations have brought out the diverse individual and social roles of basic education, including those connected with economic growth, demographic change, social equity, political participation, and personal development. In the light of these recent findings, the promotion of basic education in Uttar Pradesh and other educationally backward states is undoubtedly one of India's foremost development priorities. Unfortunately, as the field investigation reported in sections 3 and 4 brings out, the functioning of government schools in rural Uttar Pradesh is extraordinarily poor as things stand, and probably deteriorating. The central issue seems to be the absence of any kind of accountability of the schooling establishment. Reforming the schooling establishment is no easy task, and the prospects of real change seem to depend crucially on basic education in Uttar Pradesh becoming a major political issue.

Fifth, this case study highlights the crippling effects of social inequality on development achievements. In Uttar Pradesh, extreme inequalities of political power have severely distorted the priorities of state intervention and the implementation of most development programmes. The low participation of disadvantaged groups in the political process, in turn, reflects the continuing influence of sharp inequalities relating not only to class but also to caste and gender. Even the failure to achieve widespread literacy fits in this general pattern, reflected inter alia in the elitist biases of educational policy and the lack of accountability of the schooling establishment. These observations are of some general importance, given the frequent tendency to see inequality through the narrow prism of income (p.111) distribution, and to consider inequality as a secondary issue compared with the overriding objective of reducing poverty. This perspective fails to recognize (1) that social inequality is not just a matter of income distribution, (2) that social change, including the elimination of oppressive inequalities, is an integral part of development, and (3) that social inequality can be a major obstacle to the successful pursuit of a wide range of other development objectives.

Sixth, Uttar Pradesh's development experience has some bearing on current debates about economic reform in India. The current focus of economic reforms, as well as of the critiques of these reforms, is overwhelmingly on the specific question of liberalization.153 Without denying the importance of the issues that have been debated in that context, this focus is seriously inadequate. As this case study of Uttar Pradesh illustrates, the social failures that account for the persistence of endemic deprivation in India are wide-ranging and deep-rooted, and the requirements of real reform go much beyond economic liberalization.154

Finally, the transformation of development priorities and achievements in Uttar Pradesh is a challenge not only for the state but also for the public. The agency of the state is obviously central to this transformation, but the decisions of the state and the effectiveness of government initiatives are themselves contingent on the nature and content of democratic politics. In the present political climate, it would be naive to expect the government to initiate a major reorientation of development priorities on its own, or on the basis of bland expert advice. Ultimately, this is a political battle, which calls for more effective enfranchisement of disadvantaged groups as well as better articulation of social needs. The other case studies presented in this book illustrate the feasibility of substantial changes in the balance of political power in Indian states. Even the recent political history of Uttar Pradesh itself includes some hopeful signs of change. The social failures of Uttar Pradesh are quite daunting, but the potential rewards of action are correspondingly high, and the costs of continued inertia even higher.

(p.112) References

Bibliography references:

Agarwal, Bina (1994), A Field of One's Own: Gender and Land Rights in South Asia (Cambridge: Cambridge University Press).

Agnihotri, Satish (1994), ‘Missing Females: A Disaggregated Analysis’, mimeo, University of East Anglia, forthcoming in Economic and Political Weekly.

Ahmad, A. and S.C. Nuna (1990), School Education in India (New Delhi: NIEPA).

Anon (1991), ‘Note on Participation in Education—All India’, Sarvekshana, 14 (3), January–March.

——— (1992), ‘Morbidity and Utilization of Medical Services: NSS 42nd Round’, Sarvekshana, 15 (4), April–June.

Bandyopadhyay, R. (1991), ‘Education for an Enlightened Society: A Review’, Economic and Political Weekly, 16 February.

Banerjee, Sumanta (1994), ‘Obstacles to Change’, Economic and Political Weekly, 26 March.

Banerji, Rukmini (1995), ‘Urban Poverty and the Context of Primary Schooling in India’, mimeo, Spencer Foundation, Chicago.

Bara, D., R. Bhengra, and B. Minz (1991), ‘Tribal Female Literacy: Factors in Differentiation among Munda Religious Communities’, Social Action, 41 (4).

Bardhan, Pranab (1988), ‘Dominant Proprietary Classes and India's Democracy’, in Kohli (1988).

Bashir, Sajitha (1994), ‘Public Versus Private in Primary Education: Comparisons of School Effectiveness and Costs in Tamil Nadu’, Ph.D. thesis, London School of Economics.

Bashir, Sajitha, et al. (1993a), Education for All: Baseline Survey of Three Districts of Uttar Pradesh, India (New Delhi: New Concept Consultancy Services), vol. 1.

——— (1993b), Education for All: Baseline Survey of Three Districts of Uttar Pradesh, India (New Delhi: New Concept Consultancy Services), vol. 2.

Basu, Alaka Malwade (1989), ‘Is Discrimination in Food Really Necessary for Explaining Sex Differentials in Childhood Mortality?’, Population Studies, 43 (2).

——— (1992), Culture, the Status of Women and Demographic Behaviour (Oxford; Clarendon Press).

Berreman, G.D. (1963), ‘Caste and Community Development’, Human Organisation, 22 (1).

(p.113) Bhoosnurmath, Kashinath (1991), ‘People's Perception of Importance of Education; A Field Experience in Dehradun District’, The Administrator, 36.

Biswas, G. (1990), ‘Health and Nutritional Level of Children in Uttar Pradesh, with Special Reference to Allahabad’, Occasional Paper No. 56, Govind Ballabh Pant Social Science Institute, Allahabad.

Bliss, Christopher and Nicholas Stern (1982), Palanpur: The Economy of an Indian Village (Oxford; Oxford University Press).

Bose, A. (1991), Demographic Diversity of India (Delhi; B.R. Publishing).

Bose, Sugata (1993), Peasant Labour and Colonial Capital: Rural Bengal Since 1770 (Cambridge: Cambridge University Press).

Brass, Paul R. (1965), Factional Politics in an Indian State (Berkeley and Los Angeles: University of California Press).

——— (1985), Caste, Faction and Party in Indian Politics; Volume Two—Election Studies (Delhi: Chanakya Publications).

——— (1990), The Politics of India since Independence (Cambridge: Cambridge University Press).

Byres, T.J. (1988), ‘Charan Singh, 1902–87: An Assessment’ Journal of Peasant Studies, 15 (2).

Budakoti, D.K. (1988), ‘Study of the Community and Community Health Work in Two Primary Health Centres in Chamoli District of Uttar Pradesh’, M.Phil. Dissertation, Centre for Social Medicine and Community Health, Jawaharlal Nehru University, New Delhi.

Caldwell, Pat and John Caldwell (1987), ‘Where There is a Narrower Gap between Female and Male Situations: Lessons from South India and Sri Lanka’, paper presented at a Workshop on Differentials in Mortality and Health Care, BAMANEH/SSRC, Dhaka.

Chanana, Karuna (1990), ‘Structures and Ideologies: Socialisation and Education of the Girl Child in South Asia’, Indian Journal of Social Science, 3 (1).

Chanana, Karuna (ed.) (1988), Socialisation, Education and Women: Explorations in Gender Identity (New Delhi: Orient Longman).

Chandra, P. (1958), ‘Hindu Social Organisation in Village Majra’, Eastern Anthropologist, 11 (3–4).

Chen, Marty and Jean Drèze (1995), ‘Recent Research on Widows in India’, Economic and Political Weekly, 30 September.

Church, R. (1974), ‘The Impact of Bhoodan and Gramdan on Village’, Pacific Affairs, 48 (1).

CIET International (1995), ‘Nepal Multiple Indicator Surveillance: (p.114) Summary of Preliminary Results (Formal Education)’, mimeo, CIET International, New York.

Cohn, B.S. (1990), An Anthropologist among the Historians and Other Essays (Delhi: Oxford University Press).

Colclough, Christopher (1993), Educating All The Children: Strategies for Primary Schooling in the South (Oxford: Clarendon).

Crook, R.C. and J. Manor (1994), ‘Enhancing Participation and Institutional Performance: Democratic Decentralisation in South Asia and West Africa’, report to the Overseas Development Administration, January.

Danda, A.K. (1987), A Rural Community in Transition (New Delhi: Inter-India Publications).

Das Gupta, Monica (1987a), ‘Informal Security Mechanisms and Population Retention in Rural India’, Economic Development and Cultural Change, 36 (1).

——— (1987b), ‘Selective Discrimination against Female Children in Rural Punjab’, Population and Development Review, 13.

——— (1994), ‘Life Course Perspectives on Women's Autonomy and Health Outcomes’, paper presented at a meeting of the Population Association of America, Miami, May.

Dasgupta, Partha and Debraj Ray (1990), ‘Adapting to Undernourishment: The Biological Evidence and Its Implications’, in Drèze and Sen (1990), vol. 1.

Datt, Gaurav and Martin Ravallion (1995), ‘Why Have Some Indian States Done Better than Others at Raising Rural Living Standards?’, mimeo, World Bank.

Derr, Bruce W. and Susan S. Wadley (1987), ‘Changing Patterns of Employment in Karimpur: 1925–1984’, paper presented at the Wisconsin Conference on South Asia, Madison, November.

Devi, M.D. Usha (1992), ‘Research Perspective for Understanding Women's Education’, Economic and Political Weekly, 13–20 June.

Dharampal (1983), The Beautiful Tree: Indigenous Indian Education in the Eighteenth Century (New Delhi: Biblia Impex Pvt. Ltd.).

Drèze, Jean (1990a), ‘Poverty in India and the IRDP Delusion’, Economic and Political Weekly, 29 September.

——— (1990b), ‘Famine Prevention in India’, in Drèze and Sen (1990), vol. 2.

——— (1990c), ‘Widows in Rural India’, Discussion Paper No. 26, Development Economics Research Programme, STICERD, London School of Economics.

Drèze, Jean, Peter Lanjouw, and Naresh Sharma (forthcoming), (p.115) ‘Economic Development in Palanpur, 1957–94’, to be published in P. Lanjouw and N. Stern (eds.), A Sort of Growth; Palanpur 1957–94 (Oxford: Oxford University Press).

Drèze, Jean and Mrinalini Saran (1995), ‘Primary Education and Economic Development in China and India; Overview and Two Case Studies’, in K. Basu, P. Pattanaik, and K. Suzumura (eds.), Choice, Welfare and Development (Oxford: Clarendon).

Drèze, Jean and Amartya Sen (1989), Hunger and Public Action (Oxford: Clarendon Press).

——— (1995), India: Economic Development and Social Opportunity (Delhi and Oxford: Oxford University Press).

