Among the propositions so thoroughly confirmed that they may claim the status of sociological laws is that which states that marriage tends to 'protect' individuals from suicide. Durkheim noted that married people over 20 of both sexes -- whose social integration was relatively higher--were very much less likely to commit suicide (Durkheim, 1979). Durkheim's findings have been replicated in many subsequent studies. In an examination of United States suicides in 1940 Henry and Short (Henry & Short, 1954) found rates for married persons were lower than for others in the same age group. Divorce has the opposite effect. Breault (1986, cited in Lester, 1989) who examined the effects of church membership, divorce and inter-region migration on the suicide rates of the American states at different times during the period of 1933 to 1980, found that suicide rates were higher where divorce rates were higher. Many suicide studies of Western populations indicate a pattern of higher suicide rates at older ages, often fol lowing the death of or separation from a partner, with males exceeding females (Booth, 1999). In a study of suicide in Singapore, Hassan (1983: 5460) found that marriage provides considerable immunity against suicide and that, on average, divorce and widowhood provide least protection. Married people in Australia were also found to be less likely to commit suicide (Hassan, 1995). Ruzicka and Choi (Ruzicka & Choi, 1993: 108) summarize the marriage law in unqualified terms:
In all societies for which there are statistics available, suicide mortality has been considerably higher among those who are not married than among those who are married.
It is important to note that most of the literature on suicidal behavior is based on data obtained from developed Western countries (Latha et al., 1996). The impact of marital status on suicide in developing countries has received little attention. This is especially true of India where the relationship between marital status and suicide does not appear to have received systematic examination. Suicide is a growing problem in India, one whose seriousness is rarely acknowledged by policy-makers. In 1995, 89,178 suicides were recorded for the whole of India. Between 1985 and 1995 the official suicide rate for the country as a whole rose from 6.8 to 9.9 per 100,000, an increase of 69% (National Crime Records Bureau, 1997). A number of the Indian states, especially those in South India and West Bengal have suicide rates, which are high by world standards. The Union Tenitory of Pondicherry, for example, had a suicide rate in 1995 of 68.5 per 100,000. Kerala had a suicide rate of 25.9 and Tamil Nadu 15.0 per 100,00 0 (National Crime Records Bureau, 1997).
The present paper explores the impact on suicide rates of gender and marital status in India. It was hypothesized on the basis of established findings elsewhere that suicide rates for those who are married would be lower than for all other marital categories. It was also predicted that -- with two significant exceptions -- for all marital categories, male suicide rates would be higher than female rates. The predicted exceptions were for suicides by widows and widowers and for those who were divorced. Because of the traditional stigmatization of widows, it was hypothesized that their suicide rates would be higher than those of widowers. It was also predicted that the social disapproval of divorce in Indian society would result in higher suicide rates for divorced women than for men.
Our data are official suicide statistics provided by India's National Crime Records Bureau. The use of official statistics has been attacked by Douglas (1967) and Baechler (1975, cited in Danigelis & Pope, 1979) on the grounds that their validity is undermined by systematic variations between nations in the meanings …