Ma Chunhua

　　 In comparison to Europe and the United States, the Chinese Mainland, South Korea, Japan and Chinese Taiwan—one could say East Asian society in general—seem to regard family with much greater importance. During the past fifty years, the countries and regions within East Asia have seen varying degrees of change in politics, society, economy, culture and demographics. Changes in family structure are a part of social transition, which will inevitably engender changes in every aspect of family life.

　　Many researchers have observed that Chinese families have shifted to a nuclear family pattern, with the “family network” as a supplement to the nuclear unit. Some, however, have denied the existence of such a shift, maintaining that the nuclear family has always been the predominant family structure in Chinese cities, while the stem family continues to be the basic structure of Chinese families. Japan preceded South Korea, the Chinese Mainland and Chinese Taiwan in embarking on the path of modernization, and the hypothesis of a correlated shift to the nuclear family pattern in Japan was once accepted and validated by researchers. However, researchers have recently discovered that the proportion of nuclear families relative to Japanese society as a whole increased only after the 1960s.While extended families (non-nuclear families) decreased in proportion, they did not decrease in absolute number. The increase in nuclear families after the Second World War was not because the three-generation extended family structure was damaged in Japan’s economic restructuring, but rather was the result of a demographic dividend (a period where low birthrates and a shorter life expectancy in the older generation result in a high ratio of labor force to dependents within the total population). Moreover, the normative role of the linear (stem) family was not weakened in nuclear families. These families still kept close relations with their relatives, who provided strong support in child rearing and elder-care.

　　The hypothesis of the transition to the nuclear family pattern was also universally accepted in South Korea. Nuclear families composed the bulk of Korean families in pre-industrial society. However, the South Korean nuclear family did not conform to the classical stereotype of Western modernity in which the father is the sole source of income and bears the family’s living costs and his children’s education expenses. Usually, the nuclear family would live nearby—sometimes even with—relatives, but would stay financially independent.

　　Some demographers have claimed that a societal shift to the nuclear family structure is an actual fact of the process of industrialization and urbanization. Others, however, contend that it is only superficial, and that the nuclear family is simply a transitional form of the stem family. Another viewpoint is that rapid industrialization is in fact being accompanied by both types of family structure, and does not spell an immediate societal shift to nuclear families. Some researchers have even proposed looking at the family structure as a dynamic process for the family members who constitute it: these family members might experience a disintegration and reorganization throughout different stages of life. Most people spend most of their life in a traditional stem family with their parents and children, which is the most stable status.

　　Utilizing the statistics and research from the 2006 East Asian Social Survey (EASS), the research tries to analyze the trajectory of patterns of change in family structure among different East-Asian countries and regions. As the countries and regions we look at achieve different levels of economic development and modernization, we examine whether patterns in the changes of family structure take on unique differences and distinctive features and assess the relation between modernization and changes in family structure. Is the nuclear family the primary family structure in these countries and regions? If so, does the extent to which nuclear families stay connected with their kinship networks vary between locales?

　　We assessed the level of modernization in the Chinese Mainland, Japan, South Korea and Chinese Taiwan by comparing their GDP per capita and Human Development Index. Ranked from high to low, we found the sequence to be: Japan, South Korea, Chinese Taiwan and the Chinese Mainland.

　　We employed and modified the definition of nuclear family articulated by George P. Murdock and Talcott Parsons, i.e. a married couple and their children living in a single household, and serving several societal functions: reproduction, parenting, emotional stabilization, and economic production and consumption, all of which are performed completely by the couple, independent from the help or support of kinship networks. The presumption of a strict division of labor based on gender (i.e. the husband supports the family and plays the instrumental role while the wife keeps house and plays the emotional role) as presupposed in earlier Western theories of the modern family structure such as Murdock’s and Parson’s is not included here. The basic unit for analysis is the family household (all of the people living in a household, including but not limited to the family) rather than family. Altogether we collected 3,208 valid samples from the Chinese Mainland, 2,130 from Japan, 1,605 from South Korea and 2,102 from Chinese Taiwan.