Drèze, Jean and Amartya Sen (eds.) (1990), The Political Economy of Hunger, 3 volumes (Oxford: Clarendon Press).

Drèze, Jean and P.V. Srinivasan (1995), ‘Poverty in India: Regional Estimates, 1987–8’, Working Paper No. 36, Centre for Development Economics at the Delhi School of Economics.

Dube, S.C. (1958), India's Changing Villages: Human Factor in Community Development (London: Routledge and Kegan Paul).

Dubey, S.M. (1965), ‘Role of Kinship and Ideology in Rural Factionalism: A Study of Factions in a Village of Eastern U.P. with Special Reference to the Role of Patti’, Eastern Anthropologist, 18(1).

Dumont, L. and D. Pocock (eds.) (1957), Contributions to Indian Sociology (Paris: Mouton).

Dyson, Tim and Mick Moore (1983), ‘On Kinship Structure, Female Autonomy, and Demographic Behavior in India’, Population and Development Review, 9 (1).

Elder, J.W. (1970), ‘Rajpur. Change in the Jajmani System of an Uttar Pradesh Village’, in Iswaran (1970).

EPW Research Foundation (1993), ‘Poverty Levels in India: Norms, Estimates and Trends’, Economic and Political Weekly, 21 August.

Frankel, Francine R. and M.S.A. Rao (eds.) (1989), Dominance and State Power in Modern India: Decline of a Social Order, 2 volumes (Delhi: Oxford University Press).

Fukunaga, Masaaki (1993), Society, Caste and Factional Politics (New Delhi: Manohar).

Fuller, C.J. (1989), ‘Misconceiving the Grain Heap: A Critique of the Concept of the Indian Jajmani System’, in J. Parry and M. Bloch (eds.), Money and the Morality of Exchange (Cambridge: Cambridge University Press).

Ghosh, A., U.R. Ananthamurthy, A. Béteille, S.M. Kansal, V. Mazumdar, and A. Vanaik (1994), ‘Evaluation of Literacy Campaigns in (p.116) India’, report of an independent Expert Group appointed by the Ministry of Human Resource Development (New Delhi: Ministry of Human Resource Development).

Gillespie, Stuart (1995), ‘Nutrition Trends in India: Latest Data’, mimeo, UNICEF, New Delhi.

Gokhale, A.M. (1988), ‘Panchayati Raj in Nagaland’, mimeo, Department of Rural Development, Ministry of Agriculture, New Delhi.

Gooptu, Nandini (1990), ‘Caste, Deprivation and Politics: The Untouchables in U.P. Towns in the Early Twentieth Century’, paper presented at a Workshop on Labour and Dalit Movements held at the School of Oriental and African Studies, London.

Gould, Harold (1972), ‘Educational Structures and Political Processes in Faizabad District, Uttar Pradesh’, in Rudolph and Rudolph (1972).

Government of India (1986), The Teacher and Society: Report of the National Commission on Teachers (New Delhi: Government of India Press).

——— (1989), ‘Child Mortality, Age at Marriage and Fertility in India’, Occasional Paper No. 2 of 1989, Demography Division, Office of the Registrar General, New Delhi.

——— (1991a), Family Welfare Programme in India: Yearbook 1989–90 (New Delhi: Ministry of Health and Family Welfare).

——— (1991b), ‘Provisional Population Totals: Rural–Urban Distribution’, Census of India 1991, Series 1, Paper 2 of 1991, Office of the Registrar General, New Delhi.

——— (1991c), ‘Provisional Population Totals: Workers and their Distribution’, Census of India 1991, Series 1, Paper 3 of 1991, Office of the Registrar General, New Delhi.

——— (1992a), Health Information of India: 1991 (New Delhi: Central Bureau of Health Intelligence, Ministry of Health and Family Welfare).

——— (1992b), National Policy on Education 1986: Programme of Action 1992 (New Delhi: Ministry of Human Resource Development).

——— (1993a), Sample Registration System: Fertility and Mortality Indicators 1991 (New Delhi: Office of the Registrar General).

——— (1993b), ‘Housing and Amenities: A Brief Analysis of the Housing Tables of 1991 Census’, Census of India 1991, Paper 2 of 1993, Office of the Registrar General, New Delhi.

——— (1994a), Annual Report 1993–94 (Part 1) of the Department of Education (New Delhi: Ministry of Human Resource Development).

——— (1994b), Report of the Group to Examine the Feasibility of Implementing the Recommendations of the National Advisory Committee set up to Suggest (p.117) Ways to Reduce Academic Burden on School Students (New Delhi: Department of Education).

Government of India (1995), Expenditure Budget 1995–96, vol. 2 (New Delhi: Ministry of Finance).

Government of Uttar Pradesh (1964a), ‘Factionalism and Leadership Change in Bhurbaharal: A Case Study of Panchayat Elections’, Report of the Planning and Research Action Institute, Lucknow.

——— (1964b), ‘Female Leadership in Deintikar Gram Sabha’, Report of the Planning and Research Action Institute, Lucknow.

——— (1966), ‘Decision-makers in a Gramdan Village’, Report of the Planning and Research Action Institute, Lucknow.

Guhan, S. (1981), ‘Social Security: Lessons and Possibilities from the Tamil Nadu Experience’, Bulletin, Madras Development Seminar Series, 11 (1).

——— (1990), ‘Social Security Initiatives in Tamil Nadu 1989’, Working Paper No. 96, Madras Institute of Development Studies.

Gupta, Akhil (1994), ‘Blurred Boundaries: The Discourse of Corruption, the Culture of Politics and the Imagined State’, mimeo, Department of Anthropology, Stanford University.

Gupta, D.B., A. Basil, and R. Asthana (1993), ‘Population Change, Women's Role and Status, and Development in India; A Review’, mimeo, Institute of Economic Growth, Delhi University.

Hale, S. (1978), ‘The Politics of Entrepreneurship in Indian Villages’, Development and Change, 9 (2).

Haque, T. and A.S. Sirohi (1986), Agrarian Reforms and Institutional Changes in India (New Delhi: Concept).

Hardiman, David (1994), ‘The Indian “Faction”: A Political Theory Examined’, in R. Guha (ed.), Subaltern Studies I: Writing on South Asian History and Society (Delhi: Oxford University Press).

Harlan, Lindsey and Paul B. Courttight (eds.) (1994), From the Margins of Hindu Marriage: Essays on Gender, Religion and Culture (New York; Oxford University Press).

Harris, Marvin (1993), ‘The Evolution of Human Gender Hierarchies: A Trial Formulation’, in Miller (1993b).

Harriss, Barbara (1990), ‘The Intrafamily Distribution of Hunger in South Asia’, in Drèze and Sen (1990), vol. 1.

Hasan, Zoya (1989), ‘Power and Mobilization: Patterns of Resilience and Change in Uttar Pradesh Polities’, in Frankel and Rao (1989).

——— (1995), ‘Shifting Ground: Hindutva Politics and the Farmers' Movement in Uttar Pradesh’, in Tom Brass (ed.), New Farmers' Movements in India (Ilford: Frank Cass).

(p.118) Hitchcock, John T. (1975), ‘The Idea of the Martial Rajput’, in M. Singer (ed.), Traditional India: Structure and Change (Jaipur: Rawat Publications).

Horowitz, B. and Madhu Kishwar (1982), ‘Family Life—The Unequal Deal’, Manushi, 11.

Indian Institute of Management (1985), Study of Facility Utilization and Programme Management in Family Welfare (Ahmedabad: Public Systems Group, Indian Institute of Management).

International Institute for Population Sciences (1994a), National Family Health Survey: India 1992–93 (Bombay: IIPS).

——— (1994b), National Family Health Survey 1992–93, State volumes (Bombay: IIPS).

Iswaran, K. (ed.) (1970), Change and Continuity in India's Villages (New York: Columbia University Press).

Jain, A.K. and Moni Nag (1985), ‘Female Primary Education and Fertility Reduction in India’, Working Paper No. 114, Center for Population Studies, The Population Council, New York.

Jain, L.R., K. Sundaram, and S.D. Tendulkar (1988), ‘Dimensions of Rural Poverty: An Inter-Regional Profile’, Economic and Political Weekly, Nov. (special issue); reprinted in Krishnaswamy (1990).

Janakarajan, S. (1995), ‘Village Resurvey: Some Issues and Results’, paper presented at a workshop on The Village in Asia Revisited held at the Centre for Development Studies, Thiruvananthapuram, January.

Jeffery, P., R. Jeffery, and P. Lyon (1989), Labour Pains and Labour Power: Women and Child-bearing in India (London: Zed).

Jejeebhoy, S. and S. Kulkarni (1989), ‘Demand for Children and Reproductive Motivation: Empirical Observations from Rural Maharashtra’, in S.N. Singh et. al. (eds.), Population Transition in India, Delhi: B.R. Publishing.

Jena, B. and R.N. Pati (eds.) (1989), Health and Family Welfare Services in India (New Delhi: Ashish).

Jha, Praveen Kumar (1994), ‘Changing Condition of Agricultural Labourers in Post-independent India: A Case Study from Bihar’, Ph.D. thesis, Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi.

Jha, Shikha (1994), ‘Foodgrains Price and Distribution Policies in India: Performance, Problems and Prospects’, Preprint No. 134, Indira Gandhi Institute of Development Research, Bombay


forthcoming in Asia-Pacific Development Journal

Jimenez, E., M.E. Lockheed, and V. Pacqueo (1991), ‘The Relative (p.119) Efficiency of Private and Public Schools in Developing Countries’, World Bank Research Observer, 6 (2).

Kakrambe, S.A. (1983), Karamveer Bhaurao Patil and Mass Education Movement: Impact on Maharashtra (Bombay: Somaiya Publications).

Kamat, A.R. (1985), Education and Social Change in India (Bombay: Somaiya Publications).

Kantowsky, D. (1968), ‘Local Politics in Rameshvar, a Multi-caste Village in Eastern U.P., India’, Seminar Paper No. 7, in collected seminar papers on Autonomy and Dependence in ‘Parochial’ Politics, Institute of Commonwealth Studies, University of London.

Karve, Irawati (1965), Kinship Organisation in India (Bombay: Asia Publishing House).

——— (1974), Yuganta: The End of an Epoch (Hyderabad: Orient Longman).

——— (1994), ‘The Kinship Map of India’, in Uberoi (1994).

Khan, M.E. (1988), ‘Infant Mortality in Uttar Pradesh: A Micro-level Study’, in A.K. Jain and P. Visaria (eds.), Infant Mortality in India: Differentials and Determinants (New Delhi; Sage).

