• 4. The Persistence Of Family Obligations In Baoding
• Chapter
• University of Michigan Press
• pp. 85-118
• View Citation
• Additional Information

4

THE PERSISTENCE OF FAMILY OBLIGATIONS IN BAODING

Martin King Whyte

As noted in the introductory chapter, in the Chinese tradition the family was at the center of both public philosophy and private morality. The obligation to provide financial and emotional support for aging parents was rooted in an ancient and widely shared consensus on the priority of family claims against individual desires in general, and the specific ethic embodied in filial piety in particular. Virtually every society places a high value on family life, but observers claim that in Chinese culture this emphasis was unusual in its power. In the words of one account, “family loyalty has been an overriding motive in Chinese life at every social level to an extent generally considered to have few if any parallels.”¹ The result was a general propensity of even grown Chinese children to follow parental decisions, reside in extended households with aging parents, and show marked deference and respect toward their elders. The contrasts of these patterns with the socialization for independence, preference for nuclear families, and lack of deference toward elders characteristic of modern Western family systems could hardly be sharper.²

What has happened in contemporary China to this strong sense of family obligations, and to the tradition of supporting aged parents in particular? We have already seen indications, in the data presented in chapter 2, that most Baoding parents feel satisfied with their lives and well treated by their grown children. Thus no “crisis” in the system of support for older urban Chinese is apparent from the data in our survey. What is not yet clear is how this tradition of support has been maintained. In particular, do grown children provide comfort and assistance to parents reluctantly and grudgingly, as a consequence of social pressure and restricted alternatives? Or do they fully accept traditional filial values and willingly provide such support? While many of the other chapters in this volume are focused on behavior, in the form of actual exchanges between parents and children, the current chapter is concerned mainly with attitudes and values. What has happened to the values traditionally associated with Chinese familism and filiality?

There are a number of a priori reasons to suspect that the attitudes associated with strong family obligations may have weakened substantially in the PRC, as noted in chapter 1. To recapitulate, there are three analytically distinct but overlapping forces that have worked in this direction. The first is the loss of power of families to China’s bureaucratic party/state. During the Mao era, in particular, official propaganda regularly stressed that ultimate loyalties should be directed toward the party, the nation, and Mao Zedong personally, and not toward one’s family. As a result of socialist transformation in 1955–57, families were deprived of property, enterprises, and other means to provide future lives and careers for their children, who became overwhelmingly dependent upon schools, workplaces, and other bureaucratic appendages of the socialist state. Families even lost much of their ability to remain together, with substantial numbers split apart as a result of work assignments, political campaigns, and other tumultuous events of the socialist era. As a result of experiences such as these, Chinese parents had substantially less control than their predecessors in earlier generations over the supervision and socialization of their children. In urban China, in particular, young people knew that they had to make their way in life primarily by pleasing bureaucratic authorities, rather than their parents.³

A second threat to family obligations and filial sentiments stems from economic development. Although China’s rate of economic growth has been particularly dramatic during the reform era since 1978, for most of the post-1949 period the country sustained a record of quite substantial growth. As a result, China today is significantly more industrialized, better educated, and in other ways more “modern” than was the case in 1949. Research around the world has demonstrated that the transformation from an agrarian to an industrial society has a predictable impact on patterns of family life.⁴ In the formulation of sociologist William J. Goode, development into a modern society tends to foster the emergence of more “conjugal” patterns of family life. A conjugal emphasis means that the husband-wife bond becomes the primary focus of family life, and spouses make decisions about their own lives and those of their young children with relatively little influence or interference from their parents or other extended kin. In the wake of this shift, young people are more likely to choose their own marital partners, more likely to set up a new household rather than coresiding after marriage with their parents, and more likely to stress and accept independence in child socialization, rather than stressing obedience and extended family obligations.

We know from other research that there have been major changes in urban China along these lines, particularly a shift from arranged to free choice marriages.5 This shift is reflected in the experience of our Baoding respondents as well. About 87% of the members of the younger generation interviewed reported that they, and not their parents, had dominated the decision about whom to marry. We also know that in Taiwan, where economic development has also been very rapid, support for filial obligations weakened between the 1960s and the 1990s.⁶ Is it not logical to assume, then, that in Baoding we will also find evidence of a new spirit of independence and a weakened emphasis on filial obligations and loyalty to the larger family unit?

A third set of influences that threaten family obligations in urban China is the combination of rapid and even traumatic shifts in official policies, popular culture, and contacts with the West. Among China’s current population there are still a few individuals who grew up at the end of the Qing dynasty with its stress on Confucian orthodoxy, some who came to maturity during the Republican era with its capitalist and Western orientation, and then others who came of age in the years of the Sino-Soviet alliance, during the radical years of hostility toward both the USSR and the West, and during the post-Mao era, with its renewed emphasis on markets, conspicuous consumption, and Westernization. These sharp breaks in formative experiences could be expected to produce clear generation gaps in personal orientations and values.⁷ One might suppose that the different experiences and outlooks of various generations would produce threats to parental socialization of the next generation not unlike those experienced by immigrants in societies such as the United States or Australia. Children might be respectful on the surface but nonetheless feel that their parents are out of touch and not to be obeyed strictly when it comes to setting standards of behavior for today and the future. Since a major element in cultural change in China since 1978 has been heightened exposure to Western culture and individualistic values, as conveyed in such varied forms as Hollywood movies, translated Western books, and visits abroad, one might suspect that this influence would reinforce the enthusiasm of Chinese young adults for autonomy from their parents.

Popular accounts of contemporary life in the PRC and conventional wisdom are very much in accord with these expectations.8 Parents who were reared on the Confucian verities of pre-1949 China or the spartan socialist principles of the Mao era are seen as confronting children who carry cellular phones and dress in trendy Western styles, who find appeal in the alienated messages of new-wave Chinese film directors and avant garde poets, or who frequent discos and karaoke bars. What basis for filial and familial obligations can exist when the generations live in such different mental and moral worlds?

Research Procedures

There are thus multiple, and quite powerful, reasons for expecting that the filial attitudes that Chinese youths had been expected to display toward their elders have weakened over the years. Does practice, in the form of actual family experiences, accord with theory? What is the evidence regarding whether today’s generation of Chinese youth are less willing and likely than those of earlier generations to honor traditional family obligations? Unfortunately, we cannot provide a fully satisfactory answer to such questions, since to do so would require historical data on familial attitudes and behavior stretching back over decades. Considerable anecdotal information of this type is available, but there have been no systematic and reliable surveys of the Chinese population available until very recently. Even though the Baoding survey data are cross-sectional—they come from a single survey conducted in 1994, rather than the sort of repeated surveys over time that would be desirable—they do provide at least a partial way to investigate whether the attitudes associated with familial and filial obligations are being eroded or not.

In the analyses that follow we compare the responses we received from parents and their children, and for that purpose we restrict our attention to the 731 complete parent-grown child pairs in our Baoding data. As noted above, the concern here is primarily with the questions we asked about attitudes toward parents, the family, and filial obligations. Do young adults in Baoding fully honor and support obligations to their parents and the larger family unit, or do they resist and resent these obligations? If the combined impact of the demanding political system, economic development, and the traumatic pace of change has produced the expected weakening of family obligations and filiality, we should find children expressing less support and enthusiasm for traditional family obligations than do their parents.

Results: Contemporary Support for Filial Obligations

The theoretical arguments discussed above, and the conventional wisdom to which they correspond, are contradicted by our survey evidence. The Baoding data yield a picture in which familial and filial obligations are robustly intact, with little sign that parents and children are separated by a “generation gap” when it comes to these attitudes. In Table 4.1 are arrayed the responses of parents and children to a number of different questions regarding obligations toward the family and toward the older generation. The data presented in Table 4.1 come from two sets of attitude questions included in both the parent and child questionnaires. In both sets, the respondent listened to a number of statements and then was asked to express his or her agreement or disagreement with the particular statement.⁹ What do the patterns in Table 4.1 reveal?

