Although both Chinese and Indian traditions favored large extended families, such families were always rare in immigrant Singapore where neither the occupational structure, based on wage labor, or the housing pattern, characterized by small, rented quarters, favored such family forms.

In the 1980s, families were important in that most individuals as a matter of course lived with their parents until marriage and after marriage maintained a high level of interaction with parents, brothers, and sisters.

Probably the most common leisure activity in Singapore was the Sunday visit to the grandparents for a meal and relaxed conversation with brothers, sisters, in-laws, uncles and aunts, cousins, and other assorted kin.

Although the age of marriage increased in the 1970s and 1980s, reaching a mean 28.5 years for grooms and 25.8 years for brides in 1987, Singapore remained a society in which it was assumed that everyone would marry, and marriage was a normal aspect of fully adult status.

Both ethnicity and class affected the form and functioning of families. Chinese and Indian families rested on cultural assumptions of the permanence of marriage and of the household as an ongoing, corporate group whose members, bound by duty, obligation, and subordination, pooled and shared income.

The continued efforts of Indian parents to arrange the marriages or at least to influence the marital choices of their offspring and the Tamil obligation to provide daughters with large dowries reflected such cultural definitions of family and household.

In a similar manner, some Chinese combined the household with the family enterprise, practicing a traditional entrepreneurial strategy that included mobilizing the savings of all household members and allocating them in accord with a long-term plan for family success.

Such a strategy might take the form of a thriving business with branches in the major cities of Malaysia and Indonesia, or of sons and daughters employed in the Singapore civil service, a large foreign bank, or a university in Australia.

Malay families, on the other hand, gave priority to the individual and to individual interests.

They viewed relations between siblings as tenuous and saw the household as a possibly short-lived coalition of autonomous individuals linked by sentiments of mutual concern and affection.

Malays had traditionally had much higher rates of divorce and adoption than other ethnic groups, and the distinction continued in the 1980s although the divorce rate was lower than in the l940s or through the l960s.

More significantly, for the Malays divorce was regarded as a realistic and normal, although unfortunate, possibility in all marriages.

Because Malays did not define the household as a continuing body, they did not make long-range strategic plans to maximize family income and success.

In Malay families, husbands, wives, and children with jobs held separate purses and sometimes separate savings accounts.

It was thus difficult for Malays to establish family businesses as the Chinese and the Indians did.

Class affected families in a manner generally similar to many other industrialized societies.

In all ethnic groups, lower-class or working-class people tended to be dependent on kin outside the immediate household for a wide range of services, and to operate wide networks of mutual assistance and gift exchange.

Throughout the 1980s, kin provided the bulk of child care for married women working in factories.

Such relatives were paid for their services, but less than a stranger would have been paid.

The possibility of such support often determined whether a woman took a job outside the home, and thus demonstrated the relation between large numbers of kin and material comfort and security.

Substantial sums of money were passed back and forth on such occasions as the birthdays of aged parents, the birth of children, or the move into a new apartment.

Family members were a major source of information on and referrals to jobs for many unskilled or semiskilled workers.

Relations with the extended circle of relatives were not always harmonious or happy, but they were important and necessary to the welfare and comfort of most working-class families. Middle- and upper-class households were less dependent on kin networks for support.

They maintained close ties with parents and siblings, but did not need to rely on them.

Indeed their relations with their extended kin often were more amiable than those of the lower-class households, where mutual need often was accompanied by disputes over allocation of such resources as grandparents' childcare services, or of the costs of supporting elderly parents and other dependent kin.

Middle- and upper-class households spent more leisure time with people who were not their relatives and gained much of their social support from networks based on common schooling, occupation, and associational memberships.

In such families, the bond between husband and wife was close as they shared more interests and activities than most workingclass couples and made more decisions jointly.

Marriages across ethnic lines occurred, but not often. Between 1954 and 1984, intermarriage rates remained at 6 percent of all marriages.

None of the traditional cultures encouraged marriage outside the group. The Hindu traditions of caste endogamy and the Malay insistence on conversion to Islam as a condition of marriage were major barriers to intermarriage.

Shared religion encouraged intermarriage, with marriages between Malays and Indian Muslims the most common form of ethnic intermarriage.

Interethnic marriages included a disproportionate number of divorced or widowed individuals. Divorce rates in Singapore were low. Interethnic marriages were somewhat more likely to end in divorce than were marriages within an ethnic group.

During the 1980s the divorce rate for Malays fell, while it rose for the other ethnic groups.

In 1987 there were 23,404 marriages in Singapore, and 2,708 divorces, or 115 divorces for every 1,000 marriages.

The figures included 4,465 marriages under the Muslim Law Act, which regulated the marriage, divorce, and inheritance of Muslims, and 796 divorces under the same act, for a Muslim divorce rate of 178 divorces for every 1,000 marriages.

Marriages under the Women's Charter (which regulated the marriage and divorce of non-Muslims) totaled 18,939, and divorces under that law were 1,912, for a non-Muslim divorce rate of 100 per 1,000 marriages.

The differential rates of divorce for ethnic groups may have suggested greater differences than were in fact the case.

Situations that for Malay families resulted in prompt, legal divorce were sometimes tolerated or handled informally by Chinese or Indian families for whom the social stigma of divorce was greater and the barriers to legal separation higher.

For all ethnic groups, the most common source of marital breakdown was the inability or unwillingness of the husband to contribute to maintaining the household.