National O National Council Council on Famil on Family Relations Relations Love Matches and Arranged Marriages: A Chinese Replication Author(s): Xu Xiaohe and Martin King Whyte Source: Journal of Marriage and the Family, Vol. 52, No. 3(Aug, 1990), pp. 709-722 Published by: National Council on Family Relations StableUrl:http://www.jstor.org/stable/352936 Accessed:14/10/200815:42 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jspJstOr'sTermsandConditionsofUseprovidesinpartthatunless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use Please contact the publisher regarding any further use of this work, Publisher contact information may be obtained at http://www.jstor.org/action/showpublisher?publishercode=ncfr Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmIssion JStOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the cholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about jSTOR, please contact support(@jstor. org National Council on Family Relations and National Council on Family relations are collaborating with JStOR to digitize, preserve and extend access to Journal of marriage and the Family ittp://www.jstor.org
Love Matches and Arranged Marriages: A Chinese Replication Author(s): Xu Xiaohe and Martin King Whyte Source: Journal of Marriage and the Family, Vol. 52, No. 3 (Aug., 1990), pp. 709-722 Published by: National Council on Family Relations Stable URL: http://www.jstor.org/stable/352936 Accessed: 14/10/2008 15:42 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ncfr. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. National Council on Family Relations and National Council on Family Relations are collaborating with JSTOR to digitize, preserve and extend access to Journal of Marriage and the Family. http://www.jstor.org
XU XIAOHE AND MARTIN KING WHYTE University of michigan Love Matches and Arranged Marriages A Chinese Replication Data from a probability sample of 586 ever. Through some combination of structural mod married women in Chengdu, Sichuan, in the Peo- ernization, cultural Westernization, and govern ple's Republic of China, are used to examine the mental pressure, arranged marriages are increas transition from arranged to free-choice marriages ingly giving way to freedom of choice, or"love in that society. Retrospective data on mate-choice matche pple play the domi experiences reveal that the role of parents has nant role in selecting whom they will marry.(For declined sharply while young people more and the classic account of this transition worldwide ore dominate the process of spouse selection However, the transition toward free mate choice Even though the nature of the trend is in- Dears to have made little further headway in re- disputable, its implications are not. Individuals cent years, and there is still little sign of a"dating reared in the West, where youth-driven"mate culture"emerging Variations in mate-choice ex- choice has been the rule in most social strata for periences are used to test the prediction of de- centuries, assume that this shift from arranged fenders of arranged marriage that"love matches marriages to love matches is progressive and start out hot and grow cold, while arranged mar- "healthy. As young people are increasingly riages start out cold and grow hot. In a partial freed from arbitrary family dictates and controls replication of an investigation of the same ques- they are able to select partners tion conducted by robert blood in Tokyo, Japan, criteria of love and personal compatibility, and in 1959, the evidence refutes this prediction. the result should be happier marriages and in- Multiple regression analyses indicate that wives in dividuals. In the " bad old days 'of arranged mar- Chengdu love matches are more satisfied with riages, according to this view, many people found their marital relationships than their counterparts themselves stuck in marriages with persons in arranged marriages, regardless of the length decidedly not of their own choosing, mates factors that differentiate these two types of repulsive. The shift away from arranger onl the marriage, and that this difference cannot be selected on the basis of family status, wealth attributed to the influence of other background other criteria, whom they might find perso riage, then, should reduce the level of marital misery in a society Throughout the world a revolution has been tak Traditionalists in many parts of the world raise ing place in the way mates are selected. In societies questions about this set of assumptions, however where parents used to select marital partners for They point to the high divorce rates in modern their children, the power of parents is crumbling. societies characterized by freedom of mate choice evidence of the pro Journal of Marriage and the Family 52(August 1990): 709-722
Xu XIAOHE AND MARTIN KING WHYTE University of Michigan Love Matches and Arranged Marriages: A Chinese Replication Data from a probability sample of 586 evermarried women in Chengdu, Sichuan, in the People's Republic of China, are used to examine the transition from arranged to free-choice marriages in that society. Retrospective data on mate-choice experiences reveal that the role of parents has declined sharply, while young people more and more dominate the process of spouse selection. However, the transition toward free mate choice appears to have made little further headway in recent years, and there is still little sign of a "dating culture" emerging. Variations in mate-choice experiences are used to test the prediction of defenders of arranged marriage that "love matches start out hot and grow cold, while arranged marriages start out cold and grow hot." In a partial replication of an investigation of the same question conducted by Robert Blood in Tokyo, Japan, in 1959, the evidence refutes this prediction. Multiple regression analyses indicate that wives in Chengdu love matches are more satisfied with their marital relationships than their counterparts in arranged marriages, regardless of the length of the marriage, and that this difference cannot be attributed to the influence of other background factors that differentiate these two types of women. Throughout the world a revolution has been taking place in the way mates are selected. In societies where parents used to select marital partners for their children, the power of parents is crumbling. Department of Sociology, University of Michigan, 3012 Literature, Science, and the Arts Building, Ann Arbor, MI 48109-1382. Through some combination of structural modernization, cultural Westernization, and governmental pressure, arranged marriages are increasingly giving way to freedom of choice, or "love matches," in which young people play the dominant role in selecting whom they will marry. (For the classic account of this transition worldwide, see