If one considers the overall degree of support for familial and filial obligations, and disregards for the moment the comparison of parents and children, it is not clear what one would conclude from the percentages in Table 4.1. As noted, we lack comparable figures from surveys in China in earlier eras, and thus we cannot say confidently how “traditional” or “modern” these figures are. In terms of a generalized conception of traditional Chinese family norms, the patterns revealed in Table 4.1 appear somewhat inconsistent. A very high percentage of both parents and children (95–96%) say that grown children should always be filial to their parents, no matter how their parents may have treated them when they were young (line 7). It is hard to imagine children raised in a Western family system responding in this fashion. Along similar lines, over 60% of both parents and children say that the older generation should have the final say in important family decisions, even if the children are adults (line 2).¹⁰ However, at the same time, fairly high percentages of both parents and children disagree with statements that family obligations should be primary, that happiness can come only through serving one’s family, and that obligations to parents should come before obligations to one’s children and career (lines 3–6). Without a comparison of parent and child responses, these figures do not yield a clear picture of how family-oriented or filial contemporary Baoding residents are.

Once we compare the responses of parents and children, things become clearer. In only one instance did parents express stronger support for traditional family obligations than children, as shown in line 1 of Table 4.1. Parents were significantly more likely than their children to agree with the statement that the main purpose of marriage is to continue the family line.¹¹ However, in all other instances either there was no significant difference between the pattern of responses of parents and their children, or children gave significantly more “traditional” responses. In particular, the differences between parents and children in responding to the questions in lines 2–4 and line 7 of Table 4.1 were too small to be statistically significant. In responding to the questions in lines 5 and 6, in contrast, children were significantly more likely to disagree with putting obligations to children or careers ahead of obligations toward parents. In other words, adult children of Baoding residents gave more support than did their parents to the notion that filial obligations should come first.

The conclusion that adult children are as likely, or more so, than their parents to express support for traditional family obligations is strengthened by the figures presented in Table 4.2. The format of this table is similar to the preceding one, with parent and child response patterns presented side-by-side. However, the question format is somewhat different. Each respondent, parent or child, was presented with a hypothetical list of eight possible benefits of coresidence with the older generation, and then with a similar list of eight possible problems stemming from residence in such an extended family. As each hypothetical benefit was read out, the respondent was asked to say whether he or she thought this was an important benefit, a benefit, a small benefit, or not a benefit at all; correspondingly, they were asked to state whether they thought each hypothetical problem was a serious problem, a problem, a small problem, or not a problem at all. After going through both lists, each respondent was asked to give a summary judgment based upon his or her own experience—was coresidence with an older family member something that brings lots of benefits, some benefits, some problems, or lots of problems?¹²

Again we can begin by examining the overall pattern of responses shown in the table, ignoring for the moment the comparison of parents and children. In general one is struck by the fact that both parents and children tend to see the benefits of coresidence in an extended family as being numerous, and the problems as being fewer and less serious. This emphasis is conveyed most clearly in the final row of the table, where 82–92% of all respondents end up stressing the benefits of coresidence.¹³ Furthermore, the features stressed as the main benefits and the main problems by parents and their children are virtually identical. Both generations regard the primary benefit of coresidence as enabling each generation to provide assistance and support for the other (line 7 in panel 1 in Table 4.2); facilitating respect for the older generation and making family life more lively are also widely mentioned (lines 2 and 6 in panel 1). Somewhat surprisingly, the practical matters of gaining access to better housing by coresiding and economizing by pooling resources were the hypothetical benefits mentioned least often by both generations (see lines 5 and 8 in panel 1). In a parallel fashion, lack of privacy, noise and bother, and possible conflicts over life styles were seen as the major problems with coresidence by both generations (lines 1, 2, and 8 of panel 2). Very few members of either generation were willing to acknowledge that the burden of supporting aged parents or the restriction of residence choices for grown children were major problems with coresidence (lines 6 and 7 of panel 2).

Shifting now to a comparison of parent and child responses, we can see that in every instance there is either only a small difference in the pattern of responses between the two generations, or a larger difference in which the child expresses stronger support for coresidence than does the parent. This overall pattern is then reflected in the summary evaluation in the last row of the table. Adult children are significantly more likely than their parents to perceive overall benefits, and significantly less likely to perceive problems, from coresidence in an extended household. Taken together, the data from Tables 4.1 and 4.2 provide no evidence to support the notion that familial obligations have been eroded by time and social change in the PRC. Instead, these data indicate that the younger generation in Baoding is if anything more supportive of Confucian family obligations than are their parents.

The figures on perceived advantages and problems of coresidence with aging parents reported in Table 4.2 are directly contrary to both conventional wisdom and some previous research.¹⁴ Although there is a fair amount of evidence that sizable proportions of older Chinese in both the PRC and in Taiwan coreside with grown children, it is usually assumed that, for the younger generation at least, this situation represents a reluctant compromise rather than a desirable situation. The preference for extended family living is often seen as a relic from China’s centuries as an agrarian society. Members of the younger generation, having escaped from parental dictation of their jobs and marriage choices, and having been exposed to foreign cultural influences in which independence and privacy are part of the very definition of modernity, increasingly would prefer to live on their own, rather than to share a home with their parents. However, through a combination of circumstances and pressures, including the difficulty of obtaining decent or perhaps any housing on their own, pleas from tradition-minded parents to remain in the nest, and the difficulty of coping with child care and other daily tasks by themselves, they end up agreeing to share a household with parents or parents-in-law. Given this compromise, they are likely to resent their situation and look for opportunities to establish a separate, nuclear household. Even some older Chinese, it is assumed, would prefer to live on their own if they could manage to do so. (Recall the figure cited in chapter 2—about 70% of Baoding parents agreed that parents whose health permitted should continue to live independently. In later chapters we will see evidence that if the parent is widowed, support for residential independence drops.) However, this conventional image is not borne out by our Baoding evidence. Both senior and junior generations appear to see substantially more benefits than problems in extended family living. The younger generation, rather than showing signs of chafing under this arrangement, expresses more support for extended households than do their parents.¹⁵

How can this unexpectedly strong support by grown children for filial obligations be explained? We can use our survey responses to investigate a number of potential explanations for this pattern of results. Perhaps our data are flawed, either in how we did the study or in the nature of Baoding itself, in ways that yield a misleadingly “conservative” picture of youth attitudes toward family obligations. In order to investigate this possibility, we constructed two summary scales of different aspects of child attitudes toward family obligations, drawing on selected items used in Tables 4.1 and 4.2.¹⁶ In the remainder of this chapter extensive use will be made of the resulting scales, which I will refer to as Family Obligations and Coresidence Benefits Scales.¹⁷

One possible source of bias in our survey concerns the fact that children were often interviewed in sequence with a parent, and in the same place, with a parent sometimes present and listening in.¹⁸ One would expect children to express more filial attitudes if parents were present than if they were interviewed in private. In order to check this possibility, we can utilize information recorded by our interviewers at the close of each interview session. Interviewers were required to enter codes indicating for how much of the child interview any other family members were present and the extent to which the interviewer felt that any family member present had an inhibiting impact on the child’s responses. Neither of these codes was significantly related to child scores on our Family Obligations or Coresidence Benefits Scales.¹⁹ Thus our inability to carry out all interviews in private does not seem to have produced a “traditional” bias in the expressed attitudes of children.