Khan, M.E., R. Anker, S.K. Ghosh, Dastidar, and S. Bairathi (1989), ‘Inequalities between Men and Women in Nutrition and Family Welfare Services: An Indepth Enquiry in an Indian Village’, in J.C. Caldwell and G. Santow (eds.), Selected Readings in the Cultural, Social and Behavioral Determinants of Health, Health Transition Series, No, 1 (Canberra: Health Transition Centre, Australian National University).

Khan, M.E., S.K. Ghosh, Dastidar, and R. Singh (1986), ‘Nutrition and Health Practices among the Rural Women: A Case Study of Uttar Pradesh’ Journal of Family Welfare, 33 (2).

Khan, M.E., R.B. Gupta, C.V.S. Prasad, and S.K. Ghosh, Dastidar (1988), Performance of Health and Family Welfare Programme in India (Bombay: Himalaya Publishing House).

Khan, M.E., C.V.S. Prasad, and A. Majumdar (1980), People's Perceptions about Family Planning in India (New Delhi: Concept).

Khare, R.S. (1962), ‘Group Dynamics in a North Indian Village’, Human Organisation, 21 (3).

——— (1964), ‘A Study of Social Resistance to Sanitation Programmes in Rural India’, Eastern Anthropologist, 17 (2).

Kingdon, Geeta Gandhi (1994), ‘An Economic Evaluation of School Management-Types in Urban India: A Case Study of Uttar Pradesh’, Ph.D. Thesis, St. Antony's College, Oxford.

(p.120) Kishor, Sunita (1993), ‘“May God Give Sons to All”: Gender and Child Mortality in India’, American Sociological Review, 58.

Kishwar, Madhu (1995), ‘When Daughters are Unwanted’, Manushi, 86 (Jan.–Feb.).

Kohli, Atul (1987), The State and Poverty in India: The Politics of Reform (Cambridge: Cambridge University Press).

Kohli, Atul (ed.) (1988), India's Democracy: An Analysis of Changing State–Society Relations (Princeton, NJ: Princeton University Press).

Kolenda, Pauline (1983), ‘Widowhood among “Untouchable” Chuhras’, in A. Ostor, L. Fruzzetti, and S. Barnett (eds.), Concepts of Person: Kinship, Caste and Marriage in India (Delhi: Oxford University Press).

Krishnamurthy, M.G. and M.V. Nadkarni (1983), ‘Integrated Child Development Services: An Assessment’, report prepared for UNICEF, New Delhi.

Kumar, Ashok (1990), ‘The ICDS Projects in North Indian States: A Study of Selected Blocks’, National Institute of Public Cooperation and Child Development, Lucknow.

Kumar, Krishna (ed.) (1994), Democracy and Education in India (London: Sangam Books).

Kumar, Radha (1993), The History of Doing: An Illustrated Account of Movements for Women's Rights and Feminism in India, 1800–1990 (London: Vision).

Labenne, Sophie (1995), ‘Analyse Econométrique du Travail des Enfants en Inde’, M.Sc. thesis (unpublished), Department of Economics, Université de Namur, Belgium.

Lambert, Claire M. (ed.) (1976), Village Studies: Data Analysis and Bibliography—Volume 1: India 1950–1975 (Essex: Bowker Publishing Company, for the Institute of Development Studies at the University of Sussex, Brighton).

Lata, Divya (1995), ‘An Assessment of Early Child Care and Education Programmes Supported by SCF in India’, mimeo, Save the Children Fund (UK), New Delhi.

Lieten, G.K. (1994), ‘The North Indian Kulak Farmer and His Deficient Democracy’, Occasional Paper 1994, 4, Indo-Dutch Programme on Alternatives in Development, Indian Council of Social Science Research, New Delhi.

Macdorman, M. (1986), ‘Contemporary Marriage Practices in North India: Evidence from Three Uttar Pradesh Villages’, Ph.D. thesis, Australian National University, Canberra.

Mahapatra, R.K. (1994), ‘Food Crises in Kalahandi: 1960/61–1992/93’, (p.121) M.Phil. Dissertation, Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi.

Maharatna, Arup (1995), ‘Children's Work, Activities, Surplus Labour and Fertility: A Case Study of Peasant Households in Birbhum, West Bengal’, paper presented at the second workshop on Applied Development Economics, 6–10 January 1996, Centre for Development Economics at the Delhi School of Economics.

Mahendra Dev, S. (1993a), ‘India's (Maharashtra) Employment Guarantee Scheme: Lessons from Long Experience’, paper presented at a workshop on Employment for Poverty Alleviation and Food Security, convened by the International Food Policy Research Institute, Washington, DC, October 1993.

——— (1993b), ‘Social Security in the Unorganized Sector: Lessons from the Experience of Kerala and Tamil Nadu’, mimeo, Indira Gandhi Institute of Development Research, Bombay.

Majid, A. (1986), ‘Women's Contribution to Household Income Among Agricultural Labourers of Punjab’, Research Study No. 86/2, Agricultural Economics Research Centre, University of Delhi, Delhi.

Mandelbaum, D.G. (1970), Society in India (Berkeley; University of California Press).

Mari Bhat, P.N., K. Navaneetham, and S.I. Rajan (1992), ‘Maternal Mortality in India’, paper presented at a workshop on Health and Development in India held at the India International Centre, 2–4 January 1992

; also published in

M. Das Gupta, T.N. Krishnan, and L. Chen (eds.) (1995), Women's Health in India: Risk and Vulnerability (Bombay: Oxford University Press).

Marriott, McKim (1952), ‘Social Change in an Indian Village’, Economic Development and Cultural Change, 1.

Matthai, John (1915), Village Government in British India (London: T. Fisher Unwin).

Maurya, K.N. (1989), ‘An Analysis of Causative Factors Responsible for Low Utilisation of Health and Family Welfare Services’, in Jena and Pati (1989).

Mehrotra, Nidhi (1995), ‘Primary Education in Rural India: A Field Report’, mimeo, Department of Education, University of Chicago.

——— (forthcoming), ‘Primary Schooling in Rural India; Determinants of Demand’, Ph.D. thesis, University of Chicago.

Middleton, John, et al. (1993), ‘Uttar Pradesh Basic Education Project: Staff Appraisal Report’, Report No. 11746–IN, Population and Human Resources Operations Division, World Bank, Washington, DC.

Miller, Barbara (1981), The Endangered Sex (Ithaca: Cornell University Press).

(p.122) Miller, Barbara (1993a), ‘On Poverty, Child Survival and Gender: Models and Misperceptions’, Third World Planning Review, 15.

Miller, Barbara (ed.) (1993b), Sex and Gender Hierarchies (Cambridge: Cambridge University Press).

Miller, D.B. (1976), From Hierarchy to Stratification: Changing Patterns of Social Inequality in a North Indian Village (London: Oxford University Press).

Minhas, B. (1992), ‘Educational Deprivation and its Role as a Spoiler of Access to Better Life in India’, in A. Dutta and M.M. Agrawal (eds.), The Quality of Life (Delhi: B.R. Publishing).

Minhas, B.S., L.R. Jain, and S.D. Tendulkar (1991), ‘Declining Incidence of Poverty in India in the 1980s’, Economic and Political Weekly, 6–13 July.

Minturn, Leigh (1993), Sita's Daughters: Coming Out of Purdah, The Rajput Women of Khalapur Revisited (New York: Oxford University Press).

Moller, Joanne (1993), ‘Inside and Outside: Conceptual Continuities from Household to Region in Kumaon, North India’, unpublished Ph.D. thesis, London School of Economics.

Mull, Dorothy (1991), ‘Traditional Perceptions of Marasmus in Pakistan’, Social Science and Medicine, 32.

Murthy, Nirmala (1992), ‘Issues in Health Policies and Management in India’, paper presented at a Workshop on Health and Development in India, 24 January 1992

; also published in

M. Das Gupta, T.N. Krishnan, and L. Chen (eds.) (1995), Women's Health in India: Risk and Vulnerability (Bombay: Oxford University Press).

Nanda, Amulya Ratna (1991), ‘Provisional Population Totals: Workers and Their Distribution’, Census of India 1991, Series 1, Paper 3 of 1991, Office of the Registrar General, New Delhi.

——— (1992), ‘Final Population Totals: Brief Analysis of Primary Census Abstract’, Census of India 1991, Series 1, Paper 2 of 1992, Office of the Registrar General, New Delhi.

——— (1993), ‘Union Primary Census Abstract for Scheduled Castes and Scheduled Tribes’, Census of India 1991, Series 1, Paper 1 of 1993, Office of the Registrar General, New Delhi.

Narain, Iqbal (1972), ‘Rural Local Politics and Primary School Management’, in Rudolph and Rudolph (1972).

Narayana, D. (1995), ‘Education by Women as a Leading Factor in Fertility Reduction in India’, paper presented at a Workshop on Fertility Change in India and Brazil, Center for Population and Development Studies, Harvard University, April.

(p.123) National Council of Educational Research and Training (1992), Fifth All-India Educational Survey, 2 volumes (New Delhi: NCERT).

National Institute of Public Cooperation and Child Development (1992), Statistics on Children in India: Pocket Book 1992 (New Delhi: NIPCCD).

Nuna, Sheel C. (1990), Women and Development (New Delhi: National Institute of Educational Planning and Administration).

OʼHanlon, Rosalind (1985), Caste, Conflict and Ideology: Mahatma Jotirao Phule and Low Caste Protest in Nineteenth-Century Western India (Cambridge: Cambridge University Press).

Oldenburg, Philip (1992), ‘Sex Ratio, Son Preference and Violence in India: A Research Note’, Economic and Political Weekly, 5–12 December.

Olper, M.E. and R.D. Singh (1952), ‘Economic, Political and Social Change in a Village of North Central India’, Human Organisation, 11 (2).

Omvedt, Gail (1993), Reinventing Revolution: New Social Movements and the Socialist Tradition in India (London: M.E. Sharpe).

Osmani, Siddiq R. (1990), ‘Nutrition and the Economics of Food: Implications of Some Recent Controversies’, in Drèze and Sen (1990), vol. 1.

Panigrahi, Lalita (1972), British Social Policy and Female Infanticide in India (New Delhi: Munshiram Manoharlal).

Parikh, Kirit (1994), ‘Who Gets How Much from PDS: How Effectively Does It Reach the Poor?’, mimeo, Indira Gandhi Institute of Development Research, Bombay.

Park, R.L. and I. Tinker (eds.) (1963), Leadership and Political Institutions in India (Princeton: Princeton University Press).

Prabhn, K. Seeta (1995), ‘Structural Adjustment and Financing of Elementary Education: The Indian Experience’, Journal of Educational Planning and Administration, 9.