　　Within the basic division of nuclear, stem and joint family, we further divided the family structure of the surveyed countries and regions into the following categories: single-adult family, nuclear family, one-couple family without children or where children live away, stem family, joint family (two or more married couples within one household), extended family, inter-generational family (a family with a span of at least three generations between members of the household and a discontinuity between generations, e.g. a family composed of grandparents and grandchildren), family composed of unmarried siblings, family composed of unmarried but cohabiting partners, and other families.

　　Our results show that both the patterns of changes in family structure and changes in the relations between nuclear families and their kinship networks differ from the respective patterns of change accompanying modernization in Western families. First of all, East Asian society has not seen an increasing prevalence of nuclear families with greater levels of modernization. While Japan, South Korea, Chinese Taiwan and the Chinese Mainland follow a descending sequence in terms of their levels of modernization, the number and proportion of the nuclear families (regardless of whether or not one-couple families or single-parent families are included) do not follow the same order accordingly. Likewise, the proportion of stem families does not follow the reverse order. When examined across different age groups, there are no universal generalizations about the change of family structure. Rather, different trends are apparent within countries and regions: in every age group South Korean family structures take on more “modern” features, while in Chinese Taiwan they are more “traditional”, and in the Chinese Mainland there tends to be a mixture of the modern and traditional. The family structure in Japan, however, depends heavily on the age structure. As such, our results indicate that there is little correlation between changes in family structure and level of modernization in East Asia.

　　Second, earlier theories of Western modern families not only refer to a nuclear family as one composed by one married couple and their unmarried children, but also define the division of household labor between husband and wife (the couple are in an equal relationship with the husband working outside the home while the wife taking care of domestic responsibilities) as well as their relationship with their relatives (close or not close). That is, the concept of a nuclear family covers both composition of the family structure and intra-family dynamics and inter-family relations, which are inseparable. The proportion of such typical nuclear families, however, is very low in East Asia; South Korea, with 24.4% conforming to such a model, is the country with the most, while the Chinese Mainland has the lowest, with 5%. In most of the nuclear families in East Asia, both spouses work outside the home or maintain close relations with their extended families.

　　Third, we observed clear divergence in family structure between all four locales. With the exception of family structures in Japan, trends in family structure within particular age groups have no apparent correlation to the overall age structure of the population. The Chinese Mainland has the highest proportion of one-couple families to the total number of families, but the lowest proportion of single families and nuclear families in almost all age groups. South Korea, in contrast, has the highest proportion of single families and the lowest proportion of stem families in almost all the age groups. The proportions of one-couple families in Chinese Taiwan are the lowest in all the age groups. The highest proportion of stem families is found in the Chinese Mainland and Chinese Taiwan. When family structure is further subdivided, Japan has the highest proportion of empty-nest-couple families (nearly 80% of one-couple families), although not the highest proportion of one-couple families. This is likely because it was the first to enter a greying society among the four. Owing to the irregularity in the family structure in the four locales, it is difficult to make any prediction about the dynamic trend for the future of the family structure in the Chinese Mainland. However, it is comparatively certain that the proportion of empty-nest-couple families will increase, which has attracted attention from quite a few scholars. The family structure in Chinese Taiwan is even more traditional than that of the Chinese Mainland, with the highest proportion of extended families (which are similar to traditional family structures). This indicates that traditional Chinese culture and the concept of family have been better preserved in Taiwan.

　　Last, nuclear families (including one-couple families and single-parent families) and kinship networks in the four locales have shown distinctly different patterns. The level of interaction between spouses in nuclear families with their parents is highest in Chinese Taiwan and lowest in South Korea. The two, however, both have a fair amount of inter-generational support and dependence. In terms of the closeness between the nuclear families and their kinship networks, the Chinese Mainland is second only to Taiwan, but its inter-generational support and dependence ranks second to last, above Japan. Compared with the other three, ties of nuclear families and their kinship networks in Japan are more distant, possibly because of its better social security system and more abundant public goods.

　　Ma Chunhua is from the Institute of Sociology at the Chinese Academy of Social Sciences.

　　The Chinese version appeared in Chinese Social Science Digest, No.111, Mar, 2013.

　　Translated by Jiang Hong

　　Editor: Du Mei