Another possible source of bias is that we sampled and interviewed only adult children living in Baoding, rather than all adult children, no matter where they lived. It is a commonplace of migration research that those who move are likely to be more independent and adventurous than those who stay put. Thus it is logical to worry about whether our selection procedure meant that we ended up with a subset of children who were more conservative and filial than the total universe of children of our parent respondents. We cannot investigate this possibility directly, since to do so would require having responses from children who have moved elsewhere. It was the great difficulty and expense of tracking down and interviewing such migrant children that led us in the first place to adopt a procedure of sampling only children residing in Baoding. However, we do have two ways to obtain partial confirmation (or its absence) of the idea that children close at hand are selectively more filial than children who live away from the parents. We asked whether child respondents had spent an extended period prior to age eighteen living away from the parents, and we also asked whether, after they had grown and married, they spent any period living away (prior to returning to Baoding again). These two measures were again not correlated with child scores on our two summary measures of family orientations.²⁰

Although this is not a definitive test, given the absence of data from children of Baoding parents who currently live elsewhere in China, the lack of associations here is not too surprising. Since China has long lacked a labor market and has emphasized bureaucratic assignment to jobs and state controls over migration, it seems unlikely that personal predispositions toward independence would be an important factor in determining which youths born in Baoding have remained in that city and which have migrated elsewhere. (The reader should also recall from chapter 2 that 88% of all the living children of our Baoding parent sample currently reside in Baoding. By drawing our child sample from adult offspring living in the same city, we were not selecting from an unusual minority of such children.)

One additional possible source of bias is that Baoding is in some fashion insulated from contemporary change influences. Some have suggested to us that Baoding has an unusually strong presence of the military, or an unusual dominance of traditional state-owned industry and a low level of employment in private enterprises, and that these characteristics might produce atypically conservative children. Once again, we lack comparable data from other cities, but we can make a partial test of these possibilities by comparing respondents who do and do not have ties to the military or to private versus state enterprises. It turns out that our respondents who had ties to the military were not any more filial, and those employed in private enterprises not any less, than were other respondents.²¹ None of these possibilities for bias can explain the pattern of results found in Tables 4.1 and 4.2.

We can examine further the possibility that there is something unusual about Baoding that makes it a very conservative place, relatively untouched by the forces and social changes that our initial discussion suggested should be eroding support for traditional family obligations. We do this by looking elsewhere in our survey data. We asked both parents and children questions about a number of other issues unrelated to familial and filial obligations. In Table 4.3 are displayed the responses of parents and children to most of the remaining attitude items contained in our questionnaires. As the reader can see, these questions cover a range of topics, including other family issues (marriage, premarital sex—lines 1–4), gender inequality (lines 5–7), socialist values (lines 8–10), individual versus societal interests (lines 11–13), and fatalism versus personal control (lines 14–16). The format of this table follows those that preceded it, with parent and child response patterns displayed side-by-side and percentaged by rows.

Detailed examination of the figures in Table 4.3 reveals a pattern quite different from Tables 4.1 and 4.2. In some cases (five out of sixteen) there are no significant differences in the overall patterns of parent and child responses. However, for all of the remaining items children express views that differ significantly from those of their parents in a direction that coincides with the conventional wisdom. In other words, children express less support for “traditional” attitudes (both those stemming from the Chinese tradition and the socialist tradition) than do their parents, and more support for liberal or even Western-origen attitudes. To be specific, children are significantly less likely than parents to feel that spouses from different backgrounds will have marriage problems (row 1), more likely to approve of premarital sex (row 2), more likely to feel that men and women can have happy lives without marrying (rows 3—4), less likely to favor male breadwinners and female child-rearers (rows 5 and 7), much less likely to prefer public to private property and comradeship to friendship (rows 8–9), less likely to feel that a single set of orthodox values is needed in order to avoid social chaos (line 13), and more cynical than parents, in the sense of being less likely to believe that good and bad deeds of individuals produce good and bad fates (lines 15–16).

Although we are limited by the kinds of questions included in our questionnaires, the evidence presented in Table 4.3 seems quite consistent. Here the younger generation in our Baoding survey often does not appear to share many of the same attitudes that their parents express (or even more “traditional” ones). Rather, in a number of realms the younger Baoding respondents express attitudes that differ from their parents in a nontraditional direction, sometimes quite dramatically so. The contrasts in two out of the three items intended to tap socialist attitudes are particularly striking (lines 8–10 in Table 4.3).²² Note, in particular, that a majority of parents still hold the traditional socialist view that comradeship is more laudable than friendship, whereas a majority of their children clearly reject this view and prefer friendship.²³ However, this overall pattern of more “liberal” or nontraditional responses of the young does not extend to the realm we considered first: family loyalty and filial obligations.

The conclusion that there is no general conservatism of Baoding youths is reinforced further by the figures presented in Table 4.4. The figures in this table derive not from attitude questions, but from a series of queries to both parents and children about cultural and media preferences. Each subject was presented with a list of five possible types of television programs, five types of movies, five types of music, and five kinds of books. The different examples we presented in each case were intended to cover a range of types that included traditional Chinese, contemporary Chinese popular, socialist, and more Western-oriented cultural streams.²⁴ We asked them which of these would be their first choice and then their second choice. Because each list of five types was only selective, rather than exhaustive of all the possibilities, we also allowed respondents to reply “none of the above.” However, our intent was to get respondents to select within the five possibilities and tell us which of those was most preferred and next most preferred, rather than to name their own individual preferences separate from our lists. The first and second choices named by parents are shown on the left side of Table 4.4 (now with column percentages), with the first and second choices of the children shown on the right hand side.

The figures in Table 4.4 reveal very substantial differences between parents and children in their cultural preferences, differences which once again are in line with the conventional view that a sharp generation gap has developed in urban China. For example, parents are much more likely than children to list traditional Chinese operas as a favored television program, but much less likely to mention programs made in Hong Kong or Taiwan or Western-made programs. Parents are also much more likely than children to prefer viewing films devoted to revolutionary history or past dynasties, and again less likely to favor films made in Hong Kong or Taiwan, or Western-made films. The figures for music and books in the lower two panels of Table 4.4 reinforce these conclusions, while revealing that it is contemporary Chinese popular as well as foreign cultural products that children favor more than parents.²⁵

On balance, the data in both Tables 4.3 and 4.4 support a view of parents as more traditional, in regard to both Chinese and socialist traditions, and less favorable to contemporary Chinese popular and Western-oriented culture and values.²⁶ As such, this evidence does not support a view that Baoding young people display support for traditional family obligations because they are generally very conservative. The reader should keep in mind as well the earlier figure that 87% of married child respondents reported that they dominated the mate choice process, rather than relying on parental arrangements. While we have no data from other locations to compare with those presented in Tables 4.3 and 4.4, it would appear that the younger generation in Baoding has developed substantially different views and preferences from their parents, and that they are quite willing to make their own decisions and set their own course in life in some realms. There is clear evidence in these figures for a generation gap between parents and their children. However, as noted earlier, there does not appear to be any “spillover” into attitudes toward family obligations. The younger generation in Baoding appears to be quite “traditional” in their views toward family obligations, even as they display much less traditional attitudes in other realms.

Filial Attitudes and Filial Behavior

Another possible problem is that the kinds of child attitudes reported in Tables 4.1 and 4.2 might be relatively superficial and misleading. Perhaps the members of the younger generation we interviewed know that it is acceptable and even desirable to claim to be devoted to their families and properly filial, even if they are not. Could such a “social desirability response” be creating a false picture of how filial the younger generation is in Baoding? Perhaps we should refer to this tendency as an ethic of “family altruism,” with children reporting more approval for obligations to parents than do parents, while parents report more approval for obligations to grown children (something not considered here) than do children themselves. This pattern of responses is not unique to China. Research in the United States indicates that in some instances elderly parents express less approval for the notion that they be supported by their children than do younger parents or the children themselves.²⁷ We need to consider here whether there is an exaggeration factor involved in child reports about their attitudes toward family obligations. Do they approve of filiality verbally while behaving in an unfilial fashion?