Prasad, B.G., et al. (1969), ‘A Study of Beliefs and Customs in a Lucknow Village in Relation to Certain Diseases, Menstruation, Childbirth and Family Planning’, Indian Journal of Social Work, 30 (1).

Prasad, K.V Eswara (1987), Wastage, Stagnation and Inequality of Opportunity in Rural Primary Education: A Case Study of Andhra Pradesh (New Delhi: Ministry of Human Resource Development).

Raju, S. (1988), ‘Female Literacy in India; The Urban Dimension’, Economic and Political Weekly, 29 October.

(p.124) Ramasundaram, S. (1995), ‘Implementation of Family Welfare Programme: An Overview’, CSD Papers, Series 1, Centre for Sustainable Development, Lal Bahadur Shastri National Academy of Administration, Mussoorie.

Ranjan, Sharad (1994), ‘Rural Non-farm Employment in Uttar Pradesh, 1971–1991: A Regional Analysis’, M.Phil. thesis, Centre for Development Studies, Thiruvananthapuram.

Rao, Nitya (1993), ‘Total Literacy Campaigns: A Field Report’, Economic and Political Weekly, 8 May.

Rastogi, P.N. (1964), ‘Factionalism, Politics and Crime in a U.P. Village’, Eastern Anthropologist, 17 (3).

——— (1965), ‘Factions in Kurmipur’, Man in India, 45 (4).

——— (1966), ‘Polarisation at Thakurpur: The Process and the Pattern’, Sociological Bulletin, 15 (1).

Retzlaff, R.H. (1962), Village Government in India; A Case Study (New York: Asia).

Rosenzweig, Mark R. and T. Paul Schultz (1982), ‘Market Opportunities, Genetic Endowments, and Intrafamily Resource Distribution: Child Survival in Rural India’, American Economic Review, 72.

Rudolph, S.H. and L.I. Rudolph (eds.) (1972), Education and Politics in India: Studies in Organization, Society, and Policy (Cambridge, MA: Harvard University Press).

Sahay, B.N. (1969), ‘Leadership, Conflict and Development Programme in a North India Village’, in L.P. Vidyarthi (ed.), Conflict, Tension and Cultural Trend in India (Calcutta: Punthi Pustak).

Saith, Ashwani and Ajay Tankha (1992), ‘Longitudinal Analysis of Structural Change in a North Indian Village: 1970–1987’, Working Paper No. 128, Institute of Social Studies, The Hague.

Saldhana, Denzil (1994), ‘The “Socialisation” of Critical Thought: Responses to Illiteracy Among the Adivasis in Thane District’, in Kumar (1994).

Satia, J.K. and S.J. Jejeebhoy (eds.) (1991), The Demographic Challenge: A Study of Four Large Indian States (Delhi: Oxford University Press).

Sen, Amartya (1970), ‘Aspects of Indian Education’, Lal Bahadur Shastri Memorial Lecture at the Institute of Public Enterprise, Hyderabad

; reprinted

in S.C. Malik (ed.), Management and Organization of Indian Universities (Simla: Institute of Advanced Study)

, and partly reprinted in

P. Chaudhuri (ed.) (1971), Aspects of Indian Economic Development (London: Allen and Unwin).

——— (1985), Commodities and Capabilities (Amsterdam: North-Holland).

(p.125) Sen, Amartya (1992), Inequality Reexamined (Oxford: Clarendon Press and Cambridge MA: Harvard University Press).

Sen, Mala (1993), India's Bandit Queen (New Delhi: Indus).

Sengupta, S. (1991), ‘Progress of Literacy in India during 1983 to 1988’, Sarvekshana, April–June.

Shah, M. H. (1989), ‘Factors Responsible for Low Performance of Family Welfare Programme’, in Jena and Pati (1989).

Shami, N. (1992), ‘Socio-Economic Survey of Village Patna Khurd’, training assignment, Lal Bahadur Shastri National Academy of Administration, Mussoorie.

Shankar, K. (1991a), ‘Politics of Land Distribution in Uttar Pradesh’, Economic and Political Weekly, 27 April.

——— (1991b), ‘Integrated Rural Development Programme in Eastern UP’, Economic and Political Weekly, 12 October.

Shariff, Abusaleh (1996), ‘Elementary Education in India: Differentials and Determinants’, paper presented at the second workshop on Applied Development Economics, Centre for Development Economics at the Delhi School of Economics, 6–10 January.

Sharma, H.R. (1994), ‘Distribution of Landholdings in Rural India, 1953–54 to 1981–82: Implications for Land Reform’, Economic and Political Weekly, 24 September.

Sharma, Miriam (1978), The Politics of Inequality: Competition and Control in an Indian Village (New Delhi: Hindustan Publishing Corporation).

Sharma, Rita and Thomas T. Poleman (1993), The New Economics of India's Green Revolution (Ithaca: Cornell University Press).

Sharma, S.S. (1981), ‘Harijan Influentials in a North Indian Village’, Journal of Social and Economic Studies, 9 (2).

Sharma, Ursula (1980), Women, Work and Property in North-West India (London: Tavistock).

Simmons, G.B., C. Smucker, S. Bernstein, and E. Jensen (1982), ‘Post-Neonatal Mortality in Rural India: Implications of an Economic Model’, Demography, 19 (3).

Simon, Brian (1965), Education and the Labour Movement: 1870–1920 (London: Lawrence and Wishart).

Singh, Avtar (1977), ‘Community Structure and Technological Development’, Contributions to Asian Studies, 10 (43).

Singh, Baljit and Shridhar Misra (1964), A Study of Land Reforms m Uttar Pradesh (Honolulu: East—West Center).

Singh, H.A. (1969), ‘Strains in Leadership Structure: From Status (p.126) Groups to Pluralism in an East U.P. Village’, Economic and Political Weekly, 3 May.

Singh, H.K. (1958), ‘Class of People Benefited by Agricultural Programmes of N.E.S.’, Journal of Social Sciences, 1 (2).

Singh, H.N. (1993), ‘Social Background and Role Performance of Village Pradhans’, Occasional Paper No. 10, Institute of Social Sciences, New Delhi.

Singh, Jagpal (1992), Capitalism and Dependence: Agrarian Politics in Western Uttar Pradesh, 1951–1991 (New Delhi; Sage).

Singh, J.P. and R.P. Singh (1989), ‘A Study of Employment and Income in Agriculture under the Jurisdiction of IRDP in Sultanpur District’, Narendra Deva Journal of Agricultural Research, 4 (2).

Singh, K.K. (1967), Patterns of Caste Tension (Delhi: Asia).

Singh, N.K. and F. Ahmed (1995), ‘Crime and Politics: The Nexus’, India Today, 31 August.

Singh, Yogendra (1970), ‘Chanukera: Cultural Change in Eastern Uttar Pradesh’, in Iswaran (1970).

Sinha, Amarjeet (ed.) (1995), Village Visit Reports, compendium of selected field reports of the 60th Foundational Course, Lal Bahadur Shastri National Academy of Administration, Mussoorie.

Sinha, Shantha (1995), ‘Child Labour and Education Policy in India’, CSD Papers, Series 1, Centre for Sustainable Development, Lal Bahadur Shastri National Academy of Administration, Mussoorie.

Sopher, David (1980a), ‘The Geographical Patterning of Culture in India’, in Sopher (1980b).

Sopher, David (ed.) (1980b), An Exploration of India: Geographical Perspectives on Society and Culture (Ithaca, NY: Cornell University Press).

Srinivas, M.N. (1960), India's Villages, second edition (Bombay: Asia Publishing House).

——— (1962), Caste in Modern India and Other Essays (Bombay: Asia Publishing House).

——— (1966), Social Change in Modern India (Berkeley and Los Angeles: University of California Press).

——— (1987), The Dominant Caste and Other Essays (Delhi: Oxford University Press).

——— (1989), The Cohesive Role of Sanskritization and Other Essays (Delhi: Oxford University Press).

Srivastava, P.K. (1991), State Government and Politics in India (New Delhi: Mittal Publications).

Srivastava, Ravi S. (1995), ‘Beneath the Churning: Change and Resilience in Producer Strategies in Uttar Pradesh Agriculture’, (p.127) paper presented at a workshop on The Village in Asia Revisited held at the Centre for Development Studies, Thiruvananthapuram, January.

Srivastava, Ravi S. (forthcoming), ‘Agrarian Change and the Labour Process’, in P. Robb and U. Patnaik (eds.), Meanings of Agriculture (Delhi: Oxford University Press).

Srivastava, Saraswati (1976), ‘Uttar Pradesh: Politics of Neglected Development’, in I. Narain (ed.), State Politics in India (Meerut: Meenakshi Prakashan).

Stokes, E. (1975), ‘The Structure of Landholding in Uttar Pradesh 1860–1948’, Indian Economic and Social History Review, 12.

Subramanian, Shankar (1994), ‘Gender Discrimination in Intra-Household Allocation in India’, paper presented at the Second Applied Development Economics Workshop, 6–10 January 1996, Centre for Development Economics at the Delhi School of Economics, Delhi.

Tendulkar, S.D., K. Sundaram, and L.R. Jain (1993), ‘Poverty in India, 1970–71 to 1988–89’, Working Paper, ILO-ARTEP, New Delhi.

Tyagi, P.N. (1993), Education for All: A Graphic Presentation, second edition (New Delhi: National Institute of Educational Planning and Administration).

Uberoi, Patricia (ed.) (1994), Family, Kinship and Marriage in India (Delhi: Oxford University Press).

Vais, R.R. (1982), ‘An Evaluation of Antyodaya Programme in Uttar Pradesh’, Research Study No. 81/6, Institute of Development Studies, University of Delhi.

Varma, Jyotsna (1992), ‘Lanka Kachuara: A Village Divide’, training assignment, Lal Bahadur Shastri National Academy of Administration, Mussoorie.

Visaria, Leela and Pravin Visaria (1995), ‘Acceleration of Fertility Decline in Tamil Nadu since 1981: Some Hypotheses’, Working Paper, Gujarat Institute of Development Research, Ahmedabad.

Visaria, P., A. Gumber, and L. Visaria (1993), ‘Literacy and Primary Education in India, 1980–81 to 1991’, Journal of Educational Planning and Administration, 7 (1).

Visaria, Pravin and Leela Visaria (1994), ‘Demographic Transition: Accelerating Fertility Decline in 1980s’, Economic and Political Weekly, 17–24 December.

Vlassoff, C. (1980), ‘Unmarried Adolescent Females in Rural India: A Study of the Social Impact of Education’, Journal of Marriage and the Family, 42 (2).