Goode, 1963.) Even though the nature of the trend is indisputable, its implications are not. Individuals reared in the West, where "youth-driven" mate choice has been the rule in most social strata for centuries, assume that this shift from arranged marriages to love matches is progressive and "healthy." As young people are increasingly freed from arbitrary family dictates and controls, they are able to select partners according to criteria of love and personal compatibility, and the result should be happier marriages and individuals. In the "bad old days" of arranged marriages, according to this view, many people found themselves stuck in marriages with persons decidedly not of their own choosing, mates selected on the basis of family status, wealth, or other criteria, whom they might find personally repulsive.' The shift away from arranged marriage, then, should reduce the level of marital misery in a society. Traditionalists in many parts of the world raise questions about this set of assumptions, however. They point to the high divorce rates in modern societies characterized by freedom of mate choice as evidence of the problems inherent in love matches. They claim that arranged marriages have virtues that are not appreciated by people in Western societies. The contrast they draw is ilJournal of Marriage and the Family 52 (August 1990): 709-722 709
710 lustrated by the phrase, love matches start out up. After the marriage the couple will have the hot and grow cold, while arranged marriages start opportunity to get to know one another and forge out cold and grow hot. 'In other words, love common bonds. As this process occurs, com matches typically involve a very intense romantic patibility and mutual concern are likely to lead to involvement, accompanied by idealization of the a mature form of love, perhaps never as"hot ' as partner and fantasies about wedded bliss, in thethe premarital emotions experienced in a love period prior to the wedding(the"hot phase). match, but a relationship that provides a more Then after the wedding, reality sets in, and some realistic and durable bond that can survive the test combination of domestic chores, child care of time and family difficulties. In the long run, at burdens, financial anxieties, and mundane life least, the traditionalists claim, arranged marriages with a less-than-ideal real life partner leads to a give more satisfaction than free-choice marriages more or less inevitable decline in romantic feelings In Figure I we present a hypothetical picture of and satisfaction with the relationship over the the trends in marriage satisfaction that might be years. Here traditionalists can point to a con- found for love matches and arranged marriages if siderable accumulation of evidence, from both the claims of traditionalists are correct cross-sectional and longitudinal studies in But can we assume that spouses selected by Western societies, showing that marital happiness parents will have any realistic basis for developing and satisfaction ratings do tend to decline over the a compatible relationship? Critics of arranged course of a marriage(see for example, Pineo, marriages can point to cases that make this seem 1961; Renne, 1970; Hicks and Platt, 1970) quite unlikely-of a young and vital woman mar- The trajectory for arranged marriages, accord- ried off to an old and feeble man, or to a de know each other well, or perhaps not at all, and However, traditionalists argue that such cases are since they don' t have any romantic feelings for atypical and that arranged marriages are gener one another prior to the marriage, the partners in more likely to produce compatibility than are love an arranged marriage "have nowhere to go but matches. They argue that parents are concerned FIGURE 1. HYPOTHETICAL TRENDS FOR MARRIAGE SATI ION IN OVE MATCHES AND ARRANGED MARRIAGES -cH Love Matches 5-8 9.14 15-19 20-25
Journal of Marriage and the Family lustrated by the phrase, "love matches start out hot and grow cold, while arranged marriages start out cold and grow hot." In other words, love matches typically involve a very intense romantic involvement, accompanied by idealization of the partner and fantasies about wedded bliss, in the period prior to the wedding (the "hot" phase). Then after the wedding, reality sets in, and some combination of domestic chores, child care burdens, financial anxieties, and mundane life with a less-than-ideal real life partner leads to a more or less inevitable decline in romantic feelings and satisfaction with the relationship over the years. Here traditionalists can point to a considerable accumulation of evidence, from both cross-sectional and longitudinal studies in Western societies, showing that marital happiness and satisfaction ratings do tend to decline over the course of a marriage (see for example, Pineo, 1961; Renne, 1970; Hicks and Platt, 1970).2 The trajectory for arranged marriages, according to traditionalists, is different. Since they don't know each other well, or perhaps not at all, and since they don't have any romantic feelings for one another prior to the marriage, the partners in an arranged marriage "have nowhere to go but up." After the marriage the couple will have the opportunity to get to know one another and forge common bonds. As this process occurs, compatibility and mutual concern are likely to lead to a mature form of love, perhaps never as "hot" as the premarital emotions experienced in a love match, but a relationship that provides a more realistic and durable bond that can survive the test of time and family difficulties. In the long run, at least, the traditionalists claim, arranged marriages give more satisfaction than free-choice marriages. In Figure 1 we present a hypothetical picture of the trends in marriage satisfaction that might be found for love matches and arranged marriages if the claims of traditionalists are correct. But can we assume that spouses selected by parents will have any realistic basis for developing a compatible relationship? Critics of arranged marriages can point to cases that make this seem quite unlikely-of a young and vital woman married off to an old and feeble man, or to a deformed son of a wealthy family, and so forth. However, traditionalists argue that such cases are atypical and that arranged marriages are generally more likely to produce compatibility than are love matches. They argue that parents are concerned FIGURE 1. HYPOTHETICAL TRENDS FOR MARRIAGE SATISFACTION IN LOVE MATCHES AND ARRANGED MARRIAGES 6 ---- Love Matches 0--- Arranged Marriages -2 . . . . . . ..... 0-1 2-4 5-8 9-14 15-19 20-25 25+ Years Married 0 0 -_ (n o, 0) 4 - 2- 0 - - r- 710