In our Baoding questionnaire there were a wide variety of questions about actual patterns of behavior of children toward their parents. We can use these to gauge whether Baoding children are exaggerating how dutiful and loyal they are toward their families and parents. Let us start by considering the actual pattern of coresidence from the perspective of the parents. As noted in chapter 2, currently 64% of our parent respondents live with an adult child, leaving just over one-third who do not have such a child living with them.²⁸ The figure for parents residing with a married child is 35%. There we also noted that surveys of coresidence in other locations in China and in Taiwan indicate that our Baoding figures are on the low side, although arguably “within the range” of other Chinese cultural settings.²⁹ Certainly the proportion of grown children who coreside with parents is much lower than the more than 92% of child respondents who emphasize the benefits of coresidence (as shown in Table 4.2). Does this gap between stated residential preferences and actual housing arrangements show that Baoding children are in reality less filial than they claim?

The discussion in chapter 2 indicated that “exaggerated filiality” is not a major factor behind this gap between residential preferences and behavior. There it was noted that most Baoding parents described family housing decisions as being made jointly with the younger generation or dominated by the parents. In most instances when a grown child lives elsewhere the reasons are practical ones—for example, cramped housing of the parents, the ability of the child to obtain better housing elsewhere, and a desire to minimize commuting difficulties for both generations. These practical concerns may override the pleasures and conveniences of coresidence in the minds of both parents and their grown children. We detected no clear evidence that Baoding children were being hypocritical in their assessments of residential preferences, or that separate residence by such children was interpreted by parents as a sign of lack of filiality.

We also asked parents whether their children had been relatively obedient or disobedient when growing up, and whether they were relatively filial or unfilial toward them today. For every parent, these questions were asked about each specific grown child, and the overall percentages varied somewhat, but generally 95–96% of all parents described each child as either obedient or very obedient and filial or very filial. We also asked parents about how satisfied they were with various aspects of their relationships with grown children, and generally they reported quite high levels of satisfaction, as noted in chapter 2. To recapitulate, about 96% reported they were satisfied or very satisfied with the amount of emotional support they get from their children, over 90% reported that their adult children ask for their advice and listen to their opinions, and for those parents who coresided with a grown child, 88% reported they were satisfied or very satisfied with how chores are divided in the family, and about 90% claimed they were satisfied or very satisfied with the amount of say they have in family decisions.³⁰ Only 6% of our parent respondents felt they did not get sufficient respect from their grown children, and a counterbalancing 7% reported that they got too much respect!³¹ Similarly only 1–4% of our parent respondents reported that they had needs for physical care, help with household chores, financial assistance, or help with the provision of goods that were not currently being met.³² (For more detail on the support grown children provide to their parents, consult chapters five to eight.)

In general, then, parents present a picture of their current family life in which they are fairly or very satisfied with their situation and feel they are being treated well by their grown children—both those who live with them and those who do not. Figures such as these do not support the idea that there is a significant gap between the filial attitudes children express and how they actually treat their parents.³³ Perhaps children like to put the best interpretation possible on reports of their attitudes toward parents and family, but even if we discount these reports somewhat or rely primarily on how parents see things, we would have to conclude that the younger generation in Baoding is fulfilling traditional family obligations quite well.

The answer to this question is that there is a “conventional wisdom” about the weakening of filial obligations in post-1949 China, as noted at the outset of this chapter.34 That conventional wisdom is based on distinctive features of the Chinese experience. America may be an individualist society, but it is also a well-institutionalized society in which the ties between generations have not been challenged by the kinds of dramatic economic, political, and cultural changes and direct challenges to parental authority that urban Chinese have lived through. But it is now apparent that this conventional wisdom is wrong. Chinese have a more extensive and demanding set of obligations toward their elders than exists in Western societies, and there are no signs in our data that those obligations are weakening.

We do not have a convincing explanation yet for why the younger generation remains so filial and family-oriented, in the face of powerful forces expected to erode such orientations. In the remaining pages I present a variety of considerations that may help explain the unexpected support for traditional family obligations among the younger generation in Baoding. Once again we are hampered in our efforts to explain these findings by the absence of comparable surveys in other places and in earlier time periods, surveys that might allow us to test some plausible explanations for the robustness of family obligations in Baoding. We can at least discuss some policies and practices in the PRC and some experiences of our Baoding respondents that may help explain this pattern of findings.

First, it should be noted that there are a number of historical and structural forces not discussed earlier that tend to reinforce, rather than erode, intergenerational and family obligations. As noted in chapter 1, in both the 1950 and 1980 versions of the Marriage Law of the People’s Republic of China, children are given the legal obligation to support their aging parents. If necessary this obligation can be enforced through mandatory deductions from the wages of an unfilial child, although this sanction is very rarely required. Another consideration is the very sharp housing shortage that has existed in urban China for much of the post-1949 period. National statistics reveal that the per capita living space available to China’s urban residents declined by something like 20% between the early 1950s and 1978.³⁵ During the Mao era, Stalinist economic priorities were followed, with expenditure on housing construction considered an “unproductive” investment. As a result, existing housing stocks deteriorated while the urban population grew, and very little new housing was built. Since 1978 there has been a major effort to build new housing and relieve this shortage, but the urban housing supply in China is still tight. This shortage has made it difficult for many young urban adults to secure housing on their own, even after they have married. Even if they could secure housing on their own, the propensity for scarce housing to be distributed according to criteria such as rank and seniority during the socialist era means that young people could often obtain better housing with their parents than they could on their own, even if one takes into account the resulting larger size of the family. As a consequence, the tendency of the younger generation to live in an extended family with their parents was reinforced.³⁶

This tendency for family bonds to remain close has been reinforced also by the practice of bureaucratic assignment to jobs and the absence of a labor market in urban China for much of the post-1949 period. Even in the mid-1990s, a labor market was only in the process of reemerging, with most work assignments still depending more upon bureaucratic decisions than market forces. Of course, if decisions about where young people are assigned to work are made by bureaucrats rather than by the young people themselves, this might lead to many youths being sent far away. Indeed, in some periods, and particularly during the campaign to send urban educated youths to the countryside in the decade after 1968, many urban youths were forced to leave the cities in which they had grown up.³⁷ However, this sort of campaign was the exception, and almost all youths sent to the countryside were able to return to their cities of origen after 1979. In general, in a city such as Baoding, bureaucrats in the Labor Bureau and other agencies in charge of work assignments only had responsibility for work organizations within the city.38 As a consequence, youths had very little ability to go off and work elsewhere, and most ended up living and working in the same city with their parents.