(p.128) Vlassoff, C. (1993), ‘Against the Odds: The Changing Impact of Education on Female Autonomy and Fertility in an Indian Village’, paper presented at a Workshop on Female Education, Autonomy and Fertility Change in South Asia, New Delhi, 8–10 April.

Vyasulu, Vinod (1993), ‘Management of Poverty Alleviation Programmes in Karanataka: An Overview’, paper presented at a workshop on Poverty Alleviation in India held at the Institute of Development Studies, Jaipur, February.

Wadley, S.S. and B.W. Derr (1989), ‘Karimpur 1925–1984: Understanding Rural India Through Restudies’, in P. Bardhan (ed.), Conversations Between Anthropologists and Economists (Delhi: Oxford University Press).

Weiner, Myron (1991), The Child and the State in India: Child Labor and Education Policy in Comparative Perspective (Princeton: Princeton University Press).

Wiser, W.H. (1936), The Hindu Jajmani System (Lucknow; Lucknow Publishing House).

Wiser, W. and C. Wiser (1970), Behind Mud Walls, 1930–60 (Berkeley: University of California Press).

Zamora, M.D. (1990), The Panchayat Tradition: A North Indian Village Council in Transition 1947–1962 (New Delhi: Reliance Publishing House).


(*) We are grateful to Sudhir Anand, Roli Asthana, Rukmini Banerji, Sajitha Bashir, Bela Bhatia, Gaurav Datt, Keshav Desiraju, Stuart Gillespie, B.K. Joshi, Shiva Kumar, Nidhi Mehrotra, Aditi Mehta, Aya Okada, Vimala Ramachandran, V.K. Ramachandran, Jesse Ribot, Amartya Sen, Sunil Sengupta, Amarjeet Sinha, Rohim Somanathan, and R.V. Vaidyanatha Ayyar for helpful discussions, and to Geeta Gandhi Kingdon for detailed comments on a draft of this paper. We are also indebted to Shanti Rabindran and Usman Saadat for excellent research assistance. Finally, we would like to thank C.P. Sharma (Agricultural Economics Research Centre, University of Delhi), Chote Lal, and members of the Samaj Seva Sanstha in Manikpur, for their invaluable help with the field work.

(1) For some relevant indicators, see the Statistical Appendix in the companion volume Drèze and Sen (1995); on the issue of poverty decline, see Datt and Ravallion Ravallion (1995); on nutritional indicators, see the recent anthropometric evidence derived from the National Family Health Survey (e.g. Gillespie, 1995).

(2) See, for instance, Sopher (1980a), and other contributions on the cultural and social geography of India in Sopher (1980b); see also Karve (1965), Rudolph and Rudolph (1972), Dyson and Moore (1983), Caldwell and Caldwell (1987), Satia and Jejeebhoy (1991), among others.

(3) The four states mentioned earlier, taken together, had a population of 335 million in 1991, (about 40 per cent of India's total population).

(4) We have made particularly intensive use of the numerous village studies that have been carried out in different parts of Uttar Pradesh. The early work in this regard includes Olper and Singh (1952), Marriott (1952), Dube (1958), Khare (1962, 1964), Berreman (1963), Park and Tinker (1963), Rastogi (1964, 1965, 1966), Elder (1970), Yogendra Singh (1970), and Wiser and Wiser (1970). A more comprehensive list of references for the period up to the early seventies can be found in Lambert (1976). More recent field-based studies that we have drawn upon include, among others, Miller (1976), Hale (1978), Miriam Sharma (1978), S.S. Sharma (1981), Simmons et al. (1982), Macdorman (1986), Danda (1987), Khan (1988), Wadley and Derr (1989), Zamora (1990), Colin (1990), Misra (1991). Bhoosnurmath (1991), Saith and Tankha (1992), Jagpal Singh (1992), Minturn (1993), Moller (1993), Vlassoff (1993), Gupta (1994), Kingdon (1994), Ravi Srivastava (1995), and Mehrotra (forthcoming).

(5) Prior to zamindari abolition, legal ownership of land in Uttar Pradesh was vested with between 3 to 8 per cent of rural households (see Hasan, 1989, and Stokes, 1975, for various estimates). After zamindari abolition the proportion of rural households owning some land ranged from 70 to 90 per cent depending on the precise ownership criterion (see H.R. Sharma, 1994, for results of the 1953–4 round of the National Sample Survey).

(6) Ranked by their ownership of land, the bottom 40 per cent of all rural households in Uttar Pradesh owned 2.5 per cent of the total area in 1953–4, while the top 10 per cent owned 46 per cent of the area. More or less the same size distribution was observed in 1982 (H.R. Sharma, 1994).

(7) See Sharma and Poleman (1993), Ranjan (1994), and the literature cited there. The impact of agricultural change in Uttar Pradesh has also been analysed in a number of village studies, including Bliss and Stern (1982). Saith and Tankha (1992), Srivastava (1995), Drèze, Lanjouw, and Sharma (forthcoming).

(8) On this, see also Amartya Sen's contribution m this volume.

(9) Human Development Report 1994, pp. 208–9.

(10) Until the mid-eighties, Uttar Pradesh had the highest levels of infant and child mortality among major Indian states, by a long margin, and also the lowest level of life expectancy. There has been some progress in this respect in recent years, leading Uttar Pradesh to ‘overtake’ Madhya Pradesh and Orissa, where the decline of mortality has been relatively slow during the eighties.

(11) Human Development Report 1991, p. 143 Maternal mortality rates in industrialized countries such as Norway or Belgium are below 5 per 100,000 live births; this is an aspect of human well-being for which inequalities between the poorest and richest countries of the world are extraordinarily large.

(12) Gillespie (1995), based on weight-for-age data for 1992–3.

(13) On this, see the contribution by Murthi, Guio, and Drèze in this volume, and the literature cited there.

(14) We ignore the state of Arunachal Pradesh in north-east India, which has a tiny population of less than one million persons, mainly from the ‘scheduled tribes’. Arunachal Pradesh has a lower level of male literacy than the four states in question, but a higher level of female literacy (Bose, 1991, p. 62).

(15) Nuna (1990), pp. 113–14.

(16) Calculated from International Institute for Population Sciences (1994b), Table 3.6.

(17) Not infrequently, the sojourn in the educational system is so short that the child in question does not even become literate. In the village of Palanpur (Moradabad district), a survey carried out in 1983–4 showed that, in a village with 960 residents, as many as 56 adults were illiterate despite having been to school (Drèze amd Saran, 1995).

(18) That pattern applies in Kerala, where, in 1987–8, the incidence of illiteracy was about 23 per cent in the population as a whole, but under 3 per cent in the 10–14 age group (Sengupta, 1991, based on National Sample Survey data). It also applies to many Chinese provinces (Drèze and Saran, 1995). In the educationally advanced states of India outside Kerala, such as Himachal Pradesh, Tamil Nadu, and Maharashtra, illiteracy is now quite low among male adolescents (e.g. below 15 per cent in 1987–8, for the 10–14 age group), but remains high for adolescent females; see Sengupta (1991).

(19) For further discussion of gender bias in mortality as a cause of low female—male ratios in India, see chapter 7 in the companion volume (Drèze and Sen, 1995), and the literature cited there.

(20) There is a large literature on this subject. See Miller (1981), Das Gupta (1987b, 1994), Basu (1992), Khan (1988), Khan et al. (1986, 1989), Jeffery et al (1989), Harriss (1990), Minturn (1993), among others. For clear evidence of strong ‘boy preference’ in Uttar Pradesh (in contrast with much more gender-neutral patterns in South India), see the National Family Health Survey data on desired family composition; e.g. International Institute for Population Sciences (1994a, 1994b).

(21) On this issue, see Drèze and Sen (1995), chapter 7, and also the chapter by Murthi, Guio, and Drèze in this volume.

(22) In this section, we use the term ‘poverty’ in the sense that has become conventional in the literature on poverty in India, i.e. low per-capita expenditure. This conventional interpretation has to be distinguished from the notion of poverty as a failure of basic capabilities (on which see Sen, 1985, 1992).

(23) Tendulkar et al. (1993), Table A.5, based on the 43rd round of the National Sample Survey. The poverty line is meant to indicate the level of per-capita expenditure at which minimal calorie requirements are met, given the observed consumption patterns. Too much stress should not be placed on this interpretation, however, given the problematic nature of the notion of ‘calorie requirements’ (see Dasgupta and Ray, 1990, and Osmani, 1990), and related methodological problems (see EPW Research Foundation, 1993, for an overview).

(24) On this, see also the review of poverty estimates in EPW Research Foundation (1993).

(25) See, for instance, Tendulkar et al. (1993). The incidence of poverty in Kerala seems to have declined quite rapidly in recent years (see Datt and Ravallion, 1995). However, for most of the period from independence until the eighties, the incidence of poverty in Kerala appears to have been, if anything, higher than in Uttar Pradesh (see EPW Research Foundation, 1993).

(26) See Drèze and Sen (1995), Statistical Appendix; the figures apply to 1991.

(27) On this, see also Table 8 below.

(28) It should be mentioned that some of the indicators presented in Table 6 reflect not only the provision and utilization of public services, but also some contribution Of private services in the relevant fields. The operational basis of private provisioning, however, tends to be quite similar in different states, so that the observed contrasts are primarily a reflection of differences in the scope and functioning of public services. In some cases, as with schooling, another important factor is the nature of government support for private institutions geared to the provision of essential services (for instance, ‘private aided’ schools have played a major role in Kerala's outstanding experience of educational expansion).

(29) Calculated from Reserve Bank of India Bulletin, March 1993, Appendix II; the figures relate to 1990–1.

(30) See, for instance, the studies of health care in Uttar Pradesh and other north Indian states carried out by the Operations Research Group (Khan et al., 1980, 1986, 1988, 1989) and the Public Systems Group (Indian Institute of Management, 1985, Shah, 1989, Murthy 1992); also Budakoti (1988), Jeffery et al. (1989), Maurya (1989), among others.

(31) Note that there is no evidence of the situation having improved since the studies cited here have been completed. Family-planning targets have apparently been renamed ‘expected levels of achievement’ (Visaria and Visaria, 1994), but the pressure to meet these targets has not relented in the least, judging from wide-ranging discussions with health and family-planning experts (see also Ramasundaram, 1995, for a startling testimony). One of them, a senior bureaucrat, after admitting that the testimonies cited here are still entirely relevant today, and that the government's family planning programme in north India still has a tendency to antagonize a large part of the population, cheerfully added: ‘Opposition is the beauty of democracy.’

(32) Even economic growth and poverty decline in India seem to be positively related to female labour-force participation; on this, see Drèze and Srinivasan (1995).