Love Matches and Arranged marriages in China 711 about the happiness of their offspring and will the trend was the opposite, with women from ar usually be in a better position to judge com- ranged marriages eventually being much more patibility in the long run than their children are dissatisfied and having fewer love feelings than They can rationally evaluate the nature of their their counterparts from love matches. If there are own child and investigate the character of a pro- any advantages to arranged marriages, according ective spou nd then use their to the Tokyo evidence, they accrue only to wisdo husbands. In any case the of marital from that posited by the may be young and immature, and even if not they traditionalists, without the signs of steady im may be too swayed by emotions and hormones to provement(becoming "hot")suggested in the choice of tner. The hypothetica result of allowing young people to choose their in Figure I own spouses, then, may be that they will be Several features of the Tokyo study make a blinded by love and overlook areas of personal in- replication elsewhere seem worthwhile. Fir compatibility, and the latter will become apparent restricted nature of the sample used by and cause problems later on in the marriage raises questions about whether the same sort of Parents are less likely to be so blinded or so the pattern would have been found even in Japan argument goes with a more representative sample. Second, Blood did not subject his results to multivariate analysis LOVE MATCHES AND ARRANGED MARRIAGES thus leaving open the question of whether the con IN TOKYO trasts drawn between the two types of marriage could be spurious. Finally, it is not clear that this So far as we are aware, this set of arguments, and single Japanese case study can allow us to be con particularly those involving arranged marriages, clusive about the merits or demerits of the two has been systematically investigated only once. In types of marriage. For the majority of the 1959 Robert Blood conducted a survey in Tokyo, Japanese population, the peasants, free choice Japan, that was designed specifically to compare and even trial marriages had been the rule up to a love matches and arranged marriages(see Blood, century prior to Bloods survey, with arranged 1967). Blood and his associates interviewed 444 marriage being fostered subsequently through a married couples who lived in nuclear-family process ofsamuraization"(see the discussion in households in three predominantly white-collar Goode, 1963, chap 7). Then American military housing projects in Tokyo, with a further restric- occupation after World War II played some role tion that the wives in these families had to be age in undermining this newly dominant arranged 40 or younger. His research resulted in a serious marriage custom. Given this history, we might challenge to the"starting hot and cooling down suspect that cultural support for love-match and versus starting cold and heating up,argument arranged-marriage alternatives in Japan would be presented by defenders of arranged marriage, pa different and perhaps more shallow than in other cular ly in the case of wives. societies. In any case, not only social scientists Figures 2 and 3 convey important results from but also individuals in transitional societies who the blood study that are central to that challenge are debating these alternatives and facing anx Using a variety of indicators of marriage quality, ieties about finding a spouse, deserve additional such as those shown in these figures, Blood ttempts to compare the implications for marital not find the pattern of starting cold and heating relations in free-choice and arranged marriages up posited for arranged marriages. Rather, for both types of marriages the long-term trajectory downward, toward less expres A PARTIAL REPLICATION IN THE lower marriage satisfaction. For arranged- PEOPLES REPUBLIC OF CH marriage (miai husbands this decline was more In the pages that follow we report the results of a gradual than for their love-match counterparts, so partial replication of the love-match/arranged that in later stages the men who had experienced marriage comparison conducted in Chengdu, the arranged marriages were more satisfied and ex- capital of Sichuan Province, in the People's perienced more love feelings. However, for wives Republic of China. a probability sample of
Love Matches and Arranged Marriages in China about the happiness of their offspring and will usually be in a better position to judge compatibility in the long run than their children are. They can rationally evaluate the nature of their own child and investigate the character of a prospective spouse for that child and then use their wisdom and experience to arrive at the most "suitable" match. Their offspring, in contrast, may be young and immature, and even if not they may be too swayed by emotions and hormones to make a "wise" choice of a marital partner.3 The result of allowing young people to choose their own spouses, then, may be that they will be blinded by love and overlook areas of personal incompatibility, and the latter will become apparent and cause problems later on in the marriage. Parents are less likely to be so blinded, or so the argument goes. LOVE MATCHES AND ARRANGED MARRIAGES IN TOKYO So far as we are aware, this set of arguments, and particularly those involving arranged marriages, has been systematically investigated only once. In 1959 Robert Blood conducted a survey in Tokyo, Japan, that was designed specifically to compare love matches and arranged marriages (see Blood, 1967). Blood and his associates interviewed 444 married couples who lived in nuclear-family households in three predominantly white-collar housing projects in Tokyo, with a further restriction that the wives in these families had to be age 40 or younger. His research resulted in a serious challenge to the "starting hot and cooling down versus starting cold and heating up" argument presented by defenders of arranged marriage, particularly in the case of wives. Figures 2 and 3 convey important results from the Blood study that are central to that challenge. Using a variety of indicators of marriage quality, such as those shown in these figures, Blood did not find the pattern of starting cold and heating up posited for arranged marriages. Rather, for both types of marriages the long-term trajectory was downward, toward less expression of love and lower marriage satisfaction. For arrangedmarriage (miai) husbands this decline was more gradual than for their love-match counterparts, so that in later stages the men who had experienced arranged marriages were more satisfied and experienced more love feelings. However, for wives the trend was the opposite, with women from arranged marriages eventually being much more dissatisfied and having fewer love feelings than their counterparts from love matches. If there are any advantages to arranged marriages, according to the Tokyo evidence, they accrue only