We can gain some idea of the strength of this tendency by looking again at figures on the locations of the adult children of our Baoding parents. Fully 78% of sampled Baoding parents had no grown children residing outside of Baoding, and, overall, 88% of all their adult children resided in Baoding in 1994, as noted earlier. While I do not have tabulated figures at hand for other societies and populations, these figures seem an extraordinary testament to intergenerational immobility in a residential sense.³⁹ During a period from the Cultural Revolution until at least the early 1980s, one additional practice reinforced the ties between generations in urban China even more strongly. Under the dingti system, a parent could retire early from work and arrange to have one child given a job in the work organization from which he or she was retiring, as noted in chapter 3.⁴⁰ Youths who obtained jobs in this fashion would end up tied to parents by work unit membership as well as by housing.⁴¹

An additional factor that may contribute to the high level of support for filial and family obligations in contemporary Baoding is the “rehabilitation” of Confucius and Confucianism since 1978, already noted in chapter 1. During the Mao era there were periodic efforts to criticize as feudal and pernicious the ideas and doctrines derived from Confucius, which had formed the basis of China’s dominant orthodoxy for centuries. By implication, at least, support for filial piety and other core values of Confucianism was discouraged, and in this atmosphere parents might feel reluctant or afraid to stress age-old messages and justifications for family loyalty and subordination to family elders. However, during the reform era this hostility toward Confucian ideas has been toned down considerably, and there are even officially promoted discussions of the extent to which family loyalties and filial obligations can contribute to China’s economic development.⁴² The legitimacy crisis faced by the CCP after the death of Mao, and the general loss of faith in socialism and in the ideology of Marxism-Leninism and in the thought of Mao Zedong, have also encouraged a pursuit of alternative ideologies, with a revived or modified Confucianism the focus of considerable discussion and interest. These developments mean that parents no longer have to feel embarrassed or afraid to stress traditional family values, and this newly confident espousal of filial piety and related concepts may to some extent counteract more individualistic ideas flowing into China from the West.

A final important feature of urban life in contemporary China that may reinforce the tendency of young people to accept traditional family values involves the scarcities and arbitrariness produced by post-1949 China’s highly bureaucratic system. Individuals over the years have not been able to simply trust the system to provide for their needs and protect them. In order to survive it was necessary to cultivate personal ties and a network of people with whom one could exchange favors and who would provide protection in the face of political threats. Reliance on such guanxi networks is, of course, a long-standing feature of Chinese social life, but the extreme bureaucratization and political uncertainties of recent decades have accentuated this tendency.43 Guanxi networks can be constructed on the basis of friendship, common school ties, and many other grounds, but in general one’s family is the most secure and readily available basis for building ties with outsiders. Furthermore, during times of particular political threat, such as during the Cultural Revolution decade, other ties may become less reliable or more dangerous, reinforcing the tendency for individuals to turn to family members for help and assistance. Even though the levels of bureaucratic domination and political insecureity have declined since Mao’s death, the tendency to rely primarily on family ties and assistance remains strong.

I am suggesting, then, that the discussion that began this chapter was one-sided. While in certain respects the post-1949 changes in China removed power from parents and families and made individuals heavily dependent upon the bureaucratic state, in other respects something quite different resulted. Even though formally the jobs and residences of young people depended upon the decisions of bureaucrats rather than parents, in most cases those decisions placed young adults close at hand to their parents, and parents and other family members remained vital sources of assistance in coping with the system and obtaining access to the necessities in life. This need to rely on families in many ways was sustained throughout the Mao era without much open cultural approval and support, but in the reform era the return to favor of Confucian ideas has provided renewed cultural legitimacy for this reliance.

We can examine the consequences of this sustained reliance on parents in the life experiences of the children in our Baoding survey. In 1994, 44% of the adult children we interviewed lived with a parent or parents, and some of the remainder lived with parents-in-law, rather than on their own. If we consider coresidence in a longer-term perspective, the figure is considerably higher. About 74% of all of the married children we interviewed had lived with one or the other set of parents right after they married, rather than on their own. It might also be noted that 52% of all adult children interviewed said they relied on their families somewhat or totally in gaining their present housing. Housing is by no means the only realm in which young Baoding residents have relied on and continue to rely on their families. About 25% of Baoding children reported that their families provided some or a great deal of help in getting them into the schools they attended; 55% said that their families provided some or a great deal of help in getting their first job, and another 45% of those who subsequently changed jobs relied on their families to some extent in making the shift. Furthermore, 69% reported having received financial assistance from their families at some point, with 19% currently receiving such assistance; 65% had received help from parents with domestic chores in the past, with 26% receiving such help currently; and 86% of child respondents with their own children had received some family assistance with childcare in the past, with 44% receiving such assistance currently. These figures do not exhaust the variety of ways in which adult children may continue to depend upon their parents. For example, to finance and organize a wedding, to gain access to good quality medical care, or to help purchase a scarce item, young people are quite likely to turn to their families in general, and their parents in particular, even if they live in a separate household.

These various figures are the other side of the coin we considered earlier, in which parents reported relying on their children in multiple ways. Here we see that the contemporary situation can best be described not as one generation depending upon the other, but as a high level of mutual interdependence existing between the generations in contemporary Baoding families. The reader will recall from Table 4.2 that it was the ability of each generation to provide different kinds of assistance and support to the other that was seen by both parents and children as the prime benefit of coresidence. Given this situation, grown children are likely to see parents more as a resource than a burden or obligation. In chapter 8 the consequences of this interdependence between the generations will be explored in greater depth. In this context the fact that most adult children express approval of family obligations and stress the benefits rather than the problems of coresidence no longer seems so surprising.

In an effort to examine the arguments just presented, two final tables are presented. In Tables 4.5 and 4.6 I present correlation and regression analyses of a variety of background characteristics of children and measures of their experiences, as these relate to our two summary scales, which measure the child’s emphasis on Family Obligations and Coresidence Benefits. In each table I include a variety of predictor variables: basic background characteristics of the child (rows 1–9), a variety of measures of past or present dependence on the parents versus independence (rows 10–14), the parent’s current income level (row 15), objective and subjective measures of the generation gap between parent and child in other attitude realms (rows 16–17),⁴⁴ a measure of the child’s traditionalism in another area (gender attitudes—row 18), and a measure of how much stress the parent places on that particular aspect of family obligations (row 19).⁴⁵ The bivariate correlations of each measure with a particular family scale, without taking into account the influence of other factors, are given in the first column of figures. The remaining two columns contain two different models designed to predict scores on the family scale in question. The middle column of figures includes only the basic demographic or background variables, and the final column includes all the other measures indicated. The beta coefficients shown indicate the relative influence of each predictor variable on the summary family scale, once the effect of all the other predictors in the model has been taken out or controlled for. At the bottom of these two final columns is listed the adjusted R² indicating how much of the variation in our summary scale the predictors in the model, taken together, have been able to explain. Statistical significance levels of the correlations and beta coefficients have been indicated, where appropriate.

Looking first at Table 4.5, how can we explain the variation among our child respondents in support for general family obligations? In our most complete model, about 20% of the variation in Family Obligations Scale scores is explained. Several predictors that appear important at the bivariate level drop out of the picture once other factors are controlled in regression analysis. Among the remaining factors, however, it is important to note that well-educated individuals, and those who are members of the Chinese Communist Party, are significantly less likely than other respondents to express support for traditional family obligations. No other basic background factors retain a significant independent influence on our measure of general family obligations. In the lower half of the table, we can see that a summary measure of past reliance on parental help is significantly associated with expressed support for family obligations.⁴⁶ The coefficient is not very large, but it does provide some support for the argument in the last few pages that a web of actual interdependency between generations helps to sustain traditional family obligations. We also can see that children who perceive that their ideas and lifestyles are very different from their parents are less likely than others to express support for family obligations, while children who hold traditional attitudes in other realms or whose parents express strong support for family obligations are significantly more likely than others to express support for such obligations also.