(33) There is extensive ethnographic evidence on the particularly restrictive nature of gender relations in Uttar Pradesh. See, for instance, Karve (1965), Mandelbaum (1970), Sopher (1980b), Macdorman (1986), Jeffery et al. (1989), Minturn (1993), and Agarwal (1994).

(34) International Institute for Population Sciences (1994a, 1994b), Tables 3.12 and 9.14.

(35) On this, see Rosenzweig and Schultz (1982), Kishor (1993), and the chapter by Murthi, Guio, and Drèze in this volume.

(36) The instructions to census investigators state that ‘household duties’ are not to be counted as economically productive.

(37) The figures in this paragraph are calculated from Nanda (1991), Table 6, based on the 1991 Census ‘provisional population totals’.

(38) Similarly, the failure of the western region of Uttar Pradesh to achieve any kind of lead in health and educational matters compared with the eastern region, in spite of considerably higher incomes, may well relate to the particularly restricted social roles of women in the western region.

(39) As was discussed earlier, the first two also have to be interpreted as aspects of human deprivation in Uttar Pradesh. As discussed by Amartya Sen in his contribution to this book (see also Drèze and Sen, 1995), human capabilities — such as literacy or participation in society — should be seen as having both intrinsic importance, as constituents of well-being, and instrumental importance, as means of enhancing other aspects of well-being. The failure of public services, on the other hand, is mainly of instrumental importance in this context.

(40) In a study based on data from rural Uttar Pradesh, Simmons et al. (1982) find tetanus to be one of the most common causes of neonatal death. They also find that diarrhoea and low birth-weights (and not the amount of food intake) are the main causes of malnourishment among children. A sample survey carried out in 1981–2 finds that neonatal mortality due to tetanus in rural Uttar Pradesh is higher than in any other state, and five times as high as the all-India average (Biswas, 1990, p. 48).

(41) On the relatively weak link between per-capita income and child survival in India, see the chapter by Murthi, Guio, and Drèze in this volume.

(42) In many parts of rural Uttar Pradesh, it is still common for umbilical cords to be cut with unsterilized sickles, for children to be left entirely unimmunized, for cooked food to be left uncovered for long hours before consumption, and for extraordinary beliefs to be entertained about the causes of simple diseases such as tetanus or diarrhoea (personal observations). On the highly uninformed nature of many beliefs and practices relating to health care in north India and Pakistan, see also Prasad et al. (1969), Mull (1991), Minturn (1993), Khan (1988), Khan et al. (1986, 1989), among others.

(43) In the case of girls, there is the additional issue of a low value being placed on female survival. The disadvantaged position of women in Uttar Pradesh implies both (1) a low level of female well-being (evident inter alia in massive excess female mortality), and (2) a reduced ability of women to do something for themselves and for other members of the society (leading, for instance, to high mortality rates for male as well as female children).

(44) As several studies have documented, the involvement of women as doctors and teachers often has a positive influence on the functioning of health and education services, e.g. because women prefer to be treated by female doctors (Indian Institute of Management, 1985), or because parents are reluctant to send their daughters to school unless they are taught by female teachers (Gupta et al., 1993, p. 55). In Uttar Pradesh, an overwhelming majority of doctors and teachers in rural areas are male (Indian Institute of Management, 1985; Government of India, 1994a); in Kerala, by contrast, about two-thirds of all primary-school teachers are female (Government of India, 1994a).

(45) In some respects, our findings also bear remarkable similarity to those of similar field-based studies in Uttar Pradesh as well as in other educationally backward regions of India, See particularly Prasad (1987), Kingdon (1994), A. Sinha (1995), and Sinha and Sinha (1995).

(46) In Uttar Pradesh, primary, ‘middle’, and secondary schools are those with classes up to and including the fifth, eighth, and tenth grade, respectively.

(47) This regional division follows the National Sample Survey (see Jain et al., 1988, for a list of the constituent districts). The fifth region, ‘Himalayan’, is not included in our sample.

(48) See Bliss and Stern (1982) and Drèze, Lanjouw, and Sharma (forthcoming); the schooling situation in Palanpur is also discussed in Drèze and Saran (1995).

(49) In fact, both ‘villages’ are officially considered as hamlets within larger residential units recognized as villages for administrative purposes.

(50) It was a common complaint that buildings were left incomplete. Relatively minor construction jobs such as floor levelling, plastering of walls, fitting of doors and windows, and so on, were often left undone. Poor construction standards, exposure to the elements, and lack of use and maintenance, led to many buildings being abandoned altogether.

(51) Government of India (1994a), p. 31, and Government of India (1995), p. 109.

(52) In a recent survey of primary schools in three districts of Uttar Pradesh (Bashir et al., 1993a), a majority of schools reported being in possession of the teaching aids supplied by Operation Blackboard. However, ‘most of these aids and toys are lying in brand new condition in trunks, cupboards or at the head teacher's residence’ (p. 89), and only 10 per cent of teachers report using any teaching aids other than the prescribed textbooks (p. 64).

(53) The last statement is based on the fact that the times of our visits were more or less randomly distributed over the period of the day when schools are officially open. Our findings on teacher absenteeism are consistent with the District Primary Education Programme (DPEP) baseline survey for Uttar Pradesh. In the latter case, half of the teachers were absent at the time of the investigators' visits, despite the fact that the teachers had been informed beforehand of the likely dates of these visits (Dr Sajitha Bashir, personal communication).

(54) The situation appears to be somewhat better m urban areas, but even there teacher absenteeism and shirking are serious problems. In her survey in Lucknow city, Kingdon (1994) found that government schools supplied on average 3.2 hours per day of teaching to grade 8 children.

(55) Mehrotra (1995), Table 2, based on data published by the Ministry of Human Resource Development.

(56) We did observe some teaching in the primary ‘section’ of several schools that also had an upper primary or secondary section. These schools, which tend to be located in larger villages, usually have a relatively large number of teachers, supervised by a headmaster. The scope for shirking is somewhat lower in these schools than in small primary schools where, more often than not, a single teacher is actually present, and acts as his or her own supervisor. In 1986, 29 per cent of all primary schools in India had a single teacher (if any!), and another 32 per cent had only two teachers (Tyagi, 1993, p. 88). In Uttar Pradesh, two-teacher schools are effectively reduced to single-teacher schools for a large part of the year, given the widely-observed fact that, in these schools, teachers ‘frequently take turns in attending’ (Middleton et al., 1993, p. 11).

(57) According to the Fifth All-India Educational Survey, 40 per cent of female teachers in Uttar Pradesh are posted in urban areas, as opposed to 13 per cent of male teachers (National Council of Educational Research and Training, 1992, vol. 11, Table 122).

(58) Calculated from Government of India (1994a), p. 289.

(59) The proportion of female teachers in Uttar Pradesh dropped from 21 per cent to 18 per cent between 1986 and 1992–3 (National Council of Educational Research and Training, 1992, and Government of India, 1994a).

(60) According to Gupta et al. (1993), for instance, in north India ‘reluctance to have daughters taught by male teachers may begin as early as 7 to 8 years of age’ (p. 55), A recent survey carried out in Nepal also finds that employing female teachers leads to ‘lower repetition rates and higher attendance rates, especially among female children’ (CIET International, 1995, p. 1).

(61) On aspects of the social importance of female teaching in India, see Narayana (1995). On issues relating to the socialization of girl children, the content of education, and the reproduction of gender inequalities, see Chanana (1988, 1990) and Devi (1992).

(62) This is consistent with our own direct observations for the village of Palanpur in Moradabad district, which we had the opportunity to visit on several occasions in 1993 and 1994.

(63) This problem has been known for many years (see e.g. Sen, 1970, for an early discussion and Kingdon, 1996 for a recent analysis), but no action has been taken so far. According to some researchers, resistance of teachers' unions to the required changes in the schooling calendar is one major reason for this inertia (Vimala Ramachandran, former advisor to the Department of Education, personal communication).

(64) Government of India (1994a), p. 280. The last figure pertains to all recognized schools, government and private.

(65) This inspection often takes place at the office of the inspector rather than at the school. In his study of schooling in rural Andhra Pradesh, Prasad (1987) also found that school inspectors showed great reluctance to visit rural schools personally.

(66) This practice was widespread among the sample schools. Its implications for the reliability of official enrolment data are obvious enough. The long-standing problem of inflated enrolment figures in India is another example of a well-known but unaddressed flaw of the schooling system. The official figure for the gross enrolment ratio at the primary level in India is well above 100 per cent (Government of India, 1994a, p. 281), but this cheerful figure does not stand up to scrutiny; see Drèze and Sen (1995) for further discussion.

(67) National Council of Educational Research and Training (1992), vol. 1, Table 53. For a thorough discussion of different types of school management in Uttar Pradesh, and of their respective performance in Lucknow, see Kingdon (1994).

(68) This is a somewhat simplified classification, which does not do full justice to the diversity of private schooling establishments. Some private schools, for instance, are based on a mixture of pecuniary and philanthropic motives (as when better-educated but unemployed youth decide to set up an informal school, both to earn some income as well as to contribute to the advancement of their village or community).

(69) For similar observations in a low-income area in Delhi, see Banerji (1995). The high general level of parental motivation for education in our sample villages came up again and again in discussions with parents (with some important qualifications, notably relating to gender and caste, on which see section 4.3), and the performance of the schooling system has to be evaluated in that light.

(70) Kingdon (1994) finds that private-school enrolment in the city of Lucknow is also male-dominated. In the absence of official statistics on unrecognized private schools, there is little state-level information on the gender gap in private-school enrolment.

(71) Geeta Gandhi Kingdon, personal communication based on research in progress (see also Kingdon, 1996). The author's own household survey reveals that 86 per cent of primary-school children in urban Lucknow attend private schools. According to a recent survey carried out by the National Council of Applied Economic Research, the proportion of school-going children attending private schools in rural Uttar Pradesh is 27 per cent—by far the highest figure among all Indian states (Shariff, 1996).

(72) Tyagi (1993), p. 54. The corresponding all-India figures are 94 per cent and 99 per cent, respectively.

(73) A household survey conducted in Palanpur (Moradabad district) in 1983–4, at a time when the village school was non-functional, found that only three girls in the whole village were attending school in neighbouring villages (compared with twenty-nine boys). See Drèze and Saran (1995) for further discussion. Limited mobility of girls outside the village is also a serious constraint on female education beyond the primary level, given that middle and secondary schools tend to be more distant than primary schools (see Macdorman, 1986).

(74) Fifth All-India Educational Survey (National Council of Educational Research and Training, 1992, vol. 1, Table 17). The situation for predominantly scheduled-caste rural settlements (of 300 persons or more) was even worse. Less than a third of these in Uttar Pradesh had a primary school, compared to 83 per cent in South India.