to husbands. In any case the trajectory of marital feelings is quite different from that posited by the traditionalists, without the signs of steady improvement (becoming "hot") suggested in the hypothetical curve for arranged marriages shown in Figure 1.4 Several features of the Tokyo study make a replication elsewhere seem worthwhile. First, the restricted nature of the sample used by Blood raises questions about whether the same sort of pattern would have been found even in Japan with a more representative sample. Second, Blood did not subject his results to multivariate analysis, thus leaving open the question of whether the contrasts drawn between the two types of marriage could be spurious. Finally, it is not clear that this single Japanese case study can allow us to be conclusive about the merits or demerits of the two types of marriage. For the majority of the Japanese population, the peasants, free choice and even trial marriages had been the rule up to a century prior to Blood's survey, with arranged marriage being fostered subsequently through a process of "samuraization" (see the discussion in Goode, 1963, chap. 7). Then American military occupation after World War II played some role in undermining this newly dominant arrangedmarriage custom. Given this history, we might suspect that cultural support for love-match and arranged-marriage alternatives in Japan would be different and perhaps more shallow than in other societies. In any case, not only social scientists, but also individuals in transitional societies who are debating these alternatives and facing anxieties about finding a spouse, deserve additional attempts to compare the implications for marital relations in free-choice and arranged marriages. A PARTIAL REPLICATION IN THE PEOPLE'S REPUBLIC OF CHINA In the pages that follow we report the results of a partial replication of the love-match/arrangedmarriage comparison conducted in Chengdu, the capital of Sichuan Province, in the People's Republic of China. A probability sample of 586 711
FIGURE 2. HUSBANDS AND WIVES' LOVE FEELINGS IN LOVE MATCHES AN ARRANGED MARRIAGES BY LENGTH OF MARRIAGE, ToKYO, 1959 42 4.0 “ Strong 39 miai husbands 3.7 ∵. Love-match husbands ,垂非 s-m9 3.5 Love-match wives 3.3 3.2 3.1 “ Consid erable 3.0 Years Married Source: From Love Match and A e:A Tokyo-Detroit Coi Robert o. blo p 87. Copyright 1967 by Robert ood, Jr. Reprinted by permission of The Free Press, a Division of Mac
Journal of Marriage and the Family FIGURE 2. HUSBANDS' AND WIVES' LOVE FEELINGS IN LOVE MATCHES AND ARRANGED MARRIAGES BY LENGTH OF MARRIAGE, TOKYO, 1959 4.2 4.1 4.0 "Strong" 3.9 r I, ? r r r r r I ??? ?) r ?- ? ? r ? r r r_ ? r r n 3.8 - .* Miai husbands eeeee.e............................;. 3.7 - * Love-match husbands A - ? ? - e e" ..e .e eeee..e..... 3.6 3.5 3.4 - 3.3 , \ 3.2- 3.1 ' "Consid-' erable" 3.0 - Engager nent <2 2-4 5-8 9+ Years Married Source: From Love Match and Arranged Marriage: A Tokyo-Detroit Comparison, by Robert 0. Blood, Jr., p. 87. Copyright ? 1967 by Robert 0. Blood, Jr. Reprinted by permission of The Free Press, a Division of Macmillan, Inc. 5-, a w 712 * I : I , -I II _ I I I
Love Matches and Arranged Marriages in China FIGuRE 3. HUSBANDS'AND WIVES' AGGREGATE MARITAL SATISFACTION IN LOVE MATCHES AND ARRANGED MARRIAGES BY LENGTH OF MARRIAGE, TOKYO, 1959 Miai husbands Love-match wives Love-match husband 4.5 Miai wives 40 Years married ged Marriage: A Tokyo-Detroit Comparison, by robert o. Blood, Jr p.85.C 67 by Robert O. Blood, Jr. Reprinted by permission of The Free Press, a Division of Mac
Love Matches and Arranged Marriages in China FIGURE 3. HUSBANDS' AND WIVES' AGGREGATE MARITAL SATISFACTION IN LOVE MATCHES AND ARRANGED MARRIAGES BY LENGTH OF MARRIAGE, TOKYO, 1959 713 .S b< : <2 2-4 5-8 9+ Years Married Source: From Love Match and Arranged Marriage: A Tokyo-Detroit Comparison, by Robert 0. Blood, Jr., p. 85. Copyright ? 1967 by Robert 0. Blood, Jr. Reprinted by permission of The Free Press, a Division of Macmillan, Inc
714 Journal of marriage and the Family ever-married women between the ages of 22 and draw attention to the tragic case of a young 70 residing in the two main urban districts of woman driven to suicide by her unwillingness to hengdu(which together contained about 1. 2 accept the man chosen for her by her parents(see million people in 1982, or 97 o of the urban Witke, 1973). Over time a combination of increas- population of that city) was interviewed in 1987, ing wage labor in Chinas towns and cities and with a response rate of 87. 70. Sichuan is Chinas growing Western influence on Chinas culture and most populous province, with over 100 million educational system did begin to foster a people, and it is located in the Southwest. Even voice for young people in mate-choice decisions though Chengdu is the capital of the province, it during the republican era(1912-1949-see Lang may be less cosmopolitan than cities near the 1946; Levy, 1949) coast, such as Peking, Canton, or Shanghai. No After the Chinese Communists came to na claim is made here that Chengdu is representative tional power in 1949, they vigorously promoted of urban China in general. However, existing freedom of mate choice. The Marriage Law of the research on other urban areas(see Whyte and People's Republic of China, adopted in 1950, de Parish, 1984)suggests that the general features of nounced the "arbitrary and compulsory feudal he mate-choice transition process in Chengdu are marriage system.. which ignores the childrens similar to what is occurring elsewhere interests "and proclaimed that"marriage shall be Women in the Chengdu sample were first mar- based upon the complete willingness of the two ried over a 55-year period, from 1933 to 1987, and parties. Neither party shall use compulsion and no during this time period a major transformation third party shall be allowed to interfere"(quoted from arranged to free-choice marriages has been in Yang, 1959: 221). Marriage registration offices under way. The results of the Chengdu survey will were established nationwide, where couples were be used to compare Chinese love matches and ar- to be interviewed at the point of marriage to see anged marriages. Unfortunately, husbands we whether they were doing so voluntarily. A nation not interviewed, so we can only examine how mar- wide propaganda campaign was launched during riages were viewed by wives. The Chengdu survey the period 1950-1953 to try to mobilize support also uses somewhat different measures from those for, and compliance with, the Marriage Law. In employed by Blood, and of course our sample is subsequent years socialist transformation of the broader and more representative, so what follows economy eliminated the family as a production is a partial, rather than an exact, replication of the unit (until the post-Mao