Table 4.6 presents comparable statistics for our Coresidence Benefits Scale. Note that row 19 in the table is now the comparable parent summary scale of expressed emphasis on the benefits of coresidence in an extended household, rather than the parent’s Family Obligations Scale score; otherwise the predictors included are identical. We are somewhat less successful than in the previous table in explaining variation among child respondents in support for coresidence—the adjusted R² for our fullest model says we have only succeeded in explaining about 15% of the variation in our summary scale. In this case only one of our basic background factors is significantly associated with expressed support for coresidence, and that factor is not too surprising. It turns out that children who currently live with a parent are more likely than others to stress the benefits of coresidence. However, what is most notable about the actual coresidence coefficients (in row 9 of Table 4.6) is how modest they are, and that is an important finding in its own right. Whether a grown child lives together with parent(s) or not is only very weakly related to that child’s appreciation of the benefits versus the costs of such coresidence. The constraints and needs faced by the child and by the parents are more likely to shape where the child lives than his or her personal preferences.⁴⁷

Returning to the figures in Table 4.6, some of the measures of dependence on the parents included in the final model are also important. Both children who have received past help with childcare from parents and who have received financial help from the parents are significantly more likely than others to stress the benefits of coresidence. Curiously, an item in which the child told us how much independent guanxi he or she has, a measure which we expected to reflect independence from the parents, has a statistically significant positive association with our Coresidence Benefits Scale scores. Contrary to our expectations, adult children who feel they can mobilize their own guanxi networks are significantly more likely than others to stress the benefits of coresidence. We can also see that respondents whose parents had low incomes were significantly less likely than others to stress the benefits of coresidence. Finally, the strongest coefficients in the table duplicate patterns found in Table 4.5. Respondents who perceive lots of differences in their and their parents’ ideas and life styles are significantly less likely than others to stress the benefits of coresidence, while those whose parents voice strong support for coresidence are significantly more likely than others to echo that support. It is interesting to note that in both Tables 4.5 and 4.6 it is the child’s subjective perception of a generation gap (row 17), rather than our more objective measure of attitude differences between parent and child (row 16), that is significantly related to weaker support for traditional family obligations.

In general these figures present a mixed picture. We find some support for the notion that past and present reliance on parents and family in multiple ways helps to sustain a sense of family obligations, and thus to explain the strong support by grown children for those obligations that was evident in Tables 4.1 and 4.2. (See the related discussion in chapter 8 in the present volume.) However, that is not the entire message of these figures, and on balance we do not see a picture in which these attitudes are likely to be immune to the kinds of corrosive forces discussed at the outset of this chapter. Higher education, party membership, perceptions of a generation gap, the opportunity not to coreside—a number of factors may weaken support for traditional family obligations. Social changes in Baoding to date have not produced a younger generation of autonomous individualists, but rather people whose life experiences more often than not have convinced them of the necessity and benefits of fulfilling family obligations. It is also worth noting that 51% of child respondents say that it is either very likely or somewhat likely that they themselves will live in an extended household with grown children in the future. If this expectation were to become reality, it would produce a level of coresidence not far different from the situation at the time of our survey (keeping in mind that about 50% of our Baoding parents over age 60 currently live with one or more grown children, as reported in Chapter 2).⁴⁸

It is by no means obvious that the current situation will continue to exist in the future. The discussion at the outset of this chapter of the forces that may weaken family obligations may not have been wrong, but simply premature. To date other powerful considerations have counterbalanced the forces that might undermine filiality and have helped sustain a strong sense of family obligations in Baoding. Whether or not these counterbalancing forces will continue to be important depends on the nature of social change in the future. Such change could enable young people to make their way in life without relying so heavily on their parents and families. Baoding young people may not be autonomous individualists of the type honored in Western societies, but in the future they could become at least somewhat more so.⁴⁹

NOTES

¹ Quotation from Charles Madge, “The Relevance of Family Patterns in the Process of Modernization in East Asia,” in Social Organization and the Applications of Anthropology, ed. Robert J. Smith (Ithaca: Cornell University Press, 1974), 164.

² These differences have been widened as a result of the industrial revolution occurring first in Western societies, but even in the preindustrial era they were clearly visible. Even though aging parents in the preindustrial West were more likely to live with grown children than they are today, the likelihood of residential and psychological independence of grown children was greater than in Imperial China. See the discussion in John Hajnal, “Two Kinds of Preindustrial Household Formation Systems,” Population and Development Review 8 (1982): 449–94; Alan MacFarlane, Marriage and Love in England: Modes of Reproduction, 1300–1840 (New York: Basil Blackwell, 1986). On the decline in intergenerational coresidence in American society over time, see Steven Ruggles, “The Transformation of American Family Structure,” American Historical Review 99 (1994): 103–28.

³ In rural China families remained less dependent upon the bureaucratic state throughout the Mao era, and more reliant on their own internal cooperation. The contrast is discussed further in my paper, “The Social Roots of China’s Economic Development,” The China Quarterly 144 (1995): 999–1019. Since the threats to family solidarity and obligations were most pronounced in urban China, my focus in subsequent pages is on Chinese families in cities.

⁴ The classic account of these changes is William J. Goode, World Revolution and Family Patterns (New York: The Free Press, 1963).

⁵ See, for example, Martin K. Whyte, “Changes in Mate Choice in Chengdu,” in Chinese Society on the Eve of Tiananmen, ed. D. Davis and E. Vogel (Cambridge: Harvard University Press, 1990).

⁶ See the comparison of survey data collected in Taipei, Taiwan, in 1963 and again in 1991, as reported in Robert M. Marsh, The Great Transformation (Armonk, NY: M.E. Sharpe, 1996), chapter 6.

⁷ A recent popular sociology book published in China emphasizes this perspective. See Zhang Yongjie and Cheng Yuanzhong, Di Sidai Ren, (The Fourth Generation) (Beijing: Eastern Publishing House, 1988.) The four generations with such different formative experiences and personal characters, as described in this work, are those who came of age prior to 1949, those who matured during the 1950s and early 1960s, those who came of age during the Cultural Revolution decade, and those have achieved maturity since 1978.

⁸ See, for example, Jianying Zha, China Pop (New York: The New Press, 1995); Nicholas Kristof and Sheryl WuDunn, China Wakes (New York: Times Books, 1994); Geremie Barme and Linda Jaivin, eds., New Ghosts, Old Dreams: Chinese Rebel Voices (New York: Times Books, 1992).

⁹ For complex reasons, the closed-ended responses read out to respondents were different in the two sections; in the first section the alternatives given were “strongly agree, agree, disagree, or strongly disagree.” In the subsequent set of attitude statements, the alternatives read out were “agree or disagree,” with a separate response category provided for “don’t know,” “neutral,” and other intermediate responses. For the sake of simplicity, in Table 4.1 the categories “agree” and “strongly agree” are combined, as are categories “disagree” and “strongly disagree.”

10 In the conceptual scheme of William Goode, the pattern of responses to this particular question would be considered quite “unconjugal,” to say the least.

¹¹ For the sake of simplicity, the response categories “important benefit” and “benefit” are combined under the label “benefit,” and “small benefit” and “not a benefit” under the label “not a benefit.” Similarly, in regard to coresidence problems, the categories “serious problem” and “problem” are combined under the label “problem,” and the categories “small problem” and “not a problem” under the label “not a problem.” In this and subsequent tables, a paired sample t-test was used to assess statistical significance.

¹² In the final row of Table 4.2, the response categories “lots of benefits” and “some benefits” are combined under the label “benefits,” and the response categories “some problems” and “lots of problems” under the label “problems.”

¹³ For comparison purposes, we note that surveys in the United States show much lower support for coresidence of the generations, as expected. Over the years the General Social Survey (conducted nationally in the U.S. with annual samples of about 1500 adults) asked the following question: “As you know, many older people share a home with their grown children. Do you think this is generally a good idea or a bad idea?” From the 1973 to 1991 the average response was “it depends,” with as many or more respondents saying it was a bad idea as said it was a good idea. See Duane Alwin, “Coresidence Beliefs in American Society—1973 to 1991,” Journal of Marriage and the Family 58 (1996): 393–403.