(75) The Ramamurti Committee, set up by the government in 1991 to review the National Education Policy of 1986, also identified teacher absenteeism as one of the major constraints on educational development (see Bandyopadhyay, 1991). The revised National Education Policy of 1992, however, has nothing concrete to say on this issue. Nor does the problem receive any serious attention in the companion ‘plan of action’ (Government of India, 1992b).

(76) For a broader discussion of the relative efficiency of public and private schooling in Uttar Pradesh, see Kingdon (1994); see also Bashir (1994) on Tamil Nadu, and Jimenez et al. (1991) for an overview.

(77) There are other interesting examples of rent extraction based on the nexus between the schooling establishment and the local elite. In several sample villages, for instance, the headman was involved as a contractor in the construction and repair of local school buildings. There are handsome profits to be made from such activities, particularly by leaving the work unfinished or by using inferior materials.

(78) On this, see also Government of India (1986).

(79) For further discussion of this point, see chapter 5 in the companion volume (Drèze and Sen, 1995).

(80) It is worth noting that the paucity of empirical evidence is itself a reflection of the low attention which basic education has received in the literature on Indian development. Most household surveys (including the National Sample Survey) include information on literacy and school attendance, which could have been used with good effect to examine issues such as the determinants of school attendance, the circumstances that lead to children dropping out, and intra-household inequalities in schooling decisions. Few studies, however, have investigated these issues in any depth. For some recent contributions in this field, see Minhas (1992), Kingdon (1994), Subramanian (1994). Banerji (1995). Labenne (1995), Mehrotra (forthcoming).

(81) See e.g. National Sample Survey data (42nd round, 1986–7) presented in Sarvekshana, January–March 1991.

(82) Recent studies of the demand for schooling place growing emphasis on these ‘non-income’ factors (see. for instance, Coklough, 1993).

(83) The last author also presents an illuminating analysis of the political economy of the ‘poverty’ argument: ‘It should he clearly understood that acceptance of the premise that poverty compels parents to send their children to work is extremely convenient to those charged with the responsibility of reducing if not eliminating child labour because in such a case, improving the economic status of the parents becomes the focal point of attention. This is neither the responsibility of the labour or the education department and the buck can be passed elsewhere’ (pp. 32–3).

(84) On this point, see also Mehrotra (forthcoming). The quality of schooling affects not only the motivation of parents, but also that of children. And as Bashir et al. (1993b) note in their survey of primary schools in three districts of Uttar Pradesh: ‘Many parents … mentioned that the continuing of their child's education would depend on the child's interest’ (p. 6).

(85) For similar assessments, see Werner (1991), Banerji (1995), Maharatna (1995), Mehrotra (1995), S. Sinha (1995). Kerala's experience, discussed in Ramachandran's contribution to this volume, is another crucial illustration of the possibility of rapid educational expansion at an early stage of development.

(86) It is worth mentioning that literacy rates in Palanpur are very close to the corresponding figures for Moradabad district as a whole. For instance, in 1991 the rural male and female literacy rates in the district (age 7 +) were 37 and 10 per cent, respectively, compared with 37 and 9 per cent in Palanpur in 1993. This village cannot, therefore, be dismissed as a ‘basket case’, despite its astonishingly poor record of educational expansion (on the latter, see Drèze and Saran, 1995, and Drèze, Lanjouw, and Sharma, forthcoming).

(87) Even though the Kayasth population in Palanpur was quite small in 1993, this contrast is unlikely to be fortuitous. Indeed, the same pattern applies for each of the five surveys that have been conducted in Palanpur since 1957–8. Similar caste-based contrasts in educational achievements have also been observed elsewhere in India. See, for instance, the chapter on West Bengal in this volume.

(88) For instance, probit analysis of the determinants of school attendance in Palanpur indicates that children from Muslim and Jatab families are less likely to attend school than children from other families, even after controlling for per-capita income and the education of parents.

(89) On this, see also Drèze and Saran (1995), and Drèze and Sen (1995), chapter 6.

(90) Hard-boiled neoclassical analysis à la Becker might suggest that the ‘externalities’ of investment in female education are internalized through lower dowry payments. But the real world seems to work quite differently, judging from the fact that the fear of having difficulties marrying a well-educated daughter is very real for many parents in rural India (see Drèze and Sen, 1995, p. 135, and the studies cited there, and also Macdorman, 1986, and Bashir et al., 1993b, p. 17). It is quite possible, however, that as levels of male literacy in a particular community rise beyond a certain threshold, female education becomes an asset rather than a liability from the point of view of marriage prospects, because most young men aspire to marry a literate bride (see e.g. U. Sharma, 1980, Jejeebhoy and Kulkarni, 1989, and Minturn, 1993, for some relevant evidence).

(91) Blatant discrimination has not disappeared altogether. In Banda district, for instance, there have been cases of active interference of Dadu Brahmin landlords with the schooling of scheduled-caste people. For similar observations elsewhere, see Wadley and Derr (1989) and Banerjee (1994).

(92) To illustrate: ‘SC children refuse to go to school, being afraid and hurt of the scorns thrown at them’ (Shami, 1992, p. 26, about a village school in Madhya Pradesh); ‘the majority of the [school-going] children are Rajputs and the SC children are so often rebuked and hit by the teachers that they have virtually stopped going to school’ (Varma, 1992, about a Bihar village); ‘these [scheduled-caste] children often sat at the back. The teacher hardly paid any attention to them’ (Bashir et al., 1993b, pp. 20–1, on a village school in UP). See also Mehrotra (forthcoming) on peer beating, and Lata (1995), p. 32 for further examples. On the disadvantaged access of scheduled-caste settlements to schooling facilities in Uttar Pradesh (in contrast with South India, where no such bias can be observed), see National Council of Educational Research and Training (1992), vol. 1, Table 17.

(93) On the role of caste (and tribe) based peer reference in forming social attitudes, particularly with regard to education, see Sharma (1981) and Bara et al. (1991). A related issue is the elitist nature of the school curriculum and teaching methods, which may further discourage the participation of children from disadvantaged social backgrounds; on this, see Kamat (1985) and particularly Kumar (1994).

(94) Some illustrations may help. In the course of a similar investigation in Kerala in February 1995, we found that most primary schools there closely monitored school attendance, and contacted parents in the event where a child fails to turn up at school for a number of days. Similarly, most schools provided midday-meals, had an active parent-teacher association, and were supported by crèche facilities for pre-school children. None of these practices were observed in any of the schools we visited in Uttar Pradesh.

(95) Note the parallel with a similar observation reported many years ago by Iqbal Narain (1972) in an enlightening account of educational politics in Rajasthan: ‘all the villagers may be dissatisfied with a school teacher, yet if he is in the good books of the sarpanch and pradhan he is not transferred’ (p. 152).

(96) On this, see Prabhu (1995), p. 37. Over the same period, there has also been some decline in many other states, but at a much slower rate than in Uttar Pradesh.

(97) On the latter, see Tyagi (1993), p. 84. The number of primary-school teachers per 1,000 pupils declined from 26 to 17 between 1981–2 and 1992–3.

(98) On this, see Ghosh et al. (1994). The authors, after noting that the Total Literacy Campaign had made little impact in the most educationally backward states, including Uttar Pradesh, argue that this poor response is primarily due to a ‘low political commitment to the eradication of illiteracy’ in those states (p. 39).

(99) It is perhaps no coincidence that the Total Literacy Campaign has its antecedents in campaigns initiated by political and social organizations in Kerala and Tamil Nadu, one of whose primary concerns was political mobilization of illiterate people (see Rao, 1993). The link between universal basic education and political participation is also strong m the history of developed economies (see e.g. Simon. 1965, on the correlation between educational and electoral reform in Britain).

(100) For similar findings on a wide range of public services and government programmes in 8 villages of Uttar Pradesh, see A. Sinha (1995). For a detailed case study, based on Palanpur (Moradabad district), see Drèze, Lanjouw, and Sharma (forthcoming).

(101) See e.g. Singh and Misra (1964) and Haque and Sirohi (1986). On the political aspects of this failure to institute major land reforms, see also Kohli (1987) and Hasan (1989).

(102) For a contemporaneous account of the failure of Gramdan at the village level, see Planning and Research Action institute (1966). See also Church (1974).

(103) For some examples, see Shankar (1991a) and Saith and Tankha (1992).

(104) See Singh and Singh (1989), Drèze (1990a), Shankar (1991b), Saith and Tankha (1992), and A. Sinha (1995), among others. Fraudulent practices include large-scale selection of ineligible beneficiaries, disregard for the prescribed norms of consultation with villagers, and endemic extortion of bribes by bank managers, gram sevaks, and other government officials. Interestingly, the implementation flaws of recent anti-poverty schemes have much in common with those of earlier agricultural extension programmes. See, for instance, H.K. Singh (1958) on the appropriation of benefits of the National Extension Service by large landlords.

(105) Keshav Desiraju (IAS), former ICDS Director for Uttar Pradesh, personal communication. A recent field investigation in rural areas of Allahabad district found most anganwadis to be non-functional (Nidhi Mehrotra and Sangeeta Goyal, personal communication). As was mentioned earlier, this is also our own experience in the sample villages.

(106) For some relevant studies on these different initiatives, see Guhan (1981, 1990). Drèze (1990b), Mahendra Dev (1993a, 1993b), Crook and Manor (1994), Visaria and Visaria (1995), Goyal (1995), among others, also the chapters on Kerala and West Bengal in this volume.

(107) Corruption has become the working norm of most government departments in Uttar Pradesh; see, for instance, Gupta (1994). In our field work, we did encounter occasional examples of public-spirited individual initiatives. A sub-divisional magistrate in Pratapgarh, for instance, was credited for having resolutely tackled local teacher absenteeism, with some temporary success. The situation reverted back to ‘normal’, however, soon after she was transferred.

(108) Each of these examples of ‘abuse’ has been observed in one or more of our sample villages.

(109) For detailed accounts of the jajmani relation and their decline over time in various parts of Uttar Pradesh, see Elder (1970), Wadley and Derr (1989), Wiser and Wiser (1970), among others, on the provision of public services within the framework of jajmani or other patron—client relationships in pre-independence India, see also Matthai (1915).

(110) On traditional schooling arrangements in rural India, see Matthai (1915), and especially Dharampal (1983).

(111) Scholarly assessments of village society under jajmani have undergone significant change from the initial stress on reciprocity and harmony to the subsequent recognition of exploitation and oppression. See, for instance, the changing perceptions of jajmani m Karimpur, a village of western Uttar Pradesh which has been studied over a period of seventy years (Wiser, 1936, Derr and Wadley, 1987, Wadley and Derr, 1989).