reforms, at least), and he resulting proletarianization of the population Arranged marriages were the dominant tradi- combined with industrialization and the spread of ion in China for all classes for centuries, unlike education, helped to accelerate the process of the situation in Japan. Furthermore, Chinese ar- transition from arranged to free-choice marriages ranged marriages often took a more extreme form (see the fuller discussion in Parish and Whyte than that found in Chinas Asian neighbor. While 1978; Whyte and Parish, 1984) in the Japanese miai custom the young couple had ross-sectional data from the Chengdu survey ritualized meeting and at least some opportunity can be used to get some picture of the timing and to express an opinion about the partner selected extent of this transition. Responses to several per- for them, in China many people experienced what tinent, closed-ended questions are displayed by ould be called a"blind marriage, "'in which the time periods in Table 1. Women in the sample are parents monopolized the decision(with the aid of arrayed in this table in terms of the years they first hired go-betweens), and the couple did not even married, rather than by birth cohorts, and the meet until the day of the wedding. 6 time divisions that form marriage cohorts in the his arranged marriage system came under at- table are selected to match the major political tack in the early decades of this century, with turning points in recent Chinese history-the reformers and revolutionaries denouncing the per- coming to power of the Chinese Communist Party sonal misery and suicides that resulted(see Pa, in 1949, the launching of the Great Leap Forward 933/1972). One interesting voice in this chorus of in 1958, the onset of the Cultural Revolution criticism was a young fellow by the name of Mao 1966, and the beginning of the post-Mao reform Zedong, who used one of his first appearances in era print, a series of newspaper articles in 1919, to everal features of the contrasts across mar
Journal of Marriage and the Family ever-married women between the ages of 22 and 70 residing in the two main urban districts of Chengdu (which together contained about 1.2 million people in 1982, or 97% of the urban population of that city) was interviewed in 1987, with a response rate of 87.7o%. Sichuan is China's most populous province, with over 100 million people, and it is located in the Southwest. Even though Chengdu is the capital of the province, it may be less cosmopolitan than cities near the coast, such as Peking, Canton, or Shanghai. No claim is made here that Chengdu is representative of urban China in general. However, existing research on other urban areas (see Whyte and Parish, 1984) suggests that the general features of the mate-choice transition process in Chengdu are similar to what is occurring elsewhere. Women in the Chengdu sample were first married over a 55-year period, from 1933 to 1987, and during this time period a major transformation from arranged to free-choice marriages has been under way. The results of the Chengdu survey will be used to compare Chinese love matches and arranged marriages. Unfortunately, husbands were not interviewed, so we can only examine how marriages were viewed by wives. The Chengdu survey also uses somewhat different measures from those employed by Blood, and of course our sample is broader and more representative, so what follows is a partial, rather than an exact, replication of the Tokyo study. Arranged marriages were the dominant tradition in China for all classes for centuries, unlike the situation in Japan. Furthermore, Chinese arranged marriages often took a more extreme form than that found in China's Asian neighbor. While in the Japanese miai custom the young couple had a ritualized meeting and at least some opportunity to express an opinion about the partner selected for them, in China many people experienced what could be called a "blind marriage," in which the parents monopolized the decision (with the aid of hired go-betweens), and the couple did not even meet until the day of the wedding.6 This arranged marriage system came under attack in the early decades of this century, with reformers and revolutionaries denouncing the personal misery and suicides that resulted (see Pa, 1933/1972). One interesting voice in this chorus of criticism was a young fellow by the name of Mao Zedong, who used one of his first appearances in print, a series of newspaper articles in 1919, to draw attention to the tragic case of a young woman driven to suicide by her unwillingness to accept the man chosen for her by her parents (see Witke, 1973). Over time a combination of increasing wage labor in China's towns and cities and growing Western influence on China's culture and educational system did begin to foster a growing voice for young people in mate-choice decisions during the republican era (1912-1949-see Lang, 1946; Levy, 1949). After the Chinese Communists came to national power in 1949, they vigorously promoted freedom of mate choice. The Marriage Law of the People's Republic of China, adopted in 1950, denounced the "arbitrary and compulsory feudal marriage system . . . which ignores the childrens' interests" and proclaimed that "marriage shall be based upon the complete willingness of the two parties. Neither party shall use compulsion and no third party shall be allowed to interfere"(quoted in Yang, 1959: 221). Marriage registration offices were established nationwide, where couples were to be interviewed at the point of marriage to see whether they were doing so voluntarily. A nationwide propaganda campaign was launched during the period 1950-1953 to try to mobilize support for, and compliance with, the Marriage Law. In subsequent years socialist transformation of the economy eliminated the family as a production unit (until the post-Mao reforms, at least), and the resulting proletarianization of the population, combined with industrialization and the spread of education, helped to accelerate the process of transition from arranged to free-choice marriages (see the fuller discussion in Parish and Whyte, 1978; Whyte and Parish,1984). Cross-sectional data from the Chengdu survey can be used to get some picture of the timing and extent of this transition. Responses to several pertinent, closed-ended questions are displayed by time periods in Table 1. Women in the sample are arrayed in this table in terms of the years they first married, rather than by birth cohorts, and the time divisions that form marriage cohorts in the table are selected to match the major political turning points in recent Chinese history-the coming to power of the Chinese Communist Party in 1949, the launching of the Great Leap Forward in 1958, the onset of the Cultural Revolution in 1966, and the beginning of the post-Mao reform era in 1977. Several features of the contrasts across mar- 714