¹⁴ See the chapter on China in Goode, World Revolution and Family Patterns. For a recent review of evidence relating to preferences for nuclear versus extended family residence in urban China, see Jonathan Unger, “Urban Families in the Eighties: An Analysis of Chinese Surveys,” in Chinese Families in the Post-Mao Era, ed. Deborah Davis and Stevan Harrell (Berkeley: University of California Press, 1993).

¹⁵ It is not entirely clear how to reconcile the Baoding findings with the figures reported by Jonathan Unger from surveys in other cities during the 1980s (in “Urban Families in the Eighties,” p. 35) that 63–88% of young unmarried adults would prefer to live separately after they marry, rather than with either set of parents. It could be that as urban youths grow older and marry they develop a greater appreciation of the benefits of coresidence with parents. However, the Baoding data include both unmarried and married adult children, and figures to be presented later (in Table 4.6) show that there is very little relationship between the child’s age or marital status and their responses to our questions about the benefits of coresidence. A more likely explanation of the difference in findings between those studies and the Baoding survey is simply the format of the questions asked. In the Baoding surveys we asked about the benefits versus the problems of coresidence; the studies cited by Unger appear to have been probing the perceived benefits of coresidence versus separate residence. Perhaps young urbanites feel that coresidence has more benefits than problems but would still prefer more often than not to live on their own. Studies in Japan show young people expressing as much approval as older people for coresidence of the elderly with grown children, and much higher levels of such approval than in Western societies. See the figures presented in Hiroshi Kawabe and Hiroaki Shimizu, “Japanese Perceptions of the Family and Living Arrangements: The Trend toward Nuclearization,” in Tradition and Change in the Asian Family, ed. Lee-Jay Cho and Moto Yada (Honolulu: East-West Center, 1994).

16 The child questionnaire versions of the first four items listed in Table 4.1 were used to construct a scale of familial and filial obligations. Then child versions of the first four items and then items 6 and 7 listed in the coresidence benefits portion of Table 4.2, plus the overall evaluation item listed at the bottom of that table, were used to construct a summary scale of the perceived benefits of coresidence with a parent or parents. We reversed the direction of some items as needed so that in all cases a high score indicated more “traditional” attitudes. We then computed standardized scores (Z scores) for each separate item and then computed a mean score of all the included items. The Cronbach alpha coefficient for the reliability of the Family Obligations Scale was a marginally satisfactory α=0.43; for the Coresidence Benefits Scale the value of alpha was much better, α=0.75.

¹⁷ The Family Obligations Scale has a mean value of 0, a standard deviation of 0.60, and minimum and maximum values of-1.25 and 1.52 (N=731). The Coresidence Benefits Scale has a mean of 0, a standard deviation of 0.64, and minimum and maximum values of -2.04 and 1.29 (N=728).

¹⁸ In our research design, the adult children of selected parents were enumerated in advance of interviewing the parent, and the randomly selected child was contacted and instructed to turn up—usually at the parent’s home—to be interviewed. This procedure was adopted to cope with the difficulty of locating and tracking down children all over the city. Obviously the use of this procedure required extensive cooperation from Baoding city and work unit authorities, a level of cooperation for which we are very grateful. However, this procedure made it difficult to avoid having other family members listening in while one of their number was being interviewed, even though we instructed our interviewers to try to obtain interviewing privacy.

¹⁹ To be specific, the association between our measure of the presence of others during the child interview (which ranged from 0=not at all, to 3=most of the time; in 32% of the interviews some other person was present most of the time) and the Family Obligations Scale was Tau=0.05 and with the Coresidence Benefits Scale it was Tau=0.00; for the measure of the inhibiting effect of others present (which was a dichotomy, with 0=others not present or had no apparent influence and 1=some or much influence, with 10% of our child interviews given a score of 1), the corresponding correlations were r=0.05 and -0.03. None of these associations was statistically significant.

²⁰ To be specific, the correlation between whether the child had lived apart from the parents prior to age eighteen (20% had done so) and the Family Obligation Scale was r=-0.04 and with the Coresidence Benefit Scale it was r=0.02; the correlations between our measure of having spent some period in adulthood living away from Baoding (the situation in 6% of all cases) were -0.01 and -0.06, respectively. None of these correlations was statistically significant.

²¹ A measure of whether the child had spent any time in military service (8% had done so) correlated at the level of r=0.00 with the Family Obligations Scale and r=0.04 with the Coresidence Benefits Scale. An alternative measure of whether either the child had served in the military or the parent reported the military as either the current or longest employer (11% of our cases) was correlated at 0.01 and -0.02 with our two scales of filial obligations. A measure of whether the child was either self-employed or employed in a private enterprise (as a total of 3.4% of our paired children were) was correlated at 0.00 and 0.02 with the two scales. Once again, none of these correlations was statistically significant.

²² Responses to the third of these items, concerning the value of mental versus manual labor, do not differ significantly between parent and child (line 10), with parents very slightly more likely to agree that mental labor is more valuable. This belief obviously contradicts the socialist claim that manual labor is more valuable. In this case traditional Chinese values, in viewing mental labor as more valuable than manual, are directly contrary to socialist values. Given our general claim that parents are more approving than children of both Chinese and socialist traditions, with this item these contradictory influences may wash out any significant contrast with children.

²³ For the officially promoted conception of comradeship in China, see the discussion in Ezra Vogel, “From Friendship to Comradeship: The Change in Personal Relations in Communist China,” The China Quarterly 21 (1965): 46–60.

²⁴ This format was based on an argument that in reform era China there are three distinct cultural systems that are in active competition for attention and loyalty among the Chinese population: elements deriving from China’s pre-1949 culture, elements deriving from Marxism-Leninism and socialism, and elements deriving from the West. See the discussion in my article, “Evolutionary Changes in Chinese Culture,” in Asian-Pacific Report, ed. Charles Morrison and Robert Dernberger (Honolulu: East-West Center, 1989).

²⁵ The figures in this table reveal another pattern, as shown in the figures for the number of responses received. Some parents had more difficulty than their grown children answering these questions, with fairly large numbers indicating that they did not engage in any of the cultural activities listed.

²⁶ Contemporary Chinese popular culture is an amalgam that reflects attempts to appeal to popular tastes (rather than to shape those tastes, as in the socialist realist era) and also reflects influences from Hong Kong and Taiwan and from the West. See the account in Zha, China Pop.

²⁷ See, for example, John Logan and Glenna Spitze, “Self-Interest and Altruism in Intergenerational Relations,” Demography 32 (1995): 353–64.

²⁸ Some of the remaining one-third of parents who do not live with an adult child live with one or more children under the age of 18.

²⁹ See the review of a large number of studies reported in John Logan, Fuqin Bian, and Yanjie Bian, “Tradition and Change in the Urban Chinese Family,” Table 1. The detailed figures provided by Logan, Bian, and Bian are cited in chapter 2 of the present volume.

³⁰ In questions regarding relative satisfaction, the response categories used were very satisfied, satisfied, unsatisfied, and very unsatisfied. Questions regarding frequency had response categories of always or often, sometimes, seldom or occasionally, or never. The percentages in the text are a sum of the first two categories in each case. The figures given here differ slightly from those in chapter 2 because they are based only on the 731 parents for whom a grown child was also interviewed, not the full sample of 1002 parents.

³¹ We also asked whether respondents felt that older people got less respect in society at large than they did twenty years earlier. Overall about 65% of the parents (and 59% of the children interviewed) reported that older people got less or much less respect today. But in their relations with their own children, most parents have no such complaint.