(112) See e.g. P.Jha (1994), Mahapatra (1994), and Janakarajan (1995) for evidence of this problem in Bihar, Tamil Nadu, and Orissa, respectively. In the case of irrigation, of course, this decline of collective facilities was more than compensated, in due course, by private investment in irrigation devices.

(113) See, for instance, Khare (1964) and A. Singh (1977). In contrast, Elder (1970) gives an interesting account of how, in a village where jajmani was still alive and well, the landlords persuaded a Bhangi (sweeper) family to settle in the village in return for a small plot of land.

(114) Some of the early optimism in the ability of village communities to smoothly take over the task of ‘community development’ was based, perhaps, on idealized notions of harmonious village communities. See, for instance, Danda (1987) for an account of the breakdown of collective efforts in a village considered promising m the early years of community development.

(115) Ford Foundation (1992), based on the recent Krishnaswamy Report; see also Vyasulu (1993) and Crook and Manor (1994).

(116) See the chapter on West Bengal in this volume, and the references cited there.

(117) See e.g. Gokhale (1988) and Maithani and Rizwana (1992).

(118) For instance, a recent study of panchayat institutions in Dobhi block of eastern Uttar Pradesh (H.N. Singh, 1993) found that as many as two-thirds of the headmen in 82 surveyed villages belonged to the Thakur caste — the traditional landowning upper caste in that region, notorious for oppressive subjugation of the lower castes. Among the available village studies for Uttar Pradesh (see footnote 4), a large majority of those that provide information on the village headman describe him as a high-caste landlord (usually Thakur or Brahmin).

(119) See e.g. Berreman (1963) for an early study on the prospects for community development in Uttar Pradesh under conditions of extreme social and economic inequality, and Srinivas (1962) on the danger of political devolution m India leading to ‘the tyranny of the dominant caste’ (p. 32).

(120) See e.g. Chandra (1958), Dube (1958), Khare (1962), Marriott (1958), and Zamora (1990).

(121) See, for instance, village studies by H.A. Singh (1969) and K.K. Singh (1967).

(122) It is interesting to note that in two separate village studies undertaken in different districts of eastern Uttar Pradesh (Zamora, 1990, for Jaunpur, and Y. Singh, 1970, for Basti), similar political developments were observed in the period immediately following the abolition of zamindari and the institution of elected panchayats. In both villages, low-caste tenants initially asserted their political independence from the former (Thakur) zamindars but the latter returned to power within ten years, having demoralized their poorer opponents by threats of economic sanction and by exploiting factional divisions among the former tenants.

(123) See e.g. Khare (1962), Park and Tinker (1963), Retzlaff (1962), and Kantowsky (1968), for village studies in various parts of the state.

(124) See for instance Y. Singh (1970) on change and continuity in the relations between jajmans (patrons) and prajas (clients), and in their roles as political leaders and followers. A similar picture emerges from the fictionalized history of one particular village in eastern Uttar Pradesh in the popular Hindi novel Adha Gaon by Rahi Masoom Raza.

(125) On the predominance of factional considerations at the state level, see also Kohli (1987) and Brass (1965, 1990). P.K. Srivastava (1991) shows that factional battles have figured prominently in the election of Chief Ministers in Uttar Pradesh throughout its electoral history.

(126) Hardiman (1994) argues against the use of the term ‘faction’ in studies of Indian politics, on the grounds that it adds nothing to the more precise and widely applicable term ‘class collaboration’. Without disagreeing with his contention that a ‘faction’ is essentially an elite-led coalition which coopts sections of the economically exploited, we retain the use of the term since it aptly captures the specific form which class collaboration has taken in the context of village India.

(127) On this phenomenon, see e.g. Government of Uttar Pradesh (1964a, 1966), Dubey (1965), Rastogi (1965, 1966), Sahay (1969), Y. Singh (1970), M. Sharma (1978), Fukunaga (1993), among other relevant village studies.

(128) See, for instance, Khare (1962) and Hale (1978). The situation in Uttar Pradesh can be seen as an extreme form of the general relationship between public resources and private patronage in Indian political life. See Bardhan (1988) for an analysis of this phenomenon.

(129) On the use of the schooling system, both public and private, as a channel of patronage, see contributions by Narain (1972) on Rajasthan and Gould (1972) on Uttar Pradesh in Rudolph and Rudolph (1972).

(130) In fact, early studies that identified factionalism as an important feature of rural politics in Uttar Pradesh often considered it as a necessary stage in the process of political development (see e.g. Brass, 1965).

(131) On this, see particularly Hasan (1989).

(132) See e.g. Bose (1993), and the chapter on West Bengal in this volume. On anti-Brahmin movements in a number of Indian states, see Srinivas (1966), OʼHanlon (1985), and various contributions in Frankel and Rao (1989).

(133) See, for instance, Kohli (1987), Hasan (1989), and Brass (1990).

(134) For a startling case study of ‘deficient democracy’ in Uttar Pradesh, see Lieten (1994); the problem of lack of real local democracy analysed in that study has also been noted in many of the village studies cited earlier in this chapter.

(135) See, for instance, Srinivas (1966), Kakrambe (1983), OʼHanlon (1985), and Omvedt (1994) for accounts of such traditions, particularly in ‘peninsular’ India (south India and Maharashtra).

(136) See, for instance, the analysis of the role of Charan Singh, widely regarded as one of the main architects of zamindari abolition in Uttar Pradesh, in Hasan (1989) and Byres (1988).

(137) On this contrast, see Omvedt (1995).

(138) We use the term ‘dominant caste’ in the sense discussed by Srinivas (1960, 1987), i.e. that of a caste which through a combination of numerical strength and economic advantage (usually land ownership) holds political power at a local level.

(139) Why these martial castes are there in the first place is a different matter. Their influence in this region has a long history, going back at least to the Mahabharata epic, which is built around the story of warring Kshatriya families. The hold of martial castes on the land is understandable, given that land ownership has often had to be won or defended through violent means in the past, especially perhaps in this fertile region frequently exposed to raids and invasions.

(140) There have been sporadic attempts, earlier on, at caste-based mobilization based on a critique of the ideology of caste hierarchy; see e.g. Gooptu (1990) for an account of Chamar-led agitation in urban centres of Uttar Pradesh in the 1920s. These initiatives failed, however, to give rise to sustainable political organizations.

(141) On the regional patterns of class—caste relations in India, and on the history of various ‘low-caste’ movements in Maharashtra and south India, see Srinivas (1966), OʼHanlon (1985), Frankel and Rao (1989).

(142) Byres (1988) notes that the tone of Charan Singh's movement was ‘anti-Brahmin’ on the one hand, but contemptuous of the scheduled castes on the other. See also Omvedt (1995) on the anti-labour stance of the new farmers' movements of western Uttar Pradesh.

(143) Interestingly, the current Chief Minister of the state is a scheduled-caste woman (Mayawati, a BSP leader). While this is certainly a path-breaking development, its origin is also reminiscent, in some ways, of the phenomenon of unprincipled factionalism discussed earlier. Indeed, the main political force within the coalition which brought Mayawati to power is the Bharatiya Janata Party (BJP), a party staunchly apposed to ‘low-caste politics’.

(144) An extreme manifestation of this patriarchal culture is the practice of female infanticide, which used to be common among Rajputs, Jats, and other dominant landowning communities of north India, and contributed to remarkably low female—male ratios in these communities. On these issues, see Panigrahi (1972), Miller (1981, 1993a), Kishwar (1995), and also related analyses in Caldwell and Caldwell (1987), Harris (1993), Agarwal (1994), Karve (1994), among others.

(145) On this aspect of the culture of the martial castes in Uttar Pradesh, see Minturn (1993), also Hitchcock (1975). On the general links between militarism and patriarchy in different societies, see Harris (1993).

(146) On this influence, see the chapter by Murthi, Guio, and Drèze in this volume, and the literature cited there. On the contrasting status of women among Jat landowners and scheduled-caste labourers in north India, see Horowitz and Kishwar (1982) and Majid (1986).

(147) On this phenomenon of ‘Sanskritization’, see Srinivas (1962, 1966, 1989). On the history of caste-based social mobility in rural Uttar Pradesh, see also Cohn (1990) and Wiser and Wiser (1970).

(148) On the transition from bride-price to dowry among some scheduled castes of Uttar Pradesh, see e.g. Macdorman (1986) and Wadley and Derr (1989); on widow remarriage, see Kolenda (1983) and other studies cited in Drèze (1990c) and Chen and Drèze (1995).

(149) This problem has been further reinforced by the practices of patrilocal residence and village exogamy, which further isolate women from each other and from their natal families. In all these respects, the situation of women has been considerably more favourable in south India, where (in addition to higher female labour-force participation rates) the kinship system gives greater scope to female autonomy and solidarity. On the north—south contrast and other regional differences in kinship systems, see Karve (1965), Sopher (1980b), Dyson and Moore (1983), Caldwell and Caldwell (1987), Kishor (1993), Agarwal (1994), and Agmhotri (1994), among others.

(150) In 1987–8, the Gini coefficients of per-capita consumer expenditure in major Indian states were fairly evenly distributed between 0.26 and 0.33, with Uttar Pradesh (0.293) very close to the middle of this interval (see Drèze and Sen, 1995, Statistical Appendix, Table A.3). Nor are inequalities of land ownership in Uttar Pradesh likely to be particularly large compared with other states, given the relatively small proportion of landless households.

(151) For some examples (from Uttar Pradesh and elsewhere in north India), see e.g. Saith and Tankha (1992), M. Sen (1993), Kumar (1993), Srivastava (1995, forthcoming), and Bhatia (forthcoming). For further discussion of the phenomenon of endemic violence in Uttar Pradesh, see Oldenburg (1992).

(152) According to the Chief Election Commissioner, 180 of Uttar Pradesh's 425 Members of Legislative Assembly (MLAs) have criminal cases pending against them, at least 52 of which involve ‘heinous crimes’ such as rape and murder. See Singh and Ahmed (1995), who consider that Uttar Pradesh surpasses all other states (even Bihar) in this respect, and that the crime—politics nexus ‘seems to have overwhelmed politics in Uttar Pradesh’ (p. 119).

(153) As one commentator insightfully put it at a recent seminar on economic reform, the government's reform programme has three components: deregulation, fiscal adjustment, and liberalization of international trade (Dr Amresh Bagchi, National Institute of Public Finance and Policy).

(154) This general issue is explored in greater depth in the companion volume (Drèze and Sen, 1995).