Love Matches and Arranged Marriages in China TABLE 1. CHANGES IN ASPECTS OF FREEDOM OF MATE CHOICE IN CHENGDU Year First Married 1933-48 1. Traditional arranged marriage (%o) 2. Type of marriage(%o) 3. Dominant role in mate choice(%0) 30 8 5 4. Introduced to husband (o) 5. Who provided the introduction?(o) wn generatio 38 43 57684578 Parents generation 19 6. Dated husband prior to marriage?(%0) 12 17 7. Number of romances (%o) 8. Had other marital prospects(o) 9. How much in love when married?(o) 5789790 Not at all 210 age cohorts visible in Table 1 deserve comment. percentages for behavior associated with arranged First, it is apparent that a major shift away from marriages in the early cohorts. In addition, some arranged marriage has occurred with instances in political bias might be involved It may be accep- which parents dominate the proceedings declining table to admit to an interviewer that your parents from 60o to 70o in the pre-1949 period to under controlled the process if you married before 1949, 10%o today. Relatedly, those who met their but it may be unacceptable to reveal that fact husbands directly rather than through an in- about a marriage occurring within the socialist troduction, those whose introducers were peers era. We made a number of attempts to"screen rather than parents, those who had some roman- out"such biases by recomputing the figures in tic relationships and dates prior to marriage and Table 1, with those who married in rural areas and those who describe themselves as having been in those we assumed would most likely display love have all increased in comparison with the political bias(e.g. Party members)removed, and pre-1949 era. in general the extent of the shift from arranged to The extent of the change shown in the table free-choice marriages shown in the table was only might be exaggerated as a result of two kinds of slightly affected. In fact, only the elimination of biases. First, even though respondents were all women with rural roots had a detectable impact residing in Chengdu at the time of the interviews, on the results, and that was mostly for the second some of them were not when they married. Higher 1949-57, marriage cohort, rather than for the first portions of the earliest marriage cohorts than of (indicating somewhat more"actual"freedom of recent cohorts actually lived in rural areas when mate choice in Chengdu than shown in the second they married. Since the arranged -marriage column of Table 1--results not displayed here tradition remains stronger even today in Chinas So the general extent of the transformation is villages, that fact might tend to inflate the clear from the data, even if the exact percentages
Love Matches and Arranged Marriages in China TABLE 1. CHANGES IN ASPECTS OF FREEDOM OF MATE CHOICE IN CHENGDU Year First Married Item 1933-48 1949-57 1958-65 1966-76 1977-87 1. Traditional arranged marriage (%) 69 22 1 0 0 2. Type of marriage (%) Arranged 68 27 0 1 2 Intermediate 15 33 45 40 41 Individual choice 17 40 55 59 57 3. Dominant role in mate choice (%) Parents 56 30 7 8 5 Mixed 15 11 6 3 6 Respondent 28 59 87 89 89 4. Introduced to husband (o) 91 76 54 59 60 5. Who provided the introduction? (%) Own generation 38 43 75 75 74 Other 8 17 7 6 9 Parents' generation 53 40 18 19 17 6. Dated husband prior to marriage? (%) Often 12 17 24 40 48 Sometimes 6 18 27 13 21 Rarely 23 22 30 31 24 Never 59 44 18 16 7 7. Number of romances (No) None 73 29 9 5 5 One 24 63 74 66 67 More than one 3 8 18 29 28 8. Had other marital prospects (%) 4 5 2 6 9 9. How much in love when married? (%) Completely 17 38 63 61 67 26 29 22 26 19 35 20 9 11 10 9 4 4 1 3 Not at all 13 9 2 1 0 n 71 107 82 116 210 riage cohorts visible in Table 1 deserve comment. First, it is apparent that a major shift away from arranged marriage has occurred, with instances in which parents dominate the proceedings declining from 60% to 70%o in the pre-1949 period to under 10% today. Relatedly, those who met their husbands directly rather than through an introduction, those whose introducers were peers rather than parents, those who had some romantic relationships and dates prior to marriage, and those who describe themselves as having been in love have all increased in comparison with the pre-1949 era. The extent of the change shown in the table might be exaggerated as a result of two kinds of biases. First, even though respondents were all residing in Chengdu at the time of the interviews, some of them were not when they married. Higher portions of the earliest marriage cohorts than of recent cohorts actually lived in rural areas when they married. Since the arranged-marriage tradition remains stronger even today in China's villages, that fact might tend to inflate the percentages for behavior associated with arranged marriages in the early cohorts. In addition, some political bias might be involved. It may be acceptable to admit to an interviewer that your parents controlled the process if you married before 1949, but it may be unacceptable to reveal that fact about a marriage occurring within the socialist era. We made a number of attempts to "screen out" such biases by recomputing the figures in Table 1, with those who married in rural areas and those we assumed would most likely display political bias (e.g. Party members) removed, and in general the extent of the shift from arranged to free-choice marriages shown in the table was only slightly affected. In fact, only the elimination of women with rural roots had a detectable impact on the results, and that was mostly for the second, 1949-57, marriage cohort, rather than for the first (indicating somewhat more "actual" freedom of mate choice in Chengdu than shown in the second column of Table 1-results not displayed here). So the general extent of the transformation is clear from the data, even if the exact percentages 715
