
For the 50 years since Independence that the Census of India has been operating, the only data that have been disaggregated by religion are those pertaining to fertility. As the previous Registrar General of the Census of India told us, although data by caste and community are collected, they are not processed as such. Except on fertility. It would be instructive to find out why this sub-set has been the exception, especially in view of the manner in which it was presented for the 2001 Census. This time, other data sets, on education levels and distribution of workforce, are also presumably being processed by religion, but only religion-wise data on fertility has been released without taking into consideration other socio-economic parameters.
Disaggregating various kinds of data by social, economic, gender and even religious or community groupings can be a very useful and informative exercise, provided that these are appropriately contextualised and cross-related with each other. So, for instance, data on fertility rates—whether stable, declining or increasing—can only be usefully analysed when placed alongside other data on income disparities, social backwardness, education, health, and so on.
One of the standard assumptions about Muslims is that religion prevents them from practising family planning. The current controversy around the so-called (incorrect) Muslim rate of growth only underlines the hazards and irresponsibility of focusing on one variable alone, when compared across income groups it will become clear that declining fertility rates is an indication of improved economic status and low morbidity, rather than of being Muslim or Hindu. Middle-class Hindu and Muslim women have roughly the same number of children, just as a nation-wide survey of Hindu and Muslim women that we conducted in 2000-2001—at the same time as the national Census—showed that the average family size of poor Muslim women was 3.5, of Hindu women, 2.8. Not such a huge difference after all.
The findings of our own survey of 10,000 women, 80 per cent Muslim, 20 per cent Hindu, in 42 districts are an eye-opener. Let’s take the question of fertility. Why have the Census and all our demographers been silent on the delicate matter of age at marriage, arguably the single most-critical determinant of fertility for Indian women? According to the Muslim Women’s Survey (MWS), the mean age at marriage for Indian women is 15.6 years—a full two years below the legal minimum age. In rural north India, it is as low as 13.9; rural west reports 14.2 years; and rural south, 14.3. For upper-caste Hindu women in rural India, it is actually lower than that of Muslim women, except in the rural east.
Health activists, women’s groups working on reproductive health, and other social movements have repeatedly drawn attention to the importance of medical facilities and the eradication of poverty in limiting family size. But why not begin by simply observing the legal age at marriage for women? Imagine what our fertility data would look like if child-bearing were delayed by at least two years for all the women in the country. Women prefer it too.
Further probing by our survey revealed that given a choice, women would marry at 18 or later, citing—interestingly enough—safer motherhood as an important reason. Delayed marriage might also have a positive impact on educational attainment—chances of continuing in school if you’re married by 15 or 16 are remote—and, consequently, on smaller families.
The low socio-economic status of Muslims is now well-known. However there are significant regional differences. Muslims are generally poor in the north and the east, but less so in the south. Furthermore, it would be a mistake to think that socio-economic backwardness is a problem for Muslim women alone in these regions. The majority of women, Hindu and Muslim, are poor; and a closer analysis of comparative data from our survey throws up some interesting parallels. For instance, financial constraints are an important reason for girls dropping out of school; figures for Muslim girls are higher not because the community is backward or conservative, but because government, sex-segregated schools are not accessibly located. The fact that Hindu girls have better access to them makes for an improvement in their figures. There are other reasons too, to do with medium of instruction, the cost of private tuitions (almost a necessity for all students), high dropout rates for Muslim boys, which adversely impacts Muslim girls, and so on.
However, puberty and purdah, the two factors usually cited, are almost irrelevant. Of course, because Muslims as a social group are much poorer than Hindus across the country, the status of Muslim women is correspondingly worse. But, whether Hindu or Muslim, if parents have to choose between educating sons or daughters, they will choose the former. If early marriage is the norm for all women, we would do better to look into the reasons for this phenomenon that cuts across caste and community rather than isolating religion as the culprit. Here too income disparities and compelling social and sexual mores seem to be much more important, and it is now generally agreed that an improvement in economic status makes for a dramatic improvement in all round-literacy, health, morbidity and mortality, age at marriage, and obviously fertility.
If we want to see a real difference in all our gender-related Census data, from sex ratio onwards, we will simply have to look seriously at why, and how, women’s subordination is perpetuated for all women—Hindu, Muslim, Sikh, Dalit, Christian, tribal and so on. We will have to ask why the sex ratio is lowest among Sikhs; why in Kerala, which has the best social indicators for women, female workforce participation is the lowest in the country; why figures for girls’ education are the lowest in Bengal; why Rajasthan and Bihar report the highest number of minor/child marriages; why infant mortality is highest among mothers below the age of 20; why son preference persists across the board; and so on. Religion may be a convenient alibi, but gender and class inequities may turn out to be the truly disempowering factors.