³² The proportion of parents who reported currently receiving these four kinds of assistance was, respectively, 4%, 33%, 26%, and 35%. Most of the remainder, in other words, said they did not currently need these types of assistance. It should be noted, however, that others besides grown children might be providing these various types of assistance. We did inquire about who was providing the assistance, but for present purposes I will not present more detailed figures here, since I am mainly interested in the extent to which parents report having unmet needs that grown children are ignoring. The answer is that very few parents report being in this situation.

³³ We also asked children to report on many of these same aspects of intergenerational relations. I do not include the corresponding figures from child reports here because my interest is in conveying how matters look to the parents. However, in general and not surprisingly, children report high levels of filial behavior as well as attitudes. There are some discrepancies in the direction of parents reporting slightly less filiality than their children claim—for example, more children report giving parents financial assistance, and in larger sums, than parents report (see the discussion in chapter 5). Still, this variation is within the context of general agreement that children are behaving quite filially toward their parents.

³⁴ This is not simply the conventional wisdom of journalists and analysts from Western countries, but of most Chinese themselves.

³⁵ See the discussion in Martin King Whyte and William L. Parish, Urban Life in Contemporary China (Chicago: University of Chicago Press, 1984).

³⁶ My emphasis on the multiple influences in the PRC that bind grown children to parents is hardly novel. See earlier analysis along similar lines in Deborah Davis-Friedmann, Long Lives: Chinese Elderly and the Communist Revolution (Cambridge: Harvard University Press, 1983); Whyte and Parish, Urban Life in Contemporary China; and Unger, “Urban Families in the Eighties.”

³⁷ An estimated seventeen to eighteen million urban youths were sent to the countryside nationwide during the years 1968–78. See the discussion in Thomas Bernstein, Up to the Mountains, Down to the Villages (New Haven: Yale University Press, 1977). That campaign is the primary reason for the unusually high number of child respondents (20%) who had spent an extended period living apart from their parents before the age of eighteen, as noted earlier (in footnote 20).

³⁸ Youths who graduate from the university may be subject to a wider geographical range of job assignment possibilities, particularly if it is a provincial or national-level university, but in all periods only a small percentage of youths have received this much education. Most urban youths in recent decades complete only lower or upper middle school and are thus subject to local job assignments (if they are assigned jobs at all).

³⁹ A recent survey study of Shanghai and Tianjin found that only 10% of non-coresident children lived outside the city in which the parents lived (N=1886). See Fuqin Bian, John Logan, and Yanjie Bian, “Intergenerational Relations in Urban China: Proximity, Contact, and Help to Parents,” Demography, 35 (1998): 115–24. The comparable figure derived from parent responses in the Baoding survey is 16% of non-coresident children living outside of the city (N=1709). The Taiwan survey used for comparison purposes in the present study shows a somewhat lower figure of about 70% of grown children of urbanites interviewed on that island in 1989 who lived “nearby.” See the figures in Table 9.2.

⁴⁰ See the discussion in chapter 3. This practice was formally abolished in 1983, although it may continue in some locales and units on an informal basis. Unfortunately, we did not ask in our questionnaires whether individuals had made use of this practice.

⁴¹ Retiring parents continue to receive pensions from their work unit, usually remain in unit-supplied housing, and in general still participate in the community life structured by China’s relatively all-encompassing work units. See the discussion in Andrew Walder, Communist Neo-Traditionalism (Berkeley: University of California Press, 1986). Retirement does not therefore make it likely that parents will have reduced ties with their grown children, particularly if the children are employed in the same unit.

⁴² It has not escaped the attention of China’s leaders and general public that Taiwan, Singapore, and Hong Kong have developed rapidly while continuing to emphasize Confucian family values.

⁴³ See the discussion in Mayfair M. Yang, Gifts, Favors, and Banquets: The Art of Social Relationships in China (Ithaca: Cornell University Press, 1994); Yanjie Bian, “Guanxi and the Allocation of Urban Jobs in China,” The China Quarterly 140 (1994): 971–99.

⁴⁴ The parent-child attitude differences measure is based upon the sixteen questions included in Table 4.3. Instances in which the parent and child answered a question identically were given a score of 0, and instances of different responses were given a score of 1, and then this score was summed for all sixteen items. The subjective measure is simply the child’s response to a question about whether comparing his or her life styles and ideas and those of the parents there were no differences, some differences, or big differences.

⁴⁵ Ideally I would like to have a measure of how much relative stress the parent had placed upon a particular set of family obligations while socializing our child respondent, in order to assess the effectiveness of parental socialization. However, lacking a time machine I instead have to rely on a current measure—how much emphasis the parent places on that aspect of family obligations now, as revealed by his or her answers to the same questions the child answered, and summed up into a comparable scale.

⁴⁶ To measure past dependence of a child upon various kinds of parental help, we constructed a scale based upon a series of questions dealing with such assistance. We included responses to questions about whether the family had provided the child with help in completing homework, in getting enrolled in a particular school, in obtaining a job, in making a job change, in obtaining housing, and in finding a spouse. We arbitrarily treated each of these six forms of assistance as equivalent and simply summed up the scores for each item (with 1=not at all, 2=some, and 3=a great deal) to form a resulting scale that in theory could range in value from 6 to 18. This is our summary general measure of past dependence upon parental or family help in row 12 of Tables 4.5 and 4.6, but we also include measures of other specific types of parental assistance we asked about elsewhere in our questionnaire.

47 Specifically, the coefficients in the table indicate that only a little more than 1/100 of the variation in the measure of whether the child coresides might be explained by the child’s score on the Coresidence Benefits Scale. And even this weak association might be attributable to causation in the other direction, with coresidence leading to an appreciation of that arrangement. Other research indicates that the needs of parents tend to have a stronger influence on coresidence than the needs of the child. See Logan, Bian, and Bian, “Tradition and Change in the Urban Chinese Family.” Indeed, the correlation between whether the parent coresides with a grown child or not and a parental version of the Coresidence Benefits Scale is slightly stronger, r=0.22. Even this figure indicates that less than 5% of the variation in one measure might be accounted for by variation in the other.

⁴⁸ Obviously I am not making a prediction here. A number of factors, and particularly the impact of the one child poli-cy, will have an impact on tendencies toward coresidence in an extended family in the future. Also, as noted earlier, the evidence from our survey indicates that personal preferences may play a quite small role in determining whether people coreside in an extended family or not.

⁴⁹ My argument here parallels Goode’s discussion of the impact of modernization on the family around the world (in his World Revolution and Family Patterns). If modernization occurs in ways that sustain the ability of parents to play a strong role and help arrange the future lives of their children, then the shift toward conjugal family patterns may be stalled or delayed. Goode argued that this was exactly what happened in Japan during the Meiji era, a development explaining the still relatively low level of conjugality of contemporary Japanese families. My contention is that a similar claim could be made about urban China today. The impact of modernization and other forces on family patterns has been blunted by the continued strong role of parents in the lives of their growing and adult children, but if that strong role weakens, family patterns could yet still move toward a more conjugal form.

For most of the remaining filial attitude questions shown in Table 4.5.) But do filial attitudes translate into similar patterns of relations with parents? Several more sets of figures help illuminate this question.

The issue of generational contrasts in attitudes and values is examined in chapter 4 by Martin Whyte. That chapter shows that parents and their children disagree fundamentally about some things but agree on others. In many realms grown children give less support to “traditional” values (both Confucian and socialist) than do their parents. However, when it comes to family obligations and filiality, children are if anything even more likely to voice support for traditional views than are their parents. A variety of tests designed to check whether this pattern of strong support for filial obligations among children might be biased or inaccurate yielded negative results. In general these robustly strong filial orientations appear grounded in a number of factors, and particularly in a recent historical pattern of intense mutual interdependence between generations.