716 Journal of marriage and the Family for each time period may be open to some argu- indicating greater freedom of mate choice. The ment average item-to- item correlation (r) was 42. We The results in Table 1 also show that in spite of en constructed our scale by computing the stan the increasing role that young people have played dardized, or z, scores for each item and then tak recently in selecting their own spouses, very little ing the unweighted mean of these standardized in the way of a"dating culture"exists yet in scores. Values computed for this scale are plotted Chengdu.(Elements defining a dating culture in the form of a three-year moving average against might include the ability of young people to link years of first marriage in Figure 4. up romantically without adult supervision in a set From this figure we can conclude that freedom ting that is not defined as leading directly to mar- of mate choice was already increasing during the riage. In such a dating culture it is acceptable for 1930s, that this increase halted during World War young people to"try out" a variety of romantic II, and that a major further increase occurred dur- partners before progressing to the stage of prepar- ing the post-war period and the 1950s. However ing to select a spouse. See the discussion in there have been only incremental shifts toward Modell, 1983 Most women(over 90%0)never had even greater freedom for young people since that other person they considered marrying besides time. The reasons for this"stalling",of the transi their eventual husbands, and less than 30%o even tion (which contrasts with a more linear trend had had other boyfriends. Furthermore, even in toward freedom of choice in Taiwan--see Thorn- recent years generally 30-50%o of the women inter- ton, Chang, and Lin, 1989)are too complex to go viewed say they rarely or never dated their future into here, but explanations are attempted in a husbands, and even for those who did, the dates companion paper(see Whyte, 1990). Suffice it to almost always came after the decision to marry, say here that while China's communist leaders rather than prior to it(a fact that is not obvious have consistently advocated freedom for young from the table but is made clear by ethnographic people to choose their own mates, they have observations in China). In this setting, in which erected considerable barriers designed to inhibit young people have to make a vital decision about young people from developing a dating culture their lives without being able to gain experience not to mention experimenting with premarital sex first via casual dating, it is understandable that in In any case, the figures in Table 1 and Figure 4 at some cases parents are able to exert considerable test to the substantial change in the direction of influence. Depending upon which measure one freedom of mate choice that has occurred in the even a dominant lives of Chengdu women, while also leaving role in 11-43% of recent marriages, and almost enough variability in mate-choice experiences one-fifth of those who rely on introductions end even recently to allow us to compare free-choice up receiving these from their parents or other and arranged marriage senior kin. So a second conclusion that can be drawn from the table is that the transition to freedom of mate choice has not produced as much FREEDOM OF MATE CHOICE AND ability for young people to exercise free choice as MARRIAGE QUALIT currently exists in the we Has the shift toward greater freedom of mate A final pattern that becomes clear from the choice in China produced more satisfactory mar figures in the table is that the transition to free riage relations or less satisfactory ones? We ex- mate choice seems to have largely"stalled"since amine this question first with a graphic display he late 1950s. Although there have been small in- of the impact of the mode of mate choice on mar creases since then in indicators such as the percen- riage quality for Chengdu and then by subjecting tage of women with other marriage prospects and arriage quality measure multivariate in those who often dated prior to marriage nalysis to see whether any impact detected is major shifts occurred among the early cohorts. spurious or not. This exercise can only be con This pattern emerges more clearly if six of the in- ducted for wives, since we did not have the oppor dicators in Table 1(from panels 1, 2, 3, 6, 7, and tunity to interview husbands 9)are used to construct a mean Freedom of Mate The dependent variables in the Chengdu study Choice Scale. The six items indicated are not exactly comparable to those used by ranged in a consistent direction, with a high score Blood Bloods most reliable measure of marriage
Journal of Marriage and the Family for each time period may be open to some argument. The results in Table 1 also show that in spite of the increasing role that young people have played recently in selecting their own spouses, very little in the way of a "dating culture" exists yet in Chengdu. (Elements defining a dating culture might include the ability of young people to link up romantically without adult supervision in a setting that is not defined as leading directly to marriage. In such a dating culture it is acceptable for young people to "try out" a variety of romantic partners before progressing to the stage of preparing to select a spouse. See the discussion in Modell, 1983.) Most women (over 90%) never had another person they considered marrying besides their eventual husbands, and less than 30% even had had other boyfriends. Furthermore, even in recent years generally 30-50% of the women interviewed say they rarely or never dated their future husbands, and even for those who did, the dates almost always came after the decision to marry, rather than prior to it (a fact that is not obvious from the table but is made clear by ethnographic observations in China).7 In this setting, in which young people have to make a vital decision about their lives without being able to gain experience first via casual dating, it is understandable that in some cases parents are able to exert considerable influence. Depending upon which measure one chooses, parents play some or even a dominant role in 11-43% of recent marriages, and almost one-fifth of those who rely on introductions end up receiving these from their parents or other senior kin.8 So a second conclusion that can be drawn from the table is that the transition to freedom of mate choice has not produced as much ability for young people to exercise free choice as currently exists in the West.9 A final pattern that becomes clear from the figures in the table is that the transition to free mate choice seems to have largely "stalled" since the late 1950s. Although there have been small increases since then in indicators such as the percentage of women with other marriage prospects and in those who often dated prior to marriage, the major shifts occurred among the early cohorts. This pattern emerges more clearly if six of the indicators in Table 1 (from panels 1, 2, 3, 6, 7, and 9) are used to construct a mean Freedom of Mate Choice Scale. The six items indicated were arranged in a consistent direction, with a high score indicating greater freedom of mate choice. The average item-to-item correlation (r) was .42. We then constructed our scale by computing the standardized, or z, scores for each item and then taking the unweighted mean of these standardized scores. Values computed for this scale are plotted in the form of a three-year moving average against years of first marriage in Figure 4. From this figure we can conclude that freedom of mate choice was already increasing during the 1930s, that this increase halted during World War II, and that a major further increase occurred during the post-war period and the 1950s. However, there have been only incremental shifts toward even greater freedom for young people since that time. The reasons for this "stalling" of the transition (which contrasts with a more linear trend toward freedom of choice in Taiwan-see Thornton, Chang, and Lin, 1989) are too complex to go into here, but explanations are attempted in a companion paper (see Whyte, 1990). Suffice it to say here that while China's communist leaders have consistently advocated freedom for young people to choose their own mates, they have erected considerable barriers designed to inhibit young people from developing a dating culture, not to mention experimenting with premarital sex. In any case, the figures in Table 1 and Figure 4 attest to the substantial change in the direction of freedom of mate choice that has occurred in the lives of Chengdu women, while also leaving enough variability in mate-choice experiences even recently to allow us to compare free-choice and arranged marriages. FREEDOM OF MATE CHOICE AND MARRIAGE QUALITY Has the shift toward greater freedom of mate choice in China produced more satisfactory marriage relations or less satisfactory ones? We examine this question first with a graphic display of the impact of the mode of mate choice on marriage quality for Chengdu and then by subjecting a marriage quality measure to multivariate analysis to see whether any impact detected is spurious or not. This exercise can only be conducted for wives, since we did not have the opportunity to interview husbands. The dependent variables in the Chengdu study are not exactly comparable to those used by Blood. Blood's most reliable measure of marriage 716
Love Matches and Arranged Marriages in China 7 FIGURE 4. FREEDOM OF MATE CHOICE IN CHENGDU, 1934-1987 2 1930193519401945195019551960196519701975198019851990 Year of MarrIa r sample for average was used to smooth out the irregularities in trends caused by the small number of marriages in our sa each individual year of marriage. satisfaction(wri gregate indicator of marriage this procedure are displayed in Figure 5, with only quality was an structed from a weighted average of ten separate included satisfaction questions, each answered with The main feature visible in Figure 5 is that the responses on a 7-point scale. A similar measure curve for " love matches"'is consistently higher as constructed for Chengdu, consisting of the than that for arranged marriages, regardless of of the standardized scores of wives' the length of the marriage. In other words, the nses to six closed-ended marriage quality Chengdu data, even more clearly than the Tokyo s(each responded to in three categories )and data, indicate that women in free-choice mar- ne general marriage satisfaction question riages are consistently more satisfied with thei (responded to according to a 5-point scale, from marital relationship than are women in arranged very dissatisfied to very satisfied). i The resulting marriages. However, it is also clear that the plots leasure is our Marriage Quality Scale in Figure 5 do not look too much like those in In order to present results in a fashion parallel Figures 2 and 3. The pattern in which women who to those already reviewed from the Blood survey, have been married longer are generally less hed values of the Marriage Quality Scale satisfied with their marriages tha love rding to the number of years married, with wed, which is found not only in Bloods survey eperately. In order to construct such a graph, we earlier), does not appear in our Chengdu results divided the sample into three categories according Indeed for both love matches and arranged mar- to scores on the Freedom of Mate Choice Scale. riages, wives who had been married 20-24 years The middle category, totaling about 30 o of the reported having the highest-quality marriages of sample, is omitted from consideration here, thus all allowing us to compare relatively"pure"cases of This“peak’ for women married20-24 years free choice and arranged marriages. The results of cannot be explained in terms of the U-curve
Love Matches and Arranged Marriages in China FIGURE 4. FREEDOM OF MATE CHOICE IN CHENGDU, 1934-1987 -- -- Free Choice Score -2 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1! Year of Marriage 990 Note: A three-year moving average was used to smooth out the irregularities in trends caused by the small number of marriages in our sample for each individual year of marriage. quality was an aggregate indicator of marriage satisfaction (used in Figure 3), which was constructed from a weighted average of ten separate satisfaction questions, each answered with responses on a 7-point scale.10 A similar measure was constructed for Chengdu, consisting of the mean of the standardized scores of wives' responses to six closed-ended marriage quality items (each responded to in three categories) and one general marriage satisfaction question (responded to according to a 5-point scale, from very dissatisfied to very satisfied). 1 The resulting measure is our Marriage Quality Scale. In order to present results in a fashion parallel to those already reviewed from the Blood survey, we graphed values of the Marriage Quality Scale according to the number of years married, with love matches and free-choice marriages displayed seperately. In order to construct such a graph, we divided the sample into three categories according to scores on the Freedom of Mate Choice Scale. The middle category, totaling about 30% of the sample, is omitted from consideration here, thus allowing us to compare relatively "pure" cases of free choice and arranged marriages. The results of this procedure are displayed in Figure 5, with only respondents still married to their first husbands included. The main feature visible in Figure 5 is that the curve for "love matches" is consistently higher than that for arranged marriages, regardless of the length of the marriage. In other words, the Chengdu data, even more clearly than the Tokyo data, indicate that women in free-choice marriages are consistently more satisfied with their marital relationship than are women in arranged marriages. However, it is also clear that the plots in Figure 5 do not look too much like those in Figures 2 and 3. The pattern in which women who have been married longer are generally less satisfied with their marriages than those recently wed, which is found not only in Blood's survey but in a large number of Western studies (as noted earlier), does not appear in our Chengdu results. Indeed, for both love matches and arranged marriages, wives who had been married 20-24 years reported having the highest-quality marriages of all. This "peak" for women married 20-24 years cannot be explained in terms of the "U-curve" o 0 C) o 0 0 0- -1